University of Ottawa NMR Facility Web Site

Please feel free to make suggestions for future posts by emailing Glenn Facey.

Friday, June 27, 2008

Glenn is on Vacation!

The U of O NMR Facility BLOG will be quiet for a couple of weeks while I am on vacation. If you leave comments on any post, they will not appear until I return.

This photo was taken by my wife, Patty (I don't paint my nails or wear an ankle bracelet), however it reflects an activity I hope to enjoy in the next couple of weeks.

Artifacts Due To Setting the Receiver Gain Too High in 2D Homonuclear Experiments

Setting the receiver gain too high leads to very characteristic artifacts in 1D NMR spectra. If the receiver gain is set too high in 2D experiments, one can also expect artifacts in the 2D Fourier transformed data. In homonuclear experiments, setting the receiver gain too high will lead to parallel diagonal signals. This is illustrated in the COSY data in the figure below. In the left hand panel, the receiver was set correctly while in the right hand panel it was set too high. Dotted lines were drawn through the artifacts.

Tuesday, June 24, 2008

Spin-Spin Coupling Between Equivalent Nuclei

When many chemists are asked what is the 2JH-H coupling for compounds like methane, acetone, methylene chloride, dimethyl ether or DMSO, they will often return a look of confusion. "There is no coupling," they will say, "the proton spectrum is a singlet". Indeed the proton spectrum is a singlet for these compounds but 2JH-H is not equal to zero. The only reason that the coupling is not observed in the spectrum is because the chemical shifts of each proton are identical. The coupling can easily be measured by observing the spectrum of a partially deuterated isotopomer. The 2JH-H coupling constant is equal to 2JH-D multiplied by the ratio of the gyromagnetic ratios of 1H to 2H. This is illustrated in the figure below for methylene chloride.In fact, 2JH-H is -7.192 Hz not +7.192 Hz however, this cannot be determined simply by observing the spectrum. Both spectra were measured for dilute solutions with CDCl3 as solvent. The residual protons of CDCl3 were used as the chemical shift reference (7.26 ppm). The chemical shift difference between CH2Cl2 and CHDCl2 is due to an isotope effect.

Monday, June 23, 2008

COSY- 90 vs COSY- 45

Aside from the standard 1H and 13C NMR 1D experiments, 1H COSY experiments are among the most commonly used NMR techniques by organic chemists. There are many different modifications to the standard two pulse COSY experiment and often the organic chemist does not even know which one they are using. Two of the most common experiments for routine work are the gradient magnitude COSY- 90 and COSY- 45 experiments. The only difference between the two methods is the flip angle of the second pulse (90 degrees for the COSY- 90 and 45 degrees for the COSY- 45). For a concentrated sample, these experiments can be acquired in a matter of minutes. Although the signal to noise ratio is higher for a COSY- 90, the COSY- 45 is usually the preferred experiment because the diagonal signals are smaller and less intense allowing correlations between close resonances to be resolved more easily. The figure below shows magnitude gradient COSY- 90 and COSY- 45 spectra for 3-heptanone. Note the smaller diagonal responses in the COSY- 45.

Friday, June 20, 2008

APT vs DEPT-135

Both the APT (Attached Proton Test) and DEPT (Distortionless Enhancement by Polarization Transfer) sequences are very commonly used to help assign 13C NMR spectra. Both experiments yield 13C NMR spectra where the number of attached protons (the multiplicity) is encoded in the phase of the 13C NMR signals. APT spectra have quaternary carbons, and methylene carbons phased negative and methine and methyl carbons phased positive. DEPT-135 spectra show no quaternary carbons and have methylene carbons phased negative and methine and methyl carbons phased positive. A modification to the DEPT-135 method (the DEPTQ-135) will also show quaternary carbons phased negative. Although the APT and DEPT methods provide similar information, the mechanism for multiplicity selection is different for each method. The multiplicity selection in APT experiments is based on 13C magnetization dephasing during a delay equal to the reciprocal of the average one-bond 13C - 1H coupling constant. Although DEPT experiments also employ a delay related to the average one-bond 13C - 1H coupling constant, the multiplicity selection is accomplished by adjusting the final 1H pulse (a DEPT-135 uses a 135 degree 1H pulse). Another major difference between the two techniques is that the DEPT technique transfers proton magnetization to carbon giving it a sensitivity advantage over the APT method. This is illustrated in the figure below which compares the 13C APT and DEPT-135 spectra of menthol. The signal-to-noise ratio is higher in the DEPT 135 spectrum.

Thursday, June 19, 2008

11B Background Signals

Unfortunately NMR probes and NMR tubes cannot be "transparent" for all of the isotopes one may want to observe. Depending on the NMR probe, it is very common to have background signals for 19F, 23Na, 27Al, 29Si, 65Cu, 10B and 11B. These background signals must be taken into account when interpreting NMR spectra. The background signal for 11B on a Bruker AVANCE 300 with a 5 mm broadband probe is shown in the figure below with several different types of NMR tubes commonly used for routine work in our laboratory. The probe was tuned before running each spectrum and the spectra were collected with proton decoupling. The left hand panel shows the background signal for the NMR probe and the other spectra show the combined background of the probe and the indicated NMR tubes. It is obvious that the magnitude and shape of the 11B background depends on the type of NMR tube used. It should be noted that all of the major NMR tube manufactures offer quartz NMR tubes which have little (if any) 11B background signal.

Tuesday, June 17, 2008

The Available RF Field for MAS NMR Probes

In order to rotate an equilibrium magnetization vector from the z axis into the transverse plane, one must provide a pulse with an oscillating magnetic field transverse the static field, Bo, at the Larmor frequency of the nucleus being observed. This is usually provided with a vertically oriented Helmholtz coil for liquids and a horizontal solenoid coil for solids. In both cases the coils provide radio frequency fields transverse to the static magnetic field. MAS coils are solenoids oriented at 54.7 degrees from the static magnetic field. They produce an oscillating radio frequency field at the magic angle. It is only the horizontal component of this field which is capable of rotating magnetization vectors. The available radio frequency field is therefore only 82% of that compared to an identical horizontal solenoid coil.

Monday, June 16, 2008

The Effect of Molecular Alignment in a Magnetic Field

Molecules in solution tumble isotropically. The fast random motion averages out the chemical shielding and quadrupolar interactions to their isotropic scalar values. As a result, for solutions, one observes peaks at the isotropic chemical shift values. The quadrupoar interaction averages to zero and is not normally observed directly in solution (except perhaps as broad lines due to short T2 relaxation times). When molecules are fixed in a particular orientation with respect to the magnetic field, as they are for example in rigid single crystals, one observes NMR lines whose positions (frequencies) depend on the orientation of the single crystal in the magnet. In the case of a quadrupolar nucleus, the observed quadrupolar splitting will depend on the orientation of the crystal in the magnet. Like molecules in single rigid solid crystals, molecules dissolved in liquid crystalline solvents can also have a preferred orientation in a magnetic field. In this case, one is able to observe a quadrupolar splitting which is unobservable in isotropic solution. This is illustrated in the figure below for the 2H NMR spectrum of water in a liquid crystalline solvent. The solvent was prepared according to M. Ruckert and G. Otting, JACS, 122, 7793-7797 (2000). In the bottom trace, the sample is liquid crystalline and one sees a quadrupolar doublet for the spin I = 1 2H. In the upper trace at higher temperature, the system becomes isotropic and one observes a sharp singlet for the 2H resonance. Thank you to Melanie Mastronardi from David Bryce's lab for kindly providing the sample used in the figure.

Friday, June 13, 2008

The Effect of Sample Volume on Deuterium Gradient Shimming

Yesterday's post emphasized that the sample depth with respect to the probe coil has a huge effect on the success of deuterium gradient shimming. For similar reasons, the volume of the sample in the NMR tube also has a huge effect. The shimming routine will work best when the volume of the sample is similar to the volume of the sample used to create the shim map. If the volume deviates too much from that, then the shimming will be unsatisfactory. This is illustrated in the figure below where different volumes of the same sample were run. Each sample was gradient shimmed prior to acquisition.

Thursday, June 12, 2008

The Importance of Setting the Depth of Your Sample

Automated deuterium gradient shimming does a remarkable job at shimming NMR magnets and has really improved the quality of NMR data produced by novice NMR users. Despite the widespread success of this shimming method, I still hear complaints from students that "the shimming really sucks". A quick glance at the sample often explains the unsatisfactory results. There are a number of reasons why deuterium gradient shimming may not work well. One reason is that the sample tube height may not be set properly in the depth gauge such that the sample does not sit in the center of the probe coil. Deuterium gradient shimming uses a previously stored spatial map of the well shimmed magnetic field. The "shim map" is produced using a standard sample filled to the correct height and set to the correct depth with respect to the coil in the NMR probe. Any sample that does not sit in the same position as the standard sample used to produce the shim map will distort the magnetic field such that the stored shim map will not work very well. This is illustrated in the figure below. The left hand spectrum shows the result of gradient shimming for a properly positioned sample. For the spectrum on the right hand side, the NMR tube was lifted up 5 mm from a proper depth and the same gradient shimming routine was used.Take home message: Set the depth of your NMR tube properly.

Tuesday, June 10, 2008

Calibration of 13C Decoupler Pulses

Proton-carbon HMQC, HSQC, and HMBC experiments rely on knowing the duration of the 90 degree pulse for 13C for at least one power level. These pulses must be known at both high power and low power when 13C decoupling is employed during the acquisition. This calibration is conveniently done with the pulse sequence depicted in the figure below for CHCl3. For protons attached to 12C, the pulse sequence produces an out of phase signal. When no 13C pulse is applied (i.e. 0 degrees) the components of the 1H-13C doublet signal (13C satellites) are antiphase with respect to one another. When a 90 degree 13C pulse is used, the 1H-13C doublet is converted to an unobservable double quantum coherence and therefore is not present in the spectrum. When a 180 degree 13C pulse is employed, the 13C satellites will again be antiphase with one another but in the opposite sense compared to that when no 13C pulse was employed. The 13C decoupler 90 degree pulse is thus measured by collecting a series of spectra with increasing 13C pulse duration. The 90 degree pulse produces a null for the 13C satellites. This is depicted in the figure below. This must be carried out at both high and low 13C power to determine hard pulses to be used in pulse sequences and low power pulses to be used in multiple pulse decoupling schemes.

Monday, June 9, 2008

Shifts Due to Paramagnetism

The paramagnetic susceptibility of a material can be determined using NMR spectroscopy. Because of its sensitivity, proton NMR is most often used to measure the paramagnetic shift for a reference compound in the presence and absence of a paramagnetic material. Since the paramagnetic shift (in ppm) is independent of the observation frequency, it is also independent of the isotope observed. This is illustrated in the figure below for the 1H, 13C and 29Si NMR spectra of TMS in a CDCl3 solution of Cr(acac)3 and in a similar solution without the Cr(acac)3. In these spectra the high frequency peak is due to TMS in the presence of Cr(acac)3 and the low frequency peak is due to TMS in the absence of Cr(acac)3. In all three cases the chemical shift difference (in ppm) between the peaks is identical.

Friday, June 6, 2008

NMR to Determine Paramagnetic Susceptibilities

The chemical shift of an inert test compound (such as TMS) will depend on the paramagnetic susceptibility of the medium in which it is dissolved. The chemical shift difference between the test compound in the presence and absence of the paramagnetic medium can be used to determine the paramagnetic susceptibility of the medium. Furthermore, if the concentration of the paramagnetic material in the medium and the mass susceptibility of the solvent are known, the mass susceptibility of the paramagnetic material can be determined. The measurement is made by preparing two solutions: one containing a dilute test compound in an appropriate solvent and an identical solution with a known mass of a paramagnetic material added. One of the solutions is sealed in a capillary and placed in an NMR tube containing the other solution. A spectrum is then collected and the chemical shift difference between the resonances of the test compound in the presence and absence of the paramagnetic material is measured and used to calculate the paramagnetic susceptibility. (D.F. Evans, J. Chem. Soc. 2003, (1959)) The figure below illustrates this for the proton spectrum of TMS in a CDCl3 solution of Cr(acac)3 and in a similar solution without the Cr(acac)3. For another example and additional references see the very highly recommended book: S. Berger and S. Braun, 200 and More NMR Experiments p. 305 (2004).

Wednesday, June 4, 2008

Running NMR Spectra of Solids in an NMR Probe for Liquids

If you simply put a solid in a regular NMR tube and run the spectrum on a conventional high resolution NMR spectrometer, what will you get? The answer may surprise you.

One gets sharp lines in the NMR spectra of liquids (or solutions) because the rapid isotropic motion of the molecules averages out dipolar interactions, chemical shielding anisotropy and quadrupolar coupling, all of which can broaden NMR resonances. In rigid solids, there is no isotropic motion and the interactions mentioned above will broaden out the NMR resonances to the point where they will lost in the baseline of typical spectral widths employed for solution state NMR. So, for rigid solids, you are likely to see nothing. Despite what many people believe, all solids are not rigid. Many exhibit some type of molecular motion (eg. methyl group rotation, phenyl ring flips, rotation about eta bonds etc...). Some compounds called "plastic crystals" even exhibit isotropic motion. An example of such a compound is adamantane. The bottom trace of the figure below shows a conventional solution state 13C NMR spectrum of adamantane with proton decoupling. The top trace shows the spectrum of the solid acquired in exactly the same way. Although the resolution for the solid is not as good as that for the the solution, one can still resolve the two types of carbon.

Tuesday, June 3, 2008


COSY (COrrelation SpectroscopY) and TOCSY (TOtal Correlation SpectroscopY) experiments are very common. Although both experiments provide a diagonally symmetric two dimensional contour plot with the one dimensional spectrum on the diagonal and correlations off of the diagonal, they have different information content. A COSY spectrum will have off-diagonal correlations between coupled spins whereas a TOCSY spectrum will have off-diagonal correlations between all spins in a spin system. For example, consider a spin system where A is coupled to B, B is coupled to C, and C is coupled to D. A COSY spectrum will have correlations between A and B, B and C, and C and D, whereas a TOCSY spectrum will have correlations between all of the spins. This is illustrated in the figure below for 3-heptanone. In this case, there are two spin systems (one on either side of the carbonyl group). The COSY spectrum shows correlations between adjacent spins whereas the TOCSY spectrum shows correlations between all of the spins in each of the two spin systems.

Monday, June 2, 2008

The Effect of Spinning Speed on Dipolar Dephasing

Dipolar dephasing is a very simple and effective tool to help assign the 13C CPMAS spectra of solid organic compounds. The technique is based on turning off the high power 1H decoupler for a period of time immediately after cross polarization but prior to the collection of the FID. During this dephasing delay, the 13C - 1H heteronuclear dipolar interaction is averaged only by the magic angle spinning. The degree of averaging of the 13C - 1H dipolar interaction and hence the amount of dipolar dephasing, depends on the MAS spinning speed. The faster the spinning speed, the smaller the amount of dephasing for a fixed period of time. This is illustrated in the figure below. The lower trace shows a 13C CPMAS spectrum with a spinning speed of 12 kHz. The center trace shows the 13C CPMAS spectrum with a 40 microsecond dephasing delay with a spinning speed of 4 kHz. The upper trace is identical to the center trace except the rotor was spinning at 12 kHz. One can see that many of the resonances which did not survive the dephasing delay at 4 kHz are present in the dipolar dephasing spectrum at 12 kHz. When fast spinning speeds are employed, longer dephasing delays (and the complications encountered with them) are required.