Outline
of Shim Assembly (inside the magnet's room temperature bore), Probe with
V.T. Heater, sample and gas flow arrangement. Items are not to scale.Rudi Nunlist, March 3, 1994
The Variable Temperature Unit (V.T.) is used to set the sample's temperature for an nmr experiment. A temperature stabilized gas flows past the sample. Either air or nitrogen gas enters the bottom of the probe. A heater inside the probe heats the gas to the desired temperature. For high temperature operation, regular air is used. For low temperature N2 is used since the moisture in regular air would freeze.
The temperature is measured with a copper-constantan thermocouple positioned about 1 mm below the bottom of the sample tube. The control unit regulates the heater current to achieve stable temperature. The gas exits through the topof the probe.
The temperature control typically ranges from - 120º C to + 130º C. The exact limits depend on the probe's ability to sustain the temperature without damage.
The heater's ability to provide enough heat may further limit the upper range. The insulation and thermal transfer efficiency of the normally used N2 cooling circuit usually determines the low limit.
Additional limitations may be imposed by the sample spinning mechanism; spinning might become difficult, or impossible, beyond a given temperature.
V.T. units have controls for the temperature selection, the gas flow and the maximum heater power.
The main power is usually left on. The gas flow should be off when not used (instrument idle).
The gas flow is adjusted and indicated with a valve and flow meter assembly. The flow meter reads in units of 100 liters/hour. Read the bottom of the float.
The heater power is turned on with the heater power dial. The flow meter light is on when the heater power is on.. If there is no gas flow at the outlet of the unit, the light in the flow meter blinks and the heater power is switched off.
Note: If the outlet is accidentally disconnected from the probe, the unit does not turn the heater off!
The temperature is selected with the up/down arrows and the set point (NOT the actual temperature) is displayed in Kelvin. The analog Error Meter only indicates the deviation from the set point over about a 1 range.
We added an external digital thermometer to read the temperature at the thermocouple. This is a better indicator than the Error Meter.
The Bruker unit additionally has a control for the evaporator power used for low temperature operation. A red LED blinks if the nitrogen Dewar is empty.
The software temperature setting is disabled. It serves no real purpose; the data acquisition system cannot check if the requested temperature is ever reached!
Martin, Delpuech and Martin discuss several aspects of variable temperature experiments in their book "Practical NMR Spectroscopy" (Heyden & Son), page 336.
Please read the remainder of this before attempting V.T. experiments - you will learn more about how the system works and the potential errors affecting the results.
Please realize that it is your responsibility to schedule enough instrument time so that the probe can be returned to room temperature, and stay at room temperature without gas flow, before the next person uses the spectrometer.
Keeping the gas flow (and possibly the heater) on is not an acceptable short cut; the next person cannot be expected to deal with sample spinning problems or, as has nearly happened, lose a precious sample because it unexpectedly froze (or boiled) as the probe was not near room temperature. Some experiments, especially 2D, can be ruined it the temperature is not stable throughout the course of the experiment.
Bringing the probe back to room temperature may well take more than
30 minutes (depending on the final temperature the probe was at). If the
probe was at a very low temperature for a long time, (for example - 80º
for 2 hours, or longer) it may be best to bring the probe to about 50º
for some 10 to 20 minutes to speed up warming of the probe. Then keep the
gas flow on to cool the probe to room temperature.
Watch the temperature for a least another 10 minutes after reaching room
temperature - it may still drift!
The probe, and the parts surrounding it, represent a considerable thermal mass; therefore the longer the experiment, the probe was takes longer to return to ambient.
Be sure to fill out the log book, noting the temperature extremes, please!
Please be careful when handling cold and iced up connections, it is quite easy to damage glass parts by using undue mechanical force. Avoid dropping glass joints on the floor when removing the gas tubing, etc.
The ceramic spinners shatter if they are dropped. The replacement price is around $400.00! You break it - you buy it!
This is a very common arrangement.
Outline
of Shim Assembly (inside the magnet's room temperature bore), Probe with
V.T. Heater, sample and gas flow arrangement. Items are not to scale.
The spinning seems to be most reliable at relatively high rates, i.e., above 25 RPS.
In extreme situations, it might be necessary to attach a small air pump to the eject line to further increase to outflow.
During prolonged experiments at low temperatures there will be a build-up of ice at the bottom of the probe and shim assembly.
It is important to periodically check the bottom plate of the Magnet Dewar - if IT should start to ice up, the vacuum O-ring seal will leak, causing a loss of vacuum and quench the magnet! Immediately stop the experiment, use a heat gun and warm the area! Immediately inform Facility personnel!
The general operating procedures are similar for high and low temperature experiments. The first thing to do is to check that the V.T. gas connections are in place. Operating without gas flow can destroy the heater and/or the probe!
Think about the temperature you want to operate at. Be sure that your sample can withstand the temperature without freezing or boiling.
Sealed samples can explode if the pressure build-up is higher than anticipated!
Use a ceramic spinner to reduce spinning problems.
To operate at high temperature, regular air is used. The V.T. gas hose is directly connected to the probe gas inlet. The flow rate should be at least 200 l/hr.
With the gas flow on, set the heater power to the appropriate setting. Increase the set point 10 at the time until the desired temperature is reached. Readjust the heater power as needed.
After the experiment is completed, turn the heater power off. Return the set point to room temperature. Depending on the temperature last operated at, the probe may take as long as 30 to 45 minutes to cool to room temperature. The cooling can be accelerated up by removing the sample and using maximum V.T. air flow. Once the probe stays at room temperature without flow the gas flow can be turned off. Keep testing by temporarily turning the flow off until it does.
If this is not done, the next sample may reach surprisingly high temperatures!
Suggested Heater Power Settings: (If you find these to be wrong, PLEASE LET ME KNOW!!)
Heater Dial I max.(%) Probe Temp. ºC
5 5-10 30
8 40 40
9 55 80
Note: There must be no moisture in any of the parts and
assemblies where the gas flow is below freezing!
Otherwise ice will form and block the gas flow.
Low temperature operation requires the use of liquid Nitrogen. We have two ways of setting up the system. Bruker supplies an evaporator immersed into a liquid Nitrogen Dewar to generate cold gas. This is advantage for places where house nitrogen is not available. The disadvantage is that the evaporator power determines the gas flow rate. The regulation characteristics are affected by the flow rate. As a result, it might become difficult (if not impossible) to achieve good regulation at all temperatures.
The systems we built flow N2 gas through
a liquid Nitrogen Dewar. For very low temperatures, (from about - 30º
C to - 100º C) the gas flows through the liquid. For medium temperatures,
(about +25º C to - 30º C) the gas flows over the liquid. This
way, the flow rate can be much better controlled and the regulation is
better at all temperatures.
Low temperature operation is similar to high temperature except for the complications introduced by the use of lN2. In addition, sample spinning and ejection need nitrogen gas if the temperature is near or below 0 - water in the air would condense and freeze!
To set up the system, fill the lN2 Dewar. Slowly insert the evaporator into the Dewar. After the initial high pressure has subsided, remove the V.T. gas line from the probe inlet and connect the evaporator output with the short piece of hose with glass joints between the probe and the evaporator to minimize mechanical stress on the probe's heater Dewar. Select the desired temperature. Adjust evaporator and heater power.
Bruker Specifications:
Operating Time: with 10 liters liquid nitrogen (our Dewars are 25 l):
Temperature: Evaporator Time
Dial Position
100 deg.K Pos 10 2h 30
123 deg.K Pos 8 3h 30
173 deg.K Pos 6 10h
223 deg.K Pos 3 20h
273 deg.K Pos 3 20h
Evaporator Power and Flow Rate:
Power versus graduation table
Position Power (W) N2 Flow (l/hr)<
0 0 0
1 0 0
2 0.5 7
3 1 30
4 2 70
5 15 350
6 20 450
7 45 700
8 105 1050
9 130 1200
10 150 1300
When operating above 273º K, heating power should be set at minimum in order to economize on liquid nitrogen.
To set up the system, fill the lN2 Dewar. Insert the low temperature adapter. Open the quick disconnect couplers in the V.T. gas line near the probe. Attach the lines to the low temperature adapter. The input selects the low or medium range. Attach the outlet to the probe.
Performance Tests: (AM-500, 5 mm Inverse C-H Probe. July 22, 1992)
Ceramic spinner. V.T. plug on top of the shim assembly, needle valve closed. On the AM-500, spinning works best if the rate is set >= 27 RPS !
Lowest Temperatures Attained:
- Low Temperature Inlet: -102º C (lowest temperature tested)
- Medium Temperature Inlet: -41 C (lowest temperature attempted)
- Time to Temperature: (N2 Gas flow of 400 liters/hr)
Medium Range: +27 to - 36: 10 minutes.
Low Range: -43 to - 60: 2 minutes.
to - 80: 5 minutes.
to -100: 16 minutes.
- Time to Return to Higher Temperature: (N2 Gas Flow 300 l/h)
Medium Range: -100 to -10: 8 minutes
-10 to +20: 3 minutes
Heater Settings on AM-500:
Nitrogen Flow = 300 l/hr, N2 regulator set to 5 psig or greater.
Range Heater Dial I max. (%) Probe Temp º C
Low 5 5-10 -100
Medium 8 40 -10
Medium 9 50 18
Medium 9 55 24
Sample spinning often becomes difficult towards the temperature limits. The ceramic spinner should be used. Since it is heavier than the plastic ones, it does not lift as easily. Install the V.T. plug and close the needle valve. Set the spin rate to 25 RPS or higher. If the temperature is above 0 , air can be used. For lower temperatures, N2 must be used to prevent freezing of moisture in the air.
The probe tuning is temperature dependent. The probes are normally tuned at room temperature. If S/N is a problem, the probe should be re-tuned at the desired temperature. If the pulse width is critical, (Dept, many , but not all, 2D experiments), the 90º pulse(s) should be measured and the values corrected.
Probe tuning can become difficult, or impossible, at low temperatures. This is probably due to moisture freezing at the capacitors. This can problem can be overcome by carefully tuning the probe as the temperature is decreasing. Exercising the capacitors minimizes the freezing problems.
Since QNP probes are very difficult to tune, users should not attempt to do this on their own. If QNP probe tuning is necessary, consult the Facility staff, please!
Ejecting samples with the regular compressed air might not work very well. Since the shim assembly is probably quite cold, the moisture condenses. Eventually, the spinner air bearing and the probe itself will collect water. Finally, the water will probably freeze.
The eject lines on our systems are connected to an external N2 cylinder. Turn the eject selector valve to "External". Before opening the main valve on the gas cylinder, turn the regulator all the way counterclockwise (no pressure). Activate the sample eject. Now turn on the main valve. Now adjust the regulator until there is sufficient pressure for the sample to ascend.
If the pressure is set without the sample eject activated, it is possible to build up too much pressure. Since there is a reservoir, the pressure will not dissipate and, the moment the sample eject is activated, the sample will exit the probe rather violently. The sample will leave the magnet bore and land on the floor. One sample once even hit the ceiling! Both the ceramic spinner and the sample were destroyed!
If a sample must stay cold, the probe can be pre-cooled without a sample. The spinner must be at room temperature, otherwise it will collect moisture while exposed to air. This usually leads to spinning problems. If more than one sample needs to be run, use the heat gun to warm the spinner before using it again.
During "high power" decoupling experiments, the sample absorbs
rf power which leads to considerable heating.The heating depends on the
electrical characteristics of the sample itself (dielectric losses, conductivity)
and is worst in samples such as water solutions. The heating effect increases
with frequency. For a given spectrometer, carbon decoupling causes less
heating than proton decoupling with the same amount of power.
This heating can cause an error because the V.T. unit measures the temperature
of the gas before it reaches the sample. As now the sample itself is a
heater, this extra temperature increase is not easily accounted for. Obviously,
the higher the gas flow, the more heat will be removed, thus reducing the
error.
The problem of temperature calibration is an old one. Several papers have been written about this subject. The generally accepted method is to use a sample with known chemical shift vs. temperature. Ethylene glycol is used for high temperature, methanol (with "a trace" of HCl to sharpen the lines) is used for low temperature. These samples were calibrated using a variety of other methods, including melting point measurements.
The following papers contain useful discussions. Amman's paper seems to have the most reliable calibration data.
Amman, Meier, Merbach. J. Magn. Reson. 46, 319-321 (1982)
Van Geet, Anal. Chem. 42, 679 (1970)
Becker, Anal Chem. 51, 2050 (1979)
Levy. J. Magn. Reson. 37, 353 (1980)
Dickert, Anal Chem. 52, 996 (1980)
Friebolin, Org. Magn Reson. 12, 569 (1979)
Errors are possible because:
For these reasons, it is recommended that the temperature is measured with one of the standards before placing much faith in the readings. Be sure to use the same conditions (e.g., gas flow, spinning rate) for the calibration and the measurement.
The standard sample should be checked at room temperature. If it indicates a large difference between the expected temperature and the one calculated from the chemical shift it may well be that the standard is at fault.
One person once prepared an ethylene glycol sample and carefully dried it. This resulted in an impossible calculated temperature. The sample does need a trace amount of water; the literature specifies the sample to be "fresh out of the bottle". Sealed samples will keep.
Another person found problems with a sealed capillary methanol sample, having been exposed to +50º C gave impossible results as well.
To run a calibration, use a temperature standard appropriate for the range you are interested in. With a normal sample, locking will not be possible. With the high sensitivity of our probes the lines will be broadened (radiation damping). It might be preferable to prepare a sealed capillary for use in a regular 5 mm tube. Then a deuterated solvent can be used and the proton signal will be less intense.
The AM-400 and 500 have a program, TEMPCALC, which can be used to generate
a list of chemical shift vs. temperature. The PC "helpers"
have a program, VT-CALC, which does the same.
On bloch or purcell, use the "vtcalc"program.
Click Here to return to the NMR Facility
Home Page.
Click Here to back up one step.
Last Update: 9/25/96 RN.