CAN YOU IMAGE A SNOWFLAKE WITH AN SEM? CERTAINLY! William P. Wergin and Eric F. Erbe Electron Microscopy Laboratory, USDA-ARS, Beltsville, MD 20705 USA Installation of the new Oxford CT 1500 HF Cryosystem on our Hitachi S-4100 field emission (FE) SEM had just been completed and the enthusiasm in the laboratory for testing the new acquisition was running high. We scurried around the lab looking for insects, spiders or some other unsuspecting critter that we could quickly use to check out the system. Unfortunately, the pickings were slim. This was December and the bugs that hadn't been killed by the monthly visit from the exterminator were evidently lying dormant in some obscure niche. Our alternative source of specimens for testing new equipment and procedures was the so called "flower bed", a strip of unmowed turf that ran along the building that houses our laboratory. Although past attempts were made to cultivate this area, a lack of commitment and effort combined with competition from Mother Nature had turned this strip into, what we affectionately called, our "nature garden"; it may have lacked in esthetics but had always been a rich source of plant material that included such exotic plants as dandelion, thistle, pigweed etc. In desperation we opened the door and glanced along the side of the building hoping to spot at least one hardy plant huddled next to the foundation in an effort to avoid the winter weather. But alas, not only were no green plants in sight, but the bed was covered with snow, for the temperature had fallen to below freezing and snow had been falling for about an hour. The snow was beginning to accumulate and covered, not only the flower bed, but also the walk, parking lot and the cars. There would be no samples out here today unless you were interested in snow. SNOW! Why not? Would it be possible to capture a snowflake on a specimen holder, transfer it to the cold stage in the prechamber, coat and image it in the SEM? Not only would this exercise allow us to check out our new system, but we were suddenly overcome with the obsession to image a genuine snowflake with a field emission scanning electron microscope. Sure, we had seen ice crystals in the microscope before, water-ice was our most common contaminant, but a genuine snow flake crafted by Mother Nature should be a magnificent sight. We hastily devised a plan to capture our snowflake. A stub would be precooled to the outside temperature, which was below freezing and then snowflakes would be allowed to settle on the stub, which would be plunged into liquid nitrogen and transferred to the cold stage in the prechamber of the cryosystem. Because the temperature of the cold stage was -165 C, no detectible sublimation should occur. Next our snow flakes could be sputter coated and then transferred to the microscope for observation of Mother Nature's creation - that is unless sputter coating or the electron beam generated surface heating that could transform the snowflakes into microdroplets of water. It was worth a try! The first unforeseen problem involved having the snowflakes adhere to the specimen holder. The flakes appeared to bounce off the precooled holder and it became apparent that the occasional flakes that came to rest would never withstand the turbulence involved in the next step; namely, the plunge into liquid nitrogen. Would adhesive tapes solve our problem? With enthusiasm still running high, we quickly went back to the lab, covered a holder with our commonly used adhesive tape and ran back outside to precool the stub on the roof of my car, which had now become our outdoor lab bench. After what seemed to be an appropriate period for precooling, the lid was removed from the holder box and we anxiously awaited for our first snowflake to hit the stub. Again the flakes bounced off the holder. Becoming somewhat desperate we brushed some new fallen snow onto the stub and plunged it into liquid nitrogen. Failure again. The snowflakes simply would not stick to the adhesive but quickly disassociated and could be seen floating around in the bubbling nitrogen. Before our third attempt, the problem was given more careful thought. Tissue-Tek (a methyl cellulose solution) is an inert liquid that we use to mount specimens that adhesive tapes fail to accommodate. We put a thin layer of Tissue-Tek on the stub and precooled it to the outside ambient temperature. Fortunately, the Tissue-Tek remained liquid at - 3 C. Next the lid was removed and the snowflakes began to settle on the stub. Eureka! They did not bounce off, but rather gently settled on the surface of the liquid. After a few minutes the stub, containing an abundance of newly captured snowflakes, was plunged into liquid nitrogen and we proceeded to the laboratory. All went well from this point. The specimen holder was mounted on the transfer rod, evacuated, moved to the prechamber for coating and inserted onto the cold stage of the microscope. Our initial efforts to image the snowflakes was somewhat hindered by charging. However this problem was easily solved by returning the specimen holder to the prechamber and recoating. At this time the specimen holder was turned 90 degrees in an effort to coat the undersides of the delicate crystals and rods that protruded above the surface of the frozen Tissue-Tek. The results were amazing! Christmas had just passed, therefore we thought we knew how snowflakes should appear (after all hundreds of drawings had appeared on all those season's greeting cards). However, the snowflakes imaged at our first attempt, as well as those subsequently collected during the course of the winter season, provided us with an unexpected abundance of diverse images that included plates, stellar crystals, columns, needles and dendrites. This infinite variety of shapes is influenced by the atmospheric conditions during formation as well as the temperature and humidity levels that snowflakes are exposed to as they descend to earth. On a more serious note, a snowflake is actually an aggregation of two or more snow crystals. A snow crystal is a single frozen ice grain that generally results from a process know as nucleation in which atmospheric water vapor condenses on a solid particle or nucleus at temperatures below 00 C. Studies on the sizes and shapes of snow crystals are actually a legitimate endeavor because snow, which may occasionally cover up to 23% of the earth's land, supplies about one-third of the water that is used for irrigation and the growth of crops (Gray and Male, 1981). For this reason estimating the amount of water in winter snow pack is an extremely important forecast activity that attempts to predict the amount of water that may be available for the following growing season. Unfortunately these estimates can be easily confounded by the sizes and shapes of the snow crystals that comprise the snowpack. The shapes of snow crystals have been extensively studied, photographed and illustrated with the light microscope (Bentley and Humphreys, 1931; Nakaya, 1954). However light microscopic examinations are hampered by the difficulty of working with a frozen specimen, which is susceptible to sublimation and melting, and by the limiting resolution of the instrument. The low temperature SEM appears to be a viable technique for examining snow crystals at magnifications that far exceed the resolution of the light microscope. Furthermore, the ability to collect and store samples enables investigators to accumulate snow crystals from numerous locations or different time intervals so that detailed observations and comparisons can be done in a convenient and orderly manner. Unfortunately, a snow fall at Beltsville, MD is not a common event and the atmospheric conditions in our area do not favor the formation of the dendritic types of crystals. For these reasons we were not able to find and image the snow crystal that resembles that of the RMS logo. We will leave that task to cryo groups who work under more favorable conditions. But then if "no two snowflakes are alike" perhaps that feat will never be accomplished. References: 1. Gray, D.M. and Male, D.H. (1981) Handbook of Snow (Pergamon Press) pp 60-152. 2. Bentley, W.A. and Humphreys, W.J. (1931) Snow Crystals (McGraw Hill) pp 1-227. 3. Nakaya, U. (1954) Snow Crystals (Harvard University Press) pp 7-77. Figure Legends 1. Stereo pair illustrating several types of snow crystals. The crystals consist of an array of hexagonal needles, a spherical cluster of short hexagonal columns, a simple hexagonal plate and a plate with branches. 20x. 2. Stereo pair of scanning electron micrographs illustrating snow crystals of the dendritic type that contain a coating of rime. Rime accumulates on the surface of a snow crystal as it falls through a cloud of supercooled microdroplets of water. 15x. 3. SEM of a single snow crystal plate having broad branches. The nucleation center can be seen in the middle of the crystal. 75x.