A Useful Winter Emulsion

The winters in Ottawa are cold (and arguably miserable).  The cold can cause many detrimental effects on ones comfort.  One particularly uncomfortable condition usually disappears within a day or two on visiting the warm sunny Caribbean.  The proton and carbon NMR spectra below were acquired on a very useful emulsion used by many cold Canadians to keep this condition at bay.    What is the emulsion?

Happy Comfortable Holidays !!!!

NMR Tube Thickness and Signal-to-Noise-Ratio

The amount of NMR signal is expected to be proportional to the amount of sample inside the coil of the NMR probe.  As a result, the signal-to-noise ratio for samples run in NMR tubes with thick walls is expected to be lower than that for comparable samples run in NMR tubes with thinner walls due to a reduced filling factor of the NMR probe coil.  I was curious to see how much of a difference in signal-to-noise ratio there would be.  0.68 mL of  CDCl₃ (99.8 % D) was put in 5 mm NMR tubes with wall thicknesses of 0.38 mm and 0.80 mm.  The NMR tubes were New Era Entepprises NE-MP 5 (4.20 mm ID) and Norell S-300 (3.43 mm ID), respectively.  The samples are shown here:

The height of the sample column for the thick-walled tube is obviously higher due to the smaller inner diameter of the tube.  In this case, much of the sample will be "invisible" to the NMR measurement as it is outside of the active NMR probe coil volume and therefore "wasted".  Single scan proton NMR spectra were run for these samples on a 300 MHz instrument.  A third sample was prepared by removing some sample from the thick-walled NMR tube such that the column height was equal to the sample in the thin-walled tube.  The volume for this sample was 0.45 mL and it was run under identical conditions to the other two.  Care was taken to shim the magnet and tune and match the NMR probe reproducibly.  The data, processed with 0.5 Hz of line broadening, are plotted side by side in the figure below:

The 0.68 mL sample in the thin-walled tube (blue) gave a signal-to-noise-ratio of 566.  The 0.68 mL sample in the thick-walled tube (red) gave a signal-to-noise-ratio of 339 and the 0.45 mL sample in the thick-walled tube (green) gave a signal-to-noise-ratio of 369.  The difference in the signal-to-noise-ratios for the two samples in the thick-walled NMR tube may very well be the same within experimental error as the signal-to-noise-ratio is very sensitive to magnet shimming.  One would expect them to be similar based on the fact that both samples have volumes exceeding the active volume of the probe coil.  From the data, one sees a 35-40% loss in signal on going from a thin-walled to a thick-walled NMR tube.  It is instructive to look at the volume corrected signal-to-noise-ratio of the 0.45 mL sample in the thick-walled NMR tube compared to the 0.68 mL sample in the thin-walled NMR tube.  If the signal-to-noise ratio for the 0.45 mL sample is multiplied by (0.68 mL/0.45 mL), the corrected value is 557 which is very likely the same as the 566 value measured for the 0.68 mL sample in the thin-walled NMR tube within experimental error.  From these observations, one can conclude that the signal-to-noise-ratio loss is entirely due to the reduction in sample volume within the coil.         

EZNMR at the University of Ottawa

The NMR Facility at the University of Ottawa is equipped with eight NMR spectrometers and has on the order of 100 hands-on users at the graduate and post-doctoral level.  Like any university NMR facility, the users enter at varying knowledge and experience levels: from "What does NMR stand for?" to "How do I do a shearing transform for my 5QMAS data set?".  Also, the attitude of the user's supervisors varies considerably.  Some supervisors want their students to spend as little time in the NMR lab as possible by collecting all of their data in automation with a sample changer so they can maximize their time at the bench.   Others want their students to learn how to collect the best possible data and fully understand the NMR measurements they make.  There is no doubt that collecting NMR data under complete automation is incredibly time-efficient however, collecting data in this way teaches the student nothing about NMR measurements.  On the other hand, learning to use NMR spectrometers manually, at the most fundamental level, to collect the best possible data, requires a great deal of knowledge (both general and instrument-specific) and although it is the most educationally rewarding, it certainly provides less overall sample throughput.  In our facility, almost all students are first given 10 minutes of training, on how to collect NMR data under complete automation using our only fully automated instrument.  Running an NMR spectrometer in this way requires absolutely no knowledge of NMR spectroscopy.  Most users are also interested in using the other less automated instruments.  These students are provided with as much training as they desire.   The job of the NMR facility is to educate and satisfy the needs or each user.  Doing so, requires finding a "happy medium" between complete automation and complete manual spectrometer operation and using that medium as a minimum training standard.  For the last ten years or so, the "happy medium" used by the University of Ottawa to run four of it its Bruker NMR instruments is based on a customized button panel approach.  We have written button panels specific to each instrument and call the option EZNMR.  We have included EZNMR as an entry on the top TOPSPIN menu bar.  Clicking the EZNMR option opens up a button panel like the one shown below, used on our AVANCE 500 spectrometer. 

Each button either issues a command, runs a macro or runs an "au program".  Some of our students use exclusively this panel to collect their data.  The advantage to using such a system is that the student must at least learn all of the steps involved with collecting the data.  A typical EZNMR session involves simply following the button panel from top to bottom.  If the probe contains a sample, it is ejected with the EJECT button.  A new sample is lowered into the probe with the INSERT button.  The deuterium lock is established by pressing the LOCK button which prompts the user for the solvent and then establishes the lock.   The SHIM button first calls up a standard set of shims and then initiates a gradient shimming routine.  After the magnet is shimmed, the user presses any one of the green buttons depending on which NMR measurement they intend to carry out.  Pressing one of these buttons prompts the user to define a data set and then calls up a reasonable set of parameters into that data set.  If desired, the user can change the number of scans or some of the parameters by pressing the SCANS or PARAMETERS buttons.  The probe is then tuned using the TUNE button.  The START button optimizes the receiver gain and begins collecting the data.  Once started, the data can be processed at any time using the Proc 1D button or halted using the HALT button.  We use a similar button panel for the commonly used 2D NMR experiments which can be called up from the 1D panel.  It is shown here:

The advantages to using this system are:

-  It is highly customizable for the hardware of each instrument as it is based on macros and "au programs".

-  It can be added to as demands change.

-  All NMR spectrometers using EZNMR look pretty much the same so instrument specific training is less of an issue. 

-  Students can begin running NMR experiments very quickly.

-  Students are more likely to ask questions about each step and can learn at their own pace while maintaining high sample throughput.

-  Its use is entirely optional.

-  It is much more time-efficient than complete manual operation.

Students are of course encouraged to learn more about spectrometer operation than is available through the EZNMR buton panels.

19F NOESY

Two-dimensional ¹H NOESY data are routinely used to assign specific stereo-isomers based on the proton nuclear Overhauser effects (NOE's) which are strongly correlated to inter-proton distances through space.  For example, NOE's may be observed for cis- protons across a double bond but not observed for trans- protons.  The same technique can be used with ¹⁹F in fluorinated compounds to gauge the inter-fluorine distance and assign stereochemistry.  The figure below shows the ¹⁹F NOESY spectrum of a fluorine containing cobalt complex.

From the 1D-¹⁹F NMR spectrum, it is not clear which fluorine atoms are on the same or opposite sides of the four membered cobalt containing ring.  The 2D-¹⁹F NOESY spectrum, on the other hand, shows strong NOE cross peaks between fluorine C and both A and E indicating that C, A and E are on the same side of the ring.  There are also strong cross peaks between fluorine D, and both B and F indicating that D, B anf F are on the same side of the ring.

Thank you to Graham Lee (of Dr. R.T. Baker's research group at the University of Ottawa) for kindly providing the sample and sharing his data. 

Isotope Effects and the 19F - 13C HMQC Spectrum of Trifluoroacetic Acid

The ¹⁹F - ¹³C HMQC spectrum of trifluoroacetic acid is shown in the figure below.

The data were collected with a delay appropriate for a ¹⁹F - ¹³C J  coupling constant between the ¹J_F-C coupling constant of 284 Hz and the ²J_F-C coupling constant of 44 Hz.  The top and side traces are the one-pulse ¹⁹F and ¹³C spectra, respectively.  Why are the HMQC responses not at the same ¹⁹F chemical shift and why aren't they correlated to the peak in the ¹⁹F spectrum?  In order to answer these questions one must take into consideration the ¹⁹F - ^{12, 13}C isotope effects.  The chemical shift of the fluorine depends on whether it is bound to a ¹²C or a ¹³C.  The effect is largest across one bond and gets smaller over multiple bonds.  The ¹⁹F NMR spectrum for trifluoroacetic acid is shown in the figure below with and without ¹³C broadband decoupling in the upper and lower traces, respectively.

Approximately 98% of the trifluoroacetic acid is the ¹²CF₃-¹²COOH isotopomer, giving rise to a large singlet plotted off-scale in the figure. Approximately 1% of the signal is from the ¹³CF₃-¹²COOH isotoponer giving rise to a doublet with ¹J_F-C = 284 Hz and approximately 1% of the signal is from the ¹²CF₃-¹³COOH isotoponer giving rise to a doublet with ²J_F-C = 44 Hz.  All of these signals are clearly present in the lower trace of the figure.  When ¹³C broadband decoupling is applied, the doublets collapse into singlets.  The singlets from each of the isotopomers are resolved in the top trace.  The one-bond ¹⁹F - ^{12, 13}C isotope effect is 0.13 ppm and the two-bond effect is 0.02 ppm.  The figure below shows the same HMQC data with the spectrum from the top trace used as a projection.

One can see that the HMQC responses are correlated to their respective isotopomers.  These effects are also present in ¹H - ¹³C HMQC spectra, but the ¹H - ^{12, 13}C isotope effect is much smaller than the ¹⁹F - ^{12, 13}C isotope effect.  

Measurement of Long Range C H Coupling Constants

The stereochemistry of compounds is assigned very often with proton - proton NOE's by applying the 2D NOESY technique or the 1D selective gradient NOESY technique.  These methods fail, however when the distance between protons is too large to measure an NOE.  When faced with this situation, it may be possible to measure long range proton - carbon coupling constants which are able to provide the necessary information.  Three-bond carbon - proton couplings follow a Karplus relationship where the magnitude of the coupling constant is related to the dihedral angle between the carbon and the proton.  In some cases, these dihedral angles may be used to assign the stereochemistry.  Coupling constants are largest for dihedral angles of 0° and 180° and smallest for dihedral angles of 90°. The simplest way to measure the long range coupling constants is to collect a ¹³C NMR spectrum without ¹H decoupling.  These spectra can be very complicated as can be seen from the figure below showing the C2 and C3 aromatic carbons of toluene.

Extracting specific long range carbon - proton coupling constants is quite tedious.  One way to simplify matters and obtain specific carbon - proton coupling constants is to apply the selective 2D heteronuclear J-resolved technique first introduced by Bax and Freeman in 1982 (JACS 104, 1099).  This method employs a ¹³C spin echo with a selective ¹H 180° pulse applied simultaneously with the ¹³C nonselective 180° pulse. A version of this sequence is shown in the figure below with a shaped adiabatic ¹³C 180° pulse.

In this sequence one obtains a 2D spectrum with ¹³C in the F2 domain and the long range couplings to the selectively inverted proton in the F1 domain.  An example is shown in the figure below for toluene where the methyl protons were selectively inverted with a 20 msec Gaussian pulse.

All of the carbons coupled to the methyl protons are split into quartets in the F1 domain and the long range coupling constants which were very difficult to obtain from the coupled ¹³C spectrum can simply be read directly from the 2D spectrum.    

Exact Simulation of Quadrupolar Lineshapes in Solids

The NMR spectra for quadrupolar nuclei in solids contain a great deal of structural information.  The evaluation of quadrupolar coupling constants, asymmetry parameters, isotropic chemical shifts, chemical shift spans, chemical shift skews and the angles relating the electric field gradient tensor to the chemical shift tensor is typically done by simulating the NMR spectrum with suitable software and fitting the simulated spectrum to the experimental data.  Almost always, the spectra of quadrupolar nuclei in solids have been simulated using perturbation theory where the quadrupolar interaction is treated as a perturbation on the much larger Zeeman interaction.  With the recent developments in the collection of ultra-wide line NMR spectra, quadrupolar nuclei with larger and larger quadrupolar coupling constants are being studied by NMR and the perturbation approach may not be valid.  Errors between simulated and experimental spectra appear when the Larmor frequency is not significantly larger than the quadrupolar coupling constant.

Recently, a new program called QUEST (QUadrupolar Exact SofTware) has been written by Frédéric Parras from the research group of David Bryce at the University of Ottawa.  As the name implies, this program is capable of simulating exactly the spectra of quadrupolar nuclei in solids without resorting to the assumptions of perturbation theory. QUEST is able to quickly calculate accurate lineshapes regardless of the ratio between the Larmor frequency and the quadrupolar coupling constant.  It even works in cases where the Larmor frequency is much less than the quadrupolar coupling constant (i.e. NQR). The figure below shows a series of spectra calculated for a spin I=3/2 nucleus as a function of the ratio of the Larmor frequency, ν_L , to the quadrupolar coupling constant, C_Q. The spectra near the top are NMR-like and those near the bottom are NQR-like.  QUEST is a fast, graphical, easy-to-use program able to handle multiple sites, export data in Bruker format, import experimental spectra for comparison to the simulations and simulate spectra as a function of the angle of the detection coil with respect to the magnetic field. The package also includes a very helpful well written pdf manual.  The program is reported and described fully here.  To take a look at the program in action, watch these tutorial videos prepared by the author.  The complete program is available for free download here.  I highly recommend it!

60 MHz NMR on the Bench Top

The development of bench top NMR spectrometers has certainly been exciting recently!  Nanalysis (a Canadian company) has recently introduced a 60 MHz bench top NMR spectrometer.  The NMReady^TM60P is capable of running both ¹H and ¹⁹F NMR spectra.  Like the PicoSpin spectrometer, this instrument should have a high impact on the NMR scene.

Weak Lock Signals and Distorted NMR Spectra

A good ²H lock signal with a high signal-to-noise ratio is a real advantage for maintaining a stable magnetic field for long data acquisitions and also for shimming the magnet using the lock signal.  Sometimes, however it is desirable to run NMR spectra for samples with only a very small quantity of deuterated solvent and therefore a very weak lock signal.  Such may be the case when one is monitoring a chemical reaction by removing aliquots and adding a drop or two of a deuterated solvent to help with magnet shimming using the ²H lock signal.  Although one may be able to shim a magnet using a very weak lock signal (with difficulty), running the spectrum locked may not be a good idea.  Running a spectrum while locked on a very weak lock signal can lead to distortions in the spectrum.  It is often better to use the weak lock signal to shim the magnet as best you can and then run the spectrum unlocked.  This is demonstrated in the figure below. 
The figure shows two single scan ¹H NMR spectra of a sample of acetone (one drop) in CCl₄ with a drop of CDCl₃.  The spectrum on the left was acquired using the ²H lock and the one on the right was acquired unlocked.  One can clearly see the distortion in the ¹H spectrum caused by locking on a very weak ²H signal.

The Extremely Complicated 1H NMR Spectrum of Ethane

It is often incorrectly assumed that simple compounds yield simple NMR spectra. The ¹H NMR spectrum of ethane is such an example. The complexity arises when one takes into account the inequivalence between methyl groups in the mono ¹³C isotopomer which accounts for 2% of the naturally occurring ethane. In this isotopomer, one methyl group experiences a one-bond ¹H - ¹³C coupling (¹J_H-C) while the other methyl group experiences a two-bond ¹H - ¹³C coupling (²J_H-C). Also, the effects of the three-bond ¹H - ¹H coupling (³J_H-H) are exhibited in the spectrum due to the inequivalence. These couplings have a dramatic effect on the spectrum. Furthermore, there is a very small isotope effect on the ¹H chemical shifts of each methyl group due to the presence of ¹³C vs ¹²C. This effect however, is very small (~0.002 ppm) and has very little effect on the spectrum. The left panel of the figure below shows a simulation of the ¹H NMR spectrum of the ¹²CH₃-¹²CH₃ which accounts for 98% of naturally occurring ethane. As expected, the spectrum is a singlet as both methyl groups are equivalent to one another. The middle panel of the figure shows a simulation of the ¹H NMR spectrum of the ¹³CH₃-¹²CH₃ isotopomer which accounts for 2% of naturally occurring ethane. In this case the spectrum is extremely complex due to the ¹J_H-C , ²J_H-C and ³J_H-H coupling. The panel on the right shows a simulation of a scaled up representation of what one would expect for naturally occurring ethane.
The parameters for the simulations are as follows: Δδ_H between -¹²CH₃ and -¹³CH₃= 0.002 ppm, ¹J_H-C = 125 Hz, ²J_H-C = -4.67 Hz, ³J_H-H = 8 Hz and LB = 0.5 Hz.

The Major Constituents of Natural Gas

The major constituent of natural gas is methane however, other gaseous hydrocarbons are also present. One way to identify other components is to dissolve some natural gas in a solvent and examine the ¹H NMR spectrum. The spectrum in the figure below was acquired on a sample prepared by bubbling natural gas through benzene-d₆ for several minutes. The spectrum clearly shows the presence of methane, ethane, propane and water. The spectrum also indicated other impurities at much lower levels (not shown in the figure).

Protic Samples in Aprotic Solvents

The appearance of the ¹H NMR signals of protic samples in aprotic solvents depends critically on the concentration of the sample. The -OH, -NH₂ or -COOH signals can have chemical shift values and line widths over a wide range due to varying extents of hydrogen bonding and chemical exchange. The concentration can also determine whether or not one is able to observe J coupling between and -OH proton and other protons in the sample. An example of this is illustrated below. The figure shows the spectrum of methanol in deuterated acetone. The spectrum on the top is that of concentrated methanol and it consists of two singlets. In this case the methanol molecules are hydrogen bonded to one another and the -OH protons are undergoing fast exhange with one another. The spectrum on the bottom is that of very dilute methanol. It is a second order spectrum with two signals approximating a doublet and a quartet due the J coupling between the methyl protons and -OH proton, respectively. Note that the chemical shift of the -OH proton is much lower for the dilute methanol compared to the concentrated methanol. In this case the methanol molecules are not hydrogen bonded to one another and there is no (or very slow) exchange among the -OH protons between molecules allowing for the observation of the J coupling.

Double Presaturation

Presaturation is a common method of reducing the water signal in the ¹H NMR spectra of aqueous samples. Sometimes, a sample may contain more than one undesirable resonance which a user may want to presaturate. In such a case, one must presaturate at multiple frequencies simultaneously. On a two-channel Bruker spectrometer, two signals can be presaturated. This is accomplished by using both Signal Generation Units (SGUs). One of the undesirable signals is put on-resonance and is presaturated with the signal from SGU1 after which the hard pulse is given (also through SGU1). The second undesirable resonance is presaturated using SGU2. The configuration is as follows:If one has a three-channel system, one can presaturate three resonances using three SGUs. The figure below shows an example of double presaturation on a two-channel system. The sample consisted of phenylalanine dissolved in D₂O contaminated with methanol. The one-pulse spectrum in the bottom left panel shows the intense HDO and methanol signals. The double presaturation spectrum in the top left panel is on the same vertical scale as the one on the bottom left. One can see that both solvent signals have been almost completely eliminated. The spectra on the right-hand side are the same data as on the left except the vertical scale has been increased by a factor of 100. You can try this on your Bruker spectrometer using the pulse program "lc1prf2".

Sorting Out NOE's for Exchanging Rotamers

2D NOESY spectra contain cross peaks from both NOE interactions and peaks due to rotamers in slow exchange with one another on the NMR tine scale. For small molecules, the cross peaks resulting from slowly exchanging rotamers are of the same sign as the diagonal peaks. The NOE cross peaks, on the other hand, are of opposite sign compared to the diagonal peaks. When both types of correlations are present there may be more NOE correlations than expected. What follows is an example of this. The figure below shows a color coded chemical structure of a ruthenium complex with a color coded partial ¹H NMR spectrum.It is obvious from the NMR spectrum that all of the signals from the color coded protons are doubled in the spectrum. One possible explanation for this is that there is a slow rotation about the ruthenium carbon bond indicated with the red curly arrow allowing for two possible nonequivalent rotamers. This is confirmed with the 2D ¹H NOESY spectrum shown in the figure below with a 0.9 second mixing time. The spectrum clearly shows exchange peaks between corresponding pairs of ¹H signals from each rotamer.The interesting thing to note from the NOESY spectrum is that each aromatic proton (pink) from a single rotamer shows NOE correlations to the methyl groups (blue and yellow) of both rotamers - not just those from a single rotamer. With this data, it is not possible to assign the subspectrum of a single rotamer. Presumably, the assignment could be made by collecting a 2D NOESY spectrum at low temperature where the rotation was completely frozen out or by collecting a 2D NOESY spectrum with a very short mixing time where the rotation would be limited. The problem with the former approach is that the solvent may freeze at a temperature too high to stop the bond rotation. The problem with the latter approach is that the NOE's would be much reduced due to the short mixing time and collecting a 2D data set with sufficient signal to noise ratio would take a great deal of time. Another approach is to collect selective 1D gradient NOESY spectra with selective excitation of the aromatic proton from each rotamer individually. These data are shown in the figure below for two different mixing times.Each spectrum is displayed in two parts. The left-hand panel is the aromatic region with the selectively excited resonance colored red and the right-hand panel is the aliphatic region showing the NOE correlations to the methyl groups. From the spectra collected with a 2 second mixing time, one can see that the selective excitation is no longer selective due to bond rotation during the long mixing time. One can see inverted peaks for the aromatic protons of both rotamers despite the fact that the ¹H signal of only one rotamer was selectively excited. Furthermore, NOEs to the methyl signals from both rotamers are present. The spectra collected with only a 0.2 second mixing time, on the other hand, show very selective excitation. The time scale of the bond rotation is obviously longer than the 0.2 second mixing time. The spectra show only the NOEs between the selectively excited aromatic proton and the methyl groups from a single rotamer. The NOEs build up fast enough to be observed during the 0.2 second mixing time before rotation occurs. These data allow for the assignment of signals from each of the rotamers.

Thank you to Justin Lummiss of Dr. Fogg's research group for sharing this interesting system.