Over two-thirds of all NMR-accessible nuclides are quadrupolar with half-integer spin quantum numbers (I = 3/2, 5/2, 7/2, etc.). Far from being a minority interest, therefore, NMR studies of nuclei such as 7Li, 11B, 23Na, 71Ga, 87Rb (I = 3/2), 17O, 27Al (I = 5/2), 51V and 59Co (I = 7/2) are central to the characterisation and study of a wide range of microporous materials, synthetic and naturally-occurring minerals, glasses, and novel oxide materials. Although their quadrupolar coupling constants can be large (typically up to ~ 10 MHz), NMR of these nuclei is facilitated by the presence of a mI=+1/2 mI=-1/2 "central" transition which, to a first-order approximation, is not broadened by the quadrupolar interaction. The success of magic angle spinning (MAS) as a line-narrowing technique for spin I = 1/2 nuclei has encouraged its application to spectra of this type and yet, despite achieving a significant narrowing of the central transition, MAS usually fails to resolve the crystallographically-distinct sites present.
The reason for the relative failure of MAS in this case is that the broadening of the central transition arises largely from the second-order quadrupolar interaction and this has second- and fourth-rank terms in its orientational dependence, with MAS capable of only fully averaging the former. Starting in 1988, two methods were proposed for removing both the second- and fourth-rank quadrupolar broadening, and thus achieving truly high-resolution spectra of the central transition. The double rotation (DOR) [1] and dynamic angle spinning (DAS) [2] experiments both involve rapidly spinning the sample about two angles with respect to the magnetic field, either simultaneously in the case of DOR or sequentially in the case of DAS. Although these two techniques have been used with considerable success by a small number of research groups, their use has not become either widespread or commonplace, largely due to the significant additional cost and mechanical complexity of the specialised NMR probe equipment that is required.
It is not surprising, therefore, that the demonstration in 1995 by Frydman and Harwood of their novel multiple-quantum (MQ)MAS technique for removing the second-order quadrupolar broadening [3] generated such instant and enormous interest. Unlike DOR or DAS, the new experiment is purely MAS-based and can be performed on any NMR spectrometer that was equipped with a standard MAS probe. Since its introduction, Frydman's idea has become established as a "modern NMR classic"; the orginal publication has been cited over 280 times, while Frydman himself has won the 1999 Laukien Prize (US$15,000) and moved from Chicago to a prestigious new position at the Weizmann Institute, Israel.
We started working on the MQMAS NMR experiment in the summer of 1995, soon after we first became aware of the technique. The basic experiment, as originally proposed by Frydman and Harwood, yields two-dimensional lineshapes that contain unwanted dispersion-mode contributions and our initial interest was in modifying the basic experiment so that it yielded purely absorption-mode NMR lineshapes. We - and, independently, several other research groups - demonstrated that this could be achieved by performing the experiment in such a way that the two-dimensional time-domain data was amplitude-modulated as a function of the evolution time, t1. In addition, we demonstrated a novel amplitude-modulated version of the MQMAS experiment that used a "split-t1" evolution time to yield a two-dimensional NMR spectrum that, unlike those produced by other MQMAS experiments, did not require subsequent application of a "shearing" transformation. We submitted a preliminary paper on this work in December 1995; the first publication on the MQMAS experiment submitted by a UK research group [4].
In 1996, Massiot et al. showed that absorption-mode MQMAS lineshapes could also be obtained using a phase-modulated experiment in which symmetric echoes, or "shifted echoes", were acquired [5], as in NMR imaging. In two further publications in 1997, we demonstrated that our split-t1 approach was equally applicable to this type of MQMAS experiment and that, on account of its modified use of coherence transfer pathways, it yielded inherently higher sensitivity for spin I = 3/2 nuclei (a point that, only today, is gradually being recognised by the solid-state NMR community) [6, 7].
Despite its remarkably rapid acceptance as an important technique for structural studies of a broad range of solid materials, especially microporous materials, glasses, and novel metal oxides, by 1998 it was widely recognised that there were still many aspects of the MQMAS experiment that required further development. The sensitivity of the technique needed to be improved, particularly for nuclei with very large quadrupolar coupling constants. Equally, if quantitation of the technique were to be made easier, then the excitation and detection of multiple-quantum coherence as a function of the magnitude of the quadrupolar coupling needed to be made more uniform. Finally, there was still much to be understood about the theory of the MQMAS experiment and the interpretation of the resulting data, particularly concerning some of the more subtle effects of the second-order quadrupolar interaction.
This then was the background to our 1998 EPSRC application, "Multiple-Quantum MAS NMR of Quadrupolar Nuclei". We had already demonstrated that we could make original improvements to MQMAS methodology, yet Dr Steven P. Brown, the DPhil student who had worked on the split-t1 techniques, was leaving the group. Our immediate requirement at that time, therefore, was for personnel to carry on our successful MQMAS development work. Thanks to the award of GR/M12209, Dr Kevin Pike was able to join the group in October 1998 as a 24-month postdoctoral research fellow while Mr Jamie McManus joined the same month as a project student. Some of the work envisaged in the original proposal was also undertaken by Miss Sharon Ashbrook, a DPhil student supported by an EPSRC quota studentship.
Key Advances
The aim of the proposed research was to build on the early promise of the MQMAS experiment and to develop it into an even more powerful technique that could be used routinely by chemists, physicists and material scientists in both the industrial and academic communities. The research involved experimental NMR work, theoretical calculations, and computer simulations and was focused in several key problem areas.
I. Cross-polarization. Cross-polarization from 1H to the multiple-quantum coherences of a quadrupolar nucleus was used in combination with the two-dimensional multiple-quantum magic angle spinning (MQMAS) NMR experiment in order to extract high-resolution CPMAS NMR spectra [8]. The technique (which was first developed in our laboratory in 1998) was demonstrated on 23Na (S = 3/2), 17O, 27Al (both S = 5/2) and 45Sc (S = 7/2) nuclei, showing the applicability of multiple-quantum cross-polarization (MQCP) to systems with differing spin quantum number, gyromagnetic ratio, and relative nuclide abundance. The utility of this two-dimensional MAS NMR experiment for spectral editing and site-specific measurement of cross-polarization intensities was demonstrated. The possibility of direct cross-polarization to higher-order multiple-quantum coherences was also considered and three-, five- and seven-quantum cross-polarized 45Sc MAS NMR spectra were recorded. In a separate study, the physical basis of MQCP was investigated both theoretically and experimentally in static samples [9].
II. Distributions of isotropic chemical and second-order quadrupolar shifts. Mr J. McManus developed lineshape fitting procedures that allow the mean and width of the distributions of isotropic chemical shifts and quadrupolar couplings to be estimated from MQMAS spectra of amorphous samples. In a demonstration study, two-dimensional 27Al MQMAS NMR spectroscopy was used to extract the isotropic chemical shifts and quadrupolar parameters of five amorphous aluminosilicates, all of approximately mullite composition (3Al2O3.2SiO2) but of widely differing synthetic origin [10]. Three principal types of Al site were apparent in each sample, conventionally assigned to 4-, 5- and 6-coordinate Al. In some of the more anhydrous samples, two 6-coordinate Al sites were observed. Significant distributions of isotropic chemical shifts and quadrupolar parameters were evident in each of the Al sites resolved in the two-dimensional spectra and the lineshape fitting procedure was used to estimate the means and widths of these. Additional data were obtained from conventional 27Al {1H} CPMAS NMR experiments and suggested that the protons in the samples are most closely associated with particular 6-coordinate Al sites. The 27Al NMR results from the five samples were compiled and compared with those reported for other amorphous and crystalline aluminosilicates.
III. MQCP and MQMAS of amorphous materials. Our development of MQCP and our interest in analysing MQMAS lineshapes came together in this study (by Miss Ashbrook and Mr McManus) of some interesting amorphous materials [11]. Mixtures of the clay mineral kaolinite with gibbsite, when heated, form mullite, a high-temperature ceramic. The 27Al MQMAS NMR technique was used to study the local Al environment in mixtures of kaolinite and gibbsite that had been ground together for varying amounts of time. This mechanical treatment formed new chemical species which were amorphous in nature and, therefore, unsuited to analysis by X-ray diffraction. Both novel (MQCP) and established (single-quantum) cross-polarization techniques were employed to characterise the starting materials and mixtures, confirm the spatial proximity of 1H and 27Al nuclei, and "edit" the two-dimensional 27Al MQMAS NMR spectra, thereby allowing overlapping peaks corresponding to octahedral Al sites to be studied.
IV. Dipolar coupling and chemical shift anisotropy. Second-order perturbation theory predicts the existence of a "cross term" between the quadrupolar and dipolar interactions of two spin I = 3/2 nuclei. This cross term manifests itself as a broadening in solid-state NMR spectra of spin I = 3/2 nuclei which cannot be fully removed by magic angle spinning (MAS) and has an inverse dependence on the Larmor frequency, ω0. In these attributes, the second-order quadrupolar-dipolar broadening does not differ from pure second-order quadrupolar broadening. However, Mr McManus was able to show that the recently developed two-dimensional MQMAS NMR technique, designed originally to suppress second-order quadrupolar broadening, allows the two broadening interactions to be separated and quantified [12]. He then went on to show that a similar cross term arises between the quadrupolar and chemical shift anisotropy (CSA) interactions but that this only affects the appearance of STMAS (see below) spectra. Mr McManus has written a full account of this experimental and theoretical work on second-order quadrupolar cross terms in his PhD thesis [13] and it is envisaged that this will soon be submitted for publication.
V. Higher-order MQMAS experiments. The question of whether or not higher-order (five-, seven- and nine-quantum) MQMAS experiments yield isotropic NMR spectra of half-integer quadrupolar nuclei with higher resolution than the basic three-quantum MAS experiment was examined by Dr K. J. Pike [14]. The frequency dispersion was shown theoretically to be greatly increased in higher-order MQMAS spectra, but it was argued that whether or not this translates into an increase in resolution depends upon the ratio of the homogeneous to inhomogeneous contributions to the isotropic linewidth. Experimentally, it was demonstrated using three-, five- and seven-quantum 45Sc MAS NMR and three- and five-quantum 27Al MAS NMR of crystalline samples that higher-order MQMAS experiments could yield a real and useful increase in resolution but that, owing to the presence of inhomogeneous broadening in the isotropic spectra, this increase was less than the theoretically predicted value.
VI. Satellite-transition MAS NMR. First proposed by Gan in 2000, the satellite-transition MAS (STMAS) experiment offers an alternative approach to established methods such as DAS, DOR and MQMAS for obtaining high-resolution NMR spectra of half-integer quadrupolar nuclei. Unlike the multiple-quantum experiment, STMAS involves two-dimensional correlation of purely single-quantum coherences, satellite transitions in t1 and the central transition in t2, and typically yields 3 to 6 times the signal-to-noise ratio of MQMAS in the same experiment time. Dr Pike implemented this experiment in our laboratory and undertook to improve the basic experiment in a number of ways. A phase-modulated shifted-echo STMAS experiment that yields pure absorptive lineshapes was demonstrated and shown to be compatible with the split-t1 technique used in MQMAS NMR to reduce the duration of t2 acquisition and avoid shearing the final two-dimensional spectrum. The application of STMAS to nuclei with spin greater than I = 3/2 was also considered, the dispersion of isotropic shifts achieved by STMAS and MQMAS were compared, and the effects of anisotropic cross-term broadening mechanisms on linewidths in "isotropic" STMAS spectra were discussed [15].
Project Plan Review
The project proceeded essentially as envisaged originally. The one major circumstance not forseen at the time of the application was that in October 1999, after only one year of the three-year lifetime of the grant, I and my group would move from Oxford to Exeter University where I took up a new position as Reader in Magnetic Resonance. We successfully moved our NMR equipment with us as well, however, and I would judge that this major career move did not impede the progress of the research supported by this grant in any significant way.
Although I have been asked to report only on the two-year period 1/10/1999 to 30/09/01, I find this division very unnatural and am reporting on the full three-year lifetime of the grant (one Oxford year and two Exeter years).
Research Impact and Benefits to Society
The research supported by this grant and described in the 10 published papers (so far) and Dr McManus' PhD thesis is, in my view, mostly internationally competitive, while in a few places (the work on MQCP, STMAS, and quadrupolar-dipolar cross terms, for example) I would judge it to be internationally leading. There will be great benefits to society but it is the nature of this type of work that these may take decades to emerge.
Three of the publications supported by this grant were based around samples provided by Dr K. J. D. MacKenzie, then working on a James Cook Research Fellowship at the Department of Materials, Oxford University, but now Asssociate Professor at the School of Chemical and Physical Sciences, Victoria University of Wellington. This is what Dr MacKenzie has to say about this work:
"Jamie McManus has been involved with me in a collaborative research project using advanced solid-state NMR techniques to explore the chemistry of aluminium in several amorphous aluminate systems of practical significance as advanced ceramic materials. These systems pose more general fundamental questions regarding the existence of pentacoordinated aluminium species which were adressed in this work.
By being able to bring multiple-quantum and cross-polarisation experiments to bear on the problem, Jamie's work has made a considerable contribution to the advancement of our understanding of these technically important systems, which would not otherwise have been possible, since these specialised techniques were not available in my laboratory. This work has resulted in three papers in the international literature and has paved the way for future collaboration."
K. J. D. MacKenzie (k.mackenzie@irl.cri.nz) 12/12/01
Explanation of Expenditure
There was no significant variance of the expenditure from the orginal spending plans; Dr Pike was employed for 24 months as a postdoctoral research fellow and Dr McManus for 36 months as a project student. We had to use some of the consumables funds for travel since - bizarrely - our request for travel costs in the original application was refused on the advice of one referee (bizarre because EPSRC's "standard amounts" include £1000 p.a. and £500 p.a. travel money for postdoctoral and student workers, respectively!). The stipend for PhD students was also increased significantly during the lifetime of the grant but we were able to cover this from an excess arising from the lower student fees in Exeter compared with those charged by Oxford.
Further Research and Dissemination Activities
As mentioned above, there should be a further two full papers to be written based on the research described in Dr McManus' PhD thesis; this will be done when time allows. This final report will be posted on our group website (enter via http://www.ex.ac.uk/ chemweb/research/staff/scw.html).
In addition to the papers published in refereed international journals, the research supported by this grant was presented as posters and talks at conferences in Cambridge, Leipzig (Germany), Denver (Colorado), London, Edinburgh, San Miniato (Italy), Asilomar (California), Mainz (Germany), Orlando (Florida), Durham, and Chamonix (France).
Significant further research work is planned on the development of the multiple-quantum and - especially - satellite-transition MAS experiments and it is likely that we will shortly be applying to EPSRC for funds to allow us to recruit the necessary personnel.
Summary
This has been a tremendously successful research project and represents great value for money. The major points to note are: (i) 10 papers published (so far) in refereed international journals; (ii) 3 papers produced by Dr Pike and 6 by Dr McManus; (iii) Dr McManus submitted his PhD thesis within 3 years of starting his PhD and has now passed his viva voce examination; (iv) Drs Pike and McManus both now productively employed in jobs that require the experience, training, or qualifications gained during the three-year period of the grant; (v) the results of the research also presented at a significant number of national and international meetings and conferences; (vi) the reputation of UK science enhanced by top-class research that is either internationally competitive or internationally leading.
References
[1] A. Samoson, E. Lippmaa and A. Pines, High Resolution Solid-State NMR. Averaging of Second-Order Effects by Means of a Double-Rotor, Mol. Phys. 65, 1013-1018 (1988).
[2] A. Llor and J. Virlet, Towards High-Resolution NMR of More Nuclei in Solids: Sample Spinning with Time-Dependent Spinner Axis Angle, Chem. Phys. Lett. 152, 248-253 (1988).
[3] L. Frydman and J. S. Harwood, Isotropic Spectra of Half-Integer Quadrupolar Spins from Bidimensional Magic-Angle-Spinning NMR, J. Am. Chem. Soc. 117, 5367-5368 (1995).
[4] S. P. Brown, S. J. Heyes and S. Wimperis, Two-Dimensional MAS Multiple-Quantum NMR of Quadrupolar Nuclei. Removal of Inhomogeneous Second-Order Broadening, J. Magn. Reson. A 119, 280-284 (1996). [Read Abstract]
[5] D. Massiot, B. Touzo, D. Trumeau, J. P. Coutures, J. Virlet, P. Florian and P. J. Grandinetti, Two-Dimensional Magic-Angle Spinning Isotropic Reconstruction Sequences for Quadrupolar Nuclei, Solid State Nucl. Magn. Reson. 6, 73-83 (1996).
[6] S. P. Brown and S. Wimperis, Two-Dimensional Multiple-Quantum MAS NMR of Quadrupolar Nuclei. Acquisition of the Whole Echo, J. Magn. Reson. 124, 279-285 (1997). [Read Abstract]
[7] S. P. Brown and S. Wimperis, Two-Dimensional Multiple-Quantum MAS NMR of Quadrupolar Nuclei. A Comparison of Methods, J. Magn. Reson. 128, 42-61 (1997). [Read Abstract]
[8] S. E. Ashbrook and S. Wimperis, Two-Dimensional Multiple-Quantum Cross-Polarization MAS NMR of Quadrupolar Nuclei, J. Magn. Reson. 147, 238-249 (2000). [Read Abstract]
[9] S. E. Ashbrook and S. Wimperis, Single- and Multiple-Quantum Cross-Polarization in NMR of Quadrupolar Nuclei in Static Samples, Mol. Phys. 98, 1-26 (2000). [Read Abstract]
[10] J. McManus, S. E. Ashbrook, K. J. D. MacKenzie and S. Wimperis, 27Al Multiple-Quantum MAS and 27Al {1H} CPMAS NMR Study of Amorphous Aluminosilicates, J. Non-Cryst. Solids 282, 278-290 (2001). [Read Abstract]
[11] S. E. Ashbrook, J. McManus, K. J. D. MacKenzie and S. Wimperis, Multiple-Quantum and Cross-Polarized 27Al MAS NMR of Mechanically-Treated Mixtures of Kaolinite and Gibbsite, J. Phys. Chem. B 104, 6408-6416 (2000). [Read Abstract]
[12] J. McManus, R. Kemp-Harper and S. Wimperis, Second-Order Quadrupolar-Dipolar Broadening in Two-Dimensional Multiple-Quantum MAS NMR, Chem. Phys. Lett. 311, 292-298 (1999). [Read Abstract]
[13] J. McManus, Residual Broadening in High-Resolution NMR of Quadrupolar Nuclei in Solids, PhD thesis, University of Exeter, 2001. [Read Abstract]
[14] K. J. Pike, R. P. Malde, S. E. Ashbrook, J. McManus and S. Wimperis, Multiple-Quantum MAS NMR of Quadrupolar Nuclei. Do Five-, Seven- and Nine-Quantum Experiments yield Higher Resolution than the Three-Quantum Experiment? Solid State Nucl. Magn. Reson. 16, 203-215 (2000). [Read Abstract]
[15] K. J. Pike, S. E. Ashbrook and S. Wimperis, Two-Dimensional Satellite-Transition MAS NMR of Quadrupolar Nuclei: Shifted Echoes, High-Spin Nuclei and Resolution, Chem. Phys. Lett. 345, 400-408 (2001). [Read Abstract]