Artifact-Free Pure Absorption PFG-Enhanced DQF-COSY Spectra Including a Gradient Pulse in the Evolution Period
✍ Scribed by Bernard Ancian; Isabelle Bourgeois; Jean-François Dauphin; Anthony A. Shaw
- Publisher
- Elsevier Science
- Year
- 1997
- Tongue
- English
- Weight
- 277 KB
- Volume
- 125
- Category
- Article
- ISSN
- 1090-7807
No coin nor oath required. For personal study only.
✦ Synopsis
We have recently shown that artifacts due to rapid pulsing frequency shift in the v 2 dimension for the sine-modulated in gradient-enhanced double-quantum-filtered (GEDQF) hocomponents compared to cosine-modulated ones. Although monuclear COSY experiment can be suppressed by a suitdata reconstruction and processing are extremely simple, the able choice of gradients during evolution, mixing, and detecgradient-switching time and the recovery from eddy-current tion periods (1). The proposed combination 4:3:10 blocks effects limit the maximum digitization rate and hence the multiple-quantum coherences generated during the evolution spectral bandwidth in the acquisition dimension. Thus, sufperiod. However, by retaining only one coherence pathway fering from such severe instrumental constraints, this elegant (N or P), the application of the first gradient leads to two solution has not yet been routinely implemented on modern major drawbacks: (i) a reduction in sensitivity by a factor high-resolution spectrometers. of two and (ii) the inability to obtain pure-phase double-
In another key paper, the Keeler group (9) has described absorption spectra (2, 3).
a simple method for obtaining pure-absorption heteronuclear Whereas the loss in sensitivity is not a critical defect, the correlation spectra with the aid of carefully placed pulsed field phase-twist lineshape may be a severe handicap in highgradients in the pulse-sequence timing. The idea is to record resolution NMR because of the increased linewidth genertwo separate spectra with gradients of the same magnitude, ated by the magnitude calculation mode and the loss of but with alternately positive (negative) and negative (positive) information about the signs of the cross peaks. J. Keeler et signs. In the first spectrum, suitable gradients select the N type al.
(3) have recently shown that the key to retaining both or echo spectrum (coherence order /1 during t 1 and 01 during coherences (/1 and 01) during t 1 , and thus to obtaining t 2 ), whereas in the second spectrum, another gradient sequence time-domain data which are amplitude modulated in t 1 and selects the P type or antiecho spectrum (coherence order 01 can be phased along the v 1 dimension, is not to apply a during t 1 and 01 during t 2 ). Although each of these spectra will gradient during the evolution period. It is unfortunately clear show the undesirable phase-twist lineshape, they can always be that this modification is fundamentally incompatible with a combined during data processing to yield a pure-absorptioncomplete removal of multiple-quantum rapid-pulsing artimode spectrum (10, 11). Nevertheless, such an attractive alterfacts generated by the first 90Њ pulse which requires the native has not yet been used in homonuclear correlation even unavoidable application of a gradient immediately after the though phase errors due to the evolution of chemical shifts preparation period.
during the first gradient are certainly less severe than in hetero-Another original idea called the switched acquisition time nuclear spectroscopy. (SWAT) gradient method has been proposed by Hurd et al.
In this Note, we propose a variant of the Keeler method (4). It consists of sampling the time-domain signal at twice (9) with much simpler data manipulation and processing. the dwell rate, inserting appropriate alternating positive and The signals from P-type and N-type pathways are alternately negative gradient pulses between digitized points for encodcollected and added in the same data block, time-domain ing coherence pathways 0 r 01 r 01 (P type) and 0 r filtered in both dimensions, and Fourier transformed as a /1 r 01 (N type) in a single acquisition. While this method complex signal in the t 2 domain and as a real signal in the avoids collection of additional data blocks, in contrast to the t 1 domain according to the time-proportional phase-incretraditional RF phase cycling (5-8), it results in a Nyquist mentation (TPPI) scheme (11-13). Finally, the spectra are phased by standard procedures, and it is shown that the first few data points of the FIDs, which have been corrupted by