Document ExKGe20BVX7EwXgYOQxkrD45j

Engineering from the U.S. Patent Office that approval of the patent application on the process will be granted. Drive for Completion: Stauffer Chem ical js looking into new methods of clean ing up monomer from vent gas streams, but also is trying to get down to funda mentals. Executive Vice-President Har old S. Mickley says the company has had a research and development task force at work for months trying to find answers to basic questions such as: "Can you drive the polymerization reaction to comple tion?" If it is possible to achieve complete polymerization, the problems of down stream cleanup can be eliminated. Walter Schlegel, assistant process engi neer at Crawford & Russell (Stamford, Conn.), explains that major trouble spots are at four primary points in the manu facturing process: (1) slurry holding SKID-MOUNTED monomer recovery unit is headed for installation at Tenneco vinyl plant tanks, (2) dryer, (3) reactor (during rou tine cleaning between batches), (4) recov ery system for process vent gases. Putting VCM emissions on skids `Theoretically, you could blow a hurri cane of air through the plant and elimi nate the worker exposure problem, but New methods of controlling VCM emissions are being developed, that would only transfer the problem but allowable level in air still has not been set by OSHA somewhere else," says Stauffer's Mickley. In fact, the cleanup problem is compli With the final standard for workers' ex sorption from vent gases. The monomer cated, since vinyl makers will have to posure to vinyl chloride monomer sched is regenerated with steam. But vinyl pro meet Environmental Protection Agency { \ uled for promulgation next month by the ducers say they can't determine how well regulations on air quality' outside the Labor Dept's Occupational Safety and the system will work commercially until plant, as well as OSHA's standard for Health Administration, engineering firms and vinyl producers are hard at work ex it is proved in major demonstration units. worker exposure. EPA's standard, which One such unit, designed, fabricated was to have been issued on Aug. 15, still ploring alternative methods of reducing plant emissions. Because no one knows how low the and skid-mounted by Chemical Design (Lockport, N.Y.), is currently being in stalled at the new polyvinyl chloride was not ready at CH^s press time. The agency has monitored plants in many locations, has indicated that air standard will be, the best that plant man agement can do right now is to improve housekeeping procedures (e.g., tightening leaks). But they have a good idea of what they must shoot for in developing new techniques. OSHA's proposed standard is plant Tenneco Chemicals is building at Pasadena, Tex. (also see p. 80). Tenneco's laboratory work indicates that almost all the vinyl chloride monomer can be re moved from purge gas streams--with a level of 5 ppm. "possibly" attainable. quality seems satisfactory, so the cleanup approach is centered on bottling up the VCM as completely as possible within the process. Id the Plant: PVC resin is produced by four polymerization processes: suspen . "no detectable level," which translates to 1 ppm. because of present limitations of measuring equipment. The industry says it would be forced to shut down if such a standard were en forced, contending that a 1-ppm. limit is technically impossible to meet. It's claimed that 1.7-2.2 million jobs would be eliminated, as would production val ued at S65-90 billion/year. Through the Manufacturing Chemists Assn., the in dustry has recommended a maximum emission limit of 25 ppm. which would be reduced to 10 ppm. in three years. Ready for Trial: Among the techniques now under trial is activated carbon ad Whether the same result can be achieved on a commercial scale remains to be seen. Tenneco must obtain an operating permit from Texas Air Pollution Control Board; no problems are expected. Another method of controlling VCM emissions is offered in the new PVC man ufacturing process developed by Robintech (Ft. Worth). The technique, called Rocor IKC (in-kettle compound), com bines the two commonly used steps--resin making and incorporation of additives. The company is close-mouthed about details, but claims that PVC made by the process virtually eliminates worker expo sure to VCM at the fabrication point. sion (which is used in more than 75% of the tonnage now made), emulsion, solu tion and bulk. Polymerization takes place in a reactor (usual capacity: 2,000^5,000 gal., although newer units hold as much as 35,000 gal.). Reaction temperatures are 120-150F; operating pressures, 110-160 psig. The reaction proceeds to about 8590% VCM conversion, at which time un reacted monomer is removed from the unit. (All plants have VCM recovery sys tems composed of a series of compressors and condensers.) The polymerized slurry is removed from the reactor, dewatered in a centri fuge and then dried with a stream of hot sorption. Calgon, for example, has re ported successful studies of monomer ad Robintech uses the method in its Painesville, O.. plant, says it has received word air. All reactors must be cleaned, which is 66 CHEMICAL WEEK September IS, 1974 BFG26610 '1 0 0 4 T 9 S 2 Engineering one point of possible worker exposure. Gearung frequency ranges from every five batches to 50. Older reactors often are scraped manually. But C&R's Schlegel says that this cleaning method is giv ing way to high-pressure water and wa ter-detergent systems. Some plants use solvents, such as tetrahydrofuran, dimethyl formamide. or ethylene dichloride, to remove baked-on polymer and absorb unreacted monomer. But this sub stitutes one problem for another: the solvent must be recovered. The new, large (e.g., 35.000 gal.) reac tors are more easily cleaned than small ones because water jets can be positioned more effectively. Also, larger quantities of resin can be-produced before cleaning is necessaty. Perhaps the biggest source of head aches, however, is the cleaning of the huge quantity of air that has been used in drying the resin. For example, a plant may use 40,000 cu.ft./minute of air. The air may be contaminated with only 10-15 lb$./hour of monomer that must be re moved after it leaves the rotary dryer. To overcome this problem, vinyl producers are working on methods of removing monomer before the resin reaches the dryer. Chemical Design suggests another ap proach: dehydrate air from the rotary dryers to ambient air moisture level and recycle it to the dryers. A bleed of 2,0002,500 cu.ft./minute could be removed from the recycle stream for cleanup in an activated carbon adsorption system. The company says that without dehydration, an adsorption system to clean up 70,00080,000 cu.ft./minute of air contaminated with 250-300 ppm. of VCM would re quire a capital investment of almost S3 million. On the other hand, an adsorption sys tem for 2,000-2,500 cu.ft./minute of bleed air contaminated with 665 lbs./ hour of VCM would cost $600,000700,000. Dehydration equipment would add $300,000-500,000 to the cost. The final vent gas streams from vinyl plants can be refrigerated to remove most remaining VCM, but other methods are being investigated. Remote Operation: Some of the new processing methods that undoubtedly will be developed during coming months may involve remote operation of plants. Ac cording to Robert A. Allen, vice-presi dent of World-Wide Construction Ser vices (Wichita, Kan.), technology has been developed to operate new, large re actors from remote, isolated control rooms. The control areas could be 300 ft. or more from the reaction area. However, Allen told a group of engi neers at an American Institute of Chem ical Engineers conference on hazardous materials last month that remote oper ation could create new problems. Rea son: operators would no longer be close to potential trouble spots. He said these problems might be ofTset by use of closed-circuit television, highly sensitive vapor detectors and microphones that could provide audiomooitoring in the control room. Vinyl producers already are spending heavily for new monitoring devices, in addition to following general house keeping procedures recommended by the Dept, of Health. Education and Welfare's National Institute for Occupational Safety and Health. NIOSH developed these recommendations for B.F. Good rich's Louisville, Ky., plant, where the VCM-cancer link was first discovered in workers. Of the 24 cases of angiosarcoma, a rare liver cancer, that have been reported, oaly one involved a worker at a mono mer plant (in Sweden). Of the other cases, 20 have been in PVC plants, two in compounding and fabricating operations and one (in West Germany) in an aerosol pesticide container-filling operation that used VCM as a propellant. No one seems to be certain of the max imum dosage of VCM that produces can cer. Joseph Wagoner, chairman of NIOSH's occupational carcinogen com mittee, says, "I'm sure 25 ppm. will show cancer." And he warns that there is no way to demonstrate the safe level of a carcinogen (CW, July 31, p. 19). But whether OSHA decides to set the standard at the level recommended by in dustry, or adopt one that is considerably lower, vinyl producers will have to de velop production methods and VCM re covery systems that will fill the bill. Plant shutdowns would seem to be the only other alternative. Producers and representatives of labor agree that self-contained air-supply packs, similar to those used by astronauts, could not be used in vinyl plants because the workers couldn't function. And there is no substitute material available to take PVC's place for many applications. As Arthur D. Little (Cam bridge, Mass.) puts it: "It would not be possible to duplicate the exact properties of vinyl plastic products by using substi tute raw materials." Whether PVC users will be forced to face up to making do with substitute ma terials remains to be seen. 236T7CD2 r " "" \ specializing in the commercial production of organic di valent sulfur compounds can make it tor you Ur- - t- /' MERCAPTO COMPOUNDS SULFIDES -v MERCAPTO ACIDS THIOLS THIOL ACIDS -V THIO COMPOUNDS Whatever your requirements --hundreds to hundreds of thousands of pounds--check with Evans! Evans can probably make it for you! Send for catalog listing chemicals we make on a commercial scale. CHemETics, irvc. 90 Tokeneke Road Darien, Connecticut 06820 Phone: 203-655-37*1 CaBie EVANSCHEM TWX: 710-457-3356 - -- --- September 18. 1974 CHEMICAL WEEK 71 BFG26611 Top of the news Showdown on vinyl plant rule presages shutdowns Newly issued advisory reports hint at compromise in fight to curb vinyl chloride fumes, but battle may continue in court. Industry insists OSHA's proposal would be ruinous The bitter controversy over worker ex fects of the proposed standard. This re els; and there are no technology transfer posure to vinyl chlonde gas may climax port does not discount the possibility of a opportunities to permit development of oq Oct. 5 when the U.S. Dept, of Labor is level higher than the 1 ppm. implicit in design for no detectable levels in new expected to set permanent standards for "no detectable level.1' But it concludes plants. Moreover. Snell says that because the 17 U.S. plants that produce the mono that anything higher is a nsk to workers. PVC manufacture is a batch operation, mer, the 37 that polymerize it to make "Despite significant cost increases that emissions couldn't be completely elimi polyvinyl chloride resins, and the thou may occur, the high demand for PVC nated. sands of plants that mold, extrude and resins is likely to continue,'' the report Something to Live With: Those conclu otherwise shape those PVC resins into states. sions would seem to support a compro plastic pipe, vinyl flooring, house siding The Environmental Protection mise regulation that the vinyl makers and phonograph records. Agency's notification that within nine could live with--perhaps not too happily. The controversy has raged since last months it will set VCM plant emission Rut probably because of the imminence January, when medical data apparently standards that could add 4% to the cost of of the Labor Dept.'s new regulations, in established a link between workers' expo PVC. This would seem to preclude the dustry representatives were reluctant to sure to vinyl chloride monomer (VCM) engineering possibility of venting ail the discuss Snell's final report noting that and angiosarcoma, a rare liver cancer. in-plant VCM leaks to the atmosphere. they haven't had rime to examine it And last week the scores of chemical Double Negative: OSHA's proposed Nor did they have much time to react and plastic companies that have a stake goal of operating monomer and polymer to Snell's preliminary report, which con in this 53-billion/year business were plants at no detectable level of vinyl chlo tained pretty much the same data used in given some inkling of whether the gov ride exposure is "not feasible," according the final report but did not contain any ernment will stay with the 50-ppm. emer to Snell. Moreover, the firm's analysis in conclusions. OSHA issued the prelimi gency standard, adopt the proposed "no dicates that the U.S. industry would com- nary report Aug. 28, called for response detectable level" by Sept. 6. Never- limit or move to --^m theless, industry some compromise position. Whatever OSHA's vinyl standards impact: higher costs, fewer plants leaders submitted comments, and these standards are set could determine the profitability of pro ducing and process ing vinyl compounds Proposed VCM exposure standard* (ppm.) Timeweighted Maximum average Increase in average unit costs (centx/lb.) Loss of Direct productivity Total* Time needed to comply (months) Proportion of industry endangered* were generally nega tive and caustic. One point of con tention in the prelim inary report was the in the U.S. for many Monomer plants use of Firestone data years and are likely to be attacked in the federal courts, either by the companies or the labor unions in 50 15 25 15 10 2-5 "No detectable amount 0.04 0.09 0.17 - 0.03 0.03 0.15 - 0.07 10% 0.12 40% 0.32 40% 0 24 36 Not fessibie 0% 0% 0% 100% for bringing a PVC plant into com pliance with a no-detectable-Ievel stan dard. Firestone, Snell volved. Polymer plants reported, was the Conflict: As to 50 -- 0.19 0.35 0.54 20% 0-6 Negligible only company that exactly what stan dards will be promul gated next week, various indications-- 2540 15-25 15-25 10-15 "No detectable amount" 0.36 0.90 4.01J 0.49 0.70 3.07* 0.85 35% 1.60 t 50% 7.08 75%* 30 30 3048* 4% 26% 100% said it could meet such a standard. Firestone replied that studies of its poly sometimes contra dictory-can be * Vinyl chloride monomer (VCM) content, expressed es certs per million, in the air to which employees ere exposed. `With estimated range of variance for individual plants, based on partial statistical analysissof data. 'Percentage of capacity chat management would seriously merization plants in Pottstown, Pa., and found in three newly issued reports: eonsidar shutting down. 3 Based on firestone's estimates: actual compliance is not considered feasible. Source: foster D. Snail. Inc. Perryville, Md., accually indicated a Foster D. Snell's contrary judgment. final report to the Occupational Safety pletcly collapse under that restrictive Declares a Firestone executive: "Inde and Health Administration on the eco standard, with all producers indicating pendent of economic considerations and nomic impact of the agency's proposed VCM exposure standards. A principal they would have to close their plants. despite enormous equipment, engineer The study offers several reasons for ing and process changes at both facilities, conclusion: "achieving no detectable lev reaching these conclusions: technology to costing in excess of $50 million, it is not els .. . is judged not feasible, both in PVC and VCM sectors." eliminate VCM leaks and fugitive losses possible to achieve the drasuc exposure is not available; existing plants have not levels required by the proposed standard, The Occupational Health and Safety been designed for total VCM contain based on engineering controls and tech Administration's final environmental im pact statement on the physiological ef- ment; there has been no operating expe rience in maintaining very low VCM lev nology that now exist." Firestone. No. 5 in PVC output, has September 25. 1974 CHEMICAL WEEK 1*5 BFG26612 2361801U Top of the news tome differences with the rest of industry od the exposure level standards that would be manageable. Industry generally supports the Society of the Plastics Indus* try's proposal: that the present temporary emergency standard of 50 ppm. be low ered in several stages to 25 ppm. for PVC plants by Oct. 5, 1975, with further impo sition of a time-weighted average limit of 10 ppm. Standards for VCM production, says SPI, could be more restrictive--10 ppm. by October 1975, and a timeweighted average (twaj of 5 ppm. T.C. Walker, president of Firestone Plastics, says that a "safe and medically sound--as well as technologically and ec onomically feasible--standard" is 40 ppm. maximum exposure and 25-ppm. twa. Snell's analysis offers data on several alternatives, maintaining the SPI distinc tion between VCM and PVC plants {table, p. IS). Cost data indicate that VCM makers can manage with a 10-ppm. limit and a 2-5-ppm. twa. without closing any plants. Meeting such a standard would increase their average unit manu facturing cost by an estimated 0J2c/lb., which is about 4-5% of the present range (7-10e/lb.) of list prices for various grades of VCM. For polymer producers, however, some plant closings may be necessary when the present emergency standard is tightened. According to the Snell study, at an expo sure limit of 25-40 ppm. and a 15-25 twa., 4% of the existing PVC plants may have to be shut down. This might not be too disastrous, considering that 58% of the in dustry's plants are more than 10 years old, and only 8% are less than five years old. But for an exposure limit of 15-25 ppm. and a twa. of 10-15 ppm., 26% of the PVC plants could face the axe. Snell's survey indicates that with this standard, unit costs of PVC would increase 1.64/lb., or about 4-8%, at those plants that can take the needed engineering changes. At some older, smaller batch op erations, the companies say it would be impossible to reach such a low exposure level without building a new plant. At certain plants, Snell believes, the addi tional cost might be 50% more or less than the l.6/lb. average. Schedule Makers: Snell estimates it will take VCM plants up to three years to achieve what it considers the lowest achievable level. For PVC plants, the es timate is two and a half years. This is based on engineenng estimates of installation time for various kinds of apparatus to reduce exposure levels at 16 CHEMICAL WEEK September 25, 1974 PVC plants. Examples: major ventilation system improvement, 18 months: replace ment or reconditioning of reactors, 18-24 months. Overly Optimistic? Some companies contend that these estimates are overly optimistic in light of today's materials shortages and delivery delays. Observes a Firestone executive: "Individual com pany confidence in being able to attain compliance with even the SPI recommen dations has been eroding as lead times for obtaining equipment expand, as capi tal costs rise rapidly, and as the prospect for financing these projects becomes dim, mer." meet new in-plant and emission standards. Industry criticism has also centered on Snell's projection of cost data to meet projected standards. The figures are said to be heavily weighted in favor of newer plants and suspension homopolymerization units in wanner climates. Such plants could probably be most readily adapted to meet a more stringent exposure stan dard. Separate Standards: The proposed nodetectable-level standards would be ap plicable to VCM and PVC plants as well as fabricators. But the Snell study follows the SPI proposal of separate standards for VCM and PVC plants. Reasoning is based on relative case of controlling emissions at VCM plants, which are generally closed systems, oper ating in the open air. Leaks mav occur around pump packings, during samplings or maintenance operation, or at tank-car loadings. The batch operation of PVC plants involves continuous emptying and recharging of autoclaves and frequent in spections and cleaning out of these reac tors. Chance of VCM exposure in these plants ts high even in the packing areas. While separate standards would seem reasonable to manufacturers, the prac ticality of this approach is limited. It's un likely that labor groups, strongly in favor of the no-detectabie level and opposed to (he concept of a threshold level for any carcinogen, could accept different expo sures for workers in various plants. Outside Plant: If the vinyl producers solve their in-plant exposure problems, they will still have to deal with monomer gas escaping from the plant. EPA Admin istrator Russell Train said last week that his agency will set standards for vinyl chloride plant emissions by next June. And without firm data on health effects of VCM at various levels in the atmo sphere, Train is proposing that the stan dards be based on "best available tech nology." EPA's survey of air and water around 10 PVC plants indicated VCM air levels of 1 to 2 ppm. Through those emissions, the agency estimates, PVC plants are los ing monomer at the Tate of 200 million Ibs./year. With controls, EPA says emissions from PVC and VCM plants could be cut 75% and 90%, respectively. Capital out lays for these controls are put at S80 mil lion for the industry as a whole, with op erating expenses estimated at some $40 million/year. John Nelson, vice-president for manu/acturing at B.F. Goodrich Chemical, ap proves of Train's analysis, especially the statement that there is "no imminent haz ard" from VCM exposure in the atmo sphere. But if OSHA's no-detectable-level pro posal becomes the law, the vinyl chloride industry might have to re-evaluate its do mestic production. Although industry spokesmen have warned of extensive shutdowns if such a standard is adopted, OSHA's environmental impact statement merely suggests that the costs will be high but will be passed on to the consumer, and that the time needed to reach the standards would be long. But this report also suggests that "in the long run U.S. companies might find it profitable to build plants abroad." Clearly, this will depend not only on tra ditional market and raw-material logis tics but also on whatever in-plant and emission standards are put in force in the U.S. and other nations. BFG26613 Zi)l)91VZ i op ui u ic i icwa New suit for dye makers Calina Industries, a New York-based warp-knit fabrics manufacturer, filed suit this month against eight of nine big dye makers indicted for price fixing by a fed eral grand jury in July (CW Business Newsletter, July 24). The Calina suit is a class action~brought on behalf of "ail pur chasers of dyes from one or more of the defendants or their coconspirators" since early 1970, Defendants include Du Pont, Baychem (now Mobay Chemical), Allied Chem ical, American Cyanamid, BASF Wyan dotte, Ciba-Geigy, Crompton & Knowles and GAF. American Color & Chemical (jointly owned by North American Phil ips and Koppers) was the only company not named in the civil action. Like the federal complaint, Calina's suit charges that Du Pont officials began , sounding out?other dye makers on the possibility of fan across-the-board price boost in 1970. and that the defendants, along with some other firms, conspired to push through price increases shortly thereafter. .? Calina says that as a result of the al leged conspiracy, it and others in the class action have < been,' and continue to be, overcharged for the dyes they purchase. The suit seeks treble damages and urges the court to enjoin the dye makers from future conspiracy on pricing. A spokesman for Du Pont says his company has not had time to thoroughly study the Calina complaint The primary area of concentration is still the federal complaint he adds. Lawyers for Calina declined to give further details. Bitter news for Abbott Last..November, Abbott Laboratories called a press-conference to report that the company was formally petitioning the Food and Drug Administration to reverse its 1969 ban on marketing of cyclamates {CW, Nov. 21TI972rp.-lS). Spokesmen presented 16 volumes of test reports-- stacked impressively on a table--which they said would convince FDA that cyclamates were safe. Last week, however', Abbott's high hopes were jolted. FDA informed the company that it considered the test re ports inconclusive and recommended that Abbott withdraw its petition. FDA said it wanted more convincing data, such as lifetime animal feeding studies, which could require several years to conduct. "We have concluded that questions about the cancer-caustng potential of cyclamatcs still arc unresolved," FDA^ Commissioner Alexander Schmidt said. * Abbott Executive Vice-President Charles S. Brown responded that the ad ditional research requested by FDA is "neither reasonable nor required by sci entific judgment." He requested a meet ing with FDA officials as soon as pos sible. Partly in anticipation of a favorable de cision by FDA, Abbott restarted cycla mates production at its North Chicago, 111., plant in January. Most of the output so far has gone to customers in several European countries where no bans exist These customers had been supplied from pre-ban stockpiles. FDA's thumbs-down may short-circuit Abbott's plans to send similar petitions to governments of other nations that have banned cyclamates. The company had been holding off until it got a reply from FDA. Potential cydamate sales--which the company had expected to reach $16 million/year, the pre-ban level--have not been included in sales estimates for 1974 or 1975, Abbott said. No letup in price hikes The wholesale price index for chem icals and allied products climbed to a sea sonally unadjusted 158.5 in August from 148.4 in July (1967-100), the Dept, of Labor's Bureau of Labor Statistics re ported last week. The corresponding Au gust 1973 figure was 111. Leading the parade in price hikes for specific chemical commodities were in edible fats and oils, which moved from an index of 347.4 in July to 380.2 in Au gust, and plastic resins and materials, which rose from 147.5 to 160.7. Other big gainers were agricultural chemicals and products, which jumped from 131 in July to 142 in August; and in dustrial chemicals, from 155.5 to 167.8. Some CPI segments registered more modest price gains. Examples: drugs and pharmaceuticals edged up to 115.3 in Au gust from 112.7 in July; and prepared paint to 152.3 from 149.7. One commod ity that scarcely changed was paint mate rials: to 155.9 from 155.4. Among other related products, pulp, paper and allied products lose from 1533 in July to 162.9 in August; nonferrous metals, from 198.4 to 200.4; refined pe troleum products, from 239.4 to 243.9. For ail commodities, the wholesale price index moved up to 167.4 in August, 17.8% above the August 1973 level. Compliance complaint On Oct. 5, the Occupational Safety and Health Administration is slated to issue a standard for worker exposure to vinyl chloride {also see p. 66). As that date ap proaches, controversy is erupting over an OSHA-sponsored economic impact re port, issued in preliminary form late last month by Foster D. Snell. Inc. Snell polled 23 polyvinyl chloride pro ducers on the costs of complying with various hypothetical exposure standards and indicated (hat imposition of the "no detectable level" standard, proposed by OSHA in May, might well close down the entire industry. The final Snell report is scheduled to reach OSHA this week. Firestone Plastics has asked OSHA for a chance to cross-examine the Snell re searchers, charging the report misrepre sents the firm's assessment of compliance costs. Firestone's protests won't delay is suance of the standard but could raise the level of controversy. "Our biggest criticism of the report," says Firestone Plastics President Todd Walker, "is that it said Firestone was the only company with the ability to meet the no-detectable-level standard. The fact is that our whole testimony at the [OSHA] vinyl chloride hearings was directed at saying it couldn't be done." Walker also criticized Snell for coming up with "no hard data." Snell Vice-President Joseph Nemec, however, says his firm will have no re sponse until after its completed report is in OSHA's hands. According to the preliminary report, all 23 PVC companies polled said they could reach a 50-parts-per-million standard im mediately; five said they could manage 25 ppm. by the end of 1976, seven by the end of 1977 and four by the end of 1978. Meanwhile, Ralph Nader's Health Re search Group has criticized the report as appearing to accept industry's blanket as sertion that the no-detectable limit is not feasible and that it focused too much on the direct impact on the PVC makers, not on the over-all economy. "This myopic concentration on compliance costs disre gards entirely the personal, social and even industrial costs of continued dis ease," the group maintained. Odds seem high that OSHA eventually will issue a 5-pptn. standard. Neither in dustry nor labor is expected to be satis fied with such a decision. `There's a lot of pressure on OSHA from the top to ease off the no-detec- table-lcvel standard," says one AFL-CIO health official. 22 CHEMICAL WEEK September 18. 1974 SVC)26614 23613001