About a week ago, I posted some questions about single versus triple
axis gradients and probes. Below I have posted the original questions
and a summary of all of the responses (with names deleted). Thank you
to everyone who responded. We are leaning toward the single axis
gradient system but have not yet decided on which probe(s) to get. I
will post an update once we get everything put together (which will
probably take a few months).
Sincerely,
Brian
K. Brian Killday
Research Specialist
Division of Biomedical Marine Research
Harbor Branch Oceanographic Institution
5600 U. S. 1, North
Ft. Pierce, FL 34946
Phone: 561-465-2400 ext. 456
Fax: 561-461-2221
Email: killday@hboi.edu
*******************Original post:
>On January 7th, 1998 Brian Killday wrote:
Greetings AMMRLers,
We are about to upgrade our Bruker AMX-500 spectrometer to gradient
capability and I am seeking opinions on which upgrades will be the most
cost effective for our needs. We are a Marine Natural Product Chemistry
lab doing isolation/structure elucidation mainly of small molecules
(<1Kd). We are debating whether to purchase a triple axis grasp III
unit ($40,000) or if we can get by with the single axis Grasp II
($17,000). Most of the standard experiments that we do only require
single axis (COSY, TOCSY, HMQC-TOCSY, HMBC, etc.). What particular
advantages would the triple axis capability have for small molecules and
would it be worth the cost difference? Do any of these experiments run
better with triple axis gradients and are there other experiments
potentially useful to us that will run only using triple axis gradients?
We are also looking at microprobes. How do the Bruker 2.5 mm probes
compare to, for example, the Nalorac probes? Thank you in advance for
your input.
Sincerely,
Brian
***********************Responses:
We have 6 Brukers, all with triple axis
gradients.
Two are DMX/DRX, and the rest are AMX's with gradients upgrade.
Despite this, I'd recommend single axis.
If Bruker sells you just a single Acustar amplifier and waveform
generator that
fit in the analog router slots, then I would just get that.
Then I'd buy Nalorac single axis gradient probe.
(What you need is, one Acustar amplifier; waveform board that fits in
the router;
something called level adapter box (or something like that) and cables;
cabling
to go from Acustar to the probe).
Except in cases where you have 7~10 gradients as in some triple
resonance expts,
etc., single axis is sufficient.
Triple axis is handy if you want to fire along different axes to prevent
accidental
cancellation of gradients (gradient recalled echo), but the problem with
Bruker
triple axis probes is that due to the transverse windings to generate
the x,y
gradients, the z gradient's active shielding is terrible, and their
so-called
"preemphasis" adjustments do not work well for real life pulse sequences
and
cause long eddy-current ringdown times, and ultimately, you have to
shape all
the gradient pulses, which leads to headaches (although I've been doing
it this
way for over 3 years now, I wouldn't wish it on anyone), and if you ever
need
to do gradient heteronuclear coherence selections with echo/antiecho
deals where
you have to change signs of these gradients to not lose sensitivity &
get pure
phase, etc., it's a real headache, especially with AMX's due to
limitations in
waveform memory and also pulse programming syntax).
Nalorac's z-grad probe's active shielding and eddy-current ringdown is
so good,
you don't need any shaping, and makes your amx pulse programs with
gradients
much, much simpler.
___________________________________
> What particular
> advantages would the triple axis capability have for small molecules
and
> would it be worth the cost difference? Do any of these experiments
run
> better with triple axis gradients and are there other experiments
> potentially useful to us that will run only using triple axis
gradients?
We have had triple axis gradients for almost a year and haven't even
installed them. From what I hear, the only advantages they have are for
DQF-COSY and for microimaging. And I think Bruker might have a
triple-axis shimming capability (Varian only has z-axis and that seems
to be sufficient for us).
_____________________________
I have compared the basic 2D expts you mentioned on a DRX400 (single vs.
triple), and the difference is not significant enough too justify the
expense. One advantage of the triple axis is the gradient shimming,
especially if you add new probes.
_____________________________
We have the triple package and have found that it works
well. The triple gradients will make a difference for
double quantum filtered cosy in H2O solvent. In general,
gradients applied at the magic angle (all three gradients
applied at equal strength) will produce the best water
suppression. For virtually every other application a single
gradient will work quite well. I'm not sure the 3 gradient
package is worth the extra money.
If you plan to do many 3D experiements (e.g. HMQC-COSY)
where gradients are used to phase encode in t1 and t2 then
the 3 axis approach can provide benefits. Using a single
axis approach for both the HMQC and the COSY experiments can
result in unwanted gradient-recalled-echoes. If you
separate the gradient phase encode steps for the two
experiments (X axis for HMQC and Y axis for COSY) then you
are assured of isolation between the two.
___________________________________
I can only speak to the Z gradient case as very nice on an AMX500.
We do all of the experiments which you list and the results are
excellent
Our main user does mostly gradient enhanced/selected 3D experiments
on fairly large proteins and they are very happy with the results.
(Your price surely does not include the probe $.)
We use mostly a Bruker Triple (H,N,C) probe with Zgradient but also have
a 3mm Nalorac BB which seems to give the same sort of performance with
respect to gradients but with better s/n with sample limited cases.
If I had the money I would buy the triple axis gradients. I hear tell
that it is very nice for shimming and it greatly improves the water
suppression (our protein people work in 90%H2O!). I would be interested
in what people say about actual experiment use and performance so would
appreciate a copy of your replies
______________________________
When we purchased our system three years ago, the triple axis gradient
unit was an acqustar made in Fremont by xGE people. It was better
than the single axis French unit so we got the acqustar with just 1
gradient. We could upgrade to triple axis with two boards and a new
probe, but don't have any need to yet. I think that water suppression
may be slightly better with triple, but we don't do much of that work.
I imagine that the French unit has improved in the last three years.
good luck,
_____________________________
We have a single axis gradient on our 400. It works just fine for small
molecules. There are two cases where I would prefer a triple axis
gradient.
1) For magic angle gradients which provide better quantum filtering and
hence better solvent suppression in COSY, etc.
2) For diffusion measurement because convection effects are weaker in
the
x-y plane than the z direction.
This may or may not justify the extra cost.
______________________________
We also have an AMX 500 which is equipped with three-axis gradients; we
are
glad that we got them. Diffusion experiments in particular are easier
to
do with transverse gradients, since the length of the sample is less
important.
Also, we like the magic-angle gradient DQFCOSY, which does water
suppression
beautifully without saturation, this requires the three-axis gradient.
______________________________
Re - your ammrl post:
We have a similar lab to yours --- we run a plethora of gradient
enhanced hmqc/hmbc experiments on natural products. When we had this
decision several years ago we went for the single axis. Of course at
that point gradient hmbcs had just been suggested at an Asilomar ENC.
We are at the point of turning over our console and I am just starting
to think about the gradient business again. It is not clear to me that
triple axis would help much but I need to learn more about it. We have
been getting some interest here in doing larger molecules for which the
triple axis is helpful, but that is not your problem.
_______________________________
We had a single axis few year ago and had a exchange for a triple axis
gradients. For the routine non aqueous ( H2O ) sample, I think a single
axis is efficient. The main advantages of the triple axis are :
- MAS gradient experiences, mainly for water suppression. ( It worked
only in some experiments such as COSY and the results are fantastic, but
for most experiments, the results from MAS and regular gradient
experiments are very similar.
- 3D shimming, but we need a very strong signal ( ie mostly H2O/D2O
solvent).
With single Z axis, one can only gradient shim the Z's ( Z1,Z2...) using
the deuterium signal of the lock signal. ( In principal, one can do a 3D
shimming with the deuterium signal, but in real world, the time
requirement to my experiment rules out this option. )
- In most `fancy experiments, there may have artifacts resulting
from gradient echos, so the use of different gradients reduces these
artifacts.
- Extra gradients in case... Z gradient is usually about 30% stronger
than X and Y, but in our system give more eddy current than X and Y, so
the gradient recovery time is usually longer. But for small and medium
size molecules, it will not be usually a problem.
_________________________
We just installed a Bruker 2.5 mm probe on our Avance 500. We
got a dual 13C/1H probe with a single gradient and the S/N was
higher than their published values. The probe has excellent
line shape, and it is also very easy to shim. The only thing I
really know about the Nalorac probes is that when gradients are
included with the probe, there is a reduced temperature range
compared to the Bruker probe.
The question of whether to get 1 or 3 gradients will probably be
with us for a while. For the experiments you described in your
note to the AMMRL mailing list, a single gradient would probably
be sufficient, and it does not reduce the probes' sensitivity
quite as much as does a triple axis system, in general about
10%. For water suppression and certain other experiments a
triple axis gradient is definitely worth having, and there is
always the likelyhood that more experiments will be developed
utilizing all three gradients.
________________________