Jul 29, 2025
Version 2
CRAMPS V.2
- Alexander L. Paterson1
- 1National Magnetic Resonance Facility at Madison (NMRFAM), University of Wisconsin-Madison, Madison, WI, United States
Protocol Citation: Alexander L. Paterson 2025. CRAMPS. protocols.io https://dx.doi.org/10.17504/protocols.io.14egn63z6l5d/v2Version created by NMRFAM Facility
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: May 30, 2025
Last Modified: July 29, 2025
Protocol Integer ID: 219201
Keywords: Spin-1/2 CRAMPS Homonuclear Decoupling, resolution mas spectra, resolution mas spectra of system, significant homonuclear dipolar coupling, dipolar coupling, spectra, ultrafast ma, cramps purpose, significant homonuclear, resolution 1d 1h
Funders Acknowledgements:
National Science Foundation
Grant ID: 1946970
Abstract
Purpose
To obtain one-dimensional high-resolution MAS spectra of systems with broadening originating from significant homonuclear dipolar coupling, i.e., 1H and 19F.
Scope
This protocol should be followed when attempting to obtain high-resolution 1D 1H or 19F spectra when ultrafast (> 60 kHz) magic angle spinning is unavailable or impractical. This protocol can require substantial optimization and produces spectra where the chemical shifts are affected by a scaling factor; if ultrafast MAS is available, it should be seriously considered before proceeding with this protocol.
Guidelines
This protocol uses the LG4 homonuclear decoupling sequence, which may not provide the strictly optimal resolution increase in all spinning regimes. However, it should perform quite well in most spinning regimes and has a very homogeneous scaling factor, making it a robust default choice. If the resolution provided by LG4 is insufficient, consult with your facility manager about other decoupling sequences.
Homonuclear decoupling benefits strongly from rf homogeneity. “Centre-packing”, i.e., restricting the sample to the central third of the rotor, will often result in improved resolution when using this protocol. Consider the filler material carefully to avoid the inadvertent introduction of background signals.
Temperature control of the sample via VT is important to stabilize tuning and minimize sample heating from the high-power rf pulses. It is recommended even for temperature-stable samples, as long as spinning stability is not negatively impacted.
Materials
Definitions:
- MAS: Magic angle spinning
- CRAMPS: Combined rotation and multiple pulse spectroscopy
- rf: Radiofrequency
- VT: Variable temperature
- τc: Decoupling cycle length
- τr: Rotor period
Appendix:
In the case of a low-sensitivity sample, the CRAMPS optimization can be performed on a reference compound prior to running the experiment on the sample of interest. However, it is critical to not change the transmitter offset between the reference compound and the sample of interest; doing so can cause errors in chemical shift referencing.
If the sweep width is wide and the decoupling element (p10 and associated compensation pulses) is long, spurious spectra can appear in the window. These are an artifact of insufficient sampling of the FID and can be ignored.
Figure 1. 1H Hahn echo (top) and CRAMPS (bottom) spectra of alanine.
Figure 2. 1H CRAMPS spectrum of alanine during optimization. Note the wide spectral window. Only the peaks around 0 ppm are real; the rest are artifacts due to limited sampling of the FID.
Troubleshooting
Before start
This protocol directly optimizes power levels, and hence is intended for intermediate or advanced users. The maximum safe probe power must be known before using this protocol.
User is responsible for knowing the duty cycle of the probe and instrument being used.
User is responsible for knowing the maximum spinning rate of the probe and rotors being used.
This SOP has only been tested on Avance III HD spectrometers, and may not function on older systems. As of 2025-01-01, it is known to not function on Avance NEO spectrometers.
Expected amount of time SOP will use: 2 hours
Procedure
Load the sample into the probe and achieve stable spinning, temperature, and tuning.
Acquire an initial zg spectrum to ensure that the transmitter frequency, o1 or o1p, is set to a reasonable initial guess that does not overlap with any obvious peaks.
Load the dumbo.fmp pulse sequence
Run the lg4 AU program using the following initial values:
Number of points for each LG4 cycle: 120
Number of cycles in the shape file: 3
Alpha value: 70
Offset angle for the shape: 0
Set initial values:
Set the 90° pulse length, p12, and pulse power, plw12, to previously optimized values.
Set the decoupling pulse power, plw13, to 0 W.
Set the compensation pulse power, plw1, to 0 W.
Set the decoupling pulse length, p10, to the shortest allowable value. This may be between 10 µs and 20 µs, depending on your spectrometer.
Set the pre-pulse length, p14, to 0.1 µs.
Set the first compensation pulse length, p22, and second compensation pulse length, p21, to 0.1 µs.
Set p8, an additional acquisition delay period, to 0.
Set the number of acquisition points, l11, to 16.
Set the phase constants (cnst21, cnst22, cnst25) to 0.
Set cnst31 to the MAS spinning rate in Hz. This is only used for calculating ratios, not for pulse or delay timings.
Ensure that the acquisition time, aq, is within the probe safety and duty cycle limit. An initial value of 10 ms is recommended.
Acquire an initial CRAMPS spectrum with the above initial guesses. Ensure that the entire spectrum is comfortably within the spectral window; the peak positions may change as parameters are optimized. This spectrum should not be particularly different than the initial one-pulse experiment.
Enable decoupling by setting plw13 to a low value, e.g., 10 W. Save this processed spectrum using wrp and refer back to it if comparisons are needed to evaluate performance.
Optimize the transmitter offset, o1, in coarse steps (e.g., 500 Hz for 1H) to avoid overlap with resonances of interest.
Note
This will likely be on the right side of the spectrum, close to but not overlapping with the rightmost peak.
Ensure that the axial peak is sufficiently suppressed. This step can be repeated any time the axial peak exceeds the intensity of the most intense real peak.
Begin by optimizing the pre-pulse length, p14, from 0.1 µs to 1.5 µs in steps of 0.25 µs
Then optimize the pre-pulse phase, cnst25, from 0° to 360° in steps of 30°.
If necessary, repeat the optimization using smaller step sizes and a smaller search range.
Any time the axial peak is more intense than the most intense real peak, the pre-pulse phase and pre-pulse length should be refined.
Iteratively optimize the decoupling elements. This may require several loops.
Coarsely optimize the decoupling pulse length, p10, from as short as allowed to 100 µs in steps of 5 µs. The optimum pulse length will result in maximum resolution.
Note
Be cautious of overly-long pulse lengths; serious artifacts can result.
Select the pulse length range that provides the best resolution. Finely optimize p10 over this range in steps of 1 µs, again selecting best resolution.
Note that if the decoupling cycle time (τc = p9 + p10) synchronizes with the rotor period (such that τc/τr = n/2), resolution will be strongly reduced.
Once p10 is optimized, optimize the decoupling pulse power, plw13, from 1 W to the probe maximum power in steps of approximately 10% of the probe maximum power.
Safety information
Be absolutely sure of the probe maximum power before proceeding.
Select the pulse power range that provides the best resolution. Finely optimize the plw13 over this range in steps of approximately 5% of the probe maximum power.
Optimize the transmitter offset in fine steps (e.g., 200 Hz for 1H).
Return to step 9 and repeat all optimization steps with smaller step sizes. Continue until a satisfactory spectral resolution is found.
Optimize the acquisition window elements.
Adjust the number of acquired complex points, l11, from 8 to 32 in steps of 2.
Adjust the receiver delay, p8, from 0 µs to 1 µs in steps of 0.1 µs.
Optimize the compensation pulse phases. First set the compensation pulse power, plw1, to the same value as plw13, and the compensation pulse lengths, p22 and p21, to 0.3 µs.
Optimize the phase adjustment of the second compensation pulse, cnst21, from 0° to 360° in steps of 5°. Optimize for spectral resolution and not the behaviour of the axial peak.
Optimize the phase adjustment of the first compensation pulse, cnst22, from 0° to 360° in steps of 5°. Optimize for spectral resolution and not the behaviour of the axial peak.
Optimize the compensation pulse lengths.
Optimize the length of the second compensation pulse, p21, from 0.1 µs to 1.2 µs in steps of 0.1 µs.
Optimize the length of the first compensation pulse, p22, from 0.1 µs to 1.2 µs in steps of 0.1 µs.
Iterate between these two steps, reducing the step size and search range each time, until optimal resolution is achieved.
Fine-tune the pre-pulse length, p14, and phase, cnst25, over a narrow range with small step sizes, until the axial peak is optimally suppressed.
Acquire a high-quality spectrum with sufficient S/N.
Correct for the chemical shift scaling factor.
If the MAS spectrum has two sufficiently resolved peaks:
- Compare the distance between the two peaks in the MAS spectrum with the distance between the two peaks in the CRAMPS spectrum to obtain the scaling factor.
- Divide the acquisition spectral width (s swh) by the scaling factor and Fourier transform the FID.
- Calibrate the chemical shift of the CRAMPS spectrum using the chemical shifts of the MAS spectrum
- If the MAS spectrum does not have sufficiently resolved peaks, acquire a CRAMPS spectrum of a reference compound (e.g., α-glycine) using exactly the same parameters as the sample of interest.
Note
Exactly is not an understatement. Do not adjust the probe tuning; if the reference compound requires the tuning to be adjusted, choose a different reference compound.
Protocol references
Paruzzo, F. M.; Emsley, L. High-resolution 1H NMR of powdered solids by homonuclear dipolar decoupling. J Magn Reson 2019, 309, 106598. DOI: 10.1016/j.jmr.2019.106598
Coelho, C.; Rocha, J.; Madhu, P. K.; Mafra, L. Practical aspects of Lee-Goldburg based CRAMPS techniques for high-resolution 1H NMR spectroscopy in solids: implementation and applications. J Magn Reson 2008,194 (2), 264-282. DOI: 10.1016/j.jmr.2008.07.019
Example Data at 900 MHz: usnan.org/ark:/83454/e1aba32d00-eae4-4f2f-9290-da5b2a6ffb30
Example Data at 600 MHz: usnan.org/ark:/83454/e156f805af-438c-4f36-b011-25becc7eff58