Measurements of oil mists from mineral oil-based and water-mix metalworking fluids
Measurements of oil mists from mineral oil-based and water-mix metalworking fluids
Keywords:: Health and Safety, Metalworking fluids, Oil mists
The subject of oil mist concentrations and the establishment of safe working levels has been the subject of lengthy debate in the USA. The American National Institute of Occupational Safety and Health (NIOSH), in a draft Criteria Document in early 1996,had sought a reduction on the allowable oil mist concentration level from 5mg/m3to 0.5mg/m3. Their investigations had appeared to show that there were significant carcinogenic risks arising from the exposure of workers to metalworking fluids. The findings were vigorously challenged by the US Independent Lubricant Manufacturers Association (ILMA), who argued that the investigation and subsequent conclusions were fundamentally flawed. They pointed out the following inadequacies in the NIOSH study:
(1)NIOSH incorrectly used highly-specific and technical terms. The draft CD referred to the use of metalworking fluids, whereas ILMA argued that the NIOSH study involved only metal removal fluids (as opposed to metalforming fluids). Therefore the studies could not justify the inclusion of all metalworking fluids in the CD. The NIOSH evidence also appeared to implicate all types of fluid, whether neat oil,soluble oil, semi-synthetic or synthetic, in inducing respiratory symptoms.
(2)ILMA rightly contended that the composition of metalworking fluids is not static. Changes in component concentration and depletion, chemical contamination, microbial contamination, metallic contamination and housekeeping contamination occurred during use. Efforts to quantify occupational exposures must also take account and differentiate between those effects arising from the fluid and those arising from the contamination.
(3)NIOSH had not adhered to the policy which required that all evidence be subjected to a proper peer-review.
(4)The composition of metal-removal fluids had changed substantially in the past 15-20 years, and were now very much safer as a consequence. The draft NIOSH CD did not differentiate between conclusions which might have applied to products of the 1950s from those of the 1990s.
In addition to the above,NIOSH were also reluctant to associate non-fluid exposures, e.g. to alcohol or tobacco, with observed incidences of cancer. At least one study had already concluded that "current smoking was consistently the strongest and most significant predictor of respiratory symptoms".
However, as a result of the ILMA comments, there has now been a major shift on the part of NIOSH in that the central thrust of their revised CD concentrates on the non-malignant effects of exposure to such fluids, including lipid pneumonia, asthma, acute airway irritation, chronic bronchitis, and impaired pulmonary function. Peer review of the evidence had since shown that their conclusion of a possible increased risk of cancer involved workers exposed to fluids before the mid-1970s.
The latest pre-publication copy of the CD maintains a reduction of recommended exposure limit (REL) to 0.4mg/m3 for thoracic particulate mass (or 0.5mg/m3 for total particulate mass), as a time-weighted average concentration for up to a 10-hour day during a 40-hour week.
In the UK, it is considered that excessive exposure of workers to oil mists can lead to eye, nose and throat health problems, and dermatitis. The current Occupational Exposure Standards are 5mg/m3 (eight-hour time-weighted average), and 10mg/m3(15 minute tim-weighted average) (HSE Guidance Note EH 40). Like all OESs, the current level for oil mists is under review. These limits were set for neat mineral oil, and do take into account other components of metalworking fluids which may well merit lower exposure limits. There are currently no Occupational Exposure Standards for water-mix metalworking fluids. However, some engineering companies have adopted an in-house limit of 1mg/m3 for water-mix fluids as an additional protective measure for their workers.
Conclusions drawn from all past evidence are also subject to some debate since the measurement of oil mist concentration level is not straightforward, and at least some of the historical evidence may be suspect in the absence of universally-accepted and validated test procedures. Some of the information gathered in the past may be subject to influence by the particular methods of sampling and analytical quantification used. Referring to the literature, sampling methods have included the use of glass wool-filled tubes, impingers, electrostatic precipitators, denuders, and other filter media such as mixed cellulose esters, PVC, paper, and PTFE. Analytical methods have included chromatographic methods, gravimetric methods,spectroscopic methods using infra-red, ultra-violet and fluorimetry. In Europe,only those methods which have been validated to ensure compliance with the General Requirements for the Performance of Procedures for the Measurement of Chemical Agents in Workplace Atmospheres described by the Comité Européen de Normalization (CEN) should be used.
The HSE have been expending much effort on the development of suitable methods of test for both of these types of fluid. The determination of oil mists from mineral oil based fluids has now been issued as MDHS 84, dated June 1997. Although a method has been developed for the determination of oil mists from water-mix fluids, it has not yet been issued.
Mineral oil fluids
During machining processes, airborne aerosols of the fluid are generated owing to the high shear forces present, and also owing to the high transient temperatures involved. Also, in certain metal forming processes, the lubricant is deliberately applied in the form of a mist generated from a nebuliser with compressed air. The potential for exposure of the operator to oil mist is significant, and employers are required to adopt a hierarchy of measures to ensure that the actual exposure is limited to as low a level as possible, and in any case within the stipulated Occupational Exposure Limits. Although many employers have carried out monitoring exercises, investigations have been hampered by the lack of standardised and validated methods of sampling and analysis.
The main problem in measuring oil mist concentrations is that there will always be oil vapour associated with the droplets, and that the partition of oil between the aerosol droplet and vapour phase in a newly-formed aerosol will be subject to constant changes. The problems are more pronounced with fluids of lower viscosity, since these fluids normally have higher volatilities. In practice, sampling methods are less effective for fluids with viscosities below ca. 18 centistokes at 40°C(Simpson, 1997) owing to evaporative losses during the collection stage. Mist concentrations of fluids of lower viscosity may underestimate the true value owing to loss of volatile components from the sample by evaporation during sampling. Fluids with viscosities less than 18 centistokes account for some 28 per cent of the total volume of fluids currently marketed in the UK, and such methods are therefore limited in their general applicability to a certain extent. It is considered that in order to obtain meaningful estimates of the total airborne oil concentrations in situations involving the more volatile oils, it would be necessary to incorporate a back-up adsorbent tube in addition to the mist collection filter, although problems may occur with incompatibility of sampling rates.
However, although the overall concentration of oil present in the vapour form is likely to be far higher than that arising from the droplet concentration, it is the oil droplets that are considered to pose the largest health risk by most occupational hygienists, since they will tend to deposit in the lungs, whereas the vapour will be exhaled.
The particle size of the aerosol droplets is a significant factor, and tends to comprise of two ranges,depending on the method of generation. Droplets formed by condensation produce uniform and fine mists with a narrow range of particle diameters, typically 1-2 microns. Those produced by mechanical generation or nebulisation are coarser and have a wider spread of diameters, a number of surveys reporting particle sizes ranging from 2-10 microns. Droplets produced by mechanical means are more likely to shrink from loss of volatile constituents than those formed by condensation.
The test procedure which has been developed by the HSE measures aerosol mist concentrations only, and not the accompanying hydrocarbon vapour.
In principle, the test method is basically a gravimetric determination, where a measured volume of air is drawn through a pre-weighed filter contained in a lapel-mounted personal sampler. The filter is re-weighed to obtain a total inhalable particulate concentration. If a level of greater than 2.5mg/m3 is indicated by direct weighing, then the oil mist concentration is estimated by measuring the weight loss of the filter after extraction with cyclohexane. Binder-free glass fibre or mixed cellulose ester membrane filters of 0.8 micron mean pore diameter are used.
A detection limit of 0.1mg/m3 is claimed for the method, for samples drawn over an eight-hour period at a sampling rate of 2 litres/min. Preliminary investigations have indicated that there is no significant bias to the method, recovery rates of 94 to 98 per cent have been achieved using spiked fluids. The method is subject to interferences arising from spray droplets, as opposed to aerosols, in which case some form of shielding is necessary to prevent direct ingress of spray droplets. Also, in some situations, the method may be liable to interference from other aerosols which may contribute to the analytical result,e.g. aerosols formed from water-mix metalworking fluids.
Details of this method have now been published a MDHS 84, obtainable from: HSE books, PO Box 1999, Sudbury, Suffolk CO10 6FS; Tel: 01787 881165; Fax: 01787 313 995.
Water-mix fluids
This method is based on the use of an elemental marker and using flame atomic absorption spectrometry(AAS) or inductively-coupled plasma atomic-emission spectrometry (ICP) as a detection method. The method is only applicable to those situations where there is a suitable element of high enough concentration in the watermix fluid concentrate or in the water used to prepare the working solution, and which is unlikely to arise from any other source in the workplace. The working range is dependent on the quantitative detection limit of the analytical technique used to measure the marker element, the concentration of the marker element in the machine fluid sump, the sump fluid strength, and the volume of air sampled. Again, quantitative analytical detection limits of 0.1mg/m3 should be achievable. Experiments have shown that the method does not show significant bias, analytical recoveries for filters spiked with watermix metalworking fluids of 100 per cent have been achieved. Although sodium, potassium or boron are suggested as suitable marker elements, careful consideration should be given before using sodium because of the ubiquitous nature of this element in the environment. Also, measurement of sodium and potassium by AAS is subject to interferences, as is the measurement of boron by ICP.
This method has been successfully used in field trials conducted by the HSE, but has not yet been published.
David MargaroniEditor
Field trials
As part of HSE's promotion of the management and control of MWFs in the workplace, work has been carried out on developing a new air sampling method for water- mix MWFs and to examine and validate current methods for mineral oils. Gravimetric and spectrophotometric methods have been evaluated, by the Health and Safety Laboratory (HSL), for mineral oil mists and the gravimetric method (involving cyclohexane extraction) was recommended (now published MDHS 84). HSL have also developed a method for measuring exposure to mists arising from watermix MWFs. This involves the use of a marker element by which watermix MWF in air can be expressed as concentration of MWF concentrate and/or concentration of the initial aqueous MWF aerosol cloud.
In 1996/1997, HSE carried out a survey at 28 small to large scale engineering companies to collect personal air sampling data and evaluate the sampling and analysis methods. Comprehensive information on the fluids used, the processes and control methods was also collected to determine the reasonable practicability of controlling exposure in context of this new air sampling data. Air sampling data was also collected using traditional gravimetric methods for comparison with published historical exposure data. Samples were collected from the machine sumps to measure for bacteriological content, endotoxins, and fines levels. This analysis, along with information collected on the management of MWFs, should allow HSE to establish typical "fines" levels for different processes or endotoxin levels related to the history of fluid usage/management. Analysis was also carried out on selected sumps for nitrosamines, PAHs and toxic metals to establish whether significant concentration could be detected.
Personal exposures (40 results) for mineral oil mist were up to 13.2mg/m3, eight hours. TWA,with 90 per cent of results less than 2.8mg/m3, eight hours TWA. The results (275 samples) for personal exposure to watermix mist were up to 2.7mg/m3,eight hours TWA, with 90 per cent of results less than 0.8mg/m3,eight hours TWA. There is still a lot more interpretation to be carried out to look at the influence of the toxicology of MWFs and published epidemiology. High bacterial contamination and endotoxin levels were found in sumps, up to 1.9 x 108 CFU/ml and up to 1 x 106EU/ml respectively, with most sumps having high levels. Many of the companies visited were found to have poor control of fluid strength, poor sump replenishment methods, and poor control of swarf, fines and tramp oil.
These are preliminary findings, the results of all of this work will be published later in the year. This information will be used to develop risk management strategies. HSE and HSL are still processing all the collected data and information, and HSE are planning to take a package of recommendations to the Working Group on the Assessment of Toxic Chemicals (WATCH) in September 1999. HSE will, in due course, consult with industry over the development of any proposals for risk management studies.
Further information on this work can be obtained by contacting: Martin Stear, HSE (Chemical Agents Occupational Hygiene Unit), Magdalen House, Stanley Precinct, Bootle, Merseyside L20 3QZ. Tel: 0151 951 3620; E-mail: martin:stear@hse.gov.uk
Martin StearHSE Chemical Agents Occupational Hygiene Unit
