two-dimensional j-spectra with absorption-mode lineshapes

1

2

3

4

5

67

8

9

10

11

12

13

14

1516

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

YJMRE 3821 No. of Pages 7, Model 5+

12 September 2007 Disk UsedARTICLE IN PRESS

www.elsevier.com/locate/jmr

Journal of Magnetic Resonance xxx (2007) xxx–xxx

OF

Communication

Two-dimensional J-spectra with absorption-mode lineshapes

Andrew J. Pell, James Keeler *

University of Cambridge, Department of Chemistry, Lensfield Road, Cambridge, CB2 1EW, UK

Received 4 July 2007; revised 30 August 2007

DPR

O

Abstract

Two-dimensional J-spectroscopy offers the possibility of a complete separation of chemical shifts and J-couplings. However, the use-fulness of the experiment is considerably reduced by the fact that peaks in the spectra have the phase-twist lineshape. We present a simplenew spectroscopic method for recording J-spectra in which the peaks are both in the absorption mode and retain their natural intensities,albeit at the cost of a considerable reduction in the signal-to-noise ratio. No special data-processing is required. The method is tested onquinine, and the steroid dehydroisoandrosterone.� 2007 Elsevier Inc. All rights reserved.

Keywords: NMR; J-spectroscopy; Broadband homonuclear decoupling

TE

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

UN

CO

RR

EC1. Introduction

In this communication we describe a new method forrecording two-dimensional J-spectra with absorption-mode lineshapes. After tilting by 45�, such spectra offer acomplete separation of chemical shifts in one dimensionand J-couplings in the other, thus facilitating the analysisof overlapping spectra [1]. However, when recorded inthe conventional way, J-spectra show the phase-twist line-shape which is not suitable for high-resolution work on ac-count of the broad dispersive component. There are anumber of ways of eliminating this phase-twist lineshape,but these all suffer from various drawbacks. The most pop-ular method, which is to use a pseudo-echo weighting func-tion to eliminate the dispersive component, results in severeintensity distortions [2]. Other methods which rely on spe-cial data processing methods have proved to be successful,but have not achieved wide acceptance [3–10]. In contrast,our method is simple to implement, requires no special dataprocessing, and retains the natural intensities of the con-ventional one-dimensional proton spectrum. A key element

58

59

60

61

1090-7807/$ - see front matter � 2007 Elsevier Inc. All rights reserved.

doi:10.1016/j.jmr.2007.09.002

* Corresponding author. Fax: +1223 336913.E-mail address: [email protected] (J. Keeler).

Please cite this article in press as: A.J. Pell, J. Keeler, Two-dimensio(2007), doi:10.1016/j.jmr.2007.09.002

of our pulse sequence is the combination of a selective 180�pulse with a simultaneous weak gradient—a pulse sequenceelement introduced by Zangger and Sterk [11]. This inge-nious element has the effect of separately inverting eachspin while leaving all its coupled partners unaffected.

We have recently shown that this separation of shiftsand couplings, while retaining absorption-mode lineshapes,can also be obtained from a modified anti z-COSY exper-iment [12]. The experiment described in this communica-tion is an alternative to the anti z-COSY approach, andhas some advantages in terms of simplicity. However, incommon with our earlier experiment, the generation ofan absorption mode spectrum comes at the cost of a signif-icant reduction in the signal-to-noise ratio.

Our method for recording absorption-mode J-spectra isbased on the well-known idea that a P-type and an N-typespectrum can be combined to give an absorption-modespectrum [13,14]. Both the P- and N-type spectra havephase-twist lineshapes, but if the x1 axis of one of the spec-tra is reversed, and the two-spectra added together, it turnsout that the dispersive parts of the phase-twist lineshapescancel. The pulse sequence for conventional J-spectros-copy, along with the coherence transfer pathway (CTP) isshown in Fig. 1(a). We cannot really describe this as theP- or N-type spectrum, since the coherence order changes

nal J-spectra with absorption-mode lineshapes, J. Magn. Reson.

mailto:[email protected]

REC

T

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

RFt1/2 t1/2 t2

(a)

RF

G

t1/2 t1/2 t2(b)

p+1

0-1

p+1

0-1

RF

G

t1/2 t1/2 t2(d)

RF

G

t1/2 t1/2 t2

(c) φ

φ

φ

p+1

0-1

p+1

0-1

Fig. 1. Pulse sequences for recording absorption-mode J-spectra. Shownin (a) is the sequence and CTP for the conventional J-spectroscopyexperiment. The filled rectangle represents a 90� pulse, and the unfilledrectangle represents a 180� pulse. Sequence (b) is for the anti J-spectrumwhich, when combined with the J-spectrum, results in an absorption-modelineshape. The selective 180 pulse is indicated by the Gaussian profile. Inpractice, the sequences in (c) and (d) are used to record the J- and anti J-spectra respectively, so that both spectra have the same intensity. Note theuse of the BIPs in (c) and (d), which are indicated by the unfilled rectangleswith the diagonal stroke. All pulses have phase x apart from the selectivepulses, which are phase-cycled as / = [x,y,�x,�y], along with the cycle/rx = [x,�x,x,�x] for the receiver phase.

2 Communication / Journal of Magnetic Resonance xxx (2007) xxx–xxx



R

in the middle of t1. Fig. 1(b) shows the CTP of the comple-mentary J-spectrum which will have a phase-twist in the

UN

CO

(a) (b)

ω1

πJ

-πJ

0

ω2

Ω1

Ω1-

πJ

Ω1+

πJ

ω2

Ω1

Ω1-

πJ

Ω1+

πJ

Fig. 2. Simulated spectra from one of the doublets of a two-spin system. The cthe multiplets tilt in the opposite direction, is shown in (b). The peaks have thereversed to give (c), which is then added to (a) to give the absorption-mode spare grey.


ED

PR

OO

F

opposite sense: we term this the anti J-spectrum. Note thatin this sequence the coherence order is �1 during the firsthalf of t1, and +1 during the second half, opposite to theCTP in Fig. 1(a).

Normally, we can switch between the P- and N-typeCTP simply by changing the phase cycling or gradientpulses. However, it is not possible to switch between theCTPs for the J- and anti J-spectra in this way on accountof the fact that there is no mixing period in the J-spectros-copy experiment.

It might be thought that placing a non-selective 180�pulse at the end of t1 in the anti J-spectroscopy sequencewould achieve the desired effect, since this will change thecoherence order from +1 to �1. While this is true, the pulsealso has the effect of flipping the spin states of all the pas-sive spins; as a result, when the J and anti J-spectra arecombined it turns out that the dispersive parts of the line-shape do not cancel. In fact, the desired result can only beachieved by applying a selective 180� pulse to one spin,such that the spin states of all of the passive spins arenot flipped. Such a selective experiment could only be per-formed on one spin at a time, which would not lead to avery useful two-dimensional spectrum.

The key point of our new method is that we can achievethe effect of this selective 180� pulse by placing the Zang-ger–Sterk pulse sequence element at the end of t1, as shownin Fig. 1(b). This element consists of a selective 180� pulsein combination with a weak gradient. The gradient strengthis chosen such that the spread of frequencies across thesample matches the width of the whole spectrum. Thismeans that a spin at a particular chemical shift comes intoresonance with the selective pulse at a particular position inthe sample. Thus, when the signal from the whole sample ismeasured, each spin has experienced precisely the kind ofselective 180� pulse we require. It is important that theselective pulse be phase cycled so as to select that part ofthe sample which experiences a refocusing pulse.

There are other ways in which this apparent selectiveinversion of all of the passive spins can be achieved, suchas the time-reversal method of Sørensen, Griesinger andErnst [15], and the anti z-COSY mixing scheme employed

(c) (d)

ω2

Ω1

Ω1-

πJ

Ω1+

πJ

ω2

Ω1

Ω1-

πJ

Ω1+

πJ

onventional J-spectrum is shown in (a), and the anti J-spectrum, in whichcharacteristic phase-twist lineshape. The x1-axis of the spectrum in (b) is

ectrum in (d). Positive contours are coloured black, and negative contours


Andrew Pell

Sticky Note

The degree symbol is missing from the sentence beginning "The selective 180 pulse..."

UN

CO

RR

EC

T

105

106

107

108

109

110

111

112

113

114

115

NH

N

O

HO

(a) (b)

HO

H

H H

O

1

2

34

56

7

8

9

10

1112

1314

15

16

17

18

19

1

23

45

620

78

9

1011

1216

13

14

15

17

1819

Fig. 3. The structures of (a) quinine, and (b) the steroiddehydroisoandrosterone.

1.0 1.0 2.0 1.0

9.0 8.5 8.0 7.5 7.0

1.0 1.0 2.1 1.1

9.0 8.5 8.0 7.5 7.0

4.5 4.0 3.5 3.0 2.5

4.5 4.0 3.5 3.0 2.5

3.0 4.11.0 1.0 1.0

1.91.

3.1 3.30.9 0.8 0.9

1.70.

(a)

(b)

(c)

(d)

20 15 11 16 {15,16}

81 {2,5} 7

Fig. 4. Conventional spectra and projections of the absorption-mode J-spectdecoupled. The spectrum is shown in two parts: (a) and (b) are the aromatic anshown in (b) is of the region shown in (a); likewise (d) is the projection of the regthe numbers in italics, and corresponds to the labelling in Fig. 3 (a). Experime

scans per t1 increment. The selective pulse was a Gaussian of length 57.6 ms, trmaximum value of 59.5 G cm�1. The B1 field strength and length of the BIP w4496 Hz, respectively. The acquisition time in t2 was 1.8 s; 85 t1 increments w

Communication / Journal of Magnetic Resonance xxx (2007) xxx–xxx 3




in our earlier work [12]. However, both of these methodslead to the generation of cross peaks which, on accountof the refocusing employed in t1, will fall directly on topof the absorption mode multiplets we are interested in.The beauty of the Zangger–Sterk pulse sequence elementemployed here is that, since it is a true 180� pulse, no crosspeaks are generated.

The way in which an absorption-mode spectrum is gen-erated is illustrated with the simulations, shown in Fig. 2,of a multiplet from a typical two-spin system. The conven-tional J-spectrum is shown in Fig. 2(a), the anti J-spectrum

ED

PR

OO

F

1.0 1.0 1.0 1.9

6.5 6.0 5.5 5.0 ppm

1.0 0.9 1.0 2.0

6.5 6.0 5.5 5.0 ppm

2.0 1.5 ppm

2.0 1.5 ppm

04.0

1.0

92.4

0.8

17 {12,13,14} 14

18 OH 10 19

ra of quinine; the projections show a single line at the shift, i.e. they ared vinyl region, whereas (c) and (d) are the aliphatic region. The projectionion shown in (c). The assignment of the peaks in (b) and (d) is indicated by

ntal details: the J-spectrum was recorded at 500 MHz for protons, with 36uncated at 1% [20]; the accompanying gradient had a strength of 3% of theere 20 kHz and 100 ls [17]. The spectral widths in x1 and x2 were 50 andere recorded, giving a maximum value of t1 of 1.7 s.


116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155




is shown in (b), and (c) is spectrum (b) with the x1-axis re-versed. The absorption-mode spectrum resulting from thecombination of (a) and (c) is shown in (d). Rather thancombining the spectra after the double Fourier transforma-tion, it is computationally more efficient to make theappropriate combinations of the time-domain data afterthe first Fourier transformation [13,14]. The facility pro-vided on many spectrometers for processing P- and N-typedata into absorption-mode spectra is exactly what isneeded to process the J- and anti J-spectra.

The drawback with the Zangger–Sterk element is that,for a given spin, only that part of the sample which ison-resonance with the selective 180� pulse contributes tothe signal. This means that there is a considerable reduc-tion in the size of the detected signal, of the order XB/(cG

l), where XB is the bandwidth of the selective pulse, c isthe gyromagnetic ratio, G is the gradient strength, and l

is the length of the sample that lies within the receiver coil.Thus, the more selective the pulse is made, the lower thesensitivity.

UN

CO

RR

EC

T(a) (b) (c)

(f) (g) (h)

(k) (l) (m)

(p) (q) (r)

1 8

18 OH

15 11

17

Fig. 5. Cross-sections taken parallel to x1 from the tilted absorption-mode J-spwhereas (f)–(j) are 5-fold expansions. The multiplets (k)–(t) of the aliphatic princrements. Multiplets (k), (l), and (n)–(t) are plotted on a scale that is expanumbers; multiplets (q)–(s) are associated with H12, H13, and H14, but theparameters for (a)–(j) are as given in the caption to Fig. 4. Multiplets (k)–(t) wein which 45 t1 increments were recorded, giving a maximum value of t1 of 0.9


RO

OF

For this 180� pulse to lead to the removal of the split-ting due to a particular coupling it must be sufficientlyselective such that it affects only one of the coupled part-ners. The selectivity of the pulse, and hence the sensitiv-ity of the experiment, is thus set by the closest separationbetween two spins which it is desired to decouple fromone another. In their original experiments Zangger andSterk estimated that they retained about 2% of the origi-nal signal, whereas Nilsson and Morris [16] use a 9 mspulse which we estimate retains around 8% of the signal.In our experiments we have tended to use somewhatlonger pulses, trading off the improvement in decouplingbetween closely-spaced multiplets with a further reduc-tion in the signal-to-noise ratio. For the experiments re-ported here, we estimate that around 0.3% of the signalis retained.

It is also necessary to make sure that the J- and anti J-spectra have the same intensity, otherwise the dispersivecomponent of the lineshape will not be eliminated. In prac-tice, we have found it convenient to use the pulse sequences

ED

P

20 10 0 -10 Hz

(d) (e)

(i) (j)

(n) (o)

(s) (t)

5 2 7

10 19 19

16 15 16

14

ectrum of quinine. Multiplets (a)–(e) are plotted on the same vertical scale,otons were taken from a separate experiment with a smaller number of t1

nded by five times relative to (m). The assignment is given by italicizedassignment of which is unclear. Experimental details: the experimentalre taken from a spectrum in which the acquisition time in t2 was 0.9 s, ands. The number of scans per increment was 160.


Andrew Pell

Sticky Note

Please could you ensure that the mathematical expression on lines 130-131 is not broken over two lines.

Andrew Pell

Cross-Out

Andrew Pell

Replacement Text

but the assignment of which is which is unclear.

CO

RR

EC

TED

PR

OO

F

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

1.0 1.0

1.0 2.0 2.0 2.2 3.1 6.1 2.0 5.1

1.0 0.6 0.6 0.8 0.9 1.6 2.7 1.7 0.9 0.9 0.8 1.7 0.8 0.8

3.0

5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm

5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm

2.5 2.4 2.3 2.2 2.1 2.0 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 ppm

2.5 2.4 2.3 2.2 2.1 2.0 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 ppm

6 3

16β

4α 4β 7β

16α

OH

{8,1α,2α}

{11α,7α,12β}

15α 15β2β 11β 14 12α

1β

19

9

18

*

1.01.0

(a)

(b)

(c)

(d)

Fig. 6. The conventional spectrum and projection of the absorption-mode J-spectrum of dehydroisoandrosterone. In (a) is shown the conventionalspectrum of the full chemical shift range. The projection of the J-spectrum of the same region is shown in (b). Expansions of the crowded region of theconventional and decoupled spectra are shown in (c) and (d). In the projection (d), the * shows the position of the residual intensity of the strong couplingartefact. The assignment in (b) and (d) corresponds to the atom numbering in Fig. 3(b). Experimental details: the J-spectrum was acquired with 44 scansper t1 increment. The spectral widths in x1 and x2 are 100 and 2561 Hz, respectively. The acquisition time in t2 was 1.6 s; 200 t1 increments were recorded,giving a maximum value of t1 of 2.0 s. All other parameters are as given in the caption to Fig. 4.




UN

shown in Fig. 1(c) and (d) to record the required pair ofcomplementary spectra.1

In the J-spectroscopy experiment in (c), the selectivepulse and gradient combination is inserted before t1, sothat the same part of the sample is selected as for the antiJ-spectroscopy experiment shown in (d); the two spectratherefore have the same intensities. Both sequences alsouse the broadband inversion pulses (BIPs) of Smith et al.

172

173

174

1 The pulse sequences are available on the WWW at http://www-keeler.ch.cam.ac.uk.


in place of the conventional 180� pulses [17]; the specificpulse that was used is designated BIP-720-25-40. We foundthat the inclusion of these BIP pulses reduced the level ofartefacts associated with imperfect refocusing to the levelwhere phase cycling of these pulses was not required. It isnecessary to use two such pulses in the sequence so thatthe phase errors introduced by the first are refocused bythe second.

A clean anti J-spectrum will only be generated if theselective 180� pulse leaves the spins states of all the passivespins unchanged. If the multiplet from one of the passive


http://www-keeler.ch.cam.ac.uk

http://www-keeler.ch.cam.ac.uk

Andrew Pell

Cross-Out

T

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233




spins lies close to the multiplet being inverted, the spin stateof that spin will be changed by the pulse. As a result, themultiplet in the anti J-spectrum will no longer be alignedat 45�, and so will not combine with the normal J-spectrumto eliminate the dispersive component of the lineshape.This effect can be minimized by further reducing the band-width of the pulse, but this will be at the cost of a furtherdecrease in the sensitivity of the experiment. A balancetherefore needs to be struck. It does not matter if the band-width of the pulse is reduced to the point where the pulseno longer inverts the whole multiplet, since in different re-gions of the sample, different lines in the multiplet will beon-resonance.

In strongly-coupled spin systems, it is found that extrapeaks appear in the J-spectrum, which complicate theirinterpretation [18,19]. These so-called strong coupling arte-facts arise because the 180� pulse causes a mixing effectwhich can loosely be described as a transfer of magnetiza-tion from the active spin to its strongly-coupled partner.Within each multiplet, these peaks due to strong couplingoccur in pairs: one is positive, and the other is negative.For example, in the spectrum of an AB system, there aretwo pairs of extra peaks, one centred on x1 ¼ 1

2C, the second

centred on x1 ¼ �12C, where C ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiðX1 � X2Þ2 þ ð2pJÞ2

q.

When the spectrum is tilted, the peaks in each pair line upon the same x2 coordinate, and so are added when the pro-jection is computed. As the peaks have opposite signs theypartly cancel in the projection, and it can be shown thatthe residual intensity in the projection is 2 sin2 2h, wheretan2h = (2pJ)/(X1 � X2). For modest strong coupling, i.e.

UN

CO

RR

EC

(a)

(g)

(m)

(p) (q)

(h)

(k) (l)

(b) (c)

(f)

*

020 -20 Hz

*

6 3

16α7β

15β 2β

1β 9

Fig. 7. Cross-sections taken parallel to x1 from the tilted absorption-mode J-given. The positions indicated by the * in (p) and (q) show disturbances due t


ED

PR

OO

F

first order in h, the cancellation is complete. This cancella-tion will only occur if a phase-sensitive spectrum is pro-jected. Thus, we can expect the intensity of the strongcoupling artefacts to be much weaker in the absorption-mode spectrum than in the conventionally processed magni-tude spectrum.

2. Results

The method is illustrated with the spectra of quinine andthe steroid dehydroisoandrosterone, the structures ofwhich are shown in Fig. 3(a) and (b). In Fig. 4 are shownthe conventional proton spectra and the projection of theabsorption-mode J-spectrum of quinine. The aromaticand vinyl region is shown in (a) and (b), while the aliphaticregion is shown in (c) and (d). The integrals of the peaks inthe projection are in good agreement with those in the con-ventional spectrum, except in the case of the overlappingmultiplets around 1.7 ppm. This retention of the naturalintensities is a particular feature of our method.

Fig. 5 shows the cross-sections taken parallel to x1 fromthe tilted two-dimensional spectrum; the multiplet struc-tures exhibit the narrow lineshapes that are expected fromJ-spectroscopy in which inhomogeneous broadening isrefocused in this dimension.

The second test is the more complex spectrum of dehy-droisoandrosterone, shown in Fig. 6(a). There are severaloverlapping multiplets in the region 1.0–2.5 ppm (expan-sion shown in (c)), making this spectrum more difficult toassign. The projection of the absorption-mode J-spectrumis shown in (b) and (d). The integrals in the projection

(n) (o)

(i) (j)

(d) (e)16β

15α

4α 4β

11β 14 12α

8, 1α, or 2α 11α, 7α, or 12β

spectrum of dehydroisoandrosterone. The assignment of each multiplet iso the t1 noise from the neighbouring methyl group.


Andrew Pell

Cross-Out

T

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259260261262263264265266

267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311




REC

are again in good agreement with those in the normal spec-trum. There is also much improved separation in thecrowded region, as is particularly evident from comparing(c) and (d).

The multiplets taken from the cross-sections are shownin Fig. 7. It is clearly seen that, even in the crowded region,it is possible to obtain well-resolved multiplets.

3. Conclusion

The presence of the phase-twist lineshape is arguably thefactor that has limited the applicability of conventional J-spectroscopy, and so over the years a number of ap-proaches to eliminating this undesirable lineshape havebeen developed. We have shown that the use of the Zang-ger–Sterk pulse sequence element in the standard pulse se-quence makes it possible to record absorption-modespectra with natural intensities. The price that must be paidis a reduction in the sensitivity. However, the resultingspectra are of such quality that the reduction in sensitivitymay well be tolerable.

Acknowledgments

A.J.P. would like to thank the University of Cambridgefor a Domestic Research Studentship. We are grateful toDr. Andrew Phillips (Astra Zeneca, Macclesfield) for hissupport and interest in this project.

References

[1] W.P. Aue, J. Karhan, R.R. Ernst, Homonuclear broad banddecoupling and two-dimensional J-resolved NMR spectroscopy, J.Chem. Phys. 64 (1976) 4226–4227.

[2] A. Bax, R. Freeman, G.A. Morris, A simple method for suppressingdispersion-mode contributions in NMR spectra: the pseudo echo, J.Magn. Reson. 43 (1981) 333–338.

[3] M. Woodley, R. Freeman, Decoupled proton NMR spectra, J. Magn.Reson. Ser. A 109 (1994) 103–112.

UN

CO

R 312


ED

PR

OO

F

[4] M. Woodley, R. Freeman, Elimination of spin–spin splittings fromhigh-resolution proton NMR spectra, J. Magn. Reson. Ser. A 111(1994) 225–228.

[5] P. Xu, X.-L. Wu, R. Freeman, Proton NMR spectra without spin–spin splittings, J. Am. Chem. Soc. 113 (1991) 3596–3597.

[6] P. Xu, X.-L. Wu, R. Freeman, Broadband-decoupled proton spec-troscopy, J. Magn. Reson. 95 (1991) 132–148.

[7] S. Simova, H. Sengstschmid, R. Freeman, Proton chemical-shiftspectra, J. Magn. Reson. 124 (1997) 104–121.

[8] V.A. Mandelshtam, H.S. Taylor, A.J. Shaka, Application of the filterdiagonalization method to one- and two-dimensional NMR spectra,J. Magn. Reson. 133 (1998) 304–312.

[9] V.A. Mandelshtam, Q.N. Van, A.J. Shaka, Obtaining protonchemical shifts and multiplets from several 1D NMR signals, J.Am. Chem. Soc. 120 (1998) 12161–12162.

[10] V.A. Mandelshtam, N.D. Taylor, H. Hu, M. Smith, A.J. Shaka,Highly resolved double absorption 2D NMR spectra from complexseverely truncated 2D phase-modulated signals by filter-diagonaliza-tion-averaging method, Chem. Phys. Lett. 305 (1999) 209–216.

[11] K. Zangger, H. Sterk, Homonuclear broadband-decoupled NMRspectra, J. Magn. Reson. 124 (1997) 486–489.

[12] A.J. Pell, R.A.E. Edden, J. Keeler, Broadband proton-decoupledproton spectra, Magn. Reson. Chem. 45 (2007) 296–316.

[13] P. Bachmann, W.P. Aue, L. Muller, R.R. Ernst, Phase separation intwo-dimensional spectroscopy, J. Magn. Reson. 28 (1977) 29–39.

[14] J. Keeler, D. Neuhaus, Comparison and evaluation of methods fortwo-dimensional NMR spectra with absorption-mode lineshapes, J.Magn. Reson. 63 (1985) 454–472.

[15] O.W. Sørensen, C. Griesinger, R.R. Ernst, Time reversal of theevolution under scalar spin-spin interactions in NMR. Applicationfor x1 decoupling in two-dimensional NOE spectroscopy, J. Am.Chem. Soc. 107 (1985) 7778–7779.

[16] M. Nilsson, G.A. Morris, Pure shift proton DOSY: diffusion-ordered1H spectra without multiplet structure, Chem. Comm. (2007) 933–935.

[17] M.A. Smith, H. Hu, A.J. Shaka, Improved broadband inversionperformance for NMR in liquids, J. Magn. Reson. 151 (2001) 269–283.

[18] G. Bodenhausen, R. Freeman, G.A. Morris, D.L. Turner, NMRspectra of some simple spin systems studied by two-dimensionalFourier transformation of spin echoes, J. Magn. Reson. 31 (1978) 75–95.

[19] A. Kumar, Two-dimensional spin-echo NMR spectroscopy: a generalmethod for calculation of spectra, J. Magn. Reson. 30 (1978) 227–249.

[20] C. Bauer, R. Freeman, T. Frenkiel, J. Keeler, A.J. Shaka, Gaussianpulses, J. Magn. Reson. 58 (1984) 442–457.


two-dimensional j-spectra with absorption-mode lineshapes

Documents