I was asked to give a summary of the responses that I got for the question of
doing Tg's by NMR.
In response to your message, and one I received in response about
DSC - NMR T1p (T1 rho) gives an extremely sharp TG compared to DSC, in my
experience.
Don Wilson
Chevron Research & Technology
--The spectral characteristics are quite
different on either side of Tg. The reason for that is that at temperatures
below Tg local motions such as side-chain rotations, terminal group
rotations, or main chain local fluctuations dominate the observed NMR
relaxation behavior, but at temperatures above Tg, backbone motions
contribute importantly. See first two refs. in our paper : Ganapathy, Chacko
& Bryant, Macromolecules 19, 1021 - 1029, 1986 for more details. Our paper
also has many other relevant references in this regard.
Hope this helps.
chacko
You can almost use "any" solid state NMR technique to determine the Tg of
polymers. Closely above Tg motions of polymer chains will freeze, i.e. the
rates of
this motion will drop by several decades during a range of only a few
centigrade (this is
described by the WLF equation). So most solid state techniques which are
sensitive to
molecular motion (line widths, T2, T1ro, ...) will show drastic changes of data
when passing the glass transition. T1 is possibly not a good
candidate since T1 rather picks up high frequency motions of side chains
occurring also below Tg! will show drastic changes
of data when passing the glass transition. Probably one of the most simple
approach would be measuring 1H-spectra as a
function of temperature. If you plot their width at half height versus T you
will see a significant change of the linewidth just at the glass transition.
Remember that the glass transition is commonly reported as a "dynamic glass
transition", i.e. it is associated with a characteristic time scale and as such
has to be published
with this time scale. (Just the same as the Tg measured using DSC always has to
be reported
together with the heating or cooling rates). So depending on the NMR experiment
you chose
and the way of data analysis, you might obtain different values for Tg. But
this is the nature of Tg, you just have to report it together with the
experimental protocol, how you obtained it.
I hope this information is helpful
Regards
Hanno Leisen
...you will see some of the motions of Tg in NMR but not at the
'right' temperatures and there will be interference's by other
side chain motions, etc.
Charlie Dickinson
It seems to me that this is done, but I can't recall by whom at the moment. I
do recall that Stephen Cheng's group at University of Akron has done a fair
amount of variable temperature solid state NMR of polymers. Hans Spiess, early
work by Schaeffer and Stejskal, and David vanderHart might also be worth
searching. I can think of several approaches that might work, but I'll bet you
can find proven methods in the literature.
One thing to keep in mind that probably won't be discussed in the literature is
the issue of temperature calibration. It is important to calibrate your
temperature using the same conditions that you intend to use for your samples.
You should be aware of the parameters that are important for accurate VT
calibration for the hardware that you have, and make sure that you control
those well during your calibration and experiments. For example, calibrate at
a particular spinning speed, with certain air pressures/flows, and a particular
rotor material, and then hold all of these parameters/materials constant for
your samples. You will also want to get a good feeling for temperature
gradients across your sample, and may wish to consider restricting your sample
volume to keep gradients to a minimum, depending on the magnitude of the
gradient that you measure and how accurate you want to be in your TG
measurements.
There are several papers that demonstrate calibration of temperature using lead
nitrate. See SSNMR, 5 (1995) 269-271 and JMR A, 116 (1995) 215-220. Remember
that lead nitrate is very dense, and a full rotor should never be spun at
maximum speed. You can dilute it with something like silica gel if you need to
spin close to the limit of your hardware.
Best regards,
Cindy Ridenour
have a particularly good source in hand at the moment:
"Multidimensional Solid-State NMR and Polymers" by Klaus Schmidt-Rohr and
Hans Wolfgang Spiess; Academic Press, San Diego: 1994. Their Chapter 8 is
devoted to exchange NMR of polymers above Tg. Essentially the method for
measuring Tg by NMR is based on the temp-dependence of the correlation times
(and in essence has not changed since ca. 1955) [note that the measured Tg
is extremely dependent on heating/cooling rates]
Williams, et al, JACS 1955, v. 77, 3701
McCrum et al, Anelastic and Dieletric Effects in Polymer Solids, Wiley,
NY:1967.
Ferry, JD. Viscoelastic Properties of Polymers. Wiley, NY: 1980.
For more interesting expts, check out these refs:
"Glass Transition Dynamics in a Blend by 2D NMR." Chin, et al,
Macromolecules, 1992, v. 25, 3031.
"Dynamics above Tg: NMR..." Leisen, et al, J Chem Phys, 1992, v 97, 3749.
"Motions above Tg... 1D and 2D NMR" Pschorn, et al, Macromolecules, 1991, v
24, 398.
See also Schmidt-Rohr and Spiess, Phys Rev Lett 1991, v 66, 3020.
Schaeffer & Spiess, Macromolecules, 1990, v 23,3431.
Wefing et al J Chem Phys, 1988, v 89, 1234.
As long as you're looking at Macromolecules, see also:
1991, v 24, 6874 authors Zemke et al
1991, v 24, 338 Zhang, et al
Kelly L. Moran, Ph. D.
Applications Scientist
Solid State NMR
Cheers,
Dave Kiemle
NMR and MS Specialist
NMR and MS Laboratories
SUNY ESF
Analytical and Technical Services
Jahn Lab Rm. B18 & B19
Syracuse, NY 13210
(315)-470-6883
Received on Fri Mar 24 2000 - 14:53:55 MST