Full text of DOI:10.1080/10473220390237359

Applied Occupational and Environmental Hygiene, 18: 838–841, 2003
c
Copyright ꢀ Applied Industrial Hygiene
ISSN: 1047-322X print / 1521-0898 online
DOI: 10.1080/10473220390237359
Effects of Fluid Composition on Mist Composition
Eugene M. White and William E. Lucke (retired)
Milacron Marketing Company, Cincinnati, OH
proportions of each will be dependent on the particular indus-
trial processes.(9) The formation of both vapors and particulates
can cause a reduction in their removal efficiency and this factor
may exacerbate worker exposures to aerosols.(10)
In a reported study, mists of selected synthetic metal-
working fluids were generated in laboratory experiments by
two processes, nebulization (atomization) and air sparging
(bubbling).(1) Short-chain fatty acid species were determined
by in situ trimethylsilyl derivatization. Comparison of rela-
tive amounts of the short-chain acids collected from mists
generated by nebulization with those generated by sparg-
ing showed that the sparged mists had significantly higher
amounts of neodecanoic, nonanoic, and dodecanedioic acids.
Comparison of the amounts of acids collected by the resin
cartridges to amounts found on the filters showed that sig-
nificant losses of octanoic and isononanoic acids occurred
over 8 hours of collection and that only dodecanedioic acid
was not lost from the filter over a 22-hour sampling pe-
riod. In another reported metalworking mist study, contam-
inants of metalworking fluids, e.g., tramp oils, were shown
in laboratory experiments to increase the misting potential
of water-based metalworking fluids.(2) Significantly, tramp
oil contamination caused less misting in synthetic fluids than
soluble and semi-synthetic fluids.
In addition to the effects that blended chemicals have on the
misting properties of metalworking fluids, contaminants from
a variety of non-metalworking fluid sources contiguous to ma-
chiningoperationsoftenmixwithfluidsandincreasetheconcen-
tration of mists in machining areas. Many of these contaminants
are commonly referred to as tramp oils. Experimentation in lab-
oratory mist chambers has demonstrated that synthetic fluids are
less prone to the misting influence of tramp oils as compared to
semi-synthetic and, especially, soluble fluids.
This article is a report on selected studies that illustrate ex-
perimental methods that have been used to investigate factors
that influence the generation of metalworking fluid mists.(1,2)
Portions of the results from these studies were used in a presen-
tation on basic misting properties at the 2002 ACGIHꢀR Oil Mist
Symposium in Cincinnati, Ohio.
METHODS AND RESULTS
Laboratory Experiments (Phase I)
Mist Generation and Sample Collection
Keywords Metalworking Fluid Mists, Aerosols, Mist Generation,
Fatty Acids, Tramp Oil
Three synthetic metalworking fluids were used to generate
mists by the processes of nebulization (atomization) and sparg-
ing (bubbling). Five fatty acids—isononanoic, octanoic, neo-
decanoic, nonanoic, and dodecanedioic—were sampled at a flow
rate of 0.9 to 3.5 liters/minute and measured in the generated
mists to determine component distributions. Pallflex T60A20
coated glass fiber absolute filters were used to collect particles
and XAD-2 resin cartridges were utilized to collect any acids
stripped from the filter during sampling. A comparison of the
distribution of acids is given in Table I.
There are significant increases in levels of neodecanoic, non-
anoic and dodecanedioic acids in mists generated by bubbling
compared to nebulized mists. Extended sampling times were
needed to reach detection limits for chemical analysis. Use of
a second-stage trap allowed collection of sample material that
might have been lost to evaporation during sampling. Potential
sample losses are given in Table II.
Experimental data from numerous animal studies have shown
that various chemical and biological constituents of water-based
metalworking fluids have the potential to cause pulmonary
irritation.(3–5) Similarly, as a result of occupational exposures,
these constituents may aggravate existing respiratory conditions
in humans or impair normal pulmonary functions.(6–8) Many of
these constituents are environmental contaminants that enter flu-
ids through normal usage during machining activities. Various
chemicals are selected and blended into metalworking fluid for-
mulations to achieve properties that will facilitate specific ma-
chining and end-product requirements. As metalworking fluids
are sprayed onto and flood tool/workpiece interfaces, misting
of fluids occurs, in part, due to the dynamic forces involved in
machining operations. A mixture of vapor and particulates can
be found in the atmosphere of a given machining operation and
838

EFFECTS OF FLUID COMPOSITION ON MIST COMPOSITION
839
levels of mist.(14) The grinding wheel speed was adjustable from
500–25,000 rpm.
TABLE I
Normalized levels of short-chain fatty acids in nebulized and
bubbled mists
Fluid Selection
Acid
Nebulized mist (%)
Bubbled mist (%)
Widely available metalworking fluids were utilized in labora-
tory experiments: 5 synthetics, 4 semi-synthetics, and 5 soluble
oils.
Isononanoic
Octanoic
Neodecanoic
Nonanoic
40
10
9
2
4
50
14
19
20
12
Tramp Oil Preparation
Dodecanedioic
An experimental tramp oil was prepared by blending ma-
chine tool slideway and anti-wear hydraulic oils. Slideway oils
are maintenance fluids that are used to reduce friction and reduce
wear between tangential, sliding parts in machinery. These oils
have enhanced adhesion properties, i.e., tackiness, and are resis-
tant to wash-off. The 50/50 mixture consisted of the following:
(1) P-47 (G-68) Standard heavy-medium way oil: viscosity—
317 to 389 SUS at 100◦F (ASTM D 2161); (2) P-70 (HM-46)
Hydraulic oil anti-wear lubricant: viscosity—214 to 262 SUS at
100◦F (ASTM D 2161). This mixture was representative of oil
contaminants that are found in in-use metalworking fluids.
Aerosol Size Determination
Mercer cascade (7-stage) impactors were used to determine
the particle size distribution of aerosols. Data from these mea-
surements showed that aerodynamic diameter particle sizes
ranged from 0.33 to 4.6 micrometers.
Sample Extraction, Derivatization, and Gas Chromatography
(GC) Analysis
Methanol and ethyl acetate solutions were used to extract
samples that were, subsequently, derivatized in an on-line GC
injection port by coinjecting extracted fatty acids and a common
derivatizing reagent, i.e., N,O-bis(trimethylsilyl)trifluoroaceta-
mide (BSTFA), a method commonly used in the brewing indus-
try to analyze for low molecular volatile components.(11,12) The
derivatized fatty acids, trimethylsilyl esters and triethanolamine
(TEA) were separated on 2 capillary columns with DB-1701 and
Stabilwax stationary phases.
Measurements of Tramp Oil–Induced Mists
A real-time, light-scattering aerosol monitor was assembled
for the study.(13) Sampling periods were 60 minutes. Measure-
ments of mists were 1.3 mg/m3 to 3.0 mg/m3. The addition of
a 0.6 percent tramp oil mixture (described above) to a fresh
5 percent metalworking fluid mix with 125 ppm water hard-
ness caused misting comparable to that of used fluid. Overall
results show that synthetic fluids generated less mists than semi-
synthetics, and soluble oils generated more mists than any class
of fluids. Notably, increases in water hardness or fluid concen-
tration had no significant effect on generated mist concentrations
up to 500 ppm. Differences in misting between fluids were not
Tramp Oil Experiments
Mist Generator
A laboratory bench–scale generator was designed and built thought to be due to compositional factors but, instead, to the
for the purpose of mist generation experiments (U.S. patent degree of contamination of a given fluid type.
granted).(13) The generator was a 3-ft-tall, clear plastic box
that contained a 2-L fluid reservoir, variable speed electric mo-
tor with a shaft-mounted grinding wheel, a fluid pump with
spray nozzle, and an airflow baffle system. A grinding func-
Two machining sites within the same building at an indus-
tion was chosen because of its propensity to generate high
Workplace Data (Phase II)
Site Selection
trial site were selected and designated as “Area L” and “Area
M.” The distance between the areas was approximately 200–300
TABLE II
feet. Due to adequate air conditioning, these areas were consid-
Fraction of short-chain fatty acid collected by second stage
ered relatively clean. There was no physical barrier between the
resin cartridge
contiguous areas that impeded airflow between them such as a
wall or partition. Different machining fluids and separate MWF
treatment and distribution systems were utilized in each study
area.
Acid
NebulizedA (%)
BubbledB (%)
Isononanoic
Octanoic
Neodecanoic
Nonanoic
54
28
43
7
89
71
85
50
0
Fatty Acid Sampling Methods and Results
Phase II investigations utilized similar sampling methods and
apparatus that were used for the Phase I (laboratory) studies. Lo-
gistical considerations provided for the use of TSI Dust Trak and
Data RAM continuous monitors to measure particulate levels in
Dodecanedioic
0
A8-hour sampling.
B22-hour sampling.

840
E. M. WHITE AND W. E. LUCKE
TABLE III
No other MWF mist exposure standard or guideline has been
Results of workplace sampling
promulgated since the inception of the NIOSH REL.
It has been in the expressed interest of responsible indus-
tries to provide healthy and productive environments for their
workers. In this regard, over the years there have been nu-
merous improvements in MWF products, engineering controls,
anti-misting technologies, and industrial hygiene procedures to
reduce workplace mists. It is noteworthy that many of these ad-
vancements have occurred as a result of cooperative efforts be-
tween MWF manufacturers, MWF end-users, union, and federal
partners. As the science of MWF mist generation is better un-
derstood, we can expect more improvements in mist reduction
technologies. However, efforts to equivocate MWF mists and
mineral oils are confounded by the fact that these are clearly
different materials, and any considerations to pose limits on
them in a similar matter is neither scientifically sound nor in the
interest of worker health.
Level found in air
Analyte
(mg/m3)
Level in fluid
Nonanoic acid acid
Neodecanoic acid cid
Dodecanedioic acid
Triethanolamine
0.2
0.1
ND
3% in concentrate
13% in concentrate
1.1% in concentrate
2.5%/0.8% in mix
0.05/0.03
ND = no data.
both areas. Each monitor type yielded data with similar ranges of
0.1 to 0.4 mg/m3. Analytical results for the workplace sampling
are given in Table III.
TEA Sampling Methods and Results
In used machining fluids in Areas L and M, the respective
TEA concentrations in distribution systems were 2.5 and 0.8 per-
cent. Air sampling determined that concentrations of TEA in
field air samples were area dependent. Measurements of TEA
particulates in Area L averaged 0.05 mg/m3 and 0.03 mg/m3 in
Area M.
ACKNOWLEDGMENTS
The authors thank Milacron researchers Ann M. Ball, Jerry P.
Byers, Ed H. Rolfert, and Henry Turchin (formerly of Milacron);
and Oak Ridge National Laboratory collaborators, Ralph H.
Ilgner, Andi Palausky, and Roger A. Jenkins, for data sets that
are used in this article.
CONCLUSIONS
Metalworking fluid mists are present in the atmospheres of
occupational environments as a result of numerous factors.
These factors include high-speed dynamic forces involved in
machining operations, the chemical composition of fresh flu-
ids, and contaminants that enter fluids from extrinsic machinery
sources. Measurements of mists formed by sparging showed dis-
tributions of smaller particle diameters. The latter considerations
are important in the design of devices for the efficient removal
of mists in the workplace.
The role of tramp oil in contributing to the formation of met-
alworking fluid mists in the atmospheres of machining opera-
tions is significant for all types of fluids. Therefore, an effective
mist control program should include maintenance measures to
reduce the amount of fluid leakage from machinery. Also, pru-
dent housekeeping, adequate ventilation, and mist enclosures
can work together to provide a total system of mist reduction
practices.
Industry and regulatory issues about in-plant MWF mist gen-
eration are prompted by universal concerns for worker exposures
and health. In 1998, the Centers for Disease Control and Pre-
vention/National Institute for Occupational Safety and Health
(NIOSH) published a long-awaited criteria document dealing
with occupational exposures to metalworking fluid mists.(15)
This document stated the NIOSH-recommended exposure limit
(REL) for MWF aerosol exposure of 0.4 mg/m3 for thoracic
particulate mass or 0.5 mg/m3 for total particulate mass (time-
weighted average). The REL has undergone rigorous exami-
nation, discussion, and real-world application, and has been
accepted as a reasonable guideline by numerous stakeholders.
REFERENCES
1. Ilgner, R.H.; Palausky, A.; Ball, A.M.; et al.: Distribution of
Fatty Acids and Triethanolamine in Synthetic Metalworking Fluid
Aerosols Generated in the Laboratory and Field. In: The Industrial
Metalworking Environment: Assessment and Control of Metal Re-
moval Fluids—Proceedings of the American Automobile Manu-
facturers Association Metalworking Fluids Symposium II, D.A.
Felinski; J.B. D’Arcy, eds., pp. 173–178. AAMA, Detroit, MI
(1997).
2. Turchin, H.; Byers, J.P.: Effect of Oil Contamination on Metal-
working Fluid Mist. Lubr Eng 56(7):21–25 (2000).
3. Alarie, Y.: Dose-Response Analysis in Animal Studies: Prediction
of Human Responses. Environ Health Perspect 42(12):9 (1981).
4. Ball, A.M.; Lucke, W.E.: Evaluation of Sensory Irritation Potential
for Commercially Available Metalworking Fluids. Presented at the
48th Annual Meeting of the Society of Tribologists and Lubrication
Engineers. Calgary, Alberta, Canada (May 17–20, 1993).
5. Schaper, M.; Detwiler, K.: Evaluation of the Acute Respiratory Ef-
fects of Aerosolized Machining Fluids in Mice. Fund Appl Toxicol
16:309 (1991).
6. Kennedy, S.M.; Greaves, I.A.; Kriebel, D.; et al.: Acute Pulmonary
Responses Among Automobile Workers Exposed to Aerosols of
Machining Fluids. Amer J Indus Med 15:627 (1989).
7. Woskie, S.R.; Smith, T.J.; Hallock, M.F.; et al.: Size-Selective Pul-
monary Dose Indices for Metal-Working Aerosols in Machining
and Grinding Operations in the Automobile Manufacturing Indus-
try. AIHAJ 55(1):20 (1994).
8. Chan, T.L.; D’Arcy, J.B.; Siak, J.: Size Characteristics of Machin-
ing Fluid Aerosols in an Industrial Metalworking Environment.
Appl Occup Environ Hyg 5(3):162 (1990).

EFFECTS OF FLUID COMPOSITION ON MIST COMPOSITION
841
9. Nungesser, E.H.; Jaycock, M.A.; Doshi, D.R.; et al.: A Laboratory 12. Nimz, E.L.; Morgan, S.L.: On-Line Derivatization for Complex
Microcosm Model to Determine the Fate of Components from Met-
alworking Fluid Systems. Symposium Proceedings of the Indus-
Fatty Acid Mixtures by Capillary Gas Chromatography/Mass
Spectrometry. J Chromat Sci 31:145 (1993).
trial Metalworking Environment: Assessment and Control, p. 378. 13. Turchin, H.; Rolfert, E.H.: Characterization of Fluid Misting. U.S.
Dearborn, MI (November 13–16,1995). Patent 5,889,201 (1999).
10. Leith, D.; Leith, F.A.; Boundy, M.G.: Measuring the Concentration 14. Woskie, S.R.; Smith, T.J.; Hammond, S.K.; et al.: Factors Affecting
of Mineral Oil Mists. Symposium Proceedings of the Industrial
Metalworking Environment: Assessment and Control, pp. 189–
195. Dearborn, MI (November 13–16, 1995).
Worker Exposures to Metal-Working Fluids During Automotive
Component Manufacturing. Appl Occup Environ Hyg 9(9):612–
621 (1994).
11. Hawthorne, D.B.; Kavanagh, T.E.; Clarke, B.J.: Determination of 15. National Institute for Occupational Safety and Health: Criteria for a
Low Molecular Weight Organic Compounds in Beer Using Capil-
lary Gas Chromatography. Amer Soc Brewing Chemists J 45(1):23
(1987).
Recommended Standard—Occupational Exposure to Metalwork-
ing Fluids, DHHS (NIOSH) Pub. No. 98-102. NIOSH, Cincinnati,
OH (1998).