Technical Issues and
CO2-based Cleaning Systems

TECHNICAL FEASIBILITY OF SCCO2 CLEANING

COMPATIBILITY OF SCCO2 CLEANING WITH POLYMERIC MATERIALS

EFFECT OF MIXING ON SCCO2 CLEANING

Return to the first page of the SCCO2 Cleaning Technology Review

TECHNICAL FEASIBILITY OF SCCO2 CLEANING

SCCO2 cleaning has been tested on both a laboratory and pilot scale fairly extensively over the past decade. Published results of these tests have been primarily generated by United States federal government sources in the Department of Defense, Department of Energy, and Environmental Protection Agency, although a few published results that were located came from the private sector. Collectively, these tests have proven that SCCO2 has the potential to be a viable technical alternative for many cleaning applications found in the manufacturing environment — particularly in precision cleaning applications where the parts being cleaned have very small cracks or crevices that must be cleaned, as is the case with gyroscopes and instrument bearings. SCCO2 has proven to be especially suited to removing nonpolar, hydrophobic contaminants, which include many of the oils used in industrial and commercial applications. Some examples include:

• Researchers at Liquid Carbonic Supercritical of Allentown, Pennsylvania performed pilot tests with a SCCO2 unit on oil removal from rings, washers, and plates. A gas chromatagraph analysis was used to analyze oil left after cleaning. It was found that SCCO2 cleaning removed 97 to 99.95 percent of the oil, exceeding required cleaning specifications in all cases. Particulates were also successfully removed in tests on a computer disk drive’s components using SCCO2. In both tests, SCCO2 cleaned to levels below those typically obtained on the same parts using traditional solvent-based vapor degreasing. (Ref. 4)

• A research team at Los Alamos National Laboratory performed controlled laboratory-scale tests in 10-milliliter beakers and full-scale tests in a 60-liter cleaning vessel of SCCO2 cleaning on 10 different substrates, including aluminum, glass, copper, brass, stainless steel, and epoxy boards. The samples cleaned in these experiments were small coupons (test pieces) of each substrate. The removal of a wide variety of contaminants was tested under controlled conditions, including a variety of typically encountered commercial and industrial oils, and a simulated human skin oil mixture. Results showed that the cleaning efficiency varied, depending on the type of contaminant being removed more so than on the type of substrate being cleaned. Removal was most effective (typically 99+ percent) on low to moderately volatile compounds such as skin oil, mineral oil, and 3-IN-ONE (a registered trademark product), and least effective (less than 75 percent) on heavier contaminants such as silicone oil. The researchers concluded that SCCO2 cleaning is a good candidate for water-sensitive or high-temperature sensitive parts when aqueous cleaning may not be feasible. The investigators also conclude that SCCO2 cleaning is generally good for precision cleaning where the cleaning solution needs to be able to enter small crevices and cracks in a substrate. (Ref. 5)

• Tests of SCCO2 cleaning were performed for a manufacturer on metal discs contaminated with oil-type residues in a one-liter system at 82 degrees C and 2,000 psi. The tests resulted in surface carbon residue levels that were acceptable (below the specified "clean" level and comparable to what a trichloroethylene vapor degreaser could obtain). Testing also showed that test runs using "high-purity" CO2, rather than "bulk-grade" CO2, reduced carbon residue on parts by 50 percent. Interestingly, rotation (i.e. mixing) was not found to conclusively improve the cleaning efficiency in this test. The researchers hypothesized that the sample size may not have been large enough to determine the effect of mixing on cleaning efficiency. The manufacturer decided against pursuing SCCO2 cleaning due to the predicted long payback time (10 years) of the investment, and the high operating pressure of the system (which was viewed as a safety hazard). (Ref. 6)

• AT&T researchers tested a number of different cleaning alternatives to replace 1,1,1 trichloroethane vapor degreaser. SCCO2 was tested on aluminum parts contaminated with cutting oils and protective fluids. SCCO2 cleaning efficiency was more than 98 percent for several of the contaminants, however, one of the cutting oils and one of the protective fluids tested contained ionic species that were not effectively removed by SCCO2 cleaning alone. These contaminants were able to be effectively removed when a predip in a polar solvent was done and when isopropyl alcohol was added to the SCCO2 cleaning step. However, the requirement for additional additives and cosolvents to obtain acceptable cleaning results lead the researchers to eliminate SCCO2 from consideration. The process selected for the project was a counterflow vapor degreaser combined with an ultrasonic agitation unit using a cyclohexane/isopropanol azeotrope as the cleaning solution. (Ref. 7)

Some of the reasons for lack of demand in the private sector for SCCO2 cleaning were identified by researchers at Pacific Northwest National Laboratory during a SCCO2 cleaning market assessment completed in 1994. (Ref. 3) Potential barriers to acceptance identified include:

• Higher capital costs for SCCO2 systems relative to other cleaning technologies;
• Lack of awareness of SCCO2 cleaning technology;
• Substrate to be cleaned lacks compatibility with SCCO2 or with high pressures;
• A perception that SCCO2 cleaning does not remove particulates effectively;
• The requirement for a continuous process (SCCO2 cleaning is a batch process); and
• The existence of established aqueous-cleaning technologies to replace solvent vapor degreasers.

• Substrates with intricate geometry;
• Water- and/or heat-sensitive substrates; and
• Substrates that have drying times that are too long using aqueous cleaning.

COMPATIBILITY OF SCCO2 CLEANING WITH POLYMERIC MATERIALS

As discussed above, SCCO2 cleaning is compatible with a fairly wide range of substrates, including most metals and glass, and many plastics. A recent study by researchers at the University of Massachusetts's Toxics Use Reduction Institute investigated the interactions between supercritical and subcritical carbon dioxide and a number of different polymers to explore in detail the applicability of using the SCCO2 for the precision cleaning of polymers. This study was performed to address the concern that using SCCO2 to clean polymers may negatively affect the polymers since many polymers are known to undergo significant absorption of gases and vapors. The absorption of CO2 into polymers can alter a number of the properties of the polymer, including increasing the melting temperature or plasticizing the material, or other changes that affect the desired physical properties of a polymer product. (Ref. 8)

An initial series of tests were performed on nine crystalline polymers, including high-density polyethylene, polypropylene, Teflon, Mylar, and Kynar. A second series of tests were completed on 11 amorphous polymers, including Plexiglas, ABS, polyurethane, and PVC. Results of the tests showed that SCCO2 cleaning can, in most cases, be adjusted to have no detrimental effect on the crystalline polymers by optimizing parameters such as the treatment or decompression time, pressure, and/or temperature. In general, it was found that the amorphous polymers absorb CO2 to a greater extent than crystalline polymers and, therefore, experience a greater amount of plasticization, making these amorphous polymers less favorable for CO2 cleaning. The amorphous polymers showed visible bending and/or bubbling of the surface in many of the samples.

EFFECT OF MIXING ON SCCO2 CLEANING

Two studies located during the review that discuss the effect of mixing on SCCO2 cleaning resulted in somewhat different conclusions. As mentioned earlier, a manufacturer of metal discs was not able to determine that mixing had any effect on cleaning efficiency, but that a larger sample size may have given a more definitive result. (Ref. 6)

A more in-depth study specifically designed to examine the effect of fluid turbulence on SCCO2 cleaning concluded that mixing has an effect. The study was completed by researchers at Pacific Northwest National Laboratory. A SCCO2-cleaning chamber equipped with an impeller was used in the experiment. Tests were done using stainless steel coupons. One series of tests were done with the coupons contaminated with silicone oil, and the other with high-temperature oil. In both tests, the percent of contaminant removed from the coupons increased as the speed of the impeller was increased until a maximum percent removal was reached. Further increases in impeller speed did not have a significant effect on cleaning efficiency. The researchers recommended that agitation be used whenever possible for SCCO2 cleaning applications to help maximize cleaning efficiency, and that mixing rates can be optimized to minimize power costs. (Ref. 9)

Continue on to the economics page of the SCCO2 Cleaning Technology Review.
or
Return to the introduction of the SCCO2 Cleaning Technology Review