Skip to main content

beard_hallThe UNC Eshelman School of Pharmacy NMR Facility serves not only the local community, but those from outside the University as well. The interests of the users are therefore quite varied: the laboratory has been used to study small molecules and natural products; polymers; and biomolecular structure and dynamics. The Facility houses a fully automated 400MR, as well as Inova 500 and Inova 400 spectrometers.

The two 400 MHz instruments are used for routine 1H and 13C one-dimensional experiments (the 400MR can also observe  31P and 19F) and for those two-dimensional experiments often associated with small molecules (e.g., COSY, ROESY, HSQC, HMBC, etc.).

The Inova 500 is also now often used to study small molecules, but was originally designed for biomolecular NMR. It is therefore our most versatile instrument, with four channels instead of two; waveform generators on all channels; triple-axis gradients; pre-cooling of the VT air to permit low-temperature studies; and a variety of probes. It is fully capable of multi-dimensional, multi-nuclear experiments, and has recently been used to observe more “exotic” nuclei, such as 11B, 119Sn, and 129Xe.

Inova 500

  • 4 full channels
  • short 2H pulse
  • triple resonance experiments (1H, 13C, 15N, 31P)
  • relaxation experiments
  • Probes:
    • 5mm triple resonance XYZ
    • 5mm broadband PFG
    • 5mm Penta PFG
    • 3mm triple resonance PFG

Inova 400

  • 1H & 13C detect
  • 1H & 13C decouple
  • 1D experiments: APT, DEPT, INEPT…
  • 2D experiments: NOESY, COSY, DQFC, HSQC, HMBC…

400MR Spectrometer

  • 1H, 13C, 31P, 19F experiments
  • 2D experiments: NOESY, COSY, DQFC, HSQC, HMBC…
  • automated
    • SMS 100 sample changer (robot)
    • dual ProTune for automatic tuning
    • automated locking
    • gradient shimming

Fee Structure

Inova 500

  • $10/hour

Inova 400

  • $8/hour


  • $9/hour

Non-UNC Usage

  • $40/hour


  • $25/hour


  • $75/person


Inova 500

Scheduling is done at any time by the users via a UNC-based calendar system.

Inova 400

Scheduling is done at any time by the user via an UNC-based calendar system. Reserve thirty-minute blocks on Monday-Friday, 8:00 a.m.–10:00 a.m. and 3:00 p.m.–6:00 p.m. Please reserve no more than two consecutive one-hour blocks during the day. There are no time restrictions at night and on the weekends.


Available on a walk-up basis for automated usage. Please do not use more than thirty minutes at a time between 8:00 a.m. and 6:00 p.m. on Monday-Friday.

Pulse Width and Power Calculations

In order to prevent expensive damage to the probes or associated electronics, care must be excercised in setting power levels and pulse widths. For our Inova 500 spectrometer, maximum power corresponds to a setting of 63 dB. (Note that Varian follows the convention that increasing dB represents increasing power.) For 1H this represents 50 W; for 13C or 15N, however, 63 dB corresponds to an output of 300 W. Knowing this, the power can be calculated at any setting using the following equations:

dB(final) – dB(initial) = 10 log { power(final) / power(initial) }.

For example, we can calculate the power used if the 1H channel is set to 60 dB:

60 – 63 = -3 = 10 log { x/50 }

-0.3 = log { x/50 }

10 e -0.3 = antilog ( log {x/50})

50 ( 0.50 ) ~ 25 W

Since pulse widths are proportional to the voltage in the probe coil, and power is proportional to the square of the voltage (or current), we should be able to calculate the pulse width at a different power level. Since Varian uses the convention that increasing the dB increases the power, rather than attenuating the power, we have:

dB(final) – dB(initial) = 20 log { pw(initial) / pw(final) }.

For example, if the 1H pw at 51 dB is 9.4 usec, we can calculate the pw at 45 dB:

45 – 51 = -6 = 20 log { 9.4/x }

-0.3 = log { 9.4/x }

10 e -0.3 = antilog ( log { 9.4/x })

x = 18.775 ~ 18.8 usec

Thus, a change of 6 dB corresponds to a 2-fold change in pulse width, but a 4-fold change in power.

Note that for our 1H amplifier 1 W corresponds to a dB setting of ~ 46 dB, whereas 1 W would correspond to ~38 dB for our 13C / 15N amplifiers.

Gradient Mapping and Shimming Instructions

In the VNMR input window type: gmapsys

The menu buttons will change in response to this. With the left mouse button click on the one labeled: Set Params

The menu buttons will again change; click on the one labeled: Gradient,Nucleus

The buttons will change again. Depending upon whether you will make a map using 1H (which is preferred, since it has a much larger signal) or 2H, click on either: Pfg H1 or Pfg H2.

In the VNMR Status Window the following message will appear: Parameters set for pfg on H1 (or H2, depending upon which you selected above), check gradtype

When you type: gradtype? in the VNMR status window, it should respond with: gradtype=’ttt’ which corresponds to our hardware.

You might also make sure that the gradients are turned on by typing: pfgon? (normally it should respond with pfgon=’yyy’, indicating that the x,y, and z gradients are on.)

To check how many gradients will be used for mapping/shimming, type: gzsize? The default for our Inova 500 is 6; if the sample is short you may want to set gzsize=5

If you selected 2H the default settings and nt=4 can be used for mapping. If you are using 1H you should set the gain=0 and tpwr=52 (or less) so the signal is not so great that the images are distorted; nt=1 is sufficient.

Click on the button labeled: Return

The buttons will change; click on the button labeled: Go, dssh

This will result in a pair of images being produced and displayed. Type: ds(1) and then type: f full to display the first of the pair. (You may also need to type: vp=0 and then vsadj).

Click on the button labeled: Th

This will display a horizontal yellow line; with the left mouse button, position this at the top of the displayed image, and then in the VNMR input window type: th=th*0.2 The line will then be positioned at the level corresponding to 20% of the height.

With the left mouse button, position one red vertical line at the left intersection of the image and the threshold you’ve just set, and with the right mouse button position the other red vertical line at the right intersection. (Be careful not to move the left line while positioning the right.)

In the VNMR input window type: gmapsys again. The buttons will change, and again click on the one labeled: Set Params

The buttons will change; click on the one labeled: Calculate gzwin

In the VNMR Status Window a message will appear, similar to: Cursors used to set gzxwin to 29.3, tof to -769.6

Click on the button labeled: Return

The buttons will change; click on the one labeled: Shim maps

The buttons will change; click on the one labeled: Make Shimmap

In the VNMR Status Window a message similar to this will appear: Eneter mapname:[last map name]

Type in the name you wish to call this shimmap. When you hit “Enter” on the keyboard, a shimmap will be made. (When done you will see pairs of images, the number equal to gzsize + 2.)

Click on the button labeled: Return

The buttons will change; to see the shimmap, click on the menu button labeled: Display

The buttons will change; click on the one labeled: Display shimmap which will then be displayed (a set of curves equal to gzsize; the power of the polynomial corresponds to the Z shim; i.e., Z1 is a straight line, whereas Z2 is a parabola, and so forth.)

Click on the button labeled: Return

The buttons will change. Then click on the button labeled: Autoshim on Z

Gradient shimming will commence, using the map that you just made. Iterations will be performed using only gradients Z1, Z2, Z3, and Z4 until the rms error is less than one; it will then begin again using shims Z1-Z6 (or Z1-Z5, if you set gzsize=5). This is done so that Z5 and Z6, which are “weaker”, will not be “swamped” by Z1-Z4. In earlier versions of the software, one had to first type in the VNMR Input Window: gmap_z1z4=’y’ before starting Autoshimming.

When finished, clik on the button labeled: Quit

This will return you to the original experiment display from which you began.

Shimming an NMR Magnet (PDF)

Shimming Ain’t Magic (PDF)

Dr. Karl Koshlap