A Survey of Exposure to Diesel Engine Exhaust Emissions in the Workplace

Transcription

1 AIVC #13,226 Pergamon A1111. oa11p lfyg Vol. 44, No. 6, pp , 2 Crown Copyright &:> 2 Pblished by Elsevit!r Science Ltd on behah of British Occpational Hygiene Society. All rights reserved PH: S3-4878()2-8 Printed in Great Britain, //$2. A Srvey of Exposre to Diesel Engine Exhast Emissions in the Workplace JOHN GROVESt* and JOHN R. CAINt i'health and Safety Laboratory, Broad Lane, Sheffield S3 7HQ, UK; thealth and Safety Exective, Magdalen Hose, Stanley Precinct, Bootle, Merseyside L2 3QZ, UK Forty sites were visited dring a srvey of exposres to diesel engine exhast em1ss1ns. Personal and backgrond exposre to gaseos components, respirable dst, elemental carbon, organic carbon and total carbon were measred and details of control systems were recorded. The reslts show a wide spread in exposre patterns reflecting the different work practices, job categories of employees and the control methods sed. However, sites where fork-lift trcks were in se consistently prodced the highest exposres. The srvey reslts sggest that the measrement of elemental carbon cold be sed as an indicator of exposre to diesel engine exhast emissions. Crown Copyright 2 Pblished by Elsevier Science Ltd on behalf of British Occpational Hygiene Society. All rights reserved. Keywords: diesel engine exhast; elemental carbon; ventilation; fork-lift trcks; ro-ro ferries; garages; toll booths INTRODUCTION Diesel engine exhast is a complex mixtre of gases and particlates. The principal gaseos components are carbon dioxide, carbon monoxide and nitrogen oxides while the particlate fraction mainly comprises fine carbon particles formed by incomplete combstion. The carbon particles are mixed with organic vapors and gases derived from oil, nbrned fel and prodcts of combstion and, as the mixtre isses from the engine, it cools and the higher boiling organic materials condense onto the carbon particles. 'Diesel engine exhast emissions (DEEEs) are classified by the International Agency for Research on Cancer as a Grop 2A carcinogen, 'probably carcinogenic to hmans' (IARC, 1989). Bt DEEEs are not considered to be carcinogens for the prposes of the Control of Sbstances Hazardos to Health (COSHH) Reglations 1999 (HSE, 1999a) becase they are not considered to meet the criteria Received 19 Janary 1999; in final form 21 December 1999, *Athor to whom correspondence shold be addressed. Tel.: ; Fax: for Category 1 or 2 carcinogens set ot m the approved gide to classification and labelling spporting the Chemicals (Hazard Information and Packaging for Spply) Reglations 1994 (HSE, 1994). However, DEEEs are regarded as 'a sbstance hazardos to health' and for this reason are sbject to the general provisions of the COSHH Reglations (HSE, 1999a) which reqire that exposre be prevented or, where this is not reasonably practicable, adeqately controlled. Certain components of the DEEEs sch as the individal gaseos components are covered by COSHH and have Occpational Exposre Limits listed in EH4 (HSE, 1999b). The role of the varios diesel exhast fractions in casing ill health is not clear. The irritant effects of diesel exhast are well known bt mch attention has also focsed on the carbon particles and their adsorbed organics, particlarly in Germany and the USA. In a search for an index of exposre to diesel exhast, the Health and Safety Exective sponsored work (to be pblished) which reslted in a measrement method for carbon particles and adsorbed organics similar to that sed in Germany (Dahmann et al., 1996). This method was sed to carry ot a srvey of indstry with two objectives: 435

2 436 J. Groves and J. R. Cain to obtain qantitative information on workplace exposre to diesel engine exhast, and to assess the range of measres sed to control exposre to diesel engine exhasts. The otcome of that srvey is reported in this paper. An alternative method for measring elemental carbon derived from diesel exhast was pblished (NIOSH, 1998) after this srvey was carried ot. This method ses a thermooptical detector which is said to be able to correct for carbon formed pyrolytically dring analysis. Recent stdies have compared performance between different laboratories sing the colometric method nder different operating conditions (Gillemin et al., 1997) and laboratories sing the colometric and thermo-optical methods (Birch, 1998). METHODS Srvey strategy Sites were selected in the UK to provide a crosssection of typical occpational exposres to diesel exhast emissions. After preliminary enqiries, 4 premises were selected for inclsion in the srvey and were classified, on the basis of similarity of operation, into the grops shown in Table 1. Grops 1-6 were reasonably well defined categories while grop 7 comprised a miscellaneos collection of operations which did not fit neatly into the other six grops. Each site was visited to gather the following information: descriptions of the workplaces and the types of controls in operation (for example local exhast, respiratory protective eqipment); details of relevant work activities inclding range and timing of employee's work, and estimates of exposre to diesel exhast mainly by personal sampling bt also spported by backgrond samples where appropriate. Sampling and measrement At the time of this srvey, it was not clear which Table I. Sampling sites and job grops Grop Site No. Description of site 1 Amblance depot 2 Ri'Mo ferry 3 Railway repair/stations 4 Bs garage and/or bs repair 5 Vehicle testing 6 Fork-lift trcks 7 Toll booths, set-down area, tnnel repair sites l Regional amblance depot Local hospital amblance depot Ro-ro ferry loading/nloading Ro-ro ferry loading/nloading Ro-ro ferry loading/nloading Railway station Railway station Bs garage rn-in/rn-ot Bs garage rn-in/rn-ot Bs garage rn-in/rn-ot Bs garage rn-in/rn-ot Bs garage rn-in/rn-ot Vehicle testing depot Vehicle testing depot Inspection/roller test (vehicle manfactre) Warehose Warehose Trck loading depot Warehose/assembly line Toll booth Tnnel repair/maintenance Toll booth/tnnel repair Airport coach set-down area Airport cargo post Airport tnnel roadway Airport baggage handling area Ferry terminal gate cabin

3 Diesel engine exhast emissions in the workplace 437 sel 4 vey opp JiJe ops ix ing the 1cal ail imre bt ere ich re) measrement wold best represent exposre to diesel exhast. Althogh there has been a trend towards the measrement of carbon particles, we did not want to exclde measrement of gases. The emphasis of this work was the measrement of personal exposre and as the measrement of exposre to carbon reqires a pmp for sample collection we needed to measre the gases diffsively to avoid overbrdening operators with additional pmps. Conseqently, long-term diffsive colorimetric gas detector tbes were sed althogh we recognized that there was the potential for some interference from the gaseos components of the exhast and that there may be detection limit problems. Carbon monoxide was measred with Draeger 5/a-D or Gastec Dosi No. I DL tbes (detection limit abot 6 p.p.m. over 8 h); carbon dioxide was measred with Draeger 5/a-D or Gastec Dosi No. 2 D tbes (detection limit abot 65 p.p.m. over 8 h for Draeger tbe, 25 p.p.m. for Gastec tbe); nitrogen dioxide was measred with Gastec Dosi No. 9 D tbes (detection limit abot.1 p.p.m. over 8 h). The respirable fraction of diesel exhast particlate was collected onto pre-weighed 25 mm qartz filters sing a cyclone sampling at min- 1 (the qartz filters were cleaned in a frnace at 55 C for 4.5 h before se to remove residal carbon). After sampling, he filters were re-weighed (appropriately blank corrected), treated with dilte hydrochloric acid (2 µl of 1.5% w/w) to remove carbonates, then analysed for elemental and organic carbon sing a Strohlein Colomat. The Colomat ses a two-stage analysis. In the first stage, the qartz filter is placed in an i.r. frnace and heated to abot 8 C nder a stream of nitrogen. The organics adsorbed onto the carbon particles are vaporized and catalytically converted to carbon dioxide. The carbon dioxide is adsorbed into barim perchlorate and qantified sing a colometer where the reslts appear as 'conts' each of which is eqivalent to.2 µg carbon. The reslt is converted to the mass of carbon in the vaporized organic sbstances.and is referred to as 'organic carbon' (OC). This represents the organic carbon adsorbed by the carbon particles in the exhast withot discriminating between the many sbstances that may be present. After the first stage of the analysis is complete, there shold remain only carbon particles on the qartz filter. In the second stage, the frnace is again heated to 8 C bt this time nder oxygen. Under these conditions, carbon is converted to carbon dioxide and again qantified in the colometer. This reslt is referred to as 'elemental carbon' (EC) and represents the carbon particles in the diesel exhast. The OC and EC reslts can be added to provide a 'total carbon' (TC) figre. Field blank filters were analysed with each batch of samples and the sample reslts corrected appropriately. Pyrolytic formation of EC from diesel exhast particlate dring the OC measrement stage bas not been detected sing the conditions of analysis reported here. The measrement of aldehydes (acrolein, formaldehyde and acetaldehyde) was attempted at some backgrond locations sing glass fibre filters treated with 2,4-dinitrophenylhydrazine (DNPH) sampling at 2 ml min- 1 However, it became apparent that not only is DNPH nsitable for the measrement of acrolein becase of the formation of mltiple reaction prodcts bt the DNPH itself was affected at some locations by other components of diesel exhast. Conseqently, this measrement was abandoned. Personal samples were collected in the breathing zones of workers potentially exposed to diesel exhast dring their working shift. Static samples were placed in locations where personnel worked and occasionally in those areas where it was impractical to collect personal samples. Wherever possible, sampling times were at least 6 h dration to ensre collection of sfficient carbon for analysis bt some processes inevitably reslted in shorter sampling periods. All reslts reported here are representative of 8 h time weighted exposres. In smmary, each set of personal and backgrond samples provided reslts on: respirable particlate, elemental carbon, organic carbon, total carbon, carbon dioxide, carbon monoxide and nitrogen oxides. DESCRIPTION OF WORKPLACES AND CONTROL STRATEGIES Smmary descriptions of typical premises in each workplace groping are given in the following paragraphs. Amblance depots (grop 1) The amblance engines are normally rn only dring engine testing or when the vehicles are moved on or off ramps so exposre to exhast emissions is intermittent and of short dration. At one depot, the typical throghpt with three mechanics was 5-1 vehicles per day. Flexible dcting was available to connect to the exhast tailpipe bt only occasionally sed. Local exhast ventilation (LEV) was available at for inlets adjacent to each inspection pit. General ventilation was provided by partially opening shttered doors that rn the fll length of the workshop. No respiratory protection was worn. Ro-ro ferries (grop 2) Ro-ro ferries are sed to transport both freight and vehicles. When nloading the ferry, the freight is nlashed from the deck then taken by diesel-powered tgs or container lorries to the dock. When

4 438 J. Groves and J. R. Cain loading the ferry, the vehicles are marshalled into position on the deck where the freight is nloaded and shackled to the deck. Unloading and sbseqent loading can last for p to 3 h. Exhast ventilation nits are sally positioned at the stern end of the main deck bt they are of limited se when the deck is filled to near capacity. General ventilation can be provided by opening the bow and stern doors when appropriate. Drivers are recommended not to start their engines ntil all freight loads are nlashed. Railway repair/stations (grop 3) The main activities at repair and maintenance iles inclde refelling, watering rnning repairs, servicing and examination of diesel-powered railway traction nit. Night shifts tend to be bsiest, as the rolling stock is reqired for service dring the day. Operators may be exposed to diesel exhast when the engine has to be rnning for certain repair and test operations bt the time of exposre depends on the work. ndertaken. The depots vary in size and n nber of employees. At one large railway repair depot with 5 employees, general ventilation was achieved by opening the end doors of the bilding and there were roof monted axial flow extractor fans fitted above the rails. No LEV was sed dring repair of the traction engines and rcspira tory protect' on wa not sed. At a smaller depol, gener;ll ventilation was provided by opening doorways in areas where visible fme wa known lo accmlate. Three timer controlled exhast ventilation fans in the ceiling removed emissions at those times when exposre was at a maximm dring the working day. Bs garage and/ or bs repair (grop 4) There are two distinct grops of employees within a bs garage; the operations staff and tb.e engineering staff. Operations staff sch as drivers and condctors typically only spend short periods of time within the garage whereas engineering staff may spend their fll working day within the garage. The bses are sally stationed overnight in the garage and are driven ot at the start of the day dring a period known as the 'rn-ot'. The majority retrn to the depot/garage ('rn-in') in the early and late evening. The diesel engines may be rnning while the bses are being cleaned, maintained and repaired within the garage and also when the bses are started prior to leaving the depot. There is a tendency for some drivers to start a vehicle and leave it rnning to allow the engine to warm p before 'rn-ot'. Al several garages it is local policy that engines shold not be lefl rnning or, where necessary, shold not be left rnning for more than 15 min. One large garage inclded in the srvey employed 14 mechanics, 2 semi-skilled operators, 4 laborers and 3 cleaners all of whom were potentially exposed to diesel exhast. Exposre occrs in the body repair shop when vehicles are driven into and ot of the work area. General ventilation was by exhast fans, dctwork and by keeping doors open at all times dring the working day. Flexible dcting attached to the tailpipe was available at some sites. Respiratory protective eqipment is not worn nless specifically reqested by operators. Vehicle testing (grop 5) Exposre to diesel exhast occrs when vehicles are rn into the inspection booths. There was little evidence of control measres at the sites visited althogh an electric portable air mover was in se at one station to control exposre to diesel exhast dring vehicle testing. One vehicle manfactrer had attempted to control exposre by fitting exhast ft.lters to vehicle tailpipes dring testing bt the filters were freqently not retrieved after the te t and eventally lost. Fork-lift trcks (grop 6) Fork-lift trcks, many of which are diesel powered, are widely sed to load and nload goods from lorries and to move goods arond warehoses. Where nmeros fork-lift trcks are in se, there can be an accmlation of diesel exhast. At one halage firm where six fork-lift trcks were in operation, a ownber of employees complained of breathing problems allegedly linked to exposre to engine exhast. Dring the bsiest periods of the day shift, small amonts of fme were seen in the roof space of the warehose. General ventilation was provided throgh open roller shtter doors in the warehose. Toll booths, set-down areas, tnnel repair sites (grop 7) Exposre at toll booths arises from passing vehicles and depends on the weather conditions and prevailing winds. Strong winds help to disperse the exhast while it tends to linger and enter the booth in calm conditions. There is no mechanical ventilation in the majority of booths and general ventilation is provided by opening windows. The operators do not generally wear respiratory protection. Unsally, at one location, the toll booth operators work in heated pressrized booths with a spply of air passed in at roof level and ot throgh the vehicle commnication window to the sides of the booth at bench height. Cargo ports at airports are generally temporary bildings where drivers have to present varios docments. Qees of vehicles bild p at these secrity checks an.d engines are often left idling while the driver enters the bilding.

5 Diesel engine exhast emissions in the workplace 439 Exhast ventilation systems in tnnels sally consist of a series of fans at each end of the tbe. Air is drawn in from otside the tnnel at several ventilation shafts on each side of the tnnel and the air is exhasted throgh extraction dcts in the roof at either end of the tnnel and then expelled into the atmosphere throgh exhast tnnels. >< :E Cl en -vv 'tj"q'\c'li-r-oocoo r-ol.1"11"r<'loo-- NNNNNNNN MMC'lirrJc-r!MrtIM t-1vnonoov"l c--ic--ic--ic--ic--ic--ic--in RESULTS A total of 691 respirable dst samples were collected. One personal sample was omitted from analysis becase of pmp failre. A frther 5 personals and 29 backgrond samples were declared invalid becase of filter damage (prodcing negative weight gains) leaving a total of 656 respirable dst samples split between 215 personals and 441 backgrond samples. Close examination of the respirable dst reslts reqired the elimination of a frther 65 samples (26 personal, 39 backgrond) becase the sm of the elemental carbon and organic carbon analyses exceeded the respirable particlate on the filter. This probably reflects the difficlties in handling qartz filters which are prone to shed fibres dring handling. All respirable dst masses in the srvey were corrected by weighing blank filters in the sal way. A total of 189 personal q.,_nd 42 backgrond respirable dst reslts were accepted as valid. Estimates of the limit of detection for weighing qartz filters (based on three times the standard deviation of the blanks) ranged dring the srvey from abot 21 p to 12 µg for a random selection of 1 batches of samples. The median vale of 45 µg was adopted as representative of the overall limit of detection. The median air volme sampled was 714 litres. Combining this with the limit of detection for weighing gives 63 µg m - 3 as the detection limit for airborne concentration of respirable dst. Any vales reported below this concentration were assigned a vale eqal to half the detection limit as is the sal convention. The estimated detection limit for both EC and OC was 1 µg. Using the median volme of air given above, this eqates to an airborne detection limit of 14 µg m- 3 for both EC and OC. Vales reported below this concentration were allocated a vale of half the detection limit. Total carbon is simply the sm of elemental and organic carbon. Of the respirable dst samples collected, 22 personals and 244 backgronds were analysed for EC and OC (these inclded some samples which prodced invalid respirable dst reslts bt which were nevertheless analysed for carbon content). Dring analysis, nine EC and one OC analysis failed. Conseqently, reslts are reported for 217 personal and 237 backgrond EC analyses and 22 personal and 243 backgrond OC analyses. Analytical reslts were divided into personal and, bd 2' >... 8 E ;:l tl ;::i - ::... a c--i :... - ;:l... bd i:q OJ... A. :;:; ci z >< Cl en :a ci z - bd. e P.. i: en.. ;:l... bd.d -. M\\--l.f"IO \OV"lOl.f"IC-ON-.:::t' c<;-- \ \\Noooor-MV NC-r- \\C-N NNOONV\ " OON\\ oonooonv)nv) -\\\C'f V) :! ONNOOO\NN C-r""IMr-\t-('f)M \Ol"-\' \\...: N...:...:...:...: : N OOOOMV\\..- -\NV\Ol.OO --NN-("f1-N OV"lOr- MN \ M \ l.f"io -N-N-M.-N J:-l.f"IOC-\ON..-'tj" NO\l.f"l\Olfl'tj'"O\\O..- N N N,..., 'tj",... N M\Ol.f"l-OV)O\O\..-.V')'tj'".-NM ci Q)......, ;.. 1-o. <J) ;::::s '(tj... o ;>., ci..oj.;::: C en gt:::-= <B ll)g 2 8Z ::: E O - ;:::s en...c: v o.. o o.. <l'.i:!:l>li..r-.r-. NC-M...-o-NCtr).-\-\t-M NN-N c<;c<;c'i,...n<c:f'lf'l\ot-

6 """ """ Table 3. Elemental carbon smmary (µg m-3) Personal Backgrond Job grop Sampling site No. Mean Median GM GSD Min Max No. Mean Median GM GSD Min Max 29 I Amblance depot Ro-ro ferry Il l 3 Railway repair Bs garage/repair Vehicle testing II Fork-lift trcks Toll booths, tnnels, etc Totals Table 4. Organic carbon smmary (µg m -3) '.-.., < ::i - '.-?" (J!::. ::i Personal Backgrond Job grop Sampling site No. Mean Median GM GSD Min Max No. Mean Median GM GSD Min Max I Amblance depot Ro-ro ferry ! Railway repair Bs garage/repair 53! Vehicle testing II Fork-lift trcks Toll booths, tnnels, etc f21 Totals

7 Diesel engine exhast emissions in the workplace 441 backgrond samples and varios statistics calclated for each workgrop as shown in Tables 2-5. Probability.plots for respirable dst, elemental carbon, organic carbon and total carbon sggest that the data distribtions are approximately lognormal. Respirable dst Table 2 shows that the geometric mean (GM) for 189 personal exposres to respirable dst was 28 µg m- 3. Amblance depots recorded the lowest exposres (I 18 µg m- 3 ) bt exposnrel where forklift trcks were in se were nearly twice those of the other grops (369 µg m:--3 ). The fork-lift trck category also recorded the maximm respirable dst vale. Except for the amblance category, the exposres for backgrond samples are all lower than the corresponding personal samples. The geometric standard deviations (GSD) for the personal samples are arond two for each workgrop except for the ro-ro ferries with a vale of 2.7 which perhaps reflects the non-repetitive natre of the work. Two backgrond samples at site 34 (grop 7) prodced respirable masses of abot 8 mg m- 3 which are clearly nsally high. Cl) CJ ci z V)tj"M.-\\CM OCNNOOVOON V'\MOO\ON\ONOC OOC\NNC-\N MNMNNNNM \ON'o::TMM('f')1'" C--OOOC"""l'tj l.llv), C'l - Elemental rs.arbon The GM concentration for 217 personal exposres to elemental carbon was 25 µg m- 3 (Table 3). Fork-lift trcks provided the highest GM exposre (66 µg m- 3 ), ro-ro ferries were ranked second (37 µg m- 3 ) and the lowest exposres were at vehicle testing stations (11 µg m- 3 ). Backgrond exposres to elemental carbon within the work categories were very similar to the personal exposres with an overall backgrond exposre of 24 µg m- 3. Within the grops, fork-lift trcks were an exception since the geometric mean backgrond was abot 3% higher than the corresponding personal exposre. Ro-ro ferries were the only category where personal exposres exceeded backgrond exposres to any degree. The close eqivalence of personal and backgrond exposres sggests that the elemental carbon is relatively homogeneosly dispersed in the measred. atmospheres. This might indicate that those people in the same work area bt not directly involved in the process generating diesel exhast might also be significantly exposed. Organic carbon The pattern of exposre changed for organic.carbon (Table 4). Fork-lift cks agai11 gave the h.ighesl GM personal exposre (99 µg m- 3 ) bt not markedly differenl from railway repair (91 µg m- 3 ) and bs garage/repair (9 µg m- 3 ). This sggests that the latter two job grops are exposed to a higher proportion of organics adsorbed onto the [/l CJ z OJl.s c. Cl) V)MOO\NNMM...: M "i N N- N

8 442 J. Groves and J. R. Cain carbon particles Lhan che fork-lift trcks category. Whereas th_e overall means for per onal and backgrond expo re to elemental carbon were similar lhey are different for organic c<irbon al 7 µg m-3 for personals and 43 µg m-3 for backgronds (the ratio of these two reslts for organic carbon is similar to the ratio of the corresponding reslt for respirable dst). This sggests that those closest to the sorce of organic carbon receive a higher exposre to organic carbon than those more remote from the sorce. Total carbon In terms of total carbon (elemental +organic) shown in Table 5, fork-lift trcks provided the highest GM personal exposres (188 µg m-3) followed by bs/garage repair (124 1g m-3) and railway repair (I. 13 1g m-3). The overall mean of personals (12 µg m-3) was greater than for the backgronds (65 1g m-3). It is interesting to note that total carbon acconts for 49% of the overall mean personal respirable concentration in mass terms for the personal exposres. The corresponding percentage for backgrond samples is very similar at 46%. In both cases, the remaining 5% or so of the respirable mass will be made p of the elements brned off from adsorbed organics carbon dri}1g the measrement of organic carbon together with. combstible and non-combstible general workplace dst. Personals vs backgronds Generally, personal exposres will be expected to be higher than backgrond exposres since the operator will ally be closer to the sorce of the polltant. This is the case with respirable dst in this rvey (28 µg m-3 personal., 14 µg m-3 backgrond), organic carbon (7 1g m -3 personal, 43 µg m -3 backgrond) and total carbon (12 µg m3 personal 65 1g m-3 backgrond). The ratio of personal to backgrond concentration is 1.5 for respirable dst, 1.6 for organic carbon and 1.6 for total carbon. However, for reasons not yet nderstood, elemental carbon deviates from this trend and the personal and backgrond overall geometric means are essentially the same (25 µg m- 3 personal, 24 µg m- 3 backgrond). When the for sets of reslts (Tables 2-5) are ranked on the basis of highest personal exposre, fork-lift trcks rank first with bs/garage repair and railway repair consistently occpying second or third place. The ranking for amblance depots on this basis is dbios as it is based on only three personal reslts. Percentiles Personal expo re percentile aje given io Table 6. Whereas the 95th percentile for most job grops is Jess than abot 11 µg m-3 EC, the fork-lift trck grop has a corresponding percenlile of 39 µg m- 3 indicating significantly high.er exposres. Relationship between elemental and organic carbon Figre l how the relationship between elemental carbon and organic carbon for personal samples. There is clearly no straightforward relationship between these two measrements. Other work in this laboratory (HSE, 1999c) has shown that the relalive proportion of elemental and organic carbon in diesel exhast varies depending on the load on the engine. For example, when idling at low revs, one particlar engine prodced elemental carbon which was abot 3% of the total carbon emitted (i.e. organic carbon predominated). At higher revs and higher engine loads, the proportion of elemental cartion in the total carbon increased to as mch as 8% of the total carbon. This information can help to interpret tj1e reslts in Fig. I on the basis of the work.loads on the engine. The percentage of EC was calclated a a proportion of TC for the GM personal concentrations at each site where the concentrations of EC were greater than the detec- Table 6. Percentiles for personal samples Job grop concentrations (µg m- 3 ) Analysis Percentile (%) All data Respirable dst Elemental carbon Ill Organic carbon Total carbon

9 Diesel engine exhast emissions in the workplace 443 5L l' I 4 M' 3ooj ' Cii 2 E Q) E Q) I I,, II j i I 1 w 1 L; " Li :_1 " "---'... " J - I I I Organic carbon (g/m3) Fig. l. Elemental carbon vs organic carbon. tion limit. The reslts were ranked in ascending order of percentage EC. Sites 3-5, and 33 had percentages of EC above 4%.(range 4-69%). These samples shold originate mainly from engines that are working nder some load and generating higher proportions of EC. Sites 3-5 are ro-ro ferries where there wold be expected to be heavy loads on the engines dring start-p and manoevring of vehicles on deck. Sites are in the fork-lift trck category where there is likely to be mch activity reqiring freqent acceleration of the engine nder load. Site 33 is a toll booth where vehicles will have to pll away from a standing start, again ptting a heavy load on the engine. At the other end of the scale, sites 6-9, 11, 14, 32, 36 and 39 have percentages of EC p to abot 2% (range 5-21 %). These samples wold be expected to derive from lightly loaded engines. Sites 6-9 and 14 are railway repair and maintenance depots where it is likely that the engines are left idling for some time while adjstments are made and tests carried ot. Site 11 is a railway station where engines may be left rnning while trains are standing at platforms. Site 32 is in the fork-lift trck category and stands apart from the three other fork-lift trck sites. However, this may not be srprising since the work activity inclded spplying an assembly line as well as warehose dties. Site 36 is an airport coach set-down area and site 39 is in an airport baggage handling area. Between the two arbitrary percentages of 2% EC and 4% EC, there is a range of activities which wold be expected to be a mixtre of heavy and lightly loaded engine dties. The relevance of these findings depends on the risk to health associated with EC and OC since some operations will generate proportionately more of one than the other. Those operations that place heavy loads on the engine might be expected to present more problems if EC is regarded as a health risk. Lighter engine loadings, particlarly with freqent idling, will be mainly associated with OC as the dominant featre. Becase the fork-lift trck category prodced higher exposres than the other job categories, the for sites which make p this grop were examined in more detail. Table 7 smmarizes the geometric mean personal expo res at each site for the for measrements shown. All for measrements rank site 29 as prodcing significantly higher exposres than the other sites. The GM respirable dst at these for sites was significantly greater than the GM for all reslts (Table 2). The reslt for elemental carbon for sites was also higher than for Table 7. Personal reslts at fork-lift trck premises ( µ g m -3 ) Respirable EC oc TC Site No. No. GM No. GM No. GM No. GM LO Total

10 444 J. Groves and J. R. Cain the mean of all reslts (Table 3) bt site 32 was lower. A similar pattern was repeated for organic and total carbon where site 32 prodced reslts similar to the overall means. Ths, amongst the fork-lift trck category, sites confirm the significantly higher exposres than in the other work categories bt site 32 prodced reslts which were more in line with the overall mean. It is interesting to note that the ratios of elemental carbon to organic carbon change significantly between the sites in Table 7. At site 29, elemental carbon predominates which cold indicate that the engines are being worked nder some load. At sites 3 and 31, organic carbon is present at a slightly higher proportion than elemental carbon, bt at site 32, organic carbon predominates markedly perhaps indicating engine sage nder light load. Relationship between respirable dst and elemental/ organic carbon Figres 2 and 3 show the relationship between respirable dst and elemental carbon and organic carbon, respectively, for personal samples. There appears to be no direct relationship between respirable dst and elemental carbon (Fig. 2) so respirable dst cold not be sed as a srrogate for elemental carbon. Some strctre may be present in Fg. 3, probably becase organic carbon acconts for the blk of the DEEB mass in most samples bt even in this graph there is considerable scatter. Gas detector tbes Measrements with the gas detector tbes pro- dced reslts of qestionable vale becase the exposres were low and, for most samples, less than the detection limit of the tbes. To calclate the varios statistics, gas detector tbe reslts recorded as less than the limit of detection were assigned vales eqal to half the limit of detection as follows: carbon monoxide 3 p.p.m.; carbon dioxide 125 p.p.m.; nitrogen dioxide.1 p.p.m. Of the 22 detector tbes sed to measre personal exposre to carbon monoxide, 113 prodced no response; 118 of the 216 backgrond tbes deployed also prodced no response. The reslt was an overall personal GM exposre of < 6 p.p.m. Only the amblance depot category prodced measrable exposres (38 p.p.m.) bt this was based on only two reslts above the detection limit so they may not be representative. Amongst the 22 personal measrements for carbon. dioxide, 49 prodced a response less than the detection limit. A total of 116 backgrond reslts (ot of 216) were less than the limit of detection. The highest GM personal exposre was in the forklift trck category (12 p.p.m.) and the two reslts from the amblance depot category prodced an exposre of similar magnitde lending some weight to the carbon monoxide reslts for this job grop. Contrary to findings sing the other measrements, vehicle testing ranks third when sing carbon dioxide as the index of exposre (9 p.p.m.). The overall backgrond exposre (27 p.p.m.) is only jst above the detection limit bt is less than the corresponding personal exposre vale (65 p.p.m.), 5 I T " 3 6 " J 9 Respirable dst (g/m3) i.i. '------, j Fig. 2. Elemental carbon vs respirable dst.

11 Diesel engine exhast emissions in the workplace 445 he!ss de en ;be!cnit ::1 - ond be of ry Ut be 1e. tr he Jts an. k lts n ht p. ts, X Sl e- ), possibly sggesting a job-related exposre. Atmospheric carbon dioxide will of corse contrib te to these reslts. Of the 22 gas detector tbes sed to measre personal exposre to nitrogen dioxide, 176 recorded vales less than the Limit of detection; 157 of the 216 backgrond tbes prodced responses less than the limit of detection. However, the measrements did again confirm the fork-lift trck category as providing the highest GM exposre (.2 p.p.m.) spporting to some extent the conclsions drawn from other measrements. The data sggest that time weighted average gas detector lttbes are nlikely to have sfficient sensitivity to measre exposres to carbon monoxide carbon dioxide or nitrogen dioxide reslting from diesel exhast except where exposres are relatively high. DISCUSSION AND CONCLUSIONS Forty sites were visited dring the smvey. The reslts of personal exposre and backgrond sampling concentrations foj gaseos components, respirable dst, elemental carbon, organic carbon and total carbon varied between sites and between workgrops. The data analysis shows a wide spread in exposre patterns reflecting the different work practices: job categories of employees and the control methods sed. Tile 95th percentiles for personal exposre to EC, OC TC and respirable dst were: EC 17 Lg m- 3 ; OC 262 µg m -3 ; TC 416 µg m -3 ; respirable dst 73 µg m -3. In a complex mixtre sch as diesel exhast, it is difficlt to identify one component which wold be representative of exposre as a whole. However, this srvey has demonstrated that elemental carbon cold be sed as an indicator of exposre. With some exceptions, sch as the manfactre and se of carbon black and in coal mines, the diesel engine is likely to be the only significant sorce of elemental carbon in the workplace (elemental carbon is prodced only in trace amonts by petrol engines). Placing the varios sites into seven workplace grops was a pragmatic approach to categorizing the data and the Jesits obtained for each grop shold not be an taken as an indication that all premises fall into one of these grops. However, exposres in premises sing fork-lift trcks were c onsistently higher than at other workplaces. Poor working practices (e.g. failre to switch off the engine when not reqired for a sbstantial period of time or not driving the vehicle correctly), poorly maintained engines and inefficient general venti lation contribted to the high exposres at these premises. None of the measrements for carbon monoxide, carbon dioxide and nitrogen dioxide exceeded the relevant crrent UK 8 h time weighted average occpational exposre limits (5 p.p.m. for carbon monoxide, 3 p.p.m. for nitrogen dioxide, 5 p.p.m. for carbon dioxide). Where exposre to carbon dioxide was over 1 p.p.m. 8 h TWA, the control systems were generally falty, poorly maintained or inadeqately designed. While the measre- 8 --,- ----,-..,---,-.,..,.. -r ---r---- I M' 6 - E Ci (J 2.. c -e 4 (1) (J (J c: (1) 2 - tl'... D a 6 D 9 J. -,_j _ J Respirable dst (g/m3) Fig. 3. Organic carbon vs respirable dst.

12 446 J. Groves and J. R. Cain ment of carbon dioxide may not represent exposre to diesel exhast as a whole, it may be an indicator of the overall adeqacy of control measres, in particlar LEV or roof extraction systems. It is ncertain at present what wold be a sefl indicator of irritancy. Aldehydes (acrolein, formaldehyde and acetaldehyde) are a possibility bt frther research is reqired in this area. Measrement of exposre to the aldehydes and other gases sch as nitrogen dioxide derived from diesel exhast may be important in bilding p a database in a field where little information is available. However, the employer shold look for better means of control where the workforce complain of irritancy. Exposre to DEEEs shold be prevented or where this is not reasonably practicable, adeqately controlled, following a sitable risk assessment as reqired nder COSHH reglation 6. In those sitations where it is not reasonably practicable to prevent exposre to DEEEs, for example by changing the method of work, modifying the layot of the workplace or modifying the work operations, then a nmber of measres can be taken to control exposre. These inclde: the se of lower emission or more fel efficient engines; the se of cleaner fels sch as low slphr diesel fels; the se of catalytic ' converters and catalysed and non-catalysed particlate traps; the se of tailpipe exhast extraction sys-, terns and the provision of sfficient general ventilation. As exposre to DEEEs is best controlled at sorce, respiratory protective eqipment (RPE) shold be sed only as a last resort. This srvey revealed that the main means of controlling exposre to DEEEs were engineering controls (in particlar engine exhast filters and extraction fans) or work practices (e.g. reglar engine maintenance, trning the engine off when not in se, job rotation, etc.) or a combination of these and that they varied between premises. Overall, poor working practices, poorly maintained engines and inefficient, inadeqately maintained mechanical ventilation contribted to higher occpational exposre to DEEEs measred dring this srvey. In warehoses where fork-lift trcks are sed and where the highest personal exposres and backgrond concentrations were measred, control is mainly by general ventilation. Improved general ventilation as well as reglar servicing of the trcks wold redce exposre. Frthermore, where goods are being nloaded from delivery lorries, the delivery vehicles shold not enter the warehose. If this is not possible then engines shold be switched off dring loading and nloading. At bs railway locomotive depots, control of exposre is mainly by a combination of general ventilation sch as extraction fans in the roof and LEV. Many sch sites rely on general ventilation only. Ideally the contaminated air shold be exhasted to the otside with make-p air flowing into the workplace throgh open doors or spply air openings. The major drawback to sch control is the cost of heating the make-p air dring cold weather and the failre to remove diesel emissions at sorce. In the winter months, these premises conserve heat by keeping the doors sht as mch as possible so casing an accmlation of DEEEs. At several sites, engine exhast filters were attached to the tailpipe of the vehicle to remove particlate from the diesel fme at sorce. One vehicle manfactrer cold not achieve control sing this method becase the drivers of the vehicles are responsible for fitting the filters to the exhast and removing them. The filters are often misplaced as the drivers do not always retrieve them once they have been removed. There is also a high level of system damage to the filter nits and the filter elements reqire high levels of maintenance. A tailpipe exhast hose, connected to a fan which discharges the diesel exhast otside the bilding, was sed as a means of control. together with engine exhast filters at vehicle repair premises and vehicle testing sites. The tailpipe exhast hoses remove the gaseos emissions as well as the respirable particlate at sorce. Personal exposre and backgrond concentrations to the respirable particlate tended to be lower at these sites compared with other sites in the srvey. These engineering controls are, however, not common in the average motor vehicle repair and vehicle testing sites. RPE was sed to control exposre to DEEEs at only two sites and was only provided on reqest by employees. Following an adeqate risk assessment of workplace exposre to DEEEs to assess whether to prevent or control sch exposre, consideration shold be given, where appropriate, to the se of: alternative technologies where practicable sch as battery powered vehicles; alternative fels sch as 'green diesel' or compressed natral gas; low emission or more fel efficient engines; catalyst technology or particlate filters. Additional measres might inclde instigating a programme of reglar engine maintenance, reviewing the engineering controls sed, introdcing worker job rotation and providing information, instrction and training of staff on the hazards associated with exposre to DEEEs where necessary. The chosen control measres will be based on the levels of risk and exposre, the type of workplace, present work practices and cost and benefit factors. Additional information on control of diesel exhast emissions is given in an HSE pblication (HSE, 1999d). Acknowledgements-The athors wold like to thank the Health and Safety Exective for fnding the srvey and also the many HSE staff who participated in the collection and analysis of the samples.

13 he nd on Diesel engine exhast emissions in the workplace 447.vork- 1ings. )St Of and :e. In at by ca swere nove e ve :ising are and d as they I of r eloch ing, gine ide the icnd at by REFERENCES Birch, M. E, { 1998) Analysis of crbonnceos aerosols: interlaborntory comparison. tmilyst 123, Dnbmann D., Fricke, Fl.-H. and Baer, H.-. (1996) Diesel engine emission$ in workplace-ntmospheres in Germany. Occ11pwio11(ll Hygiene 3, Gillemin, M., Cachier, H., Cb.ini, C., Dabill,., Dahnm, D., Diebold, F., Fischer, A., Fricke, H.-H., Groves, J. A., Hebisch, R., Hopillart, M., Israel, G., Mattenklotl, M., Moldenhaer, W., Sandino, J. P., Schlms, C., tter, E. and Tcek, E. (1997) International rond robin tests on the measrement of carbon in diesel exhast particles. Imer11atio11al ArchiJPcs of Occpmio11al (md E11viro11111c11ra/ f/ealrh Health and Safery Exective (1994) Chemicals {Hazard Jnfo1mation and Pncknging for Spply) Reglations 1994 {SI 1994/3247) as amended by Lile Chemicals (Hazsd Information and Packaging for Spply) (amendmenl) Reglations 1996; and 1997; and 1998; and Health and Safety Exective (1999a) Geiteral COSHH ACOP (Control of Sbstances Hazardos to Health) and Carcinogens ACOP (Control of Carcinogenic Sbstances) a.nd Biological Agents ACOP (Control of Biological Agents). ln Comrol of Sbsrm:es Hazardos w Hllalrh Reglations HSE Books, Sdbry, UK, (ISBN ). Health and Safety Exective (1999b) EH4/99 Occpational Exposre Limits 1999, {ISBN ). Health and Safety Exective (1 999c) Engine exhast emissions from diesel fels. Internal report OMS/99/12. Health and Safety Exective (1999d) Control of Diesel Engine Exhast Emissions in the Workplace. In HSG 187. HSE Books, Sdbry, UK, (ISBN ). IARC (lnternati.onal Agency for Reserch on Cane.e r) ( 1 989) Diesel tmd Gasoline Eng ine Exha11sts and some Nitroarenes. IARC, Lyon, France (ISBN ). NIOSH (National Institte of Occpational Safety nd Health) (1998) Elcmenil Carbon (Diesel Particlate), la NJOSH Manal of Analytical 1\1/etliodJ'. 4th edn. Method 54. NIOSH, Cincinnati, OH. tded ites owicle rkre- ]d naery een or or ht me ols idon s ill he st E