Click Image to Enlarge
TURI cleaning tests with Binary Azeotropes.
Binary water-based Azeotropes clean as well as Propyl Bromide.
Non-Azeotropic behavior of Acetone and water.
Azeotropic behavior of Isopropanol and water.
Vapor Degreasing with Binary Azeotropes
Process Cleaning, John B. Durkee, PhD, PE and Jason Marshall
For users who perform surface vapor degreasing, solvent cleaning has undergone a profound change since about the early 1990s. The motive forces have been those categorized by the administrative name SHE (safety, health and environmental). The change produced by those forces has been to limit applications for solvent cleaning. This is because of the absence of solvents which meet practical, economic and SHE requirements.
In the 1980s, vapor degreasing applications were shared among five cleaning solvents: CFC- 113 and 1,1,1-trichloroethane (Class 1 Ozone Depleting Chemicals), methylene chloride (ACGIH TLV exposure limit 50 ppm), trichloroethylene (ACGIH TLV exposure limit 50 ppm) and perchloroethylene (ACGIH T LV exposure limit 25 ppm).
Use of these solvents has been somewhat replaced by other halogenated solvents such as HCFC-141b and HCFC-225ca/cb (Class II Ozone Depleting Chemical), and npropyl bromide (ACGIH TLV 10 ppm).
Additional replacement solvents prompt fewer SHE concerns. But application of these solvents is limited because their solubility parameters resemble those of paraffinic solvents which don’t possess significant hydrogen-bonding and polar intermolecular forces. These replacement solvents include HFC-4310-mee (200 [CEL] 3 ppm), HFE 7100/7200 (750 [AIHA] / 200 [CEL] ppm) and OS-10 (CEL 200 ppm).1
Other non-solvent cleaning technologies, such as aqueous, are replacement choices for the above solvents. But their use does not offer the chief advantage of the vapor degreasing process, which is that the last fluid contact is with pristine cleaning agent.
AzeotropesAzeotropes are blends of solvents. However, not all solvents form azeotropes. The number of components can be two, three, four or more. But for vapor degreasing work, more than two are probably impractical. Azeotropes add opportunity and complexity to vapor degreasing operations.
An azeotrope is characterized by having the same composition in the liquid as in the vapor when the liquid is boiling. In this way, an azeotrope (with any number of components) acts as if it were a single component. For a single component, the composition in the boiling liquid and in the produced vapor is 100 percent of the solvent material.
Vapor-liquid equilibrium data are plotted in Figure 1 for the case of water and isopropanol. The composition of the vapor and liquid are equal at approximately 70 mole percent water and a boiling temperature of approximately 80.1 EC. That is the azeotropic composition. As plotted in Figure 2 or acetone and water, there is no azeotrope. There is no minimum boiling temperature or equivalence between vapor and liquid compositions.
Binary azeotropes may be thought of as new solvents because they don’t require development of new molecules, synthesis routes, manufacturing capability or, generally, regulatory acceptance. Also, they’re seldom used for vapor degreasing. In past publications,2 the approach of replacing single-component solvents with four binary azeotropes has been explored. Hansen Solubility Parameters and physical properties were the targets of the single component solvent which were to be matched by the binary azeotrope.
Description of this ResearchOne solvent was chosen as a “target” to see if selected binary azeotropes could produce the same laboratory cleaning performance as demonstrated by the “target” solvent in similar laboratory cleaning experiments. Because of the high level of interest it generates, the chosen “target” solvent was n-propyl bromide.
Four soils demonstrated to have been cleaned by n-propyl bromide were chosen. That past work was completed and published by an unbiased non-commercial source of cleaning data developed by the University of Massachusetts Lowell’s Toxics Use Reduction Institute (TURI).3 The soils were:
• Castrol Quench G Oil—By the MSDS, it was identified as CASs # 64742-55-8, 64742-65-0, 8052-42-4.
• Cargill Canola Oil—By the MSDS, it was identified as CAS # 120962-03-0.
• CP Hall Co. Plasthall Eso Oil—By the MSDS, it was identified as CAS #8013-07-8.
• Soltex Polybutene 32 Coating—By the MSDS, it was identified as CAS #9003-29-6.
Four binary azeotropes were chosen for these cleaning tests from a database containing thousands of binary azeotropes. Selection was based on the components generating a substantially lower level of concern about SHE issues than the “target” solvent and their expected level of cleaning performance. The four binary azeotropes were:4
1. A mixture of 16.5 wt% water in t-butyl acetate, with a boiling point of 87.4 EC. Both components are VOC-exempt in the U. S. Exposure limits are “not listed” and 150 ppm.
2. A mixture of 5.0 wt% water in methyl acetate, with an expected boiling point of 56.5 EC. Both components are VOC-exempt in the US. Exposure limits are “not listed” and 200 ppm.
3. A mixture of 12.9 wt% water in n-heptane, with a boiling point of 79.4 EC. Exposure limits are “not listed” and 400 ppm.
4. A mixture of 51 wt% water in propylene glycol methyl ether (PGME), with a boiling point of 97.5 EC. Exposure limits are “not listed” and 100 ppm.
Experimental ProcedureThe four supplied solvents were mixed with DI water in 600 ml beakers to obtain binary azeotropes. Each was heated to its boiling point on a hot plate. Twelve pre-weighed aluminum coupons were coated with each of the four soils with a hand-held swab. Coupons were weighed again to determine the amount of oil applied. Three coupons were cleaned in each azeotrope for five minutes at the boiling point, rinsed for 15 seconds in 49 EC tap water and dried for 30 seconds using compressed air at room temperature. Coupons were weighed a third time to determine the amount of oil remaining and efficiencies were calculated and recorded.
DiscussionOne unusual experimental outcome was that the methyl acetate azeotrope performed somewhat better than anticipated via analysis using Hansen Solubility Parameters. There is no known cause of this discrepancy. U.S. patent applications have been filed covering vapor degreasing with binary azeotropes.
ConclusionsThese results are an initial validation of the selection methods for binary azeotropes described in reference 1.
The two binary water-based azeotropes (t-butyl acetate and methyl acetate) in which both components are VOC-exempt in the U.S. offer good potential to replace n-propyl bromide in cleaning via vapor degreasing technology.
Though the two binary water-based azeotropes (t-butyl acetate and heptane) were expected to clean the soils best, the n-propyl bromide also cleaned the soils well.
The heptane-water binary azeotrope is a low-cost way of getting both polar and hydrogen-bonding intermolecular forces into a non-polar paraffin solvent.
The azeotrope which was expected to be less effective (PGME) was somewhat less effective. But, it can be useful when matched to other soil materials.
Future WorkAdditional laboratory tests will be conducted using available resources.
John Durkee is the author of the book Management of Industrial Cleaning Technology and Processes, published by Elsevier (ISBN 0-0804-48887). He is an independent consultant specializing in critical and metal cleaning for contamination control. He can be reached at (830) 238-7610.
In his role as the University of Massachusetts Lowell Toxics Use Reduction Institute’s (TURI) Manager of Lab o rat o ry Testing, Jason Marshall oversees the day-to-day operation of the laboratory, helping Massachusetts companies evaluate the performance of cleaning chemistries and equipment. Recent projects include a TURA program-wide effo rt to promote adoption of alternatives to trichloroethylene (TCE) for businesses in Massachusetts. He can be reached at (978) 934- 3133.
1. Durkee, J.B., “Use of Hansen Solubility Parameters to Identify Cleaning Applications for ‘Designer’ Solvents,” Chapter 11, Hansen Solubility Parameters: A User’s Handbook, Second Edition, CRC, May 7, 2007, ISBN 0849372488.
2. Durkee, J.B., “New Solvents for Old,” Presented at the Tenth Intern ational Symposium on Pa rt i cles on Surfaces: Detection, Adhesion and Removal, Toronto, Canada, June 19-
3. The data can be found at www.cleanersolutions.org.
4. All azeotropic data can be found within Gmehling, J., Azeotropic Data, Part I, Wiley, ISBN 3-527-30833-4.