close

Вход

Забыли?

вход по аккаунту

?

Separation of Overlapping Multiplets in Two-Dimensional NMR Spectra by Selective УInjectionФ of Magnetization.

код для вставкиСкачать
tant was the finding that [2H,]-7 is transformed by the
gametophytes of the brown alga Gffordia rnitchellae
into 9, the C,, analogue of giffordeneLgl(giffordene =
(2Z,4Z,6E,8Z)-undeca-2,4,6,8-tetraene).In analogy to the
biosynthesis of the acyclic tetraene 11, which arises from 7 by
loss of a hydrogen atom at C(13) and cleavage of the C(9k
C( 10) bond, the formation of the intermediate (3Z,5Z$Z)decatetra-1,3,5,8-ene (8) must be postulated in order to account for the unusual geometry of 9. Compound 8 is
thermally labile and rearranges below room temperature into the isomeric tetraene 9 by an antarafacial [I ,7] hydrogen
shift.[g,lo]This electrocyclic reaction which otherwise occurs
only in the biosynthesis of vitamin D[‘ is now proven experimentally for the first time for the secondary metabolism
of marine brown algae as well.
(3S,5Z,8Z)-Undeca-l,5,8-trien-3-oI (12),[’ 2 p 31 isolated
from brown algae, was discussed by Moore and Jaenicke as
an intermediate common to the fatty acid metabolism and
the C, ,H,, pheromone^.^'^] A recently reported “biomimetic” synthesis of hormosirene (B) from 12 appears to support
this hypothesis.[’ ’1 In analogy to the biosynthesis of the
mushroom aroma oct-1-en-3-01, 12 could be derived from a
C,, polyene carboxylic acid.[16] Under standard conditions[’] all incubation experiments with a chain-shortened
(5Z,82)-[5,6,8,9-2H,]-deca-1,5,8-trien-3-ol
were negative.
The expected [2H,]-propenylcycloheptadiene [’H,]-lO was
not found. Thus, we rule out the unsaturated alk-1-en-3-01s
of type 12 as intermediates between the fatty acid
metabolism and the C,, hydrocarbons. Because the eicosanoids are not catabolized into the multiply unsaturated
C,, carboxylic acids 4 or 5, we presume that 9-hydroxy- or
9-hydroperoxyeicosapentaenoic acid is the intermediate. In
these compounds the functionalized C(9) of 3 and 6 corresponds to the carboxyl groups of 4 and 5 (Scheme I), the
precursors of the C,,H,, and C,,H,, hydrocarbons in
higher plants. Thus, the group of C,, compounds in the
brown algae could arise from the pool of eicosanoids according to the fragmentation C,, + C,, + C,, whereas in higher
plants unsaturated fatty acids are converted to the olefins
C,.
according to the scheme C,, + C l ,
+
Received: April 10, 1992 [Z5299IE]
German version: Angew. Chem. 1992, 104, 1261
CAS Registry numbers:
7, 142781-27-9; 10, 142864-29-7; 11, 142781-26-8; dictyotene, 22735-58-6;
arachidonic acid, 506-321-1.
I. Maier, D. G. Miiller, BiolBuN. 1986, 170, 145.
W. Boland, Biol. Unserer Zeir 1987, 17, 176.
D. G. Miiller, C. E. Schmid, Biol. Chem. Hoppe-Seyler 1988, 369, 647.
W. Boland, K. Mertes, Eur. J Biochem. 1985, 147, 83.
C. Neumann, W. Boland, Eur. J Biochem. 1990, lYf, 453.
Eicosapentaenoic acid (6) and arachidonic acid (3) are the main components of the phospholipids found in the membranes of female gametes of
E. siliculosus: C. E. Schmid, W. Eichenberger, D. G. Miiller, Biol. Chem.
Hoppe-Seyler 1991, 372, 540.
[7] Experimenrol Procedure: To a cell suspension of approximately 10’ gynogametes of E. siliculosus in 15 mL o f seawater was added 0.2 rng of
[2H6]-7in 2 pL of DMSO. After 30min at 18°C the volatile components
were collected over 24h on a filter of active charcoal (1.5 mg; CLSA filter,
CH-8405Winterthur (Switzerland)) by means of a circulating stream of air
(miniature circulation pump; Brey, D-W-8940 Memingen (FRG)) in a
closed system. The dead volume of the circulating air was about 15 mL.
The filter was eluted with CH,CI, (30 pL), and the metabolites were characterized by GC/MS. The method was modified from: W Boland, P. Ney,
L. Jaenicke, G. Gassmann in Analysis of Volatiles (Ed.: P. Schreier), de
Gruyter, Berlin, 1984, p. 371.
IS] D. G . Miiller. W. Boland, L. Jaenicke, G. Gassmann, 2. Naturforsch. C
1985, 40, 457.
[9] W. Boland, N. Schroer, C. Sieler, M. Feigel Helv. Chim. Act0 1987, 70,
1025; W. Boland, L. Jaenicke, D. G. Miiller, G. Gassmann, Experientia
1987, 43, 466.
[I]
[2]
[3]
[4]
[5]
[6]
I248
0 VCH
Verlagsgesellschaft mbH, W-6940 Weinheim, 1992
[lo] F. Naf, R. Decorzant, W. Tbommen, B. Willhalm, G. Ohloff, Helv. Chim.
Acta 1975, 58, 1016.
[Ill M. Akhtar, C. J. Gibbons, J Chem. Soc. (London) 1965, 5964.
[I21 K. Yamada, H. Tan, H. Tatematsu, M. Ojika, Tetrahedron 1986,42, 3775.
[I31 W. Boland, D. G. Miiller, Tetrahedron. Lett. 1987, 307.
[I41 L. Jaenicke, D. G. Miiller, R. E. Moore, J. Am. Chem. Soc. 1974,96,3324;
R. E. Moore, Arc. Chem. Res. 1977,10,40.
1151 W. D. Abraham, T. Cohen, J Am. Chem. Soc. 1991, 113,2313.
[I61 M. Wurzenberger, W. Grosch, Biochem. Biophys. Acta 1984, 795, 163.
Separation of Overlapping Multiplets in
Two-Dimensional NMR Spectra by
Selective “Injection” of Magnetization**
By Catherine Zwahlen, Skbastien J. E: Vincent,
and Geoffrey Bodenhausen*
Broadly speaking, multidimensional NMR spectra can be
interpreted at two different levels of sophistication. In many
cases, it is sufficient to focus on chemical shifts, but a more
thorough analysis must also address the information contained in the fine-structure of cross-peak multiplets.[’] There
are three major obstacles that may impair such a detailed
analysis: excessive complexity, poor digital resolution, and
accidental overlap. Very complicated structures can be simplified either by some means of decoupling during signal
acquisition, or by partial deconvolution of the multiplets
once they have been recorded.[’] The problem of digital resolution can be solved very effectively by “zooming in” on
multiplets with the help of “soft” two-dimensional methods
such as selective correlation spectroscopy (s0ft-C0SY).[~]
However, these methods are unsuitable when the experimentalist is confronted with overlapping multiplets. Serious cases of overlap call for some way of increasing the dimensionality of the spectra. While three-dimensional spectra
recorded with non-selective pulses have the advantage that
multiplets rarely suffer from overlap, their digital resolution
is usually very poor. Thus, it is desirable to record three-dimensional spectra with selective pulses,[41but three-dimensional experiments are relatively cumbersome and time-consuming, and it is difficult to present the outcome in a
graphically pleasing way.
Fortunately, it is possible to achieve a similar separating
effect by a “poor man’s’’ form of three-dimensional spectroscopy, which amounts to recording signals in a plane
S(w,,w,) of a hypothetical S(w,,w,,w,) data cube, where
one of the three frequency variables (i.e. w , ) is fixed. This can
be achieved by a variant of relayed magnetization transfer,
where coherence is “injected” in a selective manner at the
beginning of the evolution period of a soft-COSY sequence.
Consider a soft-COSY experiment that is set up to correlate
multiplets resonating in the vicinity of 52, in w1and 52, in w,.
If there are several spins that resonate at nearly or precisely
degenerate frequencies a,, 52,, ,... and a,, 52,. ,..., the softCOSY spectrum will consist of a superposition of crosspeaks due to coherence transfer processes A + X, A + X , ...
Now suppose that each of the nearly degenerate nuclei A,
A’, ..., has a coupling partner M, M‘, ..., and suppose that
these are not degenerate amongst each other. In such a case,
[*I Prof. Dr. G. Bodenhausen, C. Zwahlen, S. J. F. Vincent
Section de Chimie, Universitk de Lausanne
Rue de la Barre 2, CH-1005 Lausanne (Switzerland)
[**I We thank Dr. W Bermel, Karlsruhe, and Prof. H. Kessler, Garching, for
a sample of cyclo-(~-Pro’-~-Pro~-~-Pro’).
This work was supported by the
Swiss National Science Foundation and by the Commission pour
I’Encouragement de la Recherche Scientifique (CERS).
0570-0833192j0909-1248$3.50+ .25/0
Angew. Chem. Int. Ed. Engl. 1992, 31, No. 9
one may carry out experiments where the soft-COSY sequence is preceded by a selective coherence transfer step
M +A, or M' +A,etc. In this manner, the soft-COSY multiplets of the processes A + X, A + X , etc., can be unraveled.151
how a doubly selective
We have recently
homonuclear Hartmann-Hahn (HOHAHA) transfer can be
used in one-dimensional spectra. Figure 1 a shows the pulse
sequence for one-dimensional applications. In-phase magnetization I," is first excited by a self-refocusing 270" Gaussian
pulse truncated at 2.5 % with a typical duration of 30 ms for
a peak amplitude of 55 Hz.['I More sophisticated pulses such
as Gaussian cascades optimized by quaternions['I or EBURP pulses['0] may also be used. In Figure l , the doubly
selective irradiation of duration rDSIis implemented by modulating a rectangular radiofrequency (rf) pulse with a function cos(w,t), where w, = % (0,- 0,). This modulation
U
QA
V
QA
L
Qx
C
41 Qn
+
QA
QA
5
14
20
55
41
Vl
[Hzl
-
65
73
Fig. 2. Optimization of the rf amplitude v, [Hz] for a HOHAHA transfer
between the P p r o t o n (6 = 2.10) and the B"""" proton (6 =1.62) of L-proline'
'"
-$
12.8 "
Hz,"
T)
~is ~78 ms.
,
in 1 in C,D,/CDC13 (8/1) at 303 K. Since .I($is
The experiments were performed on a Bruker MSL 300 spectrometer equipped
with an Oxford Research Systems selective excitation unit.
Typically, the optimum radiofrequency amplitude v1 is
30 Hz for each sideband. If the amplitude is much weaker, it
is impossible to lock all magnetization components within
each multiplet simultaneously; if the rf amplitude is too
large, various passive spins, some of which may have chemical shifts that lie accidentally in the vicinity of the rf sidebands, will also be perturbed.
Figure 1 b shows the pulse sequence for normal selective
correlation spectroscopy (soft-COSY). The in-phase magnetization I t partly evolves into an anti-phase term 2ItZz during t , . [ ' Z 1The second and third pulses convert this term first
into 2Z:I:, then into 2ItZ:, which evolves into observable
magnetization Z
.: The method can easily be adapted for
soft-NOESY, by inserting a mixing period 5, between the
last two pulses.['. 131 Figure 1 c shows the sequence obtained
by combining Figures 1 a and 1 b; in effect, the initial pulse
of the soft-COSY experiment is applied at 0, rather than at
a, and it is immediately followed by a doubly selective
irradiation period z ~ for
~ HOHAHA
,
transfer from M to A.
The only information needed prior to setting up the experi-
b
J
selective irradiation must be optimized empirically to give
the best in-phase transfer, as shown in Figure 2 for a transfer
within the coupling network of L-proline' of cyclo(L-Pro'+
pro2-~-pro3)I.["]
Qx
Fig. 1. a) Pulse sequence for selective in-phase coherence transfer in onedimensional spectroscopy. The initial 270" Gaussian pulse, with a carrier frequency set at the chemical shift 62,. excites in-phase magnetization of spin M,
which is transferred to spin A during the subsequent doubly selective irradiation period TDSI = l/JhM,with two rfsidehands at QM and 62,. The phase of the
spin-locking field is orthogonal to the phase of the initial excitation pulse.
b) Normal selective correlation experiment (soft-COSY) using three self-refocusing 270" Gaussian pulses, the first to excite A magnetization, the latter two
for transferring magnetization from spin A to spin X. c) SPLIT-COSY sequence, consisting of a soft-COSY preceded by a doubly selective HOHAHA
to inject in-phase coherence from spin M to spin A at the beginning of the
evolution period. To suppress unwanted signals, the phases of the first pulse and
of the receiver are alternated between 0 and 180" in all three sequences.
generates two sidebands at frequencies wrf w,, which coincide with the chemical shifts QM and 0, of the two chosen
spins, provided the carrier frequency is placed at
w,, = % (0, + 0,). The optimum duration of the doubly
selective irradiation is given by the inverse of the active coupling constant, rDSI= l/JAM.
The amplitude of the doubly
Angew. Chem. In(. Ed. Engl. 1992, 31, No. 9
0 VCH
a
r,
Qu'
Fig. 3. a) Soft-COSY experiment correlating the B"" proton (at 6 = 1.23 in the
vertical w , domain) and the $'MSproton (6 = 2.79 in the horizontal (uzdomain)
of p pro line) in 1 obtained by exciting the magnetization of the $'" proton by
a simple 270"Gaussian pulse at the beginning of the evolution period. h) Same
multiplet, hut obtained by injection of magnetization from the OL proton
(6 = 4.49) to the p" proton at the beginning of the evolution period, as indicated on the graph representing the coupling network. Since in p pro line' J ( a p )
is 6.5 Hz, tm,is 154 ms. The spectral widths were 75 Hz in both w 1 and w2.The
matrices consisted of 128 x 1 K data points before and 256 x 1 K data points
after zero filling. A Lorentz-Gauss transformation (LB = - 0.3, GB = 0.05)
was applied before Fourier transformation. With two scans for each I , increment and a relaxation delay of 10 s, the overall duration required for spectrum a was about 45 min; 16 scans were used for spectrum h.
Verlagsgesellschafl mbH, W-6940 Weinhelm,1992
0870-0833/92/0909-1249$ 3 . 8 0 + .28/0
1249
ment is the chemical shift 0, of a "passive" M spin which
must be coupled to one of the A spins that we wish to distinguish. It is also desirable (though not essential) that we know
the approximate value of the coupling constant JAM
to be
able to estimate the optimum duration z ~ zl/JAM
~ ,
of the
doubly selective irradiation period.
Before tackling difficult cases with overlapping multiplets,
it appears advisable to check the efficiency of HOHAHA
injection by an application to a harmless case within the
coupling network of proli line^ of CyCfO(L-PrO'-L-Pr02-DPro3) l.[''I The agreement of the patterns and pure absorption peak shapes in the multiplets shown in Figure 3 provides
conclusive evidence that excitation by HOHAHA transfer, if
done properly, is fully equivalent to excitation with a selective pulse. Figure 3 also shows graphs of coupling networks
where the nodes represent the spins and the edges their mutual couplings.['41 To represent a soft-COSY experiment
symbolically, a bold arrow runs from the spin that is active
in t , to the one that carries the coherence in t,, while HOHAHA transfer is represented by a wavy arrow.
Now consider a case where the multiplets appear to be
hopelessly entangled. In earlier studies, the 1:3 mixture of
the ethyl esters 2 of cis- and trans-2-phenylcyclopropanoic
acid (R = COOCH,CH,, R = C,H,) was used to test the
limits of pattern recognition['41 and multiplet separation" 51
procedures.
KC
M
MI
'K
X'
KC
Xc
c
MI
K'
HwH:
Xc
X'
2
R'
The protons A"'" and A''" on the one hand, and X'"" and
Xcis on the other, have pairwise nearly the same chemical
shifts. This situation is very typical for a mixture of isomers.
In a conventional soft-COSY experiment, one obtains two
strongly overlapping multiplets due to the two transfer proceSSeSx f r m s +Afrans and x c i s +Acis (Fig. 4a). Fortunately,
the existence of nondegenerate coupling partners M'""" and
Meismakes it possible to transfer magnetization selectively to
either X'""" or XciS.A soft-COSY sequence following selective transfer may provide pure cross-peaks of either
X"""" +AfranSor Xcis-+A"", as illustrated in Figures 4c and
4 b, respectively. We dubbed this novel experiment Separation of Patterns by Lifting the Imbroglio of Transfer in Correlation Spectroscopy (SPLIT-COSY).
Figure 5 illustrates an interesting case where our technique
failed to split two overlapping cross-peaks due to an unforeseen case of degeneracy. In the normal soft-COSY spectrum
corresponding to the region shown in Figure 5, one observes
two severely overlapping multiplets due to y""""
/I""""
and
pt-,, in L-proline'. We hoped to split these by using a
selective HOHAHA transfer to y"""" from two other strongly
coupled protons of the same amino acid, Beis and atran".To
our disappointment, the SPLIT-COSY spectrum obtained in
this manner, which is shown in Figure 5, is virtually identical
to the corresponding soft-COSY spectrum. This experiment
shows that the magnetization received by y'""" is in part
transferred to p'", although the J coupling constants between
and both 6""and 6""""are vanishingly small. In
general, when two coupled spins have nearly the same chemical shift, it is not possible to transfer magnetization to one
of them selectively. Indeed, two such spins will always be
irradiated at the same time by one of the rf sidebands, so that
Fig. 4. a) Normal soft-COSY spectrum of 2 in CDCI, showing overlapping
cross-peaks Xcis+Acisand X""""+A""". The symbols A' and A' stand for the
HA protons of the cis and zruns isomers of 2. b) Clean X'" -A"* multiplet
(centered at 6 =1.16 and 2.42, respectively) obtained by SPLIT-COSY with
+A"""'
injection from M ' " (6 = 1.95) to X"' with rDS,= 130 ms. c) Clean XfrEnS
multiplet (centered a t 6 = 1.18 and 2.40) obtained by an injection from M'""'
(6 = 1.79) to X'""", with ' T =
~108
, ms. Experimental conditions were the same
as for the spectra in Figure 3.
if any magnetization is transferred to one of them, it will
automatically be transferred to the other, in the manner of
total correlation spectroscopy (TOCSY).[161
6'
vc
8'
a
-+
--f
1250
0 VCH
Verlugsgeseilschuft mbH. W-6940 Weinheim. f992
Fig. 5. SPLIT-COSY spectrum of 1 centered at 6 =1.59 in wIand 2.27 in q,
showing a superposition of two cross-peaks
+ 8""" and y'""'
+ br'""sin
L-proline'. In this case, a doubly selective HOHAHA period led to a transfer
from 6"' und6"""" (which are nearly degenerate at 6 = 3.55) to yfnnS and$'", the
latter being coupled and nearly degenerate in the vicinity of 6 = 1.59. Experimental conditions were the same as for the spectra in Figure 3.
Although the two-step transfer from dciS+ 6'""" to pi"
represents an obstacle to the separation of the two overlapping multiplets in Figure 5, the same phenomenon can be put
to use in a constructive fashion to separate the multiplets in
Figure 6a. These two multiplets originate from two different
0S70-0833/92j0909-1250$3.50+.25/0
Angew. Chem. Int. Ed. Engl. 1992, 31, No. 9
proline residues. The L-proline' multiplet can be separated
easily by a transfer from pcis to /3"""s as shown in Figure 6 b.
In order to excite only the pisproton of L-proline', magnetization was transferred from the strongly coupled pair
p + 6"""" to the strongly coupled pair pcis + fra's. The resulting multiplet is represented in Figure 6 c and corresponds
perfectly to the pattern in Figure 6 a obtained by soft-COSY.
CAS Registry numbers:
1, 70493-40-2; cis-2, 946-39-4; trans-2, 946-38-3
[l] R. R. Ernst, G. Bodenhausen, A. Wokaun, Principles ofNucleur Magnetic
Resonance in One and Two Dimensions, Clarendon Press, Oxford, 1987.
[2] J.-M. Le Parco, L. McIntyre, R. Freeman, J Mugn. Reson, 1992. 97, 553;
P. Huber, G. Bodenhausen, ibid., in press.
[3] R. Briischweiler,J. C. Madsen, C. Griesinger, 0. W. Ssrensen, R. R. Emst.
J. Magn. Reson. 1987, 73, 380; J. Cavanagh, J. P. Waltho, J. Keeler. ibid.
1987, 74, 386.
[4] J. Friedrich, S. Davies, R. Freeman, Mol. Phys. 1988, 64. 691; C.
Griesinger, 0. W Snrensen, R. R. Ernst, J. Magn. Reson. 1989, 84, 14.
[5] N. Miiller, L. Di Bari, G. Bodenhausen, J. Mugn. Reson. 1991, 94. 73.
[6] R. Konrat, I. Burghardt, G. Bodenhausen, J. Am. Chem. Soc. 1991, 113,
9135.
[7] B. Boulat, R. Konrat, I. Burghardt. G. Bodenhausen, J Am. Chem. Sac.,
1992, 114, 5412.
[8] L. Emsley, G. Bodenhausen, J Magn. Reson. 1989, 82, 21 1.
[9] L. Emsley, G. Bodenhausen, J Mugn. Reson., 1992, 97, 135.
[lo] H. Geen, S. Wimperis, R. Freeman, J: Magn. Reson. 1989, 85. 620; H.
Geen, R. Freeman, ibid. 1991, 93.93.
[Ill H. Kessler, W Bermel, A. Friedrich, G. Krack, W E. Hull, J Am. Chem.
Sor. 1982, 104. 6297.
[12] 0. W. Ssrensen, G. W. Eich, M. H. Levitt, G. Bodenhausen, R. R. Ernst,
Prog. Nucl. Magn. Reson. Spectrosc. 1983, 16, 163; U. Eggenberger, G.
Bodenhausen, Angew. Chem. 1990,102,392; Angew. Chem. inr. Ed. Engi.
1990,29, 374.
1131 H. Oschkinat, G. M. Clore, A. M. Gronenborn, J Magn. Reson. 1988, 78,
371; H. Oschkinat, W. Bermel, ibid. 1989, 81, 220.
[I41 M. NoviE, G. Bodenhausen, Anal. Chem. 1988, 60, 582; P. Pfandler, G.
Bodenhausen, J. Magn. Reson. 1988, 79, 99.
[15] L. Emsley, P. Huber, G. Bodenhausen, Angew. Chem. 1990, 102, 576;
Angew. Chem. fnr. Ed. Engl. 1990,29,517; L. Emsley, G. Bodenhausen, J.
A m . Chem. Soc. 1991, 113, 3309.
[I61 L. Braunschweiler, R. R. Emst, J. Mugn. Reson. 1983.53.521
Se2NCI, and [Se,NCI,]+[GaCl,]-,
Chloride Nitrides of Trivalent Selenium**
By RenC Wollert, Antje Hollwarth, Gernot Frenking,
Dieter Fenske, Helmut Goesrnann, and Kurt Dehnicke*
I
Qa7
I
Fig. 6. a) Multiplets obtained with normal soft-COSY of 1, centered at
6 = 1.60 in o,and b = 4.28 in w z . To the left, cross-peak correlating the p"""'
proton (at 6 = 1.62) and the a proton (6 = 4.32) of L-proline*; to the right,
cross-peak connecting the P" proton (6 = 1.58) and the a proton (6 = 4.23) of
L-proline'. b) SPLIT-COSY multiplet of L-proline' alone obtained by injection
(6 = 2.10) to B"""" via J(p"B""")= - 12.8 Hz with
of magnetization from P's
T ~ =78
, ms. c) SPLIT-COSY multiplet of L-proline' alone obtained by injection of magnetization from the two nearly degenerate protons 6"' (6 = 3.52)
= - 11.7 Hz, to the two nearly degenerate
and bL"""(6 = 3.57), with J(bCis6"""")
(6 =1.61 and 1.58, with
and strongly coupled protons Y'""~ and
J(y''""'p"") = 3.4 Hz) via J(6c'y"'ns) = 3.1 Hz and J(6tr~"sy"""s)= 6.7 Hz. The
cleanest in-phase magnetization was obtained for T ~ =~220, ms. In this case,
the double irradiation period was followed by a hard purging pulse in order to
remove antiphase terms [6]. Experimental conditions were the same as those for
the spectra in Figure 3.
The chemistry of selenium-nitrogen compounds has only
livened up in the last few years, mainly because of the development of new synthetic processes."' Nevertheless, the research in this field is often limited by the properties of the
starting materials: the explosive character of tetraselenium
tetranitride or by the low solubility of the more stable
tetraselenium dinitride.['] We report herein about a readily
accessible chloride nitride of trivalent selenium, which is soluble in organic solvents and stable, but at the same time is
very reactive and suitable as a synthetic reagent.
Diseleniumtrichloride nitride (1) is formed by the action of
tris(trimethylsilyl)amine on a suspension of selenium tetrachloride in boiling dichloromethane [Eq. (a)]. Upon cooling
2 SeC1,
Our new SPLIT-COSY technique allows one to separate
overlapping multiplets in crowded parts of COSY spectra in
order to extract hidden information. The structure of the
resulting multiplets is the same as in soft-COSY, unlike the
multiplets obtained in Selectively Znverted Soft COSY (SISCOSY).r'51 In general, samples containing mixtures of isomers, which are likely to give overlapping cross-peaks,
should become amenable to investigation, provided each isomer has at least one site with a distinct chemical shift. If
necessary, several consecutive HOHAHA transfer steps may
be used.
Received: April 7, 1992 [ Z 5285 IE]
German version: Angew. Chem. 1992,104, 1233
Angen>.Chem. Inr. Ed. Engl. 1992, 31, No. 9
0 VCH Verlagsgesellschaft mbH.
+ N(SiMe,),
------t
Se,NCl,
+ 3C1SiMe3 + C1,
(a)
1
moisture-sensitive crystals can be isolated from the red solution. 1, which when viewed against light appear red but
viewed against a background appear a shiny, metallic green.
[*I Prof. Dr. K. Dehnicke, Dr. R. Wollert, DipLChem. A. Hollwarth,
Prof. Dr. G. Frenking
Fachbereich Chemie der Universitat
Hans-Meerwein-Strasse, D-W-3550 Marburg (FRG)
Prof. Dr. D. Fenske, Dr. H. Goesmann
Institut fur Anorganische Chemie der Universitat Karlsruhe
[**I This work was supported by the Deutsche Forschungsgemeinschaft and
the Fonds der Chemischen Industrie.
W-6940 Weinheim, 1992
0570-0833/92/0909-1251$3.50+.25/0
1251
Документ
Категория
Без категории
Просмотров
3
Размер файла
469 Кб
Теги
two, dimensions, separating, nmr, selective, overlapping, multiple, magnetization, spectral, уinjectionф
1/--страниц
Пожаловаться на содержимое документа