James Holton 12/5/06 Thought y'all might get a kick out of this. Perhaps even find it useful someday... I wrote this email to Tom A. when Gary Krebs suggested we "address" heavy atom safety in the PX MOU renewal. ------------------------------------------------------------ Years ago, Rene Ricks from LBL chemical safety evaluated the use of heavy atoms at PX beamlines determined that there is no risk involved. All of the PX sample are orders of magnitude below any trigger levels. Some examples: The most common heavy atom derivative we use is selenium. There are 5x more Se samples than all other kinds of heavy atoms combined (based on the wavelength used for data collection). The US RDA for selenium in adult men an pregnant women is 55-60 ug/day. We go through 50-100 samples per day that contain ~50 mM Se (one Se atom per 100 amino acid residues at 50% solvent) and are rarely larger than 100um3 or 1 nL in volume. Assuming all samples were 100um3 and 50 mM and we screen 100/day, then eating all of the samples for all 8 PX beamlines would give you 3 ug/day of Se (so I guess I should be taking a selenium supplement). The "most poisonous metal" (according to some sources) is plutonium. I don't think too many of our users have access to it. The "second most poisonous metal" is mercury, and Hg derivatives are common. If we consider the case where all PX samples were soaked in mercury (say 50 mM bound), then you get 8 ug/day. There is no RDA for mercury. However, a pizza delivered to the beamline with 8 ounces of anchovies on it would also contain ~8 ug of mercury. Anchovies average 0.043 ppm Hg and can be as high as 0.34 ppm Hg (according to the FDA's website). If you want to consider the protein itself as poisonous, then you need to think about the most toxic substances there are: tetanus toxin, botulinal neurotoxin and shigella toxin all have LD50s around 1 ng/kg in humans (according to about.com). A 1 nL crystal contains about 500 ng of protein, so it would probably not be a good idea to eat a crystal of tetanus toxin (assuming they even exist). Bear in mind that anything this toxic is BSL3. Anyone bringing such materials out of containment I'm pretty sure is breaking a few laws before they get to the LBL gate. So, I don't think we have to worry about the samples themselves. The only potentially worrisome situation would be people bringing bottles of poisonous things with them to the beamline for doing on-site heavy-atom soaks. This is not done very often because it takes a long time and has a high risk of ruining the samples (this can be connected with the safety envelope). Still, 1 mL tube of 1M HgCl is something I would rather not have around. If I happened to drink the whole tube, I would have as much Hg in me as a bad anchovie. A potentially lethal dose would be drinking 10 mL of 1M HgCl. All that I think we can say about tubes of Hg or any other hazardous substance is that we ask our users to fill out their ESS form properly and that they inform us when the are planning to bring them so that the beamline scientist can supervise the soaks and make sure the users pack out everything that they pack in. We do not pat down users when they come to the beamline to check for them, but we ask. Do you think we need something stronger than that