600 and 800 MHz NMR Notes / News

Instructions for using the cryoprobe are now available.

Fast determination of 90 degree 1H pulse
A new automation program determines the 90 degree 1H pulse automatically in a few seconds for different power levels. It works best with samples with an intense solvent peak. Simply type p90 in Topspin.
Note that the program also executes a 'getprosol 1H', i.e. the 1H and gradient pulse lengths in the acquisition parameters of the current data set are automatically updated.

Topspin License Problems
If you get the following error when trying to start Topspin on a data station
  FLEXlm FEATURE TOPSPIN1 License server does not support this feature

The solution is to set the BRUKER_LS_LICENSE_FILE environment variable.

If you are running under Windows:
Control Panel --> System --> Advanced --> Environment Variables --> New
( under User Variables for XXXX)
fill in the dialog box
variable name = BRUKER_LS_LICENSE_FILE
value = 1700@carp.anu.edu.au

Then logout and log in

If you are running under Linux:
Create the following script to run topspin
If necessary, change the last line to the installed location of Topspin.
Make the script executable
  chmod 755 topspin
Then execute the script to start Topspin.

Software Upgrade (13 October 2004)
Peter Barron has recently done some work on the computer which drives the 800:

Currently available probes (July 2004)
TXI   H / N / C probe with automatic tuning and matching.
TBI   H / P / broadband probe with manual tuning and matching. (BB tunes down to Rh).

Shigemi tubes:
Varian Shigemi tubes are not suitable for our 800 MHz NMR spectrometer, because they have a too long segment of glass at the bottom. (The tube must not be placed lower than what is indicated by the line in the plexiglas sample-adjustment tube, since the temperature sensor in the probe is right below.) Bruker Shigemi tubes have a shorter glass segment (in fact, too short, according to Ralph Ruin) which allows the sample solution to be centered in the rf coil.

Also, do not use "Varian length" Shigemi tubes in Bruker cryoprobes. There is some danger of doing some damage to the probe. Only use ones with "B" as part of the part number.

zz, multizz
GO, 7 June 2004
Always use zz instead of zg.   zz is an AU program from Rutger Folmer (Astra-Zeneca) which
- copies the current fid/ser file to the directory ~/dumpster, i.e. one doesn't accidentally delete data so easily.
- copies the actual temperature parameters (flow rate, temp. setting, correction factors) into the current data directory (file edte).
- copies the real pulseprogram in the current data directory.
multizz is like multizg, except that it uses zz instead of zg and allows to do gradient shimming between queued experiments.

Peter Barron, July 2004
The "getprosol" command is very useful once you have set up the pulse program and channel assignments you want to use. It drags all the pulse times/pulse shapes/ pulse powers from the prosol table that the pulse program/channel assignements require. Note that pulse times for observe and decouple 1H and 13C (i.e. f1 and f2) and decouple 15N (on f3) have been entered in the table.

But what happens if your 1H pulse time (very likely) or 13C or 15N (less likely) are different to what is in the prosol table? The "getprosol" command can calculate new values on the fly.

i.e. getprosol   <nucleus>   <pulse time>   <power>

   getprosol   1H  9.2   -2.0
calculates all 1H pulse times relative to a 9.2 pi/2 @ -2.0dB and loads 13C and 15N from table

   getprosol   1H   9.2   -2.0   13C   11.1   -6
calculates both 1H and 13C, including all adiabatic pulses, and loads 15N from table.

adiabatic pulses
Peter Barron, July 2004
At 800 MHz need to use adiabatic pulses for 13C so look for experiments with "sp" in the name. "getprosol" will then load the default shapes/powers/times. (If you are doing 1H-15N experiment which is designed for doubly labelled samples, the pp name may not contain "sp", even though it contains shaped/adiabatic 13C pulses, as this would imply adiabatic 15N pulses. Note further that some pp's for 1H-15N experiments have optional parts only performed if the sample is 13C labelled as well. Whether this part is executed depends on the use of the zg options either on the command line or the parameter ZG_OPTNS
i.e. "zg -DLABEL_CN")

The above getprosol commands do not set up for 13C adiabatic decoupling which is essential for 800MHz.
After doing getprosol, check/set:

cpdprg2 =       p5m4sp180
pdcp2   =       1.5ms
pl12    =       2-3 dB less than for garp (2-3dB < value called in by
                getprosol i.e. 12-13dB)
spnam15 =       Crp40,1.5,10.1

CB, 28 May 2004
The Perl script explist lists all experiments in a directory.

Original program by Rutger Folmer. Modified by Patrik Andersson and me.

ssh, lsh
GO, 28 May 2004
The au-programs ssh and lsh store and load the current shims in a file in the data directory. Great for retrieving shims, as they tend to be sample dependent.

CB, 28 May 2004
The Perl script lpa lists the acquisition parameters in a user-friendly format. It must be started in a shell at the directory level of the experimental number (expno). For example lpa   3
Original program by Rutger Folmer. Modified by me.

ATMM Problem
GO, 15 May 2004
I noticed that after adjustment of the tuning and matching of the 1H channel with atmm, the probe gets again vastly detuned after the first 1H pulse applied. A second subsequent atmm provides stable tuning and matching. The problem is specific to the 1H channel.

Temperature Problem
GO, 15 May 2004
Peter Caha broke the connection to the BTO2000. This means that the temperature displayed in edte is ca. 20 deg. lower than the actual sample temperature. I did a temperature calibration:
A setting of 278.5 K yields 25 °C.
A setting of 290.8 K yields 37 °C.
Position 5 of the sample changer now holds the Bruker temperature calibration sample (ethyleneglycol with 20% DMSO). The program batman allows an automatic (and very time consuming) temperature calibration.

For the Bruker temperature calibration sample (ethyleneglycol with 20% DMSO):
temperature   peak separation
25 °C 1.50 ppm
37 °C 1.38 ppm
47 °C 1.30 ppm

Joel Mackay, 28 April 2004
We have sparky running fine now. It is installed in /opt/sparky and if you put
  alias sparky='/opt/sparky/bin/sparky'
in your .bashrc file, you should be able to run it fine. It allows to display the rrr files from Bruker 3D FTs.

Always leave some sample in the magnet!
GO, 21 April 2004
The sample changer requires that the magnet is left open, i.e. no cap put on top. In order to keep dust and dirt out of the probe, please make sure that there is always a sample in the magnet.

Sample volume
GO, 20 April 2004
Ralph Ruin from Bruker recommended that the top of the solution is at least 4 cm above the bottom of the NMR tube to get good shims. This corresponds to 550 microlitres.
Shigemi tubes:

Frequencies, bf2, bf3, calibrate
GO, 20 April 2004
The basic frequencies for 13Cand 15N were changed to 201.192313 and 81.076467 to yield o2p and o3p parameters which are nearly correct (within 0.2 ppm). Bruker default parameters reference to TMS instead of DSS (DSS is the commonly used standard for aqueous protein solutions), resulting in about 2 ppm deviation of the 13C frequency.
Accurate calibration values (namely the chemical shifts of the carriers, assuming that the 1H carrier is at the water frequency) can be calculated by the stand-alone program calibrate which must be run in the data directory containing the acqus file. It does not modify any files of Bruker.

GO, 19 April 04
Every user should put extensions on all user-generated files (e.g. two-letter initials) which end up in the common directories (pulse programs, shapes, parameter sets, gradients). This will make it much easier for keeping track of the files.

.bashrc, explist, pp, dd

GO, 19 April 04

Modify the .bashrc file to include your favourite aliases, paths and prompt. For a start, I copied mine, /home/go/.bashrc, to every user to give you a prompt that tells you where you are, alias pp to take you to the pulse program directory, alias dd to take you to your data directory, and access to the perl script 'explist -t' which gives you a listing of all experiments done including titles. More perl scripts will be added. This setup will also let you use $pp instead of /opt/xwinnmr/exp/stan/nmr/lists/pp and $dd for your data directory /opt/data/<username>/nmr.

Avance 800 Update

Peter Barron, April 04

The Avance 800 has finished its basic installation and acceptance testing, although the TBI probe has yet to be installed and the delivery of the Cyroprobe is still a number of months away. A loan TXI probe with XYZ gradients and ATMA is currently installed. The sample changer is operational. Hence the system is available for use, either locally or remotely.

User Accounts

The following accounts for non-ANU users have been set up:

sydney,   unsw,   wollongong

Each has the standard Bruker passwd for both login and vnc remote access. Please change these passwords on first login using the commands "passwd" and "vncpasswd". Each login has startup files for xwinnmr (.uxnmrrc and .xwinnmr-popeye/) and VNC (vncstart and .vnc/) installed.

Remote Access

See the document on Remote Access by VNC. When operating the system remotely, the following commands will be useful:

sx n insert sample from holder n of BACS 60 sample changer
bsmsdisp open shim and lock panel for manual shimming via mouse
atma automatic tuning & matching ("atma exact" for best results)
atmm manual tuning & matching via mouse
gradshim gradient shimming - see later

Probe Temperature

Ensure that edte window is open, that the airflow is set to 535 l/hr, the heater is on and the temperature is stable before proceeding with tuning and shimming. The temperature from 280 K up has been corrected and certainly should be accurate to < 0.5C.

Probe Tuning

"atma" will automatiaclly tune the probe (see bug below). "atma exact" will do a slightly better job and take a little longer. "atmm" will allow manual tuning & matching via the mouse and the wobb signal in the acquisition window - in doing this you will note some hysteresis in the adjustments i.e. when you reverse the stepper motor direction, it may take a number of mouse clicks before any adjustment occurs. Also note that when direction arrows turn red, the limit of adjustment in that direction has been reached.

Currently there is a software problem that prevents tuning of 15N via the 15N preamplifier - no wobb signal is observable. To tune 15N requires a cable change BUT the 15N tuning is very insensitive and therefore does not need checking. To avoid problems, turn 15N on f3 to off in edasp before using "atma". This is not necessary using atmm, just change to 13C once the tuning window starts.

Gradient Shimming

Shim maps for 1H 3D, 1H 1D and 2H 1D have been performed. Each login has a series of shim iteration files installed, all involving a single iteration. If you want to run multiple iterations, just repeat the single iteration. Make sure the user name under gradshim is correct in order to avoid permissions error during data acquisition. Suggested iteration files to use are:

1D 1Hi1_z15_26
3D 1Hi1_shim12_26 (or i1_shim19_26 or i1_shim27_26)
1D 2Hi1_z15_22

Note that it is not necessary to do 3D shimming on every H2O sample. However if it is done, it is then advisable to follow it with one iteration of 1D 1H shimming as this may produce a slightly better result for the on-axis shims.

Pulse Times and Power

The temporary TXI probe has strict power limits that can be used. Users must stick to these. The 15N pulse time is relatively long to ensure maximum power on the outer coil during simultaneous 13C & 15N pulses is not exceeded. (At some point in the not too distant future we will set up the PowerCheck software to minimize the chance that powers can be accidentally set to exceed probe limits). The pulse times measured during installation are:

NucleusPulse TimePowerComment
  12.1adiabatic decoupling
(cpdprg2=p5m4sp180, pcpd2=1.5ms, sp15=Crp40,1.5,10.1)
(Comment by GO:2208.34 garp yields perfectly fine decoupling with significantly less heating)

Please do not exceed these powers. It is recommended that, after setting your pulse program name, pulse times and powers are loaded from the prosol table via the "gpro" command. The prosol table has 1H, 13C and 15N pulse times, powers and shapes set, including all the recommended 13C (use adiabatic with cpdprg2=p5m4sp180, sp15=Crp40,, pcpd2=1.5ms )and 15N (garp) decoupling values, and the selective pulse shapes, pulse times and powers for 13C selective excitation, adiabatic refocussing and inversion. All rf channels and amplifiers have cortab tables so that linear behaviour can be assumed. If you checked your 1H pulse time and find it has changed, use gpro command with the following options:

    gpro 1H <pulse time> <power level> i.e. gpro 1H 10.2 -2

which will recalculate all other 1H pulse times used by the current pulse program according to the hard pulse time and power. (There is a bug at the moment - gpro creates a file "prosol_History" in the current dataset but with permissions of r-r-r- or 444 which therefore cannot be overwritten by a subsequent gpro command. To repeat gpro, you will need to find the experiment number in a shell and delete the "prosol_History" file first).


Data Storage & Transfer

Data should be collected in /opt/data/<username>/nmr and should be transferred using ftp after "tar"ing and zipping (gzip). Note that the ftp server on popeye is disabled so it will be necessary to ftp from popeye to an ftp server on your campus. I assume Gottfried will implement some policy for cleaning up data.

Don't hesitate to contact me if you have any questions or encounter any problems in using the system.

800 MHz NMR home page.