MWF Biocides Part II: Science vs. Fiction

Frederick J. Passman, Neil M. Canter, Richard Rotherham, Jerry P. Byers and Alan C. Eachus | TLT Technical Analysis March 2016

In both the U.S. and Europe, examining the research goes a long way toward setting the record straight.
 
KEY CONCEPTS
• The health risks associated with formaldehyde-condensate have been conflated with those associated with formaldehyde and are not supported by scientific data.
• Formaldehyde-condensate microbicides are an essential part of the ever-shrinking list of biocidal products that are approved for use in metalworking fluids.
• Aldehyde-based microbicides are unique in their ability to denature endotoxin and thereby reduce the risk of respiratory disease caused by endotoxin exposure.

THE NOVEMBER 2015 ISSUE OF TLT contained an article titled Biocides: Both Problem and Solution. While this timely article addressed a topic that is foremost in the minds of many metalworking fluid (MWF) formulators, resellers, fluid managers and end-users, it only scratched the surface of an issue that is both fascinating for tribologists and under scrutiny by regulatory agencies in the U.S. and other countries.

This follow-up article expands the information in the November article and is based on consensus documents (primarily ASTM standards), peer-reviewed literature and regulatory agency material to provide an update on the current status of U.S. and European regulations affecting the registration and use of microbicides in MWFs.

DDT VERSUS FORMALDEHYDECONDENSATE MICROBICIDES
The U.S. EPA and European regulatory agencies have conflated F-C microbicides with formaldehyde based on their erroneous assumption that formaldehyde- condensate molecules will completely hydrolyze to free-formaldehyde plus the other reactive intermediate(s) while in solution.

After decades of use—often applied in aerosol form in the same manner as many agricultural pesticides—Dichlorodiphenyltrichloroethane (DDT) was discovered to be bioresistant (half-life raging to 30 years), bioaccumulative and—because of its adverse impact on eggshell thickness—contributing to marked declines in the populations of various species of birds. The public outrage over DDT’s ecotoxicological properties led to it ultimately being banned by many countries. Despite its effectiveness against the insect vectors of malaria, dengue fever and typhus, DDT became the symbol of indiscriminate pesticide use. Moreover, the unequivocal link between DDT bioaccumulation and near extinction of a variety of bird populations eclipsed DDT’s public health benefits.

In the 1980s, formaldehyde (HCHO) came under regulatory pressure after it was reported that HCHO was associated with an increased cancer rate among workers in 10 high-exposure industries (IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 8, 2006, Available here).

In this IARC report, the standardized mortality ratio (SMR, a measure of the number of deaths due to all causes or a specific cause among members of a designated [exposed] population relative to the general putatively unexposed population) among HCHO-exposed workers for all cancers was 0.76 (95% confidence interval (CI: 0.69 – 0.84) and for nasopharyngeal cancers was 2.10 (95% CI: 1.05 – 4.21)). This means that while the overall incidence of cancer among exposed workers was significantly less than that of the general population, exposed workers were significantly more likely than the background population to develop nasopharyngeal cancers.

The acute toxicity (LD50) of HCHO in mice is 42 mg/kg, and its acute inhalation toxicity (LC50) is 505 mg/m3 (4h exposure; mice; Nagorny, P.A., Sudakova, Zh. A., and Schablenko, S.M., 1979. On the general toxic and allergic action of formaldehyde. Gig. Tr. Zabol. 1:27-30). Formaldehyde is currently listed by IARC as a probable human carcinogen (classification 2A).

In contrast, the acute oral toxicity of Hexahydro-1,3,5-tris (2-hydroxyethyl)- s-triazine (HTHT – the most commonly used formaldehyde-condensate – F-C – microbicide) is 560±32 mg/kg (rats). Inhalation toxicity testing depends on being able to expose test animals to vapors containing the test substance. In contrast to HCHO (vapor pressure @ 25° = 519 kPa; 3,890 mm Hg) the vapor pressure of HTHT is <0.01kPA; <0.1 mmHg @ 25°C. This means that under inhalation test conditions, there is no measurable HTHT in the air. Consequently, the acute inhalation toxicity of HTHT is undeterminable (data from HTHT’s SDS; Troy Chemical Corp.). Moreover, HTHT’s low vapor pressure means that HTHT is unlikely to cause adverse health effects due to inhalation of its vapors.

At present there are currently (as of November 2015) 27 products (biocidal chemistries) with active dossiers under the 2012 EU Biocidal Products Regulation (BPR) under Product Type 13 (Working or cutting fluid preservatives). This list of 27 products includes 11 formaldehyde condensates—including Hexahydro-1,3,5- tris (2-hydroxyethyl)-s-triazine, listed as “2,2',2"-(hexahydro-1,3,5-triazine- 1,3,5-triyl) triethanol.”

There is no evidence that there is an incremental health, safety or environmental risk posed by MWFs treated with F-C microbicides, relative to those that are either treated with alternative microbicides or those putatively formulated to be biocide-free. If science is permitted to trump fiction, then the current hysteria over the use of F-C microbicides will disappear like morning mist in the heat of a rising sun.

The assumption by U.S. EPA and the European regulatory agencies in conflating F-C microbicides with formaldehyde is derived from the traditional test methods used to measure F-C concentration in application. The first step of these methods is acid hydrolysis. The hydrolyzed sample is then reacted with a chromotrophic reagent such as 2,4-dinitrophenylhydrazine (DNPH) and formaldehyde concentration is then determined spectrophotometrically. The concentration of the original F-C product is then computed based on the molar ratio of formaldehyde to parent molecule (for example: for HTHT this ratio is 3). In MWFs only a fraction of the F-C molecule actually hydrolyses (estimates range from 1%-5% by weight). Limited actual hydrolysis and formaldehyde release has recently been demonstrated by C13 nuclear magnetic resonance (NMR) spectroscopy.

These data demonstrate unequivocally that F-C products should not be conflated with formaldehyde. Moreover, these chemical analyses are consistent with the toxicological data that show that formaldehyde is >13 to 21 times as toxic as the most commonly used F-C biocides (the acute oral LD50 for 3,3'-Methylenebis [5-methyloxazolidine] – MBO – is 900 mg/kg in rats). Based on the molar ratio of formaldehyde to parent molecule, if these products indeed hydrolyzed 100% in solution, their acute oral toxicities would be expected to be approximately one-third that of formaldehyde.


© Can Stock Photo Inc. / zhuzhu

The U.S. EPA’s 2010 IRIS report on formaldehyde listed studies that concluded that there was an increased risk of death due to leukemia, particularly myeloid leukemia, among workers exposed to formaldehyde. These studies reflected exposures to airborne formaldehyde at ≥ 5.0 mg/m3 (most notably workers installing polyurethane foam insulation). The current permissible exposure level for formaldehyde is 0.5 mg/m3 8-hour, time-weighted average (8h TWA). Studies of formaldehyde in the air around MWFs have rarely detected ≥0.5 mg/m3 and have never reported ≥0.5 mg/m3 8h TWA. Moreover, F-C chemistries represent only one of numerous potential point sources of formaldehyde—including biogenic formaldehyde (formaldehyde released as a metabolic byproduct of microbes and higher organisms including humans).

The U.S. EPA’s IRIS Report was criticized by the National Academy of Sciences (NAS) review panel who sent the document back to U.S. EPA for revision. U.S. EPA has yet to submit the revised report. Among other issues NAS questioned the study that implicated formaldehyde as an agent of leukemia. Specific confounding factors have been identified in that study (exposure to benzene, a confirmed leukemogen). There is no known biological model for formaldehyde to cause leukemia. This is supported by an inhalation study with C13 formaldehyde, which showed no C13 in bone marrow. Moreover, the dose response model that the U.S. EPA used to assess formaldehyde risk is no longer accepted. It has been superseded by the biological response model—a model that recognizes the no observable effect level (NOEL).

The substantially different toxicological profiles of F-C microbicides in contrast with formaldehyde anticipated what C13-NMR has subsequently demonstrated (Passman, F.J. “Formaldehyde Risk in Perspective: A Toxicological Comparison of Twelve Biocides.” Lub. Eng. 52 (1): 68-80; 1996). Moreover, we are not aware of any peer-reviewed report of exposure to F-C condensate treated MWF having caused a single case of nasopharyngeal cancer in the >60 years since F-C microbicides were first used in MWFs.

Moreover, regulatory pressure against the use of F-C microbicides makes no reference to the well-documented ability of aldehyde biocides to denature endotoxins. Endotoxins are biomolecules that are associated with Gram-negative bacterial cells walls. Passman has reviewed the endotoxin allergenicity and toxicity literature in two reviews of health risks associated with MWF microbe exposure (Passman, F.J. and H.W. Rossmoore. “Reassessing the Health Risks Associated with Employee Exposure to Metalworking Fluid Microbes.” Lub. Eng. 58 (7): pp: 30-38; 2002; and Passman, F.J. “Metalworking Fluid Microbes – What We Need to Know to Successfully Understand Cause and Effect Relationships.” Tribol. Trans. 51(1): 107-117; 2008). Only aldehyde-based microbicides (glutaraldehyde, formaldehyde and F-C products) denature endotoxin (Douglas, H., H. W. Rossmoore, F. J. Passman and L.A. Rossmoore. “Evaluation of Endotoxin-Biocides Interaction by the Limulus Amoebocyte Assay.” Devel. Ind. Microbiol. 31: 221-224; 1990).

U.S. EPA’s Office of Pesticide Programs (OPP) is considering restricting the maximum concentration of triazine to 500 ppm active HTHT in diluted MWF based on a misreading of Makku Linnainmaa and her coworkers (Linnainmaa et al., “Control of Worker’s Exposure to Airborne Endotoxins and Formaldehyde During the Use of Metalworking Fluids.” AIHA Journal 64:496-500; 2003): “The results of this study showed that triazine in levels over 500 ppm in MWF prevented bacterial growth in the fluid and kept the concentrations of endotoxins below 400 ng/mL.”

The observations reported in the paper showed that at ≤ 500 ppm, triazine was ineffective. The U.S. EPA read the sentence to mean that triazine was effective at 500 ppm.

Hexahydro-1,3,5-tris(2-hydroxyethyl)- s-triazine is the only triazine product currently approved by U.S. EPA for use in MWFs. The MWF end-use site for Hexahydro-1,3,5-triethyls- triazine was withdrawn more than a decade ago. In ASTM E2169 Standard Practice for Selecting Antimicrobial Pesticides for Use in Water-Miscible Metalworking Fluids (DOI: 10.1520/ E2169-12, www.ASTM.org), Table 2 lists all active ingredients approved for use as antimicrobials in MWFs as of 2012.

In short, although the current focus on F-C microbicides is real, it is driven by emotionalism and misconceptions. It is essential that STLE stakeholders recognize the difference between science and hysteria.

TYPES OF MWFs
As stated in ASTM D2881 Standard Classification for Metalworking Fluids and Related Materials, the four main types of MWFs are:
1. Straight oil. Contains petroleum oil but essentially no water, is not emulsifiable and can contain functional additives.
2. Emulsifiable oil. Frequently referred to as a soluble oil, it generally creates a macroemulsion (average micelle size is greater than 1 micron) when dispersed in water. Primary base stock is petroleum oil and contains little or no water, contains emulsifiers and other functional additives and is blended with water in its end use.
3. Semisynthetic fluid. Generally, creates a microemulsion (average micelle size is less than 1 micron) when dispersed in water. Primary base stocks are petroleum oil and water, contains functional additives and is blended with water in its end use.
4. Synthetic fluids. This fluid type is further classified in three subcategories: solution synthetic fluid, emulsion synthetic fluid and straight synthetic oils:
a. Solution Synthetic Fluid (also known as a chemical solution): Contains no petroleum oil, contains functional additives, forms a single-phase, true solution (no micelles) when further diluted with water prior to use.
b. Emulsion Synthetic Fluid: PrU.S. EPAred from natural (typically vegetable oils) or synthetic triglycerides, esters or other synthetic base stocks, contains emulsifiers and other functional additives but no petroleum oil, produces an emulsion when further diluted with water prior to use.
c. Straight Synthetic Oil: Contains no petroleum oil nor water, formulations typically prU.S. EPAred with renewable triglycerides, synthetic hydrocarbons, esters or other oil-soluble base stocks, generally combined with oil-soluble additives that contain no water and not intended to be diluted nor dispersed in water in its end use.

As noted, MWFs containing biobased base stocks are clearly placed by industry consensus in the synthetic fluid category.

Moreover, U.S. EPA has established definitions for biopersistence based on percent of biodegradation in various test systems and by various test methods. Given that there are several approved methods and that biopersistence also depends on the environment (aquatic, solid, etc.), its definition is test-method specific. For example: OECD 301 states that to qualify as readily biodegradable, a substance must be degraded by >60% within the 28-day test period. Additionally, >60% of the substance’s maximum degradation must occur within 10 days after 10% biodegradation is observed.

MICROBIAL ECOLOGY OF MWFs
For a full discussion of this topic, the reader is referred to Passman, 2006, “Chapter 9 – Microbiology of Metalworking Fluids,” In: J. Byers. Ed. Metalworking Fluids, CRC Press, New York, pp: 195-229. Microbes don’t rely on MWF impurities to thrive in in-use MWFs.

Microbes require macronutrients that contain carbon, hydrogen, nitrogen, oxygen, sulfur and phosphorous. These macronutrient elements are provided in MWF base stocks and the various performance additives formulated into finished MWFs. Given that water-miscible MWFs are diluted to end-use concentrations ranging from 2% by volume to 10% by volume in water, the diluted MWFs provide a superb medium for microbial growth. Moreover, the turbulent flow conditions that are characteristic of recirculating MWF systems provide optimal conditions for the growth and proliferation of aerobic microbes—bacteria and fungi that require oxygen. In systems with stagnant zones, anaerobic bacteria—bacteria that only grow in oxygen-free environments— feed off of the waste products produced by aerobes and off of particularly biodegradable MWF components.

The malodorous gases—commonly referred to as Monday morning odor— are invariably released after system recirculating pumps that have been left idle for the weekend are turned back on. During normal operations, microbially generated noxious gases (for example: hydrogen sulfide, low molecular weight fatty acids and various sulfur-containing alcohols) are oxidized as they diffuse through well-aerated, recirculating MWFs. When systems are left idle, these gases accumulate. Their rapid release upon restoration of recirculation—generally on Monday mornings—overwhelms the system’s ability to oxidize them before they are expelled into the facility’s atmosphere. The result is the overwhelming aroma of swamp gas. The most cost effective tactic for preventing Monday morning odor is to control microbial contamination effectively.

Make-up water is the primary source of MWF microbial contamination. It is also the primary source of the various inorganic micronutrients microbes need in addition to the aforementioned macronutrients. Consequently, the quality of water used to dilute MWFs has a major impact on microbial contamination in MWFs. Although all MWF components are made from two or more of the macronutrients listed above, more complex molecules tend to resist microbial attack better than less complex molecules do.

In recent years, an increasing number of bioresistant functional additives have been introduced to the market. A credible, bioresistant additive demonstrates one or more performance properties other than microbial contamination control. If added to a heavily contaminated MWF, it will not have any immediate effect on the microbial population. However, it will typically improve the finished formulation’s ability to withstand microbial challenges.

BIOCIDES
Biocides are products designed to control or kill one or more pests. Microbicides (antimicrobial pesticides) are biocidal products that are designed to specifically target microorganisms.

Any product that has an immediate kill effect on MWF microbial populations but does not have any other demonstrable function is a microbicide. In the U.S., all industrial microbicides must be registered with U.S. EPA’s Office of Pesticide Programs. In the EU microbicides must be registered in accordance with the Biocidal Product Regulation (BPR, EU No 528/2012). ASTM E2275 Standard Practice for Evaluating Water-Miscible Metalworking Fluid Bioresistance and Antimicrobial Pesticide Performance (DOI: 10.1520/E2275-14, www.ASTM.org) provides guidance on how to evaluate the biocidal or bioresistance performance properties of MWF components or finished formulations.

It is incorrect to assume that the substantially larger portfolio of toxicological data needed to support microbicide registration applications reflects greater hazard. Any chemical or device intended to be used to control one or more types of pests must be registered with U.S. EPA’s OPP. OPP approvals are site (pest and end-use application) specific. There are countless chemistries that are not intended for use as pesticides that are considerably more toxic than microbicides. The basis for requiring pesticides to have more complete toxicological data sets than other chemicals is primarily economic and political rather than logical. People handling concentrated inorganic acids such as hydrochloric or sulfuric acids are exposed to substantially greater risk than those handling F-C microbicides. The implementation of REACh in Europe recognizes and attempts to address this discrepancy.

A perfect example of this is a 0.5% solution of sodium hypochlorite. When purchased in the grocery section as bleach, the product has a label with minimal health and safety information. When the same product is purchased as an algaecide, it comes with a foldout label that lists numerous health and safety considerations, provides detailed information for safe handling and use, and includes a summary of the product’s toxicological profile. The only difference is the sodium hypochlorite’s intended use.

As noted above, microbicides are approved for use in specific products against specific pests. The three primary factors affecting application approvals are toxicity, exposure and efficacy. Historically, U.S. EPA has not required efficacy data, but that has changed with the 2012 revision to 40CFR158 Subpart W-Antimicrobial Pesticide Data Requirements. The BPR has always required performance efficacy data. A major challenge to the development of reliable performance data is that for many end-use sites, there are no consensus methods for testing microbicide efficacy.

The health risk associated with microbicide use is a function of toxicity and exposure. The U.S. EPA bases microbicide approvals on the risk posed to those using the products; the EU considers only toxicity. Risk decisions are inherently more complex. For example, should permissible dose concentrations be based on the potential (or likely) exposure of personnel handling microbicide concentrate or on likely exposures of personnel working in the metalworking environment? Should exposure risk be based on worst case scenarios or most common conditions? How should microbicide exposure risk be balanced against disease and biodeterioration risks associated with not using microbicides? These are complex issues that are well beyond the scope of this article. Important here is the recognition that the use of hazardous, non-microbicidal products should raise similar questions.

MWF microbicide performance is generally tested against microbes in the bulk fluid (planktonic microbes). It is important to keep in mind that only a fraction of the total microbial biomass in any MWF system is present in the recirculating fluid. For every microbe/ mL in the MWF there are likely to be ≥1,000 microbes/cm2 of system surface. These surface-associated (sessile) microbes are invariably found with biofilms. Biofilm communities share many properties in common with fixed-film biological reactors, commonly used in biotechnology to convert organic feedstocks into products. Inadequate control of biofilm communities can result in apparently cryptic, fluid stability and performance problems when the bioburden in recirculating MWFs is below detection limits but all other test results indicate biodeterioration. For more information about MWF system biofilms, read “Emerging Issues in Metalworking Fluid Microbiology: Biofilm Control,” available on the Webinar section of www.stle.org.

CLARIFICATION OF BIOCIDE TYPES
There is some confusion regarding the terms formaldehyde-condensate, formaldehyde releasing and formaldehyde-condensate, formaldehyde non-releasing. For example, the theory that nitromorpholine does not release HCHO is most likely an artifact of experimental design. As part of his doctoral thesis research, Mohammad Sondossi compared the ability of test microbes to become formaldehyde resistant after exposure to increasing concentrations of either formaldehyde or formaldehyde-condensate microbicides. Microbes exposed to either 2-(hydroxymethyl) -2-nitro-1,3-propanediol (TRIS NITRO®) or nitrobutylmorpholine did not develop formaldehyde resistance. Those exposed to formaldehyde or other F-C microbicides became formaldehyde-resistant. The flaw in the study was that TRIS NITRO is well known to hydrolyze under MWF conditions. Free-formaldehyde can be detected quite quickly after TRIS NITRO is added to end-use diluted MWF. If TRIS NITRO did not select for formaldehyde resistance, the assumption that nitrobutylmorpholine does not release formaldehyde is questionable. We are unaware of actual C13-NMR studies on nitrobutylmorpholine hydrolysis in MWFs. Consequently, the statement “nitrobutylmorpholine does not release formaldehyde” is at best speculative.

Benzisothiazolin-3-one (BIT) and other non-F-C microbicides can be effective, but they are much more formulation sensitive. Prof. Ed Bennett used to routinely test microbicide performance in >200 different MWF formulations. He often reported his results in Lubrication Engineering, TLT’s predecessor magazine. Regardless of the product, invariably there would be MWFs in which the microbicide-treated formulation would be less bioresistant than the control. In other MWFs the microbicide would have no significant effect, and in others the microbicide would be effective. The microbicides that became commercially successful were those that were effective in the greatest percentage of MWFs. The range of MWFs in which BIT is effective is substantially less than the range in which F-C products are effective.

® TRIS NITRO is a registered trademark of Dow Chemical Co., Midland, Mich.

OTHER MWF CONCEPTS
Earlier in this article, the most commonly used F-C-microbicide was listed under the chemical name Hexahydro-1,3,5- tris(2-hydroxyethyl) -s-triazine (HTHT). This biocide is not referred to as 1,3,5-triazine nor cyanuric chloride. No chlorine is present in HTHT.

Formaldehyde is also known under additional synonyms including formic aldehyde, methanal, methyl aldehyde, oxomethane and oxymethylene. Formaldehyde is not methanol, as formaldehyde is chemically known as a member of the aldehyde class, and methanol is a member of the alcohol class. It also is incorrect to refer to formaldehyde as formalin. In actuality, formalin is a 37% solution of formaldehyde gas in water usually stabilized with methanol to prevent polymerization.

ADDITIVES IN WATER-BASED MWFs
Water-based (also known as water-dilutable and water-miscible) MWFs contain a large number of specific additives due to the many functions that the fluid needs to perform. The additive classes found in the water dilutable MWF known as emulsifiable oils, semisynthetic fluids and synthetic fluids are shown below in alphabetical order:
• Antifoam additives
• Antimicrobial pesticides (biocides)
• Boundary lubricity additives
• Corrosion inhibitors
• Coupling agents
• Dyes
• Emulsifiers
• Extreme pressure agents
• Metal deactivators
• Reserve alkalinity boosters (mainly alkanolamines)
• Wetting agents.

Reference: Canter, N. (1997) “Additives for Metalworking Fluids,” in Booser, E. editor, “Tribology Data Handbook,” CRC Press, Boca Raton, FL, pp. 862-871.

NON-CHEMICAL AND NON-BIOCIDAL MICROBIAL CONTAMINATION CONTROL STRATEGIES
Pasteurization and other point-source biocidal measures are inadequate. In an MWF with 106 microbes/mL, even 99.99% kill leaves 102 survivors/mL. A percentage of these survivors are likely to settle on MWF system surfaces, proliferate and recontaminate the bulk MWFs. Point-source decontamination systems have been evaluated for >40 years. Some perform well under controlled laboratory conditions, but to date none has worked well in application.

Controlling pH does not kill bacteria. MWFs running at pH 9.2 are as likely to have high bioburdens as those running at pH 8.5. The source of this myth is unclear, but it has been pervasive since as early as the mid-1970s. It is a myth that is most likely based on culture test data. Too often culture tests are terminated after 36 to 48 hours. Many MWF microbes require 5-15 days to form visible colonies. Consequently, culture tests are commonly misinterpreted.

Fluid management best practices (for example, see Foltz, G., J., Metalworking Fluid Management and Troubleshooting, Chapter 11. In J. P. Byers, Ed. Metalworking Fluids, 2nd Ed., CRC Press, New York, pp: 253- 278; 2006) reduce but do not eliminate microbial contamination control problems. Use of quality MWFs, good water, effective condition monitoring and timely, data-driven actions are the best means for minimizing biodeterioration and microbe-associated health risks in the metalworking environment.

BIOSTABLE MWFs
As noted above, ASTM E2275 Standard Practice for Evaluating Water-Miscible Metalworking Fluid Bioresistance and Antimicrobial Pesticide Performance provides protocols for testing quick-kill and bioresistance properties of MWFs and MWF additives. It is not uncommon for a functional additive to exhibit quick-kill performance (causes the microbial population density to drop precipitously in a matter of hours). If that additive is not used for its antimicrobial properties, it can be formulated into an MWF without having a pesticide registration. However, it is misleading to suggest that substitution of a microbicide with such a product reduces the health and safety risks associated with use of MWFs.

Formulators must resist the urge to formulate with toxic molecules that serve no other function than to suppress microbial growth. There are an increasing number of molecules available that have well-documented, non-antimicrobial performance properties, do not kill microbial populations (per E2275) when added to heavily contaminated MWFs but provide good bioresistance (≥ 6 weeks of bioresistance by E2275). Additives that meet these latter three criteria are appropriate for use in formulating bioresistant MWFs. This is critical guidance to both formulators and end-users.

MEASURING MICROBES
The types of microbes most commonly recovered from MWFs include bacteria and fungi. The focus on these two groups is linked to their well-documented biodeteriogenic activity. Given that make-up water is the primary source of microbes in MWFs, it is not unlikely that algae, archaea, protozoa and viruses would also be detected if sought for. Bacteria and Archaea are kingdoms on the current, phylogenic tree of life. Algae, fungi and protozoa are all branches of the kingdom Eukaryota. Viruses are believed to have either degenerated from bacteria or to have developed from nucleic acids (see below) that survived outside the microbes from which they originated. The term microbe is the diminutive form of the word microorganism. The ASTM consensus definition of microbe is: “bacteria and other organisms that require the aid of a microscope to be seen.”

To detect microbes in MWF and metalworking systems, one must have reliable test methods. The first challenge is to obtain a suitable sample. Although ASTM D7464 (Standard Practice for Manual Sampling of Liquid Fuels, Associated Materials and Fuel System Components for Microbiological Testing; DOI: 10.1520/D7464-14, www.ASTM.org) specifically addresses fuel system sampling, the general principles it explains are equally valid for MWFs. Current industry practice is to test for microbial contamination with an agar-coated dipslide or paddle. Most commercially available paddles have two different types of growth media. A medium for growing bacteria is on one side and a medium for growing fungi is on the other side. These dipslides are a simplified version of standard plate count methods (see D5465 Standard Practice for Determining Microbial Colony Counts from Waters Analyzed by Plating Methods; DOI: 10.1520/ D5465-93R12, www.ASTM.org). In order for microbes to be detected by culture testing, they must be able to proliferate (form colonies) on the growth medium under the test conditions (incubation temperature, oxygen availability, incubation period) used. Consequently, any single culture test will only detect a fraction of the total microbial community in an MWF sample.

Although all test methods have inherent limitations, non-culture methods typically detect larger proportions of the total population. For example, the catalase test (Gannon, J. and Bennett, E.O. (1981), “A Rapid Method for Determining Microbial Loads in Metalworking Fluids,” Tribology 14: 3-6.) detects most aerobic bacteria and all metabolically active fungi but does not detect anaerobic bacteria or those aerobic bacteria that do not have a complete catalase enzyme. Endotoxin testing (ASTM E2144 Standard Practice for Personal Sampling and Analysis of Endotoxin in Metalworking Fluid Aerosols in Workplace Atmospheres; DOI: 10.1520/E2144-01, and E2657 Standard Test Method for Determination of Endotoxin Concentrations in Water-Miscible Metalworking Fluids; DOI:10.1520/E2657-11, www.ASTM.org) detects the presence of whole or disintegrated Gram-negative bacteria, but does not detect Gram-positive bacteria or fungi. Adenosine triphosphate (ATP) testing (ASTM E2694 Standard Test Method for Measurement of Adenosine Triphosphate in Water-Miscible Metalworking Fluids, DOI: 10.1520/ E2694-11, www.ASTM.org) detects all metabolically active microbes but does not detect dormant cells. Recent advances in ATP testing have made it possible to differentiate between bacterial and fungal contamination (Passman, F.J. and Küenzi, P., “A Differential Adenosine Triphosphate Test Method for Differentiating between Bacterial and Fungal Contamination in Water-Miscible Metalworking Fluids” International Biodeterioration & Biodegradation; 2014, Available here, 0964-8305.)

No individual test method can provide all of the microbiological information that might be helpful for MWF management. Method selection should be based on careful consideration of the type of information needed and the importance of timeliness. Other consensus test methods that are likely to be useful to personnel responsible for MWF condition monitoring are:
• E2563 Standard Practice for Enumeration of Non-Tuberculosis Mycobacteria in Aqueous Metalworking Fluids by Plate Count Method; DOI: 10.1520/E2563, www.ASTM.org.
• E2564 Standard Practice for Enumeration of Mycobacteria in Metalworking Fluids by Direct Microscopic Counting (DMC) Method; DOI: 10.1520/E2564, www.ASTM.org.
• D4412 Standard Test Methods for Sulfate-Reducing Bacteria in Water and Water-Formed Deposits; DOI: 10.1520/D4412-15, www.ASTM.org.

Genomic testing recently has been applied to MWF microbiology. The genomic test most commonly used is polymerase chain reaction (PCR) methods. Quantitative PCR (qPCR) purports to be able to determine both the taxonomic profile and the relative abundance of each type of microbe (operational taxonomic unit – OTU) in the original sample. Reliable qPCR data depend on the deoxyribonucleic acid (DNA) extraction step and selection of sections of ribosomal ribonucleic acid (rRNA) used as primers. Due primarily to the historical cost per test, few reports of qPCR population profiles address data variability. More recently, improved metagenomic methods such as environmental shotgun sequencing (ESS) have been developed (Tyson GW, Chapman J, Hugenholtz P, Allen EE, Ram RJ, et al, “Community structure and metabolism through reconstruction of microbial genomes from the environment,” Nature, 2004; 428:37-43). Rather than depending on rRNA primers, these methods capture a much broader range of genes, thereby providing much more intimate information about the contaminant population.

Having better information about what types of microbes make up the structure of MWF microbial populations is important, but the real question is: “What is the population doing?” This is where proteomics offers promise. Proteomic methods provide a profile of gene expression in samples by detecting enzymatic proteins. Except for the special case in which biomass accumulation plugs filters and lines, biodeterioration is the result of enzymatic activity.

By understanding the relationship between MWF conditions and gene expression, it might be possible to formulate MWFs that inhibit the expression of genes that play major roles in biodeterioration. Both metagenomics and proteomics are in their respective infancies. Moreover, in an industry accustomed to investing <$5 U.S/test, the cost of obtaining metagenomic data ($150 to $300/test) or proteomic data (thousands of U.S. dollars per test) is still prohibitive for them to be used outside limited research studies. In the future, the information that can be obtained from these emergent technologies might well change how we control microbial contamination in MWFs.

HEALTH EFFECTS OF MWF MICROBES
The earliest research on MWF microbiology was focused on the recovery of pathogenic microbes (Bennett, E.O. and Wheeler, H.O. (1954), “Survival of Bacteria in Cutting Oils.” Appl. Microbiol. 2, pp. 368-371). By the mid-1970s it was apparent that although potentially pathogenic microbes were recovered from MWFs routinely, they were not causing infectious diseases among workers (Rossmoore, H.W. (1979), “Do Metalworking Fluid Microbes Cause Disease?” The Lubricator 6(3)).

Few frankly pathogenic microbes are found in MWFs. Putatively, any microbe can become an opportunistic pathogen. The human microbiome project has reported that there are approximately 10 times as many microbial cells as there are human cells in and on the average human adult. Emerging research is beginning to show how perturbations to either the population profile or environmental niches of skin, gut, respiratory microbiomes can result in a disease state. This is different from infection—pathogenicity—by frank pathogens such as Vibrio cholerae, etc. Moreover, Ed Bennett found that dermatitis was invariably a reaction to MWF chemicals, not symptomatic of microbial infection. This issue is addressed in both Passman and Rossmoore (op. cit.) and Passman, 2008 (op. cit.). We’d cite Prof. Bennett’s papers directly, but they all predate STLE’s electronic archive.

In terms of serious infections from untreated cuts or abrasions, there are only two documented cases of necrotizing fasciitis among machinists. Given that there are an estimated 1 to 1.5 million person-years exposure to MWFs annually, the incremental risk to machinists is immeasurably small. Moreover, the risk is comparable to that posed by leaving any cut or puncture wound untreated.


© Can Stock Photo Inc. / Jezper

This is not to say that MWF microbes are not associated with disease. The effects of endotoxin exposure are well documented (see Passman, 2008, op. cit.). Between 1990 and 2000, there were several clusters of workers diagnosed with the allergenic respiratory disease, hypersensitivity pneumonitis. During that period, nearly >250 cases were reported. The total incidence of HP among machinists has yet to reach 300 cases. While HP clusters were being reported frequently, it was speculated that the microbe Mycobacterium immunogenum was the causative agent. It was also speculated that F-C microbicides either enriched for M. immunogenum or stimulated its growth by suppressing the microbes that were more commonly found in MWFs. Both speculations have subsequently been disproven.

As reported in Passman, 2008 op. cit., although Mycobacterium immunogenum undoubtedly caused HP, it is only one of a dozen MWF microbes known to do so. Speculation that Mycobacterium immunogenum was the only microbe responsible for HP among machinist discounted reports that linked HP-clusters to exposure to any of the dozen other microbes that are both routinely recovered from MWFs and known to cause HP (for example: Alternaria species, Aspergillus species, Aureobasidium pullulans, Bacillus subtilis, Bacillus cereus, Acremonium (formerly Cephalosporium) species, Cryptostroma corticale, Enterobacter agglomerans, Penicillium species, Mycobacterium immunogenum species, and Thermoactinomyces species; Passman, F.J. and H.W. Rossmoore. “Reassessing the Health Risks Associated with Employee Exposure to Metalworking Fluid Microbes.” Lub. Eng. 58 (7): pp: 30-38; 2002).

The apparent inverse relationship between M. immunogenum and other common MWF microbes turned out to be an artifact of test methodology. Although M. immunogenum is considered to be a fast-growing mycobacterium, its generation time is >4x longer than that of many other MWF microbes. It takes seven to 10 days for an M. immunogenum colony to appear on culture media. When faster growing microbes are present, their colonies expand to form a uniform covering over the growth medium (confluent growth). This masks M. immunogenum colonies that might develop later. Subsequent research demonstrated that the presence or absence of M. immunogenum was unrelated to the presence or absence of other microbes (Passman, F.J., Rossmoore, K. and Rossmoore, L. “Relationship between the Presence of Mycobacteria and Nonmycobacteria in Metalworking Fluids.” Tribol. Lub. Technol. 65(3): 52-55; 2009). By extension, this research proved false the hypothesis that F-C microbicides either enriched or selected for M. immunogenum proliferation in MWFs.

MWF AEROSOL EXPOSURE
Before the 1990s it would not have been inaccurate to report that workers are often in close proximity to the mist for an entire shift. Although mist is invariably created at the tool-workpiece interface and is often generated as used MWFs flow turbulently through return sluices/troughs, industry practices during the past 20 years have reduced employee exposure dramatically. In NIOSH-sponsored surveys, median MWF aerosol exposures fall well below the 0.5 mg/m3 REL which, in turn, is 10% of the PEL (REL: recommended exposure limit; PEL: permissible exposure limit).

CONCLUSIONS
As a family of chemicals, microbicides are diverse. They differ in their respective toxicological profiles (for example: Bronopol (BNPD) is used as a preservative for antacid tablets), modes of action, compatibilities with other MWF components, spectrum of microbes against which they are effective and persistence when used in MWFs. The F-C microbicides also are quite varied in their chemical and toxicological properties (BNPD is an F-C). Toxicologically, F-C are quite distinct from free-formaldehyde. Moreover, F-C microbicides are unlikely to hydrolyze substantially when used in MWFs. It is true that F-C microbicides are under increasing regulatory pressure. Regrettably, this regulatory pressure is not based on any sound science.

Dr. Frederick J. Passman is an STLE fellow, holds the society’s Certified Metalworking Fluids Specialist™ (CMFS) certification and heads Biodeterioration Control Associates, Inc., in Princeton, N.J. You can reach him at fredp@biodeterioration-control.com.

STLE board member and TLT columnist Dr. Neil M. Canter holds the CMFS certification and heads his own consulting company, Chemical Solutions, in Willow Grove, Pa. You can reach him at neilcanter@comcast.net.

STLE member Richard Rotherham is global business manager with Troy Corp. in Concord, Ontario, Canada. You can reach him at rotherhr@troycorp.com.

Now retired, STLE Past President Jerry P. Byers also holds STLE’s CMFS designation. He is editor of Metalworking Fluids, Second Edition, published by CRC Press and available at www.stle.org. You can reach him at jpbyers@aol.com.

Now retired, Dr. Alan C. Eachus is an STLE Life Member. You can reach him at drace.dbd@comcast.net.