On the configurational stability of chiral, nonracemic fluoro- and iodo-[D1]methyllithiums

Article abstract of DOI:10.1002/chem.201304766

Enantiomerically pure fluoro-[D₁]methyllithium and iodo-[D ₁]methyllithiums of up to 92% ee were generated by transmetalation of the corresponding stannanes with MeLi in THF at various temperatures. The intermediate halo-[D₁]methyllithiums were trapped with benzaldehyde or acetophenone already present in excess in the reaction mixture to either give halohydrins or to disintegrate to carbene. The fluoro-[D₁] methyllithiums were found to be microscopically configurationally stable within the tested range of -95 to 0°C, but chemically only stable at temperatures below -95°C due to a rapidly increasing portion disintegrating to carbene. The iodo-[D₁]methyllithiums were configurationally labile relative to the rate of addition to PhCHO at all temperatures tested (-95 to -30°C). Disintegration to carbene interfered as well. The microscopic configurational stability of enantiomerically pure fluoro-[D₁]methyllithiums prepared by tin-lithium exchange in the presence of excess benzaldehyde or acetophenone has been investigated. Depending on the reaction temperature, a portion of the generated fluoro-[D₁]methyllithiums was added to the electrophiles to give fluorohydrins; the remaining portions disintegrated to carbene and LiF (see scheme).

Full text of DOI:10.1002/chem.201304766

DOI: 10.1002/chem.201304766
Full Paper
&
Configurational Stability
On the Configurational Stability of Chiral, Nonracemic Fluoro- and
Iodo-[D1]Methyllithiums
Dagmar C. Kail,[b] Petra Malova Krizkova,[a] Anna Wieczorek,[c] and
Friedrich Hammerschmidt*[a]
Abstract: Enantiomerically pure fluoro-[D1]methyllithium
and iodo-[D1]methyllithiums of up to 92% ee were generat-
ed by transmetalation of the corresponding stannanes with
MeLi in THF at various temperatures. The intermediate halo-
[D1]methyllithiums were trapped with benzaldehyde or ace-
tophenone already present in excess in the reaction mixture
to either give halohydrins or to disintegrate to carbene. The
fluoro-[D1]methyllithiums were found to be microscopically
configurationally stable within the tested range of À95 to
08C, but chemically only stable at temperatures below
À958C due to a rapidly increasing portion disintegrating to
carbene. The iodo-[D1]methyllithiums were configurationally
labile relative to the rate of addition to PhCHO at all temper-
atures tested (À95 to À308C). Disintegration to carbene in-
terfered as well.
be generated by bromine–lithium exchange.[7] Unlabeled fluo-
romethyllithium was claimed[8,9] not to be detectable by NMR
spectroscopy at À1208C, but to give ethene, without provid-
ing experimental evidence. ICH2M (M=Li[10], ZnX[11]) were pre-
pared preferably from diiodomethane and used for syntheses.
(Iodomethyl)triphenyltin was found to furnish triphenylstannyl-
methyllithium when treated with nBuLi at À508C in Et2O and
a mixture of triphenylstannylmethyl- and iodomethyllithium in
THF.[12] Evidently, configurationally stable 1-iodoalkyllithiums
and the corresponding Grignard reagents were prepared by
Hoffmann’s group.[13] The structures of unlabeled carbenoids 1,
their aggregation, solvation, and addition[8,14] to alkenes have
been calculated extensively, especially for FCH2Li.[8] Boche et al.
Introduction
Chiral, nonracemic a-heteroatom-substituted organolithiums
are valuable reagents for synthesis.[1] Their preparation, reac-
tions, the determination of their configurational stability and
mechanisms of enantiomerization, as well as their aggregation
states have attracted much attention since their first observa-
tion[2] by Stille and Sreekumar. By far the most important are
organolithiums a-substituted with oxygen or nitrogen. We
have accessed homochiral oxygen-[3] [D1]1a, sulfur-[4] [D1]1b
and nitrogen substituted[5] [D1]methyllithiums [D1]1c and eval-
uated their micro- and macroscopic configurational stability
(Figure 1). Chloro-[6] and bromo-[D1]methyllithiums,[4] ([D1]1d)
and ([D1]1e), were found to be chemically very labile at
À788C, but configurationally stable.
À
found that the calculated barrier to planarization for FCH2
(13.6 kcalmolÀ1) which reflects configurational stability was
highest from all heteroatom-substituted carbanions (N: 11.0; O
11.0, S: 2.8; H: 3 kcalmolÀ1).[15] This finding could so far not be
corroborated experimentally because neither chiral fluoro-
methyl- nor an a-fluoroalkyllithium could be generated.
Herein, we present our results for chiral fluoro- ([D1]1 f) and
iodo-[D1]methyllithiums ([D1]1g). While a-fluoroalkyllithiums
are unknown compounds, a-fluorocyclopropyllithiums could
Results and Discussion
To access chiral fluoro-[D1]methyllithiums we decided to pre-
pare them from (R)- and (S)-tributylstannyl-[D1]methanol[3,4] of
99% ee, by tin–lithium exchange. The reactions were opti-
mized in the unlabeled series first. Reaction of tributylstannyl-
methanol (2)[16] with bis(2-methoxyethyl)aminosulfur trifluoride
(Deoxo-Fluorꢀ)[17] furnished fluoromethylstannane 3 in 68%
yield (Scheme 1).
Figure 1. Chiral heteroatom-substituted [D1]methyllithiums.
[a] P. Malova Krizkova, Dr. F. Hammerschmidt
Institute of Organic Chemistry, University of Vienna
Wꢀhringerstrasse 38, 1090, Wien (Austria)
E-mail: friedrich.hammerschmidt@univie.ac.at
[b] Dr. D. C. Kail
It was treated with MeLi (4 equiv) at À788C in THF to give
fluoromethyllithium which was trapped in situ with benzalde-
hyde (4 equiv) to yield fluorohydrin (Æ)-4 and 1-phenylethanol
[(Æ)-7a] derived from benzaldehyde by addition of MeLi
PharmBioTec GmbH, Saarland University
Campus C2.2, Room 31, 66123 Saarbrꢁcken (Germany)
[c] Dr. A. Wieczorek
Landon IP Ltd., 16 High Holborn, London WC1V 6BX (UK)
Chem. Eur. J. 2014, 20, 4086 – 4091
4086
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 1. Transmetalation of 3 in presence of benzaldehyde (4 equiv) in
THF at À788C.
Entry
Fluorohydrin (Æ)-4[a]
Stannane 3[a]
Decomposition[a,b]
1
2
3
4
1.0 (38)
1.0 (43)
1.0 (32)
1.0 (17)
0.6
–
0.7
–
1.0
1.3
1.4
5.0
Scheme 1. Preparation of fluoromethylstannanes 3 from stannylmethanols 2.
[a] The ratios of fluorohydrin (Æ)-4/starting fluoromethylstannane 3 and
decomposition were determined by 1H NMR spectroscopy of the crude
product [(Æ)-4 was standardized to 1.0]; yields are given in parenthesis.
[b] Calculated from difference between MeSnBu3 and product 4 formed.
ated by converting them with lithium (R)-1-(2-naphthyl)ethane-
thiolate[6] to diastereomeric sulfides, whose ratios were deter-
1
mined by H NMR spectroscopy. Similarly to the experiments in
the unlabeled series, MeLi was added to a mixture of fluoro-
methylstannane (R)-[D1]3 and benzaldehyde in THF at À788C
1
without and at À708C with 12-crown-4. The H NMR spectrum
of the crude product obtained by the reaction performed at
À788C indicated that only half of the starting stannane was
transmetalated and that only 30% of the fluoro-
[D1]methyllithium formed was added to benzaldehyde to give
fluorohydrin (2S)-[2-D1]4 (Table 2, entry 1).
In the case of the experiment with [12]crown-4 as ligand, all
the starting material was transmetalated and only 20% of the
generated FCHDLi was added to benzaldehyde (Table 2,
entry 2). For the sake of clarity, it is more convenient here for
the deuterated fluoro- and later for the deuterated iodohydrins
as well to give the ee value for each chiral center individually.
Scheme 2. Conversion of fluoromethylstannanes 3 to fluoromethyllithiums
1 f and their in situ trapping with benzaldehyde/ketones.
(Scheme 2). The composition of the crude product as deter-
1
mined by H NMR (400 MHz) is given in Table 1, entry 1.
Only part of the fluoromethyllithium generated in situ was
converted to fluorohydrin (Æ)-4. Evidently, the re-
maining portion decomposed to methylene. The
yield could not be improved significantly by adding
Table 2. Molar ratios of starting stannanes 3, fluorohydrins 4/5/6, and MeSnBu3 in
crude products for conversion of 3 to 4/5/6 as determined by 1H NMR spectroscopy
(400 MHz).
lithium-complexing 12-crown-4 as ligand (Table 1,
entry 2) or by replacing MeLi by MeLi·LiBr (Table 1,
entry 3). When tin–lithium-exchange was performed
with nBuLi/12-crown-4 (Table 1, entry 4) the yield was
only 17% with a high degree of decomposition
(83%), as no residual starting material could be de-
tected. In the case of adding benzaldehyde after tin–
lithium exchange and aging of the intermediate fluo-
romethyllithium for 30 sec no fluorohydrin was
formed. In all cases fluorohydrin 4 and 1-phenyletha-
nol were isolated as a mixture inseparable by flash
chromatography. In summary, our initial experiments
proved that fluoromethyllithium can be generated
and intercepted with benzaldehyde, despite being
chemically very labile.
Entry
Electrophile
T [8C] Stannane 3 Fluorohydrin 4/5/6 MeSnBu3
(R)-[D1]3
1.1
(R)-[D1]3
0
(2S)-[2-D1]4
0.3
(2S)-[2-D1]4
0.2
1
2
À78
À70
1.0
1.0
benzaldehyde
3
1.2
(S)-[D1]3
4.0
5
1.0
(1R)-[1-D1]5
0.6
3
4
À95
À30
1.0
1.0
acetophenone
3
0.5
3
6
1.0
6
5
6
À95
À78
1.0
1.0
tbutylphenylketone
0.9
0.7
In the labeled series, enantiomerically pure stannyl-
methanols (R)- and (S)-[D1]2[3a,4] were converted to
(S)- and (R)-[D1]3 in similar yields (73%) of 98% ee
consistently, when they were first reacted with
Et2NTMS[18] and then with Deoxo-Fluorꢀ. However,
(R)-[D1]3
0.2
(1S)-[1-D1]5
7
8
9
À78
À30
0
1.0
1.0
1.0
0.4
(R)-[D1]3
0.7
(1S)-[1-D1]5
acetophenone
0.3
omission
of
silylation
gave
fluoro-
(R)-[D1]3
0.1
(1S)-[1-D1]5
[D1]methylstannanes with ee values of 88–98%. The
ee values of the chiral, nonracemic [D1]3 were evalu-
0.1
Chem. Eur. J. 2014, 20, 4086 – 4091
www.chemeurj.org
4087
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
As the enantioselectivity of the addition of the chiral carbanion
(S)-[D1]1 f to benzaldehyde is zero, (R)- and (S)-configuration at
C-1 were always formed in equal amounts, resulting in a prod-
uct with zero “ee” at C-1, that is, racemic at C-1. The yields of
labeled fluorohydrins (2S)-[2-D1]4 of “ee” >95% at C-2 were
10% and 20%, respectively. The ee value was determined by
derivatization with (S)-Mosher chloride [(S)-MTPACl] and
1H NMR spectroscopy (400 MHz). Similarly, reference samples of
fluorohydrins[19] (Æ)-4 and (R)-4 (98% ee) were derivatized
(Figure 2).
fluoromethylstannane 3/fluorohydrin (Æ)-5/MeSnBu3 are col-
lected in Table 2. In all four cases (Table 2, entries 3–6) some
fluoromethylstannane 3 was recovered. Furthermore, fluoro-
methyllithium is chemically stable at temperatures ꢀÀ958C, as
the fluorohydrin and tributylmethylstannane are formed in
equal molar amounts. However, at À308C already 40% of the
fluoromethyllithium generated decomposed (ratio of fluorohy-
drin/tributylmethylstannane 0.6:1.0). These data show qualita-
tively that decomposition of fluoromethyllithium increases
with increasing temperature as expected. Unlabeled (Æ)-5
could be esterified with (S)-MTPACl under forcing conditions
(DMAP, 508C, 9 h, 1,4-dioxane) with partial resolution of the
fluorohydrin and in low yield, both of which did not influence
the determination of the ee value in the labeled series later on.
At À958C, all the fluoromethyllithium formed by transmeta-
lation was trapped by tert-butyl phenyl ketone. At À788C how-
ever, the decomposition (to carbene) already interfered with
addition to ketone. FCH2Li was only partially added to the
ketone, 31% decomposed to methylene. Surprisingly, these
data are similar to those for acetophenone, even though the
tBu group is sterically more demanding than the methyl
group. As this sterically hindered alcohol could not be derivat-
ized with (S)-MTPACl to test whether the Mosher ester could
be used in the labeled series to determine the ee value, the re-
maining experiments were performed with acetophenone as
electrophile at À78, À30, and 08C (Table 2, entries 7–9;
Figure 3).
FCHDLi was chemically unstable already at À788C and disin-
tegration increased from 60% at À788C to 90% at 08C. The
yield was very low at 08C, but all three fluorohydrins (1S)-[1-
D1]5 had 98% ee as determined by derivatization with (S)-
MTPACl and 1H NMR spectroscopy (Figure 3). These experi-
ments prove that fluoromethyllithium is microscopically config-
urationally stable at all temperatures tested (À95 to 08C), even
1
Figure 2. Signals of CH2F and CHDF groups in the H {18F} NMR spectra
(600 MHz, CDCl3) of (R)-Mosher esters derived from A) (R)-4 of 98% ee; B) (S)-
4 of 85% ee; C) mixture of (1S,2S)-[2-D1]4·MTPA-(R)/(1R,2S)-[2-D1]4·MTPA-
(R)=5:3 (enrichment of one diastereomer by PTLC); ee value (at C-2) >95%.
Table 3. Molar ratios of starting stannanes 8, iodohydrins 10, and
MeSnBu3 in crude products for conversion of 8 to 10 as determined by
1H NMR spectroscopy (400 MHz).
Fluoro-[D1]methyllithium seems to be microscopically config-
urationally stable for the short life time between generation
and addition to benzaldehyde, but chemically very labile.
Transmetalation follows a retentive course.[2,20] We assume that
the same is true for the addition of FCHDLi to benzaldehyde. It
was found by Reich and Phillips[21] and others[22] that tin ate
complexes form in quantities detectable by NMR spectroscopy
only under special conditions, e.g., presence of HMPA as addi-
tive in THF, increasing number of phenyl substituents at tin,
and low temperature (À808C). Furthermore, tributyltin com-
pounds do not favor the formation of ate complexes, which
are substantially less reactive[21] than lithium reagents. These
facts indicate that salt free species or aggregates of heteroa-
tom-substituted methyllithiums are produced, neglecting the
LiF formed by decomposition of FCH2Li.
Entry
1
T [8C]
À78
Stannane 8
Iodohydrin 10
MeSnBu3
1.0
8
14.4
10
0.7
8
5.0
10
0.45
2
3
4
5
6
7
À50
À78
À50
À95
À50
À30
1.0
1.0
1.0
1.0
1.0
1.0
(S)-[D1]8
0.0
(2S)-[2-D1]10
0.65
(S)-[D1]8
1.3
(2S)-[2-D1]10
0.44
(R)-[D1]8
3.5
(2R)-[2-D1]10
0.72
To influence the ratio of the addition of MeLi to different
electrophiles relative to transmetalation of the stannane, ben-
zaldehyde was replaced by acetophenone (reaction conditions:
stannane 3 (1 equiv), acetophenone (2 equiv), temperatures of
À95 and À308C, one drop of MeLi (2 equiv, 1m) per 5 sec) and
tert-butyl phenyl ketone (À95 and À788C). The molar ratios of
(R)-[D1]8
1.4
(2R)-[2-D1]10
0.56
(R)-[D1]8
0.1
(2R)-[2-D1]10
0.1
Chem. Eur. J. 2014, 20, 4086 – 4091
www.chemeurj.org
4088
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 4. Preparation of iodomethylstannanes 8, their conversion to iodo-
methyllithiums 1 f and in situ trapping with benzaldehyde.
stannane could be detected. This experiment indicated that
1) tin–lithium exchange is very rapid and 2) the tributylmethyl-
stannane (9) was formed by exchange of the iodomethyl for
the Me substitutent and not by quenching of intermediate
tributylstannylmethyllithium with CF3CO2D, formed by lithium–
iodine exchange. Additionally, racemic iodohydrin 10[23] was
mixed with MeLi (1.5 equiv) at À788C and quenched with
CF3CO2H after 5 min. The isolated product was identical
(1H NMR, TLC) to the starting iodohydrin and did not contain
phenyloxirane [(Æ)-11], corroborated by spiking with authentic
phenyloxirane. Clearly, the lithium alcoholate formed from the
iodohydrin, was stable toward ring closure, normally requiring
warming up to room temperature.[13a,24] As such, isolation of
the iodohydrin formed by addition of iodomethyllithium to
benzaldehyde should not pose a problem in later experiments.
Stannylmethanol 2 was converted to iodomethylstannane 8 in
89% yield, using Ph3P/NIS[25] (Scheme 4).
1
Figure 3. Signals of CH2F and CHDF groups in the H NMR spectra (400 MHz
in case of A, 600 MHz in case of B, C and D, CDCl3) of (R)-Mosher esters de-
rived from A) (Æ)-5, B) (1R)-[1-D1]5 of 90% ee, 95% D1; C) zoom of spectrum
shown in B to show the peaks of the unlabeled species marked with “o” and
those of the two (1S)-[1-D1]diastereomers marked with “x”; D) (1S)-[1-D1]5 of
98% ee.
more so than the diisopropylcarbamoyloxy-substituted methyl-
lithium.[3] It is therefore the most configurationally stable
heteroatom-substituted methyllithium.
The preparation of chiral, nonracemic iodo-[D1]methyl-
lithiums and the determination of their configurational stability
turned out to be experimentally very challenging as in the
case of fluoro-[D1]methyllithiums because of low reproducibili-
ty attributed to the chemical lability of the species involved.
Two preliminary experiments were performed before optimiz-
ing the experiments in the unlabeled series as before
(Scheme 3).
MeLi (4 equiv) was slowly added dropwise (one drop per
one second) to a stirring mixture of iodomethylstannane
(1 equiv) and excess benzaldehyde (4 equiv) in THF at À788C.
After 5 min a solution of CF3CO2H in CH2Cl2 (2m) and water
(5 mL) were added. The crude product (for ratio of compo-
nents see Table 3, entry 1) was purified by flash chromatogra-
phy to yield iodohydrin 10 (15%) always in admixture
(ca. 50%) with tin containing compounds of unknown struc-
ture, which were once removed by preparative thin layer chro-
matography in the labeled series to get an analytical sample of
homogenous iodohydrin.
To test transmetalation of iodomethylstannane (for prepara-
tion see Scheme 4), MeLi was added to a solution of 8 in THF
1
at À788C, followed by CF3CO2D 30 sec later. The H NMR spec-
trum of the crude product showed that it was a mixture of
tributylmethylstannane (50%) containing no deuterium and
two stannanes of unknown structure. No starting iodomethyl-
The mixture was routinely used for the preparation of the
(R)-Mosher esters, from which the tin-containing impurities
could by separated easily by flash chromatography. When the
reaction was performed at À508C, the yield was again 15%, al-
though the products and starting materials had different ratios
(Table 3, entry 2). These two experiments demonstrate that io-
domethyllithium is formed as an intermediate by Sn–Li ex-
change and that it was trapped by benzaldehyde to yield iodo-
hydrin 10. As expected, the competing addition of MeLi to
benzaldehyde was predominant and part of the iodomethyl-
Scheme 3. Testing of transmetalation of stannane 8 and chemical stability of
lithium alkoxide derived from 10.
Chem. Eur. J. 2014, 20, 4086 – 4091
www.chemeurj.org
4089
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
lithium decomposed in a temperature dependent proportion.
At À788C, 70% of the iodomethyllithium ended up in the io-
dohydrin, however at À508C only 45%.
We anticipated that chiral iodo-[D1]methylstannanes were
prone to substitution with excess iodide in the reaction mix-
ture, which could lead to partial racemization. To avoid this,
short reaction times at temperatures as low as possible for its
preparation from chiral stannyl-[D1]methanol and purification
were aimed for. The optimized Mitsunobu reaction[26] following
an SN2 mechanism was performed with tris(4-chlorophenyl)-
phosphine/diisopropyl azodicarboxylate/HI[27] (solution in
CH2Cl2) in a mixture of toluene/CH2Cl2 at À788C for 20 min.
The reaction was quenched with a few drops of dry methanol
and immediately applied to a silica gel column kept in dim
light, cooled to À788C in the upper part, and using an eluent
also cooled to À788C. Iodomethylstannanes (R)- and (S)-[D1]8
were obtained from the corresponding stannylmethanols (R)-
and (S)-[D1]2 in yields of 64% and 55%, respectively (iodine
has priority over tin in CIP nomenclature). Their enantiomeric
excesses determined by combining reaction with lithium (R)-1-
(2-naphthyl)ethanethiolate[6] and 1H NMR spectroscopy, were
86 and 92%, respectively. These iodo-[D1]methylstannanes
1
Figure 4. Signals of CH2I and CHDI groups in the H NMR spectra (400 MHz,
toluene-D6) of (R)-Mosher esters derived from A) (Æ)-10; B) (2S)-[2-D1]10 of
49% ee (at C-2); C) (2R)-[2-D1]10 of 61% ee (at C-2). Signals of B) and C)
broadened by deuterium coupling.
increasing temperature, as judged from the ratio of tributyl-
methylstannane and iodohydrin.
were
transmetalated
and
the
intermediate
iodo-
[D1]methyllithiums were trapped at various temperatures with
benzaldehyde already present in the reaction mixture. The
ratios of products in the crude reaction mixture of 5 experi-
ments with chiral, nonracemic ICHDLi are compiled in Table 3
(entries 3–7). As can be seen easily from entries 1–5 in Table 4,
the yields were low and part of the starting material was not
consumed, but the deuterated iodohydrins were not racemic
Conclusion
In conclusion, we present our findings on the microscopic con-
figurational stability (configurational stability relative to the ad-
dition to benzaldehyde or acetophenone) of enantiomerically
pure fluoro-[D1]methyllithiums prepared by tin–lithium ex-
change in the presence of excess benzaldehyde or acetophe-
none. Only part of the fluoromethylstannane was transmetalat-
ed as addition of MeLi to the electrophiles interfered. Depend-
ing on the reaction temperature, a portion of the generated
fluoro-[D1]methyllithiums was added to the electrophiles to
give fluorohydrins, the remaining portions disintegrated to car-
bene and LiF. Fluoro-[D1]methyllithiums were found to be con-
figurationally stable up to 08C relative to the addition to ace-
tophenone. At that temperature, carbene formation was heavi-
ly interfering, which could be avoided when keeping the tem-
perature below À958C. Iodo-[D1]methyllithiums prepared simi-
larly from tributyl-(iodo-[D1]methyl)stannanes of up to 92% ee
were found to be configurationally labile relative to the addi-
tion to benzaldehyde at all temperatures investigated (À958C
to À308C). The rate of disintegration of ICHDLi to deuterated
carbene and LiI increased significantly with increasing temper-
ature.
1
at C-2 (but at C-1) as determined by H NMR spectroscopy of
their (R)-Mosher esters.
Table 4. Generation and trapping of chiral, nonracemic iodo-
[D1]methyllithiums (R)- and (S)-[D1]1g with PhCHO in THF to give iodohy-
drins [D1]10.
Entry
[D1]1g
Conf./ee [%]
Temp.
[8C]
[D1]10
Conf./ee [%]
Yield
[%][a]
1
2
3
4
5
(S)/92
(S)/92
(R)/84
(R)/84
(R)/84
À78
À50
À95
À50
À30
(S)/49
(S)/33
(R)/61
(R)/33
(R)/28
55
27
22
22
9
[a] Mixture of iodohydrin and tin-containing impurity (1:1).
The “ee” at C-2 of the five halohydrins decreased with in-
creasing the temperature (Table 4). Surprisingly, the “ee” was
still 28% at À308C starting from an iodo-[D1]methylstannane
of 84% ee (Table 4, entry 5). Despite the short lifetime of the
iodo-[D1]methyllithium, part of the molecules enantiomerized
(Figure 4). Therefore, they are microscopically configurationally
labile, contrary to the fluoro-, chloro-, and bromomethyllithi-
ums. We assume that addition of these species to benzalde-
hyde follows a retentive course. The rate of desintegration of
iodomethyllithiums, which are also carbenoids, increased with
Experimental Section
General procedure for the generation and addition of fluo-
romethyl- (1 f) and (S)-fluoro-[D1]methyllithium {(S)-[D1]1f}
to benzaldehyde - (Æ)-2-fluoro-1-phenylethanol [(Æ)-(4)] and
(1R/S,2S)-2-fluoro-1-phenyl-[1-D1]ethanol {(2S)-[2-D1]4}
A solution of tributyl(fluoromethyl)stannane 3 (97 mg, 0.30 mmol),
(for entries 2 and 4: a stoichiometric amount of [12]crown-4 rela-
Chem. Eur. J. 2014, 20, 4086 – 4091
www.chemeurj.org
4090
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
schmidt, Chem. Eur. J. 2007, 13, 9582–9588; d) D. C. Kapeller, F. Ham-
merschmidt, Tetrahedron 2010, 66, 591–598.
[4] A. Wieczorek, F. Hammerschmidt, J. Org. Chem. 2012, 77, 10021–10034.
[5] D. C. Kapeller, F. Hammerschmidt, Chem. Eur. J. 2009, 15, 5729–5739.
[6] D. C. Kapeller, F. Hammerschmidt, J. Am. Chem. Soc. 2008, 130, 2329–
2335.
[7] a) C. Mꢂller, F. Stier, P. Weyerstahl, Chem. Ber. 1977, 110, 124–137; b) T.
Ishihara, K. Hayashi, T. Ando, H. Yamanaka, J. Org. Chem. 1975, 40,
3264–3267.
[8] For reviews see: a) G. Boche, J. C. W. Lohrenz, Chem. Rev. 2001, 101,
697–756; b) V. Capriati, S. Florio, Chem. Eur. J. 2010, 16, 4152–4162.
[9] a) See Ref. [120] in Ref. [8]; b) A. Opel, Dissertation, Universitꢃt Marburg,
1993. We could not find the experimental evidence that the author ac-
tually tried to prepare FCH2Li there.
[10] a) U. Burger, R. Huisgen, Tetrahedron Lett. 1970, 11, 3049–3051; b) G.
Wittig, F. Wingler, Chem. Ber. 1964, 97, 2139–2145; c) V. Schulze, P. G.
Nell, A. Burton, R. W. Hoffmann, J. Org. Chem. 2003, 68, 4546–4548;
d) R. Soundararajan, G. Li, H. C. Brown, Tetrahedron Lett. 1994, 35, 8957–
8960.
[11] a) Review: H. Lebel, J.-F. Marcoux, C. Molinaro, A. B. Charette, Chem. Rev.
2003, 103, 977–1050; b) R. G. Cornwall, O. A. Wong, H. Du, T. A. Ramirez,
Y. Shi, Org. Biomol. Chem. 2012, 10, 5498–5513.
[12] T. Kauffmann, R. Kriegesmann, B. Altepeter, F. Steinseifer, Chem. Ber.
1982, 115, 1810–1817.
[13] a) H. C. Stiasny, R. W. Hoffmann, Chem. Eur. J. 1995, 1, 619–624; b) R. W.
Hoffmann, Chem. Soc. Rev. 2003, 32, 225–230.
tive to MeLi or nBuLi), and freshly distilled benzaldehyde (127 mg,
121 mL, 1.2 mmol) in dry THF (1.5 mL) was cooled to the respective
temperature under argon. MeLi [Entry 1, 0.75 mL, 1.6m solution in
Et2O, 1.20 mmol; entry 3: 4 equivalents MeLi·LiBr (1.5m, in Et2O) rel-
ative to fluoromethylstannane instead of MeLi; entry 4: 4 equiv of
nBuLi (1.6m in hexanes)] was slowly added drop by drop. The mix-
ture was stirred for 10 min at bath temperature before addition of
water (3 mL). The organic phase was separated, the aqueous one
extracted with Et2O, and the combined organic layers were dried
(MgSO4). After removal of the solvent under reduced pressure, the
crude product was in some cases purified by flash chromatography
(CH2Cl2, Rf =0.24). The other experiments of Table 1 were per-
formed with similar quantities of 3 analogously, except for the
changes given in the Table. The ratios of compounds in the crude
products were evaluated by 1H NMR spectroscopy and the yields
calculated accordingly (Table 1). As 1-phenylethanol (major portion)
and (relatively volatile) fluorohydrin 4 had the same Rf, both were
isolated as a mixture. Signals of labeled or unlabeled5 fluorohydrin
1
4 could be seen in the H NMR spectrum.
Similarly, fluoro-[D1]methylstannane (R)-[D1]3 (0.105 g, 0.32 mmol)
gave
a crude mixture of (R)-[D1]3, fluorohydrin (2S)-[2-D1]4,
MeSnBu3, benzaldehyde, and 1-phenylethanol {ratio in Table 2,
Entry 1; 0.230 g, calculated yield of (2S)-[2-D1]4: 10%} at À788C.
Similarly, fluoro-[D1]methylstannane (R)-[D1]3 (0.105 g, 0.32 mmol)
[14] a) Z. Ke, C. Zhao, D. L. Phillips, J. Org. Chem. 2007, 72, 848–860; b) L. M.
Pratt, P. T. T. Tran, N. V. Nguyen, B. Ramachandran, Bull. Chem. Soc. Jpn.
2009, 82, 1107–1125; c) J. Mareda, N. G. Rondan, K. N. Houk, T. Clark,
P. V. R. Schleyer, J. Am. Chem. Soc. 1983, 105, 6997–6999.
[15] G. Boche, A. Opel, M. Marsch, K. Harms, F. Hallera, J. C. W. Lohrenz, C.
Thꢂmmler, W. Koch, Chem. Ber. 1992, 125, 2265–2273.
gave
a crude mixture of (R)-[D1]3, fluorohydrin [(2S)-[2-D1]4,
MeSnBu3, benzaldehyde, and 1-phenylethanol {ratio in Table 2,
Entry 2; 0.210 g, calculated yield of (2S)-[2-D1]4: 20%} at À708C,
and in presence of [12]crown-4 (0.226 g, 1.28 mmol).
[16] Lithium tributylstannylmethoxide[3a] generated from Bu6Sn2/nBuLi/
(CH2O)n was neutralized with 2m CF3CO2H in dry THF and worked up
extractively. Flash chromatography [hexanes/EtOAc=7:1, Rf =0.36] fur-
nished 2 (50–65%).
Full experimental procedures and spectroscopic data, copies of
NMR spectra.
[17] N. Al- Maharik, D. O’Hagan, Aldrichimica Acta 2011, 44, 65–75.
[18] M. M. Bio, M. Waters, G. Javadi, Z. J. Song, F. Zhang, D. Thomas, Synthe-
sis 2008, 891–896.
Acknowledgements
This work was supported by the Austrian Science Fund (FWF)
by Grant No. P19869-N19. We thank S. Felsinger for recording
NMR spectra and J. Theiner for performing combustion analy-
ses.
[19] Y. Akiyama, T. Fukuhara, S. Hara, Synlett 2003, 1530–1533.
[20] a) Perspective: H. J. Reich, J. Org. Chem. 2012, 77, 5471–5491; b) J. S.
Sawyer, A. Kucerovy, T. L. Macdonald, G. J. McGarvey, J. Am. Chem. Soc.
1988, 110, 842–853; c) H. J. Reich, J. P. Borst, M. B. Coplien, N. H. Phillips,
J. Am. Chem. Soc. 1992, 114, 6577–6579; d) F. Hammerschmidt, A. Han-
ninger, H. Vçllenkle, Chem. Eur. J. 1997, 3, 1728–1732.
[21] H. J. Reich, N. H. Phillips, J. Am. Chem. Soc. 1986, 108, 2102–2103.
[22] J. S. Sawyer, T. L. Macdonald, G. J. McGarvey, J. Am. Chem. Soc. 1984,
106, 3376–3377.
[23] M. Smietana, V. Gouverneur, C. Mioskowski, Tetrahedron Lett. 2000, 41,
193–195.
Keywords: chemical
enantioselectivity
transmetalation
stability
·
configuration
compounds
·
·
·
organolithium
[24] A. K. Yudin, A. Caiazzo, Science of Synthesis 2008, 37, 407–432.
[25] A. K. Bose, B. Lal, Tetrahedron Lett. 1973, 14, 3937–3940.
[26] K. C. Kumara Swamy, N. N. Bhuvan Kumar, E. Balaraman, K. V. P. Pavan
Kumar, Chem. Rev. 2009, 109, 2551–2651.
[27] F. Huber, M. Schmeißer in Handbuch der Prꢀparativen Anorganischen
Chemie, Vol. 1, 3rd. ed. (Ed.: G. Brauer), Ferdinand Enke Verlag, Stuttgart,
1975, p. 300.
[1] For reviews, see: a) Tetrahedron Organic Chemistry Series, Vol. 23, Orga-
nolithiums: Selectivity for Synthesis (Ed.: J. Clayden), Pergamon, Amster-
dam, 2002; b) Topics in Organometallic Chemistry: Organolithiums in
Enantioselective Synthesis (Ed.: D. M. Hodgson), Springer, Berlin, 2003;
c) Topics in Stereochemistry: Stereochemical Aspects of Organolithium
Compounds (Ed.: R. E. Gawley), Wiley-VCH, Weinheim, 2010.
[2] W. C. Still, C. Sreekumar, J. Am. Chem. Soc. 1980, 102, 1201–1202.
[3] a) D. Kapeller, R. Barth, K. Mereiter, F, Hammerschmidt, J. Am. Chem.
Soc. 2007, 129, 914–923; b) D. C. Kapeller, F, Hammerschmidt, J. Org.
Chem. 2009, 74, 2380–2388; c) D. C. Kapeller, L. Brecker, F. Hammer-
Received: December 5, 2013
Published online on February 27, 2014
Chem. Eur. J. 2014, 20, 4086 – 4091
www.chemeurj.org
4091
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim