reading

The cover shows Strychnos nux vomica, the tree from which semen strychnine, one of the model compounds used in many of the experiments in this book, is extracted. The picture was taken by the senior author in the botanical garden of Cape town. In addition, the 3D structure of strychnine and the HSQC pulse diagram to observe 2D (¹H,¹³C)-correlation NMR spectra is given.

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at <http://dnb.d-nb.de>.

All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

Many of our readers know the earlier books in this series, “100 and more Basic NMR Experiments” (1996), “150 and more Basic NMR Experiments” (1998) and “200 and more NMR Experiments” (2004). Since all books of this highly successful series have been sold out, we have asked ourselves whether it would be worthwhile to further increase the number of NMR experiments in a next edition. This could easily have been done, because NMR is still a very vivid field of science with a steady urge of innovation.

However, in daily discussions with our own co-workers we had to realize that even these people were not able to recall all or even the more important experiments from “200 and more”. Thus we finally decided on a “downsizing” and present in this volume “50 and more Essential NMR Experiments”, a much smaller collection. We feel that at least these should be known to the experimental NMR operator in a chemical environment. A severe cut concerns solid state NMR and NMR in structural biology, where we provide only 4 examples in this volume, knowing that especially these two fields of NMR are currently the most active ones. However, since this volume is mainly intended for organic and inorganic chemists we feel that this community is better served with a more targetted selection.

By implementing the experiments on your spectrometer, please keep in mind that some of the parameters are strongly machine dependent (e.g. power levels, spectral widths in Hz) and must be modified according to your situation. We reflect the values here as we used them to record the spectra pictured in the book.

(1) Dr. Matthias Findeisen, a physicist, replaces Siegmar Braun who retired several years ago.

(2) Nearly all experiments shown in this volume have been recorded anew, and the cited literature goes to mid 2012. There are a few entirely new experiments not present in the previous volumes. All experiments have been regrouped from a more chemical perspective.

(3) There is a new section Variants in which modifications of the actual experiments are mentioned.

(4) There is a new Question section at the end of each experiment. We hope that answering these questions will give the reader much more insight. Short answers are provided in the Appendix.

(5) The biggest change is certainly in the layout. Whereas our previous books were solid but rather dry scholarly textbooks, we have this time tried to compose a more modern book with accompanying graphics and photographs, with some historic text clips from celebrities of NMR and theire pictures. We hope that this change will help to raise interest, especially for the young people entering this fascinating field.

From the number of sold copies of our previous books we know that this information is at hand alongside numerous NMR spectrometers in the world, not counting the Chinese edition of “200 and more”. We certainly hope that the “50 Essentials” will be equally successful. We are fairly certain that every experiment works as described, but if not, please complain by email to

Finally, we have to thank many colleagues for helpful discussions and in particular Mrs. U. Zeller, who provided all the layout ideas and who was responsible for getting all this together. In addition, we thank Dr. P. Tzetkova for recording the data of experiment 3.9 and Dipl. Biochem. A. Beil for recording the data of experiment 6.7.

Quo innumerabiles libros et bibliothecas, quarum dominus vix tota vita indices perlegit? Onerat discentem turba, non instruit, multoque satius est paucis te auctoribus tradere quam errare per multos.

What is the use of having countless books and libraries whose titles their owners can scarcely read through in a whole lifetime? The learner is not instructed but burdened by the mass of them, and it is much better to surrender yourself to a few authors than to wander through many.

Chapter 1

The Organic Set of NMR Spectra

There are six NMR spectral methods, which are usually first measured on a routine basis if an organic chemist has produced a new compound. Usually, this organic set is sufficient for a complete structural elucidation, especially if additional support comes from mass spectrometry and IR-or UV-spectroscopy.

1.1 ¹H-NMR

1.2 APT-¹³C-NMR [Attached Proton Test]

1.3 COSY [COrelation SpectroscopY]

1.4 NOESY [Nuclear Overhauser Effect SpectroscopY]

1.5 HSQC [Heteronuclear Single Quantum Coherence]

1.6 HMBC [Heteronuclear Multiple Bond Correlation]

We describe therefore in this first chapter these six methods in some detail using strychnine as an example. Strychnine with its rather complicated molecular structure provides all the typical problems encountered during spectral assignments in organic chemistry. With concurrent instrumentation and about 20 mg of substance having a molecular weight around 500 Da, the total recording time of these techniques will be about 5 h.

Of course, this book offers much more, but this organic set comprises the most essential of all our essentials.

Experiment 1.1 ¹H NMR Experiment

1. Purpose

The aim of the standard ¹H NMR experiment is to record a routine proton NMR spectrum in order to get structure-related information for the protons of the sample, i. e. chemical shifts, spin–spin couplings, and intensities. Here we apply this standard procedure to strychnine and discuss different weighting functions and problems of integration.

The prospect of measuring very rapid reaction rates by NMR provided the inspiration for getting an affirmative response from Linus Pauling [then chairman of the division of Chemistry and Chemical Engineering at California Institute of Technology] that “with NMR, we could investigate the borderline between resonance and tautomerism”. For example, investigating the NMR spectrum of cycloheptatriene with temperature to see if it existed as a rapidly equilibrating mixture of cycloheptatriene and norcaradiene, or was what later would be called a “monohomobenzene”. The argument was persuasive and we soon ordered a 30-MHz Varian proton and fluorine spectrometer.

J. D. Roberts, * 1918 “A Personal NMR Odyssey” Enyclopedia of NMR, 1996, 1, 590–598.

2. Variants

Variants of this form of NMR spectroscopy include first of all excitation with different pulse angles.

However, since recent NMR instruments are sensitive enough, usually one 90° scan is sufficient to obtain a spectrum. Therefore no considerations about reduced pulse angles are necessary. Second, if a strong solvent signal is present, different forms of signal suppression are available. These are discussed in chapter 3.

3. Pulse Scheme and Phase Cycle

4. Acquisition

These data will lead to an FID digital resolution of 0.28 Hz/point for the real or imaginary part of the FID.

5. Processing

Use zero filling to si = 64K and exponential weighting with lb = 0.1 Hz, phase correction and referencing to internal TMS, which is the only acceptable reference scheme. The digital resolution of the spectrum will be with these data sw/si = 0.14 Hz/point. Before integration, perform phase and baseline correction on the spectrum accurately. A comparison of the spectra in Fig. 1.1-3 and 1.1-4 shows how a Gaussian weighting function with lb = −1 Hz and gb = 0.3 makes additional small spin couplings visible.

The appearance of an unsplit methyl signal in a CH₃CH moiety, where the chemical shift difference was large compared to the vicinal coupling constant, expected to be about 7 Hz, subsequently impressed on me the importance of MR spectral analysis. When I understood what was going on, I wrote a paper which was published in 1961 on the nature of the signals of C-methyl groups, and this explained many anomalous “coupling constants” involving methyl groups in steroids and other compounds; it anticipated later research by others on virtual coupling. Although spectral analysis is becoming almost a lost art in the midst of so-called “modern NMR”, it is in fact just as useful and necessary as it has ever been, in my own experience.

F. A. L. Anet, * 1926 “A lapsed organic chemist in the wonderland of NMR” Enyclopedia of NMR, 1996, 1, 187–190.

6. Result

A closer inspection of the integrals reveals that the integral of H-4 is too small as compared to all other integrals of the compound. Although the spectrum was recorded with only one scan, the waiting time after the receiver gain adjust command before and the actual measurement was appartently too short (see Question A).

7. Comments

The excitation pulse p1 converts the equilibrium magnetization of the ¹H nuclei into a transverse magnetization as shown in Equation (1). During the acquisition time chemical shifts and spin-spin couplings develop in the x,y plane, as shown separately in Equations (2) and (3), and are detected by the receiver in the x,y plane in quadrature mode.

8. Questions

A. Suggest a reason why the integral of H-4 is considerable smaller than the others.

B. How would one classify the aromatic spin system?

C. The intensity pattern of the signals between 4.2 and 4.0 ppm has a special name?

D. The signals at 2.35 and 1.27 ppm both belong to the methylene group of C-15 and are spin coupled to each other; however, only one has an additional spin coupling. Why?

9. Own Observations

[1] T. D. W. Claridge, “Highresolution NMR techniques in organic chemistry”, Pergamon, Oxford, 1999.

[2] I. K. M. Sanders, B. K. Hunter, “Modern NMR spectroscopy”, 2nd Edition, Oxford University Press, Oxford, 1993.

[3] H. Friebolin, “Basic one-and two-dimensional NMR spectroscopy”, 3rd Edition, Wiley-VCH, Weinheim, 1998.

[4] H. Gunther, “NMR Spectroscopy”, 2nd Edition, Wiley, Chichester, 1995.

Experiment 1.2 ATP-¹³C NMR

1. Purpose

The aim of a routine ¹³C NMR experiment is to record a ¹³C NMR spectrum with proton broad-band decoupling and data accumulation in order to get chemical shift information for structure determination. At the same time one wants to have multiplicity information. From the many schemes proposed we feel the APT (Attached Proton Test) technique is the most useful and convenient method available, especially if carried out with a chirp 180° pulse on the carbon channel to avoid phase problems at high field spectrometers.

2. Variants

Alternative methods that give information about the multiplicities are INEPT, DEPT, DEPTQ, and PENDANT, and the historic off-resonance ¹H-decoupling technique. Unlike INEPT or DEPT, the APT method yields ¹³C NMR spectra that are only enhanced by the NOE. However, APT also gives information about quaternary carbon atoms. Improved modifications of APT are known [2–4].

3. Pulse Scheme and Phase Cycle

4. Acquisition

Common values:

p1: 45° ¹³C transmitter pulse

p2: 180° ¹³C CHIRP pulse

d1: relaxation delay

d2: 1/J_CH

d3: switching delay

CPD composite pulse decoupling

Natural-abundance ¹³C NMR was not really routine until the introduction of broadband noise proton decoupling. Such decoupling removes the proton splittings from the ¹³C resonances and, in addition, gives a favorable nuclear Overhauser effect (NOE). Thus, a proton-coupled doublet 13C resonance, such as exhibited by trichloromethane, with broadband decoupling produces a singlet peak with a sixfold increase over the intensity of each of the individual doublets. There is hardly a better early example of the utility of broadband proton decoupling for ¹³C spectra than for cholesterol. Weigert had found it impossible to make sense out of the 15-MHz coupled spectrum of cholesterol, because of the jumble of resonances and the generally poor signal-to-noise ratio, even after hours of signal averaging. In contrast, with broadband proton decoupling, 25 rather well-separated resonances were observed. A challenge was thus presented of spectral assignments and was met by H. Reich and M. Jautelat in about six months. An important element in the unraveling of the resonances was D. M. Grant’s work on the steric influence of axial methyl groups on ¹³C shifts in cyclohexane rings.

J. D. Roberts, * 1918 “A personal NMR odyssey” Enyclopedia of NMR, 1996, 1, 590–598.

5. Processing

Use zero filling to si = 64K and exponential weighting with lb = 2 Hz, phase correction and referencing to either internal TMS or via the images

scale using the proton spectrum of the same sample.

6. Result

The figure shows the ¹H broad-band decoupled APT ¹³C NMR spectrum of strychnine as obtained on an DRX-600 spectrometer using a TBI probe head. Because of the inverse probe a certain number of scans had to be accumulated. Note that as usual no integration is performed, since under routine conditions the signal areas are not necessarily proportional to the number of ¹³C nuclei giving rise to that signal. Furthermore, since the d2 delay hits exactly one J_C,H value, the others will be scaled in intensity.

The signal of the solvent CDCl₃ was adjusted to be negative like the other signals of carbon atoms carrying no protons. Signals of CH and CH₃ groups are positive and signals of CH₂ groups negative.

Fig. 1.2-3 Expansion of the ¹³C NMR spectrum in the aromatic and carbonyl region

7. Comments

The APT sequence is in principle a double spin-echo experiment. In the first echo period d2 the evolution of J-coupling modulates the phases of the signals according to C_q or CH₂ groups respective CH and CH₃ groups. By using a 45° or shorter excitation pulse a part of the initial magnetization remains in the z-direction and is inverted to –z by the first 180° pulse. This could lead to a canceling of signals with long spin–lattice relaxation times, but in the second spin-echo period the 180° pulse reinverts the z-magnetization, thus eliminating this problem. In comparison with all other editing techniques APT still seems to be the most simple and efficient method, since it gives in one experiment all the necessary information on all sorts of carbon atoms. The lower sensitivity compared with polarization transfer methods such as DEPT is in practice not important for the C,H spin pair. See, however, the new DEPTQ experiment where the shortcomings of the traditional DEPT are overcome.

It was the time of the Korean War, however, and my deferments ran out. I was drafted and eventually ended up at the Army Chemical Center. where my ‘experience’ with NMR got me a transfer and assignment to help set up the Varian NMR machine (40 MHz proton and ¹⁹F, 17 MHz ³¹P) which they purchased to support chemical warfare research. There I actually learned something about NMR and even published a few papers. After Army service two options developed: go to Illinois as a graduate student with Gutowsky, or persuade the Mellon Institute to buy an NMR spectrometer. The institute itself declined, but the Dow Corning group took the plunge, at least partly because I claimed that useful ²⁹Si spectra would be possible. I chose that option, visited Varian, and confirmed that ²⁹Si spectra could be seen (with an 8.5 Mc s–1 rf unit) and immediately decided that ¹³C would be even more interesting. I soon published the first paper on ¹³C NMR spectra. The rest is another branch of personal and scientific history, except that ¹³C led to interest in spin decoupling and biological polymers, which were both to become essential links in the chain of events leading to ‘zeugmatography’, as I originally called magnetic resonance imaging.

Paul Lauterbur (1929–2007) “One Path out of Many – How MRI actually began” Enyclopedia of NMR, 1996, 1, 445–449.

8. Questions

A. What is understood by the “Ernst angle”?

B. Which parameter controls the correct phases obtained in this experiment?

C. What is meant with a WALTZ16 sequence?

D. The signals of CDCl3 show a 1:1:1 triplet with a spin coupling of 31.7 Hz. Explain and calculate from this the spin coupling found in CHCl3.

E. Looking at the spectrum and the chemical formula of strychnine, you should be able to assign at once at least two carbon atoms. Which ones?

9. Own Observations

[1] H.-O. Kalinowski, S. Berger, S. Braun, “Carbon-13 NMR spectroscopy”, Wiley, Chichester, 1988.

[2] T. D. W. Claridge, “High-resolution NMR techniques in organic chemistry”, Pergamon, Oxford, 1999.

[3] K. M. Sanders, B. K. Hunter, “Modern NMR spectroscopy”, 2nd Edition, Oxford University Press, Oxford, 1993.

[4] S. L. Patt, J. N. Shoolery “Attached proton test for carbon-13 NMR” J. Magn. Reson. 1982, 46, 535–539.

[5] J. C. Madsen, H. Bildsoe, H. J. Jakobsen, O. W. Sorensen “ESCORT editing. An update of the APT experiment” J. Magn. Reson. 1986, 67, 243–257.

[6] A. M. Torres, T. T. Nakashima, R. E. D. McClung “Improved J-compensated APT experiments” J. Magn. Reson. Ser. A 1993, 101, 285–294.

[7] U. Beckmann, W. Dietrich, R. Radeglia “ORSAT and modifications of SEFT and APT” J. Magn. Reson. 1999, 137, 132137.

Experiment 1.3 COSY

1. Purpose

The COSY (COrrelation SpectroscopY) pulse sequence generates a 2D NMR spectrum in which the signals of a normal ¹H NMR spectrum are correlated with each other. Cross-peaks appear if homonuclear spin coupling is present; thus the COSY sequence detects coupled pairs of protons (or pairs of other nuclei such as ¹⁹F, ³¹P or even ¹³C in the case of labelled proteins). Since coupled protons are usually separated by two or three bonds, the connectivity and very often a chemical structure can be derived from the COSY spectrum; however, one must be also aware of long-range spin couplings. The COSY sequence is the most important and most frequently used 2D NMR experiment.

2. Variants

Due to the importance of the COSY technique an impressive number of variants has been developed in the last 40 years. The current Bruker pulse sequence library, for example, contains more than 30 different applications, and it is impossible to discuss them all in this book. We show here a COSY sequence which includes a gradient-selected double quantum filter and the echo-antiecho scheme for the phase sensitive frequency generation in the indirect dimension. The reason for this selection is that in COSY one usually wants to see and interpret the interaction of two protons with each other. The double quantum filter suppresses singlets and reduces multiplets to AX patterns; the echo-antiecho scheme provides phase sensitive spectra.

Almost in despair, I started investigating this latter problem with the help of Gerit Alewaelers, initially on the simple case of the standard AB spectrum in liquids. This soon led to the general proposal of 2D FT NMR spectroscopy in the simple case of homonuclear COSY without phase cycling. We tried to observe the effect experimentally on ethylbenzene, spending weekends working with the rather primitive FT spectrometer available in organic chemistry at our university. Due to the bad phase stability of this spectrometer and to our total lack of practice of high-resolution NMR, Gerit Alewaeters could only observe some of the strongest cross peaks of the 2D spectrum with a signal-to-noise ratio high enough to confirm that our quantum mechanical predictions were correct (not really a surprise…..), but too low to make the new technique look very usable. Formal publication was deferred until cleaner experimental confirmation would be available, but I kept spreading the idea by personal contacts, lectures (probably for the first time at the AMPERE Summer School in Basko Polje in 1971) and by circulating ‘unpublished’ notes written in November 1971. Soon, Richard Ernst let me know that his co-worker Baumann had brought the 2D FT idea back from Basko Polje to Zürich, and that the Zürich group intended to work on it. As it turned out, one of our recurrent delights in Brussels for a number of years has been receiving news from Zürich about the progress of 2D NMR, both experimental and theoretical. It was also a pleasure to learn about the new 2D FT ideas developing in many other groups.

Jean Jeener, *1931, “Reminiscences about the early days of 2D NMR” Enyclopedia of NMR, 1996, 1, 409–410.

(1) The long-range COSY, which emphasizes connectivities caused by small spin-coupling constants. This is important in the case of allylic or W-spin couplings < 2 Hz.

(2) The COSY-45 or the E.COSY methods, which lead to slim diagonals and allow the determination of the sign of spin-coupling constants.

(3) All COSY variants may be combined with different schemes of water suppression.

(4) The phase-sensitive COSY variants can be recorded in high resolution in both dimensions which enables the extraction of digital correct spin-coupling values in the direct dimension.

3. Pulse Scheme and Phase Cycle

Common values:

p1, p3, p5: 90° ‘H transmitter pulse

p2, p4, p6: 180° transmitter pulse

d1: relaxation delay

d2: effective length of gradient

t1: evolution increment

g1, g3: gradients for Echo/Antiecho

selection

g2: gradient for double quantum selection

4. Acquisition

5. Processing

Apply zero-filling in F₁ to 1K words in order to have a symmetrical matrix of data points. Use an exponential window with lb = 3 Hz in the F₂ dimension and a squared π/2 shifted sinusoidal window in the indirect dimension. Apply complex Fourier transformation in both dimensions. Phase correction in both dimensions can be performed after the 2D transformation in order to get clean up/down patterns of the cross-peaks. Zero order phase correction of 90° is a good starting point for the F₁ dimension.

6. Result

The overview spectrum displays all relevant connectivities for strychnine. In the aliphatic expansion clearly the AX pattern can be observed, even though more complicated multipletts are coupled to each other. In the aromatic expansion this is also demonstrated.

The participation of Thomas W. Bauman at the AMPERE summer School in Basko Polje, Yugoslavia, in September 1971 was an extremely fortunate event. Being a meticulous scientist, he brought home a careful script of the lectures, among them one by Jean Jeener that attracted my attention immediately: a simple two-pulse experiment that produced revealing 2D spectra by 2D Fourier transformation of a 2D set of response signals. This was exactly the technique I had been waiting for. I had been thinking for some time about systematic computer-controlled double resonance experiments, but appreciated the complexity of the resulting 2D spectra should they follow the shape of the famous Anderson-Freeman plots.

R.R. Ernst “The success story of fourier transformation in NMR” Enyclopedia of NMR, 1996, 1, 297.

7. Comments

The two pulses given in red color are the essential COSY pulses used in the original pulse sequence. The additional pulses are used for the double-quantum filter (p3) and for phase correction due to the finite length of the gradient pulses (p2, p4, p6)

In the preparation period p of the pulse sequence we find the relaxation delay d1 and the first r.f. pulse p1 which transforms z-magnetization into transverse magnetization. The delay d2 and the 180° pulse p2 only serve to alleviate the finite length of the gradient pulse g1, which would otherwise cause a problem in phasing the 2D spectrum. Then the chemical shift develops in the evolution period e during t₁, which is written here only for proton 1, giving Equation (1). In addition, spin-spin coupling develops; thus each of the two terms with a modulation by the chemical shift will create two more terms including the spin-spin coupling.

In the mixing period m of the pulse sequence the r.f. pulse p3 creates double-quantum magnetization – 2I_1x I_2y, as in Equation (3). Again the delay d2 and the 180° pulse p4 are only for phase correction and correct for the finite length of the gradient pulse g2, which encodes the double-quantum magnetization.

The 90° pulse p5 creates antiphase magnetization from the double-quantum term as given in eq. (4)

In the detection period d chemical shift and spin–spin coupling develop once again during the acquisition time t₂, giving Equation (5).

With the pulse sequence and the echo-antiecho scheme of the gradients used, the sign of the frequencies in F₁ is determined by keeping the sine and cosine terms separate, and thus allows phase-sensitive processing.

As can be seen from the coherence pathway diagram above, the first gradient g1 acts during a period when single-quantum magnetization I⁺ is present (coherence level +1), whereas the second acts during a period when double-quantum coherence I⁺I⁺ is present. The final gradient pulse acts on I^-. Therefore the gradient ratios 30:10:50 and -30:10:50 will be successful to obtain the desired signals. All other coherences are further dephased and are not observable.

8. Questions

A. Why does the standard COSY not reveal strong cross peaks for longrange spin couplings and what is the trick, whereby the longrange COSY is functioning?

B. In which dimension one would extract spin coupling values from a high-resolution COSY and why?

C. Why are double quantum coherences only developing at higher t₁ increments?

D. In the aromatic expansion of the strychnine spectrum one observes only an AX pattern from the triplet. Why?

E. Only one of the protons H-15 displays a cross peak to H-16. Why?

9. Own Observations

[1] J. Jeener, Ampère International Summer School, Basko Polje, 1971 (proposal).

[2] W. P. Aue, E. Bartholdi, R. R. Ernst “Two-dimensional spectroscopy. Application to nuclear magnetic resonance” J. Chem. Phys. 1975, 64, 2229–2246.

[3] T. D. W. Claridge, “High-Resolution NMR techniques in organic chemistry”, Pergamon, Oxford, 1999, 155–159.

[4] J. Keeler, “Understanding NMR spectroscopy”, Wiley, Chichester, 2nd Ed. 2010.

Experiment 1.4 NOESY

1. Purpose

The NOESY (Nuclear Overhauser Enhancement Spectroscop Y) experiment is the two-dimensional equivalent of the NOE difference experiment (see chapter 3.6) and yields correlation signals that are caused by dipolar cross-relaxation between nuclei in a close spatial relationship.

The intensities of the cross-peaks under certain conditions are proportional to the sixth power of the proton–proton distances. Quantitatively, the results differ from 1D NOE difference spectroscopy, since the latter is a steady-state experiment obtained from a saturation of the energy levels, whereas NOESY is a transient experiment obtained after population inversion of the energy levels. In a qualitative way, the NOESY technique gives answers to many stereochemical problems such as exo/endo, E/Z and similar assignment questions. In NMR studies of peptides and proteins NOESY is the essential method for determining peptide conformations or tertiary structure of proteins. We show here a phase-sensitive version with two spoiling gradients during the mixing time.

2. Variants

The gradient pulses during the mixing time destroy most of the COSY artefacts present in NOESY, however, not the zero-quantum coherences. Further improvements of a gradient-supported NOESY technique have been described recently [4]. The NOESY technique has been combined with different schemes of water suppression, if biological samples have to be measured. Also, in protein NMR a 3D version with ¹⁵N editing is the standard technique. Selective 1D NOESY sequences are known if only the answer of a particular spin is needed. An echo-antiecho method as shown in chapter 1.3 for COSY is not advisable because of diffusion effects during the mixing time (encoding gradients before and after the mixing time).

A considerable drawback of the NOESY technique is the dependence of the NOE effect on molar mass and viscosity, which can change its sign and may cause it to disappear for certain conditions. The ROESY technique as described in Experiment 2.2 may be more effective in this case.

3. Pulse Scheme and Phase Cycle

Common values:

p1, p2, p4: 90° ‘H transmitter pulse

p3: 180° *H transmitter pulse

d1: relaxation delay

d2: mixing time 2, in the order of ¹H relaxation time

t1: evolution increment

g1, g2: gradients for artifact suppression

4. Acquisition

5. Processing

Apply zero-filling in F₁ to 1K real data points to obtain a symmetrical matrix of 1K×1K real data points. Use an exponential window in F₂ with lb = 5 Hz and a π/2-shifted squared sine bell in F₁.

Apply complex Fourier transformation corresponding to the States-TPPI mode of data acquisition in F₁ . Adjust the phase of the diagonal signals so that they are negative. The NOESY correlation signals will then be positive if the compound has a molar mass below 1000 (positive NOE effect). Correlation signals caused by chemical exchange will have the same phase as the diagonal signals.

6. Result

The figures show the result obtained on a DRX-600 spectrometer. Note that the phase of the diagonal signals is opposite to that of the cross-peaks as can be seen from the dotted contours. There is a wealth of information to be taken from the spectrum, which can best be studied using a molecular model or an electronic 3D file. Notice, for instance, that only one of the H-20 protons has an NOE contact with one of the H-15 protons, from which a relative assignment of the protons in these methylene groups can be derived.

Albert Overhauser, a young theoretician from the University of Illinois, had made the following prediction: in a metal, where the conduction electrons are known to be responsible for the nuclear relaxation, the saturation of the ESR resonance of these electrons should lead to an enormous increase in the nuclear polarization. […] Secondly, that Overhauser‘s audience at the meeting of the American Physical Society – where he had (in ten minutes) presented the calculations which had led to his amazing conclusion – was immediately split into two parts, which, however, overlapped: those who did not understand a single word of his demonstration, and those who did not believe a single word of his conclusions. In the first row of the sceptics who did not believe his conclusions shone all the stars of magnetic resonance: Bloch and Purcell, Rabi and Ramsey. Bloembergen was of two minds and so was I. As for the presentation itself, 1 will repeat what Van de

7. Comments

The NOESY sequence can be understood from the vector model. We consider two protons with different chemical shifts and without spin-spin coupling.

The preparation period p starts with the relaxation delay d1. The first pulse p1 of the NOESY sequence shown in red aligns all proton magnetization into the x,y-plane which is the end of the time slot p.

In the evolution period e, chemical shift and spin–spin coupling evolve during t₁. In the mixing period m the second red pulse p2 aligns the y components of the two vectors, which are by now labelled with their individual chemical shifts into the negative z-direction. This situation is called a chemical shift-encoded z-magnetization. In the second scan of the phase cycle pulse p2 aligns the y components to the positive z direction to get a FID without NOE effect. Both scans are subtracted by receiver phase to yield the difference spectrum. During the mixing time (2×d2) both protons are allowed to relax and show cross relaxation.

This pathway is shown in the coherence diagram. In addition, however, the pulse p2 can generate zero-, double-quantum-and antiphase coherences, since H,H spin coupling is also evolved during t₁. The positive and negative gradient pulses in the mixing time, which embrace the 180° pulse p3, dephase all these components except the zero-quantum coherences. Furthermore, they dephase axial signals of those protons that have relaxed during t₁ and are excited again by p2. The final pulse p4 reads the situation at the end of the mixing time and realigns the vectors into the x,y plane, where the FID is recorded.

Graaff had told me of de Broglie’s defensc of his thesis in Paris in 1924. „Never had so much gone over the heads of so many”.

The real question was of course: „Was Overhauser right?“ He was: the proof of the pudding was given the same year by Charles Slichter, a physicist from Illinois, and his student, Carver. They saturated the resonance of conduction electrons in metallic lithium and saw the enhancement of the nuclear polarization predicted by Overhauser.

A. Abragam, (1914–2011) “Time reversal, an autobiography”, Oxford University Press 1989

[1] J. Jeener, B. H. Meier, P. Bachmann, R. R. Ernst “Investigation of exchange processes by two-dimensional NMR spectroscopy” J. Chem. Phys. 1979, 71, 4546–4553.

[2] D. J. States, R. A. Haberkorn, D. J. Ruben “A two-dimensional nuclear Overhauser experiment with pure absorption phase in four quadrants” J. Magn. Reson. 1982, 48, 286–292.

[3] G. Bodenhausen, H. Kogler, R. R. Ernst “Selection of coherence-transfer pathways in NMR pulse experiments” J. Magn. Reson. 1984, 58, 370–388.

[4] R. Wagner, S. Berger “Gradient-selected NOESY—A fourfold reduction of the measurement time for the NOESY experiment” J. Magn. Reson. Ser. A 1996, 123, 119–121.

[5] T. Parella, F. Sánchez-Ferrando, A. Virgili “Quick recording of pure absorption 2D TOCSY, ROESY, and NOESY spectra using pulsed field gradients” J. Magn. Reson. 1997, 125, 145–148.

[6] M. J. Thrippleton, J. Keeler “Elimination of zero quantum interference in two-dimensional NMR spectra” Angew. Chem. Int. Ed. 2003, 42, 3938–3941.

[7] D. Neuhaus, M.P. Williamson, “The nuclear Overhauser effect in structural and conformational analysis”, 2nd Ed., Wiley-VCH, Weinheim, 2000.

8. Questions

A. Explain why the mixing time is best approached by the spin-lattice relaxation time of the spins involved.

B. What is meant by a NOESY build-up curve and how is it obtained?

C. What is the reason that build-up curves have a maximum and decay to zero at long mixing times?

D. If the ROESY technique does not have the sign problem that NOESY has, why does one not use only ROESY in all situations?

E. Which molecular entities are suitable for a distance reference?

9. Own Observations

Experiment 1.5 HSQC

1. Purpose

C,H correlation, as well as other X,H correlations, is now mainly achieved by the HSQC (Heteronuclear Single Quantum Coherence) method, because in this sequence the signals are not broadened by homonuclear H,H coupling in F₁. The HSQC scheme is included as a building block in many 3D sequences, especially for structural biology, but has become established as the standard scheme of C,H correlation in organic chemistry. In the experiment shown here (using strychnine as example) we demonstrate a combination of several features which persuade us that it is today the method of choice.

2. Variants

Since the introduction of this experiment [1], there has been a permanent development. Today, mostly a gradient-selected variant is used, with the gradients operating in the echo-antiecho method for sensitivity reasons.

Additional sensitivity improvement by a factor of images

is achieved by a double back INEPT transfer; however, this occurs only for CH groups [2–5]. In reference [6], applications for long-range spin couplings are discussed. In most cases it is desirable to obtain an editing of 2D H,X correlation spectra. This can yield a multiplicity determination in case of overlapping ¹³C signals, or reveal CH moieties in the presence of many CH₂ groups or NH₂ groups in the middle of many NH groups in proteins. This kind of multiplicity determination can be achieved by including an editing period.[7–10]

In contrast to HMQC, the HSQC experiment employs 180° pulses, which causes problems if the 180° pulses become too long (e.g., in a triple-tuned probe) but have to cover a very wide spectral range. This leads to severe phasing problems for instruments with a magnetic field above that corresponding to 500 MHz ¹H frequency. The remedy for this problem is to apply frequency-swept adiabatic 180° ¹³C pulses (see chapter 8.4), which can cover the large spectral width of ¹³C [11, 12].

The enhancement in sensitivity possible with this scheme dwarfs the enhancement obtainable with the nuclear Overhauser effect (when observing the low-y nucleus directly), and therefore it has been referred to as the Overbodenhausen experiment.

3. Pulse Scheme and Phase Cycle

Common values:

p1, p4, p7, p9, p11: 90° ¹H transmitter pulse

p2, p5, p6, p8, p10, p12: 180° ¹H transmitter pulse

p3: trim pulse

p14, p16, p18: 90° ¹³C transmitter pulse

p15, p17: 180° ¹³C transmitter pulse

p13, p19 ¹³C-CHIRP pulse for inversion

d1: relaxation delay

d2, d5: polarization delay during

INEPT or back INEPT transfer

d3, d4: editing delay for multiplicity recognition

t₁: evolution increment

g1,g2: gradients for echo/antiecho selection

td:	64K
sw:	15 ppm
aq:	3.6 s
o1:	middle of ¹H NMR spectrum
d1:	2 s
ns:	1

Sample:	3% strychnine in CDCl₃.
Time requirement:	1 h
Spectrometer:	Bruker DRX-600 with 5-mm TBI probe

td:	64K
sw:	200 ppm
aq:	1.0 s
p1:	45° ¹³C transmitter pulse 6 μs
o1:	middle of ¹³C NMR spectrum
o2:	middle of ¹H NMR spectrum
p2:	adiabatic chirped 180° ¹³C pulse [crp 60, 0.5, 20.1; 500 μs, 5.5 dB]
d1:	2 s
d2:	6.9 ms corresponding to a J_CH = 145 Hz
d3:	100 μs
CPD:	WALTZ16 sequence, individual 90° ¹H pulse:100 μs at 12 dB ns = 1024

td2:	2K data points in F₂
td1:	256 data points in F₁
sw2:	10 ppm
sw1:	10 ppm
aq1:	0.023 s
aq2:	0.19 s
o1:	middle of ¹H NMR spectrum
d1:	2 s
d2:	equal to effective duration of gradient used, here 1.05 ms
g1, g2, g3:	sinusoidal-shaped field gradients, 1 ms duration gradient ratio 30:10:50 with 0.56 T/m = 100%
rg:	One must be very careful in setting the receiver gain for this experiment. The gradient filter allows only the desired coherences to pass into the receiver; however, the double-quantum coherences develop only at higher t₁ increments because of modulation with sin (π Jt₁), cf. Equ. 2 see Question C. The receiver gain must therefore be set using a high t₁ increment to avoid overloading.
ds:	2 2
ns:	2

td2:	2K data points in F₂
td1:	256 data points in F₁
sw2:	10 ppm
sw1:	10 ppm
aq2:	0.17 s
aq1:	0.021 s
o1:	middle of *H NMR spectrum
d1:	2 s
d2:	1 s
ds:	4
ns:	8
g1, g2:	40: (−40) with 0.6 T/m = 100%

Chapter 1

The Organic Set of NMR Spectra

Experiment 1.1 1H NMR Experiment

1. Purpose

2. Variants

3. Pulse Scheme and Phase Cycle

4. Acquisition

5. Processing

6. Result

7. Comments

8. Questions

9. Own Observations

Experiment 1.2 ATP-13C NMR

1. Purpose

2. Variants

3. Pulse Scheme and Phase Cycle

4. Acquisition

5. Processing

6. Result

7. Comments

8. Questions

9. Own Observations

Experiment 1.3 COSY

1. Purpose

2. Variants

3. Pulse Scheme and Phase Cycle

4. Acquisition

5. Processing

6. Result

7. Comments

8. Questions

9. Own Observations

Experiment 1.4 NOESY

1. Purpose

2. Variants

3. Pulse Scheme and Phase Cycle

4. Acquisition

5. Processing

6. Result

7. Comments

8. Questions

9. Own Observations

Experiment 1.5 HSQC

1. Purpose

2. Variants

3. Pulse Scheme and Phase Cycle

4. Acquisition

Special values used for the spectrum shown:

5. Processing

Experiment 1.1 ¹H NMR Experiment

Experiment 1.2 ATP-¹³C NMR