This is an attempt to document the current state of play at the new APS bending magnet beamline 8BM, but you should always compare this reality with whatever is on the ground at Argonne. This beamline is under development do things like monochromator and optical control may change.
Abbreviations: ANL = Argonne National Lab, the site; APS = Advanced Photon Source, the synchrotron; 8BM = the beamline we usually use.
If you find an error in any part of this document I would really appreciate being told about it, since synchrotron trips are difficult enough without me leading people astray.
See the APS User Admin site (e.g. New User Info ) for more information on all of this.
Training and Registration: new or expired users will need to (re-)register at the user administration office at least 30 days before going to APS and do training. This information is available on-line: user registration and user safety training (here's another page on user training requirements). APS does not allow you to take the training quiz before you arrive at the Facility, but reading the course materials will help you quite a bit since you can just "test out" once you get there. Valid BNL (yellow) GERT training cards seem to suffice for the radiological part of the training at APS. So if you have one then make sure you bring it.
Note that if you don't have a valid badge, you have to stop at the visitor center before the main gate to get a pass. It's possible to arrange with user administration to leave them at the security gate for you, but they are very unreliable with this. Consequently I recommend you arrive during normal visitor hours.
Safety approval form: you need a valid active Safety Approval form in order to work at the beamline. It takes 7 days for one to be approved - apply for one here ). I'm not certain how long they last (for the experiment? for the whole year?). I make a new one for each experiment.
Foreign Nationals who have not been to ANL before need to register well in advance (at least 30 days) because they may need to be "cleared" before access is permitted. This clearance must be renewed every two years, just like at Brookhaven. You must also make sure to turn up during regular office hours for training and registration and badge issuance. Failure to do this will probably result in you being sent away from ANL at the gate - they are, if anything, even more strict than Brookhaven. You must bring your passport and I-94 etc with you to Argonne when you come.
Gate Access Form: I do not think this is required for ANL/APS access. S valid APS user card seems to suffice. Call User Admin to check (630) 252-9090 if you are uncertain. If you are a new user, you must visit the Welcome Center before going through the main gate, and they will issue you a visitor badge. Make sure User Admin has you on the list there. We once couldn't take a lab member through the gate because she wasn't on the list (this was User Admin's fault, but either way, they would not let her in).
The ID requirements listed below are for Brookhaven, but it would be wise to follow them during your time at Argonne, since historically they've been even a little stricter than Brookhaven in terms of user access..
You need a Equipment Removal Form to remove items from MSKCC (ask Madellyn to do one) through the main door. You are unlikely to need the MSKCC loading dock, since I strongly advise against taking too much stuff with you on the plane. Ship it to APS via Fedex, packed well in sturdy boxes (for which you don't need the equipment removal form).
The best/only way to get your frozen crystals to Argonne is via a Dry Shipper of which there are two in the Pavletich Lab and two in the Goldberg Lab. These use immobilized "liquid" nitrogen to keep the samples cool (technically it's not liquid, and you should indicate to FedEx that there is no liquid nitrogen in there). They can be shipped by FedEx Air. However if you use the propane popsicle method the crystals cannot be shipped by air and must be shipped ground. Call up FedEx freight and talk to them directly about it. They seem to know what they are doing, as long as you utter the words "dry shipper" and are clear that there is no liquid nitrogen in the dewar. See also this APS technical data about the DOT approval for dry shippers to/from APS.
Adhere to the instructions for shipping hazardous and non-hazardous material to and from Argonne. Note that dry shippers are considered non-hazardous since the liquid nitrogen is immobilized.
You can carry your crystals with you on the plane, but please realize that there are much stricter restrictions in place for carry-on items these days on passenger airlines. Much of our traditional toolkit is forbidden as carry-on and may yet attract suspicion as check-in luggage. Perhaps better to FedEx all that stuff.
Once you've done the safety training at administration, you also
have to undergo the beamline-specific training and sector-specific
training. The sector-specific training took all of a minute last
time, nevertheless it is a requisite. Most of the time will be spent
in the beamline-specific training (equivalent to BLOSA at Brookhaven).
Posting an SAF I can't remember how we actually did this.
It's not done in the same was as at Brookhaven.
The goniostat at 8BM is a conventional horizontal axis, which accepts
our own Supper and Huber goniometer heads. There are a fairly nice
array of cameras, although the image did wobble last time I was there
due to optical effects of shooting through the nitrogen gas stream,
making precise centering kinda "fun". A computer controlling the
system allows you to center the crystal without having to translate
the goniometer sleds yourself - it's done electronically by moving the
base onto which the goniometer head mounts. It lets you translate
along the phi axis, and also back and forth along the beam (the
microscope looks from below, so translating up/down wouldn't move the
xtal on the screen). The computer will also allow you to rotate the
goniometer head through a variety of angles.
For really coarse adjustments I still find it useful to get fairly
close using a goniometer head key before using the computer to fine-center
the crystal.
If you have problems with crystal centering, be sure to check that
the crystal is mounted firmly on the head, the head is mounted firmly
on the phi axis, and lastly that the sleds on the goniometer head
are not loose - there are two small set screws that can be adjusted
and sometimes they work loose
Since you don't manually rotate the goniostat, you don't have to
manually lock it down afterwards, either. So there's no phi lock or
retention like at X9A.
Setup->Manual Control gives you the ability to move the detector:
minimum distance 200mm, maximum distance 1000mm. You should move to
1000mm when entering the hutch to adjust the crystal to avoid damaging
the detector. It's generally not worth moving the phi axis, since
you can do this in snapshot or within the hutch for centering.
Setup->Snapshot lets you take stills. You should select
a short exposure time (5-10 seconds) and something like a a 1
degree frame. Probably used hardware binned frames (unless you
have a burning need for unbinned). Collect a new dark current each
time you change exposure times, otherwise it's not really necessary
(dark current corrects for the electronic noise on the detector).
There's no need to dezinger at the times employed. Set filename,
directory, distanceSnapshot should open the ADXV display
software to view the image when it takes it (it will take a few
seconds after you finish exposure since it has to correct the image).
If the graphics window does not open, see below under Display...
Setup->runs is the main data collection table. Here you define
the standard parameters for data collection - the first frame, number
of frames, starting phi value for the first frame, time, size of
frame etc. Again, I recommend using hardware binned images unless you
really feel you need unbinned. You can collect inverse beam
images by clicking the anomalous button and entering the number of
frames you want in each wedge. I like to do 30-40 frame wedges.
If the "positive" wedge is (e.g.) run 1, then the inverse wedge will
be 101. Run 2 will generate 102 as the inverse, etc etc. Try and keep
different run numbers for different wavelengths. Typical exposure
times are 10-30 seconds. Unless you have parameterized your crystal
lifetime it's ill-advised to expose more than 30 seconds a frame,
especially if you are doing MAD data, when radiation damage is likely
to be a factor.
Stop->After this image stops the current data collection after the
current frame (Stop->immediately is disabled). For snapshots this makes no
difference, but I regularly use this for data collection runs.
Display-status shows CCD camera status - this is useful to have on
the screen to see what the CCD is doing. Display->Follow Images
displays each frame as it comes off the detector, exceptionally useful
during data collection.
If there's a Process menu item, ignore it.
You shouldn't need to use the Exit menu item, either.
A simple ADXV manual page is available for the Daresbury
beamline 14.1 which gives you some sort of idea, but please note
that details of the implementation will differ. I have not
been able to find an online generic ADXV manual.
De-zingering should rarely be needed at 8BM since the intensity is so high,
and if you are exposing for 120 seconds on that beamline your crystal won't
last for many frames.
Check on the beam monitor screen that you have beam - you've probably
forgotten to open the hutch shutter (above the central panel).
If you cannot see the backstop shadow this is probably what it is.
Check on the data collection monitor that you in fact have data.
Check that you're looking at the right frame. If you can see a
backstop shadow you have (some sort of) beam. If you can't see any
diffraction it's time to check crystal centering and whether the
goniometer head is firmly on the phi axis.
Crystal is not moving:
Unlikely to happen at 8BM, since you don't mess with the goniostat
directly. Could be trivially checked by going into the hutch and
using the computer to move the xtal.
Diffraction sucks:
so does your crystal. Perhaps it's not centered ?
Detector intransigence:
Quantum4 detectors have been known to hang, but we experienced no problems
whatsoever from the Quantum315 in use at 8BM. Nor did we learn how to reset
it.
You should assume that hazardous waste requirements at APS are
every bit as restrictive as BNL and operate our usual trash-in,
trash-out policy. If you're generating a lot of hazardous waste you
may want to make explicit arrangements for user shipping. If you're
shipping chemicals via FedEx it's not sufficient just to drop them in a
box and send them - you must guard against spills, leakage and breaks.
The Argonne
Guest House is by far the most convenient lodging on-site and
considerably higher quality than the dorms at Brookhaven. Reservations
are on (800) 632-8990 or (630) 739-6000. Fax is (630) 739-1000.
Email is argonne-guest-house@anl.gov. For larger groups a suite may
be useful and economic (4 bedrooms, 2 bathrooms, one central living
room area).
Here is an area
map and a site
map.
Generally we rent cars when at Argonne, but if you prefer to
use a limo, try United Limousine Services on 1-800-331-9037
or 1-630-969-3865. More details are to be had at the Argonne
transportation page.
This is a list
of local restaurants. I've never eaten at the
Argonne cafeteria and it's not at APS, but here's
the menu. Most of the off-site food opportunities are
north of the lab on or near Cass Ave in the other side of I-55.
Pizza and Taco Bell are nearby. A local map might come in
handy, or use MapsOnUs or MapQuest to work out the route.
There's actually no shortage of local fast-ish food if you're willing
to hunt around a little. User admin has a list they usually give
new users (and the URL is at the top of this paragraph).
Training, BLOSA, Radiation Badges
If you have never been to APS before, or if your current user
appointment has expired, you need to go through the process of
(re-)registration and training. See above for information and links
for that. You will not be allowed onto the experimental floor
until you have completed the training . Argonne, like Brookhaven,
no longer requires short-term users to have radiation badges.
Safety Approval Forms
You must have a valid Safety Approval Form
(SAF) before working on 8BM. This is the link to apply for
the approval (looks like you need a valid badge ID to fill this
form out - although perhaps you can temporarily create one).
Searching the Hutch
This is a fairly standard synchrotron ritual. In the case of Argonne
there are two search buttons inside the hutch. Since the hutches are
large, make sure you are the only one in the hutch before beginning
the search. Press both buttons in sequence (do not confuse them
with the panic button) and exit the hutch. Close the door by holding
down the close switch. Keep holding it down while the doors close.
The huge hutch doors are hydraulically controlled and close fairly
slowly. Once interlocked, you can walk around to main panel and open
the shutter after the internal lights have stopped flashing.
Mounting xtals
Make sure that you translate the detector to at least 1000mm from the
xtal (using Setup->Manual Control from the ADXV main panel) before
mounting the xtal to avoid detector damage. That CCD has already
been scratched once. Let's try and not be the second ones to do it ;)
Taking snapshots
The interface at 8BM is the standard ADXV software you've encountered
at X25 or CHESS or other sides with Quantum detectors. It's not the
same software as you're familiar with from X9A although it is somewhat
related to it. Note that the data collection software does not
control the monochromator (wavelength). This is the same situation
as at X9A, although at 8BM everything will eventually be under
control of the one computer.
Wavelength Changes and Fluorescence scans
The optics are sufficiently stable that users can do their own
wavelength changes, although this was imperfectly implemented when we
were there and once we lost the beam (the advice at the time was to
reset the table parameters to their standard values, although on the
second trip this didn't help). In cases where you completely lose the
beam the main recourse is to call the beamline staff. Although the
optics undergo an automatic optimization after wavelength shifts, it
pays to re-do these optimizations manually since some gain in beam
intensity can be acheived (this was less of a factor on our most
recent trip). Again, the hardware setup at 8BM varies over time so
check with beamline staff before assuming things have not changed).
Starting data collection
It's a standard ADXV interface. If you've used beamlines at CHESS
you'll know how to use this software. If not, then this isn't going
to be the definitive guide but I'll try to give an outline. The main
menu is a set of buttons (setup, stop, display, exit).
Stopping data collection
On the main ADXV menu there's a Stop->After This Image button, which will
(you guessed it), stop after the current image.
Long Data Collections
CCDs exhibit noise characteristics called "zingers" derived from cosmic
radiation and natural background radiation in the fiber-optic taper.
This only becomes an issue at exposures of more than 60 seconds and
I doubt you will be using that long an exposure at APS unless you
have an implausibly high-symmetry space group that requires minimal
degrees of data. De-zingering involves collecting the same data
twice and removing zingers by comparing the two images (whether or
not they are averaged depends on the software).
Processing data
HKL2000 and regular denzo/scalepack combination are available.
The Dells they currently use for data processing are painfully slow,
especially if you use unbinned frames. We have seen some data
corruption errors over the network on the hard drives - it's worth
taking the unusual step of checking the integrity of your backups
(e.g. bzcat filename.img.bz2 > /dev/null).
Backing Up Data
Backup of data on bright beamlines on APS is a special problem.
The new Quantum 315 frames are approximately 18 megabytes, so a
complete MAD dataset on a low symmetry space group will consume
many gigabytes. No tape device is going to keep up with the data
flow on an Argonne beamline anyway. Our current ploy, aside from
taking the laptop, is to use a relatively huge Firewire drive
to do temporary backups on. Processing backups should probably
be duplicated on the laptop, as far as possible. If you're really
unlucky it might be possible to duplicate the data too ;) I recommend
using bzip2 over gzip for data frames since it seems to acheive a
higher level of compression. I DO NOT RECOMMEND USING UNBINNED
FRAMES SINCE THIS CREATES A FAIRLY SIGNIFICANT BACKUP NIGHTMARE.
Nikola's tests suggest that unbinned is slightly better than binned,
but my tests indicated that hardware binned gave a 30% greater I/sigI
and 1.5% lower R-symm than unbinned frames. Software binned frames
were the worst of all. Test it for yourself ! Each unbinned frame
is 4 times the size of a binned one: 74 Mb. That's an awful lot of
disk space for one MAD dataset.
Usual problems
Images are blank:
End of Run Forms
Don't seem to be required at Argonne.
Lab Access, liquid Nitrogen etc
Each sector at APS has a dedicated office and lab module and
usually there is a wet lab available for limited sample preparation.
Unfortunately, for NE-CAT this space is at Sector 21, and 8BM is
at Sector 8. You'll come to value that tricycle. More recently
we actually got some lab space at sector 8, which may be an ongoing
feature.
Transportation, Lodging, Food at Argonne
The best airport is Chicago O'Hare, with regular service from La
Guardia and about a 2 hour flight time. Midway is the other airport,
but it's uncommon to end up using that. Chicago is in the Central
Time Zone and one hour earlier than NYC. Argonne is approximately
30 minutes south of O'Hare, subject to the inevitable Chicago rush-hour
traffic. Take I-294 south and exit west onto I-55 (direction of St
Louis), taking Cass Ave south only a few miles along I-55.
