background-free solution boron nmr spectroscopy

Research Article

Received: 10 February 2011 Revised: 10 July 2012 Accepted: 11 July 2012 Published online in Wiley Online Library

(wileyonlinelibrary.com) DOI 10.1002/mrc.3854

Background-free solution boron NMRspectroscopyPéter Király*

The appearance of background signals arising from the NMR probe and tube is a well-known problem of boron NMR spectros-copy. Background suppression may be achieved by using DEPTH, which increases the signal-to-background (S/B) ratio. Although,
the quality of such spectra is often adequate, but in the case of rapid relaxation broadened resonances (T1<1ms), the residualbackground signals may still hamper the interpretation of the spectra. It was observed that the background signals are practicallyinvisible in solution 10B NMR. The unusual isotopic effect on the (S/B) ratio was interpreted as an inherent consequence of theinteger versus half-integer spin of 10B and 11B, respectively. The practicability of 10/11B NMR was compared for a selected set ofboron compounds covering the typical range of (S/B) ratio. The application of 11B is more favourable than 10B as long as it ispossible to achieve the desired spectral quality by using DEPTH. Otherwise, the ‘background-free’ appearance of 10B NMR spectramakes 10B a reasonable alternative of 11B DEPTH. This was found typical for compounds having relaxation broadened resonances.The variable temperature (VT) NMR study of an adduct formation process was also presented here as an example of the advantageof 10B over 11B. Copyright © 2012 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.

Keywords: NMR; 10B; 11B; background; suppression; boranes; relaxation; DEPTH

* Correspondence to: Péter Király, Department of NMR Spectroscopy, Institute ofStructural Chemistry, Chemical Research Center of the Hungarian Academy ofSciences, Pusztaszeri út 59-67, H-1025, Budapest, Hungary. E-mail: [email protected]; [email protected]

Institute of Organic Chemistry, Research Centre for Natural Sciences, HungarianAcademy of Sciences, Pusztaszeri út 59-67, H-1025, Budapest, Hungary

Introduction

Boron NMR spectroscopy has a decisive role in the structuralstudies of boron containing organic and inorganic compounds[1,2].There are two naturally occurring NMR active boron isotopes: 10B(NA=19.9%) and 11B (NA=80.1%). Both nuclei has largerspin quantum number than ½; therefore, they also have nuclearelectric quadrupole moments (I10B = 3; Q10B = 8.46 fm2 andI11B = 3/2; Q11B = 4.06 fm2). 11B is commonly regarded as moresuitable for NMR because of its higher sensitivity and betterresolution at a given external magnetic field (Table 1). A reporteddrawback of 11B is its shorter relaxation time (T1

(B11)/T1(B10) = 0.635),

which leads to somewhat broader resonances on the frequencyscale (Hz)[3]. On the other hand, the chemical shift resolution is stillbetter for 11B, because its significantly higher Larmor frequencycompensates the broadening effect on the parts per million scale.Indeed, 10B NMR applications are limited to special cases, e.g.measurement of J-couplings[4–6], mechanistic investigations basedon 10B labelling[7], relaxation studies[8–10], boron neutron capturetherapy of cancer[11–13] and the solid-state NMR of boron dopeddiamond[14,15].

The appearance of broad background signals is a well-knownproblem of 11B NMR in solution[16,17] and in the solid-state[18–20].The signal-to-background (S/B) ratio of solution 10/11B NMRspectroscopy will be discussed in this paper. Either the asymmetryof the boron environment or the molecular correlation time can bethe source of extensive line broadening, which makes theresonances of interest to comparable or even less intensive thanthe background signals. Low solubility may also result in insuffi-cient (S/B) ratio. I have recently found that the measurement of10B instead of the common 11B is a general and efficient solutionto completely get rid of the background signals. Interestingly, thisunusual aspect of 10B NMR spectroscopy seems to be neglected

Magn. Reson. Chem. (2012)

apart from the characterization of a new heterocyclic ringsystem[21] and the NMR spectroscopic investigations of metal-freecatalyst systems[22,23].

The aim of this paper was to discuss the practical aspects of 10Band 11B solution NMR experiments with respect to the (S/B) ratio.The unique isotopic effect observed for the boron backgroundsignals will be discussed in detail. The seemingly ‘background-free’ 10B NMR will be compared with the background suppressionefficiency of 11B DEPTH[24–27] for a selected set of boroncompounds (Scheme 1). An example is also presented for theapplication of 10B: the variable temperature (VT) NMR study of thecomplex formation between 2,4,6-trimethylphenyl-bis(pentafluoro-phenyl)borane (9) and pyridine. The emphasis will be on thegeneral applicability by using readily available equipments;therefore, custom build modifications aiming the physicalremoval of the boron-background producing components arebeyond the scope of this paper.

Results and Discussion

Background signals were detected in the absence of the sampletube by using either 11B or 10B NMR (Fig. 1). Interestingly, the10B NMR spectrum was ‘background-free’, but broad resonanceswere detected in the 11B NMR spectrum. The intensity and the

Copyright © 2012 John Wiley & Sons, Ltd.

P. Király

shape of the background signal may vary between spectrometersand probes, but the significant difference between 11B and 10Bremains expectedly. This observation has turned my attentionto the application of 10B instead of the common 11B. Note thathowever, this implies a reduced resolution and sensitivity, whichshould also be considered in conjunction with the (S/B) ratio.

Table 1. Selected properties of NMR active boron nuclei

10B 11B

Natural abundance, NA % 19.9 80.1

Larmor frequency at 14.1 T, nMHz�1 64.5 192.5

Nuclear spin quantum number, I 3 3/2

Nuclear electric quadrupole moment, Q/fm2 8.46 4.06

Relative receptivity, D13C 23.2 777

Scheme 1. Schematic structures of the studied boron compounds.

Figure 1. Solution boron NMR spectra were measured in the absence ofthe NMR sample tube. The background signals are readily noticeable in11B (top), but not visible in 10B (bottom).

wileyonlinelibrary.com/journal/mrc Copyright © 201

Although, the 10B NMR spectrum seemingly ‘background-free’,but strictly speaking the background signals were just notobserved. It will be shown that the ‘background-free’ appearanceof the 10B NMR spectrum is an inherent feature irrespectively ofthe sample. Therefore, its application is only limited by its naturaldisadvantages with respect to sensitivity and resolution.

A few notations are worthy to mention at this point to avoidambiguities. The term ‘sample’ refers to the compound that wasdissolved in a suitable deuterated solvent and everything else,which also contributes to the NMR spectrum, is treated as the‘background’. Accordingly, the term S/B ratio is introduced likewisethe often mentioned signal-to-noise (S/N) ratio. Although, theprecise determination of S/B is more difficult than S/N, but thedifference between insufficient, acceptable and adequate S/B ratiois obviously seen by comparing the NMR spectra. The boronbackground in solution NMR stem from two parts: the NMR tubemade of borosilicate glass and boron containing parts of the NMRprobe. The former is negligible in our system, but this may varynotably between spectrometers. However, the origin of the back-ground signal has no effect on the discussion regarding 10B,because nothing has to be adjusted to suppress the backgroundsignal (it is simply invisible). The main point is that the backgroundis the NMR signal of solid-state materials. Therefore, the S/B ratio isgiven by the proportion of the S/N ratio of the sample solutionand the S/N ratio of the boron containing solid-state materials.Indeed, the difference between solution and solid-state boronNMR reflects the S/B ratio. An isotopic effect can also be deducedby comparing the S/N of solution and solid-state 10/11B NMRspectroscopy, respectively. To the best of the author’s knowledge,the different S/B ratios of solution 10B and 11B NMR are discussedin detail for the first time.

Let us consider first the comparison of solution 11B and 10B NMR.Theoretically, the relative receptivity of 11B is 33 as compared with10B (Table 1). However, the experimentally observed S/N advantagedepends on the different 10B and 11B relaxation of the sample ofinterest. It is known[3] that the effect of the larger 10B quadrupolemoment is more than compensated by its larger nuclear spinleading to longer relaxation times (T1

(B11)/T1(B10) = 0.635) and

accordingly sharper resonances for 10B. Compound (2) has verysharp 11B (14.3Hz at half-height), but even sharper 10B (8.0Hz athalf-height) resonance. The S/N ratio and T1 relaxation of (2), (5)and (6) are given in Table 2 to illustrate the effect of relaxation withexperimental data. As the efficiency of quadrupolar relaxationincreases, the S/N advantage of 11B reduces. It is only about fivefor compounds having T1 about a few milliseconds. This featurebecomes important for compounds that are difficult to observebecause of insufficient S/B ratio. The background problem is oftena consequence of rapid quadrupolar relaxation because of largequadrupolar coupling constant (CQ). The fast relaxation allowsone to omit the relaxation delay for such compounds, becausethe acquisition time is more than enough for this purpose.

Table 2. Sensitivity advantage of 11B over 10B

(2) (5) (6)

(S/N)10B 330 34.1 26.6

(S/N)11B 4040 180 150

(S/N)11B/(S/N)10B 12.2 5.3 5.610BT1/

11BT1 (ms) 43.5/28.9 2.9/2.0 1.9/1.2

S/N, signal-to-noise.

2 John Wiley & Sons, Ltd. Magn. Reson. Chem. (2012)

Comparison of single-pulse 10B, 11B and 11B DEPTH

Accordingly, it is possible to efficiently increase the S/N by addingmore than 36000 scans within 1 h, whereas the S/B determinedquality of the spectra cannot be influenced by this way. Conse-quently, the common S/N disadvantage of 10B is reduced reasonablyfor relaxation broadened resonances, which are difficult todistinguish from the background. On the other hand, if backgroundsuppression is needed because of the low concentration of acompound having sharp resonances, then the sensitivity of 10Bmay not be sufficient as compared with 11B in conjunction withthe necessity of longer repetition rates and the increased S/Ndisadvantage of 10B.

Solid-state 11B and 10BMASNMR spectra of (1) and (8a) were alsocompared in order to illustrate experimentally the effect of CQ on theS/N. Sodium-borohydride represents an example for highlysymmetric boron environment because of its very small quadrupolarcoupling constant (CQ< 7 kHz) according to literature data[28]. Inthis case, both 11B and 10B MAS NMR spectra (Figs 2(c,d)) werereadily observed. The S/N ratio of the former was better than the

Figure 2. Solid-state 10/11B MAS NMR spectra: (a) 11B MAS NMR of (8a);(b) 10B MAS NMR of (8a); (c) 11B MAS NMR of (1) and (d) 10B MAS NMR of (1).

Figure 3. The effect of CQ on the boron signal intensity is shown by the ca

Magn. Reson. Chem. (2012) Copyright © 2012 John Wiley

latter, as one would expect. Similar experiments were also carriedout for (8a), which has much larger quadrupolar coupling constant(CQ=4.75MHz)[29]. In this case, the S/N ratio of the 11B MAS NMRspectrum (Fig. 2(a)) was not only better than the S/N of the 10BMAS NMR spectrum (Fig. 2(b)) but no signals were detected inthe latter. This is an inherent consequence of the different nuclearspin of the boron isotopes. 11B has half-integer spin (I=3/2) leadingto an intensive central-transition (CT) signal. 10B has integerspin (I=3), accordingly there is no CT available. Note that thefundamental existence of the CT for half-integer nuclei has alreadyfavoured the 11B applications over 10B because of the availability ofMQ-MAS type experiments in the solid-state[30–34].

Solid-state 10B and 11B NMR spectrum simulations were carriedout by using the SIMPSON package[35] to present the effect ofquadrupolar coupling on the signal intensity (Fig. 3.). The totalintensity (integral) of each simulated spectrum was correctedaccording to the relative receptivity (D13C) in order to take intoaccount the different nuclear properties (natural abundance, spinand giromagnetic ratio), whereas SIMPSON has provided the CQdependent shape of the powder patter. The CT of 11B obviouslypossesses much of the signal intensity, but for 10B the much lessintensity is additionally spread over the frequency scale.

The intensity of the highest peak in 11B is just reduced as thequadrupolar coupling increases, while it has a dramatic effecton 10B. Therefore, the S/N advantage of 11B over 10B increaseswith CQ. However, the boron environment has an opposite effectin solution, because the line broadening induced by quadrupolarrelaxation is more effective for 11B. Indeed, background signaltheoretically exists in 10B solution NMR, but it cannot be observedexperimentally as being spread over a very large frequency scale.This makes the solution 10B NMR spectra virtually ‘background-free’ in contrast to 11B.

A set of readily available boron compounds (1–8) was chosen tocompare 11B and 10B NMR, in particular considering the S/B ratio.The background signal intensity is negligible in the presence of(1), (2) and (3), which makes the background suppression DEPTHexperiments pointless for them (Supporting Information, Fig. S1.).These compounds have significantly sharper resonances ascomparedwith (4–8) in conjunctionwith theirmuch longer relaxationtimes (Table 3). The boron background is usually not visible next to

lculated 11B (left) and 10B (right) powder patterns.

& Sons, Ltd. wileyonlinelibrary.com/journal/mrc

Table 3. Boron T1 relaxation data at 14.1 T

Sites-(compounds) d10/11B (ppm) 10BT1 (ms) 11BT1 (ms)

(1) �36.1 10600� 200 6400� 200

(2) 0 42.8 28.3

1,3-(3) 12.2 48.8� 0.2 32.4� 0.1

6,9-(3) 10.3 17.2� 0.2 11.5� 0.1

10,8,5,7-(3) 0.1 26.5� 0.2 16.8� 0.1

2,4-(3) �35.8 75.2� 0.9 47.5� 0.4

(4) 19.5 6.9� 0.1 4.5� 0.1

(5) 16.4 2.9� 0.1 2.0� 0.1

(6) 31.4 1.9� 0.1 1.2� 0.1

(7a) 67.5 0.85� 0.01 }

(7b) 45.5 1.39� 0.1 }

(7c) 29.1 0.74� 0.1 }

(8a) 76.3 0.42� 0.05 }

(8b) 50.1 1.0� 0.1 }

(9) 69.2 0.3� 0.05 }

[9-Py] �0.2 1.5� 0.1 }

} Relaxation rates were determined only for 10B due to the low S/Bratio of the 11B NMR spectra.

P. Király

sharp resonances. Indeed, the S/B ratio is often sufficient to apply11B NMR without considering any kind of background suppression.Compound (3) is a representative example for that. If dilutedsolutions of compounds having relatively long relaxation timesand thereby sharp resonances should be measured, then theinherent S/N disadvantage of 10B makes it much less reliable inconjunction with the necessity of longer repetition rates. Theequality of absolute vertical scalewas preserved for the comparisonof 11B single-pulse and DEPTH experiments (Fig. 4.). The commonsensitivity advantage of 11B over 10B is shown only in the case of(4), but otherwise at least ten times more scans were acquiredfor each 10B NMR spectra. The background signals are just visible(after 16 scans) in the single-pulse 11B NMR spectrum of (4) andalmost completely suppressed by using the DEPTH pulse sequence

Figure 4. Boron NMR spectra of compounds (4)–(8) were acquired by usin(a) 68mM H3BO3 (4) in D2O; (b) 60mM triphenyl-borate (5) in CDCl3; (c(e) 58mM triphenyl-borane (7) in benzene-d6; (f) 38mM trimesityl-borane (8direct detection probe with the exception of (d) that was measured by using


(Fig. 4(a)). 10B NMR spectrum of (4) does not contain anybackground signals, as one would expect for any compounds(Fig. 1.). Similar behaviour of the boron signals were observedfor (5) and (6). In conjunction with the efficient backgroundsuppression of DEPTH, the peak heights of the resonances of(4), (5) and (6) drop about 21%, 24% and 30%, respectively. Thisbehaviour can be rationalized by the co-addition of two effects:relaxation under the additional radiofrequency pulses and thenon-ideal refocusing performance. The DEPTH sequence consistsof two additional p pulses in its simplest form, therefore it isgenerally longer than the single-pulse excitation. In hands of theresult of inversion recovery experiments (Table 3), it is straight-forward to estimate the relaxation induced intensity differencebetween single-pulse and DEPTH experiments: 4% (4), 8% (5) and14% (6) (Supporting Information S2 for details). Although, thesecalculated values are less than the one observed experimentally,but the trend is well reflected. The additional sensitivity loss stemsfrom the non-ideal performance of the radiofrequency pulses. Inorder to test the effect of the non-ideal probe performance, direct(Fig. 4(c)) and indirect (Fig. 4(d)) probes were compared in thisrespect. The background suppression of DEPTH is significantlyworse for the indirect probe, but there is no big difference in thequality of the 10B NMR spectra. This suggests that the efficiencyof 11B DEPTH experiment depends not only on the calibration butalso on the general performance of the NMR probe. Indeed, theacquisition of a virtually ‘background-free’ single-pulse 10B NMRspectrum is more straightforward and robust as compared withbackground suppression by using 11B DEPTH.

The advantage of 10B over 11B shows up spectacularly for therelaxation broadened resonances of (7) and (8). The commerciallyavailable triphenyl-borane (7) was packed under argon, becausein moist air it slowly converts to benzene and triphenylboroxine.The sample was handled in ambient conditions during dissolutionwith intent to observe a mixture of species because of the partialdegradation of the sample. The quality of the 10B NMR spectrumis better than that of the background signal distorted 11B NMRspectra (Fig. 4(e)). 11B DEPTH performs here nicely, but it was notpossible to completely reduce the background signal intensity.

g single pulse 11B (left), 11B-DEPTH (middle) and single-pulse 10B (right).) and (d) 55mM 4-isoquinolineboronic-acid-pinacol-ester (6) in CDCl3;) in toluene-d8. Experiments were carried out by using double resonancea triple-resonance indirect detection probe.



Among the selected set of boron compounds, the rapid relaxationmakes trimesityl-borane (8) the most challenging one with respectto the S/B ratio. Figure 4(f) shows the 11B single pulse (left),11B-DEPTH (middle) and 10B (right) NMR spectra of commerciallyavailable trimesityl-borane-(97%) dissolved in toluene-d8. Themajor component (8a) has only one resonance, which is assignedto the trigonal boron centre. However, several intensive resonancesappear in the 11B NMR spectrum because of the low S/B ratio. Theadvantage of 10B over 11B is obvious in this case. The 10B NMRspectrum also reflects the purity of the sample. The resonance ofa minor component (8b) is just visible in the 11B NMR spectra, butit is readily detected by using 10B. Furthermore, the impurity ofthe sample could also be quantified by 10B NMR.

Structural analogous of (8) have recently played an importantrole in the success ofmetal-free catalyst systems, which often calledas ‘Frustrated Lewis Pairs’ (FLPs)[36–38]. These triaril-boranes areknown to form intermolecular complexes with the appropriateLewis-bases[39–41]. FLP-based catalytic hydrogenation by using2,4,6-trimethylphenyl-bis(pentafluorophenyl)borane (9) wasreported in conjunction with the 1H, 19F and 10B NMR spectro-scopic investigation of its complex formation with quinoline[23].11B NMR was not applied in this case. An analogous systemwas chosen to compare the applicability of 10B and 11B (Fig. 5).NMR spectroscopy is an ideal tool to determine either thestructure of the borane complex or the association constant ofthe process (Kass). The thermodynamic parameters of the dative

Figure 5. The adduct formation process between (9) and pyridine.

Figure 6. Temperature dependence of the boron NMR spectra of (9) in the11B-DEPTH (middle) and 10B single pulse (right) experiments, respectively. 480respectively.


bond formation between (9) and pyridine are determined by usingthe observed temperature dependence of the ratio of free andcomplexed boranes.

The temperature dependence of 11B single pulse (left), 11B-DEPTH(middle) and 10B (right) NMR spectra of the mixture is shown inFig. 6. The single-pulse 11B NMR spectra are distorted by the strongbackground signal, which also appears in the 11B DEPTH spectra. Incontrary, the 10B NMR spectra are practically ‘background-free’facilitating the reliable integration of the resonances. The amountof free borane slightly increases at higher temperatures asexpected. However, accurate quantification by using 10B NMRcould only be achieved by taking into account the effect ofrelaxation during the excitation pulse and the transmitter/receiverdelay. This problem arises from the markedly different relaxationrates of the free and complexed states, which were determinedby using 10B inversion recovery experiment (Table 3). The relaxationinduced intensity loss was estimated for the free (31%) and thecomplexed (7%) states. The equilibrium concentrations were alsodetermined by using 1H and 19F NMR at room temperatureconfirming the necessity and accuracy of the correction (Table 4).However, the coalescence of the 1H and the 19F NMR resonanceshampered the comparison at higher temperatures. Therefore,temperature dependence of the association constant was deter-mined from only by using 10B NMR data. Enthalpy (ΔH=�26.5� 0.2kJmol�1) and entropy (ΔS=�52� 8 Jmol�1) differences werecalculated by using the Van’t Hoff plot (Fig. 7.).

presence of pyridine-d5 (0.5 eqv.) acquired by using 11B single pulse (left),and 1280 number of scans was acquired for 11B and 10B NMR experiments,

Table 4. Equilibrium concentrations

Nucleus (temperature, �C) Free borane (9)(mol dm�3)

Complex [9-Py](mol dm�3)

1H (25) 0.0377 0.022219F (25) 0.0374 0.022510B (25)} 0.0330} 0.0270}

10B (25) 0.0374 0.022610B (40) 0.0396 0.020410B (50) 0.0412 0.018810B (60) 0.0439 0.0161

} These were calculated by neglecting the effect of relaxation.


Figure 7. Van’t Hoff plot for the complex formation process between (9)and pyridine.

P. Király

Conclusions

The ‘background-free’ appearance of solution 10B NMR spectra incontrary to 11B NMR spectra was presented. The unusual isotopiceffect could be rationalized by the different integer versus non-integer spin of 10B and 11B, respectively. Therefore, S/B ratio of 10BNMR spectra is generally better than that of 11B irrespectively ofthe sample of interest. Therefore, the measurement of 10B insteadof 11B is proposed as a feasible alternative to observe completelybackground-free boron NMR spectra. Efficient suppression of thebackground signals could be achieved by 11B DEPTH experiments,which often makes the application of 10B marginal. However, ifbackground suppression is needed for relaxation broadenedresonances (T1< 1ms) then 10B NMR performs better than11B DEPTH. The comparison of single-pulse 10B, 11B and 11B DEPTHexperiments confirmed this. The advantage of 10B over 11B wasbest viewed for the VT NMR study of a complex formation process.Even the accurate quantification of the free and the complexedstates was possible by using 10B NMR. Indeed, if residualbackground signals still hamper the proper interpretation of 11BDEPTH spectrum, then single-pulse 10B NMR is a feasible alternative.

Experimental

Compounds were purchased from Aldrich Budapest, Hungaryexcept (3) and (7), which were purchased from AlfaAesar. TheNMR experiments have been carried out without further purificationof the compounds. 2,4,6-trimethylphenyl-bis(pentafluorophenyl)-borane (9) was prepared according to literature[22]. Pyridine(Ctotal

pyridine = 30mmol) was added to the solution of (9) dissolvedin toluene-d8 (Ctotal

borane = 60mmol). This has been used to calculatethe concentrations of the species [Ctotal

pyridine = (free pyridine)+(complex); Ctotal

borane = (free borane)+(complex)] by using the integralratio of free and datively bound borane derived from 10B NMR. Deu-terated pyridine ampoule was opened under nitrogen and thesample was prepared in a glow box in order to avoid water content.Solution boron NMR spectra were obtained by using 9.4 T (400MHzfor 1H; 128.3MHz for 11B and 42.9MHz for 10B) and 14.1 T (600MHzfor 1H; 192.4MHz for 11B and 64.4MHz for 10B) Varian NMR SYSTEMspectrometers (Agilent Technologies, Santa Clara, California, USA).Similar boron background signals were observed at both systemseven in the absence of the sample tube. Direct detection tunable


dual-broadband {1H-19F}/{31P-15N} and indirect detection triple-resonance 1H/13C/{31P-15N} Chemagnetics probes equipped witheither Z-gradients or XYZ-gradients were compared. Solid-stateNMR experiments were carried out at 9.4 T (400MHz for 1H,128.3MHz for 11B and 42.9MHz for 10B) by using a T3type Chemagnetics {1H-19 F}/{31P-15N} double-resonance CP/MASprobehead. Samples were packed into 4.0mmZirconia rotors usingthe standard Vespel drive tip and Teflon spacers. Solution and solid-state 10/11B NMR spectra are referenced [20] to BF3�Et2O (0ppm)dissolved in CDCl3 and powderedNaBH4 (�42.06ppm), respectively.The same low-pass filter on the low frequency channel and adeuterium selective cut-off filter (for solution experiments) wereused instead of band-selective filters for both boron nuclei in orderto match the spectrometer setup as much as it is possible. Thesame experimental parameters were used for both boron nucleito compare sensitivity: power levels were calibrated to achievep/2= 20ms in order to ensure that the excitation bandwidths arealso similar for the two nuclei; 81 920 complex points and 50 kHzspectral width using the build-in advantages of the Direct Drivereceiver technology (rof2 = 25, alfa = 10 and ddrtc = 42). This hasled to flat baseline, which eliminates the need of first order phaseand/or baseline corrections. Relaxation delay was adjusted to atleast five times the relaxation time of each compound. Noweightingfunctions were applied for sharp resonances. The same relaxationdelay (at least 1 s), the same number of scans and identical verticalscale were carefully adjusted for the comparison of 11B single-pulseand DEPTH experiments on compounds (4)–(8). Fine calibration ofthe refocusing p pulses was carried out on each sample in order tooptimize the background suppression efficiency of 11B DEPTHexperiment. In case of broad resonances (7, 8 and 9), the numberof complex points (32 k) and the relaxation delay (0.5 s) werereduced. Total experiment time was about 12 h, and 20Hz line-broadening was applied prior to FT. Boron chemical shifts werereferenced to BF3•Et2O. The standard inversion recovery experi-ment was used to estimate 10/11B T1 values.

Acknowledgements

The support of the Hungarian GVOP-3.2.1.-2004-04-0210/3.0project is gratefully acknowledged. The 2,4,6-trimethylphenyl-bis(pentafluorophenyl)borane (9) sample was a kind gift ofDr. Hasan Mehdi.

References[1] R. K. Harris, B. E. Mann, NMR and the Periodic Table, Acad. Press,

London, 1978. (ISBN: 0123276500)[2] H. Nöth, B. Wrackmeyer, Nuclear Magnetic Resonance Spectroscopy

of Boron Compounds, Springer-Verlag, Berlin, 1978. (ISBN:3-540-08456-8)

[3] J. W. Akitt. J. Magn. Reson. 1970, 3, 411–414.[4] E. J. Stampf, A. R. Garber, J. D. Odom, P. D. Ellis. J. Am. Chem. Soc.

1976, 98, 6550–6554.[5] F. Bachmann, H. Nöth, H. Pommerening, B. Wrackmeyer, T. Wirthlin.

J. Magn. Reson. 1979, 34, 237–239.[6] W. Keller, W. Haubold, B. Wrackmeyer. Magn. Reson. Chem. 1999, 37,

545–550.[7] G. M. Edvenson, D. F. Gaines. Inorg. Chem. 1990, 29, 1210–1216.[8] T. J. Marks, L. A. Shimp. J. Am. Chem. Soc. 1972, 94, 1542–1550.[9] G. H. Penner, S. I. Daleman, A. R. Custodio. Can. J. Chem. 1992, 70,

2420–2423.[10] A. Baram, P. Bendel. J. Magn. Reson. 1997, 129, 10–18.[11] L. Adams, S. Tomlinson, J. Wang, S. N. Hosmane, J. A. Maguire, N. S.

Hosmane. Inorg. Chem. Comm. 2002, 5, 765–767.[12] R. F. Barth, J. A. Coderre, M. G. H. Vicente, T. E. Blue. Clin. Cancer Res.

2005, 11, 3987–4002.



[13] M. Bonora, M. Corti, F. Borsa, S. Bortolussi, N. Protti, D. Santoro,S. Stella, S. Altieri, C. Zonta, A. M. Clerici, L. Cansolino, C. Ferrari,P. Dionigi, A. Porta, G. Zanoni, G. Vidari. Appl. Radiot. Isot. 2011,69, 1702–1705.

[14] M. Murakami, T. Shimizu, M. Tansho, Y. Takano, S. Ishii, E. A. Ekimov,V. A. Sidorov, K. Takegoshi. Diamond Relat. Mater. 2009, 18,1267–1273.

[15] M. Murakami, T. Shimizu, M. Tansho, Y. Takano, S. Ishii, E. A. Ekimov,V. A. Sidorov, K. Takegoshi. Physica C 2010, 470, S625–S626.

[16] University of Ottawa NMR Facility Blog: 11B background signals(G. Facey) http://u-of-o-nmr-facility.blogspot.com/2008/06/11-b-background-signals.html

[17] University of Ottawa NMR Facility Blog: Background suppression inliquids (G. Facey) http://u-of-o-nmr-facility.blogspot.com/2010/04/background-suppression-in-liquids.html

[18] S. Prasad, T. M. Clark, T. H. Sefzik, H.-T. Kwak, Z. Gan, P. J. Grandinetti.J. Non-Cryst. Solids 2006, 352, 2834–2840.

[19] Z. Szalay, J. Rohonczy. J. Non-Cryst. Solids 2007, 353, 295–301.[20] J. W. E. Weiss, D. L. Bryce. J. Phys. Chem. A 2010, 114, 5119–5131.[21] D. Takács, P. Király, I. Nagy, P. Bombicz, O. Egyed, Z. Riedl, G. Hajós. J.

Organomet. Chem. 2010, 695, 2673–2678.[22] G. Erős, H. Mehdi, I. Pápai, T. A. Rokob, P. Király, G. Tárkányi, T. Soós.

Angew. Chem. Int. Ed. 2010, 49, 6559–6563.[23] G. Erős, K. Nagy, H. Mehdi, I. Pápai, P. Nagy, P. Király, G. Tárkányi, T.

Soós. Chem. Eur. J. 2012, 18, 574–585.[24] M. R. Bendall, R. E. Gordon. J. Magn. Reson. 1983, 53, 365–385.


[25] M. R. Bendall, D. T. Pegg. Magn. Reson. Med. 1985, 2, 91–113.[26] D. G. Cory, W. M. Ritchey. J. Magn. Reson. 1988, 80, 128–132.[27] J. Feng, J. A. Reimer. J. Magn. Reson. 2011, 209, 300–305.[28] L. Niemelä, E. Ylinen. Phys. Lett. A 1970, 31, 369–370.[29] D. L. Bryce, R. E. Wasylishen, M. Gee. J. Phys. Chem. A 2001, 105,

3633–3640.[30] L. Frydman, J. S. Harwood. J. Am. Chem. Soc. 1995, 117, 5367–5368.[31] A. Medek, J. S. Harwood, L. Frydman. J. Am. Chem. Soc. 1995, 117,

12779–12787.[32] C. Fernandez, J. P. Amoureux. Chem. Phys. Lett. 1995, 242, 449–454.[33] R. Hajjar, Y. Millot, P. P. Man. Prog. Nucl. Magn. Reson. Spectrosc. 2010,

57, 306–342.[34] D. Carnevale, V. del Amo, D. Philp, S. E. Ashbrook. Tetrahedron 2010,

66, 6238–6250.[35] M. Bak, J. T. Rasmussen, N. C. Nielsen. J. Magn. Reson. 2000, 147,

296–330.[36] G. C. Welch, R. R. S. Juan, J. D. Masuda, D. W. Stephan. Science 2006,

314, 1124–1126.[37] G. C. Welch, L. Cabrera, P. A. Chase, E. Hollink, J. D. Masuda, P. Wei,

D. W. Stephan. Dalton Trans. 2007, 3407–3414.[38] A. L. Kenward, W. E. Piers. Angew. Chem. Int. Ed. 2008, 47, 38–41.[39] S. J. Geier, D. W. Stephan. J. Am. Chem. Soc. 2009, 131, 3476–3477.[40] S. J. Geier, A. L. Gille, T. M. Gilbert, D. W. Stephan. Inorg. Chem. 2009,

48, 10466–10477.[41] T. Voss, T. Mahdi, E. Otten, R. Fröhlich, G. Kehr, D. W. Stephan,

G. Erker. Organometallics 2012, 31, 2367–2378.


http://u-of-o-nmr-facility.blogspot.com/2008/06/11-b-background-signals.html

http://u-of-o-nmr-facility.blogspot.com/2008/06/11-b-background-signals.html

http://u-of-o-nmr-facility.blogspot.com/2010/04/background-suppression-in-liquids.html

http://u-of-o-nmr-facility.blogspot.com/2010/04/background-suppression-in-liquids.html

background-free solution boron nmr spectroscopy

Documents