But both the regulatory and end-user climates have changed, driven primarily by concerns about health, safety and the environment. This is leading the industry to reduce or eliminate the use of these classes of compounds, at least in some countries.

Problem is, maintaining fluid performance and longevity in the presence of microorganisms is more challenging when some or all of these compounds are eliminated. However, research has shown that certain primary amino alcohols, combined with nonformaldehyde releasing biocides, approaches the performance of the problematic substances.

Regulatory Bad News

The demand for longer-lasting, better performing, lower maintenance metalworking fluids has never been greater. Reducing the frequency of tank-side maintenance and extending the time between shut-downs can reduce the cost of fluid ownership and increase productivity. At the same time, regulatory requirements, health and safety considerations and customer preferences are limiting the ingredients considered acceptable, particularly for MWFs intended for use in multiple regions.

An important multifunctional additive, boric acid, has been used as a cost-effective, biostatic corrosion inhibitor and pH stabilizer since the early 1980s. However, in 2010, the European Classification, Labeling and Packaging regulation took effect, which classified boric acid as toxic to reproduction, Category 1B. Therefore, boric acid was added to the European Unions candidate list of Substances of Very High Concern, and its use is limited to less than 5.5 percent by weight in MWF concentrates.

Although boric acid can be effective below this dosage, depending on the dilution rate, its classification as toxic to reproduction has caught the attention of the industry. As a result, it is possible that some fluid producers and end users may decide not to allow any amount of boric acid.

In addition, formaldehyde-condensate biocides, in particular triazine, have been the bactericides of choice for water-dilutable MWFs for several decades. Triazine and another major class of formaldehyde condensates, oxazolidines, provide cost-effective control of prominent gram negative bacteria, such as pseudomonas aeruginosa. These biocides are also relatively stable at the alkaline pH of most MWF concentrates.

Unfortunately, formaldehyde condensate biocides also have the potential to release formaldehyde under typical use conditions, and the International Agency for Research on Cancer and National Toxicology Program consider formaldehyde to be a known human carcinogen. In view of industry concerns with formaldehyde and, by association, formaldehyde-condensate biocides, a major biocide manufacturer, The Dow Chemical Co., has decided not to reregister a product used for many years in the MWF industry. The request for cancellation was acknowledged by the United States Environmental Protection Agency in July 2014.

Dicyclohexylamine (DCHA) has been used for several years to extend fluid life. However, although exempt from regulation in Germany under TRGS 611 due to its N-nitrosamine being considered noncarcinogenic, under the Globally Harmonized System in the European Union, DCHA is classified as H301 (toxic if swallowed), H311 (toxic in contact with skin) and H410 (very toxic to aquatic life with long lasting effects). Therefore, some MWF producers and users have decided not to use DCHA in their fluids.

Searching for Alternatives

In an effort to find suitable replacements for boric acid, triazine and DCHA, Angus investigated certain amino alcohols and registered biocides, with the goal of matching or exceeding the performance of the conventional substances in semisynthetic MWF formulations. Eighteen high-oil semisynthetic formulations, a typical fluid class in Western Europe, were prepared for the study.

The formulations were divided into four groups. Three control fluids contained 0, 2.5 and 5 percent boric acid with a triazine biocide. Five fluids contained a primary amine stream (PAS) and five different formaldehyde-free biocides. Another five fluids contained DCHA and the formaldehyde-free biocides. The final five fluids contained PAS, 3-amino-4-octanol (3A4O) and the formaldehyde-free biocides.

Microbiological testing consisted of a modified ASTM E2275 procedure, using a mixed bacterial/fungal inoculum isolated from spoiled MWFs. The fluids were sampled periodically until failure, defined as two consecutive weeks above 100,000 colony forming units (CFU) per milliliter bacteria or 1,000 CFU/ml fungi.

A significant change in the pH of a fluid can also indicate microbiological degradation. Most often, pH will decrease due to the generation of acidic biodegradation products, but pH can also increase due to alkaline biodegradation products. In this study, the pH of each fluid was measured initially and weekly thereafter during microbiological testing.

Cast iron corrosion can also indicate microbiological growth. In this study, a sample of fluid was tested periodically to determine corrosion control, using a modified ASTM D4627 procedure.

In addition, some fluid components, including certain amines and acids, can stain aluminum alloys. The higher the fluid alkalinity and pH, the greater the chance of staining. Aluminum staining properties were determined by immersing coupons in the fluids for 24 hours and rating on a scale of 0 to 5, with 0 indicating similar appearance to unexposed coupons, and 5 indicating dark or white stains on the majority of the surface.

Finally, residue left by MWFs on machines and parts can create cleaning difficulties and other problems. Ideally, residues should be soft and easily removed using water. In the study, each fluid was tested in duplicate and the amount of residue estimated visually on a scale of 0 (least) to 3 (most).

Microbiology

Bacterial results showed early failure for fluids with 0 to 2.5 percent boric acid. The boron-free fluid with PAS and benzisothiazolinone (BIT) resisted bacterial attack for an additional 4 to 5 weeks, and a similar fluid with PAS and butyl-BIT (BBIT) resisted bacteria for 13 weeks. The fluid with triazine and 5 percent boric acid, as well as those with PAS and the other formaldehyde-free biocides, resisted bacterial attack through 15 weeks.

Fungal testing showed that the fluids with 0 to 2.5 percent boric acid lost fungal control early, but increasing boric acid to 5 percent prevented growth through 15 weeks. All fluids containing PAS failed early, except that with BBIT, which resisted fungi until 9 weeks.

Fluids containing DCHA and BIT, a morpholine derivative (MDM) and phenoxyethanol (PhET) resisted bacterial degradation longer than the fluids with triazine and 0 to 2.5 percent boric acid, but not nearly as long as the fluid with 5 percent boric acid. Fungal resistance of the fluids containing DCHA was better than those with triazine and 0 to 2.5 percent boric acid, but only the fluid with DCHA and BBIT approached the performance of the fluid with 5 percent boric acid.

Fluids containing 3A4O and formaldehyde-free biocides resisted bacterial growth better than those with triazine and 0 to 2.5 percent boric acid; however, only the fluid with 3A4O and BIT resisted bacterial attack as well as that with 5 percent boric acid. All fluids with 3A4O resisted fungal attack better than those with triazine and 0 to 2.5 percent boric acid. Those with 3A4O and BIT or PhET came closest to matching the performance of the fluid with 5 percent boric acid.

pH Stability

Changes in fluid pH can indicate microbiological degradation and loss of performance. The pH data taken during microbial challenge tests were generally consistent with the microbiological results.

Fluids with PAS had better pH stability than might have been predicted from the microbial data. Control fluids with triazine and 0 to 2.5 percent boric acid lost pH control much faster than the fluid with 5 percent boric acid. Most fluids with PAS and formaldehyde-free biocides had similar pH stability to the 5 percent boric acid fluid, except the one with BBIT, which lost control after 9 weeks.

Fluids with DCHA lost pH control relatively quickly, especially those with o-phenylphenol (OPP) and BIT. This could be partially due to migration of DCHA into the tramp oil or evaporative loss.

Fluids with 3A4O had better pH stability than those with DCHA, although the pH of the fluid with BIT decreased rapidly after 9 weeks. Previous studies have shown that 3A4O partitions much less into tramp oil than DCHA, which could be one reason for its improved pH stability and better microbiological control.

Corrosion & Staining

Cast iron corrosion testing of the fluids with triazine and 0 to 2.5 percent boric acid was stopped after 4 weeks due to poor microbial control, so corrosion data are not available beyond that point. The control fluid with 5 percent boric acid provided good corrosion control through 13 weeks. Fluids with PAS and the formaldehyde-free biocides lost corrosion control after 8 weeks, except the fluid with BIT, which lost control at 4 weeks.

Most of the fluids with DCHA lost corrosion control at 8 to 13 weeks, although the fluid with MDM performed a little better than the others. The fluids with 3A4O controlled corrosion through 8 weeks, and those with OPP and BBIT performed nearly as well as the 5 percent boric acid fluid through 13 weeks. The 5 percent boric acid fluid performed best after 16 weeks.

The control fluids discolored the aluminum alloys, which could be due to the combined effects of the buffering and neutralization additives, triazine and their associated salts. There is no evidence of boric acid having a negative impact on aluminum alloys.

Rinsability

Residue rinsability tests showed that the fluids with 2.5 to 5 percent boric acid had better rinsability than the fluid without boric acid. This was surprising because reaction products of boric acid (amine salts, etc.) can produce hard, sticky residues.

Most of the other fluids rinsed off easily, with the exception of those containing PAS and BIT or MDM. Apparently, the biocide influenced rinsability more than PAS, since the other PAS-containing fluids rinsed easily.

It is believed that the high oil content of the semisynthetic formulations had a beneficial effect on residue removal, which could explain why residues of the fluids with boric acid rinsed off easily. Unfortunately, this does not explain why the residues of some fluids without boric acid were more difficult to remove.

Conclusions

Matching the performance of 2.5 percent boric acid plus triazine is easy using DCHA or a primary amine combination, together with several formaldehyde-free biocides. However, matching the overall performance of 5 percent boric acid is challenging, although fluids with PAS, 3A4O and BIT came closest to doing so.

Several fluids containing PAS had similar pH stability as 5 percent boric acid, and fluids with PAS/3A4O and several biocides outperformed comparable fluids with DCHA. Finally, triazine-containing fluids stained some or all aluminum alloys, but none of the alternative fluids did, even at a dilution pH of 9.3 to 9.8.

In the final analysis, cost, performance and environmental/safety requirements will dictate whether the alternatives are viable for MWF formulators and users requiring fluids without boric acid, secondary amines and formaldehyde condensate biocides.

Patrick Brutto is Metalworking Fluids Scientist for Angus Chemical Co., Buffalo Grove, Illinois, United States. Contact him at pebrutto@angus.com.