NMR and Food Chemistry - Popcorn

Both liquid state and solid state NMR have become very important tools in the food industry for both research and quality control. Since starch is a very important biopolymer and a major constituent in many foods, a great deal of work has been done on its chemical and physical characterization. Starches are 1 -4 linked polymers of glucose. Native starches are a mixture of a linear polymer (amylose) and a branched polymer (amylopectin). Furthermore, the starches are of two main types differing in their crystal structures and water content: the A type (cereal starches) and the B type (tuber starches). In addition to A or B starches, the starch granules in plants can have some amorphous starch as well. Corn starch is of the A type.

Popcorn is a snack enjoyed by millions around the world. As a child, I remember being fascinated at watching it pop and wondering what was going on. Much study has been devoted to the physics of popping corn. Essentially, every kernel of corn is a pressure vessel. When cooked, the moisture trapped within the starchy endosperm of the kernel is superheated. When the steam pressure inside the kernel becomes high enough the hull (pericarp) of the kernel explodes and the superheated water in the starch granules suddenly vaporizes, expands and rapidly cools making a solid foam out of the starchy endosperm. Each starch granule is a bubble in the solid foam.

Although much work has been done to understand starch and much work has been done to understand the popping of corn, I was unable to find any efforts directed to the chemical changes in corn starch before and after popping. This prompted me to collect a few spectra to address the issue. The first figure below shows the starch region of the ¹³C CPMAS spectra of unpopped corn (bottom trace), popped corn (middle trace) and cooked but unpopped corn (upper trace). The spectrum of unpopped corn is a mixture of A type starch (with a characteristic three line pattern in the C1 region) and amorphous starch (with a broad distribution of overlapping lines in the C1 region). The spectra of popped corn and cooked but unpopped corn are essentially identical and characteristic of amorphous starch. The data indicate that the heating and dehydration of the corn transform the crystalline A starch into amorphous starch. The observation is consistent with published studies on the hydration of starches. Aside from this change, there is no evidence for any other chemical transformation in the starch.

The second figure shows an expansion of the C1 region highlighting the conversion of crystalline A type starch into amorphous starch.

Probe Arcing

Probe arcing can occur during the application of an rf pulse. It is the passage of a spark between a localized area of high voltage inside the probe to ground. In cases of severe probe arcing, one can hear a "snap" during the application of a pulse. If the probe is removed from the magnet and the cover is removed, one can see the arcing as a small "bolt of lightning" between the high voltage area and ground. When a probe arcs, the integrity of the pulse is affected and one will observe FIDs of irreproducible amplitude and phase. Needless to say, the quality of data collected on an arcing probe will be severely compromised. If a probe is permitted to arc for extended periods of time, some of the electronic components inside the probe can be permanently damaged. The amplifiers can also be damaged as arcing will cause mismatching and high levels of reflected power. Here are a few things you can do to help stop probe arcing.

1. Use lower pulse power levels.
2. If the location of the arcing can be found visually (i.e. you can see the spark), then the high voltage area and the path to ground can be wrapped with teflon tape.
3. Round off any sharp edges inside the probe as these are the areas of highest local voltage. In particular, any solder joints around the coil should be smooth.
4. Keep the coil and capacitors as far away from ground as possible.
5. Purge the inside of the probe body continuously with nitrogen gas.

Field Dependence of Chemical Shift Resolution in Solids

As undergraduates we learn that the chemical shift resolution in an NMR spectrum increases with increasing field strength. This is true because the number of Hz in a ppm increases with field and inequivalent resonances are more separated (in Hz) at higher field.

Will one always get better chemical shift resolution at higher field? You may be surprised to learn that the answer is - no. In certain cases for liquids, where dynamic processes are occuring at a rate comparable to the frequency difference between resonances (i.e. on the NMR time scale) one may even get less resolution at higher fields due to line width changes at higher field.

In the MAS or CPMAS spectra of solids one obtains liquid-like spectra of solid materials. In the cases where dipolar coupling is effectively removed by either or both MAS and high power decoupling, the width of the resonances often depends on a chemical shift distribution associated with a particular site. Just like chemical shielding anisotropy, this distribution increases (in Hz) with the magnetic field strength. As a result, the chemical shift resolution does not improve on going to higher field as the widths of the MAS lines increase (in Hz) as a function of field. Two examples of this are shown below. The first figure is the ²⁹Si CPMAS spectrum of the clay kaolinite at 11.7 T (top trace) and 21.1 T (bottom trace). Both spectra were acquired with MAS rates exceeding the span of the chemical shift tensors. One can see that there is little if any improvement in the chemical shift resolution at higher field.

The second figure is the ¹⁹F MAS spectrum of the perfluorinated polymer, Nafion at 11.7 T (top trace) and 21.1 T (bottom trace). The MAS rates for each spectrum were chosen such that the dipolar coupling between the fluorines was effectively averaged by the MAS and that the spinning sidebands would be coincident (in ppm) in the spectra. Again, one can see that there is little if any improvement in the chemical shift resolution at higher field.

Thank you to Victor Terskikh and Eric Ye of the National Ultrahigh-Field NMR Facility for Solids for providing the figures for this post.

The 500 MHz 1H NMR Spectrum of Butane

n-Butane is a very simple molecule. Should it not then give a very simple ¹H NMR spectrum? The figure below shows two calculated 500 MHz proton spectra for n-butane. Which spectrum most closely represents the true spectrum of n-butane?The answer is the very complicated spectrum B. The spectra were calculated with the following parameters:

Spectrum B
Spectrometer frequency = 500 MHz
δ1 = δ4 = 0.333 ppm
δ2 = δ3 = 1.271 ppm
³J₁₂ = ³J₃₄ = 7.11 Hz
⁴J₁₃ = ⁴J₂₄ = -0.07 Hz
³J₂₃ = 6.77 Hz
LB = 0.5 Hz

Spectrum A

Spectrometer frequency = 500 MHz
δ1 = δ4 = 0.333 ppm
δ2 = δ3 = 1.271 ppm
³J₁₂ = ³J₃₄ = 7.11 Hz
⁴J₁₃ = ⁴J₂₄ = -0.07 Hz
³J₂₃ = 0 Hz

LB = 0.5 Hz

The only difference between the simulations is that in spectrum B a coupling of 6.77 Hz was assumed between the two methylene groups whereas in spectrum A the same coupling was taken to be zero. The reason spectrum spectrum B is so complicated is that despite the fact that both the methyl groups and both the methylene groups are chemically equivalent, they are not magnetically equivalent. This is true for both spectrum A and spectrum B however, in spectrum A the second order effects are small based on the parameters used in the simulation.

Thank you to Adrian Dingle for inspiring me to create this post.

Distortions in QCPMG Spectra from Pulse Miscalibrations

A standard QCPMG NMR pulse sequence consists of a 90^o pulse followed by a train of 180^o pulses. Ideally, the resulting spikelet envelope should outline the static lineshape from a conventiaonal Hahn-echo experiment. If the first pulse deviates from 90^o due to incorrect calibration, the QCPMG spikelet pattern does not change significantly, the only effect is somewhat lower overall intensity (first figure). At the same time the miscalibrated subsequent pulses lead to significantly distorted spikelet patterns (second figure). The 180^o pulse misset by as little as 20^o-30^o, could produce considerable oscillations in the spikelet intensity across the envelope. This illustrates that QCPMG NMR experiments are much more sensitive to proper setup of the 180^o pulses than the Hahn-echo experiment. The QCPMG spectra shown were calculated in SIMPSON for a central transition of a spin 3/2 nucleus resonating at 295 MHz (⁸⁷Rb at 21.1 T), C_Q=10 MHz, η_Q=0.7, CS anisotropy= -200 ppm, coincidental EFG and CSA tensors, ω_RF/2π= 200 kHz.

Many thanks to Eric Ye of the National Ultra-high Field NMR Facility for Solids for contributing this post.

13C NMR with 1H and 31P Decoupling

NMR users are very familiar with the advantages of proton decoupling when observing ¹³C. The ¹³C NMR spectra of phosphorus containing compounds can be made simpler by applying ³¹P decoupling either on its own or in addition to proton decoupling. The figure below shows the ¹³C NMR spectrum of dimethyl methylphosphonate with all possible combinations of proton and ³¹P decoupling. The data collection required a triple resonance probe with the appropriate band pass filters.

31P - 13C HMQC

When most people think about HMQC or HSQC spectra they think about protons and ¹³C or protons and ¹⁵N. Although these are by far the most common spins to probe, the HMQC technique can be applied to other spin pairs as well. In earlier posts to this BLOG, this was demonstrated for protons and ¹¹B and ³¹P and ¹⁰⁹Ag. This post shows an application of the technique to ³¹P and ¹³C. Measurements made between spin pairs, where one of the spins is not a proton, require a triple resonance probe and the appropriate filters.

The figure below shows ³¹P detected ³¹P - ¹³C HMQC data for dimethyl methylphosphonate with ¹³C decoupling during the acquisition and ¹H decoupling during the entire sequence. The spectra on the top and left-hand sides of the plots are separately run one dimensional ³¹P and ¹³C spectra, respectively, both with proton decoupling. The methyl carbon of dimethyl methylphosphonate has ¹J_CP = 142 Hz whereas the methoxy carbons have ²J_CP = 6 Hz. Two HMQC spectra were run. The one on the left was optimized for 142 Hz and the one on the right was optimized for 6 Hz. Each spectrum shows a correlation according to its coupling constant. The data were collected on a 500 MHz instrument with the appropriate bandpass filters on each channel.

90 Degree Pulse Determinations

In the routine procedure of R.F. pulse optimization, distorted nutation curves can sometimes be observed, as can be seen in the calculated graph below. This is usually due to the short recycle delay time (D1) not sufficient for complete relaxation. With increasingly shorter delay time, the maximum of the nutation curve shifts to lower flip angle, which makes the curve asymmetric. Under this circumstance, the 90 degree pulse determined from the maximum of the intensity will be inaccurate. On the other hand, if the recycle delay time is not too short, say 1 to 2 times of T1, the length of 180 degree pulse can still be determined with reasonable accuracy.

Many thanks to Eric Ye of the National Ultra-high Field NMR Facility for Solids for contributing this post.

The Selective 1D Gradient NOESY

Many students run very long 2D NMR experiments to find one particular piece of information. This is especially true for 2D NOESY experiments where perhaps only a single NOE correlation is sought. Depending on the amount of compound available, 2D NOESY measurements often take hours or tens of hours to acquire and can cost alot of money in instrument user fees. Many people do not realize that there are selective 1D analogs to the non selective 2D experiments. These 1D experiments rely on shaped pulses for selective excitation and take only a small fraction of the time required to run the comparable non selective 2D experiment. Previously I have posted entries in this BLOG on the very useful 1D selective gradient TOCSY experiment. The figure below demonstrates the use of the 1D selective gradient NOESY experiment for bis(phenylthio)methane compared to the standard proton NMR spectrum. In this case, the methylene protons were selectively irradiated and the NOE's were observed on the phenyl rings. The experiment took less than 5 minutes to acquire.

The Background from a Dirty NMR Probe

Have you ever wondered why the manager of your NMR equipment gets on your case about wiping down NMR tubes before putting them in the NMR probe? The figure below should answer your question. It shows the ¹H NMR spectrum of a clean empty NMR tube inside an NMR probe before and after the NMR probe has been cleaned.

Over time, the "stuff" from your hands and residue on the outside of your NMR tubes builds up on the inside of the inserts inside the coil of the NMR probe. This "stuff" contains protons and results in a background signal in all subsequent NMR spectra. I have seen inserts of NMR probes so dirty that samples have "stuck" inside the probe. The offending gunk resembles grey-black bubble gum or perhaps dirty dried nasal effluent. Periodically NMR probes must be cleaned to remove this offensive residue. This can usually be accomplished by gently inserting and removing a cotton swab soaked in alcohol inside the coil insert. For the particular case in the figure above, the NMR probe could not be cleaned in this way and had to be disassembled by a service engineer and cleaned in an ultrasonic bath.

WIPE YOUR NMR TUBES BEFORE PUTTING THEM IN AN NMR PROBE !!

Thank you to Dr. Michael Lumsden who manages the NMR Facility at Dallousie University for suggesting this post, kindly sharing his tale and providing the figure.

Spin Echos for Uncoupled Spins

The spin echo is one of the most fundamental building blocks for NMR pulse sequences. Its main purpose is to refocus chemical shifts. The simplest spin echo is that for uncoupled spins where only the offset, Ω (i.e. the frequency difference between the carrier and the resonance) need be considered. The pulse sequence is represented in the upper portion of the figure with the vector and product operator representations below. A 90_x pulse is first given to create magnetization along the -y axis of the rotating frame. During the first delay period, τ, the magnetization rotates in the x-y plane at a rate, Ω. The 180_x pulse rotates the magnetization 180 degrees about the x axis. During the second delay period, the magnetization again rotates in the x-y plane at a rate, Ω in the same direction as during the first delay. At the end of the second delay, the magnetization is on the y axis and the collection of the FID is started. It is important to note that the echo will always have its maximum at 2τ after the 90 degree pulse regardless of its offset, Ω or the duration of τ. The value of τ however is limited by the T₂.

What is T1ρ and How is it Measured?

The time constant for the build up of magnetization along the direction of the main magnetic field, B_o, (the z axis) either after a pulse or upon initially exposing a sample to the magnetic field is called the T₁ relaxation time or spin-lattice relaxation time. It is this relaxation time which determines the rate at which a pulse sequence can be repeated. The time constant for the decay of magnetization in the x-y plane of the rotating frame of reference after a pulse is called the T₂ relaxation time, the spin-spin relaxation time or the transverse relaxation time. It is this relaxation time which determines the natural line width of a particular resonance. There is another relaxation time constant of interest to NMR spectroscopists - T₁ρ. T₁ρ is the time constant for the decay of magnetization along the radio frequency field, B₁, of an applied spin locking pulse in the rotating frame of reference. It is analogous to T₁ except it describes relaxation along the radio frequency field of the pulse (which is static in the rotating frame) rather than relaxation along B_o. T₁ρ's are of interest in ROESY, TOCSY and cross polarization experiments. The T₁ρ is measured by first applying a 90 degree pulse to an equilibrium magnetization vector. A spin locking pulse is then applied. The phase of this pulse is shifted 90 degrees with respect to the excitation pulse such that the field of the spin locking pulse is coincident with the spin vector in the rotating frame of reference. During the spin locking pulse, the large magnetization vector (which was initially polarized in B_o) decays to its equilibrium value in the much smaller field, B₁, with time constant, T₁ρ. The T₁ρ is measured by analysing the intensity of the NMR signal in spectra collected as a function of the duration of the spin locking pulse. This is illustrated in the figure below.

Ottawa U Takes Delivery of a Bruker AVANCE III 400

The Bruker AVANCE III 400 NMR spectrometer for solids was delivered to Ottawa U last week and is currently being installed. It will be available for use in February. With the addition of this instrument, the Ottawa U campus boasts 7 NMR instruments (8, including the Bruker AVANCE II 900 on the NRC campus ). For some photos of the installation, follow this link.

Weak One-bond or Multiple Bond Correlations in 1H / 13C HMQC / HSQC Spectra

Many people are quite surprised to see either unusually weak one-bond correlations or weak multiple bond correlations in their ¹H / ¹³C HMQC / HSQC spectra. These people must be reminded that there is nothing "magic" about these experiments - the responses are based solely on an assigned delay proportional a reciprocal coupling constant. The large scale success of the ¹H / ¹³C HMQC / HSQC techniques can be attributed to the fact that most one-bond ¹H - ¹³C coupling constants are very similar ( ~ 145 Hz). The pulse sequences are therefore run with a delay based on a 145 Hz coupling constant. When one-bond coupling constants are significantly different than 145 Hz then the correlation will be either very weak or absent in the spectrum. Also, if multiple bond couplings are unusually large then those multiple bond correlations may be present in the spectrum. The figure below is an example. In the 500 MHz HMQC spectrum of an alkyne (optimized for 145 Hz coupling), one can see an unusually small one-bond correlation between the terminal alkyne proton and its attached carbon. There is also a weak two-bond correlation between the terminal alkyne proton and the other alkyne carbon.

The Effect of Magic Angle Spinning and High Power 1H Decoupling on 13C Cross Polarization NMR Experiments

Cross polarization (CP), magic angle spinning (MAS) and high power ¹H decoupling are all routine methods used in solid state NMR experiments. It is useful to see the effect of each of these techniques on a solid sample. The figure below shows ¹³C cross polarization NMR spectra of glycine at 4.7 Tesla collected with various combinations of magic angle spinning and high power ¹H decoupling.The bottom spectrum was collected with neither MAS nor high power ¹H decoupling. One can see two very broad overlapping lines due to the carbonyl and methylene carbons. The broadening is due to chemical shielding anisotropy and heteronuclear dipolar coupling between the ¹³C and both ¹H and ¹⁴N. The second trace from the bottom was collected with high power ¹H decoupling but no magic angle spinning. The spectrum contains two broad resonances with very informative line shapes. The high power ¹H decoupling effectively removes the ¹³C - ¹H heteronuclear dipolar interaction. The line shapes are determined from the chemical shielding anisotropy and ¹³C - ¹⁴N dipolar coupling interactions. The second trace from the top was collected with magic angle spinning at 4.5 kHz but no high power ¹H decoupling. The spectrum apparently contains only one broad resonance with spinning sidebands. The magic angle spinning effectively removes the ¹³C chemical shielding anisotropy interaction. Although MAS does help average the ¹³C - ¹H heteronuclear dipolar interaction, the averaging is not very effective at a speed of 4.5 kHz. Also, MAS only partially averages the ¹³C - ¹⁴N heteronuclear dipolar interaction. The resonances are therefore broadened out by residual heteronuclear dipolar coupling. The methylene resonance is broadened to such an extent that it does not show up in the spectrum at all. The top spectrum was collected with both MAS and high power ¹H decoupling. One can see two very sharp resonances due to the carbonyl and methylene carbons. The ¹³C chemical shielding anisotropy and ¹³C - ¹H heteronuclear dipolar coupling interactions are effectively removed by the MAS and high power ¹H decoupling, respectively. Since MAS does not average J coupling and only partially averages dipolar coupling between a spin I = 1/2 and quadrupolar nucleus, the methylene carbon shows fine structure due to both J coupling and residual ¹³C - ¹⁴N dipolar coupling (see inset in yellow).

The BIRD Filter

Many modern NMR experiments exploit coupling interactions between protons and heteronuclei (eg. ¹³C). In such sequences the goal is to selectively observe the protons bound to ¹³C and suppress those bound to ¹²C. Since ¹³C is only 1 % naturally abundant, this means that 99% of the signal must be suppressed. One particularly simple scheme to accomplish this is the BIRD (BIlinear Rotation Decoupling) filter. The BIRD filter uses a heteroneuclear spin echo with delays equal to 1/(2¹J_CH) to align the ¹H(¹²C) and ¹H(¹³C) spin vectors along the y and -y axes of the rotating frame of reference, respectively. The 180 degree phase difference between the ¹H(¹²C) and ¹H(¹³C) spin vectors allows a 90 degree pulse to align the these vectors on the -z and z axes, respectively. At this point the ¹H(¹²C) spins are allowed to relax according to their T₁ to the null point. A final 90 degree read pulse puts the ¹H(¹³C) spins in the transverse plane for observation. The first of the two figures below demonstrates the use of the BIRD filter on the lineshape sample. The second figure shows a vector diagram explaining the sequence.

HMQC vs HSQC

Proton detected Heteronuclear Multiple Quantum Coherence (HMQC) and Heteronuclear Single Quantum Coherence (HSQC) are both NMR techniques used to correlate the chemical shift of the protons in a sample to a heteronucleus such as ¹³C or ¹⁵N via the J coupling interaction between the nuclei. Since both techniques essentially provide the same information - a correlation map between the coupled spins - students sometimes ask which of these two methods is better and which should they use routinely. The difference between the two techniques is that during the evolution time of an HMQC both proton and X magnetization (eg: X = ¹³C ) are allowed to evolve whereas in an HSQC only X magnetization is allowed to evolve. This means that an HMQC is affected by homonuclear proton J coupling during the evolution period while an HSQC is not affected as there is no proton magnetization during the evolution time. The homonuclear proton J coupling manifests itself as broadening in the X dimension. The top panel of figure below shows the 7.05 T ¹H /¹³C HMQC and HSQC spectra of menthol with an expansion of one of the resonances highlighted in yellow. One can see that the expanded cross peak of the HMQC is broader in the ¹³C dimension than that of the HSQC. The bottom panel of the figure shows the corresponding ¹³C projection spectra. One can see that the resolution is better in the projection of the HSQC compared to the HMQC. One might conclude that, due to the higher ¹³C resolution, it is always better to run an HSQC rather than an HMQC. This is definitely the case if all of the pulses are calibrated well, however since there are many more pulses in an HSQC compared to an HMQC, it is more susceptible to losses in signal-to-noise-ratio due to poor probe tuning or poor pulse calibration. My advice to students is that, if high ¹³C resolution is required, then make sure the pulses are calibrated well on a well tuned and matched probe and run an HSQC. If high ¹³C resolution is not critical then run an HMQC.

Bloch-Siegert Shifts

Bloch Siegert shifts are frequency differences between NMR signals observed in the presence and absence an rf field applied during the acquisition time. The shifts arise because the applied rf field changes the effective magnetic field experienced by nearby resonances. The resonances are always displaced away from the frequency of the irradiating field. The shift is inversely related to the difference in frequency between the irradiation and the resonance and therefore is generally not observed when heteronuclear decoupling is applied. When homonuclear decoupling is employed these shifts can become significant and are typically used to calibrate the strength of the homonuclear decoupling field. One must be aware of these effects when reporting chemical shifts in homonuclear decoupling experiments. The figure below shows the effect of applying homonuclear decoupling fields of varying strength in the 300 MHz ¹H NMR spectrum of dimethyl acetamide. One can see that the displacement of the resonances is away from the decoupling frequency and that the magnitude of the shift is inversely related to the frequency difference between the resonance and the irradiation frequency.

1H / 27Al TRAPDOR NMR of Kaolinite

TRAPDOR (TRAnsfer of Populations in DOuble Resonance) NMR (Grey and Vega, JACS 117, 8232 (1995)) is a solid state NMR technique where the effects of dioplar coupling between a quadrupolar nucleus and a spin I = 1/2 nucleus can be observed in the spectrum of the spin I = 1/2 nucleus. The technique relies on a rotor synchronized spin echo of the spin I = 1/2 nucleus with CW irradiation of the quadrupolar nucleus during the first echo delay period. The CW irradiation during a single rotor cycle behaves like an adiabatic frequency sweep as the quadrupolar frequencies vary over the course of the rotor cycle. The effects of dipolar coupling between the quadrupolar nucleus and the spin I = 1/2 nuclei, which are normally averaged by MAS, are reintroduced in the TRAPDOR measurement and the complete refocusing of the spin I = 1/2 NMR signal is prevented. The technique therefore can be used to determine whether or not a spin I = 1/2 nucleus is close in proximity to a quadrupolar nucleus. The figure below shows the ¹H / ²⁷Al TRAPDOR NMR spectrum of kaolinite at 11.7 Tesla. The top two traces are conventional rotor synchronized ¹H Hahn echo spectra acquired with MAS rates of 12 kHz and 2.8 kHz, respectively. The bottom trace was acquired at the same spinning speed as the middle trace with CW irradiation of the ²⁷Al during the first echo delay. One can see a very much reduced ¹H echo indicating the presence of heteronuclear ¹H - ²⁷Al dipolar coupling.
This technique can be used to "find" quadrupolar neuclei which are "invisible" by direct detection due to their very large quadrupolar coupling constants.

Merry Christmas

The Importance of Mixing Your NMR Samples

It is very important that your NMR samples are mixed well before NMR data are acquired. If the sample has a concentration gradient (i.e. more concentrated at one end of the sample column compared to the other) it will be very difficult to shim the magnet over the entire volume of the sample as the magnetic susceptibility is not constant over the sample volume. As a result the NMR lines may be skewed and will be much broader than necessary. This will lead to a much lower signal-to-noise ratio based on signal heights. The figure below shows partial 300 MHz ¹H NMR spectra for 2-bromobutane in CDCL₃. The spectrum on the left was acquired on a sample where 1 drop of 2-bromobutane was added to CDCL₃ in an NMR tube. The tube was gently swirled but not shaken. The magnet was shimmed using a gradient shimming routine and the data collected. The spectrum on the right was acquired on the same sample except the tube was removed from the magnet, shaken and reinserted. The magnet was reshimmed with the same gradient shimming routine and the same number of scans were collected. The difference in the quality of the NMR data is obvious.

Optimizing Decoupler Pulses for CP/MAS NMR

The ¹³C line width of protonated carbons, the signal-to-noise ratio and often the resolution in a CPMAS spectrum depends on effective high power proton decoupling during the acquisition time. A convenient sample to optimize the decoupling power or pulses for ¹³C CPMAS NMR is glycine, as the line width of the methylene carbon is very sensitive to the quality of the decoupling. The figure below shows such an optimization of the pulse widths used in TPPM (Two Pulse Phase Modulation) decouipling. The data were collected at 11.7 T.

QCPMG

The quadrupolar Carr-Purcell-Meiboom-Gill (QCPMG) sequence can be used to measure the NMR spectra quadrupolar I = n/2 nuclei in the solid state. This technique is essentially a T₂ sequence where a series of echos are collected. The entire echo train represents the time domain data and is Fourier transformed to produce a frequency domain spectrum. The QCPMG spectrum consists of spikelets separated in frequency by the reciprocal of the time separation between the echos in the echo train. The intensity envelope of the spikelets mimics the static line shape. This is analogous to the rotational echoes in the FID's of MAS data and the associated spinning sidebands in the frequency domain MAS spectra. The QCPMG technique represents an improvement in sensitivity compared to a single conventional Hahn echo as the intensity is concentrated in the spikelets rather than spread across the entire frequency span of the spectrum. The figure below shows the pulse sequence, an echo train and a QCPMG ²³Na spectrum of solid sodium sulfate at 4.7 Tesla. The ²³Na Hahn echo spectrum is also ahown as a comparison to the QCPMG spectrum. The spectrum represents the central transition only. The satallite transitions are not visible.

90 Degree Pulses for I = n/2 Quadrupolar Nuclei in the Solid State

The 90 degree pulse for an I = n/2 quadrupolar nucleus in the solid state depends on the strength of the rf pulse with respect to the quadrupolar frequency. If the strength of the pulse is much greater than the quadrupolar frequency, the pulse is non-selective and excites all transitions equally. If however is is much less than the quadrupolar frequency, then the pulse is selective to the central (m = 1/2 - m = -1/2) transition. The duration of the pulse producing a maximum signal is shorter for selective vs. non-selective pulses at a similar power level. In solution, where the quadrupolar interactions is averaged by random isotropic molecular motion or in the solid state, if the symmetry around the I = n/2 nucleus is cubic, the quadrupolar frequency is small with respect to the strength of the rf pulses and the pulses are non-selective. When the symmetry around the I = n/2 nucleus in the solid state is non-cubic, the quadrupolar frequency is significant and the pulses are very often selective to the central transition. This is illustrated in the figures below for the ²³Na MAS spectrum of a mixture of NaCl (cubic) and Na₂SO₄ (non-cubic). The first figure shows the ²³Na MAS spectrum labelling each component of the mixture. The second figure shows the effect of increasing the pulse duration. One can clearly see that the 90 degree pulse for NaCl is close to twice that of Na₂SO₄.

Before You Leave .....

This may seem to be a strange post .... a rant really...... but very important.

Over the years I have seen many students start a long acquisition (or series of acquisitions) on spectrometers and then immediately leave the lab. After a long lunch, an afternoon of playing billiards, a good night sleep or perhaps a weekend of skiing, they return to the lab and find no useful data waiting for them.

Why? ......

Well ....... perhaps the spectrometer was set up to run 4 rather than 20,000 scans, perhaps the receiver was saturated, perhaps the recycle delay was set to 1000 seconds rather than 2 seconds, perhaps the pulses were not set correctly, perhaps the spectral width was set too small, perhaps the probe was not tuned and matched, perhaps a delay was set to 10 seconds rather than 10 milliseconds. perhaps a typing error was made in the command to start the acquisition..... etc.

The NMR lab charges you for your time whether you get useful data or not, so it is important to be careful.

Before you leave the lab......

1. Double check the parameters in your experiment and for all queued experiments.
2. Query the spectrometer as to how long the experiment will take ("expt" (Bruker), "time" (Varian)) and ask yourself if the response makes sense.
3. Check the probe tuning and matching.
4. Make sure the receiver gain has been set correctly.
5. Look at the first few scans to make sure you have a signal.

The Importance of Grinding Solid Samples

When the heteronuclear dipolar coupling interaction has been removed by high power decoupling, the NMR spectra of dilute spin I = 1/2 nuclei in a single crystal give rise to relatively sharp lines. The frequencies of the lines depend on the chemical shift tensor and the orientation of the single crystal with respect to the magnetic field. Finely powdered samples have many thousands of crystallites and all orientations of the crystallites with respect to the magnetic field are represented equally. As a result, for powders, one obtains a broad powder pattern. Samples that are not ground into a powder contain many fewer crystals than crystallites in a powder and will yield spectra with partially resolved lines. The envelope of lines for all of the crystals will approximate the true powder spectrum. An example of this is shown in the figure below.

Thank you to Victor Terskikh of the National Ultrahigh Field NMR Facility for Solids. for suggesting this post and kindly providing the data for the figure.

Complexed Solvents

I was once asked by an inorganic chemist: why do I have two THF signals in the spectrum of my compound dissolved in THF-d8? Many inorganic compounds crystallize with complexed solvent molecules as a fundamental component of their structure. This is particularly true of tetrahydrofuran (THF). The complexed solvent molecules are released when the solid compound is re-dissolved in solution and can easily be detected by high resolution NMR. The figure below shows the 500 MHz ¹H NMR spectrum of an inorganic compound containing complexed THF which was re-dissolved in THF-d8. One can see the spectrum of the residual protons of the THF-d8 solvent and the spectrum of the complexed THF that was released when the solid was dissolved. The signals are separated due the isotope effect.

The Effect of the Contact Time on CP/MAS NMR Spectra

One parameter for CP / MAS data collection that must be set by the user is the contact time during which magnetization is transferred from the abundant nucleus (usually ¹H) to the dilute nucleus (e.g. ¹³C). In the case of the ¹³C nuclei in organic samples, the build up of magnetization for each type of carbon depends on the extent of the dipolar coupling to the proton network. The extent of ¹³C - ¹H dipolar coupling depends on both the degree of protonation for each type of carbon and any molecular motion (such as methyl group rotation) which may average the dipolar coupling. At longer contact times, the magnetization decays as a function of the T₁(rho) of the protons. It should be noted that cross polarization is also affected by MAS. The length of the contact time should be chosen such that all types of carbons have had sufficient time to polarize yet not so long as to loose significant magnetization due to the proton T₁(rho). For ¹³C CP/MAS an appropriate choice is usually between 1 and 10 ms. The figure below shows the effect of the duration of the contact time for the two ¹³C resonances of glycine. The 50 MHz ¹³C CP/MAS spectra were run as a function of contact time and plotted side by side. The intensities of each resonance are marked with color coded points. One can see that the carbonyl carbon builds up more slowly than the protonated carbon. An appropriate choice of contact time for glycine is 2 -3 msec.

Kinetic Experiments on Bruker Spectrometers

Students often have to monitor the progress of a chemical reaction as a function of time using NMR spectroscopy. I have written three simple programs for XWINNMR (which should work with little or no modification for TOPSPIN). Each program uses a different method to control the time allowed between collecting spectra. All are very simple and easily implemented. They should be added to the Bruker/XWINNMR/exp/stan/au/src directory. The first two programs, kinetic_ds and kinetic_t, are suitible for slow reactions where precise timing is not critical as they do not take into account the time required to initialize each acquisition. The third program, kinetic_2d avoids the problem by using a pseudo 2d approach and is suitible for faster reactions.

1. kinetic_ds
This program uses dummy scans to control the time allowed between spectra. (A dummy scan is a scan taken without turning on the receiver.) The more dummy scans, the longer the time between experiments. The user should set up the appropriate parameters and then run the program (by typing xau kinetic_ds). You will be asked for the total number of spectra to be collected, the number of scans to be collected for each spectrum and the number of dummy scans to be used in all but the first spectrum. The first spectrum will be collected in the current experiment and the others in subsequent experiments.

/* kinetic_ds */
/* written by Glenn Facey, August 24, 2005 */
/* This program will set up a kinetic run based on the use of dummy scans */
/* The user is asked for the number of spectra, the number of scans for */
/* each spectrum and the number of dummy scans for all but the first spectrum */
/* the first spectrum uses no dummy scans. */
GETCURDATA
GETINT("Enter total number of spectra",i1)
GETINT("Enter the number of scans for each spectrum",i2)
GETINT("Enter the number of dummy scans for all but the first spectrum", i3)
STOREPAR("ns",i2)
STOREPAR("ds",0)
Proc_err(0,"Kinetic Run in Progress");
RGA
ZG
TIMES(i1-1)
IEXPNO
SETCURDATA
STOREPAR("ds",i3)
STOREPAR("ns",i2)
ZG
END
QUITMSG("Data Collection Complete!")

2. kinetic_t
In this program, the user should set up the appropriate parameters and then run the program (by typing xau kinetic_t). You will be asked for the total number of spectra to be collected, the number of scans to be collected for each spectrum and the time in seconds allowed between the end of one acquisition and the beginning of the next acquisition. The first spectrum will collected in the current experiment and the others in subsequent experiments.

/* kinetic_t */
/* written by Glenn Facey, August 24, 2005 */
/* This program sets up and runs a kinetic experiment */
/* The user is asked to input the number of spectra, */
/* the number of scans for each spectrum and the time in */
/* seconds between the end of an acquisition and the */
/* beginning of the next. The program will measure the */
/* receiver gain and start the acquisitions. */
GETCURDATA
GETINT("Enter total number of spectra",i1)
GETINT("Enter the number of scans for each spectrum",i2)
GETINT("Enter the time interval (in seconds)", i3)
STOREPAR("ns",i2)
Proc_err(0,"Kinetic Run in Progress");
RGA
ZG
TIMES(i1-1)
IEXPNO
SETCURDATA
STOREPAR("ns",i2)
ssleep(i3);
ZG
END
QUITMSG("Data Collection Finished")

3. kinetic_2d
This program avoids initialization delays by collecting the data in a pseudo 2D format where each slice of the experiment is a spectrum. The program uses a pulse program called zg30kin.gf (see below) which should be put in the directory Bruker/XWINNMR/exp/stan/lists/pp (This pulse program program should be modified to suit the needs of the user). A variable delay list called kinetic must also be set up. This list contains the same number of lines as the number of spectra to be collected. Each line in the variable delay list defines the time interval (in seconds) to be allowed before each acquistion. The user must set up the appropriate parameters (including the number of scans to be collected for each spectrum) and then run the program (by typing xau kinetic_2d). You will be asked only for the total number of spectra to be collected. The program will set up a pseudo 2d acquisition. Data collection is started with the zg command. The data are processed with the xf2 command.

/* kinetic_2d */
/* written by Glenn Facey, August 24, 2005 */
/* This program sets up a pseudo 2D kinetic run */
/* using the pulse program zg30kin.gf with a Variable */
/* delay list called "kinetic". */
GETCURDATA
GETINT("How many spectra do you want to acquire?", i1)
FETCHPAR("SFO1",&d1)
FETCHPAR("DW",&f2)
FETCHPAR("SW",&d2)
FETCHPAR("SF",&d3)
XCMD("parmode 2D")
XCMD("pulprog zg30kin.gf")
XCMD("vdlist kinetic")
STOREPAR("SFO1",d1)
STOREPAR("DW",f2)
STOREPAR("SW",d2)
STOREPAR("SF",d3)
STOREPAR1("TD",i1)
STOREPAR1("SI",i1)
QUITMSG("Setup Complete!\n1. Define 'VD' List called 'kinetic'.\n2. Run the experiment with 'zg'.\n3. Process data with the 'xf2' command.")

Pulse program zg30kin.gf

;zg30kin.gf
;zg30 modified to run kinetic experiment in pseudo 2D mode
;using VD list
;avance-version (00/02/07)
;1D sequence
;using 30 degree flip angle

#include
"d11=30m"
1 vd
ze
2 d1
p1*0.33 ph1
go=2 ph31
d11 wr #0 if #0 ivd
lo to 1 times td1
exit

ph1=0 2 2 0 1 3 3 1
ph31=0 2 2 0 1 3 3 1

;pl1 : f1 channel - power level for pulse (default)
;p1 : f1 channel - 90 degree high power pulse
;d1 : relaxation delay; 1-5 * T1
;d11 : short delay for I/O

Proton Spin Pairs

In the solid state, in the absence of very fast magic angle spinning or homonuclear multiple pulse decoupling schemes, the ¹H NMR spectrum of a typical solid is a broad featureless line greater the 50 kHz in width. This is due to the homonuclear dipolar coupling interactions between the many protons present in the system. The situation is different for an isolated pair of protons. For an isolated pair of protons, there is only one dipolar interaction between the protons and the energy level diagram for the system has only three levels corresponding the combination of spin states among the two protons and the dipolar coupling between them. There are two transitions and therefore two resonances. The separation between the resonances depends on the magnitude of the dipolar coupling constant, R, and the orientation of the internuclear vector with respect to the applied magnetic field. For powdered samples where all orientations with respect to the applied magnetic field are represented, one observes a "Pake" doublet. This situation is very similar to the solid state NMR of ²H where in that case the three energy levels arise from the Zeeman states of a single ²H nucleus and their coupling to an electric field gradient.

Isolated proton pairs occur naturally in the waters of hydration of inorganic salts and the solid state ¹H NMR spectrum is a Pake doublet. The separation between the inner peaks of the Pake doublet is 3/2 R and between the two shoulders is 3R, where R is the dipolar coupling constant. The dipolar coupling constant is directly proportional to the inverse cube of the distance between the protons. Therefore from a single spectrum, one can measure the internuclear separation, r. For the case of the waters of hydration, one can measure the H-O bond length with knowledge of the H-O-H bond angle. The figure below illustrates the Pake doublet spectrum obtained for CaSO₄^. 2 H₂O. The asymmetry in the spectrum is the result of chemical shielding anisotropy and the broadening is the result of dipolar coupling to distant protons.