Document 82gDZ19gNkB9kMypBDrj1DQZm

ASBESTOS INTERNATIONAL ASSOCIATION (Limited by Guarantee) 68 GLOUCESTER PLACE, LONDON WiH 3HL, ENGLAND ~ fi MEMORANDUM TO: Asbestos Producers Advisory Panel Medical Advisory Panel FROM: Director General cc Executive Committee OUR REF.: AIA/5/FIB 7 December 198A TREMOLITE IN CHRYS07ILE The attached report on a meeting, together with a paper by Dr. Hodgs'on on Tremolite in Chrysotile, are forwarded for information. In essence, research is being conducted at the IOM* into the possibilities of testing commercial chrysotile tremolite content. This matter would appear to be of considerable significance to chrysotile miners. Ycu will be kept informed. Sir Neville Stack Encs. Notes on a Meeting Paper by Dr. A.A. Hodgson * Institute of Occupational Medecine, Edinburgh, Scotland AO 1 30C aEcIiV E D DEC i 0 1984 H. C. LEW1NS0HN, M.D. UCC 017480 NOTES ON A MEETING BETWEEN DOCTORS ELMES, HODGSON AND BROWNE REPRESENTING THE ASBESTOSIS RESEARCH COUNCIL AND SIR NEVILLE STACK, DIRECTOR GENERAL OF THE ASBESTOS INTERNATIONAL ASSOCIATION, TO DISCUSS ASPECTS OF THE PROBLEM OF TREMOLITE CONTAMINATION OF ASBESTOS FIBRE (22 November 198-4) Doctor Browne began by explaining the background of the present meeting. At the last meeting of the Scientific Committee of the A.R.C. much time had been spent in discussing recent research which had suggested that tremolite, which was a regular contaminant of chrysotile may be an important cause of some of the morbidity associated with the mining, milling and manufacture of chrysotile. He had been in correspondence with Professor MacDonald, who confirmed that research was being carried out in his Department on the adverse health effects of tremolite, both in vermiculite miners in Montana and in chrysotile miners and textile workers in Quebec and Charleston respectively, as well as examining lung fibre burdens of a wide range of cases in mesothelioma throughout Canada, to assess the importance of tremolite in causation. Some of the results of his work are being presented at the conference on "Inhaled Particles", next September. Doctor Elmes gave further details of research indicating tremolite as an important cause of asbestos-related disease. He discussed the views of the A.R.C. that an attempt should be made to produce a practical commercial test for assessing the tremolite content of chrysotile. The possibility of producing such a test method was already being explored by Doctor Hodgson in conjunction with the Institute of Occupational Medicine at Edinburgh. Some research funding was being made available by the A.R.C. but this research seems to be of great importance to the asbestos industry, particularly the mining industry and it was felt that the A.I.A. should be urged to participate. He also said that some manufacturing Companies were already beginning to frame a policy of restricting imports of chrysotile to those areas where the tremolite content was known to be minnimal. It was felt that the International Asbestos Industry should be made aware of this development. Doctor Hodgson had prepared a valuable paper for the meeting, outlining what was known about tremolite as a contaminant of chrysotile and discussing possible schemes for both identification and the separation of tremolite from chrysotile. He pointed out that a wide range of samples of fibre would be required for research on this subject. He mentioned one or two possible sources, one of which was Central Asbestos and undertook to look into the possibility of obtaining fibre from this source. But it was emphasised that collaboration from the mining industry members of the A.I.A. would be very helpful. He and Doctor Elmes gave details of main problems associated with the assessment of the tremolite content and of the health problems associated with it. The Vanderbilt talc mine was quoted. Apparently this contained up .to 50Z of tremolite but much of it in a massive and splintered form. In Zimbabwe many serpentines were incorrectly labelled tremolite. Chrysotile in the Gobi Desert was said to contain up to 17% of tremolite. In conclusion. Sir Neville Stack undertook to report the substance of the meeting to the relevant Committees of the A.I.A. UCC 017481 [Ttbl0: ITT, II! CKRYSOTIi.g A. A. HODGSON It is established that tremolite fibre has high ranking' as a carcinogen, and its biological effects are strongly suspected in its association with chrysotile, vermiculite and talc, in some but not all occurrences. Tremolite is not now mined on a commercial scale, the largest producer in Korea having closed some years ago. Exploitable deposits occur in akistan, and in northern Italy where a valuable long fibre tremolite is believed to be extracted from tine to time. '-resencc of tremolite in chrysotile There is growing concern about the tremolite content of chrysetiles in commercial use, buit as far as is knewh studies have never been made tc establish the presence cf tremolite in chrysotile on a quantitative basis. It is the pvrpcse cf these notes to outline the properties and ether relevant information concerning both cf these fibres, in order to assist towards a quantitative scheme for separat ing and identifying the tremolite. The presence cf tremolite in chrysotile shoxild be regarded as ubiquitous, even to very small quantities, because the paxagenesis cf these minerals, which ay-lies also tc vermicv.lite and talc, leads to such a conclusion, as explained later. The ubiqtiity cf tremolite is not upheld by such qualitative researches as have been made cn the minerals associated with chrysotile. 3utier I9601 found tremolite in only 8 cut of 128 sarnies of various chrysetiles, and Gcsseye and Hahn-V/einheimer 1971" found tremolite in 1 cut cf 58 samples, that particular one not having been detected by Butler. Both these investigations were made by X-ray diffraction i.aRD}, Reasons for the sparse location of tremolite in these sam.les appear to be two-fold. Although tremolite may be widespread in chrysetiles and serpentinite rocks, it is \mlikely to be distributed uniformly. The very small samples taken for XRD could well be unrepresentative, and the chance that any one sample contains tremolite in detectable amounts nay be less than one in three. Secondly, and more importantly, the direct analysis, of such samples x by XRD is hampered ty detection limits for tremolite of about 1 . Fractions of tremolite less than this would be completely unobserved. Any method for the quantitative analysis cf tremolite in chrysotile must there fore have as its first stage a means cf concentrating the tremolite, to lift its content in any residues well above the detection limits of XRD and other techniques, and if possible tc concentrate it towards ICC UCC 017482 .-.elation to TP Vs The fcllowing table sets out the concentration of tremoiite in terns cf f/ml, which will apri;T at various levels cf Tj V for chrysotile and at various assumed tremolite contents. possible tremolite content cf chrysotile f' at Tl V for chrysotile cf 2 f/ml 1 f/ml 0.5 f/r trend ite count,f/nl, is: 1 C .02 0.C 1 C.CC-5 2 0.C 1 0.4,2 0.C1 5 i .If ( .03 c .025 1C C.2f L.1( c .05 For the most part the tremolite count is negligable and is below current VI Vs for crccidclite. However tremolite is the most intransigent of all asbestos fibres and this distinction is a natter of biological signifi cance in relation to chrysotile. Chrysotile fibre inhaled into the lungs can be martially cleared by a dissolution process, leaving any associated tremolite to become concentrated in lung tissue. Thus, while at the same time natural lung clearance takes place presumably at the same rate for both typ es of fibre, the relative proportions of tremolite and chrysotile will cl'.cnge, and the effective count of tremolite at lung burden jevei will progressively increase by several, if not many, times the figures calculated above. Geology and mineralogy Serpentines and ultimately chrysotiles are products of the thermal metamorphism of both ultrabasic rocks with high magnesium contents, and siliceous dolomites containing magnesium and calcium. Tremolite is an early product cf the metamorphism of dolomites (hence tremolitic marbles}, but equally it is an accessory mineral along with talc, brucite, magnetite and etc. in the serpentinization of ultra- basic rocks. The latter are usually lov in calcium content and consequently the generation of tremolite, of which calcium is an essential part, is limited. On the other hand the dolomites have high calcium contents, and ultimate serpentinization may be expected to yield proportionately high amounts of tremolite. There is a broad belt of serpentinized ultrabasic rocks in the Northern Hemisphere stretching from Canada to Russia and Siberia. The serpentines cf Zimbabwe have a similar origin. Serpentinized dolomites are character istic of the Mediterranean area, and also of the northern and eastern Transvaal in South Africa, Chrysotile deposits, large and small, are UCC 017483 AO 1303 2 located in all these geographical areas, and it follows that these deposits might be expected to contain greater or lesser proportions of tremolite, depending on their regional siting. Mining and milling Any chiysotile deposit can be expected to contain some tremolite. Apart from its random distribution throughout the deposit, tremolite may occur in pockets, fairly easy to distinguish and hence avoided in mining operations, since it would be considered to degrade the quality of the end product. The more general distribution of tremolite throughout a chrysotile deposit cannot be avoided, and any large scale separation of small amounts of tremolite from chrysotile fibre is hardly practical or even possible in the initial stages of extraction. However asbestos milling and grading processes must to some degree bring about a differentiation between the two types of fibre. The grading of asbestos fibre is essentially a serial screening and air-lifting process, with longer fibres being lifted first and shorter fibres passing each screen to the next stage. Tremolite has a distinctly higher density than that of chrysotile, and it follows that tremolite fibres have a lesser chance of being air-lifted than have chrysotile fibres, and similarly a greater chance of being screened out. Hence there should be some concentration of tremolite in shorter grades of chrysotile, and of course in tailings. Chemistry Tremolite and chrysotile are highly distinguishable from each other in terms of chemical and thermal properties. Tremolite is the most acid resistant of all asbestos fibres, while chrysotile possesses oomplete lack of acid resistance, and indeed can be leached by water. Much of the chemistry concerning the acid resistance of asbestos fibres is well knovn^, and it is only necessary here to summarize those aspects which may be of help in formulating a scheme towards a quantitative analysis for tremolite in chrysotile. The relative acid resistance curves for asbestos fibres refluxed in 4N HC1 show that ohrysotile decomposes with a 58 % weight loss in i hr., while tremolite a 4 % weight loss under the same conditions. All the MgO and contents of chrysotile are removed leaving a siliceous residue. The tubular morphology of the chrysotile remains, but the residue is amorphous 2+ to X-ray and EK beams. Tremolite may lose up to 22 5$ of its Ca content, these ions being picked off at the ends of oation chains. The loss of other ions is negligible. These are drastic conditions, and it is possible to reduce the concentration of acid without any effect on the decomposition of the chrysotile, but with *01304 UCC 017484 1 a reduction in the removal of Ca2+ from trem' olite. Rates of reaction are proportionately decreased with reduction in acid strength, although there is little difference in reaction rates in changing from IN to 0.1N HC1. This may be an important factor in setting up a scheme to extract chrysotile if many samples are to be involved, but at the same time low acid concentrat ions will have minimal effect on the decomposition of tremolite. The kinetics of the acid decomposition of chrysotile have been reported in detail by Atkinson and Rickards 1971 5 and by Nonkman 1971 o . In these investigations the chrysotile was ground to a high surface area, which to some extent may be a disadvantage when attempting to preserve the integrity of any tremolite which may be present. The rate of reaction of H^SO^ on chrysotile is slightly greater than that of HC1 of the same normality. In this context the choice of H,SC> may be the 2- ^4 better one, because the effect of the SO. ion may be to suppress the removal of Ca2+ from tremolite, CaSO. be*i*ng insoluble. ^T Chrysotile is vulnerable to alternating acid and alkaline conditions . The reaction of acid removes Mg2+ but leaves a layer of orthosilicic acid which effectively controls the rate of diffusion of the reaction acid into the inner layers of the chrysotile. The orthosilicic acid can however be removed in a consecutive reaction with NaOH to give sodium silicate in solution. After each consecutive stage in a aeries of such extractions the residues require to be washed to remove soluble magnesium salts and sodium silicate. In suit able conditions of acid concentration this may be a way of completely remov ing chrysotile with negligable effect on tremolite residues. Tremolite will not in any way react to alkaline solutions. Talc is a frequent accessory mineral in chrysotile and its presence may interfere with the final determination of tremolite contents. Talc is insoluble in acids and therefore will accumulate in extraction residues. If the talc content is high and the tremolite content is low, the former may obscure the latter in the final analysis. One refinement to the extraction method which might be considered involves the heating of the sample specimen. Chrysotile dehydroxylates and forms a highly disorganised forsterite at about 600C; talc begins to dehydroxylate at about 700C; and tremolite remains stable up to about 950C ( a higher temperature than any other type of asbestos). The advantage of pre-heating chrysotile to 600C lies in ItB extreme vulnerability to acid decomposition, since there is little residual crystal structure left at this temperature. dO1305 UCC 017485 I Further, regarding the presence of talc, this mineral can be assessed quantitatively byIE analysis, and it has been reported that the IR spectra of many of the minerals associated with chrysotile are rendered less complex by pre-heating them to certain temperatures8 . Other -possible extraction techniques The density of tremolite is 2.9 to 3.2, that of chxysotile 2.55 It is therefore experimentally possible to differentiate between the two minerals by heavy liquid separation, using a medium of suitable intermediate specific gravity. It would be usual to apply this technique to relatively small samples and to make a final separation of the heavier mineral by centrifuge. Complete separation of tremolite from chrysotile might be difficult because of the entanglement of one type of fibre with the other. Again talc may interfere, and samples containing talc may require a secondary heavy liquid separation. The density of talc is 2.65 to 2.9, indicating a slight overlap here between the densities of talc and tremolite. However, the method may be useful as a preliminary assay of samples, to determine approximate tremolite contents. Following similar lines, it should be possible to obtain a separation of the two minerals, chxysotile and tremolite, by a bench scale flotation process. It is relevant to note in this connection that Cyprus. Industrial Minerals Corp. in TJSA have used a flotation process for the beneficiaticn of New o York state talcs, which are notorious for their high tremolite contents . Anionic surfactants will disperse chrysotile fibres and will also effect a partial, spontaneous, fibrillation of the asbestos. In principle this should lead to the release of tremolite, and permit the separation of both chrysotile and talc by a flotation method. The application of anionic surfactants to chrysotile is an essential step in the wet dispersion process for making chrysotile yarns, as outlined by Gettins and Mallon 1971^, Heron and Huggett 197111 (all four authors associated with TBA), and Otouma and Take 1975 12 . For the purpose of this present exercise it will be use ful to approach TBA to ascertain whether there is any feasibility in- applying wet dispersion and flotation techniques to the separation of tremolite from chrysotile. Identification of tremolite by UV light The Cyprus Ind. Min. Corp. patent9 , referred to above, cites an optical sorting method in which talc ore contain ing tremolite is sorted under UV radiation of wavelength 250 to 340 nm, in which the talc fluoresces white and the tremolite fluoresces orange or red. The principle slight be applied to chxysotile/tremolite mixes to give a preliminary indication of high or low levels of tremolite content. AO 1306 UCC 017486 6 Outline schemes for separating tremolite from chrysotile On face value, a chemical separation scheme appears to he absolute, but if many samples are to be investigated it will be tedious and time consuming, unless an automated means of extraction of the chrysotile can be devised. Choice of sample sizes will be important, due to the uneven distribution and variable content of tremolite in chrysotiles from different sources. A preliminary assay of samples by heavy liquid separation or by UV screening (if practical) may be necessary. The following steps are foreseen in setting up an analytical procedure. 1. Confirm experimentally that chrysotile fibre can be fully decomposed into soluble products, by using alternating acid and alkaline conditions, and by selecting an optimum reaction regime as indicated in the literature. Assess the decomposition of pure tremolite under the same conditions. 2. Determine what advantages lie in pre-milling or pre-heating the samples. 3. Investigate a procedure for the preliminary assay of samples by heavy liquid separation or by UT screening. 4. Draw up plans for an automated bench rig to decompose the chrysotile and leave a residue of tremolite. 5. Prepare an order of sample sizes (l to 10 gm ?) according to likely tremolite content, and according to choice of final quantitative assessment. 6. Investigate the interference due to other insoluble minerals, particularly talc, in the final assessment of residues. 7. Consider the choice of one or more methods for the final analysis of residues XRD for positive identification of tremolite from all samples; IR for quantitative assessment of talc in the presence of tremolite; gravimetric evaluation of large residues; EDXA (Ca content) for the evaluation of tremolite in small residues. Two further approaches should be considered: (a) If chrysotile samples contain considerably more than 1 % tremolite, it should be possible to analyse them directly by XRD/EDXA, particularly if a set of calibrated mixes has been prepared and examined beforehand. To obtain a statistically acceptable result it would be necessary to assess a number of aliquots of the unknown. (b) The feasibility of a flotation technique for separating tremolite and chrysotile should be examined as a separate issue. A flotation method would replace chemical separation, and could probably be the means of separating off any interfering talc along with the chrysotile. The T&N A 0 1 30 7 UCC 017487 2 elutriation method., as used in the testing of asbestos fibres, could probably be adapted without difficulty to the partition of chrysotile and talc from tremolite, after preparation of the samples with suitable surfactants. References 1. M. A. Butler. The physical and chemical characteristics of serpentine rocks and minerals. Phi) thesis. University College, Cardiff, 1980. 2. A. Gosseye, P. Hahn-Weinheimer. A comparative study of asbestos minerals. 2nd International Asbestos Conference, Louvain, 1971* 3. E.J.W. Whittaker (Oxford University). Advisory*communication to Cape Asbestos Fibres Ltd., 1978. 4. A. A. Hodgson. Fibrous Silicates. Royal Institute of Chemistry, 19&5* 5. A. V. Atkinson, A. L. Rickards. Acid decomposition of highly opened chxysotiles. 2nd International Asbestos Conference, Louvain, 1971* 6. L. J. Monkman. Seme aspects of the reaction of chrysotile with inorganio and organic acids. 2nd International Asbestos Conference, Louvain, 1971* 7. P. F. Holt, S. G. Clark. Nature, Lond., i960, 185, p. 257. 8. E. Martinez, R. B. Haagensen, H. L. Lovell. Investigation of the quantitative determination of minerals in serpentine and asbestos samples by infra-red spectroscopy and thermo-analysis. 1st International Asbestos Conference, Oxford, 1967* 9. US Patent 3837582. Beneficiation of New York State talc. Cyprus Mines Coxp., 1974. (now Cyprus Industrial Minerals Corp.) 10. R. B. Gettins, F. J. Mallon. Adsorption of surface active agents on chrysotile asbestos. 2nd International Asbestos Conference, Louvain, 1971 11. G. F. Heron, R. Huggett. Dispersion based textiles. 2nd International Asbestos Conference, Louvain, 1971* 12. T. Otouma, S. Take. Effect of anionic surface active agents on chrysotile 3rd International Asbestos Conference, Quebec, 1975- Proceedings of the International Asbestos Conferences are not readily available and the authors referenced above did not necessarily publish elsewhere. CopieB of the relevant papers can be obtained from AAH if so desired. A. A. H. 30.10.1984. Copies: Chairman, ARC Research Committee Institute of Occupational Medicine 5 2 UCC 017488