Surface cleaning using a carbon dioxide snow jet incorporates a high-velocity stream of small dry ice particles and gas which is directed towards a contaminated surface. The resulting interactions between the snow particles and surface contamination lead to particulate and hydrocarbon removal. Past work has demonstrated removal of micron and submicron particles and also hydrocarbon and silicone-based stains from many different samples including Si, GaAs, InP wafers, metals, ceramics, UHV components, glass, optical components, and electronic devices. Several typical applications and examples are discussed along with the experimental parameters for effective carbon dioxide snow cleaning. The above results supports previous findings that carbon dioxide snow cleaning is nondestructive, nonabrasive, and residue-free, thus making this cleaning procedure acceptable for many critical cleaning applications.

No abstract

Analyzing carbon in an electron microscope by EDS is difficult, since the pollution by hydrocarbons is responsible for a fast growing C peak when the beam is positioned in spot mode. Hydrocarbons from the chamber surfaces, vacuum pumps and sample surface migrate, react with the electron beam and form a black spot rich in carbon where the analysis is conducted. The analytical difficulty extends to N, which peak is partially overlapped by the growing adjacent C peak and strongly absorbed by the carbon layer that builds up at the surface.The Evactron Anti-Contaminator (A-C) removes Hydrocarbons from the SEM vacuum and from the sample surface [1]. The Evactron A-C is a small RF plasma device mounted on a chamber port and produces Oxygen radicals to oxidize off the hydrocarbons. Although a full clean-up of a dirty microscope chamber use for years cannot be expected in a few minutes, the following results show that the Evactron A-C makes low level carbon analysis possible just after installation.To test the improvement on removing C pollution, a pure Cu polished sample has been used. The microscope was a Gemini 982, fitted with a Noran thin window EDS. The sample had been prepared using the best standard procedure recognized by the operator to give the smallest pollution: polishing, ultrasonic cleaning and final cleaning with pure ethanol. Spectra have been acquired at 15 kV and the EDS system used to measure net intensity peaks on the elements found: C and Cu La. The spot was left static for 15 minutes, and spectra were run for 50 s. The first spectrum was acquired immediately and then three other after 3, 6 and 15 minutes. Results:The first series of spectra has been acquired before Evactron A-C cleaning. The spectrum shows immediately a C peak, and this carbon peak quickly grows up from a relative C/ CuLa intensity ratio of 0.76 % to 5.06 % after 15 minutes of static beam, as it can be seen on the spectrum in Figure 1. The Evactron A-C was then run 4 minutes to remove the hydrocarbons. The first spectrum acquired after cleaning still showed a small carbon peak ratio of 0.61 % that grew up to 1.39 % after 15 minutes. This improvement was impressive but not enough: you cannot expect to clean a sample and a large chamber that had been used for several years in 4 minutes. So a second 4 minutes cleaning was done and a third one two hours later. This gave time to the trapped hydrocarbons to diffuse back and be converted to H 2 O, CO and CO 2 by oxygen radicals from the Evactron A-C. After the third run, the EDS system did not find any carbon in the spectrum for the spectra acquired immediately and after 3 minutes. The spectrum acquired after 6 minutes and 15 minutes showed a small increase of the C intensity that allowed the software to detect C.

Critical Dimension measurements for process control in semiconductor lithography are routinely made using Scanning Electron Microscopy (CD SEM). In many situations, organic contamination of the CD SEM chamber cannot be prevented due to the outgassing of hydrocarbons present in the photoresist films used to define device structures. In other cases advantageous hydrocarbons are deposited from the room air before the wafer is brought into the machine, or there are residual deposits left over from manufacturing of the tool. The interaction of the primary beam with these hydrocarbons, resident in the SEM chamber, results in a deposition of a hydrocarbon film, whose thickness is dependant upon the total dose provided to the structure of interest. This deposited film not only reduces the available image contrast but also physically changes the size of the measured feature. In extreme cases, such changes have been reported to be as large as several nanometers 1 2 during a typical measurement sequence. Such a value approaches the entire metrology error budget for the most advanced processes.

.As the need for higher throughput of samples in Scanning Electron Microscopes (SEMs) and Focused Ion Beam (FIB) systems increases, so does the necessity to shorten pump down times between loading samples. Industry demands SEM/FIB systems to be operational 24/7 and ideally maintained in pristine condition with uncompromised image quality. Frequent venting of the SEMs or FIBs to load samples introduces moisture and contamination into the vacuum chamber, leading to much longer pump down times and decreased efficiency.The new Evactron Turbo Plasma™ De-Contaminators remove hydrocarbon (HC) contamination [1] from SEMs, FIBs and other analytical tools using a gentle, down-stream plasma afterglow process. At turbo pump pressures, Evactron cleaning becomes faster and spreads throughout the chamber. This is due to longer mean-free-paths that cause less recombination of oxygen radicals in the required three body collisions and decreased scattering to chamber walls [1]. In most cases, short plasma cleaning cycles are sufficient to remove contamination and significantly shorten pump down time, allowing for high throughput of sample processing and analysis.To demonstrate the effect of Evactron Turbo Plasma Cleaning on pump down time and HC contamination removal, the study was done using an EP model Evactron Plasma DeContaminator on a large, highly oil-contaminated 50 L vacuum chamber equipped with a 450 L/sec turbo molecular pump, 14 CFM scroll pump, MKS 972B dual range pressure gauge and a residual gas analyzer (RGA). Evactron Cleaning was done at 20 Watts for 30 minutes. RGA spectra ( Fig. 1 and 2) and pump down curves (Fig. 3) were obtained before and after contamination and plasma cleaning in order to demonstrate the effects of Evactron Turbo Plasma cleaning on pump down time and hydrocarbon removal. The chamber is considered to be in pristine condition when the HC peaks on the RGA spectrum were less than a partial pressure of 2 X 1 -10 Torr.The length of the pump down time shows dependence on hydrocarbon contamination levels [2] in SEMs and FIBs. Therefore, this time could be used as an indicator of the cleanliness of the vacuum system. The data shows that Evactron plasma cleaners significantly reduce both the pump down time of the SEMs and FIBs as well as hydrocarbon contamination, and thus help increase sample processing throughput without compromising the quality of analysis.Typically, vacuum chambers can be cleaned with the Turbo Plasma Cleaning process at turbo molecular pressures of 10 -2 to 10 -3 Torr with typical cleaning times of 2 -10 minutes to maintain pristine conditions, and returning to typical operating pressures in < 20 minutes. Current users of Evactron Plasma Cleaners report significant reduction in pump down time after using Evactron De-Contaminators as well as easier maintenance of the pristine state of cleanliness of their SEMs and FIBs. A more detailed analysis and data from multiple experiments will be presented.