Dear Colleagues,
please find the summary of helpful e-mail answers I received about
homonuclear decoupling of 31P on our INOVA spectrometer. Thanks very
much to Péter Sándor, Robert Santini and Sara Kunz.
The clean way of solving this problem is to write a pulse sequence that
has the homonuclear decoupling built in with "Explicit Acquisition".
Petor Sandor sent me such a sequence for instruments with gradient units
and wave form generator boards. Since the specific instrument, which we
use for these experiments, does not have either of them, I modified the
sequence to work without them. I am happy to send you my sequence,
parameter file etc. if you need it.
Thanks again
Ulli
______________________________________________________________________
My original posting:
Dear Colleagues,
we would like to do a homonuclear decoupled phosphorous experiment, i.e.
31P {1H/31P} on our 3 channel INOVA spectrometer. Setting the second
decoupling parameters to phosphorous (i.e. dn2='P31', homo2='y' etc.)
the instrument automatically assigns the rf channels to
1: for P31 transmission
2: for 1H decoupling
3: for P31 homo decoupling.
This makes sense, as the second channel is 1H only.
The problem now arises from the fact that the decoupling comes out of
the 3rd channel connected to the second low frequency amplifier. When we
decouple C13 instead of P31, we feed the 3rd channel through a power
combiner (basically a sophisticated T) between the duplexer and the
probe and use the corresponding band pass filters for the transmission
and decoupling lines. Things work well this way.
When I hook up the system analogously for the P31 homo case (even with
dm2='n'), I do not get any signal. With homonuclear decoupling I now
have the problem, that I cannot use the filters to rout signals and
clean them up.
My questions are the following:
- How do I prevent the transmission pulse to go into the decoupling line
and amplifier.
- Do I need to worry about the decoupling pulse to go into the
transmission line (or does the homo gating and the duplexer switching
prevent this.)
- How do I prevent the signal from getting lost in the decoupling line
(I tried crossed diodes in line in the decoupler line, still no signal,
even when dm2='n')
- The amplifiers are way too dirty, even when turned off (dn2=' ') to be
connected into the transmission line. Just upon connecting, I get
receiver overflow, unless I use gain=0. Again, in the C13 decoupled
case, the filter cleans it up enough for this not to happen.
On our 2 channel Gemini2000 instrument, we have no problem doing this
type of experiment. The console itself is doing the gating because it
uses only the one available amplifier for that frequency. The same is
true on the INOVA for homonuclear proton decoupling. So alternatively
to solving the problems above, is there a way to force the INOVA to use
the same amplifier for P31 transmission and homonuclear decoupling?
Thanks for your help
Ulli
_____________________________________________________________________
From: Péter Sándor
Let me disagree with your statement: "On our 2 channel Gemini2000
instrument, we have no problem doing this type of experiment."
I am afraid a Gemini has never been able to do 31P homodecoupling with
additional 1H BB dec. An ordinary homodecoupling, on the other hand has
never been a problem for the INOVA either.
I have, however, a simple solution for your specific problem: use the
first
channel for 31P observe as well as for homodecoupling, and channel 2 for
BB
1H decoupling. You do not need the third channel, you do not need to
recable anything or add additional hardware or add specific filters. All
you need is a pulse sequence (see attachment with a parameter file,
setup
macro and a manual file) plus a waveform generator in your 1st channel.
(I
hope in a 3 channel INOVA you have got at least one of them.)
Let me know if you need more help.
Best regards, Peter
____________________________________________________________________
>From Robert Santini:
I received your e-mail. I am responding with a couple of "first thoughts"
regarding your problem. I could be quite wrong, however. I will review
our Inova schematics if you don't get a reasonable suggestion from
another AMMRL person soon.
(a) Typically, the Varian 1H homo nuclear decoupling is done with a
built-in power combiner. If hetero-nuclear mode is specified, the combiner
is bypassed such that the output of a high power 1H transmitter is sent
directly to the probe and 1H coil. When the homo-nuclear mode is selected,
a separate, fairly low power 1H transmitter is routed by way of the
combiner to the probe. The combiner is present to co-add the 1H
transmitter pulse and the homo-nuclear decoupler pulse. This is no big
deal at 1H frequencies since the system is not particularly power limited
given the narrow 1H survey window. The transmitter duty cycle is low
enough that the combiner is not destroyed by the high peak power of the
transmitter pulse.
I have a suspicion that the software has set the homo-nuclear mode on the
3rd channel on the assumption that it is equipped with an RF combiner plus
associated RF relays, a low power 31P source, etc. This is not necessarily
the way the system is shipped from the factory. The 31P survey window
requires that all the pulse power that can be produced be applied to the
probe. In this case, the insertion loss of the combiner for the
transmitter pulse will significantly lengthen the 31P 90-degree pulse
width. If this is the configuration of your system, there will be no 31P
homo-nuclear decoupler pulse present at the probe. There should be a 31P
transmitter pulse.
b) Another possibility is that the third channel is not particularly
broadband. As I remember, on some Varian consoles, the "low cost" third
channel power amplifier does have sufficient frequency range to cover 31P and 13C.
I would have expected that the setup with an external combiner and
appropriate filters would work fine for 31P if does so with 13C.
However, if there was no significant transmitter output at 31P frequencies, you
would see a result much as you describe. This could also explain the high
noise level that you have encountered. The power bandwidth of the third
channel may be limited, but the noise bandwidth will be much greater.
C) You could have a combination of (a) and (b).
I trust that the above discussion may be of some help to you. I can think
of other possibilities to explain some of your symptoms but they are less
probable in my view. Let me know what caused your problem when you get
things running properly. If I can be of further assistance, please send
e-mail.
Best regards,
-- Bob Santini
Received on Wed Nov 07 2001 - 18:21:00 MST