Cleaning Issues

Successful snow cleaning, as for any cleaning process, requires attention to process parameters and cleaning methods. The major factors to consider include recontamination risks, moisture condensation, CO2 purity, static charge, cleaning methods, and others. By addressing these and other issues in a test program, CO2 snow cleaning becomes routine and relatively simple to scale a bench top cleaning system shown earlier to systems involving automation and process control environments.

Recontamination Issues:

Recontamination risks of any cleaning process must be investigated before implementation. There are four major sources for recontamination - particulate redeposition, carbon dioxide impurities, the cleaning equipment and fixtures, and the cleaning process.

Particle Redeposition - Once a particle is removed from the surface, the set up and system design must insure that it is swept away and cannot redeposit on the cleaned surface. Generally, high flows, or laminar flow, along with nitrogen purges, are used. These flow are passed over the surface while cleaning and assist in particle removal away from the surface. Filters and other options are needed. Make sure you always clean "into dirtier" areas and never spray towards areas already cleaned. For table top systems, just insure the stream can flow away from teh cleaning area without striking other objects.

Cleaning Equipment - Recontamination from the cleaning system itself, the fixtures and materials used to hold the sample, nozzles, and environmental chamber, can occur unless attention is paid to material selection, and surface finishes. Material selection is important. The material choice and surface finish must be specified. Cleaning fixtures and other items within the cleaning region is suggested.

Cleaning Method - The cleaning process, after removing particles and organic residues, must not allows for these particles to land back on the part that was cleaned. In other words, a particle removed from the surface by CO2 can not land back on the surface of the clean part. To insure this in critical cleaning operations, several steps should be taken, including:

• the use of HEPA filters for the recirculating nitrogen -any liberated particle will be removed by the filters
• all components will be cleaned with CO2 snow during installation and as part of the suggested operating procedure
• a filtered nitrogen flow will always blow from the clean area to the dirty area - only filtered nitrogen will be allowed to blow over the cleaned parts
• the cleaning direction will be from a clean area to a dirty area
• no uncleaned parts will ever be allowed to pass over cleaned areas of the sample
• and and many more...

CO2 Purity - This was investigated separately in two studies. First, Whitlock{1989} performed a series of tests in a class 1 cleanroom on several new wafers. Initial wafers particle populations were analyzed by a particle counter. The nozzle was held over the wafer center with the CO2 Snow stream on for 10 seconds. Then the particle population was counted again. The test used a high purity CO2 (SFC-CO2) source. The particle population increased from an average of slightly less than 16 particles before cleaning to over 80 particles after cleaning. This increase corresponds to about 1 submicorn particle per square centimeter. EDS failed to identify any element heavier than sodium suggesting that these added particles are hydrocarbon-based. High temperature baking in air reduced the particle population also suggesting a hydrocarbon-rich contaminant. The sources of these hydrocarbon particles have never been conclusively identified but source is believed twofold, hydrocarbon impurities in carbon dioxide or from the cleanroom ambient air. Improved purification methods and operating methods since then will yield better results.

Sherman{1994} studied the effects of different CO2 grades; sources included welding grade, food grade, instrument (Coleman) grade, and supercritical fluid chromatography grade (SFC). The cylinders were all gas fed except for one listed as "liquid". We analyzed new Si wafers by XPS, CO2 snow cleaned this region, and then analyzed the same area again. For comparison, we define a cleaning ratio, the final hydrocarbon concentration divided by the initial hydrocarbon concentration. Values of this ratio less than 1 imply cleaning while values greater than 1 imply a recontamination of the clean surface. Results are listed below.

We see that the values for effective cleaning (ratios less than one) are only for the SFC grades. These results imply that the SFC grades are acceptable for cleaning "clean" samples while the other grades may not be acceptable for critical applications. Nevertheless, the four grades that led to an increase in the surface hydrocarbon concentration would still remove visible carbon contamination and particles from surfaces as seen in images on the cleaning examples page (welding gas was used) and satisfy many cleaning applications outside of the semiconductor industry, precision optics or other special situations. The lesser grades may also satisfy the cleaning needs in these industries but testing is recommended.

Relative Cleaning Effectiveness

CO2 Source		Cleaning Effectiveness
SFC-1		0.36
SFC-2		0.69
SFC-3		0.58
Medical		1.53
Welding - liquid		1.65
Welding - gas		1.25
Instrument		1.18

The liquid fed welding source led to a greater contamination than the gas fed welding source. This suggests that cleaning with a liquid fed cylinder causes greater recontamination than a gas fed cylinder. Since liquid CO2 is such an excellent solvent, it is expected that the hydrocarbon contamination in a cylinder is concentrated in the liquid. Then the amount of contamination in a given amount of liquid CO2 is always greater than in the same amount of gas. On site condensers are available and claim to reduce the heavy hydrocarbon contamination by a factor of 10 to a 100 or more. You should note that the industrial gas companies list total hydrocarbon content {TOC} on many grades, but the TOC only refers to compounds with up to six carbons. All heavier hydrocarbons, such as oils, are ignored in the TOC value. Instead, these oils are reported as a percentage weight residue and are rarely reported for commercial grades except for the SFC/SFE grades.

Methods:

The main aspect of methods is to have a proper cleaning setup - a cylinder properly secured or external source with proper safety shut off connections. All fittings should be checked to make sure they are tight and no leaks are in the lines. Your sample should be secured, heated as necessary to minimize moisture condensation on the surface. As discussed in our direction (Email Us for a copy), the actual setup and operation is simple and straightforward, but proper cleaning requires attentions to the issues discussed above and in the following. Cleaning should be systematic, from a region already cleaned to regions awaiting cleaning. Never have the spray aimed at a dirty surface so that the contamination removed can land on a cleaned area. Never use a "here and there" approach to cleaning. Complete, systematic cleaning methods are needed.

Moisture Condensation:

The cold CO2 snow stream lowers the surface temperature and moisture can condense. Generally, moisture condensation does not interfere with simple cleaning in some applications, unless the moisture "freezes" or stays on the surface too long. Cleaning setups usually have a method to minimize moisture condensation - the easiest method is use a hot plate or heat lamp as part of the setup. Usual set point for the hot plate set about 35C, or a hot air gun, but note higher temperature may be necessary. The samples can be held on the hot plate by vacuum chucks or other special rigs. Generally, for samples with good thermal conductivity, moisture condensation does not occur. For samples with poor thermal conductivity, such as thick glass, an overhead hot air gun or lamp can be used. Other good choices for moisture control are dry boxes, enclosed hoods or environmental chambers that are purged or heated - all have been successfully incorporated. In critical cleaning situations, i.e. submicron particle removal from wafers or optics, moisture condensation must not occur or removal of these small particles will be hindered. Cleaning must be done in dry chambers. Applied Surface Technologies has partnered with other companies to supply proper environmental chambers and process automation.

Filtration:

Critical cleaning applications require filtering the CO2. The cylinders, hoses and all the equipment have particles and the flowing gas or liquid will transport these particles out of the nozzle and on to the sample. Generally, the cleaning process will remove these particles, but there are risks that some will stay or land nearby. For critical cleaning applications, it is imperative to have in-line filters at the point-of-use; this means placing the filter right before the nozzle. We offer three different filters as options on our units - see equipment page and price list.

Static Charge:

There is a potential for build up of static charge on surfaces during cleaning. This is caused by the ionization of a flowing CO2 gas. Obviously, this static charge buildup is not a problem for metal samples. From our experience, if the sample is grounded, static charge is not a problem. Charging is usually worse for glass samples or for electrically isolated parts on complex structures. For these cases, commercially available positive ionization sources can be obtained for charge compensation. We recommend static control devices be used in critical cleaning applications.

Feed Pressure:

The feed pressure is important to maintain in these snow spray systems. If the feed pressure decreases, stream velocity decreases and organic removal efficiency decreases or stops. With liquid fed sources, pressure reductions will lead to a higher snow percentage having a lower velocity; with gas fed, no snow output can result. The desire for reduced pressure feeds are the ease in cleaning smaller or delicate parts and minor changes to commercial units can implement this mode.

Economics:

Cost per part cleaned can range from much less than a cent to over $1.00. Factors that can influence cost are cleaning time, source selection purity, nozzle selection and other factors. Cost estimates are based upon cylinder sources and significant cost reductions can be achieved through the use of bulk delivered CO2. Equipment costs range from about $1,600.00 (shown earlier in equipment) to $5,000 for the units shown on the equipment page while automated units with environmental chambers cost much more.

Damage Risks: Carbon Dioxide Snow Cleaning can damage samples if proper care is not exercised. First and foremost, since CO2 cleaning uses a high velocity stream, the sample MUST be supported, held or fixed. Further, coatings that are poorly adhered can be blown off. As a rule, if a coating is properly deposited and bonded, it should survive cleaning. Cleaning loose fibers may pose a challenge though braids have been done. One customer even cleaned a Si based MEMS devices. Thermally sensitive material also poses challenges. Please note, most all metals, ceramics, optics, polymers, etc, can be cleaned.