NMR spectroscopy in environmental chemistry: 1H and 13C NMR parameters of tricyclic polychlorinated C10 hydrocarbons and their oxy derivatives based on two-dimensional NMR techniques

Article abstract of DOI:10.1002/(SICI)1097-458X(199905)37:5<359::AID-MRC460>3.0.CO;2-3

Two-dimensional homo-and heteronuclear NMR chemical shift correlation techniques were applied in the characterization of five tricyclic polychlorinated C₁₀ hydrocarbons, chlordene (1), heptachlor (2), rrans-nonachlor (3), α-chlordene (4) and γ-chlordene (5), which are spread globally in the environment owing to their use as insecticides. Approximate and partly contradictory ¹H and ¹³C NMR chemical shifts reported in the literature were corrected in this work. The chemical shift assignments of 1-5 were based on DQF COSY, HMQC and HMBC experiments. In addition, an INADEQUATE experiment was needed to ascertain the ¹³C chemical shifts assignment of 2. The ⁿ J(H,H)s of 1-5 were solved by computer-assisted spectral analyses. Further, complete ¹H and ¹³C NMR parameter sets of three oxy derivatives, heptachlor exo-epoxide (6), 1-exo-hydroxychlordene (7) and its acetate (8), were also determined. Compounds 7 and 8 were synthesized to be used as model compounds, and their NMR parameters are reported for the first time. By using solvent susceptibility matched symmetrical micro-NMR tubes and HMQC and HMBC experiments, 1-5 can be reliably characterized at submilligram levels by their ¹H and ¹³C NMR parameters with a 500 MHz (11.8T) spectrometer and a 5 mm diameter standard probehead. Copyright

Full text of DOI:10.1002/(SICI)1097-458X(199905)37:5<359::AID-MRC460>3.0.CO;2-3

MAGNETIC RESONANCE IN CHEMISTRY
Magn. Reson. Chem. 37, 359–364 (1999)
1
13
NMR spectroscopy in environmental chemistry: H and C
NMR parameters of tricyclic polychlorinated C10
hydrocarbons and their oxy derivatives based on
two-dimensional NMR techniques
Erkki Kolehmainen,* Jari Koivisto, Katri Laihia, Reijo Kauppinen and Jaakko Paasivirta
Department of Chemistry, University of Jyv a¨ skyl a¨ , P.O.B. 35, FIN-40351 Jyv a¨ skyl a¨ , Finland
Received 23 September 1998; revised 3 December 1998; accepted 7 December 1998
ABSTRACT: Two-dimensional homo- and heteronuclear NMR chemical shift correlation techniques were applied in
the characterization of five tricyclic polychlorinated C10 hydrocarbons, chlordene (1), heptachlor (2), trans-nonachlor
(
3), ˛-chlordene (4) and ꢀ-chlordene (5), which are spread globally in the environment owing to their use as insecticides.
1
13
Approximate and partly contradictory H and C NMR chemical shifts reported in the literature were corrected in
this work. The chemical shift assignments of 1–5 were based on DQF COSY, HMQC and HMBC experiments.
1
3
In addition, an INADEQUATE experiment was needed to ascertain the C chemical shifts assignment of 2. The
n
1
13
J(H,H)s of 1–5 were solved by computer-assisted spectral analyses. Further, complete H and C NMR parameter
sets of three oxy derivatives, heptachlor exo-epoxide (6), 1-exo-hydroxychlordene (7) and its acetate (8), were also
determined. Compounds 7 and 8 were synthesized to be used as model compounds, and their NMR parameters are
reported for the first time. By using solvent susceptibility matched symmetrical micro-NMR tubes and HMQC and
1
13
HMBC experiments, 1–5 can be reliably characterized at submilligram levels by their H and C NMR parameters
with a 500 MHz (11.8 T) spectrometer and a 5 mm diameter standard probehead. Copyright  1999 John Wiley &
Sons, Ltd.
KEYWORDS: NMR; H NMR; 13C NMR; two-dimensional NMR; tricyclic polychlorinated C10 hydrocarbons; oxy
1
derivatives; computer-assisted spectral analysis; solvent susceptibility matched micro-NMR tubes
INTRODUCTION
heteronuclear chemical shift correlation techniques,
1
3,14
15
1
13
HMQC
and HMBC, to ascertain the H and
C
Tricyclic polychlorinated C10 hydrocarbons are globally
widespread environmental toxicants owing to their use as
NMR chemical shift assignments of 1–5 and their oxygen-
containing derivatives. J(H,H)s of these rigid tricyclic
3
1
,2
insecticides and their chemical persistence. The iden-
tification of the components of chlordane (formerly one
of the most widely used pesticides) and their metabo-
lites and/or environmental degradation products is most
often based on gas chromatographic–mass spectrometric
structures can be further utilized in refining the struc-
tural vs NMR spectral relationships. Solvent susceptibility
matched micro-NMR tubes were tested for some of these
compounds especially to determine their detection limits
and to clarify the effect of resonance lineshape on their
computer-assisted spectral analysis.
3
techniques. However, for their configuration and isomer-
1
specific structure elucidation, H NMR parameters are of
predominant importance, as demonstrated in the case of
endo- and exo-epoxides of heptachlor (2). Further, the
4
EXPERIMENTAL
Compounds
biological transformation of isomers, diasteremers and
even enantiomers of toxicants can be (stereo)selective,
and their uptake, metabolism and excretion can be very
3
,4
1
13
Chlordene (1) (4,5,6,7,8,8-hexachloro-3a,4,7,7a-tetrahy-
dro-4,7-methano-1H-indene), heptachlor (2) (1-exo,4,5,6,
different. In the literature there are some H and
C
NMR data on polychlorinated tricyclic C10 hydrocar-
bons with more or less incomplete spectral analyses
7
,8,8-heptachloro-3a,4,7,7a-tetrahydro-4,7-methano-1H-
indene), trans-nonachlor (3) (1-exo,2-endo,3-exo,4,5,6,7,
,8-nonachloro-3a,4,7,7a-tetrahydro-4,7-methanoindane),
-chlordene (4) (1,2,3,5,7,8-hexachloro-1,3a,4,5,6,6a-
hexahydro-1,4-ethanopentalene), ꢀ-chlordene (5) (2,3,3a,
,5,8-hexachloro-3a,6,7,7a-tetrahydro-1,6-methano-1H-
4
–8
and/or contradictory chemical shift assignments.
The
8
˛
aim of this study was to apply two-dimensional homonu-
9
10,11
clear NMR techniques, DQF COSY , NOESY
and
1
2
INADEQUATE, and proton detected, two-dimensional
4
indene) and heptachlor epoxide (6) (1-exo-4,5,6,7,8,8-
heptachloro-2,3-exo-epoxy-3a,4,7,7a-tetrahydro-4,7-me-
thano-1H-indene) were commercial products: 1 (100%)
*
Correspondence to: E. Kolehmainen, Department of Chemistry, Uni-
versity of Jyv a¨ skyl a¨ , P.O.B. 35, FIN-40351 Jyv a¨ skyl a¨ , Finland.
E-mail: ekolehma@cc.jyu.fi
Copyright  1999 John Wiley & Sons, Ltd.
CCC 0749–1581/99/050359–06 $17.50

360
E. KOLEHMAINEN ET AL.
from EPA, Research Triangle Park, NC, USA, 2 (99.5%)
from Velsicol Chemical Manufacture Laboratory (Rose-
mont, IL) and 3 (98.3%), 4 (97.1%), 5 (100%) and
RESULTS AND DISCUSSION
Scheme 1 shows the structures of the compounds studied
and their numbering. Tables 1–4 give the H and
NMR chemical shifts and spin–spin coupling constants
for 1–8. Figures 1 and 2 show the HMQC and INAD-
EQUATE countour maps of heptachlor. In Fig. 3 are
plotted an experimental (measured in a Shigemi micro-
6
(100%) from EPA. Two derivatives which also can
1
13
C
serve as model compounds, 7 (1-exo-hydroxy-4,5,6,7,8,8-
hexachloro-3a,4,7,7a-tetrahydro-4,7-methano-1H-indene)
and its acetate 8, were synthesized by us. Compound 7
was formed from 1 by SeO oxidation in aqueous 1,4-
2
1
8
dioxane and 8 was formed as a by-product (2% yield)
of the SeO oxidation of 1 in aqueous acetic acid. Com-
NMR tube) (top) and a calculated (WIN-DAISY) (bottom)
1
7
2
1
5
00 MHz H NMR spectrum of heptachlor.
pound 7 was purified by column chromatography (silica
gel–chloroform) and 8 by crystallization from n-pentane.
The structure of the novel compound 8 was verified by
As mentioned before, a variety of NMR chemical shift
assignments have been published for these compounds,
partly contradicting those obtained in this work. In order
4–8
1
hyberresolution mass and NMR spectrometry. The H and
to obtain a reliable basis for our spectral assignments, a
1
3
C NMR spectral parameters of 7 and 8 were determined
12b
2
-D INADEQUATE (COSY-like ) correlation map was
for the first time in this work.
recorded for 2. This experiment reveals that (i) carbons
0.40 and 80.15 ppm are adjacent and that (ii) those with
6
chemical shifts of 137.51, 130.95, 60.68 and 81.92 ppm
are part of a linear system. An HMBC connectivity
between H-3a and C-5 gives the final verification for the
correct assignment for C-5 and C-6 of 2. This assignment
Spectroscopy
All NMR experiments were performed with a Bruker
Avance DRX 500 NMR spectrometer equipped with
a 5 mm diameter broad band probehead working at
5
is also in agreement with that given by ApSimon et al.
1
13
1
13
On the other hand, the H and C NMR chemical shift
500.13 MHz in H and 125.77 MHz in C. The chemi-
6
assignments for 3, 4 and 5 reported previously have
cal shift assignments are based on DQF COSY, HMQC
and HMBC measurements. The H NMR parameters
1
several deviations in comparison with our present results.
n
Further, as far as we know, the J(H,H) data for 3 have
were solved precisely by WIN-NMR and WIN-DAISY
1
8
13
not been reported previously.
software. For the C NMR chemical shift assignment
of heptachlor (2) a 2-D INADEQUATE measurement was
also necessary. Owing to the insensitivity of this experi-
ment (based on chemical shift correlations transmitted by
the direct spin–spin couplings between adjacent carbon-
Among the compounds studied, 1, 2, 7 and 8 have
the same carbon skeleton, thus forming a basis for the
comparison of substituent effects. Introduction of an elec-
0
tronegative substituent at position 1 in 1 causes a clear
1
3 nuclei with a natural abundance of 1.1%), the concen-
downfield (deshielding) shift or ˛-effect on the proton
H-1, being 2.46 ppm in 2 and 2.39 ppm in 7. When the
hydroxy substituent of 7 is substituted by an acetoxy moi-
ety as in 8, a 3.20 ppm downfield shift from the value
in 1 is observed. The corresponding ˇ-effects induced at
protons H-2 and H-7a in 2, 7 and 8 are much smaller,
0.22 ppm being the highest (at H-7a in 7).
tration of heptachlor had to be as high as possible in this
experiment: ca 300 mg of heptachlor dissolved in CDCl3,
the total sample volume being 0.6 ml. In NOESY experi-
ments, a 300 ms mixing time was used. Detailed lists of
the acquisition and processing parameters used in all of
these measurements are available on request.
The sensitivity tests were carried out using 5 mm
diameter and 15 mm bottom length symmetrical NMR
microtubes (Shigemi) which are magnetic susceptibility
matched for CDCl . CDCl was obtained from Eurisotop
1
The same trends as observed in the H NMR chemical
1
3
shifts, joint with substitution, are also apparent in the
C
NMR data (Table 2). Of course, the substituent chemical
shifts (SCS) are much larger than in the case of protons
3
3
1
3
(
water content <0.01%, 99.8% D).
because the C NMR chemical shifts generally span a
Scheme 1
Copyright  1999 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 37, 359–364 (1999)

2D NMR OF TRICYCLIC POLYCHLORINATED C10 HYDROCARBONS
361
Table 1. 1H NMR chemical shifts (ppm from CHCl
, υ D 7.26 ppm) of 1–8
3
1
υ( H) (ppm)
Proton
1
1
2
3
1
2
3
4
5
6
7
8
2.361
2.499
5.831
5.633
3.860
—
—
—
—
—
—
—
3.459
—
—
4.821
—
5.950
5.912
4.073
—
—
—
—
—
—
—
3.643
—
—
3.596
—
4.048
3.596
3.544
—
—
—
—
—
—
—
3.544
—
—
—
—
—
3.290
—
—
—
—
—
—
3.139
—
—
2.255
2.388c
3.724
4.057
—
4.279
—
3.703
3.681
3.672
—
—
—
—
—
—
—
3.298
—
—
4.753
—
5.955
5.919
3.986
—
—
—
—
—
—
—
3.241
—
—
5.563
—
5.947
5.984
3.985
—
—
—
—
—
—
—
3.282
—
2.036
—
0
a
—
3
4
5
6
6
6
7
7
7
8
1
a
3.808
3.110
4.732
2.577
2.182b
3.388
—
—
—
—
—
0
a
0
a
0
OH
—
—
—
—
—
—
1.861
a
0
0
0
H-1 is exo.
b
c
H-6 is syn to Cl-5.
H-7 is syn to Cl-8.
Table 2. 13C NMR chemical shifts (ppm from CDCl
, υ D 77.0 ppm) of 1–8
3
1
3
υ( C) (ppm)
Carbon
1
2
3
4
5
6
7
8
1
2
3
32.28
135.96
126.22
61.07
82.40
131.74
129.05
—
81.81
49.16
103.61
—
60.00
137.51
130.95
60.68
81.92
132.07
128.64
—
80.15
60.40
103.55
—
59.06
69.64
59.06
58.40
79.90
132.45
132.45
—
79.90
58.40
103.92
—
78.34
127.65
138.04
53.59
55.73
63.43
33.85
59.79
134.03
—
61.32
136.05
133.95
75.13
133.55
134.77
56.43
—
29.96
57.68
56.61
—
55.82
62.59
59.09
56.14
78.94
131.41
130.44
—
80.51
61.24
103.21
—
75.24
138.36
131.45
60.54
81.87
131.45
128.98
—
80.11
59.69
103.98
—
76.82
135.33
133.22
60.50
81.69
131.49
129.10
—
79.77
56.52
103.90
169.80
20.94
3
4
5
6
6
7
7
8
9
1
a
a
a
126.67
—
—
0
—
—
—
—
—
—
1
range more than 200 ppm whereas in the case of H NMR
the spread is generally less than 15 ppm.
which contains a double bond in its coupling route. The
vicinal coupling constant J(H-1,H-7a) also exhibits great
3
The vinyl protons at C-2 in 1 and 2, showing the two
most deshielded chemical shift values among compounds
changes upon substitution at C-1.
Regarding the computer-based spectral analysis, the
sign of the spin–spin coupling constant is also an
important parameter which always has to be taken into
account. A rule of thumb is the ‘alternate change of the
sign’ depending on the number of bonds included in the
1
–5, can be useful in determining these congeners in
their mixtures. Similarly, protonated double bond carbons
and 3 in 1 and 2 can be easily separated by HMQC
2
because all the other olefinic carbons bear substituents in
compounds 1–5.
2
3
4
coupling route: J < 0, J > 0, J < 0, etc. The best
result (the smallest difference between the experimental
and calculated spectra) in the iterative total line fitting
is obtained with a correct sign combination, as shown
As in the case of chemical shifts, compounds 1, 2, 7
and 8 form a proper subgroup (owing to their structural
similarity) to be used in estimating the substituent
effects on the spin–spin coupling constants. As a general
observation, it can be mentioned that introducing an
electronegative substituent, chlorine or hydroxy, into
1
9
in a previous paper. Among scalar couplings, the most
frequently used parameter in the structural analysis is the
3
vicinal coupling, J, exhibiting well known dihedral angle
0
20,21
position 1 in chlordene (1) causes decreases in the
dependences.
However, substituent electronegativity
absolute values of all spin–spin coupling constants.
The most significant change takes place in J(H-1,H-3)
also has an influence on these three-bond couplings, as
can be seen, for example, in the values of J(H-1,H-7a)
4
3
Copyright  1999 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 37, 359–364 (1999)

362
E. KOLEHMAINEN ET AL.
Table 3. nJ(H,H) spin–spin coupling constants of 1, 2, 3, 6,
and 8
Table 4. nJ(H,H) spin–spin coupling constants of 4 and 5
7
4
5
nJ(H,H) (Hz)
nJ(H,H) (Hz)
nJ(H,H) (Hz)
Protons
a,4
3a,5
3a,6
3a,6a
4,5
Protons
Protons
1
2
3
6
7
8
3
3.70
0.00
0.48
5.95
0.52
1.31
1.21
1.02
6.49
0.69
1.04
ꢀ16.23
0.86
7.48
—
1,6
1,7
1,7
1.04
0.64
0.00
6.83
1.73
0.89
4.61
2.17
0.46
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
3
3
3
3
3
7
,1
,2
,3
ꢀ18.44
—
2.25
—
10.30
0.11
ꢀ0.51
8.56
—
—
—
—
—
10.30
0.18
0.18
—
8.56
ꢀ0.51
—
10.37
—
—
—
2.37
—
2.24
—
2.36
0
2.37
ꢀ2.37 ꢀ1.33
0.49 ꢀ1.23 ꢀ1.26
1,7a
1,8
,3a
,7a
,OH
,2
,3
3.33
3.45
—
2.80
2.09
—
—
—
—
—
5.62
0.53
2.86
—
2.67
1.84
6.46
—
—
—
2.73
1.98
—
—
—
—
—
5.77
4,6
4,6
6,7
6,7
0
0
0
0
0
0
2.25
ꢀ2.49
1.73
9.93
5.81
—
—
—
—
4,6a
5,6
6,7a
6,8
0
0
,3a
,7a
5,6
7,7
ꢀ12.69
—
5.69
5,6a
7,7a
7,8
0.92
0
,3
2.05
6,6
1.41
4.32
0.22
ꢀ0.14
0
,3a
,7a
,OH
,3a
,7a
,OH
a,7a
a,OH
a,OH
ꢀ1.79 ꢀ1.77
0.70 ꢀ1.81 ꢀ1.86
6,6a
7 ,7a
0
0
ꢀ0.15 ꢀ0.35
0.43 ꢀ0.47
0.00
—
2.20
0.00
—
7.34
—
—
6 ,6a
7 ,8
—
—
2.18
—
0.00
0.76
—
0.00
2.16
0.00
0.00
7.35
0.00
0.00
—
7a,8
2.12
0.00 ꢀ0.12
—
8.23
—
—
7.46
—
using symmetrical, CDCl matched, Shigemi micro-NMR
3
7.54
—
—
tubes, a direct, 5 mm diameter multinuclear probehead in
1
a 500 MHz ( H) instrument and inverse (proton) detection
—
—
in heteronuclear chemical shift correlation experiments,
HMQC and HMBC. The concentrations of the samples
were 4.39 mM for 2, 3.10 mM for 4 and 3.07 mM for 5.
As measured in Shigemi micro-NMR tubes, this means
that the amount of each sample was <0.5 mg. In proton,
of 1, 2, 7 and 8. These findings can be further utilized in
refining structure–NMR parameter relationships.
1
3
composite pulse, decoupled C NMR experiments run
overnight, all carbons of 4 and 5 were clearly detected
whereas in 2 carbons 4, 6, 7 and 8 did not give reli-
able resonance lines and carbon 5 was only poorly vis-
ible. In the case of 2, HMBC improved the situation,
revealing the signals of carbons 6 and 7. In the case
Sensitivity tests
Sensitivity tests were performed for heptachlor (2), ˛-
chlordene (4) and ꢀ-chlordene (5) (representing three
different molecular structures included in this work) by
Figure 1. HMQC contour plot of heptachlor (2).
Copyright  1999 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 37, 359–364 (1999)

2D NMR OF TRICYCLIC POLYCHLORINATED C10 HYDROCARBONS
363
Figure 2. INADEQUATE contour plot of heptachlor (2).
Figure 3. Experimental (measured in a Shigemi micro-NMR tube) (top) and calculated (WIN-DAISY)18 (bottom) H NMR
1
spectra at 500 MHz of heptachlor (2).
1
3
of 4 and 5, HMQC and HMBC gave complete chemical
shift correlation data. By using an inverse, 5 mm broad
band probehead, a reliable HMQC correlation map of 2
sensitivity in a direct C measurement. This property is
1
3
22
disadvantageous in cases where C resolution is critical.
For polychlorinated hydrocarbons, the highest possible
sensitivity in carbon detection is desirable. In our expe-
rience, the optimum production of both one-dimensional
(
4.39 mM) was produced in 3 h and of HMBC in 12 h.
Unfortunately, an inverse probehead suffers from poor
Copyright  1999 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 37, 359–364 (1999)

364
E. KOLEHMAINEN ET AL.
1
13
H and C NMR spectra and two-dimensional HMQC
and HMBC correlation maps, needed for reliable spec-
tral assignments and structure characterization of these
tricyclic polychlorinated hydrocarbons, can be achieved
with a direct probehead using solvent matched, micro-
NMR tubes and inverse (proton) detection.
Acknowledgements
Erkki Kolehmainen is grateful to the Academy of Finland and Jari
Koivisto to the European Union (Contract No: ENV-CT97-0768) for
financial support.
1
The H NMR spectral lineshape measured in micro-
REFERENCES
NMR tubes differs from a pure Lorenzian line. This causes
an increase in the R-factor (percentage difference between
the experimental and calculated spectra) of a computer-
based NMR spectral analysis performed by WIN-DAISY
software. In the case of micro-NMR tubes, the R-values
were <3% whereas in the case of normal, 5 mm diameter
NMR tubes, the values were generally <0.3%. According
to the copyright owner of the WIN-DAISY software,
1. J. Paasivirta, Chemical Ecotoxicology, p. 130. Lewis, Chelsea, MI
(
1991).
2
. J. Paasivirta, Toxicol. Environ. Chem. 66, 71 (1998).
3. H.-R. Buser and M. D. M u¨ ller, Anal. Chem. 64, 3166 (1992), and
references cited therein.
. J. R. Donnelly, G. W. Sovocool and R. L. Titus, J. AOAC Int. 76,
4
5
6
1092 (1993).
. J. W. ApSimon, A. S. Chau, R. Sampson, H. Tze and A. Fruchier,
Can. J. Chem. 60, 2002 (1982).
. R. H. Cox and J. D. McKinney, Org. Magn. Reson. 11, 541 (1978).
1
8
R-values <3% are acceptable. The main reason for
increased R-values when using micro-NMR tubes is due
to broad humps in spectral lines. Unfortunately, in spite of
prolonged shimming, we were not able to remove totally
these artefacts from the NMR spectral lines. Figure 3,
describing an experimental (measured in a Shigemi micro-
7. H. G. Fitzky and S. G a¨ b, Org. Magn. Reson. 14, 1 (1980).
8
9
. N. K. Wilson and G. W. Sovocool, Org. Magn. Reson. 9, 536 (1977).
. A. Derome and M. Williamson, J. Magn. Reson. 88, 177 (1990).
10. J. Jeener, B. H. Meier, P. Bachman and R. R. Ernst, J. Chem. Phys.
9, 4546 (1979).
1. G. Wagner and K. W u¨ thrich, J. Mol. Biol. 155, 347 (1982).
6
1
12. (a) A. Bax, Two-Dimensional Nuclear Magnetic Resonance in
Liquids, p. 157. Delft University Press, Reidel, Delft (1984);
(
b) D. L. Turner, J. Magn. Reson. 49, 175 (1982).
3. A. Bax, R. H. Griffey and B. L. Hawkins, J. Magn. Reson. 55, 301
1983).
1
NMR-tube) (top) and a WIN-DAISY iterated (bottom) H
1
(
NMR spectrum of heptachlor (2), reveals that the result
of computer analysis is satisfactory.
14. A. Bax and S. Subramanian, J. Magn. Reson. 67, 565 (1986).
5. A. Bax and M. F. Summers, J. Am. Chem. Soc. 108, 2093 (1986).
1
In conclusion, modern 2-D NMR measurement tech-
niques provide a unique tool for isomer specific structure
elucidation for polychlorinated hydrocarbons at submil-
ligram levels. In addition to the structural data which
are necessary for understanding the toxicity, metabolism
and/or fate of these compounds, pushing the detection lim-
its to concentrations as low as possible opens up new pos-
sibilities from an environmental analytical point of view.
16. D. H. T. Chien and J. R. Heys, J. Agric. Food Chem. 30, 399 (1982).
1
1
7. M. Kleiman and A. Goldman, US Pat. 2 672 486 (1954).
8. WIN-NMR (version 6.0) and WIN-DAISY (version 4.0). Bruker-
Franzens-Analytik 6m6H, Bremen.
1
9. E. Kolehmainen, K. Laihia, R. Laatikainen, J. Veps a¨ l a¨ inen,
M. Niemitz and R. Suontamo, Magn. Reson. Chem. 35, 463 (1997).
20. M. Karplus, J. Chem. Phys. 30, 11 (1959).
21. C. A. G. Haasnoot and F. A. A. M. De Leeuw and C. Altona,
Tetrahedron 36, 2783 (1980).
2
2. W. F. Reynolds, M. Yu and R. G. Enriquez, Magn. Reson. Chem.
35, 614 (1997), and references cited therein.
Copyright  1999 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 37, 359–364 (1999)