Summary Boron 11 NMR Problem

From: <James.L.Simpson_at_invista.com>
Date: Thu, 17 Mar 2005 14:01:53 -0500

Thanks to all those who replied. I got some very useful information that I
was able to use to improve my spectra. I was already using quartz tubes,
but hadn't thought of the glass in the probes.

Jim Simpson
Invista
Orange, TX

Here is a summary of the information:

1. 3-coordinate Boron has T2's that broaden the signal. 4-coordinate Boron
will have more sharp, well defined resonances.
2. Quadrupole effects increase with increasing field, causing increased
line broadening.
3. Most probes have borosilicate glass inserts which can create broad
background interference. Boron concentration in the glass is about 10%.
When working with dilute Boron samples, this can have a major effect. One
way to help minimize this effect is to increase your ringdown delay to
about 1msec.
4. Because of the broad lines, it is not necessary to lock or shim.
However, I found it critical to ensure a hard pw90 pulse was being used
with a well tuned probe. I tuned the probe for every sample (usually only
minor adjustments). The pw90 was found using a 1% BF3(Etherate) samples in
Benzene.


Detailed responses follow:
~~~~
Boron that is 3-coordinate has T2?s that make it almost impossible to get
a sharp signal. 4-coordinate boron compounds have very sharp, well
defined resonances. If a chemical exchange process is going on then no
signal. Cool the sample down 20-30 C and try to acquire a signal. Try
running at 300 MHz to also avoid Quadruple moment issues.

Too high a field and too small a sample filling factor (use 10 or 15mm probes).

Alan Kook, PhD
NMR Applications Manager
Austin AI
~~~~
The big problems occur for two reasons.
1) 11B exhibits a nuclear quadrupole moment.
2) There is a high probability that the insert in your probe and the NMR
tube are boro-silicate glass. The boron concentration in the glass is
about 10%.
The first effect will give very broad lines unless the 11B atom is in a
symmetrical molecular environment. The second will produce a very fast
decaying FID where the initial intensity is high. The background FID will
swamp out more dilute boron signals in the receiver and digitizer. There
are also complications associated with receiver gating, etc.

I suppose my point is that there are fundamental issues that cannot be
solved exclusively by NMR tricks. One must exploit synthetic methods to
get the right boron environment for NMR. Then, one must eliminate the
boron signal from the probe and tube. We had boron-free probes built and
still use quartz NMR tubes.

Once these steps have been taken, boron NMR becomes much easier and a lot
more interesting.

Best regards,

-- Bob Santini

PS -- We have tried spin-echo techniques, but they are not useful except
in cases where the boron line width is already manageable. I would expect
2-D methods to be ineffective for this reason.
~~~~
I have only run a handful of 11B experiments, but I have had better luck
using quarts tubes as opposed to boro-silicate tubes. I have not had the
luxury as of yet to run 11B NMR in a glass free probe, but I expect that
this would help my results even more. Our approach has been to toss the
first few data points and then do backwards linear prediction. This
doesn't seem like a very reasonable approach for low concentration samples
though.

Don Eldred
~~~~
Boron peaks are generally pretty broad, so that is probably your main
problem with small samples. Relaxation should be fast, so a delay of
0.5s or so should be fine.

Have you tried running your standard as a separate sample? The peaks
will be broad enough that the referencing will be good enough, and it
might help you optimize resolution on your sample.

Regards,

Phil.

--
Dr Phil Dennison
NMR Facility Director (949)824-6010 (office)
Department of Chemistry (949)824-5649 (lab)
University of California, Irvine                 (949)824-8571 (fax)  
Irvine, CA 92697-2025 dennison_at_uci.edu 
USA
~~~~
Since B-11 is a quadrupolar nucleus, the relaxation times are extremely
short.  I have measured them in the low msec range.  One user had a very
unusual sample that probably formed micelles and he had an amazingly long
T1 of ~50 msec as I recall.  We use a relaxation delay of 2 microseconds
and a data acquisition time of 33 msec.  With these parameters we always
use at least 1,000 scans, which gives a total experiment time of 44
seconds..  (Depending on the spectrometer system overhead, you may have a
minimum value for the relaxation delay that will work at all.)  Because of 
the short relaxation times, you of course have relatively broad lines.
(BF3-etherate is an exception due to the symmetric environment of the B11.
 It?s width at half-height is ~7 Hz.)  Because the lines are broad, we do
not lock, spin, or shim.
Dr. Jane Strouse
Director of Instrumentation
Department of Chemistry and Biochemistry
1410 Molecular Sciences Building
UCLA
Los Angeles, CA 90095-1569
~~~~
We have been doing some water treatment improvements to our house well
water. The goal is to remove the Borom from the current 40 PPM to less than
1 PPM so we don't kill plants we water with the well water. In this effort
we have been running NMR of water samples looking at Boron using our old
NT360 with a 12mm tube. We can easily see Boron in the 1 to 40 PPM range
running 4096 scans with a total recycle time of 0.1 sec and a 90 degree
pulse. We do not lock and observe excellent sample to sample reproducibity.
Our major problem is the background Boron signal from the Boron in the glass
NMR tube and insert. We run a blank sample of pure water and subtract it out.
Woody Conover II
woody_at_acornnmr.com
~~~~
 My experiences in running 11B have been on  GE 300 MHz and an Inova 500
MHz NMR systems.  We found better results on our 500 MHz NMR system than
on a 300 MHz NMR system; quadrapolar effects actually decrease with field
strength.    You can run a very short recycle delay with 11B; your delay
values should be fine.  You may have too little boron to observe.  One
major watchout; if you are using a standard 10 mm BB probe (the same as
the one I have) then you will get a strong background signal from the
probe (likely from borosilicate glass inserts in the coil area).     My
solution to solve this was to increase my ringdown delay to around 1 msec.
   You can verify that this is or is not a problem by collecting the
spectrum with no NMR tube present and see if any of your peaks are still
present.     More elegant solutions involve using some type of echo
sequence but I never tried that.   The linewidth of 11B is fairly broad;
good luck getting this to work.
                      Mike Krejsa
Michael R. Krejsa, Ph.D.
Manager, Analytical
Cytec Surface Specialties
730 Worcester Street
Springfield, MA  01151  USA
(413) 730-2470 Phone
(413) 730-2196 Fax
~~~~
A few comments to your Boron problem. A repetition time of 0.1 sec. is
fine. Lines are broad, but T1 is equaly short. Measuring with external
standard is not necessary, you could better use the newest varian
userlib setref, and use this as external reference. You need of course a
good pw90, but the decoupling calibration as used in 13C or 31P is good
for 11B as well. When the obeserved signals are small, the probe
background- and tube-signal must be huge. Normally your signals should
be much narrower as the background signal. We use a macro setlp (see
attachment) to minimise these signals with backwards linear prediction.
Just try setlp(n) with n=3-30 and wft. It is a bit tricky but with a
little bit experience it works fine. We measured a lot of B11 1-D and
2-D spectra with our 5mm probes on Unity-500.
good luck,
Jan runsink
"********************"
"setlp  - Backward LP"
"********************"
"****************************************************"
"remove baseline_roll when alfa too large with :     
"setlp(>n) lsfid=-n ; choose 'n' to minimize lp_value"
"default value: n=3 or `sw/12500`                    "
"              if lsfid < 0 --> default=default-lsfid"
"****************************************************"
setlimit('lsfid',64000,-64000,1)
parlp
lpopt='b' lpfilt=16 lpnupts=128
while ((lpnupts+0.5) > np/2 ) do
  lpfilt=lpfilt/2 lpnupts=lpnupts/2
endwhile
if ($# < 0.5) then
  if (sw > 35000) then
    $1=trunc(sw/10000+0.5)
  else
    $1=3
  endif
  on('lsfid'):$lsfid
  if ($lsfid > 0.5) then
    $lsfid=lsfid
  endif
  if ($lsfid < 0) then
    $1=$1-$lsfid
  endif
endif
if ($1 > 0.5) then
  strtlp=$1 strtext=$1 lpext=$1
  proc='lp'
  fpmult=0.5 lvl=0 tlt=0
 else
  proc='ft' fpmult='n'
endif
~~~~
www.invista.com/email_disclaimer.shtml

This communication is for use by the intended recipient and contains information that may be privileged, confidential or copyrighted under applicable law. If you are not the intended recipient, you are hereby formally notified that any use, copying or distribution of this e-mail, in whole or in part, is strictly prohibited. Please notify the sender by return e-mail and delete this e-mail from your system. Unless explicitly and conspicuously designated as "E-Contract Intended", this e-mail does not constitute a contract offer, a contract amendment, or an acceptance of a contract offer. This e-mail does not constitute a consent to the use of sender's contact information for direct marketing purposes or for transfers of data to third parties.

Francais Deutsch Italiano Espanol Portugues Japanese Chinese Korean

www.invista.com/email_disclaimer.shtml
Received on Fri Mar 18 2005 - 10:46:59 MST