The URL can be used to link to this page

Home My WebLink AboutContract 26684crrY �E��r��Y ��` CO�ITRAGT f�(p , Between CITY OF FORT WORTH and WASTE MANAGEMENT, INC. For DISPOSAL OF CONSTRUCTION DEBRIS AND ASBESTOS- CONTAMINATED DEBRIS RELATING TO PROJECT XL PHASE I � , - �; � �,�� , � � , ;�?)��<<�. +/�'�i��U��(��(i))��;�� (� � �`��r' (C�� •.tl� C�1 �� , J �� .: � �� °�L�� i� l}°� )I�tli�� ir1�ti'? � � � . ,.r�� ,. �+.�� - _ DEPARTMENT OF ENVIRONMENTAL MANAGEMENT APRIL, 2001 City of Fo�t Worih, Texas �l�A��r �nd �„�u�nc;r c��rtenu�v�c��n DATE REFERENCE NUMBER LOG NAME PAGE 3/27/01 **C-18521 52WASTE 1 of 2 SUBJECT APPROPRIATION ORDINANCE AND AWARD OF CONTRACT TO WASTE MANAGEMENT INDUSTRIAL SERVICES, INC. FOR WASTE DISPOSAL ON PROJECT XL RECOMMENDATION: It is recommended that the City Council: 1. Approve the transfer of $14,900 in undesignated funds in the Environmental Management Operating Fund to the XL Project in the Environmental Management Project Fund; and 2. Adopt the attached appropriation ordinance increasing estimated receipts and appropriations in the Environmental Management Project Fund in the amount of $14,900 from available funds; and 3. Authorize the City Manager to execute a contract with Waste Management Industrial 5ervices, Inc. for disposal waste services at a total cost not to exceed $14,900 in connection with Project XL. DISCUSSION: Project XL stands for "eXcellence and Leadership", and was created in March 1995 as part of the Reinventing Environmental Regulation initiatives. XL projects are real world tests of innovative strategies that achieve cleaner, more cost-effective results than traditional regulatory approaches. Each project that is accepted into the program involves the granting of regulatory flexibility by the Environmental Protection Agency (EPA) in exchange for an enforceable commitment by a community to achieve better environmental results than would have been attained through full compliance with existing or anticipated future regulations. In recent years, the City has been waging a stepped up war on urban blight through enforcement of building maintenance codes and through special emphasis on property redevelopment and sustainable community issues. The City submitted its Project XL proposal to the EPA in September 1999. The proposal entitled "Asbestos Management in the Demolition of Substandard Structures as a Nuisance Abatement", sought regulatory flexibility from the EPA to proceed with the demolition of substandard structures that are not in danger of imminent collapse without prior removal of regulated asbestos containing materiaL In January 2000, the EPA issued a conditional letter of selection. On September 29, 2000, in a ceremony at City Hall, an agreement was signed by the City of Fort Worth, the Texas Department of Health, and the Environmental Protection Agency to begin Phase I of the City's XL project. In February 2001, the City received EPA approval to use 2615 Ennis Avenue as the sife for Phase I demolition. Waste Management Industrial Services, Inc. was selected for disposal based on their current. Texas Natural Resource Conservation Commission certification as an "asbestos"-receiving landfill. Additionally, they are located in very close proximity to Fort Worth, thus reducing transportation charges. The provisions in Chapter 252 of the Texas Local Government Code do not apply because it is a "procurement necessary to preserve or protect the public health or safety of the municipality's residents." � , _ C'ity of'�'o�°t i�'o�t�i, 7'exas n ue��l � �cAt1a DATE REFERENCE NUMBER LOG NAME PAGE 3/27/01 *�C-18521 52WASTE 2 af 2 SUBJECT APPROPRIATION ORDINANCE AND AWARD OF CONTRACT TO WASTE MANAGEMENT INDUSTRIAL SERVICES, INC. FOR WASTE DISPOSAL ON PROJECT�XL FISCAL INFORMATION/CERTIFICATION: The Finance Director certifies that upon approvai of recommendation No. 1 and adoption of the attached appropriation ordinance, funds will be available in the current capital budget, as appropriated, of the Environmental Management Project Fund. CB:k Submitted for City Manager's FUND ACCOUNT CENTER AMOUNT CITY SECRETARY Office by: (to) R101 531060 052200110000 $14,900.0o App�OVED Charles Boswell 8511 _ _ _ CITY COUNCIL Originating Department Head: - Brian Boerner 8079 (from) , MAR 27 2QQ� R103 538070 0521100 $14,900.00 �} Additional Information Contact: R101 531060 052200110000 $14,900.00 �"� ��`�'Q""� Brian Boerner g079 C'�:y of Fort'�Voith�?'Qxaw � - Adopted Qrdinanc� iVo, � CIiY SECR�TARY , / CONTRACT NO. �� STATE OF TEXAS KNOW ALL PERSONS BY THESE PRESENTS COUNTY OF TARRANT CONTRACT BETWEEN THE CITY OF FORT WORTH, TEXAS, AND WASTE MANAGEMENT OF TEXAS, INC. A SUBSIDIARY OF WASTE MANAGEMENT, INC. FOR DISPOSAL OF CONSTRUCTION DEBRIS AND ASBESTOS-CONTAMINATED DEBRIS RELATING TO PROJECT XL PHASE I This contract is made by and between the City of Fort Worth, a home-rule municipal corporation situated in Tarrant and Denton Counties, Texas, hereinafter "City", acting by and through Charles Boswell, its duly authorized Assistant City Manager, and Waste Management of Texas, Inc. a subs}'diary of Waste Management, Inc., hereinafter "Contractor", acting by and through �� ,q��il� � , its duly authorized �;� �S�oa Mraw��,� In consideration of the mutual promises and benefits of this contract, the City and Contractor agree as follows: 1. TERM This contract shall be effective from the date of its execution by both parties, until the completion of all Contractor services as provided in Section 2, or for a period not to exceed one (1) year from date of execution, whichever occurs first. 2. SCOPE OF CONTRACTOR'S SERVICES A. Contractor agrees to furnish on an as is needed basis all labor, materials, equipment, and superintendence necessary for the disposal of non-contaminated construction debris and for the disposal of friable asbestos-containing or asbestos-contaminated debris from the demolition of a stnicture by City under Phase I of its Project XL project. Disposal shall be at the Westside Landfill, 3500 W. Linkcrest Drive, Aledo, Texas, hereinafter Contract beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Construction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I ��, -_ -� � f� ., , i �,� ,.. � p ��U��c,�:_� ��G,�� : ��;; �l ��r r� ��:�� :� �;�1�1� r �,pn^ r��rf� �-r(�til I�� u`'J`:'i�'i�ll� uL��S � � o Page 1 "landfill." Contractor shall complete two master waste profiles for disposal at landfill, which waste profiles shall be acceptable to Contractor upon coordination with the Department of Environmental Management. Prior to the project, the generation location will be added to the master profile. For tracking and management purposes at the working face of the landfill, each load will be manifested for waste identification. B. Contractor shall adhere to the provisions of the Quality Assurance Project Plan (QAPP) applicable to disposal. A copy of the QAPP is attached hereto as Appendix A, and incorporated fizlly into this contract. C. Contractor certifies that it has and will maintain during the term of this contract, current and appropriate federal, state, and local licenses, permits, and certifications necessary for the disposal of non-contaminated constniction debris and for the disposal of friable asbestos-containing or asbestos-contaminated debris as required by this contract. D. Contractor agrees that it shall not assign, delegate, or subcontract any of the work described in this contract without first obtaining express written approval to do so from City. Contractor shall remain fully responsible for the satisfactory performance of such work and shall remain fully bound by the terms of this contract. City shall have the right to approve or reject all subcontractors retained by Contractor to perform services tinder this contract. 3. CITY RESPONSIBILITIES A. City agrees to designate a City representative to provide timely direction to the Contractor and render City decisions. B. City agrees to provide timely notice to Contractor of the start of the demolition project. C. City agrees to provide Contractor with the necessary information to properly profile the waste stream. Additionally, City will complete necessary manifests for loads of asbestos- containing material and asbestos-contaminated material. 4. COMPENSATION A. In consideration for the work performed by Contractor under this contract, City shall pay Contractor a sum not to exceed fourteen thousand nine hundred dollars ($14,900.00.) Contract beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Construction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I Page 2 � .._ ;,� ��c��-o�-,�_�� c��,��Y r.� ;�, ^ , ��(1�,� ��r��,G�C��'aG�� 1 C; J��1,�; � � ii` �,u�f il ��J �I� � �15G� �v..�L�p c Payment shall be based on the unit price charges as set forth below: 1 � Non-contaminated construction debris (C&D) _$6.50 per manifested cubic yard plus fiiel surcharge. Friable asbestos-containing or asbestos-contaminated debris =$12.00 per manifested cubic yard plus fuel surcharge. Note: Fuel Surcharge is calculated at the first of each month and fluctuates based on the cost of filel the previous year. This charge is not to exceed 2%. The fuel surcharge for Apri12001 shall be . f3� o B. City shall make payment within thiriy (30) days of receiving a correct invoice from Contractor. City is responsible for notifying Contractor of any questions concerning an invoice. In the event of a disputed or contested billing, only that portion so contested will be withheld from payment, and the undisputed portion will be paid. The City will exercise reasonableness in contesting any bill or portion thereof. No interest will accrue on any contested portion of the billing until mutually resolved. C. Contractor shall receive no compensation for delays or hindrances to the work, except when direct and unavoidable extra cost to Contractor was caused by City's failure to provide information, if any, which it is required to do. When extra compensation is claimed, a written statement thereof shall be presented to the City. 5. INSURANCE The Contractor certifies it has, at a minimum, current insurance coverage as detailed below and will maintain it throughout the terms of this contract. Prior to commencing work, the Contractor shall deliver to City, certificates documenting this coverage. The City may elect to have the Contractor submit its entire policy for inspection. I1 : Commercial General Liabilitv Insurance -$1,000,000.00 per occurrence Automobile Liabilitv Insurance -$1,000,000.00 per accident (1) This policy shall include pollution coverage; that is, it shall not have the pollution exclusion. Plus, the policy shall cover loading, unloading and transporting materials collected under this contract. Contract beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Consh-uction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I �, , ; Page 3 ... ^ � ��<<r��:0%�N, [�'t�c�i��1D �� ;�,'-1�y -,r � rr,='��;(;1�;��f�,�b7 l� I( r�ir-;.,��;; U,: Il L�. '-�., �!''-1 „�"� �'r�J� �1,��(No _,„i,p C. Worlcer's Com�ensation Insurance - Statutory limits, plus employer's liability at a minimum of $1,000,000.00 each accident; $1,000,000.00 disease - policy limit; and $1,000,000.00 disease - each employee. D. Environmental Im�airment Liability (EILI and/or Pollution Liabilitv -$1,000,000.00 per occurrence. Coverage must be included in policies listed in items A and B above; or, such insurance shall be provided under separate policy(s). Liability for damage occurring while loading, unloading and transporting materials collected under the contract project shall be included under the Automobile Liability insurance or other policy(s). NOTE: BE'I'WEEN A AND D ABOVE, ANY POLLUTION EXPOSURE, INCLUDING ENVIRONMENTAL IMPAIRMENT LIABILITY, ASSOCIATED WITH THE SERVICES AND OPERATIONS PERFORMED UNDER THIS CONTRACT SHALL BE COVERED; IN ADDITION TO SUDDEN AND ACCIDENTAL CONTAMINATION OR POLLUTION LIABILITY FOR GRADUAL EMISSIONS, CLEANUP COSTS SHALL BE COVERED. E. The following shall pertain to all applicable policies of insurance listed above: 1. Additional Insured Endorsement: "The City of Fort Worth, its officers, agents, employees, representatives, and volunteers are added as additional insureds as respects operations and activities of, or on behalf of the named insured, performed under contract with the City of Fort Worth." Reasonably equivalent terms may be acceptable at the sole discretion of the City of Fort Worth. 2. Subcontractors shall be covered under the Contractor's insurance policies or they shall provide their own insurance coverage; and, in the latter case, documentation of coverage shall be submitted to the Contractor prior to the commencement of worlc and the Contractar shall deliver such to the City. 3. Prior to commencing work under the contract, the Contractor shall deliver to the City of Fort Worth insurance certificate(s) documenting the insurance required and terms and clauses required. 4. Each insurance policy required by this agreement shall contain the following clauses: "This insurance shall not be canceled, limited in scope or coverage, or non-renewed until after thirty (30) days prior written notice has been given to the Director of Environmental Management, City of Fort Worth, 1000 Throcicmorton, Fort Worth, Texas 76102." Contract between the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Construction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I �^�--`�""--- �� ` �� '', J �I� ��,�� 2� �� y,��"�f J'� �:;��5 ��'��„ls��_j��k, � �r��(T'�� -�ij� ) ,' j�'• ti1�1 hj ': �r I V� ,/ �s�;�ilj c;;�l"_�:il,:�'�1a��1paU ,�� L��: r[ f �i.�4�".�'�'j— { �'� II �o �rL�i��'�u��p V �ni�i Page 4 5. The insurers for all policies must be approved to do business in the State of Texas and be currently rated in terms of financial strength and solvency to the satisfaction of the Director of Risk Management for the City of Fort Worth. The City's standard is an A. M. Best Key rating A:VII. 6. The deductible or self-insttred retention (SIR) affecting the coverage required shall be acceptable to the Risk Manager of the City of Fort Worth; and, in lieu of traditional insurance, alternative coverage maintained through insurance pools or risk relations groups must be also approved." 6. INDEMNIFICATION A. For purposes of this contract, the phrases "Environmental Damages" and `Bnvironmental Requirements" shall be defined as stated below: 1. Efavirofzmentczl Danaages shall mean all claims, judgments, damages, losses, penalties, fines, liabilities (including strict liability), encumbrances, liens, costs, and expenses of investigation and defense of any claim, whether or not such claim is ultimately defeated, and of any good faith settlement or judgment, of whatever kind or nature, contingent or otherwise, matured or unmatured, foreseeable or unforeseeable, including without limitation reasonable attorney's fees and disbursements and consultant's fees, any of which are incurred as a result of handling, collection, transportation, storage, disposal, treatment, recovery, and/or reuse of waste pursuant to this contract, or the existence of a violation of environmental requirements pertaining to, and iricluding without limitation: a. Damages for personal injury and death, or injury to property or natural resources; b. Fees incurred for the services of attorneys, consultants, contractors, experts, laboratories and all other costs in connection with the investigation or remediation of such wastes or violation of environmental requirements including, but not limited to, the preparation of any feasibility studies or reports of the performance of any cleamip, remediation, removal, response, abatement, containment, closure, restoration or monitoring work required by any federal, state or local governmental agency or political subdivision, or otherwise expended in connection with the existence of such wastes or violations of environmental requirements, and including without limitation any Contract beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Construction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I —'� �-- - "—�" '� Page 5 �i ��I' �; „., ;� -,,_l,��.� ;',; if �i li'�: uG�iL �1�: S�`: <<,/l!"a .� �� i�U i ��L�'��•�n��' II�''�lU� l� � (,� ��,<, �'�'ii ' lr�� ,1��;> l� _,�J �T� N �� l��a � �� attorney's fees, costs and expenses incurred in enforcing this contract or collecting any sums due hereunder; and Liability to any third person or governmental agency to indemnify such person or agency for costs expended in connection with this Agreement. 2. Efzviro�anzefztal reqzcirements shall mean all applicable present and fiiture statutes, regulations, rules, ordinances, codes, licenses, permits, orders, approvals, plans, authorizations, concessions, franchises, and similar items, of all governmental agencies, departments, commissions, boards, bureaus, or instrumentalities of the United States, states, and political subdivisions thereof and all applicable judicial, administrative, and regulatory decrees, judgments, and orders relating to the protection of human health or the environment, including without limitation: a. All requirements, including, but not limited to, those pertaining to reporting, licensing, permitting, investigation, and remediation of emissions, discharges, releases, or threatened releases of hazardous materials, pollutants, contaminants or hazardous or toxic substances, materials, or wastes whether solid, liquid, or gaseous in nature, into the air, surface water, groundwater, storm water, or land, or relating to the manufacture, processing, distribution, use, treatment, storage, disposal, transport, or handling of pollutants, contaminants, or hazardous or toxic substances, materials, or wastes, whether solid, liquid, or gaseous in nature; and b. All requirements pertaining to the protection of the health and safety of employees or the public. General Indemni fication: CONTRACTOR DOES HEREBY RELEASE, INDEMNIFY, REIMBURSE, DEFEND, AND HOLD HARMLESS THE CITY, ITS OFFICERS, AGENTS, EMPLOYEES AND VOLUNTEERS, FROM AND AGAINST ANY AND ALL LIABILITY, CLAIMS, SUITS, DEMANDS, OR CAUSES OF ACTIONS WHICH MAY ARISE DUE TO ANY LOSS OR DAMAGE TO PERSONAL PROPERTY, OR PERSONAL INJURY, AND/OR DEATH OCCURRING AS A CONSEQUENCE OF THE PERFORMANCE OF THIS CONTRACT, WHEN SUCH INJURIES, DEATH, OR DAMAGES ARE CAUSED BY THE SOLE NEGLIGENCE OF CONTRACTOR, ITS OFFICERS, AGENTS, OR EMPLOYEES, OR THE JOINT NEGLIGENCE OF CONTRACTOR, ITS OFFICERS, AGENTS, OR EMPLOYEES, AND ANY OTHER PERSON OR ENTITY. Contract beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Construction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I �, -� - �_ —� �'��G�l��!�i��:� �`�','��r(l�fc1[C! � Y �' ��° ("k�;"'r��,'�_;'��.M ,: ,11 u e;ily,.;.(h��t�� 1 �(}�5j�� IfrP,rvI;;�)�Iri, �.� ,:f�:�a;)�,�� ll UN VU1�:�L�'I�if�- I1L7UUo Page 6 C. EnvironmentalIndemnification: CONTRACTOR DOES HEREBY RELEASE, INDEMNIFY, DEFEND, REIMBURSE, AND HOLD HARMLESS THE CITY, ITS OFFICERS, AGENTS, EMPLOYEES AND VOLUNTEERS, AGAINST ANY AND ALL ENVIRONMENTAL DAMAGES AND THE VIOLATION OF ANY AND ALL ENVIRONMENTAL REQUIREMENTS RESULTING FROM THE HANDLING AND DISPOSAL OF NON-CONTAMINATED CONSTRUCTION DEBRIS AND FOR THE HANDLING AND DISPOSAL OF FRIABLE ASBESTOS- CONTAINING OR ASBESTOS-CONTAMINATED DEBRIS OCCURRING AS A CONSEQUENCE OF THE PERFORMANCE OF THIS CONTRACT, WHEN SUCH DAMAGES OR VIOLATIONS ARE CAUSED BY THE SOLE NEGLIGENCE OF CONTRACTOR, ITS OFFICERS, AGENTS, OR EMPLOYEES, OR THE JOINT NEGLIGENCE OF CONTRACTOR, ITS OFFICERS, AGENTS, OR EMPLOYEES, AND ANY OTHER PERSON OR ENTITY. D. The obligat'ions of the Contractor under this section shall include, but not be limited to, the burden and expense of defending all claims, suits and administrative proceedings (with counsel reasonably approved by City), even if such claims, suits or proceedings are groundless, false, or fraudulent, and conducting all negotiations of any description, and paying and discharging, when and as the "same become due, any and all judgments, penalties or other sums due against such indemnified persons. E. Upon learning of a claim, lawsuit, or other liability that Contractor is required hereunder to indemnify City, City shall provide Contractor with reasonably timely notice of same. F. The obligations of the Contractor under this section shall survive the expiration of this Agreement and the discharge of all other obligations owed by the parties to each other hereunder. G. In all of its contracts with subcontractors for the performance of any work under this contract, Contractor shall require the subcontractors to indemnify the City in a manner consistent with this section. H. In the event City receives a written claim for damages against the Contractor or its subcontractors prior to final payment, final payment shall not be made until Contractor either submits to City satisfactory evidence that the claim has been settled and/or a release from the claimant involved, or provides City with a letter from Contractor's liability insurance carrier that the claim has been referred to the insurance carrier. Contract beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Conshuction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I _- , I" r; ` cl^l,_. Page 7 :;��1 J��;'�� _LS P;�.�[��_ (��/Itir�J (,11�'-'' "; .. .,i� ,� ��;�r� �::i,u�i`;��;�G�� }� �,i„-.�•.,.;�'�, r.�(�q Gi�: ��`��,�:�.�?i�� Sz;�� . WARRANTY Contractor warrants that it understands the currently known hazards and suspected hazards that are present to persons, property and the environment by providing disposal services of asbestos- containing debris. Contractor further warrants that it will perform all services under this contract in a safe, efficient and lawful manner using industry accepted practices, and in full compliance �vith all applicable state and federal laws governing its activities. Contractor also warrants that it is under no restraint or order that would prohibit performance of services under this contract. S. TERMINATION A. City may terminate this contract, with or without cause, by giving 30 days written notice to Contractor, provided that such termination shall be without prejudice to any other remedy the City may have. In the event of termination, any work in progress will continue to completion unless specified otherwise in the notice of termination. B. If the City terminates this contract under paragraph A of this section, City shall pay Contractor for all services performed prior to the termination notice. C. All completed or partially completed original documents prepared under this contract shall become the property of the City when the contract is terminated, and may be used by the City in any manner it desires; provided, however, that the Contractor shall not be liable for the use of such documents for any purpose other than as described when requested. D. In the event either party defaults in the performance of any of its obligations under this contract, misrepresents to the other a material fact, or fails to notify the other party of any material fact which would affect the party's performance of its obligations hereunder, the non-defaulting party shall have a right to terminate this contract upon giving the defaulting party written notice describing the breach or omission in reasonable detail. The defaulting party shall have a thirty (30) day period commencing upon the date of notice of default in which to effect a cure. If the defaulting party fails to effect a cure within the aforesaid thirty (30) day period, or if the default cannot be cured, the contract shall terminate as of the date provided in the notice of default. Contract beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Construction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I �'J�G�'��/ �; �`?I��; ;0)�'� „�� ir �ij �r �;(�;�;^�(;��1i!`�Ur`'�1� �= •�� r:� , ..�, � `l �;,r;;`i,L''ij��,� ��5�, 0 ��:ill��.L�IIJ IV,�//\10 f Page 8 C� 9. DEFAULT A. Contractor shall not be deemed to be in default because of any failure to perform under this contract, if the failure arises from causes beyond the control and without the fault or negligence of Contractor. Such causes shall include acts of God, acts of the public enemy, acts of Government, in either its sovereign or contractual capacity, fires, flood, epidemics, quarantine restrictions, strikes, freight embargoes, and unusually severe weather. B. If at any time during the term of this contract the work of Contractor fails to meet the specifications of the contract, City may notify Contractor of the deficiency in writing. Failure of Contractor to correct such deficiency and complete the work required under this contract to the satisfaction of City within ten days after written notification shall result in termination of this contract. Contractor shall pay all costs and attorneys fees incurred by City in the enfarcement of any provision of this contract. C. The remedies provided for herein are in addition to any other remedies available to City elsewhere in this contract. 10. RIGHT TO AUDIT A. Contractor agrees that the City shall, until the expiration of three (3) years after final payment under this contract, have access to and the right to examine and photocopy any directly pertinent books, documents, papers and records of the Contractor involving transactions relating to this contract. Books, documents, papers and records of the Contractor marked as proprietary shall be kept confidential. Contractor agrees that the City shall have access during normal working hours to all necessary Contractor facilities and shall be provided adequate and appropriate workspace in order to conduct audits in compliance with the provisions of this section. The City shall give Contractor reasonable advance notice of intended audits. B. Contractor further agrees to include in all its subcontractor agreements hereunder a provision to the effect that the subcontractor agrees that the City shall, until the expiration of three (3) years after final payment under the subcontract, have access to and the right to examine and photocopy any directly pertinent books, documents, papers and records of such subcontractor, involving transactions to the subcontract, and further, that the City shall have access during normal working hours to all subcontractor facilities, and shall be provided adequate and appropriate work space, in order to conduct audits in compliance Contract beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Construction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I ��------ -------�—�—�- � ,� f ,. 1 �� iIIS�a I( ,. `r�� ) � Page 9 „ � �l ll U, i;t � ;i ��01�::, U�_��;�,�J,,�. r. �7;�,r, (.' r�. 1 „P/ II �� u ,U �'I�;�;'��Cll�i ilh,1(j� � I�r�J n,,,�f.;r..,::.ff�llq I�!,51� + 1` � S l� U] l'��`:�L'_ III�� ] r tl with the provisions of this article together with subsection C. hereof. City shall give subcontractor reasonable advance notice of intended audits. C. Contractor and subcontractor agree to photocopy such documents as may be requested by the City. The City agrees to reimburse Contractor and/or subcontractor for the cost of copies at the rate published in the Texas Administrative Code in effect as of the time copying is performed 11. INDEPENDENT CONTRACTOR It is understood and agreed by the parties hereto that Contractor shall perform all work and services hereunder as an independent contractor, and not as an officer, agent, servant or employee of the City. Contractor shall have exclusive control of and the exclusive control of and the exclusive right to control the details of all the work and services performed hereunder, and all persons performing same, and shall be solely responsible for the acts and omissions of its officers, agents, servants, employees, contractors, subcontractors, licensees and invitees. The doctrine of responcleat sicperioY shall not apply as between City and Contractor, its officers, agents, employees, contractors and subcontractors, and nothing herein shall be construed as creating a partnership or joint enterprise between City and Contactor. 12. NON-DISCRIMINATION A. During the performance of this contract, Contractor agrees not to discriminate against any employee or applicant for employment because of race, religion, color, sex or national origin, except where religion, sex or national origin is a bona fide occupational qualification reasonably necessary to the normal operation of the Contractor. Contractor agrees to post in conspicuous places, available to employees and applicants for employment, notices setting forth the provisions of the non-discrimination clause. B. Contractor also agrees that in all solicitations or advertisements for employees placed by or on behalf of this contract, that Contractor is an equal opportunity employer. C. Notices, advertisements, and solicitations placed in accordance with federal law, rule or regulation shall be deemed sufficient for the purpose of ineeting the requirements of this section. Contract between the City of Fort Worth, Texas, and Waste Management of Texas, Inc, a subsidiary of Waste Management, Inc. For Disposal Of Constniction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I j- ---- --- ------- - Page 10 � �. � �� fe 4; � D� ''�� r.'� u��;l�. lf:,l� (C'i�,�� j� f�''��° C ��^'��r �,� � +/ ;:,�� : ��,:��,�G�?��l G�J �':�-:��;�-!�:'`�?!�i�p `V'��,�� 13. GOVERNING LAW The City and Contractor agree that the laws of the State of Texas shall govern the validity and constniction of this contract, except where preempted by federal law. 14. RIGHTS AND REMEDIES NOT WAIVED In no event shall the making by the City of any payment to Contractor constitute or be constnled as a waiver by the City of any breach of covenant, or any default which may then exist, on the part of Contractor, and the malcing of any such payment by the City �vhile any such breach or default exists shall in no way impair or prejudice any right or remedy available to the City with respect to such breach or default. Any waiver by either party of any provision or condition of the contract shall not be construed or decreed to be a waiver of any other provision or condition of this contract, nor a waiver of a subsequent breach of the same provision or condition, unless such waiver is expressed in writing and signed by the party to be bound. 15. ENTIRETY This contract and any other documents incorporated by reference herein contain all the terms and conditions agreed to by the City and Contractor, and no other contracts, oral or otherwise, regarding the subject matter of this contract or any part thereof shall have any validity or bind any of the parties hereto. 16. ASSIGNMENT The City and Contractor bind themselves and any successors and assigns to this contract. Contractor shall not assign, sublet, or transfer its interest in this contract without written consent of the City. Nothing herein shall be construed as creating any personal liability on the part of any officer or agent of the City, nor shall it be construed as giving any rights or benefits hereunder to anyone other than the City and Contractor. Contract between the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Constniction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I �---�---...�_�e ti. � ; ,� � �a� e 11 ��i.�l��y �i�;;�� '��r<<_?)�r'O I^ � '�'+ r�i �;ii,�; �91�'�!�Uij,�„ r,_(i� �� ,' ��. L,.: �� ; L�;,;,�,��.��I��/'11 �J{C"�'�� ll] �I\J ��.: II iJ U g r 17. NOTICE Notices required to be made under this contract shall be sent to the following persons at the following addresses; provided, however, that each party reserves the right to change its designated person for notice, upon written notice to the other party of such change: If to City: Written notice shall be sent to: Brian Boerner, Director Department of Environmental Management 1000 Throckmorton Fort Worth, Texas 76102 If to Contractor: Questions should be directed to: Michael Gange, Environmental Supervisor Department of Environmental Management 1000 Throckmorton Fort Worth, Texas 76102 (817) 871-8504 / Fax (817) 871-6359 Mr. Al Latini Waste Management of Texas, Inc. a subsidiary of Waste Manageinent, Inc. 1601 S. Railroad Street P.O. Box 719 Lewisville, Texas 75067 (972) 315-5400 / Fax (972) 315-1296 18. VENUE Should any action, real or asserted, at law or in equity, arise out of the terms and conditions of this contract, venue for said action shall be in Tarrant County, Texas. 19. SEVERABILITY The provisions of this contract are severable; and if for any reason any one or more of the provisions contained herein are held to be invalid, illegal or unenforceable in any respect, the Contract beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Constniction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I �� o,� �G,��>> c�� r� � r% C' [' �� n P, �:. �� '(�r`�� G��r�l��,l �j 1Y �°��YI %�,r'�r�nryF r,_}�`•, ila U'Ul.%�.�'UOIf� Uf:34\�ry 12 invalidity, illegality or unenforceability shall not affect any other provision of this contract, and this contract shall remain in effect and be construed as if the invalid, illegal or unenforceable provision had never been contained in the contract. 20. MODIFICATION No modification of the contract shall be binding on Contractor or City unless set out in writing and signed by both parties. 21. AUTHORIZATION The undersigned officer and/or agents of the parties hereto are properly authorized officials and have the necessary authority to execute this contract on behalf of the parties hereto, and each party hereby certifies to the other that any necessary resolutions extending such authority have been duly passed and are now in fiill force and effect. IN WITNESS WHEREOF, the parties hereto have executed this agreement in triplicate originals in Fort Worth, Tarrant County, Texas. City of Fort Worth � Charles Boswell, Assistant City Manager Date: � � � / l� �- �, � � prove '�s to For and Legality: ;, � �� i ' � �..�" istant City Attorney V \\ A s�: � �� , % � c�%vLs G oria Pearson, ity Secretary �- /`� �,/ Waste�M�na�e�ent of Texas, I c. Name: � � ��+�+ � Title: _�J. —C'EXc-�s I'��v�S�u-+ M�►"r�*S�;i2 Date: y— / /-01 Witness: � �! - �� J . — � . � �� . " � Corporate Seal: �-I�'��I �Ontr.�act Authorization �J ��—VI Contra�t beriveen the City of Fort Worth, Texas, and Waste Management of Texas, Inc. a subsidiary of Waste Management, Inc. For Disposal Of Conshuction Debris And Asbestos-Contaminated Debris Relating to Project XL Phase I Page 13 �;��U`��OQ� �'��U��f� '';=��' ��:G:�� /6�C�1 �sl; �`��;,�,��D �� 0 --�—��___ Marsh USA Inc 4/6/01 1:07: PAGE 2/2 RightFAX <�... . . . ; ...:<:::�::�:<:.:�; `:«:.::.:::;: : ...:_ • :; :<.:.:::: ><;::: . .. . ,..:.;:.-. ..:: .:.:::. ....... r. ...... :. <. >::.:.::., . . .>.,.:� , t�n .: •" ' ` � .�*..a..... . ..... ::.....' � " . . .....:..:.: < ...:: .: " ? '.: , .. .: . �.�.>::::�....;.; CERTIFICATE NUMBER ....:............_.._.. - .......... - .. _.:::.-•-:�:��.�.�.��.:...�.�.:::��.i��;,�.�.�:....... --.......... ! :::: ,t :.: :. ,. :.:_: :��:.;:.�:. :.. R.�.s.:'..:`_` .:.:::::::.:.::'::::.:. : _..::..,'<` _.:......-::::::::::::::::::::....._..........- - .._........................._ _......... ... �- -- _........... -- € _ __�.��.��"'��.:.� ................�..._........ _........::::::::::::::._:.:_:._:::::.-:.:.:.:..:.................--............ .::::: - E_.......__ . .. ................_....__.........'.._...._......._....._........._.................::.,s,.::..�:::::..>�::. _';;;::;;:-:-:-r.;.:.::::.:::................_. :::::........_......__.............. .. _... . __ .. ................_........ ..._......................__...............-::-::::::::::::::..:..::::...:...._...................................... ..:::::::::,.:::_::::::::� • .:.:......................._..,...._..._.......... -......... -:.-.::::::::::::::::::. ::::.:...... 909 00 __........_...._ _...__.._ ..............................._...... ... ..._.................,. _--- HOU-000263 . ............. ...::.:..:..... ......... ....:....:. «:,....,...,.. . _ .: ,;:_ .. _...... ;::. ....:.......:...:.>,...:.:;;......,....,.....::::..:.:..>;:>..::;...::........,n. ..r...,.,...,:................. ,:.,....... __. pR��g THIS CERTIFICATE IS ISSUED AS A MATTER OF INFORMATION ONLY PND CONFER3 MARSH USA INC. NO RIGHTS UPON THE CERTIFICATE HOLDER OTHER THPN THOSE PROVIDE� IN THE WELLS FARGO PLAZA POLICY. THIS CERTIFICATE DOES NOT AMEND, EXTEND OR ALTER THE COVERAGE 1000 LOUISIANA, STE. 4000 PFFORDED BY THE POLICIES DESCRI9ED HEREIN. HOUSTON, TX 77002 COMPANIES AFFORDINO COVERAGE PHONE: 713/6540400 CaNPANY I-00-02 A AMERICAN INTERNATIONAL SPECIALTY LINES INS CO WSURED CQ�IPPNY Waste Management of North Texas B P.O. Box 719 1601 Waste Management Bivd. °QNP'°NY Lewisville, TX 76102�311 � COMPANY D _..... _.... _..::.....: .>. ....:::.. ...:.::... . _... :==::':el:�u:"::';° :::: .:: ::::::::::::<::;.<::::::::;:: :�:::: ;.:.:.:.:..:.... ;.:::< .::.:::.::... �;.... .�.._.,...... ;.: :. ,:... ,.. . . ....::> . . . .. ..,,.... . , .. , .. ... ��t�k��#�3�:::� ::.:::.::.:.:..:--:::..._::::::.:: :::::::::::<.: a�€�:e. �a�.�n.:_= _€��st .:�ss�ec�:�r6��.�r�i�te:..�i..; :;:..�r��s�_�€�fed:k�......_...... ....... ......_:..,::...::::.�:::::. _.......... _._--.:..._.._._..... -......._ _ . �.�e.�fi��.� e€�� .. .......... .......... ---- __..................�..:::::.::.::::::::::.............................. - ......�...._�Y..f�.....................................,.............................. .:...:.....:.:_..._...........:-::::::.:::,.::::::::::::::::::::::::::::::: I�:::::::::::::::::::::::::::.-.:..:.:f�...............- ........,......�..................._�..�.................. .. . .................... . THIS IS TO CERTIFY THAT POLIqES OF INSURANCE DE9CRIBED HEREIN HAVE BEEN ISSUED TO 7HE IN3IRED NAMED HEREIN FOR THE PaLICV PERIOD INDICATED. NQTIMTHSTANDING ANYREQUIREMENT, TERM OR CONDIIIQV OF ANY CONTRACT OR OTHER DOCUMENTIMTH RESPECTTO WHICH THE CERTIFICATE MAY BE ISSUED OR MAY PERTAIN, THE INSURANCE /1FFORDED 6Y THE POLICIES DESCRIBED HERBN IS SUBJECT TO ALL THE TERMS, C4VDITIONS AND EXCLUSI ONS OF 51CH POLIGES LIMITS SHOWN MAY HAVE BEEN REDUCED 8Y PWD CLAIMS GO TypEOFINSURANCE POLICYNUMBER POLICYEFFEGTIVE POLICYEXPIRATION LIMITS LTR DATE {MMIDDIYY) DATE (MMIDD/W) GENERAL LIA9ILITY GENERAL AGGREGATE $ COMMERCIPLGENERALLIABILITY PRODl1CTS-CGrVIP/OPAGG $ CLAlMSMPDE � OCCUR PERSONAL&A�VINJURY $ ` OVWi ER'S&CQVTRACTOR'SPROT EACHOCCURRENCE � FIREDAMAGE(Myonefire) � MED EXP (My me persmj $ AUTOM OBILE LIAHILITY CQIAHINED 9NGLE LIMIT $ ANY AUTO PLL O/Wi ED AllTOS 80DILYIN„URY $ (Per persrn) 9CHEDULED AUTQS HIRED Al1TOS 60DILYINJl1RY � (Per acddent] NQV-OWNED AUT0.S PROPERTY DAMAGE $ GARAGELIABILIN pi1TOONLY-EAACCl�ENT $ AN AUTO ONLY: - QTHER TH - - ANY AL1T0 - EACH ACdDENT $ AGGREGATE $ EXCESS LIABILITY EACH OCCl1RRENCE $ UMBRELLAFORM AGGREGATE $ OTHER THAN UMBRELLA FORM $ WORKERS COMPENSATION AND WCSTA - OTH-::;:::::::::>:i>:::<:=z:z?<:;_:=_:zri:<z-::::[:#:>:__:::<::: EMPLOVERS'LIA9ILITY TQ�YLIMITS ER ':_:£ _ EL EACH ACGIDENT $ THE PROPRIETORI INa EL DISEASEf+OIICY LIMIT $ PARTNERS+EJ(ECUTIVE OFFICERSARE: EXCL EL DISEASE-EACH EMPLOYEE $ Pollution Legal & PLS 819490400 01/01/00 01/01/02 � � Environmental Impairment Liab Claims Made & Reported Form 1,000,000 DEDUCTIBLE 500,000 DESCRIPTION OF OPERATIONSILOCATIONSIVEHICLESISPECIPL ITEMS (LIMITS MAY 9E SU9JECTTO �EDUCTI9LES OR RETENTIONS) THE CITY OF FORT WDRTH, ITS OFFICERS, AGENTS, EMPLOYEES, REPRESENTATIVES, AND VOLUNTEERS ARE INCLUDED AS ADDITIONA L INSURED AS REQUIRED BY WRITTEN CONTRACT, BUT ONLY FOR LIABILITY ARISING OUT OF THE OPERATIONS OF THE NAMED INSURED. ....._. . ...:... ...:.. . ..:.::,>:.:;;:.;:.;;::::.:;;:.;;;:;;;;;;;::::.;:,:: _,:::::;;;:.;:.;:.;::<::::«::_::><:;;::.>: :>..:<:»::>;::»::::<:::::««;::::-<.:_:::> _:_> :>._»:;;::.<::_ ;.. :::.............. _::-::...:........:..._:>;;;;:::<:;:_:<::.;:«<;;;;:;::;<::: �:.::»»::::>:::::::<:::::.;:::_:;;.««::;<>_;:<:><;_>::>:;::::»»>;.».»>::>;:>:«<>::::«:.:;-;;:�:::::::::._>._;>.;>;:::: :;:>: �`rE��tEE�s7'�::Ht?��'3�Ii, ..............: .. : . ..:..: :>.::::::< ..:...........:..: ,....,.:: <:: C1��4i`��3�1 . ... . . ........:. - ::-...:,.._....._...._ ..:.::::....:..:.............._..-:::::::::::::::::::::::::::::::::::::.:::::::::.:::::::::::::::,;;:::::::::::::::::::..:...::...:..........--:..::. - ::..:..__:............. ................._............................. ....,........................... _ ,<�. ..,..,...............,.,.._..........,....: - ................- .................. ........ - ' �, _-..-- ................_ ........ - _....... ..._..... _............._- ._..... -..................... _-, .....:::::::...:::::::::::::::::::::.::..:.:::::::::::...::::::::::: .::.:........:.:::..._._ .._........ - _...... ...._.._..::::::::::::::::::::: ;: ::..........# .::-, : ,... ...........................................,.,......... ........,..,...,........................, .......... .. .::: ...:,::::...:..:.:::::::: :.--:-::.::.::.:....- .. ::.:,.:...:. SHOULD ANY OF THE POLICIES �ESCRIBE� HEREIN BE C.4N LEQ F��R� THE EXPIR��;4Tf�R�V4T�� EOF, �� i ii�(,I^ �,�.,. L+,I f ��( � ., THe INSURER AFfOR01NG COVERACf VNLL ENDEAVQi TO�( MAl4� �Q\�,�Y� {71`� � I �O THE CITY OF FT. WORTH AT DEPARTMENT OF � �! �� s , r,, CERTIFICATE HOLOER NP,tvEO HEREIN, 8Uf FAILURE TO MNL 1R1CIj��IJI�`P�G'� 1;lA�P, � QO� IG4 ON QZ ENVIRONMENTAL MANAGEMENTADMINISTRATION P �:�� U b ��r:!',,��;��'��/��:)'�� 908 MONROE STREET7TH FLOOR LIA8ILRYOFANYKINOUPONTHEINSURERAFFOR�INGCOVEF�AGE,RSAG,�'S ii F� E AT E5 ATTN.: MICHAELGANGE � � c� ��i';;i��'�;,r'S�(��' i�r;'';/ FT. V,�ORTH, TX 76102-6311 h�aRSH usaiNc. `' —•" ev: Marlene McLoad `�� `rl.�,�-p.�,�, VM _ - - - - - - - - - - - - - - - - - - :.'::a:>tse><:�::Cs�i`�El�rt's4:::>>:; s: :_>>:�`:=::»»>:::>::>:::> - . .. ................:�;:Si�I:P"s:'trlttil. .. . .. ::.:;EII.f:lftt4Q4�"i"t:�3>�1s3:::;'::»:ir:::::::3y: �:i:�:>:�;i>3i»'ts��_��sf:;::`; :'::>3::2:>::�::X£i: - - - - ... ......................... <:�:Ra � f i 1E:55 APR 03, 2��1 TEL N0: 713-821-1557 #3623 PAGE� 2i3 CERTIFICATE OF INSURANCE "p` 4/3/2001�� ' PRO�UCER THIS CERTIFICATE IS ISSUED AS A MATTER OF INFORMATION Lockton Insurance A,�gency of Houston, Inc. ONLY AND CONFERS NO RIGHTS UPON THE CERTIFICATE 58A7 san Felipe, 17 Floor HOLDER. THIS CERTIFICATE DOES NOT AMEND, EXTEND OR Houston, TX 77057 866-260-3538 (Toll Free Phone) ALTER THE COVERAGE AFFORDED BY THE POLICIES BELOW. 866-492-1055 (Toll Free Fex) INSURERS AFFORDING COVERAGE INSURED: WASTE MANAGEMENT, INC. and InsiirPr A• Par.ific: Fmnlnvprs Ins�iranr.P C:mm�anv l/�{aste Management of North Texas Insurer B: Continental Casualtv Companv 1601 Waste Management Boulevard Insurer C: ACE American Insurance Companv PO Box 719 Insurer D: Indemnity Insurance North America Lewisville, TX 75067 Insurer E: National Union Fire Insurance Co. of PA COVERAGES THE POLIGIES OF INSIJRANCE LISTED BELOW HAVE BEEN ISSIJED TO THE INSl1RED NAMED ABOVE FOR THE POLIGY PERIOD INDIGATED. NOTWITHSTANDING ANY REQUIREMENT, TERM OR CONDITION OF ANY CONTRACT OR OTHER DOCUMENT WITH RESPECT TO WHICH THIS CERTIFICATE MAY BE ISSUED OR MAY PERTAIN, THE INuURANCE AFFORDED BY THE POLICIES DESCRIBED HEREIN IS SUBJECT TO ALL THE TERMS, EXCLUSIONS AND CONDITIONS OF SUCH POLICIES. AGGREGATE LIMITS SHOWN MAY BE EXHAUSTED BY PAID CLAIMS. �ry�R TYPE OF INSURANCE POLICY NUMBER EFFECTIVE DATE �>�IRATION LIMITS LTR DATC ; GENERAL LIABILITY EACH OCCURRENCE $ 2,000,000 ,4 X COMMCFCIAI GCNCRAL LIADILITY FIRE DAMAGE (nrrvoNeFiRe) � 1,000,000 X XCU WCLUDED x ISO FORM GG 00 0'I '10 93 GCN'L AGGFCGATC LIMITAPPLIC� PCR: x f'ROJECT .. LOCATION A X �YAUIU ALL GWNED AUTOS � ^uCl ICDULCD AUTO^u X HIREDhUTOS � X NON OWNED AUTOS � X M(:S-yU B EXCESS LIABILITYNMBRELI C X OCCURRENCE E cv�inns nnw�t I WORKERS' COMPENSATION I fl and EMPLOYERS LIABILITY HDO G19902559 1/1/2001 1/1/2002 MED EXP�rer,rer.soN� PERSONAL &ADV INJURY $ 2,000,000 GENERALAGGREGATE $ 2,000,000 PRODUGTSICOMP.OP.AGG $ 4,000,000 NED SINGLE LIMIT (EACH ACCIDENT) ISA H07686031 � 1/1/2001 � 1/1/2002 CUP-247892731 EACH OCCURRENCE XOOG 19902675 1/1/2001 1/1/2002 AGGREGATE 346 71 06 $ 5,000,0 $ WORKERS' COMPENSATION STATUTORY WLR C42982453 1/1/2001 1/1/2002 EL EACH ACCIDENT $ 1,000,00 SCF C42982532 (WI) EL DISEASE-EA EMPLOYEE $ �,OOO,OO EL DISEASE-POLICY LIMIT $ 1,000,00 REMARKS: DESCRIPTION OF OPERATIONS/LOCATIONSNEHICLES/EXCLUSIONS ADDED BY ENDORSEMENT PROVISIONS: CHECK � BLANKET W41VER OF SUBFtOCATION IS CRANTED IN FAVOR OF CERTIFICOTE HOLOER ON GLL POLICIES WHERE AN� TO THE E%TENT REQUIRED BY WRITTEN CONTRACT. LiUX i � CERTIFICATE HOL�ER IS NAMED AS AN ADDITIONAL INSURED (EXCEPT FOR WORKERS' COMP/EL) WHERE AND TO THE EXTENT REQUIRED BY WRITfEN CONTRACT. CERTIFICATE HOLDER: CANCELLATION: EHOULD ANY OF THE ABOVE DE6CRIBED f'OLICIE.�. BE CANCELLED BEFORE THE EXPIR/�TION DhTE THEREOF, THE ISSUING INSURER WILL MhIL 30 DnYS WRITTE�! NOTICE i0 THE CERTIFICATE HOLDER NAMED TO THE LEFT. EXCEPT 10 DAYS NOTICE FOR � _ _.....__...- ' city ot Fort wortn at ' Departmentof Environmental ManagementAdministration AUTHORIZED REPRESENTATIVE: � � Attn: Michael Gange �� 908 Monroe Street, 7th Floor --- -- i Fort Worth, TX 76102-6311 i i�� Ki��r'r � ��, � ��f2/�r,r�ra (``'il7l�r �'f�ii,,;�[��VUdpll � i! p Iil C. ., ��' ��'��,:� i��o �C��o l�: L'''� ��: �6��6 �PR �3, z001 � ; OLICY NUMBER: � i H DO G 19902559 TEL NO: 713-$21-1557 By: Pacific Employers Insurance Company #36�3 PAGE� �i3 COMMERCIAL GENERAL LIABILITY THIS ENDORSEMENT CHANGES THE POLICY. PLEASE READ IT CAREFULLY. ADDITIONAL INSURED - OWNERS, CONTRACTORS (FORM ;. This endorsement modifies insurance provided under the following: � MMERCIAL GENERAL LIABILITY COVERAGE PART r ' SCHEDULE LESSEES OR B) Named Insured: i Waste Management of North Texas 1601 Waste Management Boulevard �O Box 719 Lewisville, TX 75067 i Name of Person or Organization: The City of Fort Worth, its officers, agents, employees, representatives and volunteers i (If no entry appears above, information required to complete this endorsement would be shown in the Declarations as applicable to this endorsement.) WHO IS AN INSURED (Section 11) is amended to include as an insured the person or organization shown in the Schedule, but only with respect to liability arising out of "your work" for that insured by or for you. Such insurance as is afForded by this policy for the additional insured shown in the Schedule of this endorsement shall apply as primary insurance and we will not seek contribution from any other insurance of self-insurance maintained by �uch additional insured. AUTHORIZED REPRESENTATIVE: �� C'�e"' —�— Meeting America's Needs for Ezperienced and Comprehensive Environmental Management QAPP Section A August 17, 2000 Revision 1 Page 1 of 30 QUALITY ASSURANCE PROJECT PLAN AMBIENT AIR MONITORING FOR ASBESTOS DURING DEMOLITION OF SUBSTANDARD STRUCTURES IN CITY OF FORT WORTH, TEXAS (PROJECT XL) Prepared by: Environmental Quality Management, Inc. 1310 Kemper Meadow Drive, Suite 100 Cincinnati, Ohio 45240 Prepared for: City of Fort Worth Department of Environmental Management 1000 Throckmorton Street Fort Worth, Texas 76102-6311 A2 TABLE OF CONTENTS Section QAPP Section A June 2, 2000 Revision 0 Page 4 of 30 Pa�e A Project Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 of 30 Al Title and Approval Sheet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A2 Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A3 Distribution List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A4 Project Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.4.1 City of Fort Worth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A4.2 Industrial Hygiene & Safety Technology, Inc . . . . . . . . . . . . . . . . A4.3 Microscopy Laboratory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A4.4 Demolition Contractor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AS Problem Definition/Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A5.1 Background ........................................ A5.2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A6 Project Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.6.1 Technical Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.6.1.1 Demolition of Buildings . . . . . . . . . . . . . . . . . . . . . . . A.6.1.2 Air Sampling During Demolition of Buildings ...... A.6.1.3 Air Sampling During Landfilling of Demolition Debris A.6.2 Personnel .......................................... A.6.3 Project Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A7 Quality Objectives and Criteria for Measurement Data . . . . . . . . . . . . . . A7.1 Primary Project Objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . A7.2 Criteria for Acceptance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A7.3 Statistical Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A7.4 Precision and Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A7.5 Completeness . . .... . . . .. . .. . : . .. .. . .. . .. .. . . . . . . . . . . A7.6 Representativeness ......................... .. ........ A7.7 Comparability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A7.8 Analytical Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A8 Special Training Requirements/Certification . . . . . . . . . . . . . . . . . . . . . . A8.1 Field Personnel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A8.2 Laboratory Personnel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A9 Documentation and Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A9.1 Field Operations Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.9.1.1 Air Sample Documentation . . . . . . . . . . . . . . . . . . . . A.9.1.2 1Vieteorological Measurements . . . . . . . . . . . . . . . . . . A.9.1.3 Photo Documentation . . . . . . . . . . . . . . . . . . . . . . . . A9.2 Chain-of-Custody Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 2 of 30 , 4 of 30 10 of 30 11 of 30 11 of 30 13 of 30 13 of 30 13 of 30 14 of 30 14 of 30 15 of 30 16 of 30 16 of 30 17 of 30 17 of 30 18 of 30 18 of 30 18 of 30 20 of 30 20 of 30 20 of 30 ZO of 30 22 of 30 22 of 30 23 of 30 23 of 30 23 of 30 25 of 30 25 of 30 25 of 30 26 of 30 26 of 30 26 of 30 26 of 30 29 of 30 29 of 30 A2 TABLE OF CONTENTS (continued) Section QAPP Section A August 17, 2000 Revision 1 Page 5 of 30 Pa�e A9.3 Laboratory Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 of 30 < A.9.3.1 TEM Specimen Examination and Data Recording ... 29 of 30 A.9.3.2 Test Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 of 30 B Measurement/Data Acquisition . . . . . . . e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 of 28 : I� : : Sampling Design ............................................ B.l.l Sampling Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.1.2 Air Sampling Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.1.3 Particulate Loading Pilot Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.1.4 Soil Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.1.5 Moisture Content of ACM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.1.6 Water Used for Wetting Structure/Debris . . . . . . . . . . . . . . . . . . . Sampling Methods Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.2.1 Air Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.2.2 Meteorological Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.2.3 S oil S ampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.2.4 WaterSampling ..................................... Sample Custody Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.3.1 Field Chain-of-Custody . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.3.2 Laboratory ......................................... Analytical Methods Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.4.1 AirSamples ........................................ B.4.1.1 TEM Specimens Preparation . . . . . . . . . . . . . . . . . . . B.4.1.2 Measurement Strategy . . . . . . . . . . . . . . . . . . . . . . . . B.4.2 Soil Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.4.3 Moisture Content of ACM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.4.4 Water Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Quality Control Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B.5.1 Field Quality Control Checks . . . . . . . . . . . . . . . . . . . . . . . . . . . B.5.2 Analytical Quality Control Checks . . . . . . . . . . . . . . . . . . . . . . . . B.5.2.1 Quality Control Check of Filter Media . . . . . . . . . . . . B.5.2.2 Blank Contamination . . . . . . . . . . . . . . . . . . . . . . . . . B.5.3 Analytical Precision and Accuracy . . . . . . . . . . . . . . . . . . . . . . . . B.5.3.1 Replicate Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . B.5.3.2 Duplicate Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . B.5.4 Verification Counting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 of 28 1 of 28 1 of 28 6 of 28 6 of 28 7 of 28 7 of 28 8 of 28 8 of 28 8 of 28 10 of 28 10 of 28 12 of 28 12 of 28 12 of 28 14 of 28 14 of 28 14 of 28 14 of 28 16 of 28 16 of 28 16 of 28 17 of 28 17 of 28 17 of 28 17 of 28 18 of 28 19 of 28 19 of 28 20 of 28 21 of 28 Section A2 TABLE OF CONTENTS (continued) Q�P Section A August 17, 2000 Revision 1 Page 6 of 30 Pa e B6 Instrument/Equipment Testing, Inspection, and Maintenance Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 of 28 B.6.1 Field Instrumentation/Equipment . . . . . . . . . . . . . . . . . . . . . . . . . 22 of 28 B.6.2 Laboratory Instrumentation/Equipment . . . . . . . . . . . . . . . . . . . . 22 of 28 B7 Instrument Calibration and Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 of 28 B.7.1 Field Instrument/Equipment Calibration . . . . . . . . . . . . . . . . . . . 23 of 28 B.7.1.1 Air Sampling Pumps . . . . . . . . . . . . . . . . . . . . . . . . . 23 of 28 B.7.1.2 Airflow Calibration Procedure . . . . . . . . . . . . . . . . . . 23 of 28 B.7.2 Calibration of TEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 of 28 B8 Inspection/Acceptance Requirements for Supplies and Consumables .... 25 of 28 B.8.1 Air Sampling Filter Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 of 28 B9 Data Acquisition Requirements (Non-direct Measurements) . . . . . . . . . . 26 of 28 B.9.1 Precision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 of 28 B.9.2 Completeness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 of 28 B.9.3 Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 of 28 B 10 Data Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 of 28 B.10.1 Data Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 of 28 B.10.2 Data Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 of 28 B.10.3 Statistical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 of 28 C Assessment/Oversight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 of 4 C1 Assessments and Response Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 of 4 C.1.1 Performance and System Audits . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 of 4 C.1.1.1 Field Audits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 of 4 C.1.1.2 Laboratory Audit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 of 4 C.1.2 Corrective Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 of 4 C2 Reports to Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 of 4 D Data Validation and Usability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 of 3 Dl Data Review, Validation, and Verification Requirements . . . . . . . . . . . . . . 1 of 3 D2 Validation and Verification Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 of 3 D3 Reconciliation with Data Quality Objectives . . . . . . . . . . . . . . . . . . . . . . . . 3 of 3 E References ...................................................... lofl Annendices A : C � E QAPP Section A August 17, 2000 Revision 1 Page 7 of 30 A2 TABLE OF CONTENTS (continued) Comparison of the Asbestos NESHAP and the Fort Worth Method for the Demolition of Substandard Structures ISO Method 10312:1995. Ambient Air - Determination of Asbestos Fibres -- Direct-Transfer Transmission Electron Microscopy Method Standard Operating Procedure for the Screening Analysis of Soil and Sed'unent Samples for Asbestos Content ASTM Standard Test Method D 4959-00. Determination of Water (Moisture) Content of Soil by Direct Heating EPA Method 100.1 Analytical Method for the Determination of Asbestos Fibers in Water QAPP Section A August 17, 2000 Revision 1 Page 8 of 30 FIGURE5 Number Pa�e A-1 Project Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 of 30 A-2 Project Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 of 30 A-3 Sampling Data Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 of 30 A-4 Meteorological Measurement Log . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 of 30 B-1 Wind Rose for June, City of Forth Worth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 of 28 B-2 Analytical Request and Chain-of-Custody Form . . . . . . . . . . . . . . . . . . . . . . . . 13 of 28 G 1 Corrective Action Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 of 4 Q�P Section A August 17, 2000 Revision 1 Page 9 of 30 TABLES Number Pa�e A-1 Effect of Number of Days of Demolition or Land Filling and Between-Sample Variation on Statistical Power Calculations Assuming 5 Samples Upwind and 5 Samples Downwind from the Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 of 30 B-1 Air Sampling Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 of 28 B-2 Approximate Number of Grid Openings to Achieve Target Analytical Sensitivity Based on Air Volume of 3000 Liters (Direct-Transfer Preparation of TEM Specimens) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 of 28 QAPP Section A June 2, 2000 Revision 0 Page 10 of 30 A3 DISTRIBUTION LIST Brian Boerner, Environmental Management Division, City of Fort Worth, Texas Kathryn A. Hansen, Environmental Management Division, City of Fort Worth,, Texas Michael A. Gange, Environmental Management Division, City of Fort Worth, Texas Shirley Hoover, Environmental Management Division, City of Fort Worth, Texas Mike Camp, Code Compliance, Department of City Services, City of Fort Warth, Texas Brian Haggerty, City Manager's Office, City of Fort Worth, Texas Charles Boswell, Assistant City Manager, City of Fort Worth, Texas Adele Cardenas, U.S. EPA, Region VI Todd Wingler, Texas Department of Health Roger C. Wilmoth, National Risk Management Research Laboratory (NRMRL), U.S. EPA Tracy K. Bramlett, Industrial Hygiene & Safety Technology, Inc. Demolition Contractor (To Be Determined) Microscopy Laboratory (To Be Determined) QAPP Section A August 17, 2000 Revision 1 Page 11 of 30 A4 PROJECT ORGANIZATION Environmental Quality Management, Inc. (EQ) has been contracted by the City of Fort Worth, Texas, to design this Quality Assurance Project Plan (QAPP).1 The QAPP was prepared to meet U.S. Environmental Protection Agency (EPA) requirements applying to projects that include environmental measurements and is written in the format specified by EPA.�I� The project organization chart showing the relationships and the lines of communication among project participants is contained in Figure A-1. A.4.1 City of Fort Worth Brian Boerner, CHMM, will serve as the Program Manager. He has overall administrative and technical responsibility for the project. Kathryn A. Hansen, Esquire, will serve as the Project Manager. She has overall administrative responsibility for the project. As such, she will resolve any administrative problems that may occur and serve as the administrative contact with the Texas Department of Health, U.S. EPA, Industrial Hygiene & Safety Technology, Microscopy Laboratory (To Be Determined), Demolition Contractor (To Be Determined), and others. Michael A. Gange will serve as the Technical Project Officer. He has overall technical responsibility for the project. He will ensure that Industrial Hygiene & Safety Technology, Inc. Mr. Kominsky (Vice President/Director Industrial Hygiene & Safety, EQ) prepared this QAPP. He has more than 25 years of experience in the comprehensive practice of industrial hygiene and safety, of which 13 years were with the National Institute for Occupational Safety and Health (1VIOSI�. Since 1988, he has designed, irnplemented, and served as Project Manager/Principai Investigator on more than 15 asbestos research projects for EPA's National Risk Management Research Laboratory (NRMRL). Five studies (4 EPA and 1 commercial client) involved the determination of ambient concentrations of airborne asbestos in communities during demolition of buildings containing asbestos-containing materials (ACIvn and landfill of the resultant asbestos- cont�vning demolition debris, and other such projects involving fugidve emissions of asbesros. Mr. Kominsky has a Master's of Science Degree in Industrial Hygiene (University of Pittsburgh, 1973) and Bachelor of Science Degree in Chemical Engineering (University of Nebraska, 1971). He is a Certified Industrial Hygienist by the American Board of Industrial Hygiene, a Certified Safety Professional by the American Board of Certified Safety Professionals, and a Certified (Master Level) Hazardous Materials Manager by the Insdtute of Hazardous Materials Management. He is an Adjunct Associate Professor in the Department of Environmental Health at the University of Cincinnati. He has authored or co-authored more than 35 journal articies (11 ardcles related to asbestos) regarding occupational, environmental, and public health. s = s u o = c •� o � '-> � m "' = N � y- CO � � � O � c a'�o`� ��aU� N N � a � � C � Q. p � a � � ~ � � o0 � � � X N U U1 ,� X f" U F° c a� E a� � � a �� oo. 3 �� U � L � (p 'C N -� � r- o � � a� � LL � i � Om� OI� � � > Op U ° w V m` � 0 a a� � i � � i i -------�------ i i � °- O O " � CV 'p� °c ° i� ��a� Q ° o '° w �U� �i � X c� � Q u c 0 m a � S � � O O O � � L � Z � FN p� a U a � W � � � � � �� N S U c � C � O � '_^ a .N '� o -o 3O ��r C p �� O N a o ° _ �`- a�i n u-� > >- '" E op N N C � N C OQ� �� O� �;Z'N Q C Cp U O � W � o ..,_ C'J 'c � � . � Q a� 0 `c '� � O � O a' U a '_ � "' o o �o 3 m� �� 't a a c � � u.�.�= � �^ O,�a c C00 O C lL O I� ,._�`_ � �' � > o0 U s o w � ° o � � a�i � 0 � s �, � 3 •� _"� Q � � °7 � o � U � � � �- (_j m •° o ao o ��< °" �t� � O < U V�j � U U a� � U c N N L � U O � 00 �'Z aN � �, 3 ° `� � �' a�i�N °� O � Q a a�i f� � � � � � � O O L � O 1� +T- 'c � p �> cp � ;� c-- � � � w o Q N � QAPP Section A June 2, 2000 Revision 0 Page 12 of 30 � � otJ c U a� _ ° � c >. _ o � a� rnU .�v .To$ �^ 2 c °' � � .� E—� o � o— .� F� m a N 7 � ][ C O� c a a � N � �^ 0 a� c U � � c u� op = �, o °p � o N ~ 0 O .� � � �. � O .c � a � J N Q � O � N m � � <V �- < QAPP Section A August 17, 2000 Revision 1 Page 13 of 30 (IHST) implements the study in strict accordance with the QAPP and that the demolition contractor (To Be Determined) follows the Fort Worth Method. As such, he will resolve any technical difficulties with the contractors. He will also be responsible for resolving any technical difficulties associated with the laboratory analysis of the samples, data analysis and management, and preparation of the final project report. A.4.2 Industrial Hygiene & Safety Technology, Inc. Project Manager/Principal Investigator -- Mr. Tracy K. Bramlett (President, Industrial Hygiene & Safety Technology, Inc.) will serve as the Project Manager/Principal Investigator and maintain close communication with Ms. Hansen, Mr. Gange, and the demolition contractor. He will ensure that all IHST project personnel fully understand and strictly adhere to the QAPP. All IHST technical team members will be experienced professionals who possess the degree of specialization and technical competence required to effectively perform the required work. Mr. Bramlett will coordinate and supervise implementation of the QAPP including site preparation, sample collection and documentation, data management and analysis, and preparation of the project report. He will coordinate the sampling staff, ensure all equipment is calibrated properly and measurements are made in accordance with the QAPP, review and validate field data (e.g., sampling data logs), review sample custody and traceability records, and coordinate submission of the samples to the laboratory. He will ensure that any problems or potential deviations from the QAPP reported by any of the project staff are addressed immediately and receive corrective and documented action, as necessary. Prior to effecting any deviations from the approved QAPP, the potential deviation will be discussed with and approval to deviate will be obtained from Mr. Gange. A.4.3 Microscopy Laboratory (To Be Determined) Laboratory will perform the transmission electron microscopy (TEM) analysis of all samples collected. A.4.4 Demolition Contractor (To Be Determined) Demolition contractor will perform the demolition of the facilities. A5 PROBLEM DEFINITION/BACKGROUND A.5.1 Background Q�p Section A June 2, 2000 Revision 0 Page 14 of 30 In order to demolish substandard structures2 that are not in danger of imminent collapse, the City of Fort Worth currently follows the requirements established by Asbestos National Emissions Standards for Hazardous Air Pollutants (NESHAP), 40 CFR §61.145. The City of Fort Worth praposes an alternative method, hereinafter referred to as the Fort Worth Method, for the demolition of "facilities" in lieu of the current Asbestos NESHAP requirements for the demolition of substandard structures that are not in unminent danger of collapse. If left standing, the substandard facilities will within several years become structurally unsound. Instead of waiting for these buildings to reach such a structural state, the City proposes to be proactive and demonstrate that facilities with regulated asbestos-containir►g material (RACM) left in-place is at least as protective as demolition of buildings with the RACM removed.3 Based on the condition of the RACM, demolishing facilities while in substandard condition most likely is more protective of the public and environment than demolishing such facilities in danger of iinminent danger of collapse. Due to the requirements of Asbestos NESHAP, the City has only demolished facilities with RACM remauung in place when the facility is in uiuiunent danger of collapse. Preliminary data generated by both phase contrast microscopy (PCM) and transmission electron microscopy (TEM) indicates that demolition of such structures using the Fort Worth Method does not generate significant airborne fiber levels.�2� The Fort Worth Method's one primary difference from the existing Asbestos NESHAP is handling of the RACM. Prior to and during the course of demolition the facility is thoroughly wetted and demolition proceeds in a manner that allows the non-asbestos-containing building In the City of Fort Worth, a structure is considered substandard when it does not meet the standards or specifications established in the City's Minunum Building Standards Code, Ordinance No. 13743. The City of Fort Worth proposes to demolish a facility without removal of RACM with certain exceptions. Spray-applied fireproofing and large quantities of thermal system insulation (>260 linear feet) will be removed using full-containment abatement procedures. QAPP Section A June 2, 2000 Revision 0 Page 15 of 30 material (ACBM) to act as a barrier between the ACBM and the environment. By proper handling and wetting of the demolition debris, asbestos fiber release is controlled. A comparison of the Asbestos NESHAP and the Fort Worth Method is contained in Appendix A. The Fort Worth Method's primary goal is to protect the public and the environment from the release of asbestos during the demolition of buildings containing in-place RACM. It's secondary goal is to provide a alternative method for controlling asbestos that creates a cost savings for municipalities performing demolition of nuisance buildings. A.5.2 Objectives The primary objectives of this project are: 2. Determine whether the airborne concentrations of asbestos upwind (comparative environmental background�) during demolition of buildings containing in-place RACM are statistically significantly different than those concentrations downwind. 3. Determine whether the airborne concentrations of asbestos upwind (comparative environmental backgrotcnc� during land filling of building demolition containing RACM are statistically significantly different than those concentrations downwind. Environmental background is the airborne concentration of asbestos prevailing in this area that is upgradient (upwind) from the facilities being demolished or the demolition debris being landfilled. Q�p Section A August 17, 2000 Revision 1 Page 16 of 30 A6 PROJECT DESCRIPTION A.6.1 Technical Approach The project will be performed in two phases: Phase I and Phase II. Phase I--Phase I will gather data on the Fort Worth Method's ability to prevent or mulimize the release of asbestos fibers during demolition of buildings that are either exempt from the Asbestos NESHAP requirements (residential buildings that have four or fewer dwelling units) or not subject to asbestos NESHAP.S The data from Phase I will be analyzed to determine whether the Fort Worth Method is equivalent to the NESHAP method; i.e., the airborne concentrations of asbestos upwind (compar-ative environmental backgrounc� during demolition of the facilities and land filling of the resultant demolition debris are not statistically significantly different than the respective concentrations downwind. If the Phase I data supports a finding that the Fort Worth Method is equivalent to the Asbestos NESHAP method, the City will proceed with Phase II. If the methods are not found to be equivalent, the project may end with Phase I. Phase II--Phase II will gather data on various buildings subject to the Asbestos NESHAP. These buildings will include single or two-story commercial structures and/or multi-family residential structures. The ACBM that are likely to be present include, but not limited to resilient floor tile and mastic, Transite� panels, roofing materials, wallboard joint compound, wall and ceiling texture, thermal system insulation, and miscellaneous other materials. Prior to demolition of a structure, a thorough asbestos assessment of the facility will be conducted by a State of Texas Department of Health licensed Asbestos Inspector. The assessment will identify the type, quantity, location, and condition of ACBM. Prior to demolition of the facility, spray-applied fireproofing or large amounts of thermal system insulation will be removed by a State of Texas Department of Health licensed Asbestos Abatement Contractor. The type and quantity of ACBM remauung in the facility will be documented. Asbestos NESHAP regulations must be followed for demolitions of facilities with at least 260 linear feet of RACM on pipes, 160 square feet of RACM on other facility components, or at least 35 cubic feet off facility components where the amount of RACM previously removed from pipes and other facility components could not be measured before stripping. QAPP Section A August 17, 2000 Revision 1 Page 17 of 30 A.6.1.1 Demolition of Buildings The buildings will be demolished using heavy equipment only. A typical building demolition will include the following: One or more bulldozers for single-story buildings, and a combination of bulldozers, front-end loader, and track-hoes for multi-story buildings. Thoroughly and adequately wetting the structure using fire hydrant water applied with a variable rate 11-G (11 gpm) or 30-G (30 gpm) nozzle prior to, during demolition, and during debris loading. A water meter (or equivalent device) will be installed at the water hydrant to measure the volume of water used during demolition of the structure. The water will be delivered as a mist or concentrated stream, Direct high-pressure water impact of ACM will be prohibited.6 The demolition debris will be adequately wet at all times and kept wet during handling loading into containers for transport to a licensed disposal site. Collapsing structures inward (majority of the walls and interior components will be leveled on top of the building foundation) and loading debris prior to removal of the concrete slab, if present. Segregation of demolition debris to the extent feasible to reduce the amount of contaminated debris that will be treated as asbestos-contaminated waste. Debris not contaminated by the ACBM will be treated as construction debris, while all other materials will be treated as asbestos-contaminated waste. The RACM debris will be transported to a licensed disposal site in lined and covered containers. Segregation of the waste will be the responsibility of an onsite Asbestos NESHAP trained individual. Grading of the site for future use. A.6.1.2 Air Sampling During Demolition of Buildings The first study objective (see Section A.5.2) will be addressed by collecting ambient air samples at ten locations for three to five consecutive days during demolition of a facility. Five sampling locations will be located both upwind (comparative environmental background level of asbestos) and downwind of the demolition area. Meteorological conditions (such as wind Although experience demonstrates that building demolition projects typically have a minimal to moderate amount of water runoff depending on the site locations and site conditions, the City will utilize Best Management Pr•actices to control water runoff and collect storm water on the project site. Storm drain inlet protection will be used in conjunction with on-site controls (such as natural and manmade drainage channels), as necessary. QAPP Section A August 17, 2000 Revision 1 Page 18 of 30 direction and wind speed) will be determined explicitly to establish and ensure true upwind and downwind conditions during sampling (see Section B.l.l). A.6.1.3 Air Sampling During Land Filling of Demolition Debris The second study objective (see Section A.5.2) will be addressed by collecting ambient air samples at ten locations for two to three consecutive days during land filling of the demolition debris from each facility. The projected monitoring of two to three consecutive days during landfilling of the demolition debris is based on the amount of time required to landfill the debris from a given site. Five sampling locations will be located both upwind (comparative environmental background level of asbestos) and downwind of the land fill. Meteorological conditions (such as wind direction and wind speed) will be determined explicitly to establish and ensure true upwind and downwind conditions during sampling (see Section B. L 1). A.6.2 Personnel The environmental measurements (field samples and meteorological measurements) will be made by and at the direction of an ABIH-Certified Industrial Hygienist, Tracy Bramlett. He has more than 23 years of experience in the comprehensive practice of industrial hygiene with approxirnately 21 years of asbestos monitoring experience including ambient air monitoring for asbestos and licensed appropriately by the State of Texas as an air monitoring technician. Other field personnel will include industrial hygienists experienced in asbestos ambient air monitoring and related measurements. A.6.3 Project Schedule The tentative project schedule is presented in Figure A-2. The majar activities are listed sequentially, and the expected duration of each activity is presented. QAPP Section A Auwst 17, 2000 Revision 1 Page 19 of 30 Q�p Section A June 2, 2000 Revision 0 Page 20 of 30 A7 QUALITY OBJECTIVES AND CRITERIA FOR MEASUREMENT DATA The overall quality assurance objective of this project is to implement procedures for field sampling, chain-of-custody, laboratory analysis, and reporting that will provide data to satisfy the Primary Project Objective (i.e., the objective that will lead to the development of scientifically valid conclusions in the final report). Specific procedures for sampling, chain-of-custody, laboratory analysis, field and laboratory audits, preventive maintenance of field equipment, and corrective actions are described in other sections of this QAPP. A.7.1 Primary Project Objective The primary project objective is to determine the concentrations of airborne asbestos upwind (comparative environmental backgrounc� and those concentrations that are present downwind of the demolition site and the demolition debris land fill. A.7.3 Criteria for Acceptance The criteria for acceptance of equivalency of the Fort Worth Method to the e�sting Asbestos NESHAPS Method will be achieved by meeting either of the following: 1. There is not a statistically significant difference in the airborne concentrations of --- � asbestos upwind during demolition of buildings containing in-place RACM or upwind during land filling of this building demolition debris compared to those concentrations measured downwind of the respective sites. The statistical comparisons will be made at the O.OS level of significance. 2. The downwind sample average is less than 70 asbestos structures per square milliineter; the AHERA (40 CFR 763) clearance criterion. A.7.3 Statistical Power The study is designed to detect a 5-fold difference in average concentration (e.g., a 5-fold difference between the airborne asbestos concentrations upwind and downwind from the demolition or land filling sites) with high probability if such a difference actually e�sts. A false- Q�P Section A August 17, 2000 Revision 1 Page 21 of 30 positive error rate' of 5 percent will be achieved by employing a statistical significance level of 0.05 (i.e., a confidence level of 95%). A false-positive error occurs if we determine that a significant difference between the upwind and downwind airborne asbestos concentrations e�sts when, in fact, it did not. The statistical power of the upwind-downwind comparison will depend on the number of days required to demolish the buiiding or landfill the demolition debris (Table A-1). Assuming three days of demolition per facility and ten air samples collected each day (five locations upwind, five location downwind), a simple parametric comparison (e.g., t-test) between the upwind and downwind locations will have a false-negative error rate8 of appro�mately 10% for a 5-fold difference between mean concentrations and approximately 1% for a 10-fold Table A-1. Effect of Number of Days of Demolition or Land Filling and Between-Sample Variation on Statistical Power Calculations Assuming 5 Samples Upwind and 5 Samples Downwind from the Site Number of . Coefficient of Variation, % Days Difference 100 150 200 250 2-fold 0.42 0.28 0.16 0.17 2 days 5-fold 0.98 0.85 0.78 0.68 10-fold >0.99 0.99 0.95 0.93 2-fold 0.62 0.41 0.26 0.26 3 days 5-fold 0.99 0.98 0.95 0.89 10-fold >0.99 >0.99 0.99 0.99 2-fold 0.74 0.52 0.41 0.33 4 days 5-fold >0.99 0.99 0.97 0.95 10-fold >0.99 >0.99 >0.99 >0.99 ' A false-positive error rate is the probability of rejecting the null hypothesis when the null hypothesis is actually true. g A false-negative error rate is the probability of accepting the null hypothesis when the null hypothesis is actually false. Q�p Section A June 2, 2000 Revision 0 Page 22 of 30 difference (or equivalently, there is a probability of 0.90 of detecting a 5-fold difference and a probability of appro�mately 0.99 of detecting a 10-fold difference between mean concentrations at the upwind and downwind locations). A false-negative error occurs if we determine that no difference in airborne asbestos concentration e�sts, when in fact the concentrations did differ. The probability estimates assume a between-sample coefficient of variation (CV) of 250 percent, which �s environmentally conservative, and were estimated by Monte Carlo simulation using log-normal random variables. Under more optimistic assumptions regarding the variability of ineasurements (e.g., a between-sample CV of 150 percent), the probability of detecting a 5-fold difference between the mean airborne asbestos concentrations upwind and downwind using a simple comparison (e.g., t-test) would be increased to approximately 98 percent. A.7.4 Precision and Accuracy The estimated false-negative error rates assume a between-sample coefficient of variation of 250 percent. This assumption is reasonable and environmentally conservative, and is based on measured concentrations of asbestos in ambient air at other locations.�3�5� The between-sample coefficient of variation is influenced by heterogeneity in the air being measured as well as sampling and laboratory performance used to collect and analyze the samples, respectively. If the overall target precision is not achieved, the false-negative error rate will increase. The data will ultimately be analyzed using analysis of variance (ANOVA) methods, rather than simple t-tests. The ANOVA methods will provide additional statistical power in detecting differences between sampling locations and, consequently, the false-negative error rates are expected to be lower than those stated in Section A.7.2. In addition, duplicate field samples will be collected during each day of sampling and compared to results from co-located samples to evaluate the precision as well as serve as a combined check on the sample collection and analysis procedures. A.7.5 Completeness An overall measure of completeness will be given by the percentage of samples specified in the sampling design that yield usable "valid" data. Although every effort will be made to collect and analyze all of the samples specified in the sample design, the sample design is robust to Q�p Section A June 2, 2000 Revision 0 Page 23 of 30 sample loss. The loss of a small number of samples, provided that they are not concentrated at a single sampling site, will likely have little effect on the false-negative error rate. The project goal is to collect at least 95 percent of the samples specified in the sample design. A.7.6 Representativeness The sampling locations, sampling periods, and sample durations have been selected to assure reasonable representativeness. The five upwind sampling locations will be selected as locations that can be reasonably viewed as representing comparative environmental background conditions that are unaffected or influenced by activities of the demolition and land filling operations. The five downwind sampling locations will be selected as representative sites that would be influenced by potentially significant releases of asbestos from the demolition or demolition debris land filling operations. Samples will be collected each day (meteorological conditions permitting) of the demolition or land filling of demolition debris to provide estimates that are representative of the normal meteorological conditions in the area and to account for the potential day-to-day variability associated with ambient levels of airborne asbestos. A.7.7 Comparability Data collection using standard sampling and analytical methods (e.g., ISO Method 10312:1995, counting structures longer than and shorter than 5,um in length, and PCM equivalent fibers9) maximizes the comparability of the results with both past sampling results (if such exist) and future sampling results. A.7.8 Analytical Sensitivity We have selected an analytical sensitivity (i.e., the concentration corresponding to the finding of one asbestos structure during a sample analysis) that corresponds to asbestos concentrations representative of general background conceritrations in many areas of the United States. Data presented in Berman and Chatfield�6� indicate that an analytical sensitivity of 0.0005 A PCM (phase contrast microscopy) equivalent fiber is a fiber with an aspect ratio greater than or equal to 3:1, longer than 5�m, and which has a diameter between 0.2 �cm and 3.0 �cm. QAPP Section A June 2, 2000 Revision 0 Page 24 of 30 structure/cubic centimeters of air (s/cm3) is likely sufficient to detect environmental background concentrations in many areas of the United States. The analytical sensitivity selected for the ambient air monitoring is 0.0005 s/cm3 for all asbestos structures (iiiinunum length of 0.5 µm) and 0.0001 s/cm3 for asbestos structures longer than 5,um (all widths). See Section B.4 "Analytical Method Requirements." Thus, the number of ineasurements (upwind and downwind) in which asbestos structures is not observed or detected among the data collected should be minimal. QAPP Section A June 2, 2000 Revision 0 Page 25 of 30 A8 SPECIAL TRAINING REQUIREMENTS/CERTIFICATION A.S.1 Field Personnel The sampling (including equipment calibration, sample collection, and documentation) will be performed by and/or at the direction of an ABIH-Certified Industrial Hygienist (T. Bramlett). He has more than 23 years of experience in the comprehensive practice of industrial hygiene with approximately 21 years of asbestos monitoring experience including ambient air monitoring for asbestos. Other field personnel will include industrial hygienists experienced in asbestos ambient air monitoring and related measurements. A.8.2 Laboratory Personnel (To Be betermined) Laboratory is accredited by the National Institute of Standards and Technology (IVIST) National Voluntary Laboratory Accreditation Program (NVLAP) to perform Airborne Asbestos Fiber Analysis. (To Be Determined ) laboratory's NVLAP Laboratory Code No. is X��X (effective through X��X, 2000). QAPP Section A June 2, 2000 Revision 0 Page 26 of 30 A9 DOCUMENTATION AND RECORDS A.9.1 Field Operations Records A.9.1.1 Air Sample Documentation The following information will be recorded on a Sampling Data Form (Figure A-3): • Names of persons collecting the sample • Date of record • S ampling site • Location of sample • Type of sample (e.g., high volume, duplicate, field blank) • Unique sample number (identifies site, sample type, date, and sequence number) • Rotameter number and air flow reading (start/stop) • Linear regression equation and correlation coefficient for the calibrated rotameter • Sample time (start/stop) • Relevant notes describing site observations such as, but not limited to site conditions, demolition method/techniques and equipment, water application nozzle and hose diameter, water application technique (spray or concentrated stream) and approximate amount of time (e.g., hours) for each will be recorded on the reverse side of the Sampling Data Form (and on supplemental pages, as necessary). The upwind and downwind air sampling locations will be identified on a drawing of the demolition site or land fill area. At the end of each day, the samples which were collected and the corresponding data forms/drawings will be submitted to Tracy Bramlett. He will verify the data/information for completeness; any corrections will be noted and initialed on the form. A.9.1.2 Meteorological Measurements Meteorological measurements (wind direction, wind speed, relative humidity, and temperature) will be recorded on a Meteorologic Data Measurement Log (Figure A-4). QAPP Section A June 2, 2000 Revision 0 Page 27 of 30 a" � � O u i - I.n ^~ � � Q 'a� C �' ' G � � o .. .. .. .. .. .. .. .. .. .. .. � � �' S. � E--� -v � .. .. .. .. .. .. .. .. .. .. .. v � -c v � > ¢ E � u a. u p t�-. 3 v' 0 w L Ctl v tn �O c ��., Q � N � 0 �� •� � � U r' [ U a 0 .�, 0 � c 0 :�. v -� � a � � � � � � �/+ O 4+ W ' � �'vRT�1oRTx .�--�-- =�-= Weather Station Measurement Log 3ia: Datc _� / Pagc: of Invatigawr. Q�p Section A June 2, 2000 Revision 0 Page 28 of 30 W[ND WIND BAROMETRIC TEMPERATURE. RE[ATIVE TIME SPEED, MPH D[REGTION PRESSURE, la. Hg °F HUMIDITY, % Figure A-4. Meteorological Measurement Log. QAPP Section A August 17, 2000 Revision 1 Page 29 of 30 A.9.1.3 Photo Documentation A 35-mm photograph or digitized image will be taken of every sampling site. This will include the sampling station, visual debris on or in the soii. A 5-in. by 7-in. index card listing the sample number will be photographed to identify the sample and location. Other photographs or digitized images will be taken as necessary to thoroughly document the site conditions (such as "visible emissions," if such occurs) and activities. In addition, a camcorder will be used to videotape the demolition and demolition debris land filling operations. A.9.2 Chain-of-Custody Records Standard EQ sample traceability procedures described in Section B3 "Sample Custody Requirements" will be used to ensure sample traceability. A.9.3 Laboratory Records A.9.3.1 TEM Specimen Examination and Data Recording Structure counting data shall be recorded on forms equivalent to the example shown in ISO 10312:1995 contained in Append� B. A.9.3.2 Test Report The test report shall contain items (a) to (p) as specified in Section 11 "Test Report" of ISO 10312:1995. In addition, the files contauung the raw data, in Microsoft Excel format, shall be submitted. The format of these files shall be as directed by the project manager, but shall contain the following items: l. Laboratory Sample Number 2. Project Sample Number (4 Blank Lines For Project Manager to Insert Locations and Sampling Details) 3. Date of Analysis 4. Air Volume 5. Active Area of Sample Filter 6. Counting Magnification 7. Mean Grid Opening Dimension in mm 8. Number of Grid Openings Examined 9. Number of Primary Structures Detected 10. Q�P Section A August 17, 2000 Revision 1 Page 30 of 30 One line of data for each structure, containing the following information as indicated in Figure 7"Example of Format for Reporting Structure Counting Data" of ISO 10312:1995, with the exception that the lengths and widths are to be reported in millimeters as observed on the screen at the counting magnification: Grid Opening Number Grid Identification Grid Opening Identification/Address Structure or Sub-structure Number Asbestos Type (Chrysotile or Amphibole) Morphological Type of Structure Length of Structure in millimeters in 1 mm increments (e.g., 32) Width of Structure in millimeters in 0.2 mm increments (e.g., 3.2) Any Other Comments Concerning Structure (e.g., partly obscured by grid bar) � B MEASUREMENT/DATA ACQUISITION Bl SAMPLING DESIGN B.1.1 Sampling Locations Q�P ' Section B August 17, 2000 Revision 1 Page 1 of 28 Ambient air samples will be collected at 10 sampling locations each day of the demolition and each day that the respective demolition debris is placed in the landfill. Five sampling locations will be located both upwind (comparative environmental background level of asbestos) and downwind of the land fill. A wind rose for five years (1988 through 1992) of June winds (0000 to 2300 hours) is presented in Figure B-1. When considering a wind fetch of 135 degrees (from the southeast) through 202.5 degrees (from the south southwest), about 54 percent of the total hours of wind data over five months of June winds were from this 67.5 degree sector. Should the wind direction change outside of this 67 degree sector for more than a 60-minute period, the sampling will be terminated. B.1.2 Air Sampling Strategy The air sampling strategy is summarized in Table B-1. Air sampling will be conducted for 3-5 and 2-3 consecutive days during demolition and land filling of the demolition debris, respectively, if acceptable meteorological conditions exist. Acceptable meteorologic conditions include consistent upwind conditions for the environmental comparative background samples and no rain. Two sets of air samples will be collected at each sampling locations: a high-volume and a low-volume sample. A high-volume sample will be collected to achieve the target air volume (such as 3,000 liters) over the period of the demolition or land filling activities, and a second sample (low-volume) will be collected to achieve one-half of the target air volume over the same period. The second sample will be collected in the event that the first sample with the higher air volume is overloaded with particulate,10 which would preclude transmission electron microscopy lo The direct transfer analytical method (ISO 10312:1995) should not be used if the general particulate loading of the sample collection filter exceeds approximately 10 �cg/cm2 of filter surface, which corresponds to approximately 10 percent coverage of the collection filter by particulate. QAPP Section B August 17, 2000 Revision 1 Page 2 of 28 WIND ROSE PLOT STATION #03927 - DALLAS/FORT WORTH/REGIONAL AR, TX 'NORTH ' . • • - ` - . . . �. 25% ,', • � �,20% • ,� ," ' - � • i5�b �, �, . , , .� � 10% • , , � , , � � 5/�, �, � , �, ; WEST ; . ... . . . . ' . . . - , - • - , - - - - ; • - - -;- - - - - ; - - EAST, 'SOUTH . Wind Speed (K�ots) MODELER G.Schewe >21 DISPLAY 17 - 21 Wind Speed ��-is �. 70 AVG. WIND SPEED a - s $�63 Knots t- 3 ORIENTATION Direction (biowing from) DATE 4/8/00 UNIT Knots CALM WINDS 6.17% PLOT YEAR-DATE-TIME 88 89 90 91 92 June 1 - June 30 Midnight -11 PM COMPANYNAME EQ COMMENTS PROJECT/PLOT NO. Figure B-1. Wind Rose for June, City of Fort Worth. Q�P Section B August 17, 2000 Revision 1 Page 3 of 28 TABLE B-l. AIR SAMPLING STRATEGY Q�P Section B August 17, 2000 Revision 1 Page 4 of 28 TABLE B-1(continued) Q�P Section B August 17, 2000 Revision 1 Page 5 of 28 TABLE B-1 (continued) (TEM) analysis using the direct transfer method of sample preparation; otherwise, the second sample will be archived. Duplicate "Co-Located" Samples -- Two high-volume duplicate air samples (one upwind, one downwind) will be collected during each day of sampling. Each high-volume duplicate sample will be analyzed. Field B.lanks -- Two open and two closed field blanks will be collected each day of sampling. Open field blanks are filter cassettes, that have been transported to the sampling site, opened for a short-time (< 30 seconds) without any air having passed through the filter, and then sent to the laboratory. Closed field blanks are filter cassettes that have been transported to the sampling site and then sent to the laboratory without being opened. The two open field blanks will be analyzed and the closed field blanks will be archived. The closed field blanks will only be analyzed if the open field blanks show asbestos contamination. QAPP Section B August 17, 2000 Revision 1 Page 6 of 28 B.1.3 Particulate Loading Pilot Test One of the biggest difficulties associated with ambient a.ir monitoring is the paramount requirement to achieve adequate analytical sensitivity with constraints placed on sampling and analysis. Briefly, analytical sensitivity is a function of the volume of air passed through a filter, the concentration or dilution attendant to filter preparation, and the area of the filter scanned for analysis. The volume of air that can be collected is prunarily limited by the tolerable loading of total particulate collected per unit area of filter before the sample is unsuitable for analysis by the direct-transfer TEM method because of overloading. A pilot sampling test will be performed to estimate the maximum sample air volume (with an acceptable particulate loading) to achieve the specified analytical sensitivity by counting the fewest grid openings. The pilot test will consist of collecting five sets of ambient air samples over an approximately 6-hour period to achieve the following total air volumes: 3500, 3000, 2500, 2000, and 1500 liters. The sample will be examined by an experienced electron microscopist to determine if the particulate loading of the sample exceeds the criterion specified in ISO Method 10312:1995. B.1.4 Soil Sampling The Asbestos content of the soil will be determined both before and after demolition of the structure. Two samples will be collected from each side of the structure yielding a total of eight samples before and eight samples after demolition of each buiiding. The sampling locations will be randomly selected from a 5-ft by 5-ft grid system created around the building. The samples will be collected from the center of each selected grid square. The grid system will be prepared on a plot plan or sirrvlar drawing of the property. If two samples cannot be obtained from each side of the structure, samples will be obtained from the remaining sides to yield a total of eight samples. Before and after samples will be collected from the same approximate locations. The sampling locations will be marked on the site drawing. In addition, a marker (e.g., fluorescent orange painted meter pipe) will driven into the ground to mark the location of the soil sample collected before demolition of the structure. QAPP Section B August 17, 2000 Revision 1 Page 7 of 28 B.1.5 Moisture Content of ACM To determine the effectiveness of the water application process on wetting the asbestos- containing materials in the structure, 10 representative bulk samples of the asbestos-containing materials will be collected. The ten bulk samples will be collected of the asbestos-containing materials after the structure has been demolished. The water (moisture) content of each sample will be determined. A water meter (or an equivalent device) will be installed at the hydrant to measure the volume of water used during demolition of the structure. B.1.6 Water Used for Wetting Structure/Debris The water used to wet the structure and resultant demolition debris will be sampled to determine the asbestos content. That is, each sample will be analyzed for asbestos fibers greater than 0.5 ,um in length; the AHERA (40 CFR 763) definition. B2 SAMPLING METHODS REQUIREMENTS B.2.1 Air Sampling Q�p Section B August 17, 2000 Revision 1 Page 8 of 28 The samples will be collected on open-face, 25-mm-diameter 0.45-,um pore size miYed cellulose ester (MCE) filters with a 5-,um pore size MCE diffusing filter and cellulose support pad contained in a three-piece cassette with a 50-mm non-conductive cowl. This design of cassette has a longer cowl than the design specified in ISO 10312:1995, but it has been in general use for some years for ambient and indoor air sampling. Disposable filter cassettes with shorter conductive cowls, loaded with the appropriate combination of filter media of known and consistent origin, do not appear to be generally available. The filter cassettes will be positioned on tripods approximately 5 feet above the ground or at the elevation best suited to achieve an unobstructed representative air sample. The filter assembly will be attached with fle�ble Tygon� tubing (or an equivalent material) to an electric-powered (110 VAC) 1/6-horsepower vacuum pump operating at an airflow rate of approxima.tely 6.3 liters per minute to achieve the target air volume of 3,000 liters over an eight hours. Each pump will be equipped with a flow control regulator to maintain the initial flow rate of within +/- 10% throughout the sampling period. Although the pilot test described in Section B.1.3 will determine the optimal target air volume, it is anticipated that the air volume for each sample will be approximately 3,000 liters. If 110 VAC line power is not available, portable gasoline-powered generators will be used to p.ower the sampling pumps. B.2.2 Meteorological Monitoring Meteorological data will be collected to support the determination of valid sampling periods. Data validity will be based primarily on wind directions which lend themselves to maintauung the monitors in an upwind and downwind orientation from the demolition site or landfill. A combination of nearby National Weather Service (NWS) 12-hour wind direction forecasts, on-site meteorological data collection using continuous monitoring, and individual monitoring site qualitative collection data will be used to determine several monitoring objectives. Q�P Section B August 1'7, 2000 Revision 1 Page 9 of 28 These objectives include the following: 1. Forecasting a valid upcoming sampling day. 2. Providing ongoing data collection to support final data validation. 3. Checking ongoing data throughout a sampling period overall and at specific sampling locations. The primary specification for objectively determining monitored data validity and the determination of go/no-go on each sample day will be based on two meteorological conditions. The first is rainfall. If the NWS forecasts greater than a trace (0.1 inch) amount of rainfall for the upcoming 24-hour sampling period, the sampling day will be canceled. The second criterion for evaluating data validity and go/no-go situations will be wind direction. For June all winds coming from the 135 ° through 202.5 ° wind sectors (southeast through south through southwest) will be selected as valid wind flows. Wind directions prior to each sampling day will be estimated through NWS forecasts. Sampling will proceed whenever winds are generally forecast for the southwest through northeast (clockwise through northwest). Prior to the start of each day's field activities, the NWS at the Dallas-Fort Worth International Airport will be contacted to obtain 12-24 hour wind direction and precipitation forecasts. If the conditions are acceptable, the sampling day will proceed. For ongoing field activities after the decision has been made to proceed, both continuously recorded wind speed and direction will be collected as well as predetermined time spot checks at each monitoring site. A meteorological station will be installed at both the demolition site and landfill. Each station will consist of a Met One Instruments, Inc. Automet meteorological data system. It will include continuous wind speed and direction sampling and data logging over the duration of the sampling period. All data will be collected and archived in the data logger which can be checked on a routine basis with direct readout as well as downloading to a personal computer whereby related software will be used to assess hourly, daily, and period archived data. The wind station will be tripod mounted in an appropriate location away from all obstructions. The Automet sensors and their associated sensitivity are approved for use in Prevention of Significant Deterioration Monitoring Projects under U.S. EPA and will provide significant detail to the wind direction tabulations. QAPP Section B August 17, 2000 Revision 1 Page 10 of 28 The meteorological station operator will also collect wind speed data using a hand-held Dwyer wind speed meter and wind direction data using a dark-colored ribbon attached to the top of a 6-foot wooden stick. Readings will be taken once every 30 minutes over a duration of about 1 minute using the 1-minute averages to represent the period. A logbook will be kept which notes these wind directions and wind speeds as well as those noted on the Automet system data logger for the same time period. The ma.nual measurement (wind speed and wind direction) ma.de at meteorological station will also be ma.de at one other upwind and two downwind sampling locations. They will also be at the same specified 1 minute of each 30-minute time period during sample collection. The wind speed and wind direction will be noted in a logbook at each sampling location. Consistency between observations will be promoted through the use of identical hand-held instruments as well as previous training on proper observation techniques by each site attendant. If during sampling the wind directions fall outside of the acceptable range at the main meteorological station for 30 or more concurrent minutes, sampling activities will cease for the day. The samples will be archived or voided. B.2.3 Soil Sampling The soil samples will be collected using a clean metal scooping tool (e.g., a garden trowel). The samples will be collected from the center of the grid square (see Section B.1.4). To the extent feasible, each sample will represent the top 1 to 1'fi inches of soil from a 4 inch by 4 inch area. The area will be delineated using a metal template with a 1'f inch vertical flange of sufficient strength to allow the flange to be pushed into the soil. The template will be constructed of galvanized sheet metal. B.2.4 Water Sampling One water sample will be collected from each water source (i.e., water hydrant) that will be used to wet the structure and resultant demolition debris. The sample container will be an unused, precleaned, screw-capped bottle of glass or low density (conventional) polyethylene and capable of holding at least 1 liter. (Ideally, water samples are best collected in glass bottles.) Prior to collecting the sample, allow the water from the water source to run to waste for a QAPP Section B August 17, 2000 Revision 1 Page 11 of 28 sufficiently long period to ensure that the sample collected is representative of fresh water. As an additional precaution against contamination, each bottle should be rinsed several times in the source water being sampled. Two separate samples of approximately 800 milliliters each will be collected. An air space will be left in the bottle to allow efficient redispersal of settled material before analysis. The second bottle will be stored for analysis if confirmation of the results obtained from the analysis of the first bottle is required. The samples will be transported to the analytical laboratory and filtered by the laboratory within 48 hours of each sample collection. No preservatives or acids vvill be added. At all times after collection, the samples will be stored in the dark and refrigerated at about 5° C(41° F) in order to minimi�e bacterial and algal growth. The samples will not be allowed to freeze, since the effects on asbestos fiber dispersions are not known. QAPP Section B August 17, 2000 Revision 1 Page 12 of 28 B3 SAMPLE CUSTODY REQUIItEMENTS Chain-of-custody procedures emphasize careful documentation of constant secure custody of sam�les during field, transport, and analytical stages of environmental measurement projects. The sample custodian responsible for the proper chain-of-custody during this project is: Tracy Bramlett, CIH, CSP Industrial Hygiene & Safety Technology, Inc. 2235 Keller Way Carrollton, Texas 75006 Phone: 972.478.7415; fax: 972.478.7615 B.3.1 Field Chain-of -Custody Each sample will have a unique project identification number. This identification number will be recorded on a Sampling Data Form (Figure A-3) along with the other information specified on the form. After the labeled sample cassettes are recovered from the sampling trains, the sample custodian will complete an Analytical Request and Chain-of-Custody Form (Figure B- 2). This form will accompany the samples, and each person having custody of the samples will note receipt of the same and complete an appropriate section of the form. Samples will be sent to Laboratory (To Be Determined) via Federal Express Standard Overnight Service. B3.2 Microscopy Laboratory The laboratory's sample clerk will examine the shipping container and each filter cassette to verify sample numbers and check for any evidence of damage or tampering, note any changes or indication of tampering on the accompanying chain-of-custody form, and then forward the form to Tracy Bramlett. The sample clerk will log in all samples and assign a unique sample identification number to each sample and sample set. ,--i 0 s--I �t+ Q O Z C N E � I U O � C � � r � � � (� � d 0 ❑� zo QU W N � W� � � N = U �LL O az Qa z V O O 'C = O m a a� � � o Z Z m .c .. — c � u� � . .o o n o � .a o °�,' a J V t � ro � � � a U .n m J � d N d Z � ro � � �E J z m � � E '0 � V f0 o. 'o o t°' °- a °' E m � W z J � W a � w Z Fa- z O U W z 0 Q�P Section B August 17, 2000 Revision 1 Page 13 of 28 C ,p O J 0 « �i 'a a E C U U� Ua�. � � � � � '�, a � 0 4- o I- C � Q Q � �a , �L ❑ � � J T Q N � O ? � C1 +� � a`� d � o•• ❑ >. >, � � � � � 0 0 n.E � U d � � � o a tA> °' � o `a o a E � � � � � � � � � r � N � � � ai c- � a �� U ¢ 0 a� Ea �i � d � L ov O c��cE �E o != o F= ❑ C T N � d �� a T c � ._ no � � d ¢ c�a a. � � N U C � O � � � � Q U � . E �❑ N d fC � � � � _ .� .� � �s E � v v � N❑ �- � y o N o 2 aa = � ❑ � � d c � � •g •-"- •5 •"- in EE °' � ` �a �`a_ m �s .o � m E V) Z avi ui c � E � c � c a o o � o � � o cn a Z I- Z .= � .v N U B4 ANALYTICAL METHOD REQUIIZEMENTS B.4.1 Air Samples QAPP Section B August 17, 2000 Revision 1 Page 14 of 28 The 0.45-µm pore size mixed-cellulose ester (MCE) filters will be prepared and analyzed using International Organization of Standardization (ISO) Method 10312:1995 (15` Ed.), "Ambient Air - Determination of Asbestos Fibres - Direct-Transfer Transmission Electron Microscopy Method." A copy of the method is contained in Append� B. The principal objective of these analyses is to provide air concentration data of sufficient quality to support the development of conclusions regarding the effective of the Fort Worth Method to control the release of asbestos fibers during the demolition of substandard structures containing RACM B.4.1.1 TEM Specimens Preparation TEM specimens shall be prepared from the ambient air filters using the dimethylformamide (DMF) collapsing procedure of ISO 10312:1995, as specified for cellulose ester filters. DMF shall be used as the solvent for dissolution of the filter in the Jaffe washer. For each filter, a ininimum of four TEM specimen grids shall be prepared from a one quarter sector of the filter, using 200 mesh indexed copper grids. The remairung part of the filter shall be archived until further notice, in the original cassette in clean and legally secure storage, to be possibly selected for quality assurance analyses. B.4.1.2 Measurement Strategy 1. The minimum aspect ratio for the analyses shall be 3:1, as permitted by ISO 10312:1995. 2. The analyses shall be performed by a two-stage examination of the TEM specimens, as indicated in ISO 10312:1995, in order to provide data of sufficient precision for each of the structure size ranges of interest. The size ranges of structures that shall be evaluated, and target analytical sensitivities, for the anticipated collected air volume of 3000 liters, will be as shown in Table B-2. QAPP Section B August 17, 2000 Revision 1 Page 15 of 28 Table B-2. Appro�mate Number of Grid Openings to Achieve Target Analytical Sensitivity Based on Air Volume of 3000 Liters (Direct-Transfer Preparation of TEM Specimens) Approximate Number of Target Approlamate Approlumate 0.01 mm2 Analytical Magnification Area Grid Sensitivity for Examined Openings Size Range s/cc Examination mm2 Required All Structures (Minimum Length of 0.0005 20000 0.256 26 0.5 pm) Longer than 5 pm, 0.0001 10000 0.256 26 (All Widths) 3. The stopping point in the analysis for each of the examinations defined in 2 shall be 100 primary asbestos structures, or after the completion of the examination of the grid opening during which the target analytical sensitivity is achieved: A mlliunum of 4 grid openings shall be examined, in accordance with the specifications of ISO 10312:1995. 4. The structure counting data shall be distributed approximately equally among a muiullum of 3 specimen grids, prepared from different parts of the filter sector. 5. The TEM specimen examinations at approximately 20,000 and 10,000 shall be performed as independent measurements. 6. Measurement of fiber dimensions is extremely important for accurate determination of aspect ratios. Lengths and widths of fibers shall be recorded in millimeters as measured on the fluorescent screen. Where the observed width of a fiber is lower than approximately 5 mm on the screen at either of the two magni.fications used for the TEM specimen examinations, the measurement enors may seriously compromise the accuracy of the calculated aspect ratio. Accordingly, in this situation, the ma.gnification shall be increased by a factor of approxima.tely five times to obtain an accurate measurement of the width. For example, the width of a 1 mm wide fiber cannot be accurately estimated on the screen. A five-fold increase in the magnification increases the dimension to 5 mm, which can be estimated with sufficient accuracy for the purpose of this project. QAPP Section B August 17, 2000 Revision 1 Page 16 of 28 B.4.2 Soil Samples The asbestos content of the soil will be determined using a method developed by U.S. EPA, Region I: Standard Operating Procedure for the Screening Analysis of Soil and Sediment Samples for Asbestos Content (SOP:EIA-INGABED2-SOP, January 11, 1999). To more accurately quantify the amount of asbestos in the soil samples, the samples will be prepared using the procedure specified in the addendum to the method. A copy of the method is contained in Append� C. B.4.3 Moisture Content of ACM The water (moisture) content of the bulk samples of asbestos-containing materials (ACM) will be determined in accordance with ASTM Standard Test Method D 4959-00 "Determination of Water (Moisture) Content of Soil by Direct Heating." The method will be modified (as necessary) to accommodate the ACM bulk samples. A copy of the method is contained in Append'vc D. B.4.4 Water Samples The asbestos content of the water used to wet the structure and resultant demolition debris will be determined using EPA Method 100.1 "Analytical Method Determination of Asbestos in Water." A copy of the method is contained in Appendix E. Measuxement of fiber dimensions is important. Dimension of fibers is recorded as length and width if greater than 0.5 ,um in length. Reference EPA Method 100.1 for counting and sizing rules for bundles, fiber aggregates, etc. The analysis may be stopped after 100 fibers have been counted or 20 grid openings have been examined. The sample grid will be examined in a transmissions electron microscope (TEM) at a magnification of about 20,000. a BS QUALITY CONTROL REQUIREMENTS QAPP Section B August 17, 2000 Revision 1 Page 17 of 28 The overall quality assurance objective is to provide defensible data of known quality meeting quality assurance objectives. To that end, procedures are developed and implemented for field sampling, chain-of-custody, laboratory analysis, reporting, and audits that will provide results which are scientifically valid and legally defensible in a court of law. B.5.1 Field Quality Control Checks Quality control checks for the field sampling aspects of this project will include, but not be limited to, the following: • Use of standardized forms (e.g., Figures A-3, A-4, B-2) to ensure completeness, traceability, and comparability of the data and samples collected. • Calibration of air sampling equipment including pre- and post-sample calibrations using a calibrated precision rotameter. • Proper handling of air sampling filters to prevent cross contamination. • Collection of field blanks; see Section B.5.2.2. • Field cross checking of data forms to ensure accuracy and completeness. B.5.2 Analytical Quality Control Checks B.5.2.1 Quality Control Check of Filter Media Before air samples are collected, a minimum of 2 percent of unused filters from each filter lot of l00 filters will be analyzed to determine the mean asbestos structure count. If the mean count for all types of asbestos structures is found to be more than 10 structures/mm2, or if the mean fiber count for asbestos fibers and bundles longer than or shorter than 5 µm is more than 0.1 fiber/mm2, the filter lot will be rejected. By comparison, this criterion is more restrictive than that specified by EPA in AHERA (i.e., 18 structures/mm2 of filter area). Q�P Section B August 17, 2000 Revision 1 Page 18 of 28 B.5.2.2 Blank Contamination To ensure that contamination by extraneous silicate mineral fibers during sample collection and specimen preparation is insignificant (see criteria in Section B.5.21) compared with the results reported on samples, a continuous program of blank measurements will be established. This will include filter lot blanks (see Section B.5.2.1), field blanks (open and closed), laboratory blanks, and laboratory clean area blanks. Field Blanks -- Closed field blanks are filter cassettes that have been transported to the sampling site and sent to the laboratory without being opened. O�en field blanks are filter cassettes, that have been transported to the sampling site, opened for a short time (< 30 seconds) without any air having passed through the filter, and then sent to the laboratory. The number of open and closed blanks that will be collected and analyzed is presented in Table B-l. Laboratory Blanks -- Laboratory blanks are unused filters (or other sample device or container) that are prepared and analyzed in the same manner as the field samples to verify that reagents, tools, and equipment are free of the subject analyte and that contamination has not occurred during the analysis process. The laboratory will analyze at least one blank for every ten samples or one blank per prep series. Blanks are prepared and analyzed along with the other samples. If the blank control criteria (Section B.5.2.1) are not met, the results for the samples prepared with the contaminated blank are suspect and should not be reported (or reported and flagged accordingly). The preparatian and analyses of samples should be stopped until the source of contamination is found and eliminated. Before sample analysis is resumed, contamination-free conditions shall be demonstrated by preparing and analyzing two blanks that meet the blank control criteria. Laboratory blank results shall be documented in the quality control log. Laboratory blank count sheets should be maintained in the project folder along with the sample results. Laboratory Clean Area Blanks -- Clean area blanks are prepared whenever contamination of a single laboratory prep blank exceeds the criteria specified in Section B.5.2.1 or whenever cleaning or servicing of equipment has occurred. To check the clean area, an used filter is left open on a bench top in the clean area for the duration of the sample prep process. The blank is then prepared and analyzed using ISO Method 10312:1995. If the blank control criteria (see Section B.5.2.1) are not met, the area is cleaned using a combination of HEPA-filter vacuuming QAPP Section B August 17, 2000 Revision 1 Page 19 of 28 and a thorough wet-wiping of all surfaces with amended water. In addition, air samples should be taken in the sample prep room to verify clean air conditions. At least 2,500 liters of air should be drawn through a 25-mm-diameter 0.45-,um pore size MCE filter using a calibrated air sampling pump. The samples should then be analyzed using ISO Method 10312:1995. If blank control criteria are not met, sample preparation shall stop until the source of contamination is found and eliminated. Clean area sample results shall be documented in the Clean Area Blank Log. B.5.3 Analytical Precision and Accuracy Quality control checks will be performed on a routine basis to verify that the analysis system is in control. Most laboratory Quality Control Programs include frequent quality tests for both accuracy and precision. Because of the difficulty of preparing quantitative asbestos standard ma.terials, neither spiked samples nor known quantitative samples can be used on a routine basis for asbestos analysis accuracy testing. Therefore, routine quality control testing for asbestos focuses on precision checks, which involve a second count or multiple counts of a sample or a portion of a sample. B.5.3.1 Replicate Analysis The precision of the analysis is determined by an evaluation of repeated analyses of randomly selected samples. A replicate analysis will be performed on 5% of the samples analyzed to assess the precision of the counting abilities of the individual analysts. A replicate analysis is a second analysis of the same preparation, but not necessarily the same grid openings, performed by the same microscopist as in the original analysis. The conformance expectation for the replicate analysis is that the count from the original analysis and the replicate analysis will fall within a 95% confidence interval for the average count, or as follows: LCL < A1 AZ < UCL where: A1 is the original count, AZ is the replicate count, LCL is the lower confidence limit, UCL is the upper confidence limit. Q�P Section B August 17, 2000 Revision 1 Page 20 0� 28 Should either the original or replicate count fall outside the acceptance range, the grid is re- examined to determine the cause of the count variation. The 95% confidence interval is based on the Poisson distribution. For average counts less than or equal to 20 structures, Table F-1 in IS O Method 10312:1995 (see Append� C) should be used to derive the upper and lower 95% confidence limits. For average counts greater than 470 structures, a normal approximation can be used. The upper and lower confidence limits based on the normal approximation are calculated as follows: where: LCL = ,u - 1.96 x ✓�.c UCL = �,c + 1.96 x ✓,u �c is the average count, ✓,u is the definition of the Poisson standard deviation. B.5.3.2 Duplicate Analysis A duplicate sample analysis is also performed on 5°Io of the samples analyzed to assess the reproducibility of the analysis and quantify the analytical variability due to the filter preparation procedure. A duplicate analysis is the analysis of a second TEM grid preparation prepared from a different area of the sample filter performed by the same microscopist as the original analysis. The conformance expectation is similar to that for replicate analyses with one exception. That is, the count from the original analysis and the duplicate analysis should fall within a 95% confidence interval for the average count. QAPP Section B August 17, 2000 Revision 1 Page 21 of 28 B.5.4 Verification Counting Due to the subjective component in the structure counting procedure, it is necessary that recounts of some specimens be made by a different microscopist (i.e., a microscopist different than the one that performed the original analysis) in order to minimize the subjective effects. Verification counting will involve re-examination of the same grid opening by a different microscopist. Such recounts provide a means of maintainuig comparability between counts made by different microscopists. These quality assurance measurements will constitute approximately 10 percent of the analyses. Repeat results should not differ at the 5°Io significance level. QAPP Section B August 17, 2000 Revision 1 Page 22 of 28 B6 INSTRUMENT/EQUIPMENT TE5TING, INSPECTION, AND MAINTENANCE REQUIREMENTS B.6.1 Field Instrumentation/Equipment Field equipment/instruments (e.g., sampling pumps, meteorological instrumentation) will be checked and calibrated before they are shipped or carried to the field. The equipment and instruments will be checked and calibrated at least daily in the field before and after use. Spare equipment, such as air sampling pumps, precision rotameters, and flow control valves will be kept on-site to nninimize sampling downtime. Backup instruments (e.g., meteorological instrumentation) will be available within one day of shipment from a supplier. B.6.2 Laboratory Equipment/Instrumentation As part of the (To Be Determined) Laboratory's QA/QC Program, a routine preventive maintenance program is performed to minimi�.e the occurrence of instrument failure and other system malfunctions of their transmission and scanning electron microscopes. The laboratory has an internal group and equipment manufacturer's service contract to perform routine scheduled maintenance, and to repair or to coordinate with the vendor for the repa.ir of the electron microscope and related instruments. All laboratory instruments are maintained in accordance with manufacturer's specifications and the requirements of ISO Method 10312:1995. Q�P Section B August 17, 2000 Revision 1 Page 23 of 28 B7 INSTRUMENT CALIBRATION AND FREQUENCY B.7.1 Field Instrument/Equipment Calibration B.7.1.1 Air Sampling Pumps The air sampling pumps with a flow control valve will be evaluated to ensure that they are capable of maintaining a stable flow rate for a given static pressure drop; i.e., the pumps can maintain an initial volume flow rate of within +/- 10% throughout the sampling period. Prior to use the sampling pumps will be tested against the pressure drop created by a 25-mm-diameter; 0.45-,um pore size MCE filter with a 5-�cm pore size MCE backup diffusing filter and cellulose support pad contained in a three-piece cassette with 50-mm cowl at a flow rate of approxirnately 6 liters per minute at STP. B7.1.2 Airtlow Calibration Procedure A flow control valve will be used to regulate the flow rate through the sampling train during sampling. The airflow rate will be determined both before, at the mid-point, and after sampling using a calibrated precision rotameter (Manostat Model 36-546-215). The precision rotameter (a secondary calibration standard) will be calibrated using a primary standard airflow calibrator (Gilabrator electronic flow meter). A detailed written record will be ma.intained of all calibrations. It will include all relevant calibration data, including the following elements: • Gilabrator model and serial number • Rotameter model and serial number • Sampling train (pump, flow control valve, and filter) • X- and Y- coordinate calibration data , • Intercept, slope, and correlation coefficient from a linear regression analysis of the calibration data, and resulting linear regression equation that will be used to determine the sampling flow rate • Relevant calculations • Dry bulb temperature and barometric pressure • Name of person/affiliation that performed the calibration and linear regression analysis Q�p Section B August 17, 2000 Revision 1 Page 24 of 28 B.7.2 Calibration of TEM The TEM shall be aligned according to the specifications of the manufacturer. The TEM screen magnification, electron difl'raction (ED) camera constant, and energy dispersive X-ray analysis (EDXA) system shall be calibrated in accordance with the specifications in ISO Method 10312:1995, Annex B(see Appendix B). Q�P Section B August 17, 2000 Revision 1 Page 25 of 28 B8 IN5PECTION/ACCEPTANCE REQUIREMENTS FOR SUPPLIES AND CONSUMABLES B.8.1 Air Sampling Filter Media Please see Section B.5.2.1 regarding the quality control check of the filter media. QAPP Section B August 17, 2000 Revision 1 Page 26 of 28 B9 DATA ACQUISITION REQUIREMENTS (NON-DIlZECT MEASUREMENTS) B.9.1 Precision The performance goals stated in Section A.7 "Quality Objectives and Criteria for Measurement Data" assume a between-sample variability of 250% or less. The actual between- sample component of variability will be established from the data using standard variance component analyses. If the target precision is not achieved, the false-negative enor rate will be greater than the target rate of 5% and failure to obtain a statistically significant result will not provide strong evidence against the null hypothesis being tested. If a statistically significant result is obtained, the false-negative error rate is of little concern, since it is clear that a false negative error has not occurred. Laboratory precision (coefficient of variation), which is a component of between-sample variability, will be estima.ted from duplicate analyses of a subset of samples. This will provide a measure of the relative contribution of laboratory analysis to the total between-sample variability. This informa.tion will be useful for performing statistical power calculations for follow-up air monitoring surveys. B.9.2 Completeness � An overall measure of completeness will be given by the percentage of samples specified in the study design that yield usable data. The quality control criterion is > 95%. B.9.3 Accuracy Duplicate counts of a subset of randomly selected samples will be performed by having the same microscopist count two sets of grids prepared from the same sample. The results of the counts will be compared, and those which do not fall within the 95% confidence limits for a Poisson variable will be subjected to a further recount and/or a count of a third preparation in an attempt to resolve the discrepancy. A count by a second microscopist may also be employed. In addition, it is expected that a subset of samples will be selected at random by the City of Fort Worth's Quality Assurance Manager to be analyzed by an outside "second" laboratory. Q�P Section B August 17, 2000 Revision 1 Page 27 of 28 B10 DATA MANAGEMENT Commercially available computer hardware and software used to manage measurement data is controlled to ensure the validity of the data generated. Controls include system testing to ensure that no computational enors are generated and evaluation of any proposed changes to the system before they are implemented. Commercially available software does not require testing, but validation of representative calculations are required using alternative means of calculations. B.10.1 Data Assessment Sample data will be reviewed by the laboratory during the reduction, verification, and reporting process. During data reduction, all data will be reviewed for correctness by the microscopist. A second data reviewer will�also verify correctness of the data. Finally, the Laboratory Director (To Be Determined) will provide one additional data review to verify completeness and compliance with the pxoject QAPP. The City of Fort Worth's Quality Assurance Manager, will also select a random sample of the data for the same purpose. Any deficiencies in the data will be documented and identified in the data report. B.10.2 Data Management Field and laboratory data will be entered into a Microsoft Excel spreadsheet to facilitate organization, manipulation, and access to the data. Field data will include information such as sampling date, sample number, sampling site, sample description and location, sample type, air volume, and sampling period. Laboratory data will include information such as sample number, sample date received and analyzed, type of analysis, magnification, grid location, grid square area, filter type, number of grids examined, number of asbestiform structures counted, structure type (fiber, bundle, cluster, or matrix), and structure length and width. An example format for reporting the structure counting data is contained in Figure 7 of I50 Method 10312:1995 (see Appendix B). Q�P Section B August 17, 2000 Revision 1 Page 28 of 28 B.10.3 Statistical Analysis The data collected during the building demolitions and at the landfill will be analyzed using standard analysis of variance (ANOVA) techniques. The ANOVA is a formal statistical procedure that tests whether two or more groups of data are significantly different, on average. The natural logarithm of each sample concentration will be used in the comparisons. Log- transformation is used to make the variances more equal and to provide data that are better appro�mated by a normal distribution. The use of a log-transformation is equivalent to assuming the data follow a log-norma.l distribution; the log-normal distribution is commonly assumed for asbestos measurements and other environmental contaminants. Sample results reported as non- detected will be replaced by the analytical sensitivity divided by two to calculate summary statistics and to perform all statistical analyses. All statistical comparisons will be made at the O.OS level of significance. QAPP Section C August 17, 2000 Revision 1 Page 1 of 4 C ASSESSMENT/OVERSIGHT C1 ASSESSMENT AND RESPONSE ACTIONS C.1.1 Performance and System Audits C.1.1.1 Field Audits The City of Fort Worth staff will be present during the study. The City's Technical Project Officer will conduct periodic field audits. The audit will include, but not limited to the examination of sample collection and equipment calibration procedures, sampling data and chain- of-custody forms, and other sample collection and handling requirements specified in the QAPP. The auditor will document any deviations from the QAPP so that they can be corrected in a timely manner. The auditor will independently measure the flow rate of at least 50% of the air samplers in operation at the time of the audit. The relative accuracy (A%) of the audited flow rates will be established as follows: A% = 100% - RE% where RE%, the relative error, is calculated as: where: RE% _ (F-A) / A x 100 F= flow rate measured by the field crew A= flow rate measured by the auditor. The performance objective for the relative accuracy will be set between 90% and 110%. Prior to leaving the site, the auditor will debrief Tracy Bramlett regarding the results of the audit and any recommendations, if necessary. The results of the audit will be presented in a written report prepared by the auditor. QAPP Section C June 2, 2000 Revision 0 Page 2 of 4 C.1.1.2 Laboratory Audit Ms. Hoover (Quality Assurance Manager) or an independent laboratory quality assurance consultant selected by Ms. Hoover will conduct at least one quality assurance audit of the Laboratory (To Be Determined). The first audit will be conducted at the onset of the project to verify that all procedures specified in the QAPP are understood and are being followed. C1.2 Corrective Actions Sampling and analytical problems may occur during sample collection, sample handling and documentation, sample preparation, laboratory analysis, and data entry and review. Immediate on-the-spot corrective actions will be implemented whenever possible and will be documented in the project record. Implementation of the corrective action will be confirmed in writing by completing a Corrective Action Report (Figure G1). Q�P Section C June 2, 2000 Revision 0 Page 3 of 4 Originator: Date: Project Name/Number: Corrective Action Number: Description of Problem State Cause of Problem (Give Date and Time Identified) State Corrective Action Planned QA Officer Comments: (Include Persons Involved in Action) Signatures Project Manager Comments: QA Officer Project Manager Originator Figure C-1. Corrective Action Report. Q�P Section C June 2, 2000 Revision 0 Page 4 of 4 C2 REPORTS TO MANAGEMENT Effective communication is an integral part of a quality system Planned reports provide a structure to inform management of the project schedule, deviations from the approved QAPP, impact of the deviations, and potential uncertainties in decisions based on the data. The IHST Project Manager (Tracy Bramlett) will provide verbal progress reports to the City of Fort Worth's Project Manager (Kathryn Hansen). These reports will include pertinent information from the data processing and report writing progress reports and corrective action reports, as well as the status of analytical data as determined from conversations with the laboratory. Mr. Bramlett will promptly advise Ms. Hansen on any items that may need corrective action. A written report will be prepared for each field and laboratory audit. These reports will be submitted to the City of Fort Worth's Project Manager. QAPP Section D June 2, 2000 Revision 0 Page 1 of 3 D DATA VALIDATION AND USABILITY D1 DATA REVIEW, VALIDATION, AND VERIFICATION REQUIREMENTS The data will be reviewed and validated by the City of Fort Worth's Quality Assurance Manager or by an independent quality assurance consultant (To Be Determined) selected by the City of Fort Worth. QAPP Section D June 2, 2000 Revision 0 Page 2 of 3 D2 VALIDATION AND VERIFICATION METHODS The analytical laboratory will perform in-house analytical data reduction and verification under the direction of the laboratory's quality assurance manager. The laboratory's quality assurance ma.nager is responsible for assessing data quality and advising of any data rated as "unacceptable" or other notations which would caution the data user of possible unreliability. The City of Fort Worth's quality assurance manager or an independent quality assurance consultant (To Be Determined) will conduct a systematic review of the data to verify compliance with the established quality criteria in the QAPP and ISO Method 10312:1995. The data review will identify any out-of-control data points and data omissions. Based on the extent of the deficiency and its importance in the overall data set, the laboratory may be required to re-analyze the sample. Included in the data validation of a sample set will be an assessment of chain-of- custody and analyses of field quality control samples (open and closed field blanks). The precision of the data will be determined by calculating the coefficient of variation for the replicate and duplicate sample analyses as well as the analyses of the field duplicate samples. Q�P Section D June 2, 2000 Revision 0 Page 3 of 3 D3 RECONCILIATION WITH DATA QUALITY OBJECTIVES The proposed statistical methods to analyze the data and determine the significance of departures (positive and negative) from the assumptions established in the Data Quality Objectives are presented in Section B.10. Q�P Section E June 2, 2000 Revision 0 Page 1 of 1 El REFERENCES U.S. Environmental Protection Agency. EPA Guidance for Quality Assurance Project Plans — EPA QA/G-5. EPA/600/R-98/018, February 1998. 2. City of Fort Worth, Texas, Project XL Proposal "Asbestos Management in the Demolition of Substandard Structures as a Nuisance Abatement,"(September 30, 1999). Prepared by Department of Environmental Management, City of Fort Worth, Texas 76102. 3. U.S. Environmental Protection Agency. Ambient Airborne Asbestos Levels in Alviso, California. Prepared by John R. Kominsky and Ronald W. Freyberg. Contract No. 68- CO-0048, Apri127, 1995. 4. U.S. Environmental Frotection Agency. Ambient Air Monitoring at the Moss Landing Harbor District: Moss Landing, California. Prepared by John R. Kominsky and Ronald W. Freyberg. EPA Contract No. 68-D2-0058, October 17, 1990. 5. Kominsky, 7.R., R.W. Freyberg, J.A. Brownlee, D.R. Greber. AHERA Clearance at Twenty Abatement Sites. EPA/600/52-91/028, August 1991. 6. Berman, D.W. and Chatfield, E.J. (1990). Superfund Method for the Determination of Asbestos in Ambient Air. Part 2: Technical Background Document. Prepared for the Environmental Services Branch of the U.S. Environmental Protection Agency. Washington, D.C. EPA/540/2-90/OOSb. Q.�p Appendix A June 2, 2000 Revision 0 Appendix A Comparison of the Asbestos NESHAP and the Fort Worth Method for the Demolition of Substandard Structures Table - 1 Comparison of the Asbestos NESHAP and the Fort Worth Method for Demolition of Substandard Structures • , . ..� o • a •. �. . e �.,, ..._...., �. �. . D. _ � � - - R� �:x s�"`'. '� • '� A Unilcai $taEes A r A Unilni States �� Egencymental Pr.�tx6.n `'�M� C �H �geno�mental Protecticn � �OTJ'�'�O}ZTH a� ASBESTOS ASSESSMENT Not required. Full AHERA Level Asbestos t=u�l AHERA Level Asbestas Assessment. Assessment. DEMOLITION NOTIFICATION LVritten notification as early as possible Wntten notificalion at least ten working WriUen notifir.a�ion at least TWO ��rorkiny before, but not later [han ihe following days be(ore work begins. days before vrork beyins. working day. REMOVAL oF RncM PRIOR To RACM� not removed prior to demolition. Remove RACM unde� full containment if Rf\CNi not removeci prior to demolition. DEMOLITION there is: Note: SPRAY-ON FIREPROOfING 1. At Ieas[ 80 linear meters (260 linear AMD LARGE QUANTITIES Of feet) on pipes or at least 15 square THERh1AL SYSTEM meters (160 square (eet) on other INSULATION WILL BE facility componenls; or ADDRESSED UNDER FULL 2. At least 1 cubic meler (35 cubic (eet) CONTAINMENT CONDITIONS. off facility components where the length or area could not be measured previously. Adequatety wet asbeslos-containing waste material. Afterwetling, sea� in leak-tight containers while weL I( materials will not fit into containers without additional breakage, put materials in leak-tight wrapping. Label containers or wrapped materials using OSHA compliant waming labels. EMIs51oN5 coNTRo�s ouRiNG Discharge no Visible Emissions (rom Oischarge no Visible Emissions from Discharye no Visible Emissions Fro�n oEMouiloN RACM or asbestos-containing waste RACM or asbestos-containing waste R,4CM or asbestos-containing waste material. materiaL matenal. HANOUNG PROCEDl1RE5 FOR Adequatel� wet the portion of the facility Adequately wet asbestos-containing Adequately w«;t THE FACILITY during oEMo��71oN AssEsios- that contains RACtiI during Ihe wrecking waste material at all times after the wrecking operation CON7AINING WASTE MATERIAL operation. demolilion and keep wet during handling and loading for Iransport to a disposal Adequatety wet DEMOIITION DEBRIS � Adequateiy wet asbestos-containing site. at all hmes aker demolition and keep waste material at aIi times aker wet duriny handling and loading for demo�ition and keep wet during handling Asbestos-containing waste materials iransport �o a disposal site. and Ioading for transpoR to a disposal demolished in place do not have to be site. seaied i� leak-tight conlainers or WASTE MATERIALS TO BE wrapping, but may be transported and DISPOSED IN BULK WITHIN Asbestos-containing waste materials do disposed o( in bulk. TRAILERS COVERED WITH TARPS. not have to be sealed in leak-tight containers or wrapping, but may be Note: Does not apply to Category I Non- transported and disposed of in bulk. Friable ACM waste and Calegory il Non-Friable ACM waste that did not become crumbled, pulverized, or reduced to powder. TRANSPORTATION OF tvlark vehicles used to �ransport Mark vehicles used [o transport tilark vehicles uszd to Iransport DEMOUTION ASBESTOS- as6estos-containing waste material asbestos-containing waste material asbestos-cont�ining waste inatenal CONTAINING WASTE MATERIAL during ihe loading and unloading of during the loading and unloading of during the loadiny and unloading of waste so that signs are visible. waste so lhat signs are visible. was�e so �hat signs are visible. Ntanifest RACM shipments. Manifest RACM shipments. Manifest RAC(vi shipmenis. DISPOSAL OF DEMOLITION �eposit all asbestos-coNaining waste Deposit all asbestos-containing waste Deposil all asbestos-containing waste ASBESTOS•CONTAINING material as soon as practicaf at a waste material as soon as practical at a waste material as soon as praclical a[ a waste WASTE MATERIAL disposal site approved for asbestos disposal site approved for asbestos disposal site approved for asbestos disposal, uNess it is Category I Non- disposal, uniess it is Calegory I Non- disposal, unless it is Cateyory I�lon Friable ACM that is not RACtiI Friable ACM that is not R,4CM. Friable ACM lhat is not RACM SITE SUPERVI510N DURING At least one �ep�esentative trained in the At least one �epresentative ltained in the At Ieas� one representatrie trained in the DEMOLITION NESHAP shall be present on-site. NESHAP shall be present on-site. NESHAP shall be pres-e�t on-si[e. RecoRos Maintain waste disposal records (or at Maintain wasle disposal records for al tvlaintam waste disposal records tor a[ MAINTENANCE least t�•ro years. leasl hvo years. least hvo years. S70RMWATER MANAGEMENT Not specifed Not specified. Comply with Chapter 12.5, Article Iil, "Storm Water Protection," Gode of the City of Fort Worth. Use best � management practices to controi runoff as necessary. OUTDOOR AIR MONITORING OSHA monibring of �mor'Rers. OSHA monitoring of workers. AREA SAMPLWG TO BE PERFORNIED AT ALL FOUR CORNERS OF THE JOB SITE. OSHA monitoriny of svo�kers. WETTING PROCEDURES Adequately we[. Adequately weL Utilize fire hose equipped with variable rete nozzle to allow for °mistin " ' � ' � I �' � � QAPP Appendix B June 2, 2000 Revision 0 Appendix B ISO Method 10312:1995 Ambient Air - Determination of Asbestos Fibres -- Direct-Transfer Transmission Electron Microscopy Method S � i INTERNATIONAL STANDARD. iso 10312 First edition 1995-05-01 Ambient air — Determination of asbestos fibres — Direct-transfer transmission electron microscopy method Air ambiant — Determination des fibres d'amiante — Methode de microscopie electronique a transmission directe Reference number ISO 10312:1995(E) ` Printed for you by Document Center lnc., 111 Industrial Road Suite 9, Belmont, CA 94002-40d4 Phone: 650-591-7600 Fax: 650-591-761' (�0 10312:1995(E) Contents Page 1 Scope ....................................................... ............................. ....... 1 2 Normative references ..................................................................... 2 3 Definitions ................................................................................. 2 4 Principle ..............:...................................................................... 3 5 Symbols of units and abbreviations ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4 6 Reagents ................................................................................... 5 7 Apparatus .................................................................................. 5 8 Air sample collection ............................................................... 10 9 Procedure for analysis ............................................................. 11 10 Performance characteristics .................................................. 18 11 Test report ............................................................................. 19 Annexes A Determination of operating conditions for plasma asher ,..... 22 B Calibration procedures ........................................................... 23 C Structure counting criteria .......................................... 25 D Fibre identification procedure ................................................ 33 E Determination of the concentrations of asbestos fibres and bundles longer than 5 µm, and PCM equivalent asbestos fibres ....... 42 F Calculation of results .............................................................. 43 G Strategies for collection of air samples ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 47 H Methods for removal of gypsum fibres ................................. 4g JBibliography ............................................................................ 49 � ISO 1995 All rights reserved. Unless otherwise specified, no part of this pubiication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the publisher. International Organization for Standardization Case Postale 56 • CH-1211 Geneve 20 • Switzerland Printed in Switzerland 4 � ; , r � ISO ISO 10312:1995(E) Foreword ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies (ISO member bodies). The work of preparing International Standards is normally carried out through ISO technical committees. Each member body interested in a subject for which a technical committee has been established has the right to be represented on that committee. International organizations, governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization. Draft international Standards adopted by the technical committees are circulated to the member bodies for voting. Publication as an International Standard requires approval by at least 75 % of the member bodies casting a vote. International Standard ISO 10312 was prepared by Technical Committee ISO/TC 146, Air quality, Subcommittee SC 3, Ambient atmospheres. Annexes A, B, C, D, E and F form an integral part of this International Standard. Annexes G, H and J are for information only. N ISO 10312:1995(E) Introduction „ �� This Internationai Standard is applicable to the determination of airborne asbestos in a wide range of ambient air situations, including the interior atmospheres of buildings, and for detailed evaluation of any atmosphere in which asbestos structures are likely to be present. Because the best available medical evidence indicates that the numerical fibre concentration and the fibre sizes are the relevant parameters for evaluation of the inhalation hazards, a fibre counting technique is the only logical approach. Most fibres in ambient atmospheres are not asbestos, and therefore there is a requirement for fibres to be identified. Many airborne asbestos fibres in ambient atmospheres have diameters below the resolution limit of the optical microscope. This International Standard is based on transmission electron microscopy, which has adequate resolution to aliow detection of smail fibres and is currently the o�ly technique capable of unequivocal identification of the majority of individual fibres of asbestos. Asbestos is often found, not as single fibres, but as very complex, aggregated struc- tures which may or may not be also aggregated with other particles. The fibres found suspended in an ambient atmosphere can often be identified unequivocally, if a sufficient measurement effort is expended. However, if each fibre were to be identified in this way, the analysis would become prohibitively expensive. Because of instrumental deficiencies or because of the nature of the particulate, some fibres cannot be positively identified as asbestos, even though the measurements all indicate that they could be asbestos. Subjective factors therefore contribute to this measurement, and consequentiy a very precise definition of the procedure for identifica- tion and enumeration of asbestos fibres is required. The method specified in this International Standard is designed to provide the best description possible of the �ature, numerical concentration, and sizes of asbestos- containing particles found in an air sample. This International Standard is necessarily complex, because the instrumental techniques used are com- plex, and aiso because a very detailed and logical procedure must be specified to reduce the subjective aspects of the measurement. The method of data recording specified in this International Standard is de- signed to allow re-evaluation of the structure counting data as new med- ical evidence becomes available. Ali of the feasible specimen preparation techniques result in some modification of the airborne particulate. Even the coilection of particles from a three-dimensional airborne dispersion onto a two-dimensional filter surface can be considered a modification of the particulate, and some of the particles in most samples are modified by the specimen preparation procedures. However, the procedures spec- ified in this International Standard are designed to minimize the disturb- ance of the collected particulate material, and the effect of those disturbances which do occur can be evaluated. This International Standard describes the method of analysis for a single air filter. However, one of the largest potential errors in characterizing asbestos in ambient atmospheres is associated with the variability be- tween filter samples. For this reason, it is necessary to design a replicate sampling scheme in order to determine this International Standard's ac- curacy and precision. iv 6 , , INTERNATIONAL STANDARD o ISO ISO 10312:1995(Ei Ambient air — Determination of asbestos fibres — Direct-transfer transmission electron microscopy method 1 Scope 1.1 Substance determined This I�ternational Standard specifies a reference method using transmission electron microscopy for the determination of the concentration of asbestos structures in ambient atmospheres and includes measurement of the lengths, widths and aspect ratios of the asbestos structures. The method allows deter- mination of the type(s) of asbestos fibres present. The method cannot discriminate between individual fibres of the asbestos and non-asbestos analogues of the same amphibole mineral. 1.2 Type of sample The method is defined for polycarbonate capillary-pore filters or cellulose ester (either mixed esters of cellu- lose or cellulose nitrate) filters through which a known volume of air has been drawn. The method is suitable for determination of asbestos in both exterior and building atmospheres. 1.3 Measuring range The range of concentration which can be determined is 50 structures/mm2 to 7 000 structures�mm2 on the filter. The air concentrations represented by these values are a function of the volume of air sampled. There is no lower limit to the dimensions of asbestos fibres which can be detected. In practice, microscopists vary in their ability to detect very small asbestos fibres. Therefore, a minimum length of 0,5 µm has been defined as the shortest fibre to be incorporated in the reported results. 1.4 Limit of detection The limit of detection theoretically can be lowered in- definitely by filtration of progressively larger volumes of air and by extending the examination of the speci- mens in the electron microscope. In practice, the lowest achievable limit of detection for a particular area of TEM specimen examined is controlled by the total suspended partictilate concentration. For total suspended �particulate concentrations of ap- proximately 10 µg/m , corresponding to clean, rural atmospheres, and assuming filtration of 4 000 litres of air, an analytical sensitivity of 0,5 structure/I can be obtained, equivalent to a limit of detection of 1,8 structure/I, if an area of 0,195 mm2 of the TEM specimens is examined. If higher total suspended particulate concentrations are present, the volume of air filtered must be reduced in order to maintain an acceptable particulate loading on the filter, leading to a proportionate increase in the analytical sensitivity. Where this is the case, lower limits of detection can be achieved by increasing the area of the TEM speci- mens that is examined. In order to achieve lower limits of detection for fibres and bundies longer than 5 µm, and for PCM equivalent fibres, lower magni- fications are spe,cified which permit more rapid ex- amination of larger areas of the TEM specimens when the examination is limited to these dimensions of fi- bre. The direct analytical method cannot be used if the general particulate loading of the sample collection filter exceeds approximately 10 µg�cmZ of filter sur- face, which corresponds to approximately 10 % cov- erage of the collection filter by pa�ticulate. if the total suspended particulate is largely organic material, the limit of detection can be lowered significantly by using an indirect preparation method. 1 ISO 10312:1995(E� � ISO = 2 Normative references The following standards contain provisions which, through reference in this text, constitute provisions of this Internationai Standard. At the time of pubiica- tion, the editions indicated were valid. All standards are subject to revision, and parties to agreements based on this International Standard are encouraged to investigate the possibility of applying the most re- cent editions of the standards indicated below. Members of IEC and ISO maintain registers of cur- rently valid International Standards. ISO 4225:1994, Air quality — Gene�a! aspecis — Vo- cabulary. ISO 4226:1993, Ai� quality — General aspects — Units of ineasurement. ISO Standard Handbook No. 2:1993, Quantities and units. ISO Standard Handbook No. 3:1989, Siatistical Meth- ods. 3 Definitions For the purposes of this International Standard, the following definitions apply (see also ISO 4225). 3.1 acicular: The shape of an extremely slender crystal with cross-sectional dimensions which are small relative to its length, i.e, needle-like. 3.2 amphibole: A group of rock-forming ferromagnesium silicate minerals, ciosely related in crystal form and composition, with the nominal for- mula: Ao or �BZC5T8022(OH,F,CI)2 where A=K,Na B= FeZ+, Mn, Mg, Ca, Na C= AI, Cr, Ti, Fe3+, Mg, FeZ+ T= Si, AI, Cr, Fe3+, Ti In some varieties of amphibole, these elements can be partially substituted by Li, Pb or Zn. Amphibole is characterized by a cross-linked double chain of Si-O tetrahedra with a silicon:oxygen ratio of 4:11, by co- lumnar or fibrous prismatic crystals and by good prismatic cleavage in two directions parallel to the 2 crystal faces and intersecting at angles of about 56° and 124°. 3.3 amphibole asbestos: Amphibole in an asbestiform habit. 3.4 analytical sensitivity: The calculated airborne asbestos structure concentration in asbestos structures/litre, equivalent to counting of one asbestos structure in the analysis. The method in this International Standard does not specify an analyticai sensitivity. 3.5 asbestiform: A specific type of mineral fibrosity in which the fibres and fibrils possess high tensile strength and flexibility. 3.6 asbestos: A term applied to a group of silicate minerals belonging to the serpentine and amphibole groups which have crystallized in the asbestiform habit, causing them to be easily separated into long, thin, strong fibres when crushed or processed. The Chemical Abstracts Service Registry Numbers of the most common asbestos varieties are: chrysotile (12001-29-5), crocidolite (12001-28-4), grunerite asbestos (amosite) (12172-73-5), anthophyliite asbestos 'i77536-67-5), tremolite asbestos (77536-68-6) and actinolite asbestos (77536-66-4). 3.7 asbestos structure: A term applied to any con- nected or overlapping grouping of asbestos fibres or bundles, with or without other particles. 3.8 aspect ratio: The ratio of length to width of a particle. 3.9 blank: A structure count made on TEM speci- mens prepared from an unused filter, to determine the background measurement. 3.10 camera length: The equivalent projection length between the specimen and its electron dif- fraction pattern, in the absence of lens action. 3.11 chrysotile: A fibrous mineral of the serpentine group which has the nominal composition M9aS�z�S(OH)4 Most natural chrysotile deviates little from this nomi- nal composition. In some varieties of chrysotile, minor substitution of silicon by AI3+ may occur. Minor sub- stitution of magnesium by AI3+, FeZ�, Fe3+ Niz+ Mn2+ and Co2+ may also be present. Chrysotile is the most prevalent type of asbestos. 3.12 cleavage: The breaking of a mineral along one of its crystallographic directions. 9 1 - o ISO t . 3.13 cleavage fragment: A fragment of a crystal that is bounded by cleavage faces. 3.14 cluster: A structure in which two or more fi- bres, or fibre bundles, are randomly oriented in a connected grouping. 3.15 d-spacing: The distance between identical ad- jacent and parallel planes of atoms in a crystal. 3.16 electron diffraction: A technique in electron microscopy by which the crystal structure of a speci- men is examined. 3.17 electron scattering power: The extent to which a thin layer of substance scatters electrons from their original directions. 3.18 energy dispersive X-ray analysis: Measure- ment of the energies and intensities of X-rays by use of a solid state detector and muitichannel analyser system. � 3.19 eucentric: The condition when the area of in- terest of an object is placed on a tilting axis at the intersection of the electron beam with that axis and is in the plane of focus. 3.20 field blank: A filter cassette which has been taken to the sampling site, opened, and then closed. Such a filter is used to determine the background structure count for the measurement. 3.21 fibril: A single fibre of asbestos, which cannot be further separated longitudinally into smaller com- ponents without losing its fibrous properties or ap- pearances. 3.22 fibre: An elongated particle which has parallel or stepped sides. For the purposes of this Interna- tionai Standard, a fibre is defined to have an aspect ratio equal to or greater than 5:1 and a minimum length of 0,5 µm. 3.23 fibre bundie: A structure composed of paraliel, smaller diameter fibres attached along their lengths. A fibre bundle may exhibit diverging fibres at one or both ends. 3.24 fibrous structure: A fibre, or connected grouping of fibres, with or without other particles. 3.25 habit: The characteristic crystal growth form, (or combination of these forms), of a mineral, inciud- ing characteristic irregularities. 3.26 limit of detection: The caiculated airborne asbestos structure concentration in structures per li- ISO 1Q312:1995(E) tre, equivalent to counting 2,99 asbestos structures in the analysis. 3.27 matrix: A structure in which one or more fi- bres, or fibre bundles, touch, are attached to, or par- tially concealed by, a single particle or connected group of nonfibrous particles. 3.28 Miller index: A set of either three or four inte- ger numbers used to specify the orientation of a crystailographic plane in �elation to the crystal axes. 3.29 PCM equivalent fibre: A fibre of aspect ratio greater than or equal to 3:1, longer than 5 µm, and which has a diameter between 0,2 µm and 3,0 µm. 3.30 PCM equivalent structure: A fibrous structure of aspect ratio greater than or equal to 3:1, longer than 5 µm, and which has a diameter between 0,2 µm and 3,0 µm. 3.31 primary structure: A fibrous structure that is a separate entity in the TEM image. 3.32 replication: A procedure in electron microscopy specimen preparation in which a thin copy, or replica, of a surface is made. 3.33 selected area electron diffraction: A tech- nique in electron microscopy in which the crystal structure of a small area of a sampie is examined. 3.34 serpentine: A group of common rock-forming minerals having the nominal formula Mg3s�2�5��fi�q 3.35 structure: A single fibre, fibre bundle, cluster or matrix. 3.36 twinning: The occurrence of crystals of the same species joined together at a particular mutual orientation, such that the relative orientations are re- lated by a definite law. 3.37 unopened fibre: An asbestos fibre bundle of large diameter which has not been separated into its constituent fibrils pr fibres. 3.38 zone-axis: The line or crystaliographic direction through the centre of a crystal which is parallel to the intersection edges of the crystal faces defining the crystal zone. 4 Principle A sample of airborne particulate is collected by draw- ing a measured volume of air through either a 3 ISO 103'12:1995(E) capillary-pore polycarbonate membrane fiiter of maxi- mum pore size 0,4 µm or a cellulose ester (either mixed esters of cellulose or cellulose nitrate) mem- brane fi�ter of rnaximum pore size 0,45 µm by means of a battery-powered or mains-powered pump. TEM specimens are prepared from polycarbonate filters by applying a thin film of carbon to the filter surface by vacuum evaporation. Small areas are cut from the carbon-coated filter, supported on TEM specimen grids, and the filter medium is dissolved away by a solvent extraction procedure. This procedure leaves a thin film of carbon which bridges the openings in the TEM specimen grid, and which supports each particle from the original filter in its original position. Gellulose ester filters are chemically treated to collapse the pore structure of the filter, and the surface of the collapsed filter is then etched in an oxygen plasma to ensure that ali particles are exposed. A thin film of carbon is evaporated onto the filter surface and smail areas are cut from the fiiter. These sections are supported on TEM specimen grids and the filter medium is dis- solved away by a solvent extraction procedure. The TEM specimen grids from either preparation method are examined at both low and high magni- fications to check that they are suitable for analysis before carrying out a quantitative structure count on randomly-selected grid openings. In the TEM analysis, electron diffraction iED) is used to examine the crystal structure of a fibre, and its elemental composition is determined by energy dispersive X-ray analysis (EDXA). For a number of reasons, it is not possible to identify each fibre unequivocally, and fibres are clas- sified according to the techniques which have been used to identify them. A simple code is used to re- cord, for each fibre, the manner in which it was clas- sified. The fibre classification procedure is based on successive inspection of the morphology, the electron diffraction pattern for a selected area, and the qual- itative and quantitative energy dispersive X-ray ana- lyses. Confirmation of the identification of chrysotile is done only by quantitative ED, and confirmation of amphibole is done oniy by quantitative EDXA and quantitative zone axis ED. In addition to isolated fibres, ambient air samples of- ten contain more complex aggregates of fibres, with or without other particles. Some particles are com- posites of asbestos fibres with other materials. Indi- vidual fibres and structures that are more complex are referred to as "asbestos structures". A coding system is used to record the type of fibrous structure, and to provide the optimum description of each of these complex structures. The two codes remove the re- quirement to interpret the structure counting data from the microscopist, and allow this evaluation to be made later without the requirement for re- 4 � examination of the TEM specimens. Several levels of analysis are specified, the higher levels providing a more rigorous approach to the identification of fibres. The procedure permits a minimum required fibre identification criterion to be defined on the basis of previous knowledge, or lack of it, about the particular sample. Attempts are then made to achieve this min- imum criterion for each fibre, and the degree of suc- cess is recorded for each fibre. The lengths and widths of all classified structures and fibres are re- corded. The number of asbestos structures found on a known area of the microscope sample, together with the equivalent volume of air filtered through this area, is used to calculate the airborne concentration in asbestos structures/litre of air. 5 Symbols of units and abbreviations 5.1 Symbols of units (see aiso ISO 4226 and ISO No, 2) eV = electron volt kV = kilovolt I�min = litres per minute µg = microgram (10-6 gram► µm = micrometre (10—s metre) nm = nanometre (10-9 metre) W = watt 5.2 Abbreviations DMF Dimethyiformamide DE Electron diffraction EDXA Energy dispersive X-ray analysis FWHM Full width, half maximum HEPA High efficiency particle absolute MEC Mixed esters of ceilulose PC Polycarbonate PCM Phase contrast optical microscopy SAED Selected area electron diffraction SEM Scanning electron microscope STEM Scanning transmission electron microscope TEM Transmission electron microscope o ISO (SO 10312:1995(E) UICC Union Internationale Contre le Cancer 6 Reagents During the analysis, unless otherwise stated, use only reagents of recognized analytical grade and water (6.11. WARNING — Use the reagents in accordance with ; the appropriate health and safety regulations. , 6.1 water, fibre-free. A supply of freshly distilled, fibre-free water, or an- other source of fibre-free, pyrogen-free water shall be used. 6.2 Chloroform, analytical grade, distilled in glass, preserved with 1 % (VM ethanol. 6.3 7-Methyl-2-pyrrolidone. 6.4 Dimethylformamide. 6.5 Glacial acetic acid. 6.6 Acetone. 7 Apparatus 7.1 Air sampling — Equipment and consumable supplies 7.1.1 Filter cassette Field monitors, comprising 25 mm to 50 mm diam- eter three-piece cassettes, with cowls which project less than 2 cm in front of the filter surface shall be used for sample collection. The cassette shall be loaded with either a capillary pore polycarbonate filter of maximum pore size 0,4 µm or an MEC or cellulose nitrate filter of maximum pore size 0,45 µm. Either type of fiiter shall be backed by a 5 µm pore size MEC or cellulose nitrate filter, and supported by a cellulose back-up pad. When the filters are in position, an elas- tic cellulose band or adhesive tape shall be applied to prevent air leakage. Suitable precautions shall be taken to ensure that the filters are tightly clamped in the assembly, so that significant air leakage around the filter cannot occur. Representative filters from the filter lot shall be ana- lysed as specified in 9.7 for the presence of asbestos structures before any are used for air sample col- lection. 7.1.2 Sampling pump The sampling pump shail be capable of a flow-rate sufficient to achieve the desired analytical sensitivity. The face velocity through the filter shall be between 4,0 cm/s and 25,0 cm/s. The sampling pump used shall provide a non-fluctuating airflow through the fil- ter, and shali maintain the initial volume fiow-rate to within ± 10 % throughout the sampling period. A constant flow or critical orifice controlled pump meets these requirements. Flexible tubing shall be used to connect the filter cassette to the sampling pump. A means for calibration of the flow-rate of each pump is also required. 7.1.3 Stand A stand shall be used to hold the filter cassette at the desired height for sampling, and shall be isolated from the vibrations of the pump (7.1.2). 7.1.4 Variable area flowmeter A calibrated variable are a flowmeter with a range of approximately 1 I/min to 10 I(min is required for cali- bration of the air sampling system. The variable a�ea fiowmeter shali be cleaned before use to avoid transfer of asbestos contamination from the flowmeter to the sample being coilected. 7.2 Specimen preparation laboratory Asbestos, particularly chrysotile, is present in varying quantities in many laboratory reagents. Many buiiding materials also contain significant amounts of asbestos or other mineral fibres which may interfere with the analysis if they are inadvertently introduced during preparation of specimens. It is most important to en- sure that, during preparation, contamination of TEM specimens by any extraneous asbestos fibres is min- imized. All specimen preparation steps shall therefore be performed in an environment where contamination of the sample is minimized. The primary requirement of the sample preparation laboratory is that a bla�k determination shall yield a result which will meet the requirements specified in 9.7. A minimum facility considered suitable for preparation of TEM specimens is a laminar flow hood with positive pressure. How- ever, it has been established that work practices in specimen preparation appear to be more important than the tape of clean handling facilities in use. Prep- aration of samples shall be carried out only after ac- ceptable blank values have been demonstrated. NOTE 1 It is recommended that activities invo�ving ma- nipulation of bulk asbestos sampies not be performed in the 5 ISO 10312:9995(E) same area as TEM specimen preparation, because of the possibilities of contaminating the TEM specimens. 7.3 Equipment for analysis 7.3.1 Transmission electron microscope A TEM operating at an accelerating potential of 80 kV to 120 kV, with a resolution better than 1,0 nm, and a magnification range of approximately x 300 to x 100 000 shall be used. The ability to obtain a direct screen magnification of about x 100 000 is � ISO " - a necessary for inspection of fibre morphology; this magnification may be obtained by supplementary op- tical enlargement of the. screen image by use of a binocular if it cannot be obtained directly. It is also required that the viewing screen of the microscope be calibrated such that the lengths and widths of fibre images down to 1 mm width can be measured in in- crements of 1 mm, regardless of image orientation. This requirement is often fulfilled through the use of a fluorescent screen with calibrated gradations in the form of circles, as shown in figure 1. a 5 6 7 Figure 1— Example of calibration markings on TEM viewing screen 6 9 a ISO . For Bragg angles less than 0,01 rad, the TEM shall be capable of performing ED from an area of 0,6 µm2 or less, selected from an in-focus image at a screen magnification of x 20 000. This performance require- ment defines the minimum separation between parti- cles at which independent ED patterns can be obtained from each particie. If SAED is used, the performance of a particular instrument may normally be calculated using the following equation `s A= 0,785 4 x( M+ 2 OO00583) � i where A is the effective SAED area, in square micrometres; D is the diameter, in micrometres, of the SAED aperture; M is the magnification of the objective lens; CS is the spherical aberration coefficient, in miilimetres, of the objective lens; 8 is the maximum required Bragg angle, in radians. It is not possible to reduce the effective SAED area indefinitely by the use of progressively smaller SAED apertures, because there is a fundamental limitation imposed by the sphericai aberration coefficient of the objective lens. If zone-axis ED analyses are to be performed, the TEM shall incorporate a goniometer stage which per- mits the TEM specimen to be either a) � b) rotated through 360°, combined with tilting through at least + 30° to — 30° about an axis in the plane of the specimen; tilted through at least + 30° to — 30° about two perpendicular axes in the plane of the specimen. The analysis is greatly facilitated if the goniometer permits eucentric tilting, although this is not essential. If EDXA and zone-axis ED are required on the same fibre, the goniometer shall be of a type which permits tilting of the specimen and acquisition of EDXA spec- : tra without changing the specimen holder. The TEM shall have an illumination and condenser lens system capable of forming an electron probe of � diameter less than 250 nm. NOTE 2 Use of an anti-contamination trap around the specimen is recommended if the required instrumental performance is to be obtained. ISO 10312:1995(E) 7.3.2 Energy dispersive X-ray analyser The TEM shall be equipped with an energy. dispersive X-ray analyser capable of achieving a resolution better than 180 eV (FWHMi on the MnKa. Since the per- formance of individuai combinations of TEM and EDXA equipment is dependent on a number of ge- ometrical factors, the required performance of the combination of the TEM and X-ray analyser is speci- fied in terms of the measured X-ray intensity obtained from a fibre of smali diameter, using a known electron beam diameter. Solid state X-ray detectors are least sensitive in the low energy region, and so measure- ment of sodium in crocidolite shall be the perform- ance criterion. The combination of electron microscope and X-ray analyser shall yield, under rou- tine analytical conditions, a background-subtracted NaKa integrated peak count rate of more than 1 count per second (cps) from a fibre of UICC crocidolite, 50 nm in diameter or smaller, when irradiated by an electron probe of 250 nm diameter or smaller at an accelerating potentiai of 80 kV. The peak/background ratio for this performance test shall exceed 1,0. The EDXA unit shall provide the means for subtraction of the background, identification of elemental peaks, and calculation of background-subtracted peak areas. 7.3.3 Computer Many repetitive numerical calculations are necessary, and these may be performed conveniently by rela- tively simple computer programmes. For analyses of zone-axis ED pattern measurements, a computer with adequate memory is required to accommodate the more complex programmes invoived. 7.3.4 Plasma asher For preparation of TEM specimens from MEC filters, a plasma asher, with a radio frequency power rating of 50 W or higher, shail be used to etch the surface ` of collapsed MEC filters. The asher shali be supplied with a controlled oxygen fiow, and shall be modified, if necessary, to provide a valve to control the speed of air admission 'so that rapid air admission does not disturb particulates from the surface of the filter after the etching step. NOTE 3 It is recommended that filters be fitted to the oxygen supply and the air admission line. 7.3.5 Vacuum coating unit A vacuum coating unit capable of producing a vacuum better than 0,013 Pa shail be used for vacuum de- position of carbon on the membrane filters. A sample 7 ISO 10312:1995(E) holder is required which wili allow a glass microscope slide to be continuously rotated during the coating procedure. NOTE 4 A mechanism which also allows the rotating slide to be tilted through an angle of approximately 45° during the coating procedure is recommended. A liquid ni- trogen cold trap above the diffusion pump may be used to minimize the possibility of contamination of the filter sur- faces by oil from the pumping system. The vacuum coating unit may also be used for deposition of the thin film of gold, or other calibration material, when it is required on TEM specimens as an internal calibration of ED patterns. 7.3.6 Sputter coater A sputter coater with a gold target may be used for deposition of gold onto TEM specimens as an integral calibration of ED patterns. Other calibration materials are acceptable. Experience has shown that a sputter coater allows better control of the thickness of the calibration material. 7.3J Solvent washer �Jaffe washer) The purpose of the Jaffe washer is to allow dissoi- ution of the filter polymer while leaving an intact evaporated carbon film supporting the fibres and other particles from the filter surface. One design of Giass Petri dish (0700mmx15mr � a washer which has been found satisfactory for vari- ous solvents and filter media is shown in figure 2. In general, either chloroform or 1-methyl-2-pyrrolidone has been used for dissolving polycarbonate filters and dimethylformamide or acetone has been used for dissolving MEC or cellulose nitrate filters. The higher evaporation rates of chloroform and acetone require that a reservoir of 10 ml to 50 ml of solvent be used, which may need replenishment during the procedure. Dimethylformamide and 1-methyl-2-pyrrolidone have lower vapour pressures and much smaller volumes of solvent may be used. It is recommended that all washers be used in a fume hood, and when speci- mens are not being inserted or removed, the Petri dish lid shall be in place during the solvent dissolution. The washer shall be cieaned before it is used for each batch of specimens. 7.3.8 Condensation washer For more rapid dissolution of the filter polymer, or if difficulties are experienced in dissolving the filter polymer, use a condensation washer, consisting of a flask, condenser and cold finger assembly, with a heating mantle and means for controlling the temper- ature. A suitable assembly is shown in figure 3, using either acetone or chloroform as the solvent, depend- ing on the type of filter. Dimensions in centimetres Flar.imn micmernno NOTE — Solvent is added until the meniscus contacts the underside of the stainless steel mesh bridge. Figure 2— Example of design of solvent washer (Jaffe washer) tainiess steel mesh ridge (50 mesh) ; .� o ISO Condenser Specimen ISO 10312:1995(E Water drain Cold finger E-- Cold water source ID nostaticaily controiled �g mantle Figure 3— Example of design of condensation washer 7.3.9 Slide warmer or oven Use either a slide warmer or an oven for heating slides during the preparation of TEM specimens from MEC or cellulose nitrate filters. It is required to main- tain a temperature of 65 °C to 70 °C. 7.3.10 Ultrasonic bath An ultrasonic bath is necessary for cleaning the appa- ratus used for TEM specimen preparation. 7.3.11 Carbon grating replica A carbon grating replica with about 2 000 parallel lines per millimetre shali be used to calibrate the magni- fication of the TEM. 7.3.12 Calibration specimen grids for EDXA TEM specimen grids prepared from dispersions of calibration minerals are required for calibration of the EDXA system. Some suitable calibration minerals are riebeckite, chrysotile, halloysite, phlogopite, woilas- tonite and bustamite. The mineral used for calibration of the EDXA system for sodium shall be prepared using a gold TEM grid. 7.3.13 Carbon rod sharpener The use of necked carbon rods, or equivalent, aliows the carbon to be evaporated onto the fiiters with a minimum of heating. 7.3.14 Disposable tip micropipettes A disposable tip micropipette, capable of transferring a volume of approximate�y 30 µl, is necessary for the preparation of TEM specimen grids from MEC filters. 7.4 Consumable supplies 7.4.1 Copper electron microscope grids Copper TEM grids with 200 mesh are recommended. Grids which have grid openings of uniform size such that they meet the requirement specified in 9.6.2 shali be chosen. To facilitate the relocation of individual grid openings for quality assurance purposes, the use of grids with numerical or alphabetical indexing of indi- vidual grid openings is recommended. 9 ISO 10312:19951E) 7.4.2 Gold electron microscope grids Gold TEM grids with 200 mesh are recommended to mount TEM specimens when sodium measurements are required in the fibre identification procedure. Grids which have grid openings of uniform size such that they meet the requirement specified in 9.6.2 shaii be chosen. To facilitate the relocation of individuai grid openings for quality assurance purposes, the use of grids with numerical or alphabetical indexing of indi- vidual grid openings is recomme�ded. 7.4.3 Carbon rod electrodes Spectrochemically pure carbon rods, shall be used in the vacuum evaporator (7.3.5) during carbon coating of fiiters. 7.4.4 Routine electron microscopy tools and suppiies Fine-point tweezers, scalpel holders and blades, mi- croscope slides, double-coated adhesive tape, lens tissue, gold wire, tungsten filaments and other routine supplies are required. 7.4.5 Reference asbestos samples Asbestos samples, shall be for preparation of refer- ence TEM specimens of the primary asbestos min- erals. The UICC set of minerals is suitable for this purpose. 8 Air sample collection The desired analytical sensitivity is a parameter that shall be established for the analysis prior to sample collection. It is defined as the structure concentration corresponding to the detection of one structure in the analysis. For direct transfer methods of TEM speci- men preparation, the analytical sensitivity is a function of the volume of air sampled, the active area of the collection filter, and the area of the TEM specimen over which structures are counted. If total airborne dust levels are high, it may be necessary to terminate sampling before the required volume has been sam- pled. If this happens, the analytical sensitivity required can be achieved only by counting structures on more grid openings, or by selective concentration of asbestos structures using an indirect TEM specimen preparation technique. Select the sampling rate and the period of sampling to yield the required analytical sensitivity, as detailed in table 1. Before air samples 10 � ISO are coliected, unused filters shall be analysed as de- scribed in 9.7 to determine the mean asbestos struc- ture count for blank filters. Air samples shall be coliected using filter cassettes (7.1.1). During sampling, the cassette shall be sup- ported on a stand (7.1.3) which is isolated from the vibrations of the pump (7,1.2). The cassette shall be held facing vertically downwards at a height of ap- proximately 1,5 m to 2,0 m above ground/floor level, and shali be connected to the pump with a flexible tube. Measure the sampling flow-rate at the front end of the cassette, both at the beginning and end of the sampling period, using a calibrated variable area flowmeter (7.1.4) temporarily attached to the inlet of the cassette. The mean value of these• two mea- surements shall be used to calculate the total air vol- ume sampled. Basic strategies for monitoring environmental sources of airborne asbestos are described in annex G. After sampling, a cap shall be placed over the open end of the cassette, and the cassette packed with the filter face-upwards for return to the laboratory. Field blank filters shall also be included, as specified in 9.7, and submitted to the remaining analytical procedures along with the samples. NOTES 5 In table 1 a collection filter area of 385 mm2 is assumed, and the TEM grid openings are assumed to be 85 µm2 square. The limit of detection is defined as the upper 95 % confidence limit of the Poisson distribution for a count of 0 structures. In the absence of background, this is equal to 2,99 times the analytical sensitivity. Backgrounds that are different from 0 observed during analysis of blank filters will degrade the limit of detection. 6 The analytical sensitivity S, expressed in number of structures per litre, is caiculated using the foilowing equation: S kA rV 9 where A� is the active area, in square miliimetres, of sample collection filter; Ay is the mean area, in square millimetres, of grid openings examined; k is the number of grid openings examined; V is the volume of air sampled, in litres. - o ISO Table ISO 10312:1995(E) — Examples of the minimum number of grid openings required to achieve a particular analytical sensitivity and limit of detection Analytical Limit of Volume of air sampled (litres) sensitivity detection structures�l structures�l 500 1 000 2 000 3 000 4 000 5 000 0,1 0,30 1066 533 267 178 134 107 0,2 0,60 533 267 134 89 67 54 0,3 0,90 356 178 89 60 45 36 0,4 1,2 267 134 67 45 34 27 0,5 1,5 214 107 54 36 27 22 0,7 2,1 153 77 39 26 20 16 1,0 3,0 107 54 27 18 14 11 2,0 6,0 54 27 14 9 7 6 3,0 9,0 36 18 9 6 5 4 4,0 12 27 14 7 5 4 4 5,0 15 22 11 6 4 4 4 7,0 21 16 8 4 4 4 4 10 30 11 6 4 4 4 4 9 Procedure for analysis 9.1 General The techniques used to prepare TEM specimens are different for polycarbonate and cellulose ester filters. The preparation method to be used shali be either 9.3 or 9.4, depending on the type of inembrane filter used for air sampling. Cleaning of the sample cassettes before they are opened, preparation of the carbon evaporator, criteria for acceptable specimen grids, and the requirement for blank determinations are identical for the two preparation techniques. TEM examination, structure counting, fibre identification and reporting of results are independent of the type of filter or preparation technique used. The ability to meet the blank sample criteria is de- pendent on the cieanliness of equipment and sup- plies. Consider all supplies such as microscope slides and glassware as potential sources of asbestos con- tamination. It is necessary to wash all glassware be- fore it is used. Wash any tools or glassware which come into contact with the air sampling filters or TEM specimen preparations both before use and between handling of individual samples. Where possible, disposable supplies should be used. 9.2 Cleaning of sample cassettes Asbestos fibres can adher to the exterior surfaces of air sampling cassettes; and these fibres can be inad- vertently transferred to the sample during handiing. To prevent this possibility of contamination, and after ensuring that the cassette is tightly sealed, wipe the exterior surfaces of each sampling cassette before it is placed in the clean facility or laminar flow hood. 9.3 Direct preparation of TEM specimens from polycarbonate fitters 9.3.1 Selection of filter area for carbon coating Use a cleaned microscope slide to support represen- tative portions of polycarbonate filter during the car- bon evaporation. Double-coated adhesive tape is used to attach the filter portions to the glass slide. Take care not to stretch the polycarbonate filters during handling. Using freshly cleaned tweezers, remove the polycarbonate filter from the sampling cassette, and place it on to a second cleaned glass microscope slide which is used as a cutting surface. Using a freshly cleaned curved scalpel blade, cut the filter by rocking the blade from the point, pressing it into contact with the filter. Repeat the process as necessary. Several such portions may be mounted on the same micro- scope slide. The scalpel blade and tweezers shall be washed and dried between the handling of each filter. Identify the filter portions by writing on the glass slide. 9.3.2 Carbon coating of Fiiter portions Place the glass slide holding the filter portions on the rotation-tiiting device; approximately 10•cm to 12 cm ISO 10312:1995(E) from the evaporation source, and evacuate the evaporator chamber (7.3.5) to a vacuum better than 0,013 Pa. The evaporation of carbon shall be per- formed in very short bursts, separated by a few sec- onds to allow the electrodes to cool. If evaporation of carbon is too rapid, the strips of polycarbonate filter will begin to curl, and cross-linking of the surface will occur. This cross-linking procedures a layer of polymer which is relatively insoluble in organic solvents, and it will not be possible to prepare satisfactory TEM specimens. The thickness of carbon required is de- pendent o� the size of particles on the fiiter, and ap- proximately 30 nm to 50 nm has been found to be satisfactory. If the carbon film is too thin, large parti- cles will break out of the film during the later stages of preparation, and there will be few complete and undamaged grid openings on the specimen. Too thick a carbon film will lead to a TEM image which is lack- ing in contrast, and the ability to obtain ED patterns will be compromised. The carbon film thickness should be the minimum possible, while retaining most of the grid openings of the TEM specimen intact. 9.3.3 Preparation of the Jaffe washer Place several pieces of lens tissue, as shown in figure2, on the stainless steel bridge {7,1.3) and fill the washer (see 7.3.7) with chloroform (6.2) or 1-methyl-2-pyrrolidone (6.3) to a level where the meniscus contacts the underside of the mesh, re- sulting in saturation of the lens tissue. 9.3.4 Placing of specimens in the Jaffe washer Using a curved scalpel blade, cut three 3 mm square pieces of carbon-coated polycarbonate filter form the carbon-coated filter portion. Select three squares to represent the centre and the periphery of the active surface of the filter. Place each square of filter, carbon side up, on a TEM specimen grid, and place the grid and filter on the saturated lens tissue in the Jaffe washer. Place the three specimen grids from one sample on the same piece of lens tissue. Any number of separate pieces of lens tissue may be placed in the same Jaffe washer. Cover the Jaffe washer with the lid, and allow the washer to stand for at least 8 h. NOTE 7 It has been found that some polycarbonate fil- ters wiil not compietely dissolve in the Jaffe washer, even after exposure to chloroform for as long as 3 d. This prob- lem is more severe if the surface of the filter was over- heated during the carbon evaporation. It has been found that the problem of residual undissolved filter polymer can be overcome in several ways: a) condensation washing of the grids, using chloroform as the solvent, after the initial Jaffe washer treatment, can iF o ISO often remove much of the residua( fiiter medium in a period of approximately 30 min. To carry out this pro- cedure, transfer the piece of lens tissue supporting the specimen grids to the cold finger of the condensation washer (7.3.8�, which has achieved stable operating conditions. Operate the washer for approximately 30 min after inserting the grids; b) used in a Jaffe washer, 1-methyl-2-pyrrolidone has been found to be a more effective soivent than chloroform for polycarbonate filters. This solvent is more effective if the lens paper is not used and grids are placed di- rectly on the stainless steel mesh of the Jaffe washer. A dissolution period of 2 h to 6 h has been found to be satisfactory. After dissolution is complete, remove the stainiess steel mesh from ihe Jaffe washer and allow the grids to dry. 1-methyl-2-pyrrolidone evaporates very slowly. If it is required to dry the grids more rapidly, transfer the stainless steel bridge into anoiher Petri dish, and add water (6.1) until the meniscus contacts the underside of the mesh. After approximately 15 min, remove the mesh and allow the grids to dry. If it is desired to retain water-soluble particle species on the TEM grids, ethanol may be used instead of water (6.1) for the second wash; c) a mixture of 20 % 1,2-diaminoethane [ethylenediamine] and 80 % 1-methyl-2-pyrrolidone, used in a Jaffe washer, completely dissoives polycarbonate filters in 15 min, even if the surface oT the filter has been overheated. To use this solvent, place the grids directly on the stainless steel mesh of the Jaffe washer, do not use the lens paper. After a period of 15 min, transfer the stainless steel bridge into another Petri dish, and add water (6.1) until the meniscus contacts the underside of the mesh. After approximately 15 min, remove the mesh and allow the grids to dry. If it is desired to retain water-soluble par- ticle species on the TEM grids, ethanol may be used instead of water (6.1) for the second wash. 9.3.5 Rapid preparation of TEM specimens from PC filters TEM specimens can be prepared rapidly from PC fil- ters, if desired, by washing for approximately 1 h in a Jaffe washer, followed by washing for 30 min in a condensation washer using chloroform as the solvent. The alternative filter dissolution procedures described in note 7 may also be used. 9.4 Direct preparation of TEM specimens frorn cellulose ester filters 9.4.1 Selection of area of filter for preparation Using clean tweezers, remove the filter from the filter cassette, and place it on a cleaned microscope slide. Using a clean, curved scalpel blade, cut out a portion of the filter. o ISO 9.4.2 Preparation of solution for collapsing cellulose ester filters Mix 35 ml of dimethylformamide (6.4), and 15 ml of glaciai acetic acid (6.5) with 50 m) of water (6.1). Store this mixture in a clean bottle, The mixture is stable and suitabie for use for up to 3 months after prepara- tion. 9.4.3 Filter coliapsing procedure Using a micro�ipette with a disposabie tip (7.3.14), place 15 µl�cm to 25 µi/cm2 of the solution prepared in 9.4.2 on a cleaned microscope slide, and using the end of the pipette tip, spread the liquid over the area to 'be occupied by the filter portion. Place the filter portion, active surface upwards, on top of the sol- ution, lowering the edge of the filter at an angle of about 20° so that air bubbies are not created. Remove any solution not absorbed by the filter by allowing a paper tissue to contact the liquid at the edge of the filter. More than one filter portion may be placed on c,ne slicJe. Piace the slide either on a thermostatically controlled slide warmer (7.3.9) at a temperature of 65 °C to 70 °C, or in a� oven (7.3.9) at this temper- ature, for 10 min. The filter coliapses slowly to about 15 % of its original thickness. The procedure leaves a thin, transparent polymer film, with particles and fi- bres embedded in the upper surface. 9.4.4 Plasma etching of the filter surface The optimum conditions and time for plasma etching (see 7.3.4) have been determined experimentally from the recovery of fine chrysotile fibrils on 0,8 µm pore size MEC filters. The conditions required in a particular plasma asher shall be established using the procedure specified in annex A. Place the microscope slide holding the collapsed filter portions in the plasma ashe�, and etch for the time and under the conditions determined. Take care to ensure that the correct conditions are respected. After etching, admit air slowiy to the chamber and remove the microscope slide. Adjust the air admission valve of the plasma asher such that the time taken for the chamber to reach at- � mospheric pressure exceeds 2 min. Rapid air admis- � sion may disturb particulates on the surface of the etched filter. 9.4.5 Carbon coating Coat the microscope slide hoiding the collapsed filter portions with carbon as specified in 9.3.2. (SO 10312:1995(E) 9.4.6 Preparation of the Jaffe washer Piace several pieces of lens tissue on the stainless steel bridge, and fill the washer with dimethylformamide (6.4) or acetone (6.6) to a level where the meniscus contacts the underside of the mesh, resulting in saturation of the lens tissue. 9.4.7 Placing of specimens in the Jaffe washer Place the specimens in the Jaffe washer as specified in 9.3.4. Specimens are normally cleared after ap- proximately 4 h. 9.4.8 Rapid preparation of TEM specimens from cellulose ester fiiters An alternative washing procedure may be used to prepare TEM specimens from cellulose ester fiiiters more rapidly than can be achieved by the Jaffe washing procedure. After the specimens have been washed in a Jaffe washer for approximately 1 h, transfer the piece of lens tissue supporting the eng��monc �n thg rnir�l f�nnar nf a rnnrir�ncatin(i washer (7.3.8) operating with acetone as the solvent because dimethylformamide shall not be used in a condensation washer. Operate the condensation washer for approximately 30 min. This treatment re- moves all the remaining filter polymer. 9.5 Criteria for acceptable TEM specimen grids Valid data cannot be obtained unless the TEM speci- mens meet specified quality criteria. Examine the TEM specimen grid in the electron microscope at a sufficiently low magnification (x 300 to x 1 000) for complete grid openings to be inspected. Reject the grid if a1 the TEM specimen has not been cleared of filter medium by the filter dissolution step. If the TEM specimen exhibits areas of undissolved filter me- dium, and if at least two of the three specimen grids are not cleared, either additional washing with solvent shali be carried out, or new speci- mens shall be prepared from the filter; b) the sample is overioaded with particulate. If the specimen grid exhibits more than approximately 10 % obscuration on the majority of the grid openings, the specimen shall be designated as overloaded. This filter cannot be alanysed satis- factorily using the direct preparation methods be- cause the grid is too heavily loaded with debris to allow separate examination of individual particles by ED and EDXA, ancl� obscuration of fibres by 13 ISO 'i0312:9995(E� other particulates may lead to underestimation of the asbestos structure count; c) the particulate deposits on the specimen are not uniformly distributed from one grid opening to the next. If the particulate deposits on the specimen are obviously not uniform from one grid opening to the next, the specimen shall be designated as non-uniform. This condition is a function either of the air sampling conditions or of the fundamentai nature of the airborne particulate. Satisfactory analysis of this filter may not be possible unless a large number of grid openings is examined; d) the TEM grid is too heavily loaded with fibrous structures to make an accurate count. Accurate counts cannot be made if the grid has more than approximately 7 000 structures/mm2; or e} more than approximately 25 % of the grid openings have broken carbon film over the whole grid opening. Since the breakage of carbon film is usually more frequent in areas of heavy deposit, counting of the intact openings can lead to an underestimate of the asbestos structure count. NOTE 8 If the specimens are rejected because un- acceptable numbers of grid openings exhibit broken carbon replica, an additional carbon coating may be ap- plied to the carbon coated filter, and new specimen grids prepared. The larger particies can often be sup- ported by using a thicker carbon film. If this action does not produce acceptable specimen grids, this filter can- not be analysed using the direct preparation methods. If one or more of the conditions described in b), c>, d) or e) exists, it may not be possible to analyse the sample by this method. 9.6 Procedure for structure counting by TEM 9.6.1 General The examination consists of a count of asbestos structures which are present.on a specified number of grid openings. Fibres are classified into groups on the basis of morphological observations, ED patterns and EDXA spectra. The total number of structures to be counted depends on the statistical precision de- sired. In the absence of asbestos structures, the area of the TEM specimen grids which must be examined depends on the analytical sensitivity required. The precision of the structure count depends not only on the totai number of structures counted, but also on their uniformity from one grid opening to the next. Additional structure counting will be necessary if greater precision is required. 14 � ISO In order that the estimate of the structure density on the sampling filter shall not be based on the small area represented by one specimen grid, grid openings shall be examined on two of the three specimen grids prepared. Then combine the results in the calculation of the structure density. Structure counts shail be made at a magnification of approximately x 20 000, and shall be terminated at the end of the examination of the grid opening on which the 100th asbestos structure is observed, except that the count shali be continued until a minimum of 4 grid openings have been examined. Otherwise, the structure count shall continue to that number of grid openings at which the specified analytical sensitivity has been achieved. NOTE 9 The normal range for the number of grid openings which should be examined is from 4 to 20. If in- sufficient air has been sampled through the filter, the cal- culation in 9.6.4 can indicate that an impracticaily large number of grid openings should be examined. When this situation occurs, a larger value of analytical sensitivity may have to be accepted. 9.6.2 Measurement of inean grid opening area The mean grid opening area shall be measured for the type of TEM specimen grids in use. The standard de- viation of the mean of 10 openings selected from 10 grids should be less than 5%. As an optional proce- dure, or if the 5% standard deviation criterion cannot be demonstrated, the dimensions of each grid open- ing examined in the TEM shall be measured at a cali- brated magnification. 9.6.3 TEM alignment and calibration procedures Before structure counting is performed, align the TEM according to instrumental specifications. Calibrate the TEM and EDXA system according to the procedures specified in annex B. 9.6.4 Determination of stopping point Before structure counting is begun, calculate the area of specimen to be examined in order to achieve the selected analytical sensitivity. Calculate the maximum number of grid openings to be examined using the foilowing equation: Af k A9VS where is the number of grid openings to be ex- amined, rounded upwards to the next highest integer; o ISO Af is the area, in square miilimetres, of sam- ple filter; A9 is the area, in square millimetres, of TEM specimen grid opening; V is the volume of air sampled, in litres; is the required analytical sensitivity, ex- pressed in number of structures per litre. 9.6.5 Generai procedure for structure counting and size analysis Use at least two specimen grids prepared from the filter in the structure count. Select at random several grid openings from each grid, and combine the data in the calculation of the results. Use a form similar to that shown in figure 4 to record the data. Insert the first specimen grid into the TEM. NOTE 10 In order to facilitate quality assurance mea- surements which require re-examination of the same grid opening by differeni microscopists, the grid shouid be in- serted into the specime� holder in a standard orientation with the grid bars parallel and perpendicular to the axis of the specimen holder. This will provide scan directions par- allel to the edges of the grid opening. It should be ensured that all microscopists begin scanning at the same starting point on the grid opening, and that they use similar scan patterns. This procedure permits rapid relocation of fibrous structures for further examination if necessary. Select a typical grid opening and set the screen mag- nification to the calibrated value (approximately x 20 000?. Adjust the sample height until the features in the centre of the TEM viewing screen are at the eucentric point. Set the goniometer tilt angle to zero. In column 1 of the data recording form, record the number or letter used to identify.the grid. In column 2, record the identification of the particular grid open- ; ing. Position the specimen so that the grid opening is positioned with one corner visible on the screen. Move the imags by adjustment of only one translation control, carefully examining the sample for fibres, until the opposite side of the grid opening is encountered. Move the image by a predetermined distance less than one screen diameter, using the other transiation control, and scan the image in the reverse direction. Continue the procedure in this manner until the entire grid opening has been inspected in a pattern similar to that shown in figure 5. When a fibrous structure is detected, assign a sequential number to the primary structure in column 3, perform the identification pro- cedures required as detailed in annex E, and enter the appropriate compositiona) classification on the struc- ture counting form in column 5. Assign a ISO 10312:1995(E) morphological classification to the structure according to the procedures specified in annex D, and record this in column 6. Measure on the TEM viewing screen the length and width of the image of the primary structure, in millimetres, and record these mea- surements in columns 7 and 8. For a disperse cluster or matrix, assign a compositional classification and a morphological classification to each structure compo- nent, measure the length and width, and enter the data in columns 4 to 8. Use column 4 of the data re- cording form to tabulate the sequential number of total structures taking into account structure compo- nents, if non-asbestos fibres are observed, note their presence and type, if known. After a fibrous structure has been examined and measured, relocate the orig- inal field of view accurately before continuing scan- ning of the specimen. Failure to do this may cause structures to be overiooked or counted twice. Con- tinue the examination until the compietion of the grid opening on which the 100th asbestos structure has been recorded, or until the number of grid openings required to achieve the specified analytical sensitivity, calculated according to 9.6.4, have been examined whichever occurs first. The data shali be drawn ap- proximately equally from a minimum ot two grids. Regardless of the value calculated according to 9.6.4, fibrous structures on a minimum of four openings shalt be counted. 9.6.6 Measurement of concentration for asbestos fbres and bundles longer than 5 µm Consider improving the statisticai validity for meas- urement of asbestos fibres and bundles longer than 5 µm by additional examination at a lower magnifica- tion, taking account only of the longer fibres and bundles. Perform this extended examination for fibres and bundles longer than 5 µm in accordance with the procedures specified in annex E. Use a magnification of approximately x 10 000 for counting all asbestos fibres and bundles longer than 5 µm, or approximately x 5 000 if only fibres and bundles within the diameter range 0,2 µm to 3,0 µm are to be counted. Continue the count until completion of the grid opening on which 100 fibres and bundles have been recorded, or until a sufficient area of the specimen has been ex- amined to achieve the desired analytical sensitivity. Only those structures which are identified as, or are suspected to be, either chrysotile or one of the amphibole minerals will be reported in either the original or the extended TEM examination. Other ma- teriais, such as gypsum, cellulose fibres, and filter artifacts such as undissolved filter strands, will not be included in the fibre count. This restriction is intended to ensure that the best statistical validity is obtained for the materials of interest. 15 ISO 10312:1995�E) � ISO TEM asbestos structure count (page of ) Report number : .................................................... Sample number : .................................................... Filename : .............................................................. Sample description : .............................................. ................................................................................. ................................................................................. Preparation date : .................. By: ........................ Analysis date : ....................... By: ........................ Computer entry date: ..,,....... By : ........................ 16 Airvolume : ...................................................... litres Sample filter area: 2 ............................................ mm Magnification: ........................................................... Grid opening dimension : .................................... µm Level of analysis (C) : .................................................... (A) : ............................... - _ Grid Number of Grid structures Class Type of Length Width opening structure Comments primary total mm mm Figure 4— Example of structure counting form J � o ISO ISO 10312:1995�E) e� First pass 1'EM field of view � - - -....-�. E--- Grid opening Figure 5— Exampie of scanning procedure for TEM specimere examination 9.7 Blank and quality control determinations Before air samples are collected, a minimum ot two unused filters from each filter lot of 100 filters shall be analysed to determine the mean asbestos struc- ture count. If the mean count for all types of asbestos structures is found to be more than 10 structuresJ mmZ, or if the mean fibre count for asbestos fibres and bundles longer than 5 µm is more than 0,1 fibre/ mm2, reject the filter lot. To ensure that contamination by extraneous asbestos fibres during specimen preparation is insignificant compared with the results repo�ted on samples, es- tablish a continuous programme of blank mea- surements. At least one field blank shall be processed along with each batch of samples. In addition, at least one unused filter shall be included with every group of samples prepared on one micros�ope slide. Initially, and also at intervals afterwards, ensure that samples of known asbestos concentrations can be analysed satisfactorily. Since there is a subjective component in the structure counting procedure, it is necessary that recounts of some specimens be made by different microscopists, in order to minimize the subjective effects. Such recounts provide a means of maintaining comparability between counts made by different microscopists. Variability between and within microscopists and between laboratories shall be characterized. These quality assurance measurements shall constitute approximately 10 % of the analyses. Repeat results should not differ at the 5% signif- icance level. 17 ISO 10312:1995(E) 9.8 Calculation of results Calculate the results using the procedures detailed in annex F. Prior to the TEM examination of the speci- mens, the level of analysis was specified. Before the results are calculated, the compositional and morphological classifications to be included in the re- sult shall be specified. The chi-squared uniformity test shall be conducted using the number of primary asbestos structures found on each grid opening, prior to the application of the cluster and matrix counting criteria. The concentration result shall be calculated using the numbers of asbestos structures reported after the application of the cluster and matrix counting criteria 10 Performance characteristics 10.1 General It is important to use this analytical method in con- junction with a continuous quality control programme. The quality control programme should include use of standard samples, blank samples, and both interlabo- ratory and intralaboratory a�alyses. 10.2 Interferences and limitations of fibre iden�ification Unequivocal identification of every chrysotile fibre is not possible, due to both instrumental limitations and the nature of some of the fibres. The requirement for a calibrated ED pattern eliminates the possibility of an incorrect identification of the fibre selected. However, there is a possibility of misidentification of fibres for which both the morphologies and the ED patterns are reported on the basis of visual inspection only. The only significant possibilities of misidentification occur with halloysite, vermiculite scrolls or palygorskite, all of which can be discriminated from chrysotile by the use of EDXA and by observation of the 0,73 nm (002) reflection of chrysotile in the ED pattern. As in the case of chrysotile fibres, complete identifi- cation of every amphibole fibre is not possible due to instrumental limitations and the nature of some of the fibres. Moreover, complete identification of every amphibole fibre is not practical due to the limitations of both time and cost. Particles of a number of other minerals with compositions similar to those of some amphiboles could be erroneously classified as amphibole when the classification criteria do not in- clude zone-axis ED techniques. However, the re- quirement for quantitative EDXA measurements on all fibr�s as support for the random orientation ED technique makes misidentification very unlikely, par- 18 � ISO ticulariy when other similar fibres in the same sample have been identified as amphibole by zone-axis methods. The possibility of misidentification is further reduced with increasing aspect ratio, since it is rare for the minerais with which amphibole may be con- fused to display an asbestiform habit. 10.3 Precision and accuracy (see ISO Standard Handbook No. 3) 10.3.1 Precision The analytical precision that can be obtained is de- pendent upon the number of structures counted, and also on the uniformity of the particulate deposit on the original filter. Assuming that the structures are ran- domly deposited on the filter, if 100 structures are counted and the loading is at least 3,5 structures/grid opening, computer modelling of the counting proce- dure shows that a coefficient of variation of about 10 % can be expected. As the number of structures counted decreases, the precision will also decrease approximately as �N , where N is the number of structures counted, in practice, particulate deposits obtained by filtration of ambient air samples are rarely ideally distributed, and it is found that the precision is correspondingly reduced. Degradation of precision is a consequence of several factors, such as: a) non-uniformity of the filtered particulate deposit; b) distorsion of the fibre distribution by application of the structure counting criteria; c) variation between microscopists in their interpre- tation of the fibrous structures; d) variation between microscopists in their ability to detect and identify fibres. The 95 % confidence interval about the mean for a single structure concentration measurement using this analytical method should be aRproximately ± 25 % when 100 structures are counted over 10 grid openings. 10.3.2 Accuracy There is no independent method available to deter- mine the accuracy. NOTE 11 It has been demonstrated that, after polycarbonate membrane filters have been coated with carbon, particulate material is transferred to the TEM specimens without measurable losses. However, if the fil- ters are heavily loaded by particulate material, some of this may be lost before they are coated with carbon. Good comparability between the capillary-pore polycarbonate pro- . o ISO ISO 10312:1995(EJ cedure and the cellulose ester filter procedure has been demonstrated for laboratory-generated aerosols of chrysotile asbesios. 10.3.3 i�teriaboratory a�d intralaboratory analyses Interlaboratory and intralaboratory analyses are re- quired in order to monitor systematic errors that may develop among microscopists when using this method. These analyses should be designed to test both the overall method and the performance of indi- vidual microscopists. Repeating preparation of TEM grids from different sectors of a filter, followed by examination of the grids by a different microscopist, is a test for the reproducibility of the whole method. However, non-uniformity of the particulate deposit on the filter may lead to differences which are not related to the performance of the microscopists. Verified fibre counting (counting of asbestos structures on the same grid opening of a TEM grid by two or more op- erators, followed by resolution of any discrepancies) may be used both as a training aid and to determine the performance of different microscopists. The use of indexed TEM grids as described in 7.4.1 and 7.4.2 is recommended in order to facilitate relocation of specific grid openings. 10.4 Limit of detection The limit of detection of the method can be varied by choice of the area of the collection filter, the volume of air sampled and the area of the specimen examined in the TEM. It is also a function of the background of asbestos structures on unused filters. A limit of de- tection shall be quoted for each sample analysis. NOTE 12 In practice, the lowest limit of detection is fre- ' quently determined by the total suspended particulate con- ' centration, since each particle on the filter must be separated from adjacent ones by a distance large enough for the particle to be identified without interference. Particulate loadings on sampling filters greater than 25 µg�cm2 usually preclude preparation of TEM specimens by the direct methods. If the analysis is to be performed wiih an acceptable expenditure of time, the area of the - specimen examined in the TEM for structures of all sizes is limited in most cases to between 10 and 20 grid openings. , In typical ambient or building atmospheres, it has been found that an analytical sensitivity of 1 structure�l can be achieved. In some circumstances, where the atmosphere is ' exceptionally clean, this can be reduced to 0,1 structure/I or lower. For fibres and bundles longer than 5 µm, the re- duced magnifications specified permit larger areas of the TEM specimens to be examined with an acceptable ex- penditure of time, resulting in proportionately lower limits . of detection. If no structures are found in the analysis, the upper 95 % confidence limit can be quoted as the upper boundary of the concentration, corresponding to 2,99 times the analytical sensitiviry if a Poisson distribution of struc• tures on the fiiter is assumed. This 95 °/a confidence limil for 0 structures counted is taken as the detection limit. Since there is sometimes contamination of unused samples filters by asbestos structures, this should also be taken into account in the discussion of limits of detection. 11 Test report The test report shall inciude at least the following in- formation: a) reference to this International Standard; b) identification of the sample; c) the date and time of sampling, and ali necessary sampling data; d) the date of the analysis; e) the identity of the analyst; f) any procedure used that is not specified in this International Standard or regarded as optional; g) a complete lisiing of the structure counting data (the following data should be included: grid open- ing number, structure number, identification cate- gory, structure type, length and width of the structure in micrometres, and any comments concerning the structure); h) a statement of the minimum acceptable identifi- cation category and the maximum identification category attempted (refer to tables D.1 and D.2); i) a statement specifying which identification and structure categories have been used to calculate the concentration values; j) separate concentration values for chrysotile and amphibole structures, expressed in number of asbestos structures per litre; k) the 95 % confidence interval limits for the con- centration values, expressed in number of asbestos structures per litre; 1) the analytical sensitivity, expressed in number of asbestos structures per litre; m) the limit of detection, expressed in number of asbestos structures per litre; n) compositional data for the principal varieties of amphibole, if present; 19 IS0 10312:1995(E) 0) items g) to m) for asbestos fibres and bundles longer than 5 µm; � ISO = p) items g) to m) for PCM equivalent asbestos fibres and bundles. An example of a suitable format for the structure counting data is shown in figures 6 and 7. Sample analysis information (page 1 i Laboratory name Report number Sampie: 456 Queen Street Ashby de la Zouch Exterior sample 1991-09-09 Date Air volume: 2 150,0 litres Area of coilection filter: 385,0 mmZ Levei of analysis (chrysotile): CD or CMQ Level of analysis (amphibolei: ADQ Magnification used for fibre counting: x 20 500 Aspect ratio for fibre definition: 5�� Mean dimension of grid openings: 95,4 µm Initials of analyst: JMW Number of grid openings examined: 10 Analytica) sensitivity: 1,968 structures�l Number of primary asbestos structures: 13 Number of asbestos structures counted: 26 Number of asbestos structures > 5 µm : � Number of asbestos fibres and bundles > 5 µm : 10 Number of PCM equivalent asbestos structures: 3 Number of PCM equivalent asbestos fibres: 5 Figure 6— Example of format for reporting sample and preparation data 20 o ISO Sampie analysis information (pages 2 and following) Laboratoriy name Report number Sample: 456 Queen Street Ashby de la Zouch Exterior sample 1991-09-09 TEM asbestos structure count - Raw data Date ISO 10312:1995(Ei Number of Grid Grid �ructures Identifi- Structure Length Width Comments opening cation�► type primary total µm µm A F4-4 1 1 CD F 1,7 0,045 2 2 CMQ B 2,6 0,09 3 3 ADQ F 4,0 0,15 Crocidolite E3-6 4 4 CD MC+O 3,5 1,3 E5-1 5 CD MD43 7,5 5,0 5 CD MB 7,7 0,30 6 CNiQ MF 5,6 0,045 7 CD MB 5,1 0,30 8 CD MF 1,7 0,045 B F4-1 6 CD CD+O 6,5 3,0 9 CD CB 3,5 0,15 10 CD CF 3,5 0,045 11 CMQ CR+O 2,6 1,9 G5-1 7 CD CD31 6,1 3,2 12 CD CB 5,6 0,3 13 CMQ CF 4,0 0,045 14 CMQ CB 3,2 0,090 E4-4 8 15 CD B 1,5 0,23 9 16 AD F 8,7 0,15 C G4-4 10 CMQ CD42 25 5,6 17 CMQ CB 15 0,15 18 CMQ CF 9,4 0,045 19 ADQ CF 3,6 0,30 Tremolite 20 CM CF 4,2 0,045 E4-4 No libres E5-6 11 ADQ CD+3 9,4 2,5 21 ADQ CF 7,1 0,30 Amosite 22 ADQ CF 6,2 0,10 Crocidolite 23 CM CB 5,1 0,2 24 CM CR+O 3,3 1,8 F4-1 12 25 CMQ MC10 3,7 2,1 13 26 CD CC+O 7,4 0,5 1) Identification codes listed in tables D.1 and D.2. Figure 7- Example of format for reporting structure counting data 21 1�0 10312:1995(E) � ISO Annex A (normative) Determination of operating conditions for plasma asher A.1 General During the preparation of TEM specimens from an MEC or cellulose nitrate filter, the spongy structure of the filter is collapsed, into a thinner film of polymer by the action of a solvent. Some of the particles on the surface of the original filter become completely buried in the polymer, and the specimen preparation procedure incorporates a plasma etching step to oxidize the surface layer of the polymer. Particles buried by the filter collapsing step are then exposed so that they can become subsequently affixed to the evaporated carbon film without altering their position on the original filter. The amount of etching is critical, and individuai ashers vary in performance. Therefore, the plasma asher (7,3.4) shall be calibrated to give a known amount of etching of the surface of the col- lapsed filter. This is carried out by adjusting the radio-frequency power output and the oxygen flow- rate, and measuring the time taken to completely oxidize an uncoilapsed cellulose ester filter with 25 mm diameter of the same type and pore size as those used in the analysis. 22 A•2 Procedure Place an unused cellulose ester filter, with 25 mm di- ameter, of the same type as that being used, in the centre of a glass microscope slide. Position the slide approximately in the centre of the asher chamber. Close the chamber and evacuate to a pressure of ap- proximately 40 Pa, while admitting oxygen to the chamber at a rate of 8 ml/min to 20 ml/rnin. Adjust the tuning of the system so that the intensity of the plasma in maximized. Measure the time required for complete oxidation of the filter. Determine operating parameters which result in complete oxydation of the fiiter in a period of approximately 15 min. For etching of collapsed filters, these operating parameters shali be used for a period of 8 min. NOTE 13 Plasma oxidation at high radio-frequency pow- ers wili cause the filter to shrink and curl, foliowed by sud- den violent ignition. At lower powers, the filter will remain in position and wi�i slowly become thinner until it is nearly transparent. It is recommended that a radio-frequency power be used such that violent ignition does not occur. When multiple filters are etched, the rate of etching is re- duced, and the system should be calibrated accordingly. ��7 Annex B (normative) Calibration procedures B.1 Calibration of the TEM B.1.1 Calibration of TEM screen magnification The electron microscope should be aligned according to the specifications of the manufacturer. Initially, and at regular intervals, calibrate the magnifications used for the analysis using a diffraction grating replica (7.3.11). Adjust the specimen height to the eucentric position before carrying out the calibration. Measure the distance on the fluorescent viewing screen occu- pied by a convenient number of repeat distances of the grating image, and calculate the magnification. Always repeat the calibration after any instrumental maintenance or change of operating conditions. The magnification of the image on the viewing screen is not the same as that obtained on photographic plates or film. The ratio between these is a constant value for the particular model of TEM. 6.1.2 Calibration of ED camera constant Calibrate the camera constant of the TEM when used in ED mode. Use a specimen grid supporting a carbon film on which a thin film of gold has been evaporated or sputtered. Form an image of the gold film with the specimen adjusted to the eucentric position and se- lect ED conditions. Adjust the objective lens current to optimize the pattern obtained, and measure the di- ameters of the innermost two rings either on the flu- orescent viewing screen or on a recorded image. Calculate the radius-based camera constant, aL., for both the fluorescent screen and the photographic plate or film, using the following equation: u _ aD 2,0 h�+k2+12 where ' z is the wavelength, in nanometres, of the incident electrons; L is the camera length, in millimetres; ISO 10312:1995(E) is the unit cell dimension of gold, in nanometres (= 0,407 86 nm); D is the diameter, in millimetres, of the (hk!) diffraction ring. Using gold as the calibration material, the radius- based camera constant is given by �1, = 0,117 74D mm•nm (smallest ring) aL = 0,101 97D mm•nm (second ring) B.2 Calibration of the EDXA system Energy calibration of the EDXA system for a low en- ergy and high energy peak shall be performed regu- larly. Calibration of the intensity scale of the EDXA system permits quantitative composition data, at an accuracy of about 10 °/a of the elemental concen- tration, to be obtained from EDXA spectra of refer- ence silicate minerals involving the elements Na, Mg, AI, Si, K, Ca, Mn and Fe, and applicable certified reference materials. If quantitative determinations are required for minerals containing other elements, ref- erence standards other than those referred to below will be required. Well-characterized mineral standards permit calibration of any TEM-EDXA combination which meets the instrumental specifications of 7.3.1 and 7.3.2, so that EDXA data from different instru- ments can be compared. Reference minerals are re- quired for the calibration; the criteria for selection being that they should be silicate minerals with ma- trices as close as possible to those of the amphiboles or serpentine, and that smali individual fragments of the minerals are homogeneous in composition within a few percent. Determine the compositions of these standards by electron microprobe analysis or chemicals methods. Crush fragments of the same selected minerai stan- dards and prepare filters by dispersal of the crushed material in water and immediate filtration of the sus- pensions. Prepare TEM specimens from these filters according to the procedures specified in clause 9. These TEM specimens can then be used to calibrate any TEM-EDXA system so that comparable composi- 23 ISO 10312:19951E) tional results can be obtained from different instru- ments. NOTES 14 The microprobe analysis of the minerai standards are carried out by conventional techniques which can be found in annex J. The mineral is first embedded in a mount of poly(methyl methacrylate) or epoxy resin. The mount is then ground and polished to achieve a flat, polished surface of the mineral fragment. This surface is then analysed, using suitable reference standards, preferably oxide standards of the individual elements wherever these are available. It is necessary to take into account the water concentration in the minerals, which in the case of chrysoti�e amounts to 13 % by mass. This water content may vary due to losses in the vacuum system. 15 Aqueous suspensions of mineral standards should be filtered immediately after preparation, since aikali and alkali earth metals may be partially leached from minerals con- taining these elements. Express the results of the electron microprobe ana- lyses as atomic or mass percentage ratios relative to silicon. X-ray peak ratios of the same elements rela- tive to silicon, obtained from the EDXA system, can then be used to calculate the relationship between peak area ratio and atomic or mass percentage ratio. The technique was described by Cliff and Lorimer (see annex J, reference [8]). The X-rays generated in a thin specimen by an inci- dent electron beam have a low probability of interact- ing with the specimen. Thus, mass absorption and ftuorescence effects are negligible. In a silicate min- eral specimen containing element i, the following equation can be used to perform quantitative analyses in the TEM: C' — k. x A' _� Cs� As where C; is the concentration or atomic percentage of element i; CS; is the concentration or atomic percentage of silicon; 24 � ISO A; is the elemental integrated peak area for element i; AS; is the elemental integrated peak area for silicon; k; is the k-ratio for element i relative to sili- con. For a particular instrumental configuration and a par- ticular particle size, the value of k; is constant. To incorporate correction for the particle size effect on peak area ratios (see annex J, references [35� and [36], extend the Cliff and Lorimer technique by ob- taining separate values of the constant k; for different ranges of fibre diameter. It is recommended that 20 EDXA measurements be made for each range of fibre diameters. Suitable ranges of fibre diamete� are: < 0,25 µm; 0,25 µm to 0,5 µm; 0,5 µm to 1,0 µm; > 1,0 µm. Insert the TEM grid into the transmission electron microscope, obtain an image at the calibrated higher magnification of about x 20 000, and adjust the spec- imen height to the eucentric point. If the X-ray detec- tor is a side-entry variety, tilt the specimen towards the X-ray detector. Select an isolated fibre or particle less than 0,5 µm in width, and accumulate an EDXA spectrum using an electron probe of suitable diam- eter. When a well-defined spectrum has been ob- tained, perform a background subtraction and calculate the background-corrected peak areas for each element listed, using energy windows centred on the peaks. Calculate the ratio of the peak area for each specified element relative to the peak area for silicon, All background-subtracted peak areas used for calibration shall exceed 400 counts. Repeat this procedure for 20 particies of each mineral standard. Reject analyses of any obviously foreign particles. Calculate the arithmetic mean concentration to peak area ratio, k; (k-ratio), for each specified ele- ment of each mineral standard and for each of the fi- bre diameter ranges. Periodic routine checks shall be carried out to ensure that there has been no degra- dation of the detector performance. These k-ratios are used to calculate the elemental concentrations of un- known fibres, using the Cliff and Lorimer relationship. � o ISO ISO 10312:1995(E) Annex C (normative) Structure counting criteria � C.1 General In addition to isolated fibres, other assemblages of particles and fibres frequently occur in air samples. Groupings of asbestos fibres and particles, referred to as "asbestos structures", are defined as fibre bun- � dles, clusters and matrices. The numerical result of a ; TEM examination depends largely on whether the .; analyst assigns such an assemblage of fibres as a _ � single entity, or as the estimated number of individual fibres which form the assemblage. It is therefore im- portant that a logical system of counting criteria be defined, so that the interpretation of these complex structures is the same for all analysts, and so that the numerical result is meaningful. Imposifion of specific structure-counting criteria generaliy requires that some interpretation, partially based o� uncertain in- formation on health effects, be made of each asbestos structure found. It is not the intention of this � International Standard to make a�y interpretations based on health effects, and it is intended that a clear separation shall be made between recording of struc- .. ture counting data, and later interpretation of those data. The system of coding specified in this Interna- tional Standard permits a clear morphological de- scription of the structures to be recorded in a concise manner suitable for later interpretation, if necessary, by a range of different criteria, without the necessity for re-examination of the specimens. In particular, the coding system is designed to permit the dimensions of each complex fibrous structure, and also whether these structures contain fibres longer than 5 µm, to be recorded. This approach permits later evaluations of the data to include considerations of particle respirability and comparisons with historical indices of asbestos exposure. Examples of the various types of morphological structure, and the manner in which these shall be recorded, are shown in figure C.1. _ C.2 Structure definitions and treatment Each fibrous structure that is a separate entity shall be designated as a primary structure. Each primary structure shall be designated as a fibre, bundle, clus- ter or matrix. C.2.1 Fibre Any particle with parallel or stepped sides, of mini- mum length 0,5 µm, and with an aspect ratio of 5J1 or greater, shall be defined as a fibre. For chrysotile asbestos, the single fibril shall be defined as a fibre. A fibre with stepped sides shall be assigned a width equal to the average of the minimum and maximum widths. This average shali be used as the width in determination of the aspect ratio. C.2.2 Bundle A grouping composed of apparently attached parallei fibres shall be defined as a bundle, with a width equal to an estimate of the mean bundle width, and a length equal to the maximum length of the structure. The overall aspect ratio of the bundle may have any value, provided that it contains individual constituent fibres with aspect ratios equal to or greater than 5/1. Bun- dles may exhibit diverging fibres at one or both ends. C.2.3 Cluster An aggregate of two or more randomly oriented fi- bres, with or without bundles, shall be defined as a cluster. Clusters occur as two varieties. C.2.3.1 disperse cluster (type D): Disperse and open network, in which at least one of the individual fibres or bundles can be separately identified and its dimensions measured; C.2.3.2 compact cluster (type C1: Complex and tightly bound'network, in which one or both ends of each individual fibre or bundle is (are) obscured, such that the dimensions of individual fibres and bundles cannot be unambiguously determined. In practice, clusters can occur in which the character- istics of both types of cluster occur in the same structure. Where this occurs, the structure should be defined as a disperse cluster, and then a logical pro- cedure should be followed by recording structure components according to the counting criteria. The procedure for treatment of clusters is illustrated by examples in figure C.2. 25 ISO 10312:1995�E) 26 a) Disperoe clueter (type D) c► Disperse matrix (type D) �. � Fibres Bundles .,a. �' ' -► ��• ��..� .� � �,�:; � ' � �.l' . b) Compact cluster Itype C) d� Compact matrix (type C) Figure C.1 — Fundamental morphological structure types e o ISO 5µm 5µm I � 5µm 5µm ISO 10312:1995(E) Count as 1 compact cluster containing more than 9 fi- bres (all fibres shorter than 5 µm) Record as CC+O Count as 1 disperse cluster consisting of 5 fibres, 4 of which are longer than 5 µm Record as CD54, followed by 5 fibres, each recorded as CF Count as 1 disperse cluster consisting of 4 fibres, 2 of which are longer than 5 µm, and 2 cluster residuals, each containing more than 9 fibres Record as CD+2, followed by 4 fibres, each recorded as CF, and 2 cluster residual, each recorded as CR+O Count as 1 disperse cluster consisting of 3 fibres, 2 bundles, 1 of which is longer than 5 µm, and 1 cluster residual containing more than 9 fibres Record as CD+1, followed by 3 fibres, each recorded as CF, 2 bundles, each recorded as CB, and 1 cluster residual recorded as CR+O Figure C.2 — Examples of recording of complex asbestos clusters 27 130 10312:1995 jE) C.2.4 Matrix One or more fibres, or fibre bundles, may be attached to, or partially concealed by, a single particle or group of overlapping nonfibrous particles. This structure shall be defined as a matrix. The TEM image does not discriminate between particles which are attached to fibres, and those which have by chance overlapped in the TEM image. It is not known, therefore, whether such a structure is actually a complex particle, or whether it has arisen by a simple overlapping of par- ticles and fibres on the filter. Since a matrix structure may involve more than one fibre, it is important to define in detail how matrices shail be counted. Matrices exhibit different character- istics, and two types can be defined. C.2.4.1 disperse matrix (type D►: Structure con- sisting of a particle or linked group of particles, with overlapping or attached fibres or bundles in which at least one of the individual fibres or bundles can be separately identified and its dimensions measured. C.2.4.2 compact matrix (type C): Structure con- sisting of a particle or linked group of particles, in which fibres or bundies can be seen either within the structure or projecting from it, such that the dimen- sions of individual fibres and bundles cannot be un- ambiguously determined. In practice, matrices can occur in which the charac- teristics of both types of matrix occur in the same structure. Where this occurs, the structure should be assigned as a disperse matrix, and then a logical pro- cedure should be followed by recording structure components according to the counting criteria. Exam- ples of the procedure which shall be followed are shown in figure C.3. C.2.5 Asbestos structure larger than 5 µm Any fibre, bundle, cluster or matrix for which the largest dimension exceeds 5 µm. Asbestos structures larger than 5 µm do not necessarily contain asbestos fibres or bundles longer than 5 µm. 28 � ISO � � C.2.6 Asbestos fibre or bundle longer than 5 µm An asbestos fibre of any width, or bundle of such fi- bres, which has a length exceeding 5 µm. C.2.7 PCM equivalent structure Any fibre, bundle, cluster or matrix with an aspect ra- tio of 3/1 or greater, longer than 5 µm, and which has a diameter between 0,2 µm and 3,0 µm. PCM equiv- alent structures do not necessarily contain fibres or bundles longer than 5 µm, or PCM equivalent fibres. C.2.8 PCM equivalent fibre Any particle with parallel or stepped sides, with an aspect ratio of 3/1 or greater, longer than 5 µm, and which has a diameter between 0,2 µm and 3,0 µm. For chrysotile, PCM equivalent fibres will always be bundles. C.3 Other structure counting criteria C.3.1 Structures which intersect grid bars A structure which intersects a grid bar shall only be counted on two sides of the grid opening, as illus- trated in figure C.4. Record the dimensions of the structure such that the obscured portions of compo- nents are taken to be equivalent to the unobscured portions, as shown by the broken lines in figure C.4. For example, the length of a fibre intersecting a grid bar is taken to be twice the unobscured length. Structures, intersecting either of the other two sides shall not be included in the count. C.3.2 Fibres which extend outside the field of view During scanning of a grid opening, count fibres which extend outside the field of view systematically, so as to avoid double-counting. In general, a rule should be established so that fibres extending outside the field of view in only two quadrants are counted. The pro- cedure is illustrated by figure C.S. Measure the length of each of these fibre by moving the specimen to lo- cate the other end of the fibre, and then return to the original field of view before continuing to scan the specimen. Fibres without terminations within the field of view shall not be counted. � o ISO I I 5µm � I I 5µm L � 5µm ISO 10312:19951E) Count as 1 compact matrix, with all fibres shorter than 5 µm Record as MC+O Count as 1 disperse matrix consisting of 1 fibre shorter than 5 µm Record as MD10, followed by 1 fibre recorded as MF Count as 1 disperse matrix consisting of 5 fibres, all longer than 5 µm Record as MD55, followed by 5 fibres, each recorded as MF Count as 1 disperse matrix, consisting if 3 fibres, 1 of which is longer than 5 µm, and 1 matrix residuai con- taining 3 fibres Record as MD61, followed by 3 fibres, each recorded as MF, and 1 matrix residual recorded as MR30 ( � 5µm Figure C.3 — Examples of recording of complex asbestos matrices 29 ISO 10312:1995(E) Scan direction ; � .�� -• • . :. :::� Grid opening " • • `.�„_i Figure C.4 — Example of counting of structures which intersect grid bars Scan direction Count OIIIIII�'IIiND 4—TEM field of view Do not count � I�— Grid openi�g, Do not count I � ISO p `�!l11i�Y' ; i Figure C.5 — Example of counting of fibres which extend outside the field of view 30 , o ISO (SO 10312:1995(E) C.4 Procedure for data recording C.4.1 Generai The morphological codes specified are designed to facilitate computer data processing, and to allow re- cording of a complete representation of the important features of each asbestos structure. The procedure requires that the microscopist classify each primary fibrous structure into one of the four fundamental categories: fibres, bundles, clusters and matrices. C.4.2 Fibres On the structure counting form, a fibre as defined in C.2.1 shail be recorded by the designation "F". If the fibre is a separately-counted part of a cluster or ma- trix, the fibre shall be recorded by the designation "CF", or "MF", depending on whether it is a compo- nent of a cluster or matrix. C.4.3 Bundles On the structure counting form, a bundle as defined in C.2.2 shall be recorded by the designation "B". If the bundle is a separately-counted part of a cluster or matrix, the bundle shall be recorded by the desig- nation "CB", or "MB", depending on whether it is a component of a cluster or matrix. C.4.4 Disperse clusiers (type D) On the structure counting form, an isolated cluster of type D as defined in C.2.3 shall be recorded by the designation "CD", followed by a two-digit number. The first digit represents the analyst's estimate of the total number of fibres and bundles comprising the structure. The digit shall be from 1 to 9, or designated as "+" if there are estimated to be more than 9 component fibres or bundles. The second digit shall � represent, in the same manner, the total number of fibres and bundles longer than 5 µm contained in the structure. The overall dimensions of the cluster, in two perpendicular directions representing the maxi- mum dimensions, shall be recorded: In order of de- creasing length, up to 5 component fibres or bundles shall be separately recorded, using the codes "CF" (cluster fibre) and "CB" (cluster bundle). If, after ac- counting for prominent component fibres and bun- dles, a group of clustered fibres remains, this shall be recorded by the designation "CR" (cluster residuall. ' If the remaining clustered fibres are present as more than one localized group, it may be necessary to re- cord more than one cluster residual. Do not record more than 5 cluster residuals for any cluster. A cluster residual shall be measured and assigned a two-digit number, derived in the same manner as specified for the overall cluster. Optionaliy, if the number of com- ponent fibres and bundles in either the original cluster or the cluster residual is outside the range 1— 9, ad- ditionai information concerning the number of com- ponent fibres and bundles may be noted in the "comments" column. C.4.5 Compact clusters (type C) On the structure counting form, an isolated cluster of type C as defined in C.2.3 shall be recorded by the designation "CC", followed by a two-digit number. The two-digit number describing the numbers of component fibres and bundles shall be assigned in the same manner as for clusters of type D. The overall dimensions of the cluster in two perpendicular di- rections shall be recorded in the same manner as for clusters of type D. By definition, the constitutent fi- bres and bundles of compact clustsrs cannot be sep- arately measured; therefore, no separate tabulation of component fibres or bundles can be made. C.4.6 Disperse matrices (type D) , On the structure counting form, an isolated matrix of type D as defined in C.2.4 shall be recorded by the designation "MD", foliowed by a two-digit number. The two-digit number shail be assigned in the same manner as for clusters of type D. The overall dimen- sions of the matrix in two perpendicular directions shall be recorded in the same manner as for clusters of type D. In order of decreasing length, up to 5 component fibres or bundles shall be separately re- corded, using the codes "MF" (matrix fibre) and "MB" (matrix bundle). If after accounting for promi- nent component fibres and bundles, matrix material containing asbestos fibres remains, this shall be re- corded by the designation "MR" (matrix residuaq. If the remaining matrix fibres are present as more than one localized group, it may be necessary to record more than one matrix residual. Do nat record more than 5 matrix residuals for any matrix. A matrix re- sidual shall be measured and assigned a two-digit number, derived in the same manner as specified for the overail matrix. Optionally, if the nurnber of com- ponent fibres or bundies in either the original matrix or the matrix residual is outside the range 1— 9, ad- ditional information concerning the number of com- ponent fibres and bundles may be noted in the "comments" column. C.4J Compact matrices (type C) On the structure counting form, an isolated matrix of type C as defined in C.2.4 shali be recorded by the 31 I�O 10312:19951E) designation "MC", followed by a two-digit number. The two-digit number shali be assigned in the same manner as for clusters of type D. The overall dimen- sions of the matrix in two perpendicular directions shall be recorded in the same manner as for clusters of type D. By definition, the constitutent fibres and bundles of compact matrices cannot be separately measured; therefore, no separate tabulation of com- ponent fibres or bundles can be made. C.4.8 Procedure for recording of partially obscured fibres and bundles The proportion of the length of a fibre or bundle that is obscured by other particulates shall be used as the basis for determining whether a fibre or bundle is to be recorded as a separate component or is to be considered as a part of a matrix of type C or part of a matrix residual. If the obscured length couid not possibly be more than one-third of the total length, the fibre or bundle shall be considered a prominent feature to be separately recorded. The assigned length for each such partially obscured fibre or bundle shall be equal to the visible length plus the maximum possible contribution from the obscured portion. Fi- bres or bundles which appear to cross the matrix, and for which both ends can be located approximately, shall be included in the maximum of 5 and recorded according to the counting criteria as separate fibres or bundles. If the obscured length could be more than one third of the total length, the fibre or bundle shall 32 � ISO '� be considered as a part of a compact matrix of type C or part of a matrix residual. C.5 Special considerations for counting of PCM equivalent structures Use 3/1 as the minimum aspect ratio for counting of PCM equivalent structures. This aspect ratio definition is required in order to achieve comparability of the results for this size range of structure with historical optical measurements, but use of this aspect ratio definition does not significantly affect the ability to interpret the whole fibre size distribution in terms of a minimum 5/1 aspect ratio. Some applications may require that a count be made of PCM equivalent structures only. The coding system permits discrimi- nation between PCM equivalent structures that con- tain fibres and bundles longer than 5 µm and those that do not. NOTE 16 In general, clusters and matrices will yield fewer components as the minimum dimensions specified for countable fibres are increased. Thus, it may be found that a particular structure yields a higher number of com- ponent fibres and bundles in a count for all fibre sizes than it does at a reduced magnification when only fibres and bundles longer than 5 µm are being counted. However, the requirement that component fibres and bundles be recorded in decreasing length order ensures that the data are con- sistent for a particular structure, regardless of the size cat- egory of fibres being counted and the magnification in use. a o tS0 Annex D (normative? Fibre identification procedure D.1 General The criteria used for identification of asbestos fibres .. may be selected depending on the intended use of the measurements. In some circumstances, there can be a requirement that fibres shall be unequivocally identified as a specific mineral species. In other cir- cumstances, there can be sufficient knowledge about the sample, so that rigorous identification of each fi- � bre need not be carried out. The time required to perform the analysis, and therefore the cost of analy- sis, can vary widely depending on the identification criteria considered which are to be sufficiently defini- tive. The combination of criteria considered definitive for identification of fibres in a particular analysis shal► be specified before the analysis is made, and this combination of criteria shall be referred to as the "level" of analysis. Various factors related to instru- mental limitations and the character of the sample may prevent satisfaction of all of the specified fibre ; identification criteria for a particular fibre. Therefore, a record shall be made of the identification criteria ` which were satisfied for each suspected asbestos fi- bre included in the analysis. For example, if both ED and EDXA were specified to be attempted for defini- , tive identification of each fibre, fibres with chrysotile morphology which, for some reason, do not give an ED pattern but which do yield an EDXA spectrum corresponding to chrysotile, are categorized in a way - which conveys the level of confidence to be placed in the identification. D.2 ED and EDXA techniques D.2.1 General Initially, fibres are classified into two categories on the basis of morphology: those fibres with tubular morphology, and those fibres without tubular morphology. Further analysis of each fibre is con- ducted using ED and EDXA methods. The following procedures shouid be used when fibres are examined by ED and EDXA. The crystal structures of some mineral fibres, such as chrysotile, are easily damaged by the high current densities required for EDXA examination. Therefore, ISO 10312:1995(E) investigation of these sensitive fibres by ED should be completed before attempts are made to obtain EDXA spectra from the fibres. When more stable fi- bres, such as the amphiboles, are examined, EDXA and ED may be used in either order. D.2.2 ED techniques The ED technique can be either qualitative or quanti- tative. Qualitative ED consists of visual examination, without detaifed measurement, of the general char- acteristics of the ED pattern obtained on the TEM viewing screen from a randomly oriented fibre. ED patterns obtained from fibres with cylindrical symme- try, such as chrysotile, do not change when the fibres are tilted about their axes, and patterns from randomly oriented fibres of these minerals can be interpreted quantitatively. For fibres which do not have cylindrical symmetry, only those ED patterns obtained when the fibre is oriented with a principal crystallographic axis closely parallel with the incident electron beam direc- tion can be interpreted quantitatively. This type of ED pattern shall be referred to as a"zone-axis ED pattern". (n order to interpret a zone-axis ED pattern quantitatively, it shall be recorded photographically and its consistency with known mineral structures shall be checked. A computer program may be used to compare measurements of the zone-axis ED pat- tern with corresponding data calculated from known mineral structures. The zone-axis ED pattern obtained by examination of a fibre in a particular orientation can be insufficiently specific to permit unequivocal iden- tification of the mineral fibre, but is is often possibie to tilt the fibre to another angle and to record a dif- ferent ED pattern corresponding to another zone-axis. The angle between the two zone-axes can also be checked for consistency with the structure of a sus- pected mineral. For visual examination of the ED pattern, the camera length of the TEM should be set to a low value of approximately 250 mm and the ED pattern should then be viewed through the binoculars. This proce- dure minimizes the possible degradation of the fibre by the electron irradiation. However, the pattern is distorted by the tilt angle of the viewing screen. A camera length of at least 2 m shouid be used when 33 (SO 10312:1995(E) � ISO . the ED pattern is recorded, if accurate measurement of the pattern is to be possible. It is necessary that, when obtaining an ED pattern to be evaluated visually or to be recorded, the sample height shall be properly adjusted to the eucentric point and the image shali be focussed in the plane of the selected area aperture. If this is not done, there may be some components of the ED pattern which do not originate from the selected area. In general, it will be necessary to use the smallest available ED aperture. For accurate measurements of the ED pattern, an internal calibration standard shali be used. A thin coating of gold, or another suitable calibration mate- riai, shall be applied to. the underside of the TEM specimen. This coating may be applied either by vac- uum evaporation or, more convenientiy, by sputtering. The polycrystalline gold film yields diffraction rings on every ED pattern and these rings provide the required calibration information. To form an ED pattern, move the image of the fibre to the centre of the viewing screen, adjust the height of the specimen to the eucentric position, and insert a suitable selected area aperture into the electron beam so that the fibre, or a portion of it, occupies a large proportion of the illuminated area. The size of the aperture and the portion of the fibre shall be such that particles other than the one to be examined are excluded from the selected area. Observe the ED pattern through the binoculars. During the observa- tion, the objective lens current should be adjusted to the point where the most complete ED pattern is ob- tained. If an incomplete ED pattern is still obtained, move the particie around within the selected area to attempt to optirnize the ED pattern, or to eliminate possible interferences from neighbouring particles. If a zone-axis ED analysis is to be attempted on the fibre, the sample shall be mounted in the appropriate holder. The most convenient holder allows complete rotation of the specimen grid and tilting of the grid about a single axis. Rotate the sample until the fibre image indicates that the fibre is oriented with its length coincident with the tilt axis of the goniometer, and adjust the sample height until the fibre is at the eucentric position. Tilt the fibre until an ED appears which is a symmetrical, two dimensional array of spots. The recognition of zone-axis alignment condi- tions requires some experience on the part of the operator. During tilting of the fibre to obtain zone-axis conditions, the manner in which the intensities of the spots vary shouid be observed. if weak reflections 34 occur at some points on a matrix of strong reflections, the possibility of twinning or multiple diffraction ex- ists, and some caution should be exercised in the se- lection of diffraction spots for measurement and interpretation. A full discussion of electron diffraction and multiple diffraction can be found in the references by J.A. Gard C»7 P.B. Hirsch et al C14J and H.R. Wenck C42J included in annex J. Not ali zone-axis patterns which can be obtained are definitive. Only those which have closely spaced reflections corresponding to low indices in at least one direction should be re- corded. Patterns in which all d-spacings are less than about 0,3 nm are not definitive. A useful guideline is that the lowest angle reflections should be within the radius of the first gold diffraction ring (111), and that patterns with smaller distances between reflections are usually the most definitive. . Five spots, closest to the centre spot, along two intersecting lines of the zone-axis pattern shall be se- lected for measurement, as shown in figure D.1. The distances of these spots from the centre spot and the four angles shown provide the required data for anal- ysis. Since the centre spot is usually very overex- posed, it does not provide a well-defined origin for these measurements. The required distances shall therefore be obtained by measuring between pairs of spots symmetricaily disposed about the centre spot, preferably separatecl by several repeat distances. The distances shail be measured with a precision of better than 0,3 mm, and the angles to a precision of better than 2,5°. The diameter of the first or second ring of the calibration pattern (111 and 200) shall aiso be measured with a precision of better than 0,3 mm. Using gold as the calibration material, the radius- based camera constant is given by �L. = 0,117 74D mm•nm (first ring) �1, = 0,101 97D mm•nm (second ring) D.2.3 EDXA measurements Interpretation of the EDXA spectrum may be either qualitative or quantitative. For qualitative interpretation of a spectrum, the X-ray peaks originating from the elements in the fibre are recorded. For quantitative interpretation, the net peak areas, after background subtraction, are obtained for the X-ray peaks originat- ing from the elements in the fibre. This method pro- vides quantitative interpretation for those minerals which contain silicon. � y Q �JD t � Spol � • Spo � , Spot 5 � Snot 2 Spot 1 � 1 ISO 10312:1995(E) Figure D.1 — Example of ineasurement of zone-axis SAED patterns To obtain an EDXA spectrum, move the image of the fibre to the centre of the screen and remove the ob- jective aperture. Select an appropriate electron beam diameter and deflect the beam so that it impinges on the fibre. Dependi�g on the instrumentation, it may be necessary to tilt the specimen towards the X-ray detector and, in some instruments, to use the Scan- ning Transmission Electron Microscopy (STEM) mode of operation. The time for acquisition of a suitable spectrum varies with the fibre diameter, and also with instrumental factors. For quantitative interpretation, spectra should have a statistically valid number of counts in each peak. Analyses of smali diameter fibres which contain sodium are the most critical, since it is in the low en- ergy range that the X-ray detector is least sensitive. Consequently, it is necessary to acquire a spectrum for a period that is sufficiently long for the sodium to be detected in such fibres. It has been found that satisfactory quantitative an analyses can be obtained if acquisition is continued until the background sub- tracted silicon Ka peak integral exceeds 10 000 counts. The spectrum should then be manipulated to subtract the background and to obtain the net areas of the elemental peaks. After quantitative EDXA classification of some fibres by computer analysis of the net peak areas, it may be possible to classify further fibres in the same sample on the basis of comparison of spectra at the intrument. Frequently, visual comparisons can be made after somewhat shorter acquisition times. D.3 interpretation of fibre analysis data D.3.1 Chrysotile The morphological structure of chrysotile is charac- teristic, and with experience, can be recognized read- ily. However, a few holder minerals have a similar appearance, and morphological observation by itself is i�adequate for most samples. The ED pattern ob- tained from chrysotile is quite specific for this mineral if the specified characteristics of the pattern corre- spond to those from reference chrysotile. However, depending on the past history of the fibre, and on a number of other factors, the crystal structure of a particular fibre may be damaged, and it may not yieid and ED pattern. In this case, the EDXA spectrum may be the only data available to supplement the morphological observations. D.3.2 Amphibo{es Since the fibre identification procedure for asbestos fibres other than chrysotile can be involved and time- consuming, computer programmes, such as that de- veloped by B.L. Rhoades (see annex J, reference 35 ISO 10312:1995(E� [32]), are recommended for interpretation of zone- axis ED patterns. The published literature contains composition and crystailographic data for all of the fibrous minerais likely to be encountered in TEM analysis of air samples, and the compositional and struciural data from the unknown fibre should be compared with the published data. Demonstration that the measurements are consistent with the data for a particular test mineral does not uniquely identify the unknown, since the possibility exists that data from other minerals may also be consistent. It is, however, unlikely that a mineral of another structural class could yield data consistent with that from an arnphibole fibre identified by quantitative EDXA and two zone-axis ED patterns. Suspected amphibole fibres should be classified ini- tially on the basis of chemical composition. Either qualitative or quantitative EDXA information may be used as the basis for this classification. From the pubiished data on mineral compositions, a list of min- erals which are consistent in composition with that measured for the unknown fibre should be compiled. To proceed further, it is necessary to obtain the first zone-axis ED pattern, according to D,2.2. �t is possible to specify a particular zone-axis pattern for identification of amphibole, since a few patterns are often considered to be characteristic. Unfortu- nately, for a fibre with random orientation on a TEM grid, no specimen hoider and goniometer currently available will permit convenient and rapid location of two preselected zone-axes. The most practical ap- proach has been adopted, which is to accept those low index patterns which are easily obtained, and then to test their consistency with the structures of the minerals already preselected on the basis of the EDXA data. Even the structures of non-amphibole minerals in this preselected list shall be tested against the zone-axis data obtained for the unknown fibre, since non-amphibole minerals in some orientations may yield similar patterns consistent with amphibole structures. The zone-axis ED interpretation shall include all min- erals previously selected from the mineral data file as being chemically compatible with the EDXA data. This procedure will usually shorten the list of minerals for which solutions have been found. A second set of zone-axis data from another pattern obtained on the � ISO same fibre can then be processed, either as further confirmation of the identification, or to attempt elim- ination of an ambiguity. In addition, the angle meas- ured between the orientations of the two zone-axes can be checked for consistency with the structures of the minerals. Caution should be exercised in ratio- nalizing the inter-zone-axis angle, since if the fibre contains c-axis twinning, the two zone-axis ED pat- terns may originate from the separate twin crystals. In practice, the full identification procedure will normally be applied to very few fibres, unless precise identification of all fibres is required for a particular reason. D.4 Fibre classification categories It is not always possible to proceed to� a definitive identification of a fibre; this may be due to instru- menta� limitations or to the actual nature of the fibre. In many analyses, a definitive identification of each fibre may not actually be necessary if there is other knowledge available about the sample, or if the con- centration is below a levei of interest. The analytical procedure shall therefore take into account both in- strumental limitations and varied analytical require- ments. Accordingly, a system for fibre classification is used to permit accurate recording of data. The classi- fications are shown in tables D.1 and D.2, and are di- rected towards identification of chrysotile and amphibole respectively. Fibres shall be reported in these categories. The general principle to be followed in this analytical procedure is first to define the most specific fibre classification which is to be attempted, or the "level" of analysis to be conducted. Then, for each fibre examined, record the classification which is ac- tually achieved. Depending on the intended use of the resuits, criteria for acceptance of fibres as "identified" can then be established at any time after completion of the analysis. In an unknown sample, chrysotile will be regarded as confirmed only if a recorded, calibrated ED pattern from one fibre in the CD categories is obtained, or if measurements of the ED pattern are recorded at the instrument. Amphibole will be regarded as confirmed only by obtaining recorded data which indicates ex- clusively the presence of amphiboles for fibres clas- sified in the AZQ, AZZ or AZZQ categories. 36 : 1: o ISO Cetegory TM CM CD CQ CMQ con NAM Category UF AD AX ADX AQ AZ ADQ AZQ AZZ AZZQ NAM ISO 10312:1995(E) Tabie D.1 — Classification of fibres with tubular morphology Oescription Tubular Morphology, not sufficiently characteristic for classification as chrysotile Characteristic Chrysotile Morphology Chrysotile SAED pattern Chrysotile composition by Quantitative EDXA Chrysotile Morphology and composition by Quantitative EDXA Chrysotile SAED pattern and composition by Quantitative EDXA Non-Asbestos Mineral Tabie D.2 — Classification of fibres without tubular morphology Description Unidentified Fibre Amphibole by random orientatio� SAED (shows layer pattern of 0,53 nm spacing) Amphibole by qualitative EDXA. Spectrum has elemental components consistent with amphibole Amphibole by random orientation SAED and qualitative EDXA Amphibole by Quantitative EDXA Amphibole by one Zon�axis SAED pattern Amphibole by random orientation SAED and Quaniitative EDXA Amphibole by one Zone-axis SAED pattern and Quantitative EDXA Amphibole by two Zone-axis SAED patterns, with consistent interaxial angle Amphibole by two Zone-axis SAED patterns, with consistent interaxial angle, and Quantitative EDXA Non-Asbestos Mineral D.4.1 Procedure for classification of fibres with tubular morphology suspected to be chrysotile Occasionally, fibres are encountered which have tu- bular morphology similar to that of chrysotile, but which cannot be characterized further either by ED or EDXA. They may be non-crystalline, in which case ED techniques are not useful, or they may be in a position on the grid which does not permit an EDXA spectrum to be obtained. Alternatively, the fibre may be of organic origin, but the morphology and compo- sition may not be sufficiently definitive enough to be disregarded. Accordingly, there is a requirement to record each fibre, and to specify how confidently each fibre can be identified. Classification of fibres wili meet with various degrees of success. Figure D.2 shows the classification procedure to be used for fi- bres which dispiay any tubular morphology. The chart is self expianatory, and every fibre is either rejected as a non-asbestos mineral (NAM), or classified in some way which by some later criterion could still contribute to the chrysotile fibre count. Morphology is the first consideration, and if this is not similar to that usualiy seen in chrysotile standard samples, designate the initial classification as TM. Regardiess of the doubtful morphology, examine the fibre by ED and EDXA methods according to figure D.2. Where the morphology is more definitive, it may be possible to ciassify the fibre as having chrysotile morphology (CM}. 37 ISO 10312:19951E) FIBRE WITH TUBULAR MORPHOLOGY ( Is �bre morphology characteristic of that displayed by reference chrysotile� NO Examine by SAED Pattern not � Chrysotile chrysotile pattern Pattern not present or indistinct TM Examine by quantitative EDXA Composition not that Chrysotile of chrysotile composition No spectrum ] TM YES Examine by SAED Chrysotile Pattern not pattern chrysotile Pattern not present or indistinct CM Examine by quantitative EDXA ( Chrysotile Composition not that composition of chrysotile No spectrum CM � Examine by quantitative EDXA � Composition not that Chrysotile of chrysotile composition No spectrum NAM CD CDQ Figure D.2 — Classification chart for fibre with tubular morphology 38 � � ISO o ISO � For classification as CM, the morphological character- istics required are the following: a1 the individual fibrils should have high aspect ratios exceeding 5/1, and be about 30 nm to 40 nm in diameter; b) the electron scattering power of the fibre at 60 kV to 100 kV accelerating potentiai should be sufficiently low for the internal structure to be visible; c) there should be some evidence of an internal structure suggesting a tubular appearance similar to that shown by reference UICC chrysotile, which may degrade in the electron beam. Examine every fibre having these morphological char- acteristics by the ED technique, and classify as chrysotile by ED (CD) only those which give diffraction patterns with the precise characteristics shown in figure D.3. The relevant features in this pattern for identification of chrysotile are as foliows: a1 the (002} reflections should be examined to de- termine that they correspond closely to a spacing of 0,73 nm; b) the layer line repeat distance should correspond to 0,53 nm; c) there should be "streaking" of the (110) and (130) reflections. Figure D.3 — Chrysotile SAED pattern ISO 10312:1995(Ej Using the millimetre calibrations on the TEM viewing screen, these observations can readily be made at the instrument. If documentary proof of fibre identification is required, record a TEM micrograph of at least one representative fibre, and record its ED pattern on a separate film or plate. This film or plate shall also carry calibration rings from a known polycrystalline sub- stance such as gold. This calibrated pattern is the only documentary proof that the particular fibre is chrysotile, and not some other tubular or scrolled species such as halloysite, palygorskite, talc or vermiculite. The proportion of fibres which can be successfully identified as chrysotile by ED is variable, and to some extent dependent on both the instru- ment and the procedures of the operator. The fibres that fail to yield an identifiable ED pattern will remain in the TM or CM categories unless they are examined by EDXA. In the EDXA analysis of chrysotile there are only two elements which are relevant. For fibre classification, the EDXA analysis shall be quantitative. If the spec- trum displays prominent peaks from magnesium and silicon, with their areas in the appropriate ratio, and with only minor peaks from other elements, classify the fibre as chrysotile by quantitative EDXA, in the categories CQ, CMQ, or CDQ, as appropriate. D.4.2 Procedure for classification of fibres without tubular morphology, suspected to be amphibole Every particle without tubular morphology and which is not obviously of biological origin, with an aspect ratio of 5/1 or greater, and having parallel or stepped sides, shall be considered as a suspected amphibole fibre. Further examination of the fibre by ED and EDXA techniques will meet with a variable degree of success, depending on the nature of the fibre and on a number of instrumental limitations. It will not be possible to identify every fibre compietely, even if time and cost are of no concern. Moreover, confir- mation of the presence of amphibole can be achieved only by quantitative interpretation of zone-axis ED patterns, a very time-consuming procedure. Accord- ingly, for routine samples from unknown sources, this analytical procedure limits the requirement for zone- axis ED work to a minimum of one fibre represen- tative of each compositionai ciass reported. In some samples, it may be necessary to identify more fibres by the zone-axis technique. When analysing samples from well-characterized sources, the cost of identifi- cation by zone-axis methods may not be justified. The 0,53 nm layer spacing of the random orientation E� pattern is not by itself diagnostic for amphibole, However, the presence of c-axis twinning in many fi- 39 1�0 10312:1995(E) bres leads to contributions to the layers in the pat- terns by several individual parallel crystais of different axial orientations. This apparently random positioning of the spots along the (ayer lines, if also associated with a high fibre aspect ratio, is a characteristic of amphibole asbestos, and thus has some limited diag- nostic value. If a pattern of this type is not obtained, the identity of the fibre is still ambiguous, since the absence of a recognizabie pattern may be a conse- quence of an unsuitable orientation relative to the electron beam, or the fibre may be some other min- eral species. Figure D.4 shows the fibre classification chart to be used for suspected amphibole fibres. This chart shows all the classification paths possible in analysis of a suspected amphibole fibre, when examined sys- tematically by ED and EDXA. Two routes are possible, depending on whether an attempt to obtain an EDXA spectrum or a random orientation ED pattern is made first. The normal procedure for analysis of a sample of unknown origin will be to examine the fibre by random orientation ED, qualitative EDXA, quantitative EDXA, and zone-axis ED, in this sequence. The final fibre classification assigned will be defined either by successful analysis at the maximum required level, 40 � ISO '' or by the instrumental limitations. Any instrumental limitations which affect the quality of the results shall be noted. Record the maximum classification achieved for each fibre on the counting sheet in the appropriate column. The various classification catego- ries can then be combined later in any desired way for caiculation of the fibre concentration. The complete record of the results obtained when attempting to identify each. fibre can also be used to re-assess the data if necessary. In the unknown sample, zone-axis analysis will be re- quired if the presence of amphibole is to be unequivocally confirmed. For this level of analysis, at- tempt to raise the classification of every suspected amphibole fibre to the ADQ category by inspection of the random orientation ED pattern and the EDXA spectrum. In addition, examine at least one fibre from each type of suspected amphibole found by zone-axis methods to confirm their identification. In most cases, because information exists about possible sources of asbestos in close proximity to the air sampling ►o- cation, some degree of ambiguity of identification can be accepted. Lower levels of analysis can therefore be accepted for these situations. r� �► o ISO ISO 10312:1995(E) � FIBRE WITHOUT TUBULAR MORPHOLOGY Does (hro EDXA spectrum �how elementa Examine by undom comistent with emphiboteT oriant�don SAED YES NO P�ttem not proaent l.�yar p�ttem with NAM or indistinct 0,53 nm �pecing No spectrum Pettem definftaty � UF nnt amphibole rype UF NAM � Does quentitative EDXA give fi6ra compoailion eonsietant with emphi6ole) Does EDXA apactrum Does EDXA spxtnim showelementsconcistent showelementsconaietent YES NO with emphtbole7 with emph(bole7 NAM NO YES YES NO NAM AX A�X P1F'N A� No spectrum No spectrum I� Lt:ane•axis SAEO pattem coneirient with emphibole] UF AD YES NO No pattern An Uniquety emphibo�e solution �s in zoncazia SAED � • •^••• � pattem con�ittent wi[h � Ooes quantitalive EDXA Doas quantitative E�Xl1 amphihole) Is tst zone-axis SAED give fibre composition give fibre eomposition pettern coneiatent with consietent with consi�tent with NO YES emphibole7 � amphi6o�eT amphibole? NAM � VES NO YES NO NO VES No panem AZ �M NAM ADQ � No pattem u� Aa ryqM le 1st zone-axie SAED panern conaistent with Due 2nd zone-axie SAED Is 1st tono-exis SAED emphi6olaT PaKern �nd interexiei engle comietent with patternconsi�tentwith NO VES emphi6ole7 Are 2nd ione-exia SAED amphi6ole7 pattem end interexiei NAM YES NO engle coneiatent with YES NO - amphibole) � VES NO No pettern �•—� Uniquely amphibole tolution No pattern AOn No pattem An Uniquely emphibola solution Ara 2nd:one•axi� SAED peKern end interaxiel enp�e conoistenl wAb emphi6olei NO ' VES No pettern �n C Uniquely amphibole tolution AZQ I Uniquely amphibole wlution Uniquely emphibole solution Uniquely amphibole solution Figure D.4 — Classification chart for fibre without tubular morphology 41 ISO 10372:19951E) Annex E (normative) � � ISO ' Determination of the concentrations of asbestos fibres and bundles longer than 5 µm, and PCM equivalent asbestos fibres In order to provide increased statistical precision and improved analytical sensitivity for those asbestos fi- bres and bundles longer than 5 µm, it may be decided to perform additional fibre counting at a lower magni- fication, taking account only into fibres and bundles within this dimensional range. The result shall be specified as "number of asbestos fibres and bundles longer than 5 µm", For this examination, use a mag- nification of approximately x 10 000, and continue to assign a morphological code to each structure ac- cording to the procedures specified in annex C. Re- cord fibres and bundles only if their lengths exceed 5 µm. Record cluster and matrix components only if their lengths exceed 5 µm. It may also be decided to provide increased statistical precision and improved analytical sensitivity for fibrous structures longer than 5 µm, with diameters between 0,2 µm and 3,0 µm, which have historically been the basis of risk estimation in the occupational environment (PCM equivalent asbestos fibres). Use a magnification of approximately x 5 000 for this ex- tended fibre count. The result shall be specified as "number of PCM equivalent asbestos fibres". Asbestos structures within this dimensional range do not necessarily incorporate asbestos fibres or bundles longer than 5 µm. !�a Continue the extended sample examination until 100 asbestos structures have been counted, or untii a sufficient area of the specimen has been examined to achieve the desired analytical sensitiviry calculated according to table 1. The grid openings examined shall be divided approximately equally between a minimum of two specimen grids. � NOTES 17 The specimen area corresponding to the area of filter examined in the PCM fibre counting methods is 0,785 mm2, and is equivalent to approximately 100 grid openings of a 200 mesh grid. 18 Some National Standards require that asbestos fibres longer than 2,5 µm, with diameters between 0,2 µm and 3,0 µm be counted. Use a magnification of x 5 000 for counting fibres within these dimensional ranges. 19 The minimum aspect ratio for definition of a fibre in PCM fibre counting methods and in some National Stan- dards is 3�1. Use of a 3�1 aspect ratio is permitted in this International Standard, if this aspect ratio is mentiored in the test report. The test reports shall include all of the items listed in clause 11. � F.1 General Annex F (normative) Calculation of results The results should be calculated using the procedures specified below. The results can be conveniently cal- culated using a computer programme. F.2 Test for uniformity of distribution of fibrous structures on TEM grids A chek shall be made using the chi-square test, to determine whether the asbestos structures found on individual grid openings are randomly and uniformly distributed among the grid openings. If the total number found in k grid openings is n, and the areas of the k individual frid openings are designated A� to Ak, then the total area of TEM specimen examined is i=k A=�A; r=, The fraction of the total area examined which is rep- resented by the individual grid opening area, p;, is given by A;/A. If the structures are randomly and uni- formly dispersed over the k grid openings examined, the expected number of structures falling in one grid opening with area A; is np;. If the observed number of structures found on that grid opening is n;, then X2 _ �'' �n� — nP��2 L, nP� ;=t This value shall be compared with significance points of the chi-square distribution, having (k— 1) degrees of freedom. Significance levels lower than 0,1 % may be cause for the sample analysis to be rejected, since this correspond to a very inhomogeneous deposit. If the structure count fails this test, the precision of the result will be uncertain, and if new air samples cannot be collected, additional grid openings may be exam- ined or the sample may be prepared by an indirect method. ISO 10312:1995�E) F.3 Calculation of the analytical sensitivity Calculate the required analytical sensitivity S, ex- pressed in number of structures per litre, using the following equation: � S �9v where Af is the area, in square millimetres, of sam- ple collection filter; A9 is the area, in square millimetres, of TEM specimen grid opening; k is the number of grid openings examined; V is the volume of air sampled, in litres. F.4 Calculation of the mean and confidence interval of the structure concentration In the structure count made according to this Inter- national Standard, a number of grid openings have been sampled from a population of grid openings, and it is required to determine the mean grid opening structure count for the population on the basis of this small sample. The interval about the sample mean which, with 95 % confidence, contains the population mean, is also required. F.4.1 Calculation of the mean structure concentration Calculate the mean structure concentration C, ex- pressed in number of structures per litre, using the following equation: C=Sn 43 ISO 10312:1995(E) � ISO ^ where S is the analytical sensitivity, expressed in number of structures per litre; is the total number of structures found on ail grid openings examined. F.4.2 Calculation of confidence intervals The distribution of structures on the grid openings should theoretically approximate to a Poisson distri- bution. Because of fibre aggregation and size- dependent identification effects, the actual structure counting data often does not conform to the Poisson distribution, particularly at high structure counts. An assumption that the structure counting data are dis- tributed according to the Poisson distribution can therefore lead to confidence intervals narrower than are justified by the data. Moreover, if the Poisson distribution is assumed, the variance is related only to the total number of structures counted. Thus, a par- ticular structure count conducted on one grid opening is considered to have the same confidence interval as that for the same number of structures found on many grid openings. However, the area of sample actualiy counted is very small in relation to the total area of the filter, and for this reason, structures shall be counted on a minimum of four grid openings taken from different areas of the filter in order to ensure that a representative evaluation of the deposit is made. At high structure counts, where there are adequate numbers of structures per grid opening to allow a sample estimate of the variance to be made, the distribution can be approximated to a Gaussian, with independent values for the mean and variance. Where the sample estimate of variance exceeds that implicit in the Poissonian assumption, use of Gaussian statis- tics with the variance defined by the actual data is the most conservative approach to calculation of confi- dence intervals. At low structure counts, it is not possible to obtain a reliable sample estimate of the variance, and the distribution also becomes asymmetric but not neces- sarily Poissonian. For 30 structures and below, the distribution becomes asymmetric enough for the fit to a Gaussian to no longer be a reasonable one, and estimates of sample variance are unreliable. Accord- ingly, for counts below 31 structures, the assumption of a Poisson distribution shall be made for calculation of the confidence intervals. 44 F.4.3 Example of calculation of Poissonian 95 9�o confidence intervals For total structure counts less than 4, the lower 95 % confidence limit corresponds to less than 1 structure. Therefore, it is not meaningful to quote lower confidence interval points for structure counts of less than 4, and the result shall be recorded as "less than" the corresponding one-sided upper 95 % confidence limit of the Poisson distribution, as fol- lows: 0 structure = 2,99 times the analytical sensitivity 1 structure = 4,74 times the analytical sensitivity 2 structures = 6,30 times the analytical sensitivity 3 structures = 7,75 times the analytical sensitivity For total counts exceeding 4, the 95 % confidence interval shall be calculated using the values shown in table F.1. Table F.1 gives the upper and lower limits of the two-sided Poissonian 95 % confidence interval for structure counts up to 470. F.4.4 Example of calculation of Gaussian 95 % confidence intervals Calculate the sample estimate of variance s2 using the following equation: i=k �, �� — nPr� 2 sz= '=i (k-1) where n; is the number of structures on the ith grid opening; n is the total number of structures found in k grid openings; p; is the fraction of the total area examined represented by the ith grid opening; k is the number of grid openings examined. If the mean value of the structure count is calculated to be n, the upper and lower values of the Gaussian 95 % confidence interva► are given respectively by and n ts � k+� k e o ISO n ts � k i �� k where 1„ is the upper 95 % confidence limit; � n is the lower 95 % confidence limit; is the total number of structures in all grid openings examined; r is the value of Student's test (probability 0,9751 for (k — 1) degrees of freedom; s is the standard deviation (square root of sample estimate of variance►; k is the number of grid openings examined F.4.5 Summary of procedure for calculation of results In summary, structure counting data shall be calcu- lated as foliows: No structures detected The structure concentration shall be reported as less than the concentration equivalent ot the one- sided upper 95 % confidence limit of the Poisson distribution. This is equal to 2,99 times the analyt- , ical sensitivity. From 1 to 3 structures When 1 to 3 structures are counted, the result shali be reported as less than the corresponding nna-girigrl i innPr 95 % confidence limit for the Poisson distribution. These are 1 structure = 4,74 times the analytical sensi- tivity 2 structures = 6,30 times the analytical sensi- tivity 3 structures = 7,75 times the analytical sensi- tivity From 4 to 30 structures The mean structure concentration and the 95 % confidence intervals shall be reported on the basis of the Poissonian assumption, using the values shown in table F.1. ISO 10312:1995(E) More than 30 structures When more 30 structures are counted, both the Gaussian 95 % confidence interval and the Poissonian 95 % confidence interval sliall be cal- culated. The larger of these two intervals shall be used to express the precision of the structure concentration. When the Gaussian 95 % confi- dence interval is selected for data reporting, the Poissonian 95 % confidence interval shall also be mentioned. F.5 Calculation of structure length, width, and aspect ratio distributions The distributions all approximate to logarithmic- normal, and therefore the size range intervals for cal- culation of the distribution shall be spaced logarithmically. The other characteristics required for the choice of size intervals are that they should allow for a sufficient number of size classes, while still re- taining a statistieally.valid number of structures in each class. Interpretation is also facilitated if each size class repeats at 10 intervals, and if 5 µm is a size class boundary. A ratio from one class to the next of 1,468 satisfies all of these requirements and this value shall be used. The distributions, being approximately logarithmic-normal, when presented graphically, shall be plotted using a logarithmic ordinate scale and a Gaussian abscissa. F.5.1 Calculation of structure length cumulative number distribution This distribution ailows the fraction of the total num- ber of structures either shorter or longer than a given lenath to be determined. It is calculated using the following equation: i=k �n; C�P�k = �_ p x 100 �nr r=� where C(P)k is the cumulative number percentage of structures which have lengths less than the upper bound of the kth class; n; is the number of structures in the ith length class; P is the total number of length classes. 45 ISO 10312:1995(E) � ISO �' 9 F.5.2 Calculation of structure width cumulative number distribution This distribution allows the fraction of the total num- ber of structures either narrower or wider than a given width to be determined. It is calculated in a similar way to that used in F.5.1, but using the structure widths. F.5.3 Calculation of structure aspect ratio cumulative number distribution This distribution allows the fraction of the total num- ber of structures which have aspect ratios either smaller or larger than a given aspect ratio to be de- termined. It is calculated in a similar way to that used in F.5.1, but using the structure aspect ratios. Table F.1 - Upper and lower (imits of the Poissonian 95 � rnnfirlanrn �.,*a...,� „E �........� Structure Structure Structure . Y. v. y VVV••` count Lower limit Upper limit count Lower limit Upper limit Lower limh Upper limit count � 0 3,689�� 46 33,678 61,358 92 74,164 112,83 1 0,025 5,572 47 34,534 62,501 93 75,061 113,94 2 0,242 7,225 48 35,392 63,642 94 75,959 115,04 3 0,619 8,767 49 36,251 64,781 95 76,858 116,14 4 1,090 10,242 50 37,112 65,919 96 77,757 117,24 5 1,624 17,669 51 37,973 67,056 97 78,657 118,34 6 2,202 13,060 52 38,837 68,192 98 79,557 119,44 7 2,814 14,423 53 39,701 69,326 99 80,458 120,53 8 3,454 15,764 54 40,567 70,459 100 81,360 121,66 9 4,115 17,085 55 41,433 71,591 110 90,400 132,61 10 4,795 18,391 56 42,301 72,721 120 99,490 143,52 11 5,491 19,683 57 43,171 73,851 130 108,61 154,39 12 6,201 20,962 58 44,041 74,979 140 117,77 165,23 13 6,922 22,231 59 44,912 76,106 150 126,96 176,04 14 7.654 23,490 60 45,785 77,232 160 136,17 186,83 15 8,396 24,741 61 46,658 78,357 170 145,41 197,59 16 9,146 25,983 62 47,533 79,482 180 154,66 208,33 17 9,904 27,219 63 48,409 80,605 190 163,94 219,05 18 10,666 28,446 64 49,286 81,727 200 173,24 229,75 19 11,440 29,671 65 50,164 82,&48 210 182,56 240,43 20 12,217 30,889 66 51,042 83,969 220 191,89 251,70 21 13,00 32,101 67 51,922 85,088 230 201,24 261,75 22 13,788 33,309 66 52,803 86,207 240 210,60 272,39 23 14,581 34,512 69 53,685 87,324 250 219,97 283,01 24 15,378 35,711 70 54,567 88,441 260 229,36 293,62 25 16,178 36,905 71 55,451 89,557 270 238,75 304,23 26 16,9B3 38,097 72 56,335 90,673 280 248,16 314,82 27 17,793 39,284 73 57,220 91,787 290 257,56 325,39 28 18,606 40,468 74 58,106 92,901 300 267,a1 335,96 29 19,422 41,649 75 58,993 94,014 310 276,45 346,52 30 20,241 42,827 76 59,880 95,126 320 285,90 357,08 31 21,063 44,002 77 60,768 96,237 330 295,36 367,62 32 21,886 45,175 78 61,657 97,348 340 304,82 378,15 33 22,715 46,345 79 62,547 98,458 350 314,29 388,68 34 23,545 47,512 80 63,437 99,567 360 323,77 399,20 35 24,378 48,677 81 64,328 100,68 370 333,26 409,71 36 25,213 49,840 82 65,219 101,79 380 342,75 420,22 37 26,050 51,000 83 66,111 102,90 390 352,25 430,72 38 26,890 52,158 84 67,003 104,00 400 361,76 441,21 39 27,732 53,315 85 67,897 105,11 410 371,27 451,69 40 28,575 54.469 86 68,790 106,21 420 380,79 462,18 41 29,421 55,622 87 69,664 107,32 430 390,32 472,65 42 30,269 56,772 88 70,579 108,42 440 399,85 483,12 43 31,119 57,921 89 71,474 109,53 450 409,38 493,58 44 31,970 59,068 90 72,370 110,63 460 418,92 504,04 45 32,823 60,214 91 73,267 111,73 470 428,47 514,50 1) The one-sided upper 95 % confidence limit for 0 structures is 2,99. 46 ^ o ISO ISO 10312:1995(E) � G Annex G (informative) Strategies for collection of air samples G.1 General An important part of the sampling strategy is a state- ment of the purpose of the sampling programme. A sufficient number of samples should be collected so that the site is well characterized to the precision and accuracy desired, and also ensure that sample filters appropriately loaded for TEM analysis are obtained from all of the sampling locations. G.2 Air sample collection in the outdoors environment Weather conditions restrict the ability to collect satis- factory air samples in the outdoors environment, and whenever possible, sampling should be carried out in low-wind, low-humidity conditions. Detailed records of the weather conditions, windspeed and direction during the sampling period should be made. All avail- able information concerning local topography, and the types and positions of sources should be recorded. Sequential multipoint sampling is necessary to pro- vide adequate characterization of complex sites and sources. It is recommended that multiple samples are taken upwind and downwind of the site, with a mini- mum of two samples in the downwind position ex- pected to experience the maximum airborne concentration. The locations of the samplers should be carefully recorded. G.3 Air sample collection inside buildings Air samples are often collected inside buildings in which asbestos-containing construction materials are present, in order to determine whether these materi- als contribute to the asbestos fibre concentration in the building atmosphere. The optimum positions for collection of air samples can only be determined after a complete survey of the building to establish air movement patterns. Multiple samples should be col- lected in the area where asbestos building materials are present, and control samples should be collected in an adjacent area where no airborne asbestos fibres would be expected. The intakes for air conditioning systems are frequently used as the collection lo- cations for control samples. Whenever possible, static samples should be taken over a period exceeding 4 h during normal activity in the building, at face ve- locities of between 4 cm/s and 25 cm/s. 47 ISO 10312:1995(E) � ISO � Annex H (informative) Methods for removal of gypsum fibres It is common to find fibres of calcium sulfate (gypsum) in airborne particulates collected in buildings and urban environments, and particularly in samples collected where demolition or construction work is in progress. The fibres are readily released when plas- ters and cement products are disturbed. In some cir- cumstances, particles of caicite or dolomite collected on an air filter can react with atmospheric sulfur diox- ide, to form long fibres of gypsum. Gypsum fibres can give rise to high fibre counts by both optical and electron microscopy. The gypsum fibres are often. 2 µm to 6 µm long, with aspect ratios greater than 10�1. Sometimes, these fibres appear similar to amphibole asbestos fibres, and in some samples they can be morphologically very similar to chrysotile. In the TEM, the larger fibres have high contrast and at high magnification often exhibit a characteristic mottled appearance which changes under electron beam irradiation. Some gypsum fibres, however, are not easily discriminated from asbestos without ex- amination by EDXA, TEM specimens which contain many such gypsum fibres require an extended exam- ination time in the TEM, because it is necessary to 4$ examine each of these fibres by EDXA before it can be rejected. It is possible to remove gypsum fibres selectively by water extraction. A Jaffe washer (7.3.7), or a condensation washer (7.3.8), should be prepared, but using a water (6.1) as the solvent. The TEM speci- mens, which have been previously prepared and ini- tially examined in the TEM, shouid be placed in the washer to allow dissolution of the fibres. If a Jaffe washer is used, the treatment time can be reduced by heating the washer to 90 °C to 100 °C for a few minutes. If a condensation washer is used, the gypsum fibres wiil be dissolved by treatment for ap- proximately 10 min. The effect of this treatment is to remove the gypsum fibres, leaving carbon replicas f7.3.11) �vhich are readily distinguished from asbestos fibres. NOTE 20 This procedure should be used only when ex- amination of the untreated TEM specimen grids shows the gypsum fibres to be isolated from any asbestos fibres present. Losses of asbestos fibres may occur if matrices of gypsum and asbestos are exposed to this procedure. Annex J (informative) Bibliography [1] Asbestos International Association (1979): Ref- erence method for the determination of asbestos fibre concentrations at workplaces by lrght microscopy (membrane filter method). AIA health and safety publication, recommended technical method No. 1(RTM1). Asbestos Inter- nationai Association, 68 Gloucester Place, > London, W1H 3HL, England. [2] BRADLEY, D.E. (1961): Replica and shadowing techniques. In: Techniques for Electron Microscopy, Blackwell Scientific Publications, Alden, Oxford, D.H. Kay (Ed.), pp. 96-152. [3] BuROErt, G.J. and Roo�, A.P. (1982): Membrane-filter, direct transfer technique for the analysis of asbestos fibres or other inorganic particies by transmission electron microscopy. Environmental Science and Technology, 17, pp. 643-648. [4] CAMPBE�L, W.J., BLAKE, R.L., BROWN, �.L., CATHER, E.E. and SJOBERG, J.J. (1977): Selected silicate minerals and iheir asbestiform varieties. Mineralogical definitions and identification- characterization. Information circular 8751. United States Department of the Interior, Bu- reau of Mines, Washington, D.C. [5] CHATFIELD, E.J. (1986): Asbestos measurements in workplaces and ambient atmospheres. In: Electron microscopy in forensic, occupational, and environmenta/ hea/th sciences (S. Basu and J.R. Millette, Eds.). Plenum Press, New York, pp. 149-186. [6] CHATFIEID, E.J. (Editor> (1987j: Asbestos fibre measurements in building atmospheres. Ontario Research Foundation, Sheridan Park Research Community, Mississauga, Ontario, Canada, L5K 163. [7] CHATFIELD, E.J. and LEw�s, G.M. (1980): Devel- opment and appiication of an analytical tech- nique for measurement of asbestos fibers in vermiculite. In: Scanning Electron Microscopy(1980/1, (O. Johari, Ed.), SEM Inc., AMF O'Hare, Chicago, Illinois 60666, USA. ISO 10312:1995(E) [8] CLIFF, G. and LORIMER, G.W. (1975): The quan- titative analysis of thin specimens. Jou�nal of Microscopy, 103, pp. 203-207. [9] DEER, W.A., HowIE, R.A. and ZusSMAN, J. (1963): Rock-formrng minerals. Longmans, London. [10] Federal Register (1987): Asbestos-containing materials in schools. U.S. Environmental Pro- tection Agency. Vol. 42, No. 210, October 30, 1987, pp. 41826-41905. [11] GaRo, J. A. (Editor) (1971): The Electron Optical lnvestigation of Clay's. Mineralogical Society, 41 Queen's Gate, London S.W. 7. [12� GaZE, R. (1965): The physical and molecular structure of asbestos. Annals of the New York Academy of Science, Vol. 132, pp. 23-30. [13] HAWTHORNE, F.C. (1983): The crystai chemistry of the amphiboles. Canadian Mineralogist Vol. 21, part 2, pp. 173-480. [14] HiRscH, P.B., HowiE, A., NICHOLSON, R.B., PASHLEY, D.W. and WHELAN, M.J. (1965): Elec- tron microscopy of thin crystals. Buttenivorths, London, pp. 18-23, [15] HOLLAHAN, J.R. and BELL, A.T. (Editors) (1974}: Techniques and applications of p/asma chemis- try. Wiley, New York. [16] International Centre for Diffraction Data (1987): Powder diffraction file. International Centre for Diffraction Data, 1606 Park Lane, Swarthmore, Pennsylvania 19081, USA. [17] international Mineralogical Association (1978): Nomenclature of amphiboles (compiled by B.E. Leake); Canadian Mineralogisi, Vol. 16, p. 501. [18] International Organization for Standardization (Organisation internationale de normalisation) (1993): ISO 8672:1993, Air quality — Determi- nation of the number concentration of airborne 4J ISO 10312:1995tE) � ISO � inorganic fibres by phase contrast optical microscopy — Membrane filier method. [19] JAFFE, M.S. (1948): Handling and washing fragile replicas. J. Applied Physics, 19, p, 1187. [20] JoY, D.C., ROMIG, Jf. and GOLDSTEIN, J.I. (Edi- tors) (1986): Princip/es of analytica/ e/ectron microscopy, Plenum Press, New York and London. [21� LE�oux, R.L. (Editor) (1979): Short course in mineralogica! techniques of asbestos determi- nation. Mineralogical Association of Canada, Department of Mineralogy, Royal Ontario Mu- seum, 100 Queen's Park, Toronto, Ontario Canada M5S 2C6. tional Safety and Health, 4676 Columbia Park- way, Cincinnati, Ohio 45226, USA. [29] NATRELLA, M.G. (1966): Experimental statistics: National Bureau of Standards Handbook 91. U.S. Government Printing Office, Washington, D.C. 20402. [30] ORTiz, L.W. and IsoM, B.L: (1974): Transfer technique for electron microscopy of inembrane filter samples. American lndustrial Hygiene As- sociation Journal, 35, 7, pp. 423-425. � [31] PEARSON, E.S. and HARTLEY, H.Q. (1958): Biometrica tables fo� statisticians, Vol. 1, Cambridge University Press, 32 East 57th Street, New York, N.Y. 22, USA. [22) LEVADIE, B. (Editor) (1984): Definitions for [32] RHOADES, B.L. (1976): XIDENT-A computer asbestos and oiher health-related silicates. rechnique for the direct indexing of eleciron dif- ASTM Special Technical Publication 834. Ameri- fraction spot patierns. Research Report 70/76. can Society for Testing and Materials, 1916 Dept. of Mechanical Engineering, Univ. of Race Street, Philadelphia, Pennsylvania 19103, Canterbury, Christchurch, New Zealand. USA. [23] MICHAEL, J.R. and WIIIIAMS, D.B. (1987): A consistent definition of probe size and spatial resolution in the analytical electron microscope. J. Mic., 147, pp. 289-303. [24] MICHAELS, �. and CH�ssICK, S.S. (Editors) (1979): Asbestos: Properties, Applications and Hazards, Vol, 1, Wiley, New York. [25] National Bureau of Standards Special Publication 506 (1978): Workshop on asbestos: definitions and measurement methods. U.S. Government Printing Office, Washington, D.C. 20402. [26] National Bureau of Standards Special Publication 619 (1982): Asbestos standards: material and analytica/ methods. U.S. Government Printing Office, Washington, D.C. 20402. [27] National Institute for Occupational Safety and Health (1989): N/OSH Method 7400, Revision #3, 5/15/89. U.S. Department of Health and Hu- man Services, Public Health Service, Centers for Disease Control, National Institute for Occupa- tional Safety and Health, 4676 Columbia Park- way, Cincinnati, Ohio 45226, USA. [28] National Institute for Occupational Safety and Health (1989): NIOSH Meihod 7402, Revision #1, 5�15/89. U.S. Department of Health and Hu- man Services, Pubiic Health Service, Centers for Disease Control, National Institute for Occupa- 50 [33] R�N�, S.J. (1980): Identification of amphibole fi- bers, including asbestos, using common elec- tron diffraction patterns. ln: Electron Microscopy and X-ray Applications to Environmenta! and Occupational Healih Analysis, (Ed. P.A. Russell), Vol. II, Ann Arbor Press, Ann Arbor, Michigan 48106, USA. [34] RUSSE�L, P.A. and HUTCHINGS, A.E. (1978): Electron Microscopy and X-ray Applications to Environmental and Occupational Health Analy- sis. Ann Arbor Science Publishers Inc., P.O. Box 1425, Ann Arbor, Michigan 48106, USA. [35] SMALL, J.A., HEINRICH, K.F.J., NEWBURY, D.E. and MYKLEBUST, R.�. (1979): Progress in the devel- opment of the peak-to-background method for the quantitative analysis of single particles with the electron probe. Scanning EFectron Microscopy/1979/II, (O. Johari, Ed.). SEM Inc., AMF O'Hare, Chicago, illinois 60666, USA. [36] SMAI.L, J.A., STEEL, E.B. and SHERIDAN, P.J. (1985): Analytical standards for the analysis of chrysotile asbestos in ambient environments. Analytical Chemistry, 57, pp. 204-208. [37] SMITH, J.E. and JORDAN, M.L. (1964): Math- ematical and graphical interpretation of the log- normal (aw for particle size distribution analysis. J. Colloid Science, 19, pp. 549-559. [38] SPURNY, K.R., STOBER, H., OPELIA, H. and WEiss, G. (1979): On the evaluation of fibrous � ,7 o ISO particles in remote ambient air. Science of ihe Total Environment/1979/II, pp. 1-40. [39] SPURNY, K.R. (Editor) (19861: Physical and chemical characterizaiion of individual airborne particles. Wiley, New York. [40] STEEL, E.B. and SMALL, J.A. (1985): Accuracy of transmission electron microscopy for the analysis of asbestos in ambient environments. Analytical Chemisfry, 57, pp. 209-213. [41� SrEE�, E.B. and WYLIE, A. (1981): Mineralogical characteristics of asbestos. In: Geology of ISO 10312:1995(E) Asbestos Deposits, (P,H. Riorden, Ed.), SME-AIME, pp. 93-101. [42] WEtvK, H.R. (Editor) (1976): Electron microscopy in mineralogy. Springer-Verlag, New York. [43] Ya�a, K. (1967): Study of chrysotile asbestos by a high resolution electron microscope. Acta Crysiallographica, 23, pp. 704-707. [44] ZUSSMAN, J. (1979?: The rnineralogy of asbestos. In: Asbesios: Properties, Applications and Haz- ards, John Wiley and Sons, pp. 45-67. 51 ISO 70312:1995(E►� ICS 13.040.20 Descriptors: air, quality, air pollution, tests, determination, particle density (concentrationl, asbestos, microscopic analysis. . Price based on 51 pages Q�P Appendix C June 2, 2000 Revision 0 Appendix C ._ _ Standard Operating Procedure for the Screening Analysis of Soil and Sediment Samples for Asbestos Content SOP:EIA-INGASED2.�OP Asbestos in Sedimejnts/Seils 1/11/99 ' Page 1 oP 7 EPA, xeqion I 8tanaard Operating Proaodure for the screeninq Analyaia oi Boil aad 8�dimant samples !or Asbestos Cant��t Prapared for: Offic• et Eavironm�atal Measurem�at Aad Evaluatioa �.8. EPA, &�gibtt I R�vised by: G�`'vl saott Cliitord, at, Inv�atigation� And Analyaia Un C, EME � , �--�--�—�_ `,. 'r_ � � Approvsd by: Ag�nea Ya a ove, Ph.D., QA pffiaer,} ti nd Analysis unit, o8H$' Approv�d Dy : 1 a 4! �7 l. oDert . Maxiiald, M�nagsr, =nve�tiga�ians Attd Aanlyeia IInit, OEKE . 50P:EIA-INGAS$D2.5pP Asbeatos in SEdime�CsJsails 1/il/99 Page 2 of 7 � 6 = This SOP details the sample analysis protocol far determining the ashastos cantent of soil and sediment samp�es. Sample clean ug with a sieve washing is followed by stereo microscope and polarized light microscope examination. This protacol is a semi-quantitative methvd, used anly for visually astimating pexcentage levels of asbestos in a soil or sedimeht sample. � 2. Pur ae�s To ensure that the pr�tocols for analysis af asbestoe in sedime�t/soil samples are eonsietently app1ied by all analysts. 3 a spou� mnQ Aoolicatian• This protocol is not a�'reference�' method, It was de�e2oped ou� oi nece�eity to facilitate finding asbestos fiber� in a soil or muc9 (sediment) sample that does not contain any obvious asbestos filoers or asbestos-containing buiZding or product materiais when examined�dry (cr t�ret) using a stereo microscope at lOX or 2oX magnificatian. It;has been used for asbestos content in order to help delineate contaminated area�. It has proven to be an extremely sensitive method capable of f ir�ding very small amounts of asbestoB fibers in a soil or sediment matrix. Glasaware ar othex materials, suppliea, ete,, mentioned below in the pracedures are those being used in this laboratory. other mate�ials may be substituted as long as the primary purpose of the protocol is followed, name2y: to find asbestos fibers in the sample. Identification of fibrous components is accomplished by the routine polari$ad Ligbt Mioroaacpy {pi,Id} (with dispersion staining) method. (See EPA approved method references on page 5.) 4.0 Deiiaitions: PLM - Polarized Light Micrdacopy 5.o salt and Sa st Wa in s: Asbestos fibers ca� have serious effects on your health if inhaled. Sample containers should be irtitially opened in the HEPA fil'kered haod. Care mue� be taken vhen handling any unknown samples to prevent airborne a'sbestos. PLEASE AECYGL� THIB PAPER saP:ESA-ix�ASED2.SOP Asbestos in SedimentsjSoils 1/Ilf 99 Page 3 of 7 YTI PROC �8 1. A representative portion of the sample is removed from the sample container after thorcughly mixing for homogeneity. Because asbestas fibers us�ally cannot be seen because of the composftion of the sample matrix such �s the dirt, sand, mud, vegetation, water, etc., steps must be taken to cle�n up the sample to tha poirit where the asbestas fibers, if any, may be seen using the stereomicrascope at 1oX to 20 X magnification. A atereomiarosoope ie mandatorg �or this prat000i. 2. To eliminate interfering particles, a 16mm ID by 150mm long, goo� quality : PYREX or KIMAX test tube (not a fragile disposable tube) is used to gemave partfone of the well-mixed soilJsediment sample from several places in the sample container by pushing it into the sample ta accumulate a sample depth of about 2.5 inches (65mm) in Ehe test tube. A glase or plastic stirring rod is used to push the sample down into the tube and fiber-free (tapj wate� is added for shaking purpases. The soil and water mixture is shaken vigorausly ta loose� and separate the fines a�d other components of the sample and:the cantehts of the test tube are then poured into a 3 inch ID� b0 mesh Q25Q mlcrometers) sieve. This serv�s to eliminate, or greatly reduce, co�loidal material, ffne sand, si2t and other han-fibrous particulates from the sample. More water is added to the tube, shaken and dumped into the sieve. �epeat this step until the tube is clean. The sample in the sieve is then rinsed until clean (clear water running through the sievej with a fairly fine, Qr�ssuri2ed stream af water from a plastic wash bottle. All of the material remaining in the sieve is then washed from the s$eve screen using a stream of water from the rinse bottle into a square plastic weighing dish of about looml liquid capacity. Use just enough water to compl�tely cover the sample in the dish abaut 1/8th inch or so for examina�ion with t�e stereo micrascope. � After the cleaned sample is transferred to the weighing dish for exa�ination, thoraughly rinse the sieve and test tube under running tap water (pr�ferably aerated ta minimize splashittg) and carryover will not be a problem from sample to sample. It is a good idea to carry aut all washing af the sample fines over a plastic dishpan or other container set into the sink basin in order;to capture the fines and keep them from clogging the eink drain trap. A;fter a settling periad, the overlying water may be poured off and the fines/�ud disposed of separately. - PLEAHE xECYCLE T$Z8 PAPER NOTE: Since the purpase of a significant amount techriique, �11 porti SOP:EIA-INGASED2.S P Aebeetos in Sedime�ts/soils 1/11/99 . Page 4 of 7 the test is to find out if the eoil ar mud samp�e contains of asbestos (>1�) that can be i.dentified using �he PLM „e „s .,.- ----�- ---- -- - • - 3. After examining them for asbestos fibers, floating pieces of o�ganic material such as rvats, sticke, leaves, etc., may be removed to get°a better view af the rest of the eample in the dish. Frequently, root structWres found in surfaca soils wi11 trap asbestos fibers durinq the shaking proces� at�d are a good placa to loak for the fibers. The eample is then caraf��lY and systematically examined under the stereo miero5cope at lOX-2oX magnification for visible asbestas f ibers and fiber bundles. A good, bright, fo�used l�ght source such as a Nieholas transformer-base external illuminatar is v�ry helpful here. The fibers tend to stand out, shine, flash, etc., in the clean water matrix. Poking and stirring the eample �uith forcepe and/or dissectiMg needles will help to Iocate the fibers. If no fibers are seen, gantly shaking the weighing dish to redistribute particles will sametimes turn up previ4usly hidden fibers when scanning the sample a second time. Suspect fibers are removed with sharp forceps and placed upan a clean micrascape slide. ' 4. After picking as many suspect fibers or other material from the�sample as necessary to determine its �ontent, the slide preparation is allowed to dry and prepared for PLM analysis using an appropriate high-dispersion refractive�index liquid and coverslip. 5, Next, th� slide preparation is examined with a polarized light microscop� (PLM} witb dispersion staining to identify any fibers found. Standaz�d, EPA appraved PLM procedures are used to iflentify any asbestos fibers fourid as to specific type and form. The identification of asbestos using Pi.t�i is rapid and unequivocal due to the unique optical crystallographic praperties of � morpholagy, refractive indices, elongation, angle or extinction, dispersion and hirefrinqenae. Slide preps are examined for each sampla with suspect fibers to _ confirm the presence of asbestas. 6. Yf asbestos fibers are identified, return to the sieved sampla uhcier the sterea microscope, observe the remaining asbestos fibers and bundles of Pibers, and make a vi5uai estimate of ths pexcentage asbestoe content in the i�hole sampl. inciuding tne material prevfously x�shod tbrouqh the aieve. (This is all based ott asbestos fibers seen using loX ta 20X magnification unde� the stereamicroscope.) Obviously, many of the finest fibers pass through�the sieve and the finest ones remaining can't be seen at 20X magnificatian, but., �his protocol is nat meant to be used ar a auantitative met od It is useful, however, ta determine whether or not the soil or aediment is contaminated with significant amounts of asbestos. (> than 1� by volume}. PLEABE RECYCLE THIS PAPER SOP;EIA-IN�ASED2.SbP Asbastos in Sedime�tsjSails 1/�1/99 Page 5 of 7 7. As a rule, if asbestos fibera and/or fiber bundles can't be �ou7►d relatively qui.ckly and eaeily (one to two minutes) in the cleaned up�sample under the stereo microscope, the percent asbestos contQnt is most likely less than 0.1� anc3 certainly less than 1.0� : One should be able to find;asbestos fibers in a sample containing more than 1.0� in a f�w seconds up to a one miriute examination under the stereo microscape. 8. Usually, as much of the original sample as possible is returned�to the ariginal sample container. If saving the fines is required fer further examination by other methods, use individual beakers to catch the sieve washings c�ntaining the sand and sflt components. 9. The sample containers are then resealed before storage, disposal or return to the organization that requested the analyses. 10. If further information is required ori this protocol or if anyor►e has found a better way to find and estfmate aebestos fibers in muds or soils, please contact: Scott clifford USEPA, REGION 1 60 Westview St. Lexington, MA or Dan Boudreau 02421 Telephone; (781) 860-4300 APPRQVED EPA BULX ANAI�X�I� PROTaCOL: 40 CFR PART 763, SUBPART F, APPENDIX A � or, "ASBESTOS IDENTIFICATION'�- Walter C. McCrone, 1987, McCrone Rese�rch Institutet Chicago, Ill. See also, ",�,sbestos Content In Bu].k �nsulation Samples• Visual Estimates and W�ight Comx�ositian"- US EPA, EPA-560/5-88-011, September, 1988. PLEABE RECYCLE THIB PAPEA SQP:EIA-INGASED2.S�OP Asbestos in Sedim�nts/Soils lJll/99 Page 6 vf 7 7.0 Ahalvtical Procedura Addemdum for �er� Aaouret� OuantitAtiou: . ADDENDUI�i TO "PRaTOCOL �OR SC&EENINa BOIL AND SEDIME�T SAispLEB FOR�AB88BT08 CONTEPI'P IIBED BY 2S� O.B. ENFIRONMENTAL PAOTBCTION AQENCY� REGZON I I,�$QRi�TORY" Addeadum deted: August 194? This addendum can be used to more accurately quantitatQ the volume �� ashestas in soil and sediment samples. It fe meant to give the analyst a good visual ;; estimate of the fine materials volume which pass through the mesh sieve relative to the ariginal sample volume analyzed. It must be used in conjunction t�ith the above mentioned protocol. $r�mDle Pr�paratian Anel�tfaal ProaeQure• 1. Transfer a well mixed portion of homoganized soil into the�plastic weigh dish. Cover the bottom of the dish with a thin {icmp layer. Quantitatively transfer the soil/sediment into the test tuba using a wide mouth funnel. Fiber-frea (tap) water can be used to help wash fines into the test tube. A glass or plaetic stirring rad�is used ta push the sample down into the test tube end and to break-up� any sail clumps. Wash the test tube sides down with a stream of w�:ter. Let the sailJwater mix settle such that the volume of material'in the test tube can be measured. Measure from the bottcm of the�test tube ta the top of the settled soil with a rttler and record the value (i.e., 4.5cm). After the soil volume measurement, cantinue the analytical'procedure (i.e., shake vigorously to loosen and separate the fines, �pour contents into sieve for clean-up, transfer sample componen� left in sieve to weigh dish for examination, etc.) After complete examination and determination of asbestos c�ntent of the sample portion in the weigh dish which did not pass through the sieve (using stereo microscope and P?,M),� the sample in the weigh dish is quantitatively transferred back into the test tube and allowed to settle. After settling, the volume of material in the tes� tube is again measured and reoarded (i.e., 2.Ocm). . PLEASF RECYCL� THIB PAPE& SOP:EIA-INGASEB2.SQP Asbestos in Sedime�ts/Soils 1%11%99 Page � of 7 • Determine the asbestos content of the sample as follows: $ asbestos in samp2e Volume of sample which portion which did not X did not pass through pass through the sieve siev� Initial sample volume PLEABE &EdYCLE TSIB PAPER ' ' � ' ' i� Q�P Appendix D June 2, 2000 Revision 0 Appendix D ASTM Standard Test Method D 4959-00 Determination of Water (Moisture) Content of Soil by Direct Heating Q��M Designation: D 4959 — 00 .� �� Standard Test Method for Determination of Water (Moisture) Content of Soil By Direct Heating� This standard is issued under the fixed designation D 4959; the number immediately following the designation indicates the year of original adoption or, in the case of cevision, the year of last revision. A number in parentheses indicates the yeaz of last reapproval. A superscript epsilon (e) indicates an editorial change since the last revision or reapproval. 1. Scope 1.1 This test method covers procedures for deternuning the water (moisture) content of soils by drying with direct heat, such as using a hotplate, stove, blowtorch, etc. 1.2 This test method can be used as a substitute for Test Method D 2216 when more rapid resuits are desired to expe- dite other phases of testing and slightly less accurate results are acceptable. 1.3 When questions of accuracy between this test method and Method D 2216 arise, Method D 2216 shall be the referee method. 1.4 This test method is applicabie for most soil types. For some soils, such as those containing significant amounts of halloysite, mica, montmorillonite, gypswn, or other hydrated materials, highly organic soils or soils that contain dissolved solids, (such as salt in the case of marine deposits), this test method may not yield reliable water content values. 1.5 The values stated in SI units are to be regarded as standard. 1.6 This standard does not purport to address all of the safety concerns, if any, associated with its use. It is the responsibility of the user of this standard to establish appro- priate safety and health practices and determine the applica- bility of regulatory limitations prior to use. 2. Referenced Documents 2.1 ASTM Standarzls: D 653 Terminology Relating to Soil, Rock, and Contained Fluids2 D 2216 Test Method for Laboratory Determination of Water (Moisture) Content of Soil and Rock2 D 3740 Practice for Minimum Requirements for Agencies Engaged in the Testing and/or Inspection of Soil and Rock as Used in Engineering Design and Construction2 D 4753 Specification for Evaluating, Selecting, and Speci- fying Balances and Scales for Use in Testing Soil, Rock, and Related Construction Materials2 � This test method is under the jurisdiction of ASTM Committee D-l8 on Soil and Rock and is the direct responsibility of Subcommittee D18.08 on Special and Construction Control Tests. Current edition approved Macch 10, 2000. Published Apri1 2000. Originally published as D 4959 — 89. Last previous edition D 4959 — 89 (1994). 2 Annua! Boak oJASTM Standards, Vol 04.08. Copyright O ASTM, 100 Barr Harbor Drive, West Conshohocken, PA 19428-2959, United States. 3. Terminology 3.1 Definitions All definirions are in accordance with Terms and Symbols D 653. 3.2 Definitions of Terms Spec�c to This Standard: 3.2.1 direct heating—a process by which the soil is dried by conductive hearing from the direct application of heat in excess of 110°C to the specimen container, such as provided by a hot plate, gas stove or burner, heatlamps, or other heat sources. Direct application of heat by flame to the specimen is not appropriate. 3.2.2 water (moisture) content—the ratio, expressed as a percentage, of the mass of water in a given mass of soil to the mass of the solid particles. 4. Summary of Test Method 4.1 A moist soil specimen is placed in a suitable container and its mass is determined. It is then subjected to drying by the application of direct heat until dry by appearance, removed from the heat source, and its new mass is determined. This procedure is repeated until the mass becomes constant within specified limits. 4.2 The difference between the masses of the moist speci- men and the dried specimen is used as the mass of water contained in the specimen. The water content (expressed as a percentage) is determined by dividing the mass of water by the dry mass of soil, mulriplied by 100. For a given soil type and specimen size, the time to achieve a constant dry mass can be noted and used to estimate drying time for subsequent tests of the same soil type using the same size specimen and drying apparatus. 5. Significance and Use 5.1 The water content of a soil is used throughout geotech- nical engineering practice both in the laboratory and in the field. The use of Test Method D 2216 for water content determination can be time consuming and there are occasions wheri a more expedient method is desirable. Drying by direct heating is one such method. Results of this test method have been demonstrated to be of satisfactory accuracy for use in field control work, such as in the determination of water content, and in the determination of in-place dry unit weight of SOI�S. 5.2 The principal objection to the use of the direct heating �� D 4959 for water content determination is the possibility of overheat- ing the soil, thereby yielding a water content higher than would be deternuned by Test Method D 2216. While not eliminating this possibility, the incremental drying procedure in this test method will minimize its effects. Some heat sources have settings or controls that can also be used to reduce overheating. Loose fitting covers or enclosures can also be used to reduce overheating while assisting in uniform heat distribution. 5.3 The behavior of a soil when subjected to direct heating is dependent on its mineralogical composition, and as a result, no one procedure is applicable for all types of soils or heat sources. The general procedure of this test method applies to ail soils, but test details may need to be tailored to the soil being tested. 5.4 When this test method is to be used repeatedly on the same or similar soil from a given site, a correction factor can usually be determined by making several comparisons between the results of this test method and Test Method D 2216. A correction factor is valid when the difference is consistent for several comparisons, and is reconfirmed on a regular specified basis. 5.5 This test method may not be appropriate when precise results aze required, or when minor variations in water content will affect the results of other test methods, such as borderline situations where small variations in the measured water content could affect acceptance or rejection. 5.6 This test method is not appropriate for specimens lrnown to contain flammable organics or contaminants, and other test methods should be utilized in these situations. Nore I-The quality of the results produced by this test method is dependent on the competence of the personnel performing it and the suitability of the equipment and facilities used. Agencies that meet the criteria of Practice D 3740 aze genetally considered capable of competent and objective testing/sampling/inspec6on. Users of this test method aze cautioned that compliance with Practice D 3740 dces not in itseif ensure reliable results . Reliable results depend on many factocs; Practice D 3740 provides a means of evaluating some of those factors. 6. Interferences 6.1 When testing sand and gravel size particles, additional care must be taken to avoid the possibility of particle shatter- ing. 6.2 Due to the localized high temperatures in the soil during testing, the physical characteristics of the soil may be altered. Degradarion of individual particles may occur, along with vaporization, chemical transition, or loss of organics. There- fore, specimens used in this test method should not be used for other tests subsequent to drying (see Note 2). NorE 2-The subsequent use of specimens dried by direct heating in other test methods is discouraged 7. Apparatus 7.1 Direct Heat Source Any source or heat that can be directed to the soil specimen to raise the specimen temperature to or above 110°C. Commonly used sources include electric, gas, butane or oil-fired stoves, and hotplates, blowtorches, heat lamps, hair driers, space heaters, etc. Heat sources that directly apply open flame to the specimen may cause extreme degra- dation of the specimen along with oxidarion of and depositing of soot in the specimen and should not be used. 7.2 Balances A balance having a minimum capacity of 2 Kg, and meeting the requirements of Specification D 4753 for a balance of 0.1-g readability. 7.3 Specimen Containers-Suitable containets made of ma- terial resistant to corrosion and a change in mass upon repeated heating, cooling, and cleaning. One container is needed for each water content determination. 7.4 Container Handling Apparatus-Gloves or suitable holder for moving hot containers after drying. 7.5 Miscellaneous (as needed) Mixing tools such as spatu- las, spoons, etc.; eye protecrion, sach as safety glasses or goggles; cigarette papers, and lrnives. 8. Hazards 8.1 Container holders or gloves are recommended for han- dling hot containers. Some soil types can retain considerabie heat, and serious bums could result from improper handling. 8.2 Suitable eye protection such as safety glasses or goggles is recommended due to the possibility of particle shattering during heating, miacing, or mass deternunations. 8.3 Highiy organic soils, and soils containing oil or other contaminants may ignite during drying with direct heat sources. Means for smothering flames to prevent operator injury or equipment damage should be available during testing. Fumes given off from contaminated soils or wastes may be toxic, and should be vented accordingly. 8.4 Due to the possibility of steam e�cplosions, or thermal stress shattering of porous or brittle aggregates, a vented covering over the sample container may be appropriate to prevent operator injury or equipment damage. This aiso pre- vents scattering of the test specimen during the drying cycle while aiding in uniform heating of the specimen. 9. Samples 9.1 Perform the water content determinarion as soon as pracrical after sampling to prevent water loss and damage to potentially corrodible containers. 9.2 Prior to testing, store samples in non-conodible airtight containers at a temperature between approximately 3 and 30�C and in an area that prevents direct exposure to sunlight. 10. Test Specimens 10.1 Select a representative portion of the total sample. If a layered soil or more than one soil type is encountered, select an average portion or individual portions of each, and note which portion(s) were tested in the report of the results. 10.1.1 For bulk samples, select the test specimen from the material after it has been thoroughly mixed. The mass of moist material selected shall be in accordance with Table 1. TABLE 1 Test Specimen Masses Sieve Size Retaining More Than Minimum Mass ot 10 % of Sampie, mm Moist Specimen, gA 2A (No. 10) 200 to 300 4.75 (No. 4) 300 to 500 19.0 (No. �/a) 500 to 1000 "Larger specimens may be used and are encouraged. Generally, inherent test inaccuracies are minimized by using speCimens with as large a mass as practicai. �� D 4959 10.1.2 For small samples, select a representative portion in accordance with the fol(owing procedure: 10.1.2.1 For cohesionless soils, mix the material thoroughiy, then select a test specimen having a mass of moist material in accordance with Table L 10.1.2.2 For cohesive soils, remove about 3 mm of material from the exposed periphery of the sample and slice the remaining specimen in half (to check if the material is layered), prior to selecting the test specimen. If the soil is layered, see 10.1. Breaking or cutting of cohesive samples to approximately 6 mm particles speeds drying and prevents crusring or over- heating the surface while drying the interior. 10.2 Using a test specimen smaller than the minimum mass indicated in Table 1 requires discretion, though it may be adequate for the purpose of the test. Note a specimen having a mass less than the previously indicated value in tHe report of results. NorE 3—When woridng with a small sample containing a relatively large coazse-grained particle, it may be appropriate not to include this particle in the test specimen, depending on the use of test results. If this is done, such exclusion should be noted in the report of the results. 10.3 When the result of a water content determinarion by the use of this test method is to be compared to the results of another method, such as Test Method D 2216, obtain a second specimen during selection of the specimen for this comparison. Take precautions to obtain a specimen that represents the same water content as closely as possible. Protect the comparison specimens from water loss by transporting and storing the specimens in sealed containers. A correction factor can be determined for use on subsequent water content deternvnarions on the same soii types from the same site when the difference is relatively constant using several comparisons. Check the correc6on factor on a regular, specified basis. Recognize that different technicians, heat sources, and such may result in different correcrion factors. 11. Conditioning 11.1 Prepare, process, and test all specimens as quickly as possible to minimize unrecorded moisture loss. 11.2 Cut or break up the soil into small size aggregarions to aid in obtaining more uniform drying of the specimen, taking care to avoid any loss of soil. 113 If the specimens are not being tested immediately, place the specimens in containers that can be closed and stored in an area not exposed to direct sunlight, to prevent loss of moisture prior to initial mass determinations. 12. Procedure 12.1 Determine the mass of a clean, dry container, and record. 12.2 Place the soil specimen in the container, and immedi- ately deternune and record the mass of the soil and container. 12.3 Apply heat to the soil specimen and container, taking care to avoid localized overheating. Continue heating white stimng the specimen to obtain even heat distribution. Continue applicarion of heat until the specimen first appears dry. A comparatively uniform color should result. Avoid localized bumt or darkened appearance of any part of the soil by intemuttent mixing and stirring. 123.1 Experience with a particular soil type indicates when shorter or longer initial drying periods can be used without overheating. NmE 4—A piece of dry, light-weight paper or tissue, st�ch as cigarette paper, placed on the surface of the appazently dry soil �vill curl or ripple if the soil still contains significant water. 12.4 After an initial hearing period has been completed (soil appears dry), remove the container and soil from the heat source and cool to aliow handling and prevent damage to the balance. Determine and record the mass of the soil and container. 12.5 Retum the container and soil to the heat source for an additional applicarion of heat. 12.6 With a small spatula or knife, carefully stir and mix the soil, taldng care not to lose any soil. 12.7 Repeat 123 through 12.5 until the change between two consecurive mass deternunations would have an insignificant effect on the calculated water content. A change of 0.1 % or less of the dry mass of the soil for the last two determinations should be acceptabie for most specimens. 12.8 Use the final dry mass determination in calculating the water content. 12.9 When routine testing of similar soils is contemplated, the drying times and number of cycles may be established and correlated for each heat source and used for subsequent determinations. When pre-detemuned drying times and cycles are utilized, periodic verification in accordance with the procedure in 12.7 should be performed to assure that the results of the final dry mass determination are equivalent. 13. Calculation 13.1 Calculate the water content of the soil as follows: w=[(M� — MZ)l(MZ — M�)] X 100 = M, fM� X 100 (1) where: w = water content, %, MI = mass of container and moist specimen, g, M2 = mass of container and dried specimen, g, M� = mass of container, g, Mw = mass of water, g, and MS = mass of solid particles, g. 14. Report 14.1 Report the following information: 14.1.1 Identification of the sample (material) being tested, by locarion (boring number, sample number, test number, etc.), 14.1.2 Water content of the specimen to the nearest i%, 14.13 Indication of the test specimen mass, including a note if less than the minimum indicated in Table 1, 14.1.4 Indication of test specimens containing more than one soil type (layered, and the like), 14.1.5 Indication of any material (size and amount) ex- cluded from the test specimen, 14.1.6 Initial mass of test specimen prior to drying, and the mass after the incremental drying periods, 14.1.7 Identificarion of the type of direct heat source, drying settings, drying times, and number of cycles used, when standardized drying is utilized, and 14.1.8 Identification of comparison test(s) if performed, the q�'i D 4959 method of test utilized and any correcrion factors appiied (see Note 5). No�rE 5—Water content determinations conducted in accordance with Test Method D 2216 or other methods may be recorded on the same report. This is not a mandatory requirement, bui may be convenient when the results of the two methods are to be compared. 15. Precision and Bias 15.1 Precision Test data on precision is not presented due to the nature of the soil materials being tested by this test method. It is not feasible and too costly at this time to have ten or more agencies participate in a round-robin testing program. Also, it is not feasible or too costly to produce multiple specimens that have uniform physical properties. Any variation observed in the data is just as likely to be due to specimen variation as operator or laboratory testing variation. 15.2 The precision of this test method is operator- dependent, and is a function of the care exercised in perform- ing the steps of the procedure, giving particular attention to careful control and systemaric repetition of the procedures used. 15.2.1 Subcommittee D18.08 is seeldng any data from users of this test method that might be used to make a limited statement on precision. 153 Bias=There is no accepted reference value for this test method, therefore, bias cannot be determined. 16. Keywords 16.1 acceptance tests; compaction control; density; direct hearing; laboratory moisture tests; moisture content; moisture control; quality controi; rapid method; soil moisture; test procedure SUNIlVIARY OF CAANGES In accordance with Committee D-18 policy, this section identifies the location of changes to this standard since the last edition (1994) that may impact the use of this standard. (1) Editorially revised the title. (2) Editorially revised 1.2. (3) Revised Secrion 2 to include Practice D 3740. (� Revised Section 5 to include Note 1 referencing Practice D 3740. (S) Revised the numbering of existing notes. (� Revised precision and bias statement to conform to D-18 policy. , (� Added Summary of Changes. The American Sociery for TesGng and Maferials takes no position �especfing the validity of any patent rights asserted in connection with any item menfioned in this standard. Users of this standa�d are ezpress/y advised that delermination of the validity of any such patent rights, and the dsk of ininngement of such rights, are entirety their own responsibiliry. This standard is sub%ect to revision at any Ume by the responsible technical commiBee and must be reviewed every frve years and if not revised, eitherreapp�oved or withdrawn. You� comments are invfted ei�he� forrevision of this standard orforaddifionai standards and should be addressed to ASTM Headquarters. Your commen(s will receive careful aonsideratiort at a meeting of fite responsible technical committee, which you may attend. If you feel that your comments have not received a fai� hearing you should make you� views known to the ASTM Committee on Sfandards, at the address shown below. This standard is copyrighted byASTM, 100 Ba�rHa�bo��nve, PO Box C700, West Conshohocken, PA 19428-1959, United Sfates. Individual reprints (single or multiple copies) of this standard may be obtained by contactlng ASTM at the above address or at 610-832-9585 (phoneJ, 610-832-9555 (fax), o� service@astm.org (e-mail): or through the ASTM website (www.astm.org). t ' r, i �� ' ' � � � ,' QAPP Appendix E August 17, 2000 Revision 1 Appendix E EPA Method 100.1 Analytical Method for the Determination of Asbestos Fibers in Water ANALYTICAI `�ETh'OD FCR �ET�R�Ii`!M'� IGv OF ASBESTOS FI3�RS IN '�1A � cr� ►�y Eric J. Chatfield and �!. Jane �i11on Flectron Optical Laboratory Department of F,pplied Physics Ontario Research Foundation Sheridan Park Research Corr�nunity Mississauga, Ontario, Canada LSK 193 Contract 6$-03-2717 Project Officer J. `�acArthur Long Analytical Chemis*ry 3ranch Environmental Research labora�ory Athens, Georgia 30613 0 ENVIRONMENTA� RESEFRCH L�BORATO�Y OFFICE OF RESEARCH AND DE,IEIOpM�uT U.S. E�VIROMwENTAI, PROTECTIO'v aGE*aCY ATHE�S, GEORGIA 30613 r�vaoouuc e. NATIONAL TECHNICAL fNFORMATION SERVICE U$ GEPip��f�T 0� :G�YI4CE coa��•nr... ... ..... PB83-260471 III IIIIIIIIII!I!Ilillllllllll� IIII III � � �"' TECHNICAi REPORT DATA ; I'1� �cr rfOJ IHz1/v� An�u' n�i Uie !i t:'ici' bi Jurr r�.�n�.lcfinf'1 i�+L'�C.�i NO 2. ]. RECIPIENT'S ACCESSIf>MNC EPA=600L4=84-04 i.._._. _- -. ---1--------------------- �'�H 3 ? 6 0 4% i_--- .-- J 7��E aNl)yUBiiT�E � 5 fiEPORT 047E-----^' Anal tical Method f�r De�ermination of Asbestos Sz—P��tb��-�3— Y 6. P� HFORMING Op<.l�N124TION C�OE Fibers in Water .\t,THJA�:i) � 8. PEfaFORMiNG OR�iAN124TION REPOR7 �.0. Eric J. Chatfield and h1. Jane Dillon 9�FRFpRh1iN� OriG4�1iZATlON NAME ANO 40DRFS5 10. PROGRqAA E�El ENT NO. Departm?nt of Applied Physics CBNCIA Ontaric Research FOU�iddt�Oii .11.CONTRACT'GA4NTN0. — Sheridan Park Research Corrmunity 68-03-2717 Mississauga, Jntario, Canada L5K 1B3 � 12. 5PONSOA�NG ACENCV Nqb1E AND ADOpE55 1J. TYPE OF REPORT ANO PER100 COVERED Environmental Research Laboratory--Athens GA Final, 10/78-9/31 Office of Research and Development 14.SPONSORING AGENCYCODE U.S. Environmental Protecti�n Agency EPA%o00/O1 Athens, Georgia 30613 15. SUPFI.EMENT.>qY N(�TE� 16 ABS'RACT • An analytical method for measurement of asbestos fiber concentration in water samples is described. Initially, the water sample is treated with ozone gas and ultraviolet light to oxidize suspended organic materials. The water sample is then filtered through. a 0.1 riicrometer pore siz� capillary-pore polycarbonate filter, after which the filter i� prepared by carbon extraction replication for examination in a transmis�ion electron microscope (TEM). Fibers ar� classified using selected area electron diffraction and energy dispersive X-ray analysis. Measurement of characteristic features on a recorded and calibrated selected area electron diffrac- tion patt^rn is specified for precise i�+entification of chrysotile. Quantitative determination of the chemical composition, and quantitative interpretation of at leas one calibrated zone azis selected area diffraction pattern a�e specified for precise identification of amphibole. Amphibole identif7catien procedures and generation of the standard reporting format specified for the fiber count resii'ts are achieved using two computer programs �that are integral to the analytical m�thod. ��, KEV WOHDS 4NiJ OOCUMENT ANA[.YSIS ,�, J ^^ DESCHIPTORS b.iDEtdTIPIEPSiQPEN ENOEO TERMS c. COSATI I��rld;Gruup f3, C)I;;TqIBVTION S7ATEMENT 19 SECUPITY CLASS !TItiS Rep��rf1 21. NO OF PAGES UNC�.ASSIFIED 2�7 RELEASE TO PUBI I C �o. secuR�Tr c�Ass �rhrr nax�� za PfiICE UNCLASSIFIED EPA Form 2220•1 (9•7j) j � � - -- - - - -- ._ . .. � . _�- � � Pj �(. � J� i . . � ��.��y .. �---, ii l- COr� l itUl r�UMHkrt : U2t�1,1 b4 )� PA(,,: .F t!y - ACCt55I�� U,�rnE�Ft; ..... ` ...................PBb3—�bU.u71 ii3 - COLLECiiO�v CUUE; 5 00 — �•�ar�aGt�-it.+i Cu�E: WG . F=; 87 — NKC;CESS a�T�� Cu: UU dla-HE�iSTHatlur� FEE: OOuOu ��: d8 — F�k��DUC t r�sa;�a�t�; H L�9 — HECE:IN f irr�E; 1 d9a—LUa�r UuE Uu f: dlt,- I�tA,vS��CiIU;,; ft� B1�A—JUF't/Pk1uR/SUF'Ek: t�ll- HETUKn;S; U �1�'�1—RETUkiJ UAIE.; ci13- F'KuGESSI��G: � Hlv- Pai. P�Tt°�itQ�: u � tlt7— FJF2.'•�/pr�lCE: 1�111 r 818- arnv�U�aCE : 1 1 � 1 ��y- p�d�I(:nitu�v-1: u�.i26 �Zu• P�bLICaTIurv-1: b21• �I�•�IT:�TIu�:: U d23- i'C tlI�v: UOU 82u- SfJCh; OODU t32UA—STUCK TYF't CODES: h EsZS— NaGE5/SNkETS; UUZ76 fi2o- F�C t'i�c 1 Ct CUJt : A 1 S— ti2i— uU"ESfIC �HlGt: Uuuu�Ou b28— fOREIG�� Pt2IGE: OOODUOo a24- 4G1IJ'•i CUUES: tsp d33- ��w PkiCE CU►�t;: AUI fi3a- �u;��tStIG t'ri1GE; OOOuODU d35- fUREIG�� PKlCE: OOOU000 � b3b- ACTLJ�� C;OJES: MA it37• nEL�ASaHtl,i �f CU: C • d38- h:F i�RI��T: A tssy- auUl i!u",a� 1�vFu: n • nuq• PkIr�t NC: � dqi- �C DUE-t�.. _��.. a,,.��n . 64�� SuUKG� UkUEk: n 9yZa-GE,��+EnaTt �ua: � 04�;�-SuNN�iErt SkG CD: � DISCLAIMER A The i�formation in this document has been funded wholly or in part by the United States Environmental Protection Agency under Contract No. 68-03-2717 to Ontario Research Foundation. It has been subject to the Agency's peer and administrative review, and it has been approved for publication as an EPA document. Mention or trade names or conunercial products does not constitute endorsement or recorrffnen- dation for us�. ii --�►----- - - - _ . FOREWORD Nearly every phase of environmental protection depends on a capability to identify and measure specific pollutants in the environment. As part of this laboratory's research on the occurrence, movement, transformation, impact, and control of environmental contaminants, the�,analytical Chemistry Branch develops and assesses new techniques for identifying and measuring chemical constituents of water and soil. A 3-year study was conducted to develop improvements in the analytical method for determination of asbestos fiber concent.rations in water samples. The research produced an improved sample preparation and analysis method- ology, a rapid screen�ng technique to reduce analysis cost, and a new rererence analytical method for asbestos in water. The analytical method for de�ermining asbestos fibers in water is perceived as representing the curre.�t state-of-the-art. William T. Donaldson Acting Director Environmental Research l,aboratory Athens, Georgia iii � � PREFACE The Preliminary Interim Method for Determining A�oestos in Water was issued by the U.S. Environmental Protection Aqency's Environmental Research laboratory in Athens, Georgia. Thz method was based on filtration of the water sample through a sub—micrometer pore size membrane filter, followed by preparation of the filter for direct examination and counting of the fibers in a transmission electron microscope. Two alternative techniques were specified: one in which a ceilutose ester filter was prepared 5y dissolution in a condensation washer; and another known as the carbon—coated Nuclepo►-eR techn�que which used a polycarbonate fil�er. In January 1980 th� method �Nas revised (EPA-600/4-80—Oo5) to eliminate the condensation washer ��?roach, and a suggested statistical treatment of the fiber count data was incorporated. ?he analytical method published here is a furiher refinement oT the re�rised interim method. Major additions include the introduction of ozo�e—ultraviolet light oxidation prior to filtration, complete specification of techniq�.,es to be used for fiber identification and fiber counting rules, and incorporation of reference standard dispersions. A standardized reporting format has also been introduced. The major deletion is �he low temperature ashing technique for samples high in organic material content; ashing is not required for the analysi5 of drinking water and drinking water supplies when samples are treated using the ozone—uTtraviolet oxidation tech�iGue. The "field—of—view" approach for examination also has been deleted from the method. If a sample is too heavily loaded for examination of en.tire grid openings. a more reliable result is obtained by preparation of a new fi!ter using a smaller volume of water. iv aasr�acT .an analyticai �ethod for mea5urement of asbestos fiber concentration in water samples is described. Initialiy, the water samplQ is treated witn ozone gas and �ltraviolet light to oxidize suspended organic materials. The water sample is then fiitered through a O.I �m porP size capillary-pore polycarbonate filter, after which the filter is prepared by carbon extraction replication for examination in a transmi5sion electron microscope ;TE��). Fibers are classified using selected area electron diffraction (SAED1 anJ energy dispersive X-ray anal�sis (EDXA). ��easurement of characteristic features.on a recorded and catibrated SAED pattern is specified for precise identification of chrysotile. Quantitative determinatiun of the chemical composition, and quantitative interpretation of at least one cal;brated zone axis SAED pattern are specified for precise identification of ampnibole. Amphibole identification procedures and generation of the standard reporting format specified for the fiber count results are achieved using two cem�uter programs which are integral to the analytical method. This analytical method is a further development of �he interim method issued in 198C, and incorporates results of research performed under Contract 68-03-2717 under sponsorship of the U.S. Environmental Prote.ction AQency. This report covers a period from October 197E to September I981 and�the work was cc�pleted as of September 2981. 0 �y � CONTENiS FOREWORD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i i i PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i v ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v FIGURES ................................ x TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x i 1. 2. 3. 4. 5. 6. � SCOPE AND APPIICATION . . . . . . . . . . . . . . . . . . . . SUMMARY OF METHOD . . . . • • • • • • • • • oEFINtTtONS, uNiTS ANO�aBBREVIATI0N5� . . . . . . . . . . . . . 3.1 Definitions . . . . . . . . . . . . . . . . . . . . 3.2 Units . . . . . . . . . . . . . . . . . . . . . 3.3 Abbreviations . . . . . . . . . . . . . . . . . . . EQUIPMENT ANO APPARATUS . . . . . . . . . . . . . . . 4.I Specimen Preparation�Laborato�y . . . . . . . . . . 4.2 Instrumentation Requirements . . . . . . . . . . . 4.2.1 Transmission Electron�Microscope . . . . . 4.2.2 Energy Dispersive X-ray Analyzer ..... 4.2.3 Computer . . . . . . . . . . . . . . . . 4.2.4 Yacuum Evaporator . . . . . . . . . . . . . 4.2.5 Ozone Generator . . . . . . . . . . . . . . 4.3 Apparatus, Supplies and Reagents . . . . . . . . . . SAMPLE COILECTION AND PRESERVATION . . . . . . . . . . . . . . 5.1 Sampie Container . . . . . . . . . . . . . . . . . . 5.2 Sample Collection . . . . . . . . . . . . . . . . 5.3 Quantity of Sample . . . . . . . . . . . . . . . . . 5.4 Sample Preservation and Storage . . . . . . . . . . PROCEOURE . . . . . . . . . . . . . . . . . . . . . . . � 6.1 Cleanliness and Contamination Cantrol . . . . . . . 6.2 Oxidation of Organics . . . . . . . . . . . . . . . 6.3 Filtration . . . . . . . . . . . . . . . . . . . . . 6.3.1 General . . . . . . . . . . . . . . . . 6.3.2 Filtration Procedure . . . . . . . . 6.4 Preparation of Electron Microscope•Grids ...... 6.4.1 Preparation of Jaffe Washer .. . . 6.4.2 Selec�ion of Filter �rea for Carbon•Coating. 6.4.3 Cerbon Coating of the Nucleoore Filter ... 6.4.4 Tran;fer of the Filter to Electron Microscope Grids . . . . . . . . . . . . 6.5 Examination by Electron Microscop�/ . . . . . . . 6.5.1 Microscope Atiqnment and Magnification Calibration . . . . . . . . . . . . . . vii / 1 1 2 4 5 5 6 6 8 8 9 9 9 15 15 16 16 lb 17 17 17 20 20 22 23 24 24 26 27 28 28 7. 10. 6.5.2 Calibration of EDXA System 6.5.3 Grid Preparation Acceptability � � � � � 6.5.a ?rocedure for Fiber CounYing . . . . � � 6.5.5 Estimation of Mass ConcPntration 6.6 Fiber Counting Criteria . . . . . � � � 6.6.1 Fiber Counting Method , . , � � � � � � � 6.6.2 Fibers �ihich Tour.h Grid Bars �� 6.6,3 Fibers Which Extend Outs�de the.Field�of �� 'Jiew . . . . . 6.6.4 Fibers with Stepoed�Sides. . . � � � � � 6.6.5 Fiber Bundles . � � � � � � � , 6.6.6 Aggreqates of Randomly Oriented�Fihers .�� 6.6.7 Fibers Attached to Non-Fsbrous Oebris �� 6.7 Fiber ldentification Pr�cedures � � 6.1.1 General . . , , , , , , , , � � 6.7.2 SAED and E��A�Techniques . � � 6.7.3 Analysis of Fiber :dentification �ata� � � � 6.7.4 Fiber Classification Categories � � � 6.7.5 �rocedure for Classification of Fibers '�1ith� Tubular Horphology, Suspected to be Chrysotile . . , . 6.7.6 Procedure for Classification of�Fibers � � Without Tubular Morqhology, Suspected to be Amphibole . , , , 6.8 Blank and Control Determinations � � � � � � � . . . . . . . . . 6.8.1 Blank Determinations . . � 6.8.2 Control Samples . � � � � � � � � � CALCU�P.TION OF RESULTS . � � � � � � � � � � � ' ' 7.1 Test for Unifor�nity of Fiber•Oeposit•on Electron� � Microscope Grids . . . %.2 Calculation of the `�ean and Confidence.:nterval �f• � the Fiber Concentration , , 1.3 Estimated Mass Concentration . � � � � � � � � � � 7.4 fiber Length, '�Jidth, �iass and Aspect�Ratio. � � � � � QiStributions . . . . 1.4.1 Fiber Length Cumulative�Numher Oistribution. 7.4.2 Fiber Width Cumulative Number Distribution . 7.4.3 Fiber Length Cumulative Mass Distribution 7.4.4 Fiber AsAect Ratio Cumulative Number � Distribution 7.4.5 Fiher Mass Cumulative.P�umber Distribution � 7.5 ;ndex of Fibrosity ' REPORTING � � � � � � � � ' ' ' ' ' ' ' ' ' LIMITATIONS OF�A�CURACY � � � � � � � � � � � � � � � � ' ' ' ' 9.1 Errors and Limitations.of tdentification� �� 9.2 Obscuration . . . . . . . � � � � � ' 9.3 Inadequate DispersTon . . � � � � � � � � � ' ' 9.4 Contamination , . � � � � � � � � � ' ' ' ' 9. 5 Freez i ng . . . . . . . • . . . . . . . • . . • . PRECISION AND ACCURACY� , . . � � � � � � � � ' ' ' ' ' I0.1 Gcnera 1 . . . . . . . , ' . . . . . . . . . . . . . 10.2 PreCisio�� . . . . . . . . . . . � . . . . . . . . viii 29 30 "sl 34 35 35 35 36 36 37 37 33 ?a 38 39 43 45 45 50 54 54 54 55 55 56 59 61 6I 62 62 62 53 63 E3 65 55 66 66 66 66 67 67 67 � � 10.2.1 Intra-Laboratory Comparison Using Environmental Water Sources . . . . 67 10.2.2 Inter-laboratory Comparison of Filters Prepared Using Standard Oispersions and Environme�tal 'rlater Sources . . . . 67 10.3 Accuracy . . . . . . . . . . . . . . . . 71 1C.3.1 Intra- and Inter-Laboratory Comparison of Standard Dispersions of Asbestos f i bers . . . . . . . . . . . . . . . . . 71 SELECTED BI6lI0GRAPHY . . . . . . . . . . . . . . . . . . . . . . . 74 APPENDIX A- TEST DATA AND COMPUTER LISTINGS - FOR FIBER IDENTIFICATION . . . . . . . . . . . . . . 77 APPENO[X B- TEST DATA ANO COMPUTER LISTINGS FOR pATA PROCESSIP�G ANO REPORTING. . . . . . . . . . 176 Nurrb er l. 2. 4. Sa. 58. 6. 7, 8. 9. 10. 11. I2. 13. 14. 15. 16A. 168. 17. 18. s FIGURES ?dQp Calibration Markinqs on TEM Viewing Screen . . . . . . . . . . . 1 Diagram of Ozone—UV Equipment . . .,., lg Ozone—U'/ Oxidation of Water Samples in Glass Bottles� ... �� 19 'uclepore Oissolution Technique , , , , , , , , , , , , , , , , ?:� Jaffe '�lasher Oesign . . . . . . . . . . . . . . . . . . . 25 Jaffe Washer in Use . . . . . . . . . . . . . . . . 25 Condensation '�lasher . . . , . , . , , , , , , , � � � �3 Sheet For Recording Water Sam41e Oata .... �� 32 Sheet For Re�ording Fiber Classificatio�i and Measurement��ata .� 33 Counting of Fibers Which Overlap Grid Bars . . . . . . . . . . 35 Counting of Fibers Which Extend Outside the field o" View . 36 Counting and Measurement of Fiber Bundles ...... 37 Counting of Fiber Agqregates . . . . � � � � � 37 Counting and Measurement of Fibers.Attached to•Von—Fibrous. •�� Debri s . . . . . . . . . . . . 38 �"�easurement of�Zone Axis•SAED Patterns . . . . . . � � 41 Ctassificat�on Chart for Fiber With Tubular '4orphology. �� 47 TEM Microqraph of Chrysotile Fibril, showing Morphoiogy .�� 4g TEM +licrograoh of UICC Canadian Chr•ysotile Fiher 3fter Ther�nal � Deqradation by clectron 3eam Irradiation .... . �q SAEO Pattern of Chrysotile Fiber •,�ith Oi�gnosric Features Labelled.50 ClasSification Chart for Fiber �lithout Tubular �Horphology .... 52 x L � w � Number I. 2. 3. 4. 5. 6. 7. 8. 9. 10. TABLES Paoe Limitation of Analytical Sensitivity by VoTume of"Water Sample Fi 1 tered . . . . . . . . . . . . . . . . . . . . . . . . 21 Sflicate Mineral�Standards . . . . . . . . . . . . . . 29 Classification of Fibers With Tubular �`torphology ........ 46 Classificat1on of Fibers Without Tubular t�orphology ...... 46 Levels of Analysis for Amphibole . . . . . . . . . 53 Intra-Laboratory Comparison of Environmental�Water Sampies ... 68 Inter-laboratory Comparison: StanCard Dispersions . . . . . . . 69 Inter-Laboratory Comparison: Environmental Water Samples ... �0 Inter- ar.d Intra-Laboratory Comparison: Chrysotile ...... 72 Inter- and Intra-Laboratory Comparison: Grocidolite ...... 73 xi � ' ' _ �' "f_�_LL.�—�� y_tww�.___ - ,�� ..:L4.`�iiY:a+�� i..'9r` W�' _ � _ _ AN��YTICAI. METH00 FOR OETERMINA7ION OF ASBESTOS FIBERS IN ;�ATER 1. SCOPE AND APP�ICATION � 1.1 This method is applicable to drinking water and drinking water suoplies, and should he used when the bes? available analytical procedure is required. I.2 The method determines the numerical concentration of asbestos fibers, the length and width of each fiber, and the estimated mass concentration of asbestos in the water. Fiber size and aspect ratio distributions are also determined. • 1.3 The method permits, if required, identification of alt mineral fibers found in water. In particular, chrysotile can be distinguished from the amphiboles, and fibers of specific amphiboles can be identified. 1.4 The analyti.:al sensitivity which can be achieved depends primarily on the amount of other particulate matter which is present i� the sample. This limits the proportion of the sample which can be mounted for examination in the electron microscope. In drinkina water which meets the �:J'�JA turCidity criterion of 0.1 yTJ� an asbestos concentraLion of O.OI miilion fibers per liter (�1F;.) can be detected. The contamination 1eve1 in ti�e taboratory environment may degrade the sensitivity. 7he analytical ;ensitivity for the determination of mass concentration is a function of the preceding parartteters and also depends on the si_e distribution of the fibers. In low turbidity drinking �Nater the analytical sensitivity is usually of the order of 0.1 nanogram per liter (ng/L). 1.5 It is beyond the scope of this document to provide detailed instruction in electron microscopy, electron diffraction, crystailograohy or X—ray fluorescence techniques. It is assumed that those performing this anatysis will be sufficiently knoKledgeable in these fields to understand the soecialized techniques involved. 2. SUMMARY OF METH00 Water collected in a polyethylene or qlass container is treated with ozone and ultraviolet light to oxidize organic matter. After mil� ultra5ound treatment to disoerse the `ibers uniformly, a�ncwn volume of the water is filtered through a 0.1 ,�nic��meter (�m) pore size '" �ucTeporeR polycarbonate filter. A carbon coating is then apoiied in vacuum Lo the active surface of the fitter. The carbon layer coats and retains in position the material which has been collected on the filter "� surface. A smaJl portion of the carbon—Coated filter is placed on an . electron microscope grid and the polycarbonate filter matP��al is removed _ by disso)ution in an organic solvent. The carbon film co.�taining the original perticulate, supported on the electron microscope grid, is then ," � examined in a transmission electron microscope (TEM) at a magnification of about 20,000. In the TEM, selected area electron diffr�ction (SAE�) is used to examine the crystal strutture of a fiber, and its elemental ` composition is determined by energy dispersive X—ray analysis (EDXA). Fibers are tlas�ified according to the techniques which have been used *o identify them. A simple code is used to record for each fiber the degree to which the identification attempt was successful. The fiber classification procedure �is based on successive inspection of the morphology, the selected area electron d�ffraction pattern� and the qualitative and quantitative energy disRersive X—ray analyses. Confirmation of the iderttification of chrysotile is only by quantitative SAEO. and confirmation of amphibole is only by quantitative EDXA and quantitative zone axis SAED. Several levels of anaiysis are specified, three for chrysotile and four for amphibole� defined by the most specific fiber classification to be attetnpted for all fibers. The procedure permits this target classification to be defined on the basis of previous kno��ledge, or lack of it, about the particular sample. Attempts are then made to raise the classification of all fibers to this target classification, and to record the degree of success in each case. 7he lengths and widths of atl identifled fibers are recorded. The number of fibers found on a known area of the microsco�e sampie, together with the equivalent volume of water `iitered chrough this area, are used to calculate the fiber toncentration in MfL. The mass concentration is ralculated in a similar manner by surtmation of the volume of the identified fibers, assuming their density to be that of the bulk material. �. OEFINITIONS, UNITS AND ABBREVIATIONS 3.1 Definitions Acicular — The shape shown by an extremely slender crystal with smali cross—sectional dimensions. Amphibole — A group of rock—forming ferromagnesian silicate minerals, closely related in crystal form and composition and having the general formula: A2_385(Si,A1�8022(OH)z, where A� Mg, Fe+2, Ca, Na or K, and B� Mg, Fe+z�Fe+3 or A1. Some of these eleme�ts may also be substituted by Mn, Cr, Li, Pb, Ti or tn. It is characterized by a cross—iinked double chain of Si-0 tetrahedra with a silican:oxyqen ratio of 4:11, by columnar or fibrous prismatic crysta�s and by good prismatic cleavage in r two directions parallel to the crystal faces and intersecting at angles of about 56° and 124°. Amphibole Asbestos - Amphibole in an asbestiform habit. °� , Analytical Sensitivity - The calcuTated concentration in �L equivalent to countinq of one fiber. � Asbestos - A commercial term applied to a group of silicate minerals that readily separate into thin, strong fibers that are fiexible, heat resistant and chemically inert. Aspect Ratio - The ratio of length to w��th in a particle. Camera length - The equivalent projection length bet�een the sample and its electron diffraction pattern, in the absence of lens action. Chrysotile - A mineral of the serpentine group: Mg3S�z05(OH)4, It is a highly fibr�us, silky var��ety of serpent�ne, and constitutPs the most important type of asbestos. Cleavage - The breaking of a mineral along its crystallo9raphic planes, thus reflecting crystal structure. Cleavage rragment - A fraqment of a crystal that is bounded by � cleavage faces. d-Spacing - The separation between identical adjacent and parallel planes of atoms in a crystal. Diatom - A microscopic, sinqle-celled ptant of the class Bacillariophyceae, which grows in both marine and fresh water. Oiatoms secrete walls of silica, called frustules, in a great variety of fonns. Electron Scattering Power - The extent to which a thin layer of a substance scatters electrons from their original path directions. Energy Dispersive lf-ray Analysis - Measurement of the energies and intensities of X-rays by use of a solid state detector and multichannel analyzer system. Eucentric - The condition when an object is placed with its center on a rotation ur tilting axis. Fibril - A single fiber, which cannot be seGarated into smaller d components without losing its fibro��s properties or appearances. � 3 � r� Fiber - A particle which has parallal or stepped sides, an aspect ratio equal to or yreater than 3:1� and is greater than 0.5 um in length. Fiber Aggregate - An assembly of randomly oriented fibers. � Fiber Bundle - A fiber composed of parallel, smaller diameter fibers attaehed along their lengths. � Habit - Th? characteristic crystal form or combination of forms of a mineral, including characteristic irregularities. Miller Index - A spt of three or four integer numbers �sed to specify the orientation of a crystallographic plane in relation to the crystal axes. Replication - A procedure in electron microscopy specimen preparatian in which a thin copy, or replica, of a surface is made. Selected Area Electron Diffraction - A technique in electron microscopy in which the crystal structure of a small area of a sample may be examined. Serpentine - A roup of common rock-forming minerals having the . formula: �Mg,Fe)3 Si205(OH)4. Unopened Fiber - Large diameter asbestos fiber which has not been separated into its constituent fibrils. Zone A�cis - That line or crystallographic direction through the center of a crystal which is parallel to the intersection edges of t�e r.rystaT f aces defining the crystal zone. 3.2 Units eV - electron volt 3 glcm - grams per cubic centimeter kV - kiiovolt ug/L - mfcrograms per liter (10-6 grams per liter) um - micrometer (10-6 meter) Mfl - Million Fibers per Liter ng/L - nanograms per liter (10�9 grams per liter} nm - nanometer (IO�g meter) 4 NTU - yephelometric Turbi�ity Unit pPm - parts per million 3.3 Abbreviations `� � AWWa - American Water Works Association EDXA - Energy Dispersive X-ray Analysis '� HEPA - High Efficiency Particle Absolute SAEO - Selected Area Electron Diffraction 5EM � - Scanning Flectron Microscope STEM - Scanning Transmission Electron Microscepe TEM - Transmission Electron Microscope UICC - Union InternationaTe Contre le Cancer (internatio�al Union Against Cancer) UV - Ultraviolet 4. EQUIPME�vT AND APPARATUS 4.1 Specimen Preparation Laboratory Asbestos, oarticulariy chrysotile. �s present in small quantities in practically all laboratory reagents. Many building materials also contain significant amounts of asbestos or other mineral fibers which may interfere with analysis. It is therefore essentiai that all specimen preparation steps be performed in an environment where contaminaLion of the sample is minimized. The primary requirement of the sample preparation laboratory is that a btank determination using knvwn fiber-free water must yield a result which will meet the requirements specified in Section 6.8.I. Preparation of samples should be carried out only af ter acceptable blank values have been demonstrated. The sample preparation areas should be a separate clean room with no asbestos-containing materials such as flooring, ceiling tiles, fnsulation and heat-resistant products. The work surfaces should be stainless steel or plastic-laminate. The room should be operated under oositive pressure and have absolute (HEP.A) filters� electrostatic precipitation, or� eauivalent, in the air supply, .q lamfnar flow hood is recommenGed for samp)e manipulation. It is recortanended that a suppty of disposable labor�tory coats and � disposable overshoes be obtained to be worn in th�� clean room. � This will reduce the levels of dust� and particutarly asbestos, which might be transferred inadvertently by the operator into the � � � clean area. Normal electrical and water services are required. An �. air extract (fume hood) is required to remove surplus ozone from the area near the ozone generator. � 4.2 Instrumentation Requirements 4.2.1 Transmission Ele� cron Microscope A transmission electron microscope having�an accelerat:ng potential of a minimum of 80 kV, a resolution better than 1.0 nm� and a magnification range of 300 to 100,000 is required. The ability to obtain a direct screen magnifi— cation of at least 20,000 is necessary. An overall magnification of about 100,000 is necessary for inspection of fiber morphology; this magnification may be obtained by supplementary optical enlargement of the screen image by use of a binocular if it cannot be obtained directly. It is also required that the viewing screen be calibrated.(as shown in Figure 1) w9th concentric circles and a millimeter scale such that the lengths and widths of fiber images down to 1 mm width can be measured in increments of 1 nan. .� for Bragg anqles less than 0.01 radians the instrument must be capable of performi�g selected area electron diffraction from an area of 0.6 um or iess, selected from an in—focus image at a screen ma�nification of 20,Q00. This perfornr ance requirement defines the minimum separation between particles at which independent diffraction patterns car. be � obtained from each. The capability of a particular instrument may normally be calculated using the foltowing relationship: A where: A D M CS 9 _ � (M + 2000 CS93)2 � Effective SAED area in �m2 � Diameter of SAED aperture in um � Magnification of objective lens � Objective lens spherical aberration coeffitient in rtm . Maximum Bragg angle in radians � Although almost all instruments of current manufacture �eet these requireme�ts, many older instruments which are sti11 in service do not. It is obviously not possible to reduce the area of analysis indefinitely by use of apertures 0 � / 6 \ .. \\\ \\S � �\\ �� \\ ° ��, � �\ � \\ \ \ \ / \ \ `. 3 \ \\\•.\ � , 2 �\ .�\ \ \ � `�\, �� \, '� / `� �,` ,�\ '�1 i �� ,, ,,,� , \ �\ �� ` ,, � �; ; , . ; �; i .' � � ' �� i (�� • 3� Z ,,,,, „� � � � � i I ' z � •, i I � � � ! � � i3 ° jS �6 � 7 I � � , \ � � i � i � ! � t � � ' \ � � � � / ' • � \ `\ ,\\ 1\\� \� i / / . / �• ' � / / / • . j' . ' � � t � /' ,� i / Figure 1. Calibration �harkings on TEM Viewing Screen. 7 � r' V � y _. . - . .. . . _ . . .- .. _ .. � . - .�.�-�.�.a.� smaller in diameter than those soecified by the manufacturer, since there is a fundamentai limitation imposed by the sRherical.aberration coefficient of the �� objective lens. If zone axis SAED analyses are to be performed, it is required that the electron microscope be fitted with a goniometer stage which permits either a 360° rotation combined v+ith tilting througF� at lea�t +30° to -�0°, or tilting through at least +30 to -30 around two perpendicular axes in the plane of the sample. The work is greatly facilitated if the goniometer permits eucentric tilting. It is also essential that the electron microscope have an illuminatian and condenser lens system capable cf forminq an electron probe smaller than 100 nm in diameter. Use of an anti-contamination trap around the specimen is recommended if the required instrumental performance is to be obtained. 4.2.2 Energy Dispersive X-ray'Analyzer An energy dispersive X-ray analyzer is required. Since the performance of individual combinations of equipment is critically dependent on a number of geometrical factors, the req��ired performance oF the combination of electron microscope and X-ray analyzer is specified in terms of the measured X-ray intensity from a small diameter fiber, using a known electron beam diameter. X-ray detectors are generally least sensitive in the low energy region, and so measurement of sodium in crocidoli�e is selected as the performance criterion. The combination of electron microscope and X-ray analyzer must yield a ba�kground- subtracted NaKa peak integral count rate of more than 1 count per second (cps7 from a�0 nm diameter fiber of UICC crocidolite irradiated by a 1G0 r.m diameter electron probe at an acce]erating potential of 80 kV. The eq��ivalent peak/background ratio should exceed 1.0. The EDXA equipment must prov+.de the means for sub�raction of the background, identification of elemental oeaks, and calculation of net peak areas. 4.2.3 Computer ^ Many repetitive m�merical calculations are necassary, and these can be performed conveniently by relatively simple Cemputer Nrograms. For analy5e5 of zone dxis diffrattion pattern measurements, a compute� facilit; with minimum available memory of 64K words is reQuired to accommodate � 8 . : -. . � ...� -. . . ._ . „-'--�'-�-----�� ---.,--.�----� the more complex programs involved. Suggested orcgram listings for standardized data reporting and fiber identification routines are inclu�2d as part of this analytical procedure. (Appendices A and 6). �� 4.2:4 Vacuum Evaporator A vacuum evaporator capable of producing a vacuum better than 10-4 Torr (0.013 Pa) is required for vacuum deposition of carbon on to the polycarbonate filters. A sample holder is desirable which allows a 51 x 75 mm glass microscope slide to be tilted and ro`ated during the coating procedure. Use of �'iquid nitrogen cold trap abovQ the diffusion pur�p will minimize the possibility of contamination of the filter surfaces by oil from the pumping system. The vacuum evapo�ator may also be used for deqosition of the thin film of �old, or other reference material, required on e?ectron microscope samples for catibration of electron diffraction patterns. For gold deposition� a sputter coater may allow better control of the process, and is therefore recommended. 4.2.5 Ozor,e Generator An�ozone generator, in Combination with ultraviolet light irradiation, is used for the oxidation of orqanic material in water samples. This orocedure �s necessary on all water samples. The generator should be capable of generating at least 400 g of ozone p��r day at a concen+.ration of at least l� by weight when sup•�tied wi�h dr� oxygen, ihe ozone generator Modei GL-1 (PCI Ozone Corporation, 2 Fairfield Crescent, West Ca;dwe'1, New Jersey 07006) or equivalent has been found to meet the requirements of this analytical technique. � 4.3 Apparatus� Supplies and Reagents 4.3.1 Gas Supply to Ozone Generator The ozone generator can be supolied by either compressed air or oxygen. The input gas must be regulated to the pressure specified by the generator manufacturer. It is recommended that oxygen be provided in order to reduce the possibility of acid formation in the sample. 4.3.2 Gas-Line Orying Tube The ozone generator operates more efficiently when supplied with dry axygen. An in-line drying tube. filled with a desiccant, followed by a 0.2 �m pore size polytetra- fluoroethylene filter to prevent particulate from the desiccant entering the ozone generator is recorrmended. ., 6.5.4 Procedure for Fiber Countinq The number of fibers to be counted depends on the statistical precision desired. In the absence of fibers, the area of the electron microscooe grids ��hich must be examined deRends on the analytical sensitivity required. For statistical reasons, d;scussed in Section 7.2, the fibers on a minimum of 4 orid openings must be counted. The prscision of the fiber ccunt depends not oniy on the total ntimber of fibers counted, hut als� on their uniformit,y from one grid ope�ing to the next. In practice, it has been found that terminat'on of the fi5er count at a minimum of 100 fibers or 20 grid openings, Nhichever occurs first, yield5 results which usually require no further refinement. Additional fiber countiny will be necessary if greater G�ecision is required. At least three grids prepared from the filter must be used in the fiber count. Several grid ooenings are to be ; selected from each grid, and the data are all incorporated in the calculation of the results. This permits the ° measurements to be spread across a diameter of �he original filter, so that any gross deviations from a uniform deposition of fibers should be detected. Figures 7 and 8 show specimen fiber counting raw data sheets which represent the minimum standard of data reportinq for this analytical procedure. Figure 7 shcws page 1 of the raw da:a t�bulation, ��hich contains all specimen preoaration details. Figure 8 is a continuation sheet for the fiber classification ard �ezsurement data; several of these sheets may be required for ana�ysis of a sample. Select a typical qrid opening from one of the grids. Set the magnification to the calibrated higher value (abo��t 20,000). Adjust the sample height until the featur�s in the center of the screen are at the eucentric point. Check that the goniometer tiit is set at Iero. Reduce the maanification to the lower calibrated value of about 2,000. �easure hoth dimensions �f the grid opening imag� in millimeters, using the markings on the fluarescent screen. In columns 1 and 2 specify the se4uential number of the grid opening, and its dimensions. 7hese t�No columns are not used aqain unt��l fiber counting is commenced in the next grid opening to be examinea. adjust the magnificati�n to the upper calibrated vatue, close to 20,000, �nd position the grid o�ening So that one corner is visible on the screen. Move �he imaqe by adjustment of only one translation control, careful'�• examining •'.he sample for fibers, ��ntil the opposite side of the ooening is encountered. Hove the imaqe by one screen •�idth ��sing the 31 ` " Q ... f � .... .,.._.sil: peaks and about 130 eV wide. Compute the ra�io of the peak area for each specified element relative to the peak area for silicon. Repeat the procedure for about 20 particles �' of each mineral standard. Analyses of any obvious forei9n particles should be rejected, and the data from any one standard shouid be reasonabty s��lf-consistent. Calculate thQ arithmetic mean peak area ratios for each specified � element of each mineral standard. These values are required initially as input for the fiber identification program, and apart from occasional routine checks to ensure that there has been no degradation of the detector resolution, the calibration need not be repeated unless there has been a change of instrumental operating conditions. � 6.5.3 �rid Preparation Acceptability Inser*. the specimen grid into the electron microscope and adjust the magnification to a value sufficiently iow (300 - 1000) so �tiat complete qrid openings can be inspected. Examine at ieast 10 grid openings to evaluate the fiber and total particulate loadings, the uniformity of the particulate depos�t, and the Pxtent to which the carbon film is unbroken. T!�e grid must be rejected from further analysis if; a) the grid is too heavily loaded with fibers to perf orm an accurate count. RccuraFe counts cannot be performed if the grid has mo�e than about 50 fibers per grid opening. A new grid pr�paration must be made using either a smaller volume of w�ter or a suitable volume of the water diluted with fiber-free water; b) the overall distribution of the deposited d�bris 15 noticeably non-uniform. A new qrid preparation must be made, paying particular attenti�n to proper particulate dispersal and filtration procedures; c) the grid is too heavily loaded with debris to allow examination of individual particles by SAEO and EDXA. A new grid preparation must be made using either a smaller volume of water or a ditution of the original water sampTe;. d) a large proportion of the grid ope�ings have br�ken carbon film. Since the breakaqe is usually more frequent in areas of heavy deposit, counting of the intact openings could lead to biased results. Therefore, a new grid preparation must be made from a more completely dispersed sample, a reduced volume of Sample, or alternatively. a thicker carbon film �nay be necessary to support the larger particles. 30 b.5.2 Calibration of EDX� S�stem The purpose of the calibration is to enable auantitat:ve composition data, at an accuracy of about 10`= of the eTemental Concentrdtion, to be obtained from EOXA spe�tr� of silicate minerals involving the elements sodium, . � magnesium, aluminum, silicon, potassium, calcium, mancar.ese and iron. If quar��itative det�rminations ar� reGuired �or minerals containing other elements, suitable caliCration information may be incorporated in the computer analysis. The well-characterized standarc�s recommended �er�it � calibration of any TEM-EOXA combinatian which meets the instrumental sDecifications of Section 4.2, so that data from different instruments can be compared. The standar��s used for talibration, and tne elements which thpy represent, are shown in Table 2. TABLE 2. SILICATE ��1INERAL STANOAROS Elements � Min?ral Standard Na, Fe, Si Riebeckite Mg, Si Chrysotile A1, Si Na�loysite K, Si Phlogo�ite Ca, Si '�lollastonite M�, Si ( Bustamite The compositions of these standards have been determined by microprobe analysis, and the TE�4� qrids were prepared from fragments cf the same selected mineral specim�ns. They permit the corrputer Froqram of Appendix A to 5e used with any TEM-EDXA system. Place the first qrid into the microscope, forn an inaae at the calibrated hiQher maanification of about 20,�00, ard adjust the specimen height to the eucentric poir.t. Tilt the specimen towards the X-ray detector as reouired by `_he instrument geometry. Select an isolated fiber or particl� less than 0.5 �.,m in width, and accumulate an E�XA soectrum usinq an electron orobe of suitable diameter. �Jhen a wel' defined spectrum has been obtained� oerfo r•n an aopro�ria:e baCk�round SubtrdCtion and obtain the net oeak areas for each element listed, using ener�y windows centered on the 29 � � ,r_.". _�. _ _.__. _ .- -- �- - - -- . _ _ . .. .—.r..._._._. _ .. _ .. _ s --� - CONOENSER SPECIMEN � COLO wATER SOURCE ADAPTER COlO FINGER ---�, WATER DRAIN THERMOSTATICA�IY CONTROLLED HEATING MANT�E Figure 6. Condensation Washer. 6.5 Exami�ation by Electron Microscopy 6.5.1 Microscope alignment and Maqnification Calibration • Align the electron microscnpe accordinq to the specifications of the manufacturer. Initially, and at regular intervals, carry out a calibration of the two magnifications used fo� the anatysis (aooroximately 2Q,OG0 and 2,000) using a diffraction gratino replica. The catibration shoutd always be re�eated after any , instrumental maintenance or change of operatinq conditions. The maqnification of the screen image is not ;he same as that obtained on photographit plates or film. The ratio bet�reen these is usually a constant value for the instrument. It is most important that hefore the maanification calibra!ion is carried out the sa��le heiaht is adjusted so that the sample is in the eucentric position. �•28 6.4.4 Transfer of the Filter to E'ectron '�icrosc�oe �ri�5 Remove the glass slide carryinq the filter strios f,•�� the evaporator, and using the tecr,nique des�ribed in 5.a.2 cu� four oieces slightly less th�n abcut 3,m� x 3 mrr in sizp from each filter strip. The souare of filter should Fit within the circumference of an electr�n micrr,scooe grid. Three cf the fitter pieces are to be orA�.ir�d on 200 mesh ; copper grids, and un?ess the an3l�sis is to be for chrysotile only, a fcurth piece should be �repare� on a 20� mesh gold grid. Tne specimens prepared on c000er grids are used f�r fiber counting and most EOXA er.amir.ations. The preGaration on the gold arid �s intende� `or EDXA •�ork on `ibers containing sodium. Place a piece ��f t�e carbon-coated fi�ter, carbon s�de �:o, on to the shiny Side of an e'.ectron ^irroscope gri��ng fine tweezers, ��ck up the orid and �ilter together ard place quick;y on to the �hloroform-saturated ;ens tissue in the Jaffe '�asher, as shown in Fiqure SE. It is �mportant that the sampi� be placed on the t�ns tissue quickly, since hesitation whil� the sample is exposed to chlorororm vapor will cause it to curl. This is a simplified technioue which does not �nvolve drop�inq of chl�roform on to the samples. Some ccmponents of the �olycarbon�:te fil�ers now avai?able dissolve in chl�ro`orm cnly very slowly. ConsPaue�t)y, tne grids rr.��st 5e left in the Jaffe '�lasher for lonoer than 4 days, and the solvent r�ust he replaced every day. Depend'ng on the particutar 'ot number �f tne fi�tzrs, even :his period may be instrff,cient to y»ld sat'sfactorv qrids clear of undissotv�d olastic. In this ev�nt, or if a more rapid samp'e preoaration is desired, after a ninimum period of 3G minutes in the laffe Washer the lens paper supporting the grids mey be transferred to the condensation washer as iliustrated in Figure 6. The condensation washer should ther. be op�rated for ;, period of between 30 and 60.minutes, after which the ;rids will have been clear?d of residual pTastic. The rate of condensation in the washer is not critical, orovided that chloroform drias rapidly from the cold finger for the �Nhole of the washing period and the condensation level is above the samples. �urinq the d?Ssolut�on, it is reCommend?d that tnP orids not be al�owed to dry since this has been found t� oreatly increase the time re�uired for Complete dissolution of the polycarbonat2. 27 the plastic petri dish. Press the scalpel point on the filter at the beginning of the desired cut, and rock the blade downwards while maintaining pressure. It will be found that a clean cut is obtained without stressing of the filter. The process should be repeated alonv atl four directions to remove 3 rectangutar portion from the active filtration area of the filter. This filter portion should be selected from along a diameter of the filter, and should be about 3 mm wide by� a minimum of 15 mm lonq. Areas close to the perimeter of the active filtration area should be avoided. 6.4,3 Carbon Coating of the Nuclepore Filt�r The ends of the selec*.ed filter strips should be attacha� to a glass microscope slide using double-sided adhesive tape. This must be performed carefully to ensure that the filter strips lie flat on the slide and are not stretched. the filter strips can be identified by using a wax pencil on the glass slide. Af*.er inserting the necked carbon rods into the vacuum evapor�:or, place the qlass slide on the sample rotation and tilting device. The separation between the sample and the tips of the carbon rods should be about 1.5 cm to 10 cm. If desired, the amount of carbon to be evaporated can be monitored instrumentally so that a thickness of about 30 nR to 50 nm is deposited on the filter strips. Alternatively, a porcelain fragment will serve as a simple carbon deposition monitor. Place a small drep of silicone diff usion.pump oil on the surface of a clean fragment of white glazed porcelain. Locate the porcelain in the evaporation chamber with the oil droplet towards the carbon rod5 and at a distance from the carbon rods eaual to that separating the rods from the filter strips. Carbon will not deposit on the oil drop whereas it does on tFe other areas of the porcelain. With experience, the correct thickness can be monitored visually by observation of the contrast between the darkened areas of the porcelain and the unc�ated areas under the oil drop. Pum� down the evaporation chamber to a vacuum better than ��- Torr (0.013 Pa). Use of a liquid nitrogen cold trap above the diffusion pump wi�l minimize the possibility of contamination of the filter surfaces by oii from the pumping system. Continuously rotate and tilt the olass slide holding the filter strips, whi�e the carbon is evaporated in intermittent bursts, allowing the rods to cool between each eveporation. This proc�dure is necessary to avoid overheatina of the filter strips. Overheating tends to cross-link the polycarbonate which Lhen becomes difficult to dissolve in chloroform. 26 G1.A55 PE'RI DISH --� ELECTRON MICROSCOPE 1 IOOmm x 15mm 1 SPECIMENS . ST STcEL MESH . BR�OGE ! 50mes�1 � . � � ,. � :•:� - .s'�'r /{} (R ,� �.1.�_ , �y '� .i '" C� I J 25 P4RiiclES -C.iH80N � ' �_OOT NG ,�� 1 �O�rC�RBCtu_TE - �,��R � CLEi.TAON� � a N�CAOSCOpE ;,R�O 6.4.1 � CHLOROFORM Figure 4. Nuclepore D9ssolution Technique Preparation of Jaffe Washer f-- G�a80N ...��__�- U `- Gptp Prepare th� Jaffe '+lasher as illustrated in Fiaure 5A. The staintess steel mesh is formed into a bridoe slightty less than 1 cm high, and placed in a 10 cm diameter glass oetri dish with a tight fitting lid. A narrow strip of lens cleaning tissue is placed over the bridqe with each end of the tissue extending beyond the bridoe to the base of the petri dish. The other dimensions of the stainless steel bridqe and the lenqth of the tens tissue are not cri*.ical, but those specified in Figure SA have been found to be satisfactory. After the assembly is complete, fill the petri dish with cFloroform to a level just below that of the horizontal surfac? of the stainless steel 5ridqe. It may be found that the chloroform contacts the underside surface of the stainless steel mesh; this is not critical. Cover the petri dish with the lid and the Jaffe �Iasher is ready for use. Each time the Jaffe '�Jasher is used, the lens tissue and solvent shoutd be discarded and replaced with new lens tissue and fresh sOlvent. Appropriate precautions should be taken �Nhen handlinq c�loroform. 6.4.2 Selection of Filter Area for Carbon Coaring Polycarbonat•e filters are easily stretched c!urinq handlino, and tuttinq of areas for further preparation must be performed with qreat care. The best method is to use a curved edqe scalpel btade to cut �he filter while it is in 24 and the vacuum is applied, the differential pressure across the M�llipore filter wilt be insufficient to ovErcome the surface tension of the water in the filled areas. Thus no filtration will take place through the correspondina areas of the P�ucla�ore filter, and a grossly non-uniform deposit of particulate will be.obtained. c) Add the required volume of sample water to the filtration funnel. Disposable plastic beakers and pipets provide a�eans �f ineasurinq the required samp?e volume �Nithout introducing problems of sample cross-contamination. The reservoir may not be sufficiently large to accommodate the total vol�me to be filtered. In this case more of the sample may be added d�ring the.filtration, but this should be done carefuily and only when the reservoir is more than half full. In this way the addition will not disturb or affect the uniformity of particulate already deposited on the Nuclepore filter. Do not rinse the sides of the funnel, and avoid other manipulation5 which may disturb the particulate deposit on the filter. d) �iSassemble the filtration unit, and transfer the �yuclepore filter to a labelled, clean petri dish. Since the i�Juclepore filters are mor� zasily handled Nhile they are still wet, Tt is recommended that the strip of filter to be �sed for TE� sam�le �reparation should Ce cut as described in Section 6.4.2 before the filter is dried. P1ace th2 cover loose�y over the dish to limit any dep�sition of dust onto the filter. Dry the F"ilter under an infra-red heat lamp for a short time before closing the petri dish completely. Discard the ��illioore filter. 6.4 Preparation of Electron ��icroscope Grids Preparation of the grid for examination in the electron n��croscope reauires a high degree uf manual dexterity and is a critical step in the procedure. The objective is to replicate the filter surrace by depos�tion of a carbon film and then to dissolve away the filter i+self with a minimum of particle movement and breakage of the carbon film. The filter dissolution procedure is illus*.ra�ed in Figure 4. 23 � it is difficult to ensure that a unifcrm deoosit of particulate will be obtained on the filter. Samples of hign solids content, or of high fiber content, may require filtration of volumes less than these. Such samples should be diluted with fiber-free Nater so that the volumes filtered exceed the minima specified. Dilutions should be made by transferring a known volume of the sample to a disposable plastic beaker and making up to a known volume with fiber-{ree water. The mixture should be stirred vigorously before sub-sampling takes place. 6.3.2 Fiitration Procedure a) The sample must be filtered irtvnediately after the ozone-UV and ultrasonic bath treatment. If for any reason the sar�ple has been stored for �ore than a few hours after these treatments, it is recommended that ozone-UV oxidation be repeated for � short period of about 15 minutes, followed by an additior�al 15 minutes in the ultrasonic bath. � b) Assemble the filtration base and turn on the vacuum. The upper surface of the filtration base (both the glass frit and the ground mating surface) must be dry before the membrane filters are installed. Ptace a 0.45 ;,m pore size type HA Millipore filter on the giass frit. If the filter appears to become wet by capillary ac�ion on residual water in the glass frit it must be discarded and replaced by another filter. Place a 0.1 um pore size Nuclepore filter, shiny side up, on top of the Miliipore filter. If the ' Nuclepore filter becomes folded it must be � discarded and replaced. The matinq surface of the reservoir component of the filtration apparatus (the funnel) should be dried by shaking off any surplus water and draining on paper tow•el or tissue. The funnel should be positioned on the filters and firmiy clamoed, taking care not to disturb the filters. The vacuum should not he released until the filtration has been completed. It is necessary to com;nent on the �se of filtration equipment which is still w?t after washing� since improper orocedures at this point can very seriously compromise the results. If the qlass frit is wet when the �iillipore filter is avplied to it, capillary action will result in some areas of the Millipore filter structure being filled by water. When the Nuclepore filter is applied to the surface of the �tillipore filter 22 � _. TABLE 1. LIMITATION aF ANALYTICAI SENSITIVITY BY VOLUME OF WATER SAh1PLE FILTERED Volume Filtered (mL) � 1 Anal�tical Sensitivity Using 25 mm Diameter Using 47 mm Diameter (Fibers�Liter} Filter2 Filter3 0.1 0.6 1.5 x 10� I 0.5 2.8 3.0 x 106 1.0 5.7 1.5 x 106 2.0 11 0.8 x 106 I 5. Q 28 3. 0 x 105 20 57 1.5 x 105 � I 25 142 6.0 x 104 50 285 3.0 x 104 100 ( 570 1.5 x 104 I iConcentration corresponding to 1 fiber detected in 20 grid openings of nominal 200 mesh grid (approximately 80 rm square grid ooenings) 2Assuming Active Filter Area of 1.99 cm2 � 3Assuming Active Filter Area of 11.3d cm2 21 J , . _...._.. sufficient to produce a mixing action in the liquid but should not splash sample out of the container. It is not easy to indicate when oxidation is complete, but this treatment as described has been found to be adequate for all water samples so far handled. When pxidation is complete, remove the UV lamp and quartz pipet, re-cap the bottle and place it in the ultrasonic bath for a period of 15 minutes. This allows part?culate released from the oxidized organic materials and the container surfaces to be unifo rnly dispersed throughout the sample. The water level in the bottle may have fallen, due to evaporation during the oxidation procedure. The loss of volume should be noted and can be accounted for if it is significant. The sample should be filtered immediately after it is removed from the ��ltrasonic bath. 6.3 Filtration 6.3.1 Generai The separation of suspended particulate by filtration of the sample through a membrane filter is a critical step in the analytical procedure. The objective is to produce a Nuclepore filter on which the suspended solids from the sample are distributed uniformly, with a ninimum of overlapping of particles. The volurre to be filtered depends on the diameter of the filtration equipment in use, the total suspended solids content of the sample, and in some samples the volume depends on the fiber concentration present. Table 1 shows the limitation of the anaiytical sensitivity as a function of the volume of water filtered. In practice, it is usually found that the concentration of suspende� solids limits the filtration volume. The maximum particulate lo�ding on the filter which can be tolerated is about 2� ug/cm , with an oDtimum value of about 5 ug/Cm Where the ConCentration of suspended solids is known, the maximum vo�;,��e which can be used may be estimated. Usually, however, nothing is known about the sampTe and the best procedure is to prepare several filters using different volumes of the sample. It has been found that suitable filter samples display a faint coloration �f the surface, and with exper�ence over-loaded filters usuaily can be recognized. ihe determination of a suitable volume to filter is usually a matter of triat and error in the analysis of samples of relatively low total suspended solids but high asbestos concentration. No attempt should be made to filter sam�le volumes less than 10 mL for 25 r�n diameter equipment, and 50 mL for 41 mm diameter equipment. If smaller volumes are filtered 20 � . V�, �' � __ ,. .�' _ �-� . , • . :�,• -_ � � ,� . . ..,- ....... _._ �.�w-'�� . . ...�.. ..�,� -�.....! .+u � - - - • - -- - _ .- _ . w ._.:;�..,w Figure 3. Gzone-UV Oxidation of Water SarrQles in Glass Bottles. The ozone supply line has been split into two lines to permit simultaneous oxidation of two samples. A valve and a filter holder are incorporated in each of the supply lines to the samples. � An air extrdct to r?move surplus ozone is required. If it is necessary to check that the ozene generator is functioning within the specifications, the output can be verified by normal chemical methods. A suitable technique is to bubble the ozone through a solution of potassium iodide and to titrate the displaced iodine with sodium thiosulfate solution, using starch as an indicator. Before the ozone-UV treatment, place each polyethylene or glass bottle co�taining the water sample in the ultrasonic bath for a period of 15 minutes. Mark the level of the liQuid in the sample bottle using a waterproof felt marker. The quartz pipets should be thoroughly washed before each use, and �nstalled on the ozone supply as indicated so that the tip is close to the bottem of the sample bottle. The UV lamv is also thoroughly washed and then irrmersed in the sample and switched on. At an ozone concentration of 4� in oxygen, tr�at each sample with - about 1 liter/minute of gas for approximately 3 hours. At other ozone concentrations, adjust the oxidation time so that each sample receives about 10 grams of ozone. The gas flow rate should be t, 19 --r-..._..._.._..-- � ( Alii 'Exipn�, .� • l\ va��E va�vE � ��lieN f �. �O�DEN �Tr� I � a��.r— (�•2 TI�IIM� '�Olw�fi u�"stAv�OlEi . ' �SMP � iAMo:E . °f "".c5 a� OU6Pi: �" � o P�PgT .v�rER .. -o ? :1 7 0� ��0 54MP�� : � 0 )SmTI �, :. �1 I! � =—'_ 11 C�OIING � � vIGY g P � —'� OZ7HE OENEAa7p4 GAS•LiNE �RriqG tUB� 9 %i�TEp � � l.1MP TR1NS• i7RNERi ; �^ aRESSc4E L') aEGu;.a.OA OXrGEN SUPv�r C��Np�R o/C �OwER SUPP�� v6R1�915 iqpNSfOPMEP I�v 7UT��T "eLECTaiCal �UNCT:ON Figure 2. Diagram of Ozone-UV Equipment. <� 18 � _.: .v.,,.,r� . ; _. _. At all times after collection, it is recommpnded that th� sam.ples should be stored in the dark and refriqerated at about C C in order to minimize bacterial and algal growth. The samoles should not be allowed to freeze, since the effects on asbestos fiber dispersions are not known. Before the sample hottles are opened, the �xterior surfaces shou'd be thoroughly washed and then rinsed in fiber-free �rater to avo�d inadvertent contamination of the sample by material which may be attached to the bottles. 6. PROCEDURE 6.1 Cleanliness and Contamination Control It is most imoortant that all giassware an� apparatus be cle�ned thoroughly in order to minimize the po5sibility of specimen contamination. All phases of the specimen pre�aration should be conducted in the clean room facilities or in a laminar ftow hood, Glassware should be cleaned in an ultrasonic bath usinq a d?teroent solution. After this, it should be rinsed three times using fiber-free water. After drying, equipment should be stored in clean containers and covered using alur�inum foil or parafilm, ail glassware must be washed by the above proced�re before each use. 6.2 Oxidation of urganics Oxidation of the hioh molecular weioht organic ccmponents in �ater samoles prio�• to filtration has been found necessary if orec;se results are to be obtained. asbes�os fibers have an affinity fcr these organic material5. ThrPe separate effects have been identified which result from this affinity and which give rise tc serious errors if this oxidation is not carried out: a) asbestos fibers associated with organic materials tenh to adhere �o the container walls; b) asbestos fibers tenc' to aqgregate with organic materials; c) fibers embedded in org�nic material are not transferred to the TEM specimen. All three effects give rise to low results. Before suh-samoles ar� taken from the bottte it is necessary to ensure that all t'�e particulate material is in suspension. The oraanic material and associated fibers must be released from the container walls. �his can be achieved by treating the water sample in the orioinal cotlection container usinq the ozone-ultraviolet (ozenP-U'/) technique to oxidize the organic materials. However, if a sample is known to be free of or�anic interferences [he ozone-L� oxida[ion may not he required. The equipment should be assembled as shown .n Fi�-ures 2 and 3• 17 � 5.2 Sample Collection -�- - - - - _ - . -....._ �-.._._.�. .._ - .. .. _ .. . _ �- - ..- � -� It is beyond the scoQe of this procedure to furnish detai?ed � instructions for field sampling; the qeneral princioles of obtainina water samples apply. However, some specific considerations apoly to asbestos fibers hecause they are a special tyDe of particulate matter. These fibers are small, and in water � range in length from 0,1 ;,m to 20 um or more. 5.3 Because of the ranae of sizes there may be a vertical distribution af particle sizes in large bodies of water. This distribution may vary with depth depending upon the vertical distribution of temperature, the water current oattern and the local meteoroloaical conditions. Sampling should take place according to the objective of Lhe analysis. If a representative sample of a water supply is reQuired, a carefully designated set of samvles should be tatcen representing the vertical as well as the horizontal distritiution and thPse samples should be composited for analysis. When sampling from a faucet, remove ali hoses or fittinqs and altow the water to run to waste for a sufficiently long period to ensure �hat the samp�e collected is representative of fresh water. Faucets or vaives shouid no' '�e adjusted until all samples have been colle�*ed. If possibl�, sampling at hydrants and at the ends of distribution systems should be avoided. As an additional precaution aqainst contaminaiion, before collecti�n cf the sample, each bottle may be rinsed several times in the source water being sampled. In the case of deoth sampling in bodies of water, this rinsing may compromise the results and should be omitted. Quantity of Sample Two separate samples of approximately 800 milliliters each required. An air space must be left in the b�ttle to allow efficient redispersal o; settied material before analysis. second bottle is stored for analysis if confirmation af the obtained from the analysis of the first bottle is required. 5.4 Sample Preservation and Storage are The results Samples must be transported to the analytical laboratory as soon as possible after collection. 4o preservatives should be added during samplinq; the addition of acids should be particularly avoided. If the sample cannot be qiven ozone-UV treatmant and filtered within 48 hours after arrival at the analytical laboratory, amounts (1 milliliter per liter of sample) of a pre-filtered 2.71'o solution of inercuric chloride sufficient to give a final concentration of 20 ppm of inercury may be added, to prPvent bacterial growth. Appropriate care should be taken when handling mercury compounds. 16 r„_ ._.,.___.__._.__ ._ .__._.-- -- -._ � . . __.._ _-_.__ �---.-- 4.3.23 Routine laboratory Supplies Routine laboratory supplies and lab�Nare are r2Quir2d. 'he general suoolies include a deteropnt For cleaning ,, apparatus, markinq pens for lab�lling olass and plastic ' apparatu5, q1a55 miCrosCope slides, lens pap?r (r�r preparation of Jaffe Washer and linin4 of TE� ar�d storace dishes), lint free tissues. General latware i�c?udes s�,c� items as graduated cylinders, beakers of sever�l s�zes, � pipets. Whenever po55ible, disoosable plastic labware �s recommended to avoid the problems of contamination from new glassware and Cro55-COntemtndti�n between sarples. , 5. S�t4PLE COL;.ECTION ANO PRESERVATIOP� 5.1 Sample Container The sample container will be an unused, orp-cleaned, screw-capped bottle of giass or low density (conventional) aolyethylene and capable of holdinq at least 1 liter. It is rec�cmQrded that the ; use of polyproRylene bottles be avoided sincP �rob�ems of , part�culate being released into water samples have been observed. ideally, :�ater samples are best collec�ed in qiass bottles. However, glass can have significant levels of asbestos on the surfaces and therefore requires careful claaning before use. Glass is also difficult to ship because of possibl� breakage throu�h droopinq or freezi�q. Because of these di;advantaces, polyethylene bottles are more conveni�nt to use and therefore are reco�^mendeG. ihe bottles should first be rinsed twice b� fil�ino aooroximately one third full with fiber-free water and shaking vioorously for 30 seconds. After discardina the rinse war2r, the bottles should then be filled with fiber-free water anrl treated in an ultrasonic bath for 15 minutes, followed by several rinses «ith fiber-free water, It is recorr,mended that blank determinations be made on �ne bottles before sample collection. The followinq m►_thod has been found satisfactory for these dete r.ninations. A pre-washPd bottle containing approximately 800 milliliters of fiber-`re� water is processed as described for preRaration of samoles, includi^� ozone-UV and uitrasonic treatments. When usina �olyethylene bottles, 1 bottle in each batch or a minimum of 1 bcttle in each ?� is tested for backoround level. when �sinq qlass bottles, the ris�c of asbestos contamination from the 5ottle is greater and a�+�nimur^ of 4 bottles in each 24 are exar�ined far backoround tevel. �dditional blanks mav be desirable when samplirg waters sus�ecte�t of contair+ing very low levels of asbestos, or ;vhen �dditicnal confidence in the bettie blanks is des�red. 15 . . . . . -- _ --n 4.3.17 Carbon Grating Replica `�', A carbon grating replica with about 2000 parallel lines oer rrm (Cat. No. 10020, Ernest F. Fuliam, Inc., Schenectady, N.Y, 12301) or equivalent is required for calibration of the magnification of the TEM. �� 4.3.18 Chloroform Spectro rade chlorofor-m, distilted in glass (preserv�d with 1� (v/v� ethanol, 8urdick 3 Jackson Laboratories Inc., Muskegon, Michiqan 49442) or equivalent, is required for the dissolution of the polycarbonate filters. 4.3.19 Petri Dishes Oisoosable ptastic oetri dishes (Millipore Cor�. Cat. No. PO 10 047 00) or eq�i�r3lent, are useful for storage of sample filters and ;petimen qrids. If charge build—up on these dishes is experienced, it has been found that rinsing them with a weak detergent solution will reduce the problem. 4.3.20 Quartz Pipets Quartz pipets are used to bubble ozone through the liquid sample. These pipets are formed by h�ating quartz tubing and drav+ing it to a tip of approximately 0,35 mn inside diameter. The pipet should be sufficiently long to reach within 1 inch of the bottom of the sample bottle, to create good mixing of the liquid during oxidation. 4.3.21 Mercuric Chioride Solution A 0.01 molar solution of inercuric chloride may be required for preservation of vrater samples. This is orepared by dissolving 2.71 g of reagent grade mercuric chloride in 100 mL of fiber—free water. The solution is then filtered tv+ice through the same 0.1 um pore size Nuclepore filter, using the filtration apparatus described in Section 4.3.6 and a conventional filtration flask. � 4.3.22 Routine EleCtron MiCroSCopy Prepardtion Supplies Electron microscooy preparation supplies such as scalpels, disposabte scaloel blades (curved cutting edge), double—sided adhesive tape, sharp point tweezers and soecimen scissors are required. These items are available from most EM supply houses. � 14 4.3.13 Ultrasonic 8ath An ultrasonic bath is requ�red `or dispersinq particu�atz in sample containers and for general cleaning of equipment. The size of unit selected is unimportant, and should be related to the volume of work in proqrQss. Bransonic t�odel B-52 (Branson Cleaning FQuipment Compan�, Parrott Drive, Shelton Connecticut 06a84) has a pc��er of 200 watts at a frequency of 5Q kNz and has been found to meet the requirements. 4.3.14 Carbon Rod EleCtrodes Spectrochemically pure carbon rods arp required for use in the vacuum evaporator during carbon coating of filters. Type AGKSP, �yational Spectroscopic Electrodes, manufactured by Union Carbide, or equivalent, have beer found to meet tfie requirements. 4.3.15 Carbon Rod Sharpener This device is used to sharpen the carbon rods to a neck of 3.6 mm long ard 1.0 mm diameter. The use of necked rods, or equivalent, allows the carbon layer to be applied with a minimum of P�eating of the polycarbonate membrane. The ;harpener, Cat. No. 1204, Ernest F. Fullam, Inc., Schenectady, N.Y. 12301, or equivalent� meAts the requirements. 4.3.16 Standards a) Reference Standard Fiber Suspensions. Glass amooules of stable concentrated chrysotile or a�ahibole fiber dispersions, (Electror. Optical laboratory, Ontario Research �ound�tion, Sheridan Park, �i�ssissauga, �,,, -� Ontario, Canada L5K 183) can be used to establish quality assurance in analyticai programs. The refPrence suspensi�ns of known mass and numerical � fiber concentrations are used to generate control sam�les for inclusion in analytical programs. b) Reference Silicate Mineral Standards on TEM Grids. For calibration of the EDXA system, reference silicate mineral standards are reouired (Electron Optical laboratory, Ontarin Research �ounda`ion, Sheridan Park, Mississauga, Ontario, Canada LSK 183). c) Asbestos 3ulk Material. Chrysotile (Canadian!, Chrysotile (Rhodesian), Crocido�ite, Amosite. UICC (Union tnternationale Contre le Cancer) Sta�dards. Available from Duke Standards Ccm�any, �45 Sherman Avenue, Palo Alto, CA 9a306. 13 .. -- -.�.�...�� __. ._. _ . . _ ..-� �._._._---_ ... _...__ _ - -- --_.-� -,t.a,, important. 8ecause of recent changes in the formulation of Nuclepore polycarbonate filters which have degraded their solubility in chloroform, a more complex dissolution �f procedure may be required. The additional steps in the preparation are more easily completed if the original washer desiqn is followed. This original design is illustrated in Figure 5A. Figure 56 shows samples being . placed on a Jaffe 'rlasher oF this design. Alternatively, methylene chloride may be substituted for chloroform, but because this has a higher vapor pressure it is then necessary to ensure that the Jaffe washer is tightly sealed to avoid exczssive evaporation. 4.3.11 Condensation Washer A condensation wash�r may be useful if TEM specimens are required more quickly than is possible if the Jaffe Washer is used alone to dissolve some batches of Nuclepore polycarbonate filters. A condensation washer consists of a system Hith controlled heating, controlled refluxing� 3nd a cold finger for holding the electron microscope sample �rids. Fiqure 6 shows one model of the condensation washer Cat. No. I6950, Ladd Research Industries, Inc., P.O. Gox 901, Burlington, Vermont 05401) which has been found satisfactory. 4.3.12 Electron Microscope Grids Specimen grids of 200 mesh and 3 mn diameter are required in both copper and gold. The grid oneninqs should be approximatety 80 um square. The fiber count result obtained is pr000rtional to the �nean area of the openings examined. Theref ore, it is important that an accurate measurement of the dimensions of each grid openir,g can be obtained. Since there is a wide range of quality in the available copper specimen grids, these should be examined carefully to establlsh the degree of unifornrity of beth the grid openings and the grid bars. Copper speci�nen grids Cat. N0. SPI N302CC and 3020T, SPI Supplies Oivi�ion of Structure ProSe, Inc., P.O. 8ox 342, West Chester� PA 19380, or equivatent, have been found to meet the requirements. In addition, these qrids have a mark at the center openins. This reference can be used to indicate the location of openings which have been examined. Alternatively, finder grids may be substituted if re—examination of specific grid openings is to be required. Gold specfinen grid� Cat. No; 216I2, Ernest F. Fu11am, Inc., P.O. Box 444, Schenectady, N.Y. 12301, or equivalent', have been found to meei the requirements for gold grids. 12 or equivalent has been found to be suitable. `�hen using the larger diameter equipment it is necessary to filter proportionately larger volumes of water. 4.3.7 Filtration Manifold When a number of samples are to be fittered, several filtration units can be operated simultaneously from a single vacLum source by using a mul•�iple port filtration H manifold (Millipore Corqoration� Cat. No. XX26 047 35) or equivalen't. The manifold should include valves to permit each port to be opened or closed independently. 4.3.8 Vacuum Pump A pump is required to provide a vacuum of 20 kPa for the filtration of water samples. A water jet pump (Edwards Nigh Vacuum Inc., Grand Island, NY 14072, Cat. No. C1-0046-01-000-female connection or O1-0039-01-000--male connection} or equivalent has been found to provide 5ufficient vacuum for a 3-port filtration manifold and also incorporates a non-return valve to prevent back-streaming. 4.3.9 Membrane Filters The diameters of the membrane filters should be matched t� the diameters of the filtration apoaratus in use. For filtration of water samples, two types of filters are r�quired: - polyCarbonate Capillary-p�re membrane fiTters� 0.1 um pore size (Nuclepore Corooration, 7035 Commerce Circle, Pleasanton, California 94566) or equivalent, are used to collect the suspended material•from a water samole. - mixed esters of cellulose membrane filters, 0.45 um pore size Type HA (Millipore Corporation� Bedford, �hA 01730} or equivalent, are used as a support filter placed between the glass frit of the filtration apparatus and the polycarbonate filter. 4.3.10 J�ffe '�Jasher A Jaffe Washer is used for dissolution of Vuclepore filters. Several desi9ns of Jaffe '�IashE� have been used which are modifications of the original design. Provided that the polycarbonate filter can be comptetely dissolved, artd that the materials used .in the different designs of washer are demonstrably free of mineral fiber contamination, the precise design is not considered 11 . . _ _-..... —, _ �� A stainless st?el pressure filtration assembly (Millipore Corpor�tion, Bedford MA 01730, Cat. No. XX40 047 00) with z ,� 0.2 �;m pore size fluoroporeR filter (Millipore Corporation, ',' Cat. No. FGLP 047 00} in the normal filter position and silica gel in the reservoir have been found to be � satisfactory for this purpose. ��, 4.3.3 In—Line Gas Filtration Assembly A filter is placed in the ozone line immediately before the gas enters the samole. A 25 mm stainlass steel gas line filter holder (Millipore Corporation, Cat. No. XX40 025 00) or equivalent with a 0.2 t,m pore S1ZE Fluoropore fitter (Millipore Corporation, Cat. No. FGLP 025 00) or equivalent is used in each ozone supply line to ensurP that the ozone entering the sample is particle—free. 4.3.4 Ultraviolet Lamp A submersible short wavelenqth (254 nm) ultraviolet lamp is required for the ozone—UV axid�tiun treatment of water samples. A 6 inch Pen—RayR ultraviolet lamp (Part No. 90—OOQ4-11) and power supply modei SCT-4 {Ultra—Violet Products Inc., 5100 Walnut Grove Avenue, San Gabriel, California 91778) or equivalent have been found to meet the requirem�nts of this analytical technique. 4.3.5 Source of Known Fiber—�ree Water For blank determinations, final �vashinq of analytical equipment, and dilution of some samples, a source of water whfch is free of both particles and fibers is required. Fresh double—distilled water from a glass distillation apRaratus (MEGA—PURET�� manufactured by Corning and available from all authorized Corning laboratory Supply Oealers) or equivalent is preferable, and has been found to meet this requirement. De—ionized water, filtered through a O.l um pore size Nuclepore polycarbonate filter`has also been found to be satisfactory, but the filtration assembly itself tends to contribute some oarticles to the fiitrate. 4.3.6 filtration Apparatus The water sample is filtered through a membrane filter of either 47 mm diameter or 25 ��n diameter. The filtration assembly shouid be chosen to suit the si2e of filter in use. A glass frit support is required in order to obtain a uniform deposit on the fitter. The reservoir must be easily cleaned in order to prevent sample cross— contamination. A 47 rtm analytical filter holder (Hillipore Corporation, Cat. ��o. XX10 047 00) or a 25 mm analytical filter hotder (Millipore Corporation, Cat. No. XX10 025 00). 10 � ASBESTOS aNA�YStS - aATER SAMPL� DATA " . SAMPLE: JOB; PREP: By _ Date - - LOUNT: 8y Date - - INSTRUMEYT: MAGhIFICATI0N5: GriG : DIL'JTIONS: 0 ' ( 1 Volume Taken (mLj 2 Volume Taken (mL) • COCE ' I i PRCCESS: By Oate - - _ Count Final Volume ;ml; _ I Final Volume (mL) , � FINAL PREPARATjON FILTRATION: Vol. filtered (mL) Active Ar?a (cm2) ; — � I C�MMENTS: (for inctusion in computer print-out; rormat in 5 lines of 50 characters) � I I i � � I � � i � ' �I � I i t fI8E6 CLASSIFICAT10�lS: I COI;NT, NAM ;�t CN C� CQ C"Q CDO 0F . AD ;�X AOX .aQ ApQ aZQ aZ? ; PRQCESS'I FfBER tYPE I CLaSS;Ffw1�tON ; FI?EF� 'YP� i CLMS�.'=(�r�iGt� j NOTES: Pr�paration: — i ; � � Examination: � i , I ' I I I Figure 7. Sheet for Recording Water Sample Data. :i? ..._.�: :. :. a � ; i L � I � i ! ' I I I I � � r i � I � � I I I � i I i i � � . , I � i + . � I v � � ' � i ' I I I I � I , ' � � � � � ' `,� �{►�' i i (( I � �; I I; � � ! �i , ' , L ( �( I � ' : '� � ' ) ' � ' � � � j � ;.'. � .. � � � I I = :� ,i � ,�I, { ,, � �� , N { I � I , � � � � ; I I , � � � 11 � i ; �; a i � , ' ' � i ' � I I i I � ._ . 1 Y , � � I , _ � . � � I I I I � I � � � I I i��:, � : Q ` i i � I i i 4 ( ,; ; o �,; 1 � 1� i ;, � �, I, , i. W i ! i ( � 1 I . � . ! d • : � � i ( I I I' �( I i i t �.; �+.� � f;, . �,, � o , , ;.,.��. ,,;.� , . , i � ; � i � i , i . i � . , � a�i �� ' °; � �'� ' � i I I ' � � � I i � ' � � I I �`�' ' 3`"i I i (' I � ' � I j I � I I i I � � � � � � � � r � � i � v+ �! • ' � i �-� ( � , I + � � I I I i j ( , • , • � � Ji ' LL I J`� � � � I � � i I � 1 1 � I � I i I � � ' ' Z7 u� ' x� . � ¢ � � t � � i � � � i � � 4 ; � ro �: . xi p� ; � , � C i � � Q � , ( i I 1 � I � I I ! � • • •r "�' � ' ' � � I � � i � � j ( � � ' � ' O ( � I � �� { � � � i � ~ �� 1 I .. � ,n' : i i Q , , � � � 1 f � � � ( I � i I ' j CJ a , ; i � Q i � � ! , � ' , i � I � � � �J V- �� i ( i O i � � � � � � � � ' � i ' ' � ( . I I � i , ~ ro N , ! Q , ! � 1 I I I � I � ' ( � I ' � I '- � ; � I � � � � ' I � , I ' �-� v .,, ; ,^� � � I � � I � I � ! � � i � ( � � i.a a, :� i ... � � ; � � � � i � �--C-- i---- j i� , , '�' = I ' ' ( ' t I ' � � ' �i �v � � � I � I ��' � ( � I i i I I' !�? •� � �. � � � � � .� �, ; � i � � I i ( � i I ; I � ' o � �' � � � (� � I � I � � � i i� ; � ��; i �; � � -_l__�..-�._...L_L I � � ; i L . , �, ' ' � ' ' � � : ' � � � � ! ; � I � � ; ' ; I � , ; � � � � � i , i y � �-; i i �� I � i . 1� I i! � � i � y I i N � , . I dF � � I ; ' � ' � I � r • I ( t ; ; I ' � � i � i f i • I � a c i I j i � � I ; i ' • � a, ; �� i�� , I �� J '� � I Y' � x�%� X 1!. ( X X f X X' X X�:� Z):' X X f X X X X � X Y. � X I+C � a.' � X( Q1 1 O 1 y= I � � 1 I I � i I � i � . � . : G: �' v I j � ( i � I I ' i � I i I � ' I ty � . � � � � { � � � � � i �� ' : � - '� � Z � � _ � � ! I I ' I I �� I i � , � � . 33 �-..o....----- -- L� other translation control, and then scan the ir.,aqe in the reverse direction. Continue in tJiis manner until the entire grid opening has been inspected. '�lhen a fiber is detected, classify it accordinq to the �rocedures descrihed in Section 6.�, and then insert the apRro�riaLe classification on the data sheet. °�easure the lenoth and width of the fiber imaqe in millimeters and record these i� the appropriate columns of the data sheet. Do not record fibers of obvious biological oriqin or diatom fragments. Continue the examination until 100 fibers have been recorded in all classif;cation categories of interest, or until 20 grid openinQs have been inspected. The data should be drawn approximately equally from the three grids. In all samples, fibers on a�ninimum of 4 grid openings must be counted. Fibers less than 0.5 �,�m in length will not be incorporated in the fiber concentration catculation. 6.5.5 Estimation of Mass Concentration If the primary objective of the analysis is to determine the mass concentration, the fiber countinq should be approached in a different manner. The number of fibers which must be counted in order to achieve a reliable estimate of the mass concen�ration depends primarily upon the range of the fiber diameter distribution. ?he mass concentration measurement is most s�nsitive to fibers of iarge diameter, which unfortunately are among those which occur infrequently. When the diameter distribution is narrow, such as that £ou�d in the case of chrysotile fibrils, then the mass concentration has approximately the same precision as that of the number concentration. However, the mass concentration may be actually meaningless when calculated from a low number of fibers observed during a routine fiber count, if these fibers have a broad distribution of widths. If the mass concentration is the pri,nary interest, and the precision required is greater than is possib'e from the normal fiber count, a dif`erent approach to the fiber count must be used. Initially, establish the largest width of fiber which can be detected on the grid by a cursory survey, at a reduced magnif�cation, of a laroe number of grid openinqs (about 50). Calculate the volume of this fiber. Adjust the magnification to a value such that a width of 1 mm on the screen corresponds to 10� of the width of the previously selected larqe fiber. Carry out � routine fiber count for a minimum of l00 fibers, recordino only fiber images greater than 1 mm in width. Continue countinq until the total volume of fibers is at least 10 times the volume of the originally selected lar^ye `iber. The precision and accuracy of this technique has not been 34 � Y investfqated fully, but for samples with broad width distributions it is capable of yieldino significantly more precise mass determinations than are obtainable by the conventional fiber count. 7he remaining problem concerns the assumption that the � widths also represent the thicknesses cf the fihers. �leasurements of particle thicknesses can be made separately, using the shadow castino techniaue. Before the filter is carbon coated, apply a vacuum coating of _ platinum-carbon or gold to the active surface of the Nuclepore filter at an anqle of 45°, In the TEM, the fibers wi11 then display shadows on the carbon film which approximate to their thicknesses. Suitable techniques for shadowing are described in the �aoer by D.E. 8radley included ,n the Selected Bibliography. b.6 fiber Counting Criteria 6.6.1 Fiber Counting Method F�ber counting with this analytical �nethod will be performed only by the qrid opening technique. If a specimen grid is too heavily loaded for examination of entire grid openings, a more reliable result is obtained hy preparation of a new filter, usinq a smaller volume of sample. 6.6.2 Fibers Which Touch Grid Bars A fiber which intersects a grid bar will be counted onty for two sides of the grid openinq, as illustr�ted in Figure 9. The length r�f the fiber will be recorded as twice the visible lenqth. Fibers in�ersecting either of the other two sides wilt not be included in the count. �ri .�'�'^� �„!: : ,: ��F'� � - � --'t: � / r �� 0 NO7 COUN7 �t � ' i , � � \ ` ( , �. " � COUNT p$ I �� . 1 �.. � i ' % � I,�`��. � i JO N07 , .� \t:,.. � � \ LCUNT �`, • I COUNT 4S/ `� /''� � '�� _',;irr.'-`_.'��.�- Figure 9. Counting of Fibers Which Overlap Grid sars. 35 � 6.6.3 This procedure ens��res that the numerical count will be accurate, and that the best average estimate of length has been made. Fibers Which Extend Outside the Field of View Ouring scanning of a grid openinq, fibers which extend outside of the field of view must be counted systematically to avoid double—counting. In qeneral, a rule must be established so that fibers extending outsi�e the field of view in only two quadrants are counted. Fibers without terminations in the field of view must not be counted. The procedure is illustrated by Figure 10. ihe lenqth of each fiber counted is established by moving thP sample, and then returnin to the original field of view before scanning is continue�. Figure 10. Counting of Fibers Which Extend Outside the Fi•eld of View. 6.6.4 Fibers with Stepped Sides ' , A fiber with stepped sides will he assianed a�Nidth mid-way betrreen the minimum and maximum widths. � 6.6.5 Fiber B�ndles a fiber bun�le conoosed of ��ny �arallel fib�rs �il? be counte� as a sinqle fiber of a��idth enual •`.o an �s�imate of the mean bund�e �Nidth. �igure 11 shows exom�'es or the procedure. S � � Figure 11. Counting and `�easurerrent c,f =iber Bundl�s. =ach bundle `o be counted as one fiber wi:h di^�ensions as indica�ed '�y arrca�s. 6.6,6 Aggregates of Randomly Qriented Fibe�s The structure of an a4areaate of randomty oriented fihers may be sufficien�ly visible that the constitue^t fib2rs can be counted. This is illustrated in F�Qure 12. In this C�u�- as e COUNT GS 3 =igure 12. Counting or �iber Aggr?c,at2s. 37 ., � � case individual fibers will be recorded. '�here the fiber aggregate is too large and complex to count each individual fiber, the identification and aqgreqate dimensions will be reCorded, but it will not be incorporate� in the fiber � count and mass calculations. 6.6.7E Fibers Attached to �on-Fibrous Debris _ � A fiber may be attached to, or partially concealed by, a y particle of non-fibrous debris. If two ends are visible which appear to be the ends of a single fiber, the fiber will be counted. Where only one end of a fiber is visible, � the fiber will be counted as a sing�e fiber having a lenath equal to twi�e the visible length, except where this would _ place the concealed end outside of the particle. In this � case the length will be recorded as the visihle length plus the extension of it to the opposite side of the particle. Examples of the procedure are shown in Figure 13. There may be more than one fiber attached te a single particle of debris; each one should be counted. If an assembly of fibers and particles is too complex to treat in this way, the overall dimensions should be recorded, but the assembly should not be incorporated in the fiber count and mass calculations. � � � i / � �� Figure 13. Counting and Measurement of Fibers Attached to Non-Fibrous Debris. 6.7 Fiber ld�ntification Procedures 6.7.1 General Before it is incorporated into the fiber count, each particle xith an aspect ratio of 3 to 1 or greater and not of obviously biological oriqin must be identified according , to defined criteria. It is recoqnized that eccnomic - considerations•usually preclude unequivocal identification of every fiber reported. In this analytical method, the � 38 requirement for unequivocal iden`ification is limited to a small oroportion of the fibers in order to demonstra�e �he presence of the particular species. '�he proport�on of fibers examined tor unevui�oc31 identification �'ll be stated in the �nalytical result. T;�e remainder of the fibers are then classified on the hasis of cryst��l�qrdoh�C or chemical s�milarity, or bo�h, to the identified fiCers. If on later examination ir is considered r,ecessary t,o perfor� a more complete znd riqorous i�entification, additional fibers may be examined in �ore detail to conf�r�r conciusions based on the fiber c?assification data. In general, it witl be fo��nd that for various instrumen�al reasons it may be impossible to identify a specific fiber completely, even though the fiber ,�ay be of a��e11- characterized variety. It is, nevertheless, important tc record the degree to �hich the procedures were successf��l in classification or i��:�tification of a particular fiber. 6.7.2 SAEO and EOXA Techniques Fibers are initially classified into t��o cateoories on the basis of morphology: those fibers �Nith tubular morphology, and tho5e fibers without tubular morpholoq�. Further analysis of each fiber is conducted using S�E� and EDXa methods. Althouvh the precise technioues and classification procedures are specified in Sections 6.7.d and 6.7.5, some genPral guidance on the use of SF.ED and EOXA methods is given here. The Crystal Str�CtUr° of some miner.jl fi5ers, suc� as chrysotile, is easily degraded by the high current densities required for cDXA examination. iherefore, SAED investiaation of these sensitive fibers must be completed before attem�ts are made to obtain EDXA spectr�. '�hen examining mere stable fibers, such as tfie anphiboles, the order of work is unimpartant. The SAEO technique can be either oualitative or quantir,ative. Qualitative SAED consists of visual examination of the pattern obtained on the �icroscooe screen from a randomly oriented fiber. SAED patterns obtained from fibers with cylindrical syrrmetry, such as chrysoti'e, are an exceptTon since they are not sensitive to axial tilt, and �atterns from randomly oriented ribers can be interpreted quantitative'y. For non-cylindricat fibers, auantitative (zone axis) SAED reauir?s aliqnment oT the fiber so that a ori^cipal crystallographic axis is parallel to the electron beam, The pattern is then recorded and its consistency wi:h known mineral s`_ructure� is checked by a computer program, Thp SPED pattern obtained from one zone axis may not be sufficiently 39 SpeCific to identify the mineral fiber, but it is often possible to tilt the fiber to another angle and to record a different zone axis pattern. The angle between the two axes can atso be checked for consistency with the structure of a suspect2d mineral. For visual examination of the SAED pattern, the camera lenqth of the TEM should be set to a low value and the SP.ED pattern then should be viewed through the binoculars. This procedure minimizes the irradiation and possible degradation of the fiber. However, the pattern is distorted by the tilt anqlP of the viewing screen. For recording purposes, a camera length of at least 2 meters must be used if accurate measurement of the pattern is to be possible. It is of extreme impor.tance that, when obtaining an SAED pattern for either recording or visuai evaluation, �he sample height be properly adjusted to the eucentric point and the imaqe be focussed in the plane af the selected area aperture. If this is not done there may be some components of the SAEO pattern which do not originate from the selected area. It will 5e found in general that the smallest SAED aperture will be necessary. - For accurate measurements of the SAED pattern, an internal calibration standard is required. A thin coating of gold, or other calibration material, must be applied to the underside of the TEM specimen. This,coating can be applied eit�her by vacuum evaporation or, more conveniently, by sputterinq. The polycrystalline gold film yields diffraction rings o� every SAED pattern and these rings provide the required calibration in°ormation. To form an SAED pattern, move the image of the fiber to �he center of the scresn and insert a suitabte selected area aperture into the electron beam so that the fiber, or a portion of it, is in the illuminated area. The size of the --.. -- apert��re and the portion of the fiber should be such that particles other than the one to be Pxamined are excluded from the selected area. Observe the diffraction pattern with the binocular attachment. If an incomplete diffraction pattern is obtained, move the particle around in the selected area to attempt to get a clearer diffraction pattern or to eliminate possible interf erences from neighboring particles. If a zone axis SAED analys1s is to be attempted on the fiber, the sample must be in the appro�riate holder. The most convenient holder allows complete rotation of the sample and Single axis tilting. Rotate the sample until the fiber image indicat�s that the fiber is oriented ��ith its length coincident with the tilt axis of the goniometer, and adjust the sample height until the fiber is at the 40 eucentric oosi�ion. Tilt the fiber until a pattern apoears which is a symmetrical, two dimensional array of soots. The recbgnition of zor,e axis alignment conditions reGuires some experience �n the part of the operator. During tiltir.g of the fiber to obtain zone axis conditions, the ni�nner in which the intensities of the spots vary shculd be observed. If wea� reflections occur at so�e points on a matrix of strong reflections, the oos5ibitity of multiple diffraction exists, ard some caution should be exercised in selection of diffraction spots for r��easurement. A full discussion �f electron diffraction and multiple diffraction can be found in the references by J,A. Gard, P.B. Hirsch et al, and H.R. Wenk, included in the Selected Bibliography. Not all zone axis patterns which can be obtained are useful or definitive. Only those which have �losely-spaced refl�ctions corresponding to low indices in at least one direction should be recorded. Patterns in which all d-spacings are less than about 0.3 nm are not useful and are usually very wasteful in computer time. A useful guideline is that the lowest angle reflections should be �ithin the radius of the first gold diffraction ring {lIl), and that patterns with smaller distances between reflections are usually the most defin�tive. Five sp�ts� closest to the center spot, along two intersecting line5 of the tone dxis pattern must be seleCted for measurement, as �llustrated in Figure 14, • SPOTI SPOT 3 �p�T 2 � ez e� � SP07 4 e3 � � Oa • � SPOT 5 • Figure 14, Measurement of Zone Axis SAEO Patterns. • 41 ti � The distances of these spots from the center soot and ahe four angles shown are the input for tne comouter proqrar�. Since the center spot is usually very over-exposed, it �ees not form a 5uitable or;gin fcr measurerrent. The reauire� distances must therefore be obtained by measuring betwe?n pairs of spots symmetrically disposed about the center spot, preferably separated by several repeat distances. The distances must be measured with a precision of better tt�an 0.3 mm, and the angles be'ter than 2.5°, The diameter of the first or second rino of the calibration pattern (211 and 200) must also be measured with the same precislon. . The camera constant (aL) required for the computer program i� yiven by: where: aL = a� �+ k2+�2 a� Wavelength of t`�e inci�ent electrons L� Effective camera lengtn in rtm 0 a= Unit cell dimension in Argstroms D� Diameter of the (h, k, 1) diffraction rings in millimeters h, k, 1= Miller indices of tne sc�:terina olane of the crystal. Using gold, the camera constant is qiven by: ;�l � 2.3548 0 (.`irst rinq) aL � 2.0393 0 (second ring) Analysis of a fiber by EOXA is required in this analytical procedure. Interpretation af the EDX� spectrum may be eithPr qualitative or quantitative. For qualitative interpretation of a speCtrum, the elements oriqinating from the fiber are recorded. For quantitative interpretation, the net peak areas, after backqround subtraction, are obtained fnr the elements originatino f�cm the fiber. As discussed in Section 6.5.2, this method provides f�r quantitative interpretation for those minerals which contain silicon. 42 To obt�in an EDXA spPctrum move the imaqe uf tho fib?r to the center of the screen and remove the objective a�er�ur•�. Select an appropriate electron beam dia�eter and deflert the spot to impin9e on the fiber. De�ending on the instrumentation, it may be necessary to tilt the samoie ar� in some instruments to use Sranninq Transmisci�n =1�c�ron , Microscopy (STEM) moda of operation. The ti�e for acouisition of a suitabie spectrum varies �Nith the fiber diamet��, and also with instrumental factors, For quantitative interpretation, spectra shoutd hove a " statistically valid number of counts in each peak. A.nalyses of small diameter fibers which contain sodium are the most criticat, sinc�� it is in the lew eneroy range that the X—ray detector is �east sensitive. �ccor�ingly, it is necessary to acquire a spe�trum for a sufficiently long period that the presence of sodium can be detected in such fibers. It has been found that satisfactory quantitative analy5es ca� be obtained if acquisition is �ontinued until the background-subtracted silicon Kx peak integral exceeds 10000 counts. The spectrum should then 5e manipulatzd to subtract Lhe background and t� obtain the net areas of the elemental peaks. After quantitative EDXA classification of some fi5ers b� computer analysis of the net oeak areas, it may be oossible to classify further fibers in the same sample o� the basis of compar�son of spectra at the instrument. Freouently, visual Com�arisons can be made aft�r so�eNhat Shcrter acquisition times. 6.7.3 Analysis of Fiber ldentification Cata Since the fiber identification procedur� can 5e involved and time—consuming, a Fortran computer program has been provided, the listing of which is given in Appendix A. This program permits the EDXA and zone axis SAE� measurements to be ��mpared aoainst a library of compositional and structural �ata for 226 minerals. The mineral library includes fibrous s�ecies which have been listed by several authors, tooether with other �inerals which are known to be similar to amphibote in either their compositions or somE aspects of their crystalloqraphy, AddiLional minerais may be added to the )ibrary if the� are thouqht to be of concern in particular s�tuations. Rejection of a mineral by the program indicates that ei*her the compositional or crystallo�raphic datd for ;he �ineral in the library are inconsistent with the TPaSurem?r,�5 Tade on the unknown fiber. Denonstration that the �easuremerts are consistent with the data for a particular test �irera? does not uniauely identify the unknown, since tha - possibility exists that data from oth?r minerals �ay aiso 43 � be consistent. It is, however, very unlikely that a mineral � of another structural class could yield data consistent with that from an amphibol� fiber identified uniquely by quanti- ` µ tative EDXA and two zone axis SAED patterns. The computer program classifies fibers initially on the basis of chemical composition. Either Qualitative or auantitative EDXA information may be entered. The procedure using qualitative EDXA consists of entering the list of elements which originate from the particle. For quantitative EGXA (silicon-containing minerals only), the list of elements and the areas under the corresponding X-ray emission peaks, after hackground correction, form the inpu} data for the computer program. The width of the fiber is also required as input into the proqram. The program will select from the file a list of minerals which are consistent in composition with that measured for ttie unknown fiber. To proceed further, it is necessary to obtain the first zone axis SAED pattern, according to the instructions in Section 6.7.2. It would be attractive to specify a partic�lar zo:�e axis pattern to be obtained for confirmation of amphibole, particularly if such a pattern could be considered characteristic. Unfortun�tely, for a fiber with random orientation on the qrid, no specimen holder and aoniometer currently available will permit convenient and repid location of two pre-selected zone axes. 7he most practical approach has been adopted, which is to accept those low � index patterns which are easily obtained, and then to test their consistency with the structures of the minerals already pre-selected on the basis of the EDXA data. Even the structures of non-amphibole minerals in this pre- selected list must be tested against the zone axis data obtained for the unknown fiber, since non-amphibole minerals may yield similar patterns consistent with amphibole structures in some orie�tations. The zone axis SAEO interpretation part of the proqram will cons':der all minerals previously selected from the file as being chemically compatible wit�� the EDXA data. It wi11 then return a second and usually reduced �ist of minerals for which solutions have been found. A second set of zone axis data from another pattern obtained on the same fiber can then be processed either as further confirmation or �o attempt elimination of an a�biguity. Im.addition, the angle measured between the orientations of the two zone axes car be entered into the computer to be checked for consistency with the structures of m',nerats. Caution should be exercise� in rationalizing the inter-zone axis anqle, since if the fiber contains c-axis twinnin the two zone axis SAED patterns may originate from the,separate twin crystals. 44 In pr�ctice, th� full proqram will �ormall�� 5e aoplio� �� very few fibers, unless precise identification of �11 fibers is reouired. 6.7.4 Fiber Classification Categories It is not always possible to proceed t� a definiti��e identification of a fiber; this may be due to instrumental limitations or to the actual nature of the fiber. In many analyses a definitive identification of each fiber may not actually be necessary if there is other knowiedqe available about the sample, or if the concentration is belo�N a level of interest. The analytical procedure must therafore take account of both instrumental limitations and varied analytical require�ents. Accordinqly. a system of fiber classification has been devised to oermit accurate recording of data. The classifications are shown in Tahles 3 and 4, and are directed tewards identification of chrysotile and amphibole respectively. Fiber; wi11 be reported in these categories. 7he general princi:ple to be followed in this anal�tical procedure is first to define the most soecific fiber classification (target classification} which is to be attempted. Then, for each fiber A;<amined, the classifica- tion which is actually achi2ved i5 recorded. Qepending on the intended use of the results, criteria for acceotance of fi'bers as "identified" can then be established at any tim� after completion o` the anaiysis. In an unknown sample, chrysotile wilt be revarded as confirmed only if a recorde�, calibr�ted SAED pattern from one fiber in the CO category is ohtained. Amphiboie �r�ill be regarded as confirrr,ed onTy by obtaining recoroed da�a which yields exclusively amphibole solutions for fibers classified in the AZQ, All or AllQ categeries. 6.7.5 Procedure for ClasSifiCation of Fibers �ith Tuhular Morphology, Suspected to be Chrysotile Many fi5er5 are encountered which have tubular morpholoa,� similar to that of chrysotile, but whith de°y further attempts at characterization by either SAED or EGX�, They may be non-crystatlire, in which case �aED techniaues ar� not useful, or the� may be in a position on the qric.' ;�h�ch does not permit an EDXA saectrum to be obtained. Alternatively, the fiber may be of organic oriQin, but not sufficiently definitive that it can be disreaarded. Classification attempts will meet with varicus degrpes cF success. Figure 15 shows the classification oroced�,re to be used for fibers which �isplay an� tub��l8r ;��r�noloay. a5 � TM CM CD CQ CMQ COQ NAM TABLE 3. ClASSIFICATION OF FISERS '�ITH TUBULAR MORPHOLOGY - Tubular Morphology not sufficiently characteristic •for classification as chrysotile - Characteristic Chrysotile Morphology - Chrysotile SAED pattern - Chrysotile composition by Quantitative EDXA - Chrysotile Morphology and composition by Quantitative EDXA - Chrysotile SAED pattern and composition by Quantitative E�XA - Non-Asbestos Mineral UF AD AX ADX AQ AZ AOQ AZQ All TABLE 4. CLASSIFICATION OF FIBERS WITNOUT TUBULAR �10RPHOLOGY - Unidentified Fiber . Amphibole by random oripntation S�EO (shows layer pattern of 0.53 nn spacing) � - Amphibole by qualitative ED�f1. Spectrum �as ele�ental components consistent with amphibole - Amph'bole by random orientation SAcD and Qualitative EOX,� � - Pmphibole by Quantitati�e E�XA - Amphibole by one Zone Axis SAED - Amphibole by random orientation SAED and Quantitative EDXA - Amphibole by one Zone Axis SAED pattern and Quantitative EOXA - �Amphibole by two Zone �xis Sr�ED patterns with consistent inter-axial angle n istent qZZQ _ Amphibole by two Zone Ax�s SAED patterns, co s inter-axial anole dnd Quantitative EDXA ,yqM - Non-Asbestos Mineral � 46 FIBER WITH TUBULQR MORPHOIOGY IS fiDer morphotogy �naraCt2ristiC o•` that displayed by re�2re�ce t�rysotilp7 cxar�ine �y SAEO Pa[ter� nut I ChrpSo[ile ��y50tile Od[tern �at_ern nut pr?sent or indiSLir.Ci TM � Ezamine �y auanti[at��re E�X� ompo�i[�on not Chr�So:ile Chdt o` Chry50Cile CCmD051C7C� ��o ScecCrum NaM TM CQ i I � I c�cdmine �y �rED �hry50C��^ �3CL2r,•1 ^OC �at:ern C'1rJ5GL'•�2 °�Cte�n na( �r�sent Or indi5[�nc; CM c,�am�n? �Y quanti:ati•�a �;,.�,1 Ch�ySOti�e coTposi[�on E,camine py �uancira;ive E1x:� '�JmDoS�•_'Cn �OC :�ry;�C�'.? tr�at of �!�rysotile ;,;mooS�C'on '10 SDeCcrum � �o � ICOmDOS1'ip� n0; that �f CnrysOtile Yo SDec _r;un CM LNAM Figure 15. Classification Chart for Fiber '�lith Tubular ��lorphology. 47 ,, � c The chart is se1` explanatory, an� essentially every fi�er is either rejected as a non-asbestos mineral (NP.M), or Classified in some way which could still contribute to the chrysotile fiber count. Morphology is the first consideration, and if this is not similar to that usually seen in chrysotile standard samples, the initial classification is TM. Regardless of the doubtf��l morphology, the fiber ��ill still be examined by SAEO and EDXA methods according to Figure 15. Where the morphology is m�re def�;�itive, it may be possible to classify the fiber as having Chrysotile morphology (CM;. 7he morpho?�gical characteristics reauired will be: a) the individual fibri:s shoul� have high aspect ratios exceeding 10:1 and be about 40 nm in diameter; b) the electron scattering power of the fiber at 60 to 100 kV accelerating potential should be sufficiently low for internal structure to be visible; and c) there shoutd be some evidence of internal structure sugqesting a tubular apDearance similar to that shown in Figure 16A, which may degrade in the electron beam to the appearance shown in Figure 16B. Every fiber having these morphological characteristics will be exa�ined by the SAEO technique, and only those whith give diffraction patterns with the precise characteristics of Figure 17 will be classified as chrysotile 5y SAED (CD}. 7he relevant features in this pattern for identificati�n of chrysotile are indicated. The (002) reflections should be examined to determine that they correspond approximately to a spacing of'0.73 nm, and the layer line repeat distance should correspond to 0.53 nm. There should also be "streaking" of the {110) and (130) refTections. Using the millimeter calibrations on the microscope vie��ing screen,. these observations can readily be Tade at the instrument. A TEM micrograph of at`least one representative fiber will be recorded, and its SAED p,attern will aiso be recorded on a separate film or plate. This ptate will also carry cali- bration rings from a known polvcrystalline substance such as gold. This calibrated oattern i� the only documentary proof that the particular fiber is chrysot�Te and not some other tubular or scrolled s�ecies such as ha�toysite, paly- gorskite, talc or vermiculite. Th? proportion of fibers which can be sutcess`ully identified as chrysotile by SAED is variable, �nd to sone extent dependent on both the instru�ent and the precedures of �he operator. The fiber� that fail to yield an ider.tifiable S�ED,pattern �Nill reTain in the TM or G� categories unless they are examined by cDxH. 48 Figure 16A. TEM Micrograph of Chrysotile Fibril, showing htorphology. �--t 0.�5 .um Figure 166. TEM tiicrograph of UICC Canadian Ctirysotilz Fiber aft�r Thermal Degradation by Electron Beam Irradiation. 49 � � � h � � .0.53nm / • .' � tt0. � 1�0 � i � Figure 17. SAEQ Pattern of Chrysotfle Fiber with Diagnostic Features Labelled. Necessary criteria are the presence of 0.73 nm spacing for the 002 reflections, 0.53 nm spacing for the laver line repeat and characteristic streaking of the 110 and 130 reflections. 6.7.6 In the EDXA analysis of chrysotile there are only two elements which are relevant. Far fiber classification, the EDXA analysis must be auantitative. If the soectrum displays prominent peaks from magnesium and silicon, with their ar�as in the aporopriate ratio, and �ith onty minor peaks from other elements, the fiber will be classified as chrysotile by ouantitative EDX�, in the cateqories C�, CMQ or COQ, as appropriate. For chrysotile analyses there are essentially three possible levels of analysis: 1. morphotoqical and SAEO discrimination only (Target classification CD); 2. in addition, EDXA of only those fibers unclassified by SAED (Target classification COj; 3. EDXA in addition to SAED on ail fibers (Target classification COQ). Procedure for Classifi�ation of Fibers Without Tubular horphology, Suspected to be Amphibole Every particle without tubular morphology and which is not obviously of biological �rigin, with an aspect ratio c� 3 to 1 or greater and having parallel or stepped sides, will be considered as a suspected amph��bale fiber. Further examination of the fiber by SAE� and E�XA technipues w11� 50 meet with a variable dPqree of success, dependinq on th� nature of the fiber and on a number of instrument�l limitations. It will not be posSiblA to identify every fiber completely, even if time and cost were of no concern. ��oreover, confirmation of the presence of amphibole can he achieved only by Quantitative interoretation of zene axis SAEO patterns, a very time-consuming procedure. � Accordingly, for routine samples fron unknown sources, this analytical procedure limits the requirement for zone axis SAED work to a minimu� of one fiber representative of each compositional class reported. In some samples, it may be necessary to identify more fibers by the zone axis , technique. When analyzing samoles from well-character=zed sources, the cost of identification by zone axis mP�'nods may not be justified. The 0.53 nm layer spacinq of the ran�om orientation SAc"D pattern is not by itself diagnostic for amphibo�e. However, the presence of c-axis twinning in many fibers leads to contributions to the layers in the patt2rns by several individual parallel crystals of different axial orientations. This aRoarently random positioning of the spots along the layer lines, if also associated �ith a high fiber aspect ratio, is a charac*eristic of amphibole asbestos, and thus has some limited diaanostic value. If a pattern of this type is not obtained, the identity of the fiber is stilt amb�Quous, since the absence of a recognizable pattern may be a consequence of an unsuitable orientation relative to the electron bea�, or the fiber may be some other mineral species. Figure 18 shows the fiber classification chart for suspected amphibole fibers. This �har� 5hOw5 all the classification paths possible in analysis of a suspected amphibole fiber, when examined systematically by SAED and EDXA. Initially two routes are possible, depending on whether an attempt to obtain an EDXA saectrum or a rardom orientdtion SAEO pattern is made first. The normal procedure for a��alysis of a sample of ur,known oriqin wiil be to examine the fiber by random orientation S�E�, qualitative EDXA, quant;�ative ECXA, and zone axi5 :,'.�D, in this sequence. The final fiber classification assioned wi;l be defined either by successful analysis 3� the target 1eve1 cr by the ir�strumPnta; limi�3�ion5. Th� maximum classif�Ca��on achieved for each fiber will be recorded en tne �ounting sheet in the approqriate column. T�e var�o�s classification cateoories Can then he comhin�d �n any desired k�ay for calculation of the f�b?r concentration, and a camplete record or the results frcm each fiber is m3intained for r?�s�essment of the data if nP�essary. cl� � . t10CP wITMOUi 'VBUL�� YCRvy7lJ'vY � b�1 "Mr ::�1 �r�v •M� .Irinfv \ t.r t . . .1[: . .� �.o-�xy o� .. ' ., nc• � . � I ii � ..�e• . I�Y�V 'q :Nt:.v� �.�n .�.��- .'�G'. 'eY. I�y � �Ut" ' t • -(Jf. Y?VI » � ( . Jw� :•a,.�,�if�.t :�� I •�a�� :aw��•�o, v-� � .'r .. uan•xl.� I . b.v : .:�t:•�r'. •'��.t � lill�w� 9 .. �:S �C ! w-.po . I HpY ! .. ' �. .. � . � � ;s0� i t::tr � . � .. w � � .�ar � p •�':• � � ' •AY' � �oYl I I ti �cec:•� 1'�-'�� ( _ �t1�:'^`-E: ": .��. . _ . ' ' �A• �, � 1 ., I .. V�yI .. . �) �� I JYiY � ,--� � .. . „�, .o . ; K,� � ;�- I �� :�� V� ; _Z� � T. �. . _�: i ,:�:::.'`=:, ::,.. . . :•a : .o _� ,r �.� „ : .«., ---, i q JYI � �iZ� � ��Z� .. ai +�c^ x'< . a� � A[: Figure 18 � :J�f .. � .. .� 1 iR _0�. 1 _or� �`n r�: . . -.t:• r . a� fG ��v .. '��. . . . �rl . �• `�� 1 j -• � � . .. � � • �. � � _, �T : o, � V� �- � � � �� •- . .. �:� . t :_ • .` 4Ji �T- 1 • I -� �� :1�•• �,;,�, � �`_. (:- ^. � , �� , " �.. :•a : � {::: ' �.n •i. . � . v�^ •- VL rYl ij� �L. ..t'„• . ' G� - Classification Cnart for Fiber '�lithout Tubular �1or�hology. Bold Lines indicate the Preferred Paths. 52 Level of Analysis 1 2 3 4 Dependinq on the �articular situat�on, tour leve?s n° analysis can be defined in r_his analytical procedure, and these are shown in Tabie 5. In the routine ��nknown samole, a level 3 analysis will be required if the presence of amphibo'e is to be confirmed. For this level of analysis, attemuts wil' be made to raise the classification of every fibe�� to the AGQ category, fn addition, at least one fiber from each type of suspected amphibole found will be examined by zone axis S�E� methods to confirm the identification. tABLE 5. LEVELS OF ANALYSIS FOR �MPHi80LE Apptication Routine monitoring of known and well-charact- eri2ed sources for one mineral fiber type. Routine monitoring of known and well-charact- erized sources where discrimination Letween two or more amphiboie fiber types is required, Routine samples from uncharacterized sources in which presence or absence of amphibole is to 5e confirmed. Samoles where precise identification of all amphibole fibers i; an impor�:ant issue. Ta rge t C�a55ifiCdtlOn for all Fibers ADX ADQ ADQ azQ 53 Required Classification fer Confirm�tion of Amphiboie in e Proportion of th2 Fibers Not Applicable P�oi Applicable All, AZQ or AllQ - Sol�!tions must inc;ude only amphiboles. AZLQ - Sol�tions must incluc� on;y am�hibol?s. w � , , 6.8 Blank and Control Oeterminations To ensure that contamination by extraneous fibers during sa�o'.e preparat�on is insignificant �ompared with the results r?port2d on Sample5, it i5 necessary t� e5tabli5h a continuous program of bldnk measurPments. Initially, and at intervals durinq an analytica' program, it is also necessary to ensure that sampies of known fiber concentrations ca� be ar��lyzed satisfactorily, 6.8.1 Blank Determinations At least one blank determinat�on will be made aiong wi�h every group of samples prepared at any one time. For the blank determination, a 0.1 am yuclepore filter will 5e prepared by filtration of 100 m� of o2e�e-UV treated fiber-free water if using �5 rrm diameter equipment, and 500 ml treated water if using 47 mm diameter equipment. If the samples have been preserved with mercuric chloride, an equivalent amount of the solution should be added to the water used for the preparation of the blank. This blank filter will be carbon coated at the same time as the group of samples, a�d solvent extracted in the same Ja`fe '�asher. All aspects of the sample preparation will then be i,aentir.al to those for the actual samples. Al1 fibers on 20 grid openings of the b�ank sample •�ill be recorded. 'he mean fiber concentration for the blank must Ge tess than 0.05 Mf� or less than lo of the lowest individual value reported in the samples concerned, whichever is the arPater value. If a value higher than these criteria is encountered, satisfar.tory blank values must be �enonstra=ed before further analyses are carried out. If it is sus�ectec �`�at s3mp'es could have been centaminated during the original preparation, the dupticate bottles should be used for the repreparation of the samples concerned. 6.8.2 Control Samples Control samples must be incorporated into sample analysis programs in order to demonstrate that the expected results can be produced from samples of known fiber concentration. Such reference suspensions can be prepared using ampoules of stable fiber dispersions listed in Section 4.3.10. It is recommended that the range of fiber concentrations found in the real samples shoutd be simulated using the roference suspensions. The sealed am�oules of fiber dispersions - become unstable ��hen they ar? opened, and the fiber concentration value sho��ld not be rel'ed upon for more than 8 hours after opening. �ccordingly, it is recommended that, upon opening a disDersion concentra�e ampoule, several reference sus�ensions of �ifferent fiber _ r conCentrations be prepared in sample bottles. These 54 botties can t�ien be stored for �reoar,jtio� and ana�ysis atong with water sampies of urknown fiber conrontrations. 7. CALCUI:aTION OF RESULTS The result� are conveniently calculated usino a c�m��er �rogram, rhe A listin� of which is provided in Appendix B. The Re:hods by wh;ch the calculations are made are described below. 7.1 Test for Uniformity of Fiber Deposit on Electron �icroscooe Grids A check must be made using the chi-square test, to deter-mine whether the fibers found on individual grid ooenings are randomly and uniformly distributed among the qrid openings. If the tota' number of fibers fcund in k grid ooeninvs is n, and the �reas or the k individual grid openings are designated Al to AK, ttien the total area examined is i = k A = �� A� i - 1 The fraction of the total area examined which is reo�-esented by the individual qrid opening area, ��, is qiven by A�/A, If the fibers are randomly a�d uriformly diseersed over the k gr�d opening5 counted, the ex�eCted number of fibers faliinq in one vr?o opening with area Ai is r,N�. If the observed n�mb��r fcund on � that grid opening is n�, then: i = k _ � �nl _ npi�2 X2 np� i=1 This value is comp�red with significance poir,ts of the :t2 distribution, having (k - 1) deqrees or freedom. Sianificanc� levelc lower than O.lo are cause for tne sample analysi�s to be rejected, since this corresoonds to a very inhOm0o2�POt�S de�os�t. If this occurs, a new filter should Ce prepared, oayina �:ore attention to both unifor�n dispersal of the sus�ension a�d the filtration procedure �s descri�ed in Secti�n 6.3.2. �� , � 7.2 Calculation of the �ean and Confi�ence Int�rval of the Fiber Concentration � I n the f i ber count, a max imum of 20 gri d open i ngs ha��e be�n ;a�-o i ec from a population of grid openinos, and it is requ�red to Cet?^'�^� the mea� grid ooeniny fiber count for the pcpularion on ti�e bas�s of this samplinc. The interval about the samole Tean,�which, ��:':r 954 confidence, :ontains the populatian mean, is a�so recuirea. The distribution of fibers on the qrid openin�s should theoreti- cally approximate a Poisson distribution. Beta��se o` fi�er aggregation and size-dependent �dent�ricatior..efreCts, the 1i�er count �ata often do �ot conform to the Poisson distribution, particularly at high fiber co�nts. Sim�;e assum�tion of a P�i55nn di;tribution �nay thereferro lead to confidence inte��va�5 narr�wer than are justified by the data. Moreover, if a Poisson distrihution is assumed, the variance is fixed in r�lation to the total number cf fibers counted. Thus a qarticular f�ber c^unt ` conducted on one grid opening is considered to havz the same confidence intzrval as that °or the sar.,e n;,mb�r of fibers f�urd cn many grid openings. However, the area of sample actually c�.un:eC is very small in relation to the total area of �he filter, and f�r . this reason fibers must be counted on a mininum of a orid CO?ninp5 taken from different areas of the filter in order to PnS�r° representative evaluation of the depasit. At high fiber counts, where there are adeauate numbers of �ibers per grid o�ening �o alloH a saRpl� 2stimate of the vari�nce �o Ce made, :he distributioii can be approx��rated to Gaussian, �«i�h independent values for the :rean and variarce. �.ihere the �a�ole estimate of variance exceeds that implicit in the Poisscrian as�umption, use of Gaussian statistics with the varia�ce Cefine� by the actual data is the mo�t conservative approach to calc�lation �f ton.`idence intervals. F�t low fiber counts, it is not possibte Lo obtain a r�liable sa:r.cie estimate of th� variance, and the distribution also becoR�s asymmetric, bui not necessarity Poissonian. For 3C fibers and below, the distribution becomes sufficiently r,syrm��^etric that the G�3ussian fit is no lon�er a reasonable one, and sar.ple variar.ce estimates are unreliable. �ccordingly, for fiber cou�ts be'cw 3: fibers, the assumption of a°ois5on �istri5ution r+us: Ce �a�� .`�r ca;culation of the confiden�:e intervals. For total fiber caunts less than 5, the lower 95: confi�e�ce va�,e �orresponds to one fiber or less, and in addition, t^�e uooeY ?��: confidence value correspondino to a fiDer count of zero is ?.�� fibers. Therefore, it is not meaninqful to ouote lower �Jnf;�crc� intervat points for fiber ccunts of les5 than 5, anr! the �esu�� should be spetified as "less than" the Corros�onding ?o�sscn �cD?!" 95'; confidence value. �5 56 For fiber counts highe� than 30, the samvle estimate of variarce is a`so calculate�, and the lar�er of th_e two corfidence in`erv�1S is selected. For calculation �f N��son 95:; conr"idence �ntervals, Table 40 cr the reference by E.S. rE;�rson and H.G. Hartley should be used, with an extension to an axpectation of 100. Fcr Tore than 100 fibers� the Poisson distrib�tion can be accurately approxiTateo , by a Gauss9an distribution, stitl using the Foisson varianc� est�mate. For counts of more than 30 f��bers, the 95'� cenride�c2 interval based on a sample estimate of variance is calcutated usinc the Student's "t" distribution. For the two—sided Student's "t" � calculation, k values of grid opening fiber count are compared w�t:� � the expected values for the areas of the grid openinos concerned. In surmnary, fiber counting data will be reported as follows: �yo fibers detected The value witl be reported as 1?�s than 369�= of the concentration eouivatent t� one fiber. 1 to 4 fibers When 1 to 4 fibers are counted, the result will be reperted aS less than the CorreSponding upper 95� Confidence limit (Poisson). 5 to 30 fibers ��ean and 95e confidence intervals �i11 be reported on the basis of the Poisson assu�ption. More than 3Q fibers �hen more than 30 fibers are counte�, both :he Gaussian 95�; confidence intervai �nd the Poisso� 95�: co;.fidence intervai will be calculated. The larger of these 2 intervals :v'tl be selected for data reporting, �hen tne Gaussian 95'_ confidence interval is selected for data reporting, the Poissan intervat will also be noted. Fiber counts performed on less than 4 Qr�o openin5s y�elc ��ery N;,e 95� confiaence intervals when �sing Gaus�ian s�atistics. This �s be�ause �he value of Student's "t" is very large for 1 and Z �egrees of freedem. �ccordingly, fiber counts Tust not be �ace rn 1e55 than 4 grid openings. 57 r a The samvie estimate of variance S2 is first calculated: t = k � (n� - np�)� 52 _ i = 1 (k - 1) where: n� � N•�mber af fibers on the i'th grid o�ening n = Total number of �'ibers found in k grid oper.ings p� = Fraction uf the total area exa�nineci reoresented by the i'th grid opening k � Number of qrid openings for the 95� confidence interval, the valuA of t0.975 �s obtained from table5 for (k - 1) degrees of freedom. If the mean value of fiber count is calculated to he n, the uooer and lower values �f the 95o confidence interval are given by: where: �� _ � + ts � n� = " -Jk n� • Up�er 95': confidence timit n� � lower 95'; C��fi�ence limit n �'"lean number of fiber, per grid ooening s � S�andard �eviation (s4��are root of sampl? esti�na_e of variance) k � Number c° grid ooeninqs ,. 58 7he fiber Cancentration in MF� which corresponds to counting of one fiber is giver b�; where: AfxR� � A x V x 1000 Af = Ef�?ctive filtration d►'Pd of filter �nembrane in mm used for filtration of livuid samole A = Total area ex3mined in mm2 v = Original volume of sample fittered (mL1 Rp � Oilution ratio of originat sample The mean concentration in ,�+FL is obtained by multiolyinq the mean number of fibers ver orid ooenirg by kC. To ot�ain the uaQer and lower 95'� confidence l�mits f�r the concentration !in .HFL) mul*i�?y the vatues n� and n� hy kC. 7.3 Estimated Mass Concentration The mass ef each amphibole fiber in micrograms is calcula�ed usinQ the relationship; - where: M = �x'�lZxDx10�6 « � Ma55 in microvrams � . � Length i n ::m '�J = Width in :,m � � Density of fiber in q�c�3 59 � , � b � For chrysotile, the mass may be calc:alated us�r.g the relat'.cnsni� for a cylinder: M = QX�Xw2XoXio-� The estimated mass concAntratio� is then giver. by: where: i=n M� = C x� M� x 106 i-1 M,. � Mass concentration in ;.gJL C = fiber concentration in MFL, ��hich corresponrs to counting of one fiher M� = Mass of the i'th fiber, in micrograms n = Total number of fibers found in k orid openi�gs The densities to be assumed are as follows: Chrysotile 2.55 g/cm3 Crocidolite 3.31 g/cm3 Cummingtonite 3.28 g/crn3 Grunerite 3.52 g/cm3 Amosite 3,43 g/cm3 Antho�hyllite 3.00 g/cm3 Tremolite 3.00 g/cT3 Actinolite 3.10 g/cm3 Unknown Amqhibole 3.20 glc�n3 � 60 7.4 Fiber Length, Width, Mass and Aspect Ratio Distributicns The ��istributions all approximate to togar;thmic-normal, and so the size range interva�s for calculation of the distribution must be spaced logarithmically. The other characteristics reouired for the choice of size intervals are that they should al;ow `or a sufficient number of size classes, while siill retaining a statistically-valid number of fibers in each class. Interpretation is also facilitated if each size class repeats at decade intervals. A ratio from one class to the next of 1.468 setisfies all �f these requirements. The other constraints are that the length distribution should include 0.5 um as one interval point� since this is the minimum length to be cou�ted in the method� and the minimum aspect ratio is by definition ?.0. The resulting size classes for the various distributions can be seen in the exam�le shown in Appendix 6. The distributions, being approximately logarithmic-normal, must be plotted u5ing a logarithmic cr�ina�e scale and a Gaussian abscissa. 7.4.1 Fiber Length Cumulative Number Oistribution This distrib!�tion allows the fraction of the total number of fibers either ;horter or ionger than a qiven lenvth to be determined. It is calculated using the relationship: where: i=k � n� �(N)k i= ry X 100 � �i i=1 �(.ry�R = CuR�lat;ve number percentage of fibers which have lengths tess than the upper bound of the k'th class � n� = PJumber of fibers in the i'th length class� N � Total number of length classes bl J 0 � M 7.4,2 Fiber 'Nidth Cumulative Number Distrihution This distri�ution allows the fractipn of the to��l number o` fibers either narrower or wider than a given width to be � determined. It is calcutated in a;imilar way to `hat �sed in 7.4.1 fpr the len5th distribution. � 7.4.3 Fiber Lenqth C��mulative Mass Distribution `' This distribution allows the fraction of the total mass incorporated in fibers either shorter or longer than a given length to be determined. It is computed using the relationship: i = k j = n� � � 1 �w� ` i = 1 .l = 1 �(M�k i= N j=;ii x 100 �\ ' � lj,��2 /J i =_1 j = i �here: C�M�k = Cumulative mass percentage of fib�rs which have lengths less than the upper bound of the k'th class n� _`Jumber of fibers in the i'th length class 1� = Length of the j'th fiber in the i'th length class w� _'�Jidth of the j'th fiber in the �'th length class N = total ►�umber of length classes 7.4.4 Fiber Aspect Ratio Cumulative '�umber Oistribution This distribution ailows r.he fraction of the total nur�ber of - fibers which have aspect ra�ios either smaller or larger than a given aspect ratio to be determined. It is 62 • calculated in a Similar way t� that used in 7,4.1 fGr �he length distribution. 1,4.5 Fiber Mass Cumul�tive Number DisLribution This distribution allows the fraction of the totat numb?r ° of fibers which have masses eTther smaller or larQer than a given mass to be determined. It is calculated by placing the f�bers into logarithmically-spaced mass cate9ories, after which the cumulative frequency distribution is obtained in a similar way to that used in 7.4.1 for the � length distribution. 7.5 Index of Fibrasity It is possibte to discriminate between amphibole asbestos fiters and amphibole cleavage fragments on the basis of the distribution of their aspect ratios. The concept of fibrosity in a miner�l embodies a high median aspect ratio, together with a large �pread of aspect ratios above the median value. A single number can be used to describe the fibrosity of a mineral fiber dispersion, and in many cases the value can be used to state if the mate,�ial i; or is not asbestos. The fibrosity index can be defined thus: F-R9 wher� R is the median ot the asoect ratio distribution and q is `he geome}r?c standard d2viation uf the aspect rat�o dis�ribution above the median. The value of g is obtained from th�t portion of the distribution lying between one and two geometric standard deviations above the median. Meaningful values of the index of fibrosity can be obtained for most waterborne fioer �isoersions if jmore than 50 fibers h;ve been measured. The fibr�sity index as defined above has values exceedino lOG for waterbcrne dispersions of asbestos. Values below SC indicate a distribution CharacteristiC of cleavage fragm�nts, or one fror� which the hiqh aspect ratio fibers have been ;electively removed. 8. REPORTING The computer progr3m provided in �ppendix B satis`ies all of �he reporting requirements for this ana�ytical method, 3nd i� is reccrr�ended that this format be used. The size classifications used �nust be the sa�e as those in Appendix B. 63 4 . 8.1 Before tre fiber count data can be processed to qive con�entr�tion values, a d�cision must be �ade as to which fiber classifications ar� to be considered adequate as identification cf the fiber species in question. This decision Nill devend on how much is known about the partiCular source from �Nhich the samole was collected. for a sample from a completely uncharacterized source, tne fol�owing procedure will be used to accumulate the classified fibers: a) Confi!�ned Amph�bole: AllQ t AZQ + All (solutions must inclJde onl� amphiboles) b) Amphibole Best Estimate*: AllQ + AZQ + All + aZ + ADQ + a4 c) d) e) fj Suspected Amphibole: A�X + AX + AO Confirmed Chrysotile: CDQ + CD Chrysccile Best �stimate*: COQ + CD + CMQ + CC Suspected Chrysotile: CM *NOTE: Best estimate can be reported only if some fibers ar� also reported in the confirmed cateqory, otherwise all fiber classifications �nust be reoorted as suspected amqhibole or chrysotile. 6.? The concentration in t�tFL, tooether with 95� confidence intervals, will be reported for the groupings in Section 8.1 (a) to (f). 8.3 Two significant fig�res will normaily be used for cencentrations greater than 1 MfL, and one significant figure for concentrations less than 1 MFL. 8.4 For confirmation of chrysotile, a micrograph and a calibrated diffraction pattern will be provided from a typical fiber. The identification features in Figure 17 must be visible on the diffraction pattern. For confirmation of amphibole, either (1) or (2) or (3} below ,ust be provided for a typical fiber of each amphibole variety reported. The data provided must yield solutions which irc�ude only amphibole. 1) A micrograph, a calibrated zone axis SAED patt�rn, and an EOXA spectrum togecher with peak area �neasurements and EDXA calibration data; 64 2) a microqraph, and two calibrated zone axis S��C patterns Nith a�easurement of the anQular r��ta';on between the two o�:tterns; � 3) A micronraph, tw� ca ibra�e� zo�� axiS S�cD oatt�rns wilh a �:easurement of tne ang��iar rot��ticn ��c•�vA�n 'he g two patterns, and an ED"4 spectrum tc�ether with r.eak area reasurements and EDXA calibratio,: data. 8.5 Tabula`_e the length, width ar�d aspect ratio di5tributions. S.6 Report `he estimatr_d mass concentration in :,g/L, for each of the grouaings in Section 3.1 (a) to (f). 8.7 One significant figure will normally be used for reporting mass concentration. 8•8 Report the concentration.in MFL co�resoonding to one fiber detected. S.9 Report the total number of fibers courtted in eacn of the groupirgs in Section 8.1 (a) to {f;. 8.10 Report the XZ value for ea�h of the groupings in Section 8.I (al to (f). , 8.11 Report the nur�ber of fiber aggregates not included in the fiber count 8.12 Report an� specia' circumstances :�r cbserva�ions s�ch as aggr=gation, presence of orq�nic mat�rials, amount o� �ebris, presence of othe; fibers and their probable ident;ty :f known, 9. IIHITATIONS OF ACCURACY 9.1 C;•rors and Limitations of Idertification Complete identification of every Chrysotile fiber is not possible, due to both instrumental limitations and the nature of some of the fibers. The requiremen� for a calibrated SAED pattern eliminates the possibility of an incorrect identification of the fiber selected. However, there is a possibility of misidentifica*ion of other chrysotile fibers for which both morphology and SAED pattern are reported on the basis of visual tnspection only. The cn)� significant possibilities of misidentification occur with halloysite, ve r,niculite scrolls or pal��gorskite, all of :�hich can be discrimina�ed from chrysotile by the ::se of �DXA and bv observation of the 0.73 rm (002) refiection or chryso_ile in the �AED pattern. .,, As in the case of chrysotile fibers, cemplets id2ntif;cstion or � every amphibote fiber is not possible due to instr�mental 65 � limitations and the nature of some of the fibers. i�oreover, complet� identification of every amqhibole f?ber is usually not practical due to limitat�ons of both time and cost. Particles of number of other minerals having comp�sitions similar to those of some amphiboles could be erroneously classified as amphibole ��hen the classification criteria do not include zone axis SAED techniques. However, the requirement for quantitative F.�xA measurements on all fiber; ds supoort for the random orientatior� SAEa technique makes misidentification very unlikely, part�cularl.� when other similar fibers in the same sample have been identified as amphibole by zone axis methods. The possibility of misidentifiCation is further reduced ��ith increasing aspect ratio, since many of the minerals with which amphibole may be confused do not display its prominent cleavage parallel to the c-axis. 9.2 Obscuration If large amounts of other materials are present, some asbestos fibers may not be observed because of physical overlapping. 7his will resuit in low values for the reported asbestos content. 9.3 Inadequate Dispersion If the initiaT water sample contains organic material which is incompletely oxidized in the ozone-UV treatment, it will not be possible to disperse any fibers associated with the organic material. This may lead to adhesion of some fibers to the container walls and aliquots taken during filtration w�lt then not be representative. It may also 1ea� to a larae proportion of fiber aggregates which are either not tr•ansferred during the replication and filter dissolution step or which cannot be counted during the sampie examination. The result obtainEd from such an analysis w,ll be low. The sample will also be inadeauately disoersed if it is not treated in an ultrasonic bath prior tc fi)tration, and therefore instructions regarding this treatment must be followed closely. 9.4 Contami nat ion Contamination by irttroduction of extraneous fibers during the analysis is an importa�t source of erroneous results. particularly for chrysotile. The possibility of contamination, th�refore, . should always be a consideration. 9.5 Freezinq 7he effect of freezing on asbestos fibers is not known but there is reason to suspect that fiber breakdown could occur and result in a higher fiber co�nt than was present in the oriQinal sample. 1'herefore, the sample should be transported to the lsborator� and stored under conditions that will avoid frest�ng. � 10. PRECISION aND �CCUaaCr 10.1 Generai The precision that can be obtained is dependant upon �he nu��er o.` fibers counted, and on the uniformit� of particulate dev�sit on the original filter. If 100 f�bers are co�nted �nd tfe loadino is at least 3.5 fihers/grid squdre, c�mputer modeling of the countinq , � procedure shows that a relative standard deviatior, of about 1C'� can be expected. As the number of fibers counted decreases, the precision wil) also decrease approximately as r�y where N is the number of fibers counted. In actual practice, some degradation from this precision will be observed, fhis deqradation is a . consequence of sampte preparation errors, non—unifo mity of che filtered par+iculate depo�it, and fiber identificaticn variability between operators and between instr�ments. the 95'o confidence interval about the mean for a single fiber cnncer�tration measurement using this analytical method should be about � 25:; when about 100 fibers are counted over 20 grid ooen�ngs. Fcr these conditions the precision of the cempu�ed mass concentration is generaily lower than the precision f�r the fiber number concentration. The precision to be expected for a singte determination of mass concentration is critically dependent on the fiber width distribution. For a result oased on �easurement of a minimum of about 100 fibers, the 95'� confidence interval about the mean computed mass concentration may vary between {25"'.. dnd =60�:. If better precision is reauired for a mass dete w.nination, the alternative counting method described in Section 6.5.5 should be used. 10.2 orecision 10.Z.1 Intra—Laboratory Ccnparison Using cnvironmental �ater Sources Table 6 shows the results obtained from analysis of 10 replicate samples from each of 8 water sampling locations. �our of these locations were associated with a source of chrysotile and four associated with a source of amph�bole. It can be seen that the relative standard deviations of the number concentrations range betw2en 13� and 22`S. The corresponding relative standard deviations for the �ass concentrations range between 29°� and 69`�. 10.2.2 Inter—Laboratory Comparison of Filters Pr�pared ;isi�g Standard Dispersions ard Environmental 'nater Sources Tables 1 and 8 show the fiber c�unting results ob�ained �vhen seCtors of filters prepared in the ORF Labora�cry �;ere distributed to six laboratories considered experi�nc�d in asbestos analysis by the identification and countinQ techniques incorporated in this manual. The sampl=s as 6-7 i � 7 I I � � O � I ' C L L 7 � 1 ' , u' I � v Q� �'1 I � 7 S o I I� _ a .Q. ..�. v T .T.. ..-. ^ � � I Z � I 1 ^�� I I � a�L � i - ' i ' L 9 � ' ' I � ^ � � : .a .t � ✓ "� O � � CD T I ^ .r T � �.ii ^ y) Q �^I .O N Q � V � J � � ✓1 O � � � I � s I a N o i c '-�ni i-vi ti ON I �p v 9 '?�., 1 � `� ' I � 9 � O C � � ^ � O+ !�+ ' W o e � .- > , � � � o ' Q~ C � i J y ..� .� ^ •�1 I ' � `,V,, ,fl ( �L 9 � � �' 1 ^...) :�. Q �� O f'! ! P i � O O O I S .n -� � I � tJ L ' � Q s ' � ' y � Z I i V� C ' \ I1 = � � ./'t � I W �, e I c� �. .- v 7� .n Z� � E o 0 0 0 �_ � o e.� O � 1 i ; p o Z � 7 q ^ 5 f � � r� � ~ W �� , � � � I � I � � s�n � i _� o ; ;, r' � � � ( .. I ` I J _. -, O o J � 1 a x -+ '� ! " '^, � T . N � �... I a cr � � I x c . . a , < v �v .n � Q i � - � CY O �� Q C I... J..�i I r` �O ^J .t1 ' -- "f ` y � c J L Y. � O '� .-r I a I � �l � N ' .��. 9 C N ' U J�I ; I I � y = , i � �( � � v _ � n �, i ��, � � T � c � .. � . z ' i � �r � v . .� r �1+ .-� i Q � ^J' ' •r .� — �' � : � i � � � � I . ¢ 3 n J � v 1 � � J ? � � � � � � � ¢ , d � � � I = � 1 ' I = L �^ � t i � . a ' y L c . - J J O L .. O � 7 > > s W � J ! � o a O � ... � ..J � a " .+ — — .. CII I � v - i .� v 7 � Q -- � .+ .+ J � ✓ �+ _ ~ � � I k � y j : � � � i � � � � � z .= � �t � � ._ i � : i I � 1 I � � � y 1 � � � y I �+ I � ' � = J i � � � ' r i � I � C � � � ( 2 , � N ( J � � ; � f � „ � ; - . ----�_ . � •- � J J - I �' `0 r � � � • � � �O 'n 'n "� v ! ' j — � ' i -� • 2 i � � ' i� _ �----�„___,� f • �1 � - � j I � � ' � t a i 7 ; ! � "�� s � � � ' . s . a+ yf _` � � � �,� � � i, - c � � . , �- �' � o � j O r N � !i � � v_ N ^ � ' _ _ . _ . o � .+ I " s – a � e _ • � � , -� � T , "' - - a � _ I < N , � - ' � I ' � � � Q � �1 � . ctw � , � � � _ �, _ _ � � ; cn , . C1 • . , n f � ' �'_____ � � • CL , L o° e � � �c � �, � . Q ' ' y� o' : ^ � 1' x o � . ' � � � � � Z . � � ' Z � Q ' � � ' +.... ! ti � , y : i H ' '� � I a� � � ' � ^? '_' v c _ .., i � y a _ Y : a ' . � � o ^ � �, � � � .�. � .. . .+i �� _ 0 ' � � ( c.�i .�y. ' a � , . � . r.' � � � . N , - � . . – l ... ¢ a �, ,� c��, '� � i, � � 1 V � T � O � � � }� ry � � ' Q ' ( 2 �' r1 1 . r [1„ I I � J I a N^ o , -• 4�' 'v U . i� � c° ,'�� ^ ^ � _ . y :v n � � > � I � � ! r O ! `� ' F— . ' � ,, � � � � ., ^ � ; ' a+ c � i I � _. ^ b � � •� .n � � � =�u - � .n Q � g` . i � ' . t � � _ �— ------_-- ' � � � ; � � W = ' _, i i� ^ ^ ti O � ,� . F- �' �� ?�t � ^ _ . i � � � � � � � " = j � � V (v �. .� � s• _ •'• � ! 1 0 . .r� • v ^, n ,•�, � -; 'v > (� � v � T � " Q P '� � ... i 'T O � . j' ' I"'�__ I W � 1 � S � . .J I o f -r c N � a . O . ^ �, Q ' � � °� �' � a a � • F— ' , � � � - � � � ' p • , i ' 1 � n . ' . � � - . 'J • O .. � . i . , ; O � j 7 • q � Y • ' � � � �' C .A y / L _ � i � J O � L � N � ` O � L � � . .� c � '^ ' � � ? � L � J �. J .-.� �� i .n S .� � T F • 69 r.. 0 N � � L. 6! � ^" �D � N OO 41 Cl u ..•. C t� t� t� 1� .� ,Q C � � � � ? Z U �1'1 4! � � .-+ c a� �o 6J �a. C � � y.� • C N N N d � �C N W C . •r � .-� W t� CO J 9 C 4- L � N O O� � � � 1 a u C> C �p O N N N L � O t"') Q+ N � ++ u'� o'+ r� v �n W C � H � • Q v 3 p n O er J C� C �D a0 � a 4i O �t� t0 �D f" � .-� Z W � Z y. � O N 'C � L OJ y^ c . a ..� � a� c a� r� rn �n L''! E ti O Q N N f"'f a U Z Z � � � C � a rn c`^'� tD Q� J tJ � � � I Li C N M w 'J'+ 0 I � � ��4 N N N N c.i � " � . � %v C ta.. � � � � �" N O p a.i Q� � ... .-� O� U � .-. � r'n O � � i 3.� �-n O �"� CV �+ u'� 4 C O� � � Q -� p C 1-. t� .-. O � �„) �C 1� ct �"' � � � .-� �-+ N N Z � W J m �t �, L O ~ ...� N P'9 Q 'O L O � A ...7 U o W N .� �p ^"� Q� Q 'J n � � � i � � • C' � > � N i .-�i � U v O t� r� r+ N �C1 t0 rn .-. � Q� Q ^ C cT O �-+ N .� r� •r. 1 1 h Q � s � M N � O N �O tD � O O .--� ch cn �A - .-. distributed were identified 5y numher �n)y, ;� 'ahie i i� can be seen that the r,lative siandard �eviations for �he Six results on each of the stan�ard dispers;�n fi'te�s ��c rot exceed 27n. In Table 8, the environmental ��ater sources used to prepare the filter samoles containe� similar types of suspended materials as those ��sec to generate the intra-laboratory results in Tab',e 6. Tne relative Standard deviations dp not exceed 29";, which appears higher than the valves obt3ined for the intra�laboratory res�lts. However� when the 6 inter-laboratory results are compared xith Lhe 10 intra-laboratory values, there is no stat�stic�il� Significant difference to indicate that there has 5e�n an� degradation of precision. 10.3 Accuracy I0.3.1 Intra— and Inter-�aboratory CCmparison of Standard Dispersions of Asbestos Fibers Tables 9 and 10 show the results obtained between two taboratories when stable aQueous fiber disoersions of kncwn ma55 concentrations �ere analyled. The fibor concentrations reported displayed no siqnifi�ant differpnce between vaiues frcn the t��o laboratories. The rela�iv� standard deviation of the mean fiber concentr3tion was 17', for chrysotile and 16�: for croci�o)ite. The corres�oneing relative standard deviat�ons `or the �ass concentration were 16�� for chrysotile, and 37': for cr,c��o�ite. ;he hipher variability for crocidolite �s a conSeCuence of the lcw statistica� reliability of the �arqe diameter fi5er counts. The computed mean mass COnC2ntratiCn for chrysotile was about 4b�� hi�her tha� ��e Kn�wn mass concentration. This �ay be a conseouence o.` the d�ff�cu;�v of diameter measurernent for sinqle chrysotile fi�rils ;r the assumption of the bulk value for the �ensi�y. The computed mean value for �ass concentra�ion f�r tne crocidolite sample was 67.4 �q/L, which is very close t� the known concentration of 50 �g1L. 71 z 0 � � W a N W J 0 �- cr O m r -� aC � v �u J r Z O O /� in r � — G U V O� > � � O O N h- �--� O Q J O � W Z � �. V G J Q W � � � W O ►�- Z � � � r J � Q W Q J � a r ti � � ' i .� I , w- c � o � I I 0 �, c� r� r� ... p ^+ � N � � � `�' � I � �� I , ' zs � ` � � I I � � I � N C ( I a o i � �y � ' ' 7 A J N �A N � O O� I C� � ' I � � � � a.+ Gf � � Q l'��'1 I � � � �O :V �a c c --� � � � ^ i I � y ' : +� C � � I IWU I , �' � I q > L J y t� �-+ � rC �f1 .T.� l0 �1'1 Q N C CD N N N O U •r � ; 1 1 1 a..� ,+d � � � � � � L � ( �D C �D a� 4� .-» .-r ai ' o u � � p :.� :.1 � O'� L d .G � C O � � A � N .-� .-� CJ a � N M N v, � � L h i, ^' i ( O � � J 6 Z w p � tn J CC S W �. � U Q � � U C W b Z t+.. O W !-- s .• a S� 0 U C.� � � � � c f" J CC Q1 O � G C J 6!) �' Q �-. � � � O W Q Q � 1 � J W a N"' W Z oG .-. LL O � .-� y W tn J �- GJ ,..� H = a c a, "' c i O 7 I � v � �n ,n r� � , rn rn ^ � .n �n cv .-. ,� � E L � d I 2 � � � � � � i I i i . � ' � I N C ( E ,Q I � iJ � ( � '� �o -r � O et -S p y L \ ' � Q � ' u ar Q+ C Q� � � � �• � t� �.!'f � � 9 C 3 � pJ � � � N � � G I � •� V � I � H O W t� i I i I ' , � � i ^ � � A � I L ( ( i � :' � � i � _ � i r. � � ' � N ; C d .�-r �^ �-. � N ti I i O u u � � i i i � , i � ro � � � � I N 6� I � �+ r+ � M ^ i , " � ... ,� � �� 1 j � �, i O ( ' � i � I � .1�+ � � � i '�'% , i L � I a . G L� C cC f"� tp ( � 'a ( v v � x � � y .-N .�r .-. I E � i � y ' � � N M I � Q tQ � � I � � I o I.r L ""' i v O a ro � I � � . 73 w � � ti SELECTED B:BLIOGRAPHY Anderson� C.N. and J.M. Long (1980). Interim Method for Determining Asbestos in Water, Report EPA-600/4-80-005. U.S. E�vironmental Protec�;on Ager,cy, Athens, Georgia. Availabie through National Technical Infornation Service, Springfietd, Virginia 22161. Asher, I.M. and P.P. McGrath (E�itors)(1976). Symposium on E?ectron ��icro- scopy of Microfibers. Proceedings of the First F0� Offi�2 of Science Sur*r�er Syrr�osium, Penr�sylvania State University. U.S. Government Printing Office, Washington, O.C. 20402. Bates, R.L. and J.A. Jackson (1980). Glossary of Geology, Second Edition. American Geological Instituta, Falls Church, Virgir�,ia 22041. Beaman, O.R. and D.M. File (1976). Quantitative �etermination of Asbestos Fiber Concentrations. Anal. Chem. 48(1): 101-110. Bradley, O.E. (1965). Replica and Shadowing Techniques. In: Techniques ,'cr Electron Microscopy. Biackweil Scientific Publications, Alden Press, Qxford, O.N. Kay (ed.) 96-152. CampbeTl, W.J., R.�. Blake, L.L. Brown, E.:. Cathe~, and �.J. Sjoberg (19%�l. Selected Silicate �inerals and Their Asbestiform Jarieties. Bureau of ;�ines Inforn►ation Cir�ular 8751. U.S. Bureau of Mines, Avondale �esearch Center, 4900 LaSalle Road, Avondale, Md 20782. Chatfield, E.J. (19�9). Preparation and Analysis of Particulate Sanples by Electron Microscopy, with Special Reference to Asbestos. Scanning Electron Microscopy/1979/1, SEM Inc., AMF 0'Hare, Chicago, Illinois 60666. 563-578. Chatfield� E.J. and M.J. Oillon (2979). A National Survey for Asbestos Fibres in Canadian Orinking Water Supplies. Health and Welfare Canada Re�ort 79-EHD-34. Information Directorate, Oepartment of �ational Health and Welfare, Rrooke Claxton Building, Ottawa, Canada K1A OK9. Chatfield, E.J. and R.'a. Glass (1976). I�proved Hethodology for Q2te�ination of asbestos as a'�ater Pollutant. Ontario Research Foundation Re?ort, Mississauga, Ontario, Canada. Chatfield, E.J., R.W. G1ass and M.J. Dilton (1978). Preparation of Water Samples for Asbestos Fiber Counting by Eiectron Nicroscopy, Report EPA-E00/4-78-�11. U.S. En'vironmental Protection Agency, Athens, Ge�rgia. Available through the Nati�nal Techrical Information Serv�ce, S�ringf�etd, Virginfa 22161. � 74 Chopra. K.S. (1978). Interlaboratory ��easurements of Amphibole and Chrysotile Fiber Concentrations in Water. Jcurnal of Testing and E�al�ation, Vol. 6, No. 4, 241-241. Cook, P.M., I.B. Rubin, C.J. Maggiore, and '�.J. Nicholson (19i5). X-ray �� Diffraction and �lectron Beam Analysis of Asbestiform Minerals in Lake Superior Waters. Proc. In*er. Conf. on Environ. Sensing and Assess�ent. Fublished 1976 IEEE, New York� N.Y., 2(34-1): 1-9. Deer, W.A., R.A. Howie and J. Zussman (1966). Rock-Forming Minerals, Part 2, � Chain Silicates. Longmans, �ondon, England, 99-192. Gard, J.A. (Editor} (1971). The Electron-Optical Investigation of Clays. Mineralogical Society, 41 aueen's 6ate, London, S.W. 7. Hirsch, P.B., A. Howie, R.B. Nichotson, D.'�. Pashley and �.J. '�helan (1:E5). Electron Microscopy of Thin Crystals. Butterworths, London, 1fi-23. Kay, O.N. (1965). Techniques fp.r Ei�ectron �Nicroscopy. Blackwe'11Scientific � Publications. Alden Press, Oxford. / Leake, B.E. (1978�. Nomenclature of Am�hiboles. The Canadiar Hineralogis�, 16, 501-520. l�e, R.J. (1978), Basic Concepts of Electron Diffrac�ion and Asbestos Ioentification us9ng S�0 Part I: Current Methods of Asb2stos identification using SAO. SEM/1978/I (ed. 0. Johari) A�HF 0'Hare, Chicago, Illinois, 677-b86. Lee, R.J, (1?78). Basic Concepts of Electron Diffraction ard Asbestos Identificatio�; Jsing S�D. Part II: Single C��ys�al and SA�. I5?d, 68�-��. Lee, R.J,, J.S. Lally and R.M. Fisher (198i). Design of a Laboratory for Particulate Analysis, EPA 600;4-51-032. National Technical Information Service, U.S. Dept. of Cornmerce, Sprin9field, Va 22161. Ledoux, R.L. (Editor) (1979). Short Course in Mineralogical Tecnniques of Asbestos Oetermination. Mineralogical Association of Canada, �/o OepartTent of Mineralogy, Royal Ontario Museum, 100 Queen's Park, Toronto, Ontario, Canada, MSS 2C6. Lishka, R.J., J.R. Millette, a�d E.F. McFarren (1975). �shestos anal�sis 5y Electron Microscopy. Proc. AWWA �ater Quality Tech. Conf, American water '�lorks Assoc., Den•rer, Colorado, XIV-1 - :tI'I-2. McCrone, '�1.C. and I.M. Stewart (1974). Asbestos. �,mer. Lab. 6(4) 13-13. Michaels, L. and S.S. Chissick (Edi:�rs) (19791. Asbestos (Volume 1) Properties, Applications and Hazards. John Wiley and Sons, New York, ;�.Y. y 75 � �r f� �� Millette, J.R, arid E.F. McFarren (1976). EDS of '�aterbarne Asbestos F;bers in TEh, SEM and ST"cM. Scar.ning Electron Microscopy/19i6/I (ed. �. Jonari) IIT Research Institute, Chicaga, Illinois fi0616, 451-460. National Bureau of Stan�ards Special Publication 506 (1977). Proceedings of �lorkshop on Asbestos: Definitians and Measurement �iethods. Held at ��BS, Gaithersburg, M0, (Issued �ovember 1978). U.S. Government Printing Office, Washington, D.C. 20402. - National Bureau of Standards Spec'al Publication ��o. 619 (1981). Proceedings of the NBS/EPA Workshop on Asbestos Standards: Materials and AnalyticaT Methods. Held at NBS, Gaithersburg, MD, (Issued 1982). U.S. Government Printing Offi,ce, Washington, D.C. 20402. Vatrella, M.G. (1966). Experimental Statistics: P�ational Bureau of Standards Handbook 91. U.S. Government Printing Office, '�Jashington, D.C. 20402, Pearson, �.S. and H.O. Hartley (1958). Biometrika Tables for Statisticians, Vo�ume I. Cambridge University Press, 32 East 57th Street, New York, 22, N.Y. Russell, P.A. and A.E. Hutchings (1978). Electr�n Microscopy and X-ray Applications to Environmental and Occupational Health Analysis. Ann Arbor Science Publishers Inc., P.O. Box 1425, Ann ArCor, Michigan 481G6. Samudra, A.Y. (197�). Optimum Procedure for Asbestos Fibers Identification frcm Sele.:ted Area Electron Diffraction Patterns in a Modern Analytical Electron Microscope Using Tilted Specimens. Scanning Electron ��licrosco�y, 1917/I (ed. 0. �Johari). IIT Research Institute, Chicago, Illinois 60616, 385-392. ' Siegrist, H.6. and A.G. '�lylie (1980). Characterizing and Discriminating the Shape of Asbestos Particles. Environmental Research, 23, 348-361. Smith, J.E. and M.�. Jordan (2964). Mathemati�al and Graphical Interpretation of the ��q-Norm?1 Law for Particle Size Distribution Analysis. J. Colloid Science 19, ��+9-559. Spurny, K.R., W. Stober, H. Opiela and �. Weiss (1979). On the Evaluation of Fibrous Particles in Rem�te :mbient Air. The Science of the Total Environne�t 11, 1-40.. Steen, D. (1981). Oetermination du Taux de Contamination de L'Air Par L'Amiante. Institut Universitaire de '"�dec7ne du Travail et d'Hygiene Industrielle, Route de-la Clochatte, CH1052, Le Mont-sur-Lausanne, Switzerland. Wenk, H.R. (Editor) (19;6). Electron �licroscopy in �lineralogy. Springer-Verlag, New York. 76