NMR spectroscopy of organolithium compounds, part XXXIII: trans-2,3-dimethylcyclopropyllithium aggregation in diethylether: An equilibrium between three different complexes of comparable energy and the influence of LiBr on aggregate structure

Article abstract of DOI:10.1002/poc.3800

The aggregate formation of trans-2,3-dimethylcyclopropyllithium (2) was studied in diethylether (DEE). With the help of the isotopic fingerprint method three clusters, a monomer, a dimer, and a fluxional tetramer in the ratio 1.00:0.53:0.22 were identified at 187?K. In the presence of 1 equivalent of LiBr cyclopropyllithium (1) forms in DEE at 163?K a mixed dimer and a mixed tetramer, while in THF a mixed dimer dominates. Under the same conditions 2 and 2,2,3,3-tetramethylcyclopropyllithium (3) form in the solvent mixture DEE/THF (1:1) mixed dimers, as does 2 also in THF.

Full text of DOI:10.1002/poc.3800

Received: 18 September 2017
DOI: 10.1002/poc.3800
Revised: 28 November 2017
Accepted: 1 December 2017
R E S E A R C H A R T I C L E
NMR spectroscopy of organolithium compounds, part
XXXIII*: trans‐2,3‐dimethylcyclopropyllithium aggregation
in diethylether: An equilibrium between three different
complexes of comparable energy and the influence of LiBr
on aggregate structure†
1
2
Thomas Fox | Harald Günther
1
University of Zürich, Institute of
Abstract
Chemistry, Zürich, Switzerland
2
The aggregate formation of trans‐2,3‐dimethylcyclopropyllithium (2) was
studied in diethylether (DEE). With the help of the isotopic fingerprint method
three clusters, a monomer, a dimer, and a fluxional tetramer in the ratio
University of Siegen, Chemistry
Department, Siegen, Germany
Correspondence
1.00:0.53:0.22 were identified at 187 K. In the presence of 1 equivalent of LiBr
Harald Günther, University of Siegen,
Chemistry Department, Adolf‐Reichwein‐
Straße, Siegen, Germany.
cyclopropyllithium (1) forms in DEE at 163 K a mixed dimer and a mixed
tetramer, while in THF a mixed dimer dominates. Under the same conditions
Email: guenmr@chemie.uni‐siegen.de
2
and 2,2,3,3‐tetramethylcyclopropyllithium (3) form in the solvent mixture
Funding information
Fonds der Chemischen Industrie,
Frankfurt, Germany
DEE/THF (1:1) mixed dimers, as does 2 also in THF.
KEYWORDS
13
1
6
C NMR, H NMR, Li NMR, aggregation, cyclopropyllithium, exchange, NMR
1
| INTRODUCTION
been found, as for instance, for n‐propyllithium in
[
7]
cyclopentane
and for mixed aggregates of methyl-
[
8a]
[9]
The formation of different aggregates like monomers,
dimers, tetramers, and larger species in ethereal and
lithium with LiI
or LiBr
and butyllithium with
[
8e]
[10]
LiBr
as well as for cyclopentadienyl systems.
At
hydrocarbon solvents is
organolithium compounds. Sometimes, as for example
for n‐butyllithium in diethylether (DEE)
phenyllithium in tetrahydrofuran (THF),
a
well‐known feature of
low temperature, mostly below −50°C, the lifetime of
these aggregates is long enough that the individual struc-
tures can be characterized by NMR spectroscopy, using
[
2]
[
3]
or
[4,5]
1
6/7
13
31 [11]
a particular
nuclei like H, Li, C, and P.
structure dominates to the extent that other structures
are not detectable, while quite often two different
complexes are in fast and reversible equilibrium as in
the case of n‐butyllithium in THF a tetramer and a
We have recently reported that cyclopropyllithium (1)
forms static tetramers in DEE and dimers as well as
[
1]
fluxional tetramers in THF. In a fluxional aggregate,
intra‐aggregate exchange of lithium cations is still fast
on the NMR timescale while inter‐aggregate exchange is
slow. For trans‐2,3‐dimethylcyclopropyllithium (2) and
for 2,2,3,3‐tetramethylcyclopropyllithium (3), on the other
hand, only dimers were observed in the solvent mixture
DEE/THF (1:1) as well as for 3 also in DEE.
[4]
dimer or in the case of iso‐propyllithium in hydrocar-
[6]
bon solvents a tetramer and a hexamer. In addition,
equilibria between larger numbers of complexes have also
*
[1]
For part XXXII, see
.
6
[7]
†
Dedicated with best wishes to Professor Dieter Seebach on the occasion
We now found, using the NMR‐friendly Li isotope
th
1
13
of his 80 birthday.
as well as H and C NMR, that 2 forms in DEE at
J Phys Org Chem. 2018;e3800.
wileyonlinelibrary.com/journal/poc
Copyright © 2018 John Wiley & Sons, Ltd.
1 of 11
https://doi.org/10.1002/poc.3800

2
of 11
FOX AND GÜNTHER
1
87 K three different aggregates in the ratio 1.00:0.53:0.22,
frequency solvent signal of [D ]DEE at 14.5 ppm. How-
10
a rather unusual result for alkyllithium compounds. In
cases where two aggregates coexist, a third aggregate of
higher or lower aggregation number may often be pres-
ent, but its concentration is usually too low to allow struc-
tural characterization.
ever, it was possible to measure the chemical shifts but
13
6
13
not the C, Li coupling constants of the C resonances
[
13]
from a 2D‐HMQC spectrum
(see later Figure 5).
The aggregation number n of the 3 aggregates was
therefore derived from a sample of an equimolecular mix-
ture of C‐1 deuterated and C‐1 non‐deuterated 2
(
2 /2 = 1:1) with the help of the isotopic fingerprint
d
2
2
| RESULTS
[
8a, 11c, d]
method.
The deuterium induced isotope effect
[
8,9a]
over 2 bonds leads, as is well documented,
frequency shift of the Li resonance in the order of 5 to
to a high‐
.1 | The solution structure of 2 in DEE
6
A 0.2 M solution of 2—used as mixture of S and R enan-
tiomer—in [D ]DEE shows at 263 K in the 58.88‐MHz
Li NMR spectrum a slightly broadened signal at δ 2.12
10 ppb. In the spectrum of the mixture 2 /2 = 1:1, a mono-
d
mer is then characterized by a doublet formed from the
10
6
6
Li resonances in the deutrated (d) and non‐deuterated
with a half width of 0.75 Hz that further broadens at lower
temperatures until at 187 K, 3 sharp resonances with
relative intensities of 1.00:1.23:1.15 emerge at 2.22, 2.08,
and 2.02 ppm (Figure 1). The high temperature Li signal
thus arises from 3 different aggregates that are in fast
inter‐aggregate exchange.
(h) compound. For a dimer, we have in the region of slow
exchange at low temperature, following Pascals's triangle,
a 1:2:1 triplet according to the clusters (dd), (dh) and (hd),
and (hh). The observed multiplets are “shift multiplets”
and not to be confused with similar multiplets caused by
first‐order spin,spin coupling. As has been pointed out
6
[
2c]
The most interesting question is now the determina-
tion of the aggregation number n of the individual
clusters. This is usually solved by observing the one‐bond
recently by Reich, the method is related to experiments
[
14]
by Brown
who used similar lithium reagents to deter-
7
mine the cluster size from the number of Li signals
obtained after mixing and ligand exchange. Experiments
using this technique with lithium enolates have been pub-
6
13
Li, C coupling constant of the particular aggre-
[
4,7,12]
[12]
gate,
where the empirical Equation 1
[
15]
lished recently.
Provided that deuterium‐induced iso-
À
Á
6
1
6
13
tope shifts for a particular Li signal can be measured,
the isotopic fingerprint method has the advantage that—
except for the deuteration—all components of the mixture
have an identical ligand structure.
For the Li NMR spectrum of the Li labeled mixture
2d/2 = 1:1, we observed at 177 K a quintuplet, a triplet,
and a doublet (Figure 2). The triplet and the doublet result
from a dimer and a monomer, respectively, while the
J Li; C
½Hzꢀ ≈ ð17±2Þ=n
(1)
obs
relates the coupling to the number n of equivalently
6
coupled Li nuclei and thus to the aggregation state. In
6
6
the present case, this analysis was prevented by the
13
overlap of the C‐1 signals, expected by the results in the
[1]
mixed solvent DEE/THF close to δ 17, with the low‐
6
FIGURE 2 Isotopic fingerprints for the 3 Li NMR signals of
2
d
/2 = 1:1 at δ 2.02, 2.08, and 2.22 (cf. Figure 1) at 177 K;
solvent [D10]DEE (M = monomer, D = dimer, FT = fluxional
6
FIGURE 1 58.88‐MHz Li NMR spectrum of 2 in DEE at various
2
6
2/1
temperatures; relative intensities at 187 K 1.00:1.23:1.15, Gaussian
resolution was used for this spectrum
tetramer); isotope shifts Δ Li( H) are 3.4 ppb (M), 5.9 ppb (D),
and 5.0 ppb (FT)

FOX AND GÜNTHER
3 of 11
quintuplet characterizes a fluxional tetramer. A dimer and
a fluxional tetramer was already observed for 1 in THF,
[1]
but a monomer was in this case not found. In the flux-
ional tetramer with intra‐aggregate exchange, the number
of “next” neighbors is 4 compared with 3 in the static tet-
6
ramer and thus provides the Li environments dddd,
(
dddh,ddhd,dhdd,hddd),
(ddhh,dhdh,dhhd,hdhd,hddh,
hhdd), (hhhd,hhdh,hdhh,dhhh), and hhhh that lead to a
6
1
:4:6:4:1 quintuplet for the Li resonance. Fluxional tetra-
mers were found, for example, for tert.butyllithium in ben-
7
13
[16]
zene with a Li, C splitting of 10.7 Hz that corresponds
6
13
[11d]
6
13
to a Li, C coupling of 4.1 Hz,
and with Li, C cou-
plings of 4.1, 4.3, and 3.98 Hz for tert.butyllithium in
[
2j,17]
cyclopentane,
pentyllithium in DEE, respectively.
mer was also reported for phenyllithium and butyllithium
and for sec.pentyllithium and tert.
[2j,6a]
A fluxional tetra-
[8b,e]
[18]
in DEE,
for vinyllithium in THF, and as intermedi-
ate for the Li exchange between the homodimers (BuLi)2
and (PhLi) in a mixture of both compounds in DEE/
2
[8e]
TMEDA (tetramethylethylenediamine).
The Li NMR results are supported by the 400‐MHz H
NMR spectrum, where a triplet at −2.22 ppm for 1‐H at
6
1
2
63 K changes at 187 K to 3 different, partly overlapping
1
FIGURE 3 A, 400‐MHz H NMR signals of 1‐H of 2 at 187 K;
triplets between −2.30 and −2.45 ppm. The triplet struc-
ture arises because the two coupling constants J(1‐H,Hsyn)
and J (1‐H,Hanti) (see Scheme 1) are of similar magnitude
1
6
3
B, 400/58.88‐MHz H, Li HOESY spectrum of 2 at 187 K
(M = monomer, D = dimer, FT = fluxional tetramer)
(
3
(
c. 5.5 and 6.3 Hz) (Figure 3A). An additional low‐intensity
triplet of an unidentified species (<5%) is observed at
2.53 ppm. Further resonances, however broadened by
signal overlap, are observed for 2‐H (−0.15 ppm), 3‐Ha
at 263 K, that is in the region of fast inter‐ and intra‐aggre-
gate exchange, turns into a complex line spectrum for the
slow exchange domain at 187 K (Figure 4).
−
The assignment for the different carbon resonances
s
13
1
(
(
0.50 ppm), and the methyl groups (1.0–1.2 ppm)
Figure 3B, F ). The assignments for δ(H ) and δ(H ) were
was achieved by a C, H‐HMQC spectrum (Figure 5) that
also proved the signal overlap with the low‐frequency
DEE signal for δ(C‐1) and allowed the indirect measure-
1
s
a
made at 253 K by nuclear Overhauser effect (NOE)
difference spectroscopy
nances δ(4‐H) and δ(5‐H) followed from H, H‐COSY spec-
troscopy
spectrum
ment of the 1‐H resonances and confirmed the
[19]
13
and those for the methyl reso-
ment of the C‐1 resonances of the aggregates close to
1
1
13 ppm. The remaining cross peaks did not show enough
fine structure to allow assignments for individual aggre-
gate signals but identified the δ‐region of the different car-
bon resonances.
[
20]
6
1
at 263 K. Finally,
a
Li, H HOESY
[11e,21]
at 187 K (Figure 3B) allowed the assign-
1
H
assignment by cross peaks of the monomer and the dimer
Li signals with 2‐Hs.
Starting at high frequency with the spectrum recorded
at 187 K (Figures 4 and 5), the signal group around
24.5 ppm is assigned to C‐5. Aside from an unidentified
broadened signal centered at 24.8 ppm, 2 singlets at
24.52 and 24.48 ppm can be assigned to the dimer and
the tetramer. The signal at 24.42 ppm then comes from
the monomer. At ca. 21.5 ppm, 2 signals for 2 are
superimposed and accompanied by a singlet. They can
be assigned to C‐4 in the three aggregates and are presum-
ably due to dimer/tetramer and monomer. More compli-
cated is the signal group between 19.8 and 20.8 ppm
that originates from C‐2 and C‐3 where we expect 6 sig-
nals from the different aggregates. The HMQC spectrum
yields δ(C‐3) > δ(C‐2), and a detailed assignment could
be based on the comparison with the spectra recorded
6
Further evidence for the coexistence of 3 aggregates
13
came from the C NMR spectrum. Not unexpected, the
13
1
rather simple C{ H} spectrum of the DEE sample of 2
SCHEME 1 Structures of compounds 1 to 3

4
of 11
FOX AND GÜNTHER
1
3
FIGURE 4 100.61‐MHz C NMR spectra of 2 at 263 K, and
87 K; the C‐1 resonance at ca. 13 ppm is covered by the [D10
1
]
DEE solvent signals at 14.5 ppm; at x signals of an unidentified
1
3
FIGURE 6 A, Partial 100.61 MHz C NMR spectrum of 2 with
assignment (FT = fluxion tetramer, M = monomer, D = dimer);
temperature 187 K, Gaussian resolution enhancement; B, the same
species
d
spectral region for 2 /2 = 1:1 with relative signal intensities given in
brackets; temperature 187 K
2
13 2/1
13
13
Δ C( H), on C‐2 and C‐3. As shown in the dia-
grams of Scheme 2, these shifts are restricted to the car-
bons of the ligand that is deuterated at C‐1. By contrast,
6
2
6
2/1
6
the isotope shifts for Li, Δ Li( H) affect all Li next
neighbors and in a fluxional aggregate all lithium cations
2
13 2/1
of the cluster. Consequently, the Δ C( H) data are the
same for all ligands in a particular aggregate, and we
13
expect for the C resonances of C‐2 and C‐3 6 additional
1
3
1
FIGURE 5 100.61/400 MHz C, H‐HMQC spectrum of 2 at
1
3
1
90 K; DEE = solvent signal of [D10]DEE; no C decoupling was
1
3
2
used during t . For a higher resolution of C domain between 19
and 21 ppm, see Figure 6
for the C‐1 deuterated species where deuterium‐induced
13
C isotope shifts exist.
In Figure 6, we compare the spectral C region
between 19.8 and 20.8 ppm (Figure 6A) with the same
region in the spectrum of the mixture 2 /2 (1:1)
1
3
d
SCHEME 2 Schematic diagrams of the deuterium‐induced
1
3
6
(Figure 6B). This spectrum shows also the signals caused
isotope shifts for C and Li in a dimer; the isotope effect works
6
by the deuterium induced two‐bond isotope shifts,
for both Li nuclei but only for the carbons in one ligand

FOX AND GÜNTHER
5 of 11
1
3
13
1
TABLE 1
C NMR frequencies and relative intensities of the signals of C‐2 and C‐3 in the 100.61 MHz C{ H} spectra of Figure 6
Spectrum 6(A)
Signal
1
2
3
4
5
6
Frequency (ppm)
Relative intensity
20.882
1
20.766
1
20.315
1
20.269
1
20.051
1
20.025
1
‐‐‐
‐‐‐
‐‐‐
‐‐‐
Spectrum 6(B)
Signal
1
2
3
4
5
6
7
8
Frequency (ppm)
Relative intensity
20.762
1
20.681
2
20.608
1
20.306
1
20.273
1
20.187
2
20.024
2
19.927
2
resonances due to the deuteration at C‐1, that is a total of
2 signals.
Usually,the Cisotopeeffectover2bonds, Δ C( H),
causes shifts of ca. 100 ppb to low frequency (high field).
However, signal overlap, indicated by double intensity,
reduces in the present case the total number to 8
resonances. In addition, there are unexpected frequency
differences for the signals A(1) and B(1) that should arise
from δ(C‐3) in the non‐deuterated molecules of 2 as well
deuterated molecules of
2
50% of signal B2
1
(20.681 ppm). This results in a shift of 201 ppb to lower
frequency (high‐field) that is not an isotope‐induced shift.
The isotope‐shifted signal for the fluxional tetramer is
13
2 13 2/1
2
13 2/1
then B3, and the isotope effect Δ C( H) for C‐3
amounts to 73 ppb. For C‐2, the fluxional tetramer signal
A5 (20.051 ppm) is now shifted to B7 (20.024 ppm) that
has double intensity. Its isotope‐shifted pendant coincides
with the monomer signal B8. The result is that the tetra-
mer C‐2 signal for the non‐deuterated molecules in the
2d/2 mixture is again shifted to lower frequency (high
field), however only by 27 ppb. The isotope effect
as of those in the mixture 2 /2. In order to analyze these
d
observations, we reproduce the frequencies and relative
intensities of the signals of C‐2 and C‐3 in Table 1. A similar
analysis for the C‐4 and C‐5 resonances is prevented by
strong line overlap.
2
13 2/1
Δ C( H) for C‐2, that is the frequency difference B7‐
B8 amounts to 97 ppb. The complete result is collected in
If we regard the ligand 2 as a structural rigid unit,
that is a building block in all clusters and consider—as
outlined earlier—the fact that the deuterium induced
Table 2.
The interesting observation is that the frequency
differences Δδ for the C signals of the H‐species in the
samples with pure 2 and with a (1:1) mixture of 2 /2,
13
13
13
isotope effect on C‐2 and C‐3 is identical for all 2d
d
molecules in a particular cluster, we should see pairs of
C NMR signals at the same frequency, 1 signal in spec-
trum 6(A) for the aggregates of 2 and 1 signal in spectrum
are for the monomer, as expected, practically zero and
for the dimer with maximal 9 ppb well in the order of
the experimental error. By contrast, those for the fluxional
tetramer are with values of 27 ppb (C‐2) and 201 ppb (C‐3)
significantly larger. Because the isotope shifts for all
aggregates are comparable, the basic ligand structure is
apparently unchanged by aggregate formation, and the
aggregate dynamic in the fluxional tetramer, where four
ligands participate in the dynamic process, does not
change the magnitude of the isotope shifts. But what
is the origin of the chemical shifts of C‐2 and C‐3 in
the non‐deuterated ligands of the fluxional tetramer in
13
6
2
(B) for the non‐deuterated molecules in the mixture 2d/
. In addition, the 6 signals of the isotope‐shifted reso-
nances from the 2 molecules in the mixture 2 /2 are
d
d
expected in spectrum 6(B) at lower frequencies (higher
field).
As the inspection of Figure 6 shows, signal pairs of the
same frequency are observed only for the monomer and
the dimer. For the monomer, we have A2/B1 (20.766/
2
0.762 ppm) and A6/B7 (20.025/20.024 ppm). One of the
signals at B2 and signal B8 must then be the isotope
the mixture 2 /2? A convincing explanation, how the
deuterated species could affect the chemical shift of
d
2
13 2/1
shifted monomer signals in 2 . This yields Δ C( H)
d
data of 81 ppb for C‐3 and 97 ppb for C‐2.
For the dimer, we have two closely spaced signals in
both spectra. The pair A3/B4 (20.315/20.306 ppm) is
assigned to C‐3 and the pair A4/B5 (20.269/20.273) to C‐
2
13 2/1
TABLE 2 Isotope shifts Δ C( H)/ppb and chemical shifts not
induced by deuteration, Δδ/ppb, for the resonances of C‐2 and C‐3
2
. The isotope‐shifted signals are then superimposed at
Monomer
Dimer
Fluxional Tetramer
2
13 2/1
signal B6 leading to isotope effects Δ C( H) of
1
2
2
2
Δ
Δδ
Δ
Δδ
Δ
Δδ
19 ppb for C‐3 and 86 ppm for C‐2.
C‐2
C‐3
97
0.001
97
−0.004
119
27
For the fluxional tetramer, we are now left with signal
86
0.004
81
0.009
73
201
A1 (20.882 ppm) and in the 2 /2 mixture for the non‐
d

6
of 11
FOX AND GÜNTHER

FOX AND GÜNTHER
7 of 11
the non‐deuterated ones is not at hand, and we believe
that a small difference in the sample temperature is
responsible. For both spectra, the temperature was
controlled by the methanol thermometer and could not
be measured inside the sample tube, a well‐known draw-
back in low‐temperature NMR. For separate measure-
ments, experimental errors σ may thus lead to small
temperature differences, in our case 187 ± σ K. For a
dynamic system like the fluxional tetramer, a tempera-
ture‐induced shift of 0.2 ppm as for δ(C‐3) and only
SCHEME 3 Schematic structures for a static tetramer (1•LiBr)
and a dimer (1•LiBr) in solution
2
0
.03 ppm as for δ(C‐2) is thus not surprising. Because
similar shifts for the static aggregates, monomer and
dimer, are absent, the finding for the tetramer underlines
again its fluxional nature. The complete spectral data are
collected in Table 3.
temperature was inserted in the pre‐cooled magnet. On
[
22]
the basis of the X‐ray results,
we can tentatively assign
the 2 smaller singlets to a static mixed tetramer (1•LiBr)2
that has 2 different lithium environments, Li1(C,Br,Br)
and Li2(C,C,Br), also found for CH Li/LiBr (1:1) in
3
2
.2 | Aggregate formation in the presence
[
9a]
DEE.
The larger signal comes than forms a mixed
of lithium bromide
dimer 1•LiBr where only 1 lithium environment exists
2.2.1 | Cyclopropyllithium (1)
(
Scheme 3).
A completely different Li spectrum was observed
6
For crystals obtained from 1‐bromocyclopropane (4) by
reaction with lithium metal, a mixed tetramer (1•LiBr)
was found in the solid by X‐ray crystallography.
a Li singlet at δ 0.67 (rel. to 1.0 M LiBr in THF) was
measured at 230 K for a 0.2 M sample prepared from 4
and Li metal in DEE.
caused line broadening until finally at 163 K three singlets
at δ 0.96, 0.43, and 0.20 with an intensity ratio of
[
23a,b]
when the solvent was changed to THF.
A 1.6 M
2
[22]
sample showed aside from a large singlet at δ 0.85 two
smaller ones at δ 0.26 and 0.07 (1.0:0.4:0.09). The large
signal was identified as belonging to a mixed dimer 1•LiBr
Later,
6
13
6
[23a]
by the quintuplet in the C spectrum with a one‐bond
Lowering the temperature
1
13
6
13
coupling J( C, Li) of 10.9 Hz observed for C‐1 as well
1
6
as H, Li HOESY correlations for 1‐H and 2,3‐H . Conse-
s
1
1
quently, H, H NOESY peaks with 1‐H dominated for
2,3‐Ha (Scheme 1). The origin of the smaller signals
1
.00:1.07:1.71 were observed (Figure 7). Further measure-
1
13
ments ( H, C NMR, or 2D spectroscopy) were prevented
by rapid crystallization, and recording the Li spectrum
was only possible when the sample prepared at room
13
6
6
remains unknown because one‐bond C, Li coupling
could not be observed.
2
.2.2 | Trans‐2,3‐
dimethylcyclopropyllithium (2)
Mixed dimer formation of 2•LiBr was found exclusively
when
1‐bromo‐trans‐2,3‐dimethylcyclopropane
was
reacted with lithium metal in DEE/THF (1:1). The isoto-
pic fingerprint method yielded for a 1:1 mixture of 2
d
1
6
and 2 a doublet in the H‐decoupled Li NMR spectrum
h
2
6
2/1
(
Figure 8A, Δ Li( H) = 4.2 ppb). The intensity changes
of the doublet lines observed when the NOE of H on Li
was suppressed during H decoupling (Figure 8B) showed
1
6
1
that the NOE affects the low‐frequency signal of the
mixed dimer 2 •LiBr. The high‐frequency signal thus
h
results from aggregates 2 , in accord with the known
d
6
high‐frequency deuterium‐induced isotope effect for Li.
C{ H} NMR spectrum, a quintuplet with
J( C, Li) = 10.1 Hz was observed at 187 K (Figure 8C),
13
1
In the
1
13
6
6
FIGURE 7 58.88‐MHz Li NMR spectrum of cyclopropyllithium
1
in the presence of LiBr (1:1) without H decoupling at various
and a HOESY spectrum showed cross peaks for Li/1‐H
[23a]
temperatures in DEE
; external reference 1.0 M LiBr in THF
and Li/2‐H that agrees with a proposed structure where
s

8
of 11
FOX AND GÜNTHER
6
as for 1•LiBr was observed in the Li NMR spectrum with
3
1
signals at δ 0.82, 0.28, and 0.07 in the intensity ratio of
.0:0.23:0.03. Again, only the dominating signal at δ 0.82
could be identified as belonging to a mixed dimer 2•LiBr
13
1
13
6
by a C quintuplet at δ16.6 with J( C, Li) = 11.0 Hz.
2
.2.3 | 2,2,3,3‐
Tetramethylcyclopropyllithium (3)
1
6
13
H‐NMR, Li‐NMR, and C‐NMR gave similar results as
6
described above for 2. The 58.88 MHz Li NMR spectrum
showed in DEE/THF (1:1) the Li signal at 0.69 ppm (rel.
to ext. LiBr in THF) and the isotopic fingerprint displayed
a doublet ( Δ Li( H) = 5.6 ppb) indicating the formation
of a mixed dimer 3•LiBr (Figure 8D). This was supported
2
6
2/1
13
by the quintuplet observed for the C signal of C‐1 at δ
3
1
13
6
2.09 with a coupling of J( C, Li) = 10.8 Hz (Figure 8E).
Two HOESY cross peaks Li/1‐H and Li/4,4′‐H suggest a
structure similar to 5.
3
| DISCUSSION
6
FIGURE 8 Isotopic fingerprint in the 58.88‐MHz Li NMR
The most remarkable aspect seems to us the finding that 2
forms in DEE three aggregates of comparable energy and
that an alkyllithium system like 2 exists as a stable mono-
mer in the absence of additional complexation agents.
Especially the formation and high concentration of a
monomer, where DEE coordination apparently reduces
the tendency for the formation of higher aggregates,
are of interest. Such an effect is mostly seen in THF.
Monomer formation has been found for phenyllithium
2
6
2/1
spectrum of 2•LiBr and 2
d
•LiBr (1:1); Δ Li( H) = 4.2 ppb:
1
A, with NOE through H decoupling and B, with NOE suppression;
1
3
8
temperature 178 K, 0.2 M in [D10]DEE/[D ]THF (1:1); C, C NMR
1
13
6
quintuplet observed for C‐1 in 2•LiBr at 187 K, J( C, Li) = 10.1 Hz;
D, isotopic fingerprint in the 58.88‐MHz Li NMR spectrum of a 1:1
6
mixture of 3
h
•LiBr and 3
d 8
•LiBr 0.2 M in [D10]DEE/[D ]THF (1:1) at
1
2
6
2/1
1
5
60 K with NOE suppression during H decoupling; Δ Li( H) =
1
3
.6 ppb; E, C NMR quintuplet observed under the same conditions
1
3
1
13
6
as in D for C‐1 in 3•LiBr, J( C, Li) = 10.8 Hz
[
8b, 12]
in dilute THF solution
and several alkyllithium
in THF or in the presence of complexa-
tion agents like pentamethyldiethylenetriamine or
[
12,24]
compounds
the Li‐Li distance of the double lithium bridge is parallel
to the C‐2/C‐3 bond of the three‐membered ring as shown
in Scheme 4. In the solvent THF, the same signal pattern
1
13
6
hexamethylphosphoramide with J( C, Li) values of 14
1
13
6
to 17 Hz. J( C, Li) data around 17 Hz are abundant for
[
4]
halogen‐lithium carbenoids,
but the formation of
higher aggregates is here prevented by the substituent X
in the C‐1(Li,X) fragment. Recently published examples
of monomer stabilization by THF were found for
[
25]
trisolvated 1‐penylcyclopropyllithium
iso‐propyllithium.
and trisolvated
In DEE/THF (1:1), the methyl‐
substituted cyclopropyl compounds 2 and 3 form only
[
26]
[
1]
dimers.
6
From the relative intensities of the Li NMR signals of
at 187 K (Figure 1), one derives, considering the different
2
number of lithium cations in monomer (M), dimer (D),
and fluxional tetramer (FT), a ratio of 1.00:0.53:0.22 and
concentrations of 12.6% (FT), 30.3% (D), and 57.1% (M).
o
With the van't‐Hoff equation, ΔG = RT ln K, free energy
SCHEME 4 Assumed structure for the mixed dimer of 2•LiBr;
shown is the S enantiomer of 2
differences between these aggregates of 1.4 (D/FT),
2.3 (M/FT), and 1.0 (M/D) kJ/mole are calculated.

FOX AND GÜNTHER
9 of 11
TABLE 4 Difference between dimer chemical shifts (ppm) for saltfree samples (homodimers) and samples that contain LiBr (mixed
dimers); negative values stand for shielding
1
3
6
C
Li
Compound/
Solvent
Temp./K
Saltfree
Temp./K
+LiBr
Diff.
Temp./K
Saltfree
Temp./K
+LiBr
Diff.
1
2
3
/THF
173
−8.02
163
−8.60
−0.58
173
1.67
163
0.85
−0.82
/DEE/THF
/DEE/THF
163
175
17.50
31.40
187
160
16.45
32.09
−1.05
187
175
1.74
2.10
178
160
0.72
0.69
−1.02
−1.41
0.69
[8]
Noteworthy is also the fact that 3 forms in pure DEE
As far as the isotopic fingerprint method
is
[1]
only dimers, apparently due to severe steric hindrance.
In this respect, one has to remember that steric strain in
organolithium aggregates is more easily eliminated than
in rigid structures because these systems are mostly in
chemical equilibrium with molecular arrangements
where steric strain may be minimized or completely
absent.
concerned, our results for the mixed aggregates show that
in the absence of independent information, an isotopic
fingerprint doublet may be misinterpreted as resulting
from a monomer. Only in homo‐aggregates or in cases
where deuteration is possible for all ligands, the interpre-
tation of the shift multiplets is straightforward.
Because of the chirality of 2, diastereomeric dimers
4
4
| EXPERIMENTAL
(
R,R + S,S) and (R,S + S,R) were found in the solvent mix-
[1]
13
ture DEE/THF. This caused signal splitting in the
C
.1 | Synthesis and sample preparation
13
NMR spectrum for various C resonances. In the present
case, such splittings were not observed. Apparently, the
lifetime of these complexes in DEE is, in the temperature
range of our measurements, too short to allow their
detection by NMR. On the other hand, THF as the
stronger donor as compared with DEE may stabilize these
dimers to the extent that detection by NMR is possible.
In contrast to our results obtained for methyl‐ and n‐
[
1]
Salt‐free samples of 2 were prepared as described by
reaction of lithium‐6 powder with bis‐(trans‐2,3‐
dimethylcyclopropyl)‐mercury via trans‐2,3‐dimethylcy-
clopropylbromide following the procedure known for the
synthesis of dicyclopropylmercury.
proceeded as outlined in detail in.
that contained lithium bromide, the corresponding
organic bromides—cyclopropyl bromide for 1, trans‐2,3‐
dimethylcyclopropyl bromide for 2, and 2,2,3,3‐
tetramethylcylopropyl bromide for 3—obtained from the
geminal bromides
hydride, or, for the deuterated species, with tributyltin-
[
28]
Sample preparation
[
1]
For samples
[8e,9a]
butyllithium,
the number of different mixed
complexes with LiBr is drastically reduced for the
cyclopropyllithium systems 1 to 3.
[
29,30]
The large high‐frequency (downfield) shifts for δ(C‐1)
in the methyl‐substituted mixed aggregates with respect
by reduction with tributyltin-
13
[31]
to the C‐1 resonance of 1•LiBr—25.1 ppm for 2•LiBr
and 40.7 ppm for 3•LiBr (Table 3)—are a consequence
of the deshielding β‐effects of 2 and 4 methyl groups,
respectively. They exceed the average β‐effect for 1 methyl
deuteride —were treated under argon with excess
lithium‐6 powder or lithium chips followed by filtration
from the metal excess into an NMR tube that was degasses
at −90°C and sealed. Deuterated solvents, [D ]DEE and
10
[27]
group of +9.4 ppm considerably, but they are not much
[D ]THF, carefully dried were used throughout.
8
different from the corresponding variation in the saltfree
[1]
samples (25.5 and 39.4, respectively). Also, the compar-
4
.2 | NMR spectra
13
ison between δ( C) for C‐1 in the homodimers and the
mixed dimers, shown in Table 4, does not yield a clear
trend for a possible bromide substituent effect through
the double lithium bridge on δ(C‐1). On the other hand,
A Bruker Analytische Meßtechnik NMR spectrometer
AMX 400 was used together with the standard Bruker soft-
ware to record 1‐dimensional and 2‐dimensional NMR
6
1
13
notwithstanding small temperature effects for the Li
spectra at 400.13 MHz ( H), 100.61 MHz ( C), and
6
13
chemical shifts, these data (Table 4) clearly indicate a
58.88 MHz ( Li). All C spectra were measured with pro-
ton decoupling. Low temperature spectra were controlled
with the Bruker low‐temperature unit which was cali-
6
bromide shielding effect for Li. For 1, one finds shielding
for the mixed dimer compared with the homodimer of ca.
[
32]
0
.8 ppm, for 2 of ca. 1 ppm, and for 3 of ca. 1.4 ppm.
brated with the methanol thermometer. Solvent signals
Because we compare only dimers in the same solvent, a
bromide shielding effect on δ( Li) in the mixed aggregates
seems well established.
which were referenced to TMS served as internal stan-
6
1
13
dards for H and C, while an external reference of LiBr
6
(if not stated otherwise 0.1 M in THF) was used for Li.

10 of 11
FOX AND GÜNTHER
ACKNOWLEDGEMENTS
[8] a) O. Eppers, H. Günther, Helv. Chim. Acta 1990, 73, 2071; b) O.
Eppers, H. Günther, Helv. Chim. Acta 1992, 75, 2553; c) K.
Bergander, R. He, N. Chandrakumar, O. Eppers, H. Günther,
Tetrahedron 1994, 50, 5861; d) H. Günther, J. Braz. Chem. Soc.
1999, 10, 241; e) B. Böhler, H. Günther, Helv. Chim. Acta
We are indebted to Prof. H. Ihmels, Siegen, for making
available technical assistance and to the Fonds der
Chemischen Industrie, Frankfurt, Germany, for financial
support.
2
015, 98, 427.
9] a) T. Fox, H. Hausmann, H. Günther, Magn. Reson. Chem. 2004,
2, 788; b) S. Desjardins, K. Flinois, H. Oulyadi, D. Davoust, C.
Giessner‐Prettre, O. Parisel, J. Maddaluno, Organometallics
003, 22, 4090.
[
4
ORCID
2
Harald Günther http://orcid.org/0000-0002-2685-1175
[
10] a) W. Bauer, G. A. O'Doherty, P. v. R. Schleyer, L. A. Paquette,
J. Am. Chem. Soc. 1991, 113, 7093; b) M. R. Sivik, W. Bauer,
P. v. R. Schleyer, L. A. Paquette, Organometallics 1996, 15, 5202.
REFERENCES
[
1] T. Fox, H. Günther, J. Phys. Org. Chem. 2017, 30, 3746. https://
[11] For reviews see: a) H. J. Reich, Ref. 2c; b) D. Li, I. Keresztes, R.
doi.org/10.1002/poc.3746
Hopson, P. G. Williard, Acc. Chem. Res. 2009, 42, 270; c) H.
6
,7
Günther, High resolution Li NMR of organolithium com-
pounds, in Encyclopedia of Nuclear Magnetic Resonance, (Eds:
D. M. Grant, R. K. Harris) Vol. 5, John Wiley & Sons, Ltd,
[
2] For reviews see: a) V. Capriati, F. M. Perna, A. Salomone,
Dalton Trans. 2014, 43, 14204; b) V. H. Gessner, C. Däschlein,
C. Strohmann, Chem. – Eur. J. 2009, 15, 3320; c) H. J. Reich,
Chem. Rev. 2013, 113, 7130; d) Z. Rapaport, I. Marek, The Chem-
istry of Organolithium Compounds, John Wiley & Sons, Ltd.,
Chichester, U.K. 2006; e) J. B. Grutzner, Application of NMR
in Carbanion Chemistry, in Encyclopedia of Nuclear Magnetic
Resonance, (Eds: D. M. Grant, R. K. Harris) Vol. 9, John Wiley
6,7
Chichester 1996, 2807; d) H. Günther, High resolution Li
NMR of organolithium compounds, in Advanced Applications
of NMR to Organometallic Chemistry, (Eds: M. Gielen, R.
Willem, B. Wrackmeyer), Wiley‐VCH Verlag GmbH, Weinheim
1996, 247; e) W. Bauer, ‘NMR of organolithium compounds:
general aspects and application of two‐dimensional
heteronuclear Overhauser effect spectroscopy (HOESY)’ in
Ref. 2f, p 125.
&
Sons, Ltd., Chichester, U.K. 2002, 481; f) A.‐M. Saspe, P. v. R.
Schleyer (Eds), Lithium Chemistry—a Theoretical and Experi-
mental Overview, Wiley, Chichester, U.K 1995; g) D. B.
Collum, Acc. Chem. Res. 1993, 26, 227; h) L. M. Jackman, J.
Bortiatynski, in Advances in Carbanion Chemistry, (Ed: V.
Snieckus), Jai Press, Greenwich (Connecticut) 1992, 45; i) W.
Bauer, P. v. R. Schleyer, in Advances in Carbanion Chemistry,
[12] W. Bauer, W. R. Winchester, P. v. R. Schleyer, Organometallics
1987, 6, 2371.
[
13] a) L. Müller, J. Am. Chem. Soc. 1979, 101, 4481; b) A. Bax, R. H.
Griffey, B. L. Hawkins, J. Magn. Reson. 1983, 55, 301; c) S.
Braun, H.‐O. Kalinowski, S. Berger, 150 and More Basic NMR
Experiments, Wiley‐VCH, Weinheim 1998, 384; d) H. Günther,
NMR Spectroscopy, 3rd ed., Wiley‐VCH, Weinheim 2013, 389.
(
8
Ed: V. Snieckus), Jai Press, Greenwich (Connecticut) 1992,
9; j) R. D. Thomas, in Isotopes in the Physical and Biomedical
Sciences, Eds. E. Buncel, J. R. Jones, Elsevier, Amsterdam,
991, Chap 7, 367; k) H. Günther, D. Moskau, P. Bast, D.
1
[
14] L. M. Seitz, T. L. Brown, J. Am. Chem. Soc. 1966, 88, 2174.
Schmalz, Angew. Chem. 1987, 99, 1242 Angew. Chem. Int. Ed.
Engl. 1987, 26, 1212; l) R. O. Bach (Ed), Lithium: Current Appli-
cations in Science, Medicine, and Technology, John Wiley &
Sons, New York 1985; m) J. L. Wardell, Alkali Metals, in Com-
prehensive Organometallic Chemistry, (Eds: G. Wilkinson, F. G.
A. Stone, E. W. Abel) Vol. 1, Pergamon Press, Oxford 1982 43;
n) M. Schlosser, Polare Organometalle, Springer, Berlin 1973;
o) T. L. Brown, Pure Appl. Chem. 1970, 23, 447. p) T. L. Brown,
Acc. Chem. Res. 1968, 1, 23.
[15] L. R. Liou, A. J. McNeil, A. Ramirez, G. E. S. Toombes, J. M.
Gruver, D. B. Collum, J. Am. Chem. Soc. 2008, 130, 4859.
[
[
16] S. Bywater, P. Lachance, D. J. Worsfold, J. Phys. Chem. 1975, 79,
148.
2
17] R. D. Thomas, M. T. Clarke, R. M. Jensen, Organometallics
986, 5, 1851.
1
[
[
[
[
18] W. Bauer, C. Griesinger, J. Am. Chem. Soc. 93, 115, 10871.
[
[
3] L. D. McKeever, R. Waack, J. Chem. Soc., Chem. Commun. 1969,
19] Ref. 13c, p 111; ref.13d, p 346.
7
50.
20] Ref. 13c, p 353; ref. 13d, p 296.
4] a) D. Seebach, R. Hässig, J. Gabriel, Helv. Chim. Acta 1983, 66,
21] Ref. 13c, p 413; ref. 13d, p 355.
3
1
08; b) J. Heinzer, J. J. M. Oth, D. Seebach, Helv. Chim. Acta
985, 68, 1848.
[22] H. Schmidbaur, A. Schier, U. Schubert, Chem. Ber. 1983, 116,
938.
1
[
5] L. M. Jackman, L. M. Scarmoutzos, J. Am. Chem. Soc. 1984, 106,
627.
6] a) R. D. Thomas, R. M. Jensen, T. C. Young, Organometallics
987, 6, 565; b) H. Günther, NMR Spectroscopy, 3rd ed.,
4
[23] a) O. Eppers, H. Günther, unpublished; b) H. Günther, in
Recent Developments in Molecular Spectroscopy, (Eds: B.
Jordanov, N. Kirov, P. Simova), World Scientific, Singapur
[
1
1
989, 457.
24] G. Fraenkel, W. R. Winchester, J. Am. Chem. Soc. 1988, 110,
720.
Wiley‐VCH, Weinheim 2013, p 466.
[
[
7] a) G. Fraenkel, A. M. Fraenkel, J. J. Geckle, F. Schloss, J. Am.
Chem. Soc. 1979, 101, 4745; b) G. Fraenkel, M. Henrichs, J. M.
Hewitt, B. M. Su, M. J. Geckle, J. Am. Chem. Soc. 1980, 102,
8
[25] Q. Dufroi, L. Vendier, M. Etienne, Chem. Commun. 2016, 52,
3
345.
6781.

FOX AND GÜNTHER
11 of 11
[
26] G. Barrozino‐Consiglio, G. Hamdoun, C. Fressigné, A.
Harrison‐Marchand, J. Maddaluno, H. Oulyadi, Chem. – Eur.
J. 2017, 23, 12475.
How to cite this article: Fox T, Günther H. NMR
spectroscopy of organolithium compounds, part
XXXIII: trans‐2,3‐dimethylcyclopropyllithium
aggregation in diethylether: An equilibrium
between three different complexes of comparable
energy and the influence of LiBr on aggregate
structure. J Phys Org Chem. 2018;e3800. https://doi.
org/10.1002/poc.3800
[
[
27] Ref. 13d, p 418.
28] G. F. Reynolds, R. E. Dessy, H. H. Jaffé, J. Org. Chem. 1958, 23,
1
217.
[29] S. Singh, R. E. Robertson, Can. J. Chem. 1977, 55, 2587.
[30] P. S. Skell, A. Y. Garner, J. Am. Chem. Soc. 1956, 78, 5430.
[31] D. Seyferth, H. Yamazaki, D. L. Alleston, J. Org. Chem. 1963, 28,
7
03.
[
32] a) A. L. van Geet, Anal. Chem. 1970, 42, 679; b) ref. 13d, p 81 ff.