.
Angewandte
Communications
lithiated intermediate [(Sp)-3]2·Et2O does not feature coordi-
nation of the lithium centers by the chiral additive that had
been used in its synthesis. Thus, despite its role in the
asymmetric ortho-lithiation step, TMCDA apparently does
not play a role for the aggregation of the resulting lithiated
species.
TMCDA afforded an e.r. of 85:15 coupled with nearly
quantitative conversion and a short reaction time (entry 6).
Interestingly, the substitution of Et2O for MTBE under
identical conditions resulted in a decrease of stereoselectivity
(e.r. = 67:33; Table 2, entry 8).
If the reaction mixture was stored for longer periods at
À788C, crystallization of the stereomerically enriched aggre-
gate occurred. This could be proven by separation of the
crystals from the mother liquor. Quenching them afforded the
highly stereomerically enriched product (Rp)-6 (e.r. > 99:1) in
an yield of 68% (entry 7).[19] The combination of sBuLi and
catalytic amounts of TMCDA resulted in a decrease of
conversion and a slightly lower stereoselectivity comparable
to the respective stoichiometric runs described above
(entries 9, 10).
Quantum-chemical calculations were performed to eluci-
date the proposed selectivity and catalytic activity of TMCDA
during the ortho-lithiation cycle. For the abstraction of the
two prochiral ortho-hydrogen atoms, the respective stereo-
differentiating transition states were computed in the pre-
lithiation complex [2·TMCDA·iPrLi].[20] In fact, the deproto-
nation leading to (Sp)-configuration in the lithiated inter-
mediate is favored by 7 kJmolÀ1 in comparison to the other
transition state. This difference in energies corresponds
satisfactorily to the enantiomeric ratios observed in the
products after quenching of metalation runs with stoichio-
metric amounts of TMCDA (e.r. = 86:14).
In the possible prelithiation complex [2·TMCDA·RLi],
iPrLi seems to provide the ideal combination of sterical
demand and reactivity to necessitate a highly stereoselective
ortho-lithiation. In comparison, primary alkyl lithium nBuLi
seems to be too small and tBuLi too bulky to induce the high
selectivities that are observed when using iPrLi.[20]
This prompted us to investigate the stereoselectivity of the
ortho-metalation of 2 using sub-stoichiometric amounts of
TMCDA. While the use of catalytic amounts of chiral lithium
amides in stereoselective deprotonation reactions has
attracted considerable attention,[16] similar examples for
workable deprotonation reactions involving organolithium
reagents with substoichiometric amounts of chiral ligands are
still comparatively scarce:[17] To the best of our knowledge,
only one example of a stereoselective catalytic deprotonation
reaction utilizing alkyl lithium bases has been previously
reported. OꢀBrien and co-workers developed a (À)-sparteine/
bispidine-based asymmetric catalytic lithiation of N-Boc
pyrrolidine or phosphinoboranes, albeit in the presence of
stoichiometric amounts of an achiral auxiliary bispidine,
which supplants sparteine from the lithiated substrate.[18]
Thus, we investigated the catalytic stereoselective ortho-
lithiation of 2 (Table 2). We found that 0.2 equiv TMCDA and
1.3 equiv iPrLi at À788C in a mixture of pentane/Et2O
resulted in a near-quantitative conversion of 2 after 96 h, and
upon quenching with benzophenone, in an e.r. of 81:19. If only
0.05 equiv TMCDAwas used, the e.r. decreased to 71:29 while
the conversion remained nearly quantitative (Table 2,
entries 1 and 3). This indicates that at low temperatures in
Et2O, iPrLi is capable of slowly metalating 2 even in the
absence of the coordinating diamine.
This non-stereoselective background reaction obviously
competes with the stereoselective metalation mediated by
sub-stoichiometric amounts of the chiral auxiliary. Interest-
ingly, the reaction did not proceed quantitatively with
catalytic amounts of TMCDA if no Et2O was added to the
reaction mixture, but only to the mole fraction of the additive
(Table 2, entry 4). Et2O is obviously essential for the promo-
tion of the catalytic activity of TMCDA. Consequently, the
stereoselectivity of the reaction could be improved by low-
ering the amount of Et2O in the reaction solvent: Performing
the lithiation in a 12:1 pentane/Et2O mixture with 0.2 equiv
For TMCDA to be catalytically active, it has to be
released from the enantiomerically enriched ortho-lithiated
species (Sp)-3·TMCDA to be available for the formation of
further pre-lithiation complexes with unreacted starting
material and iPrLi. From complex [2·TMCDA·iPrLi], another
stereo-differentiating lithiation step can occur. The molecular
structure of the lithiated intermediate suggests a possible
mechanism for the release of TMCDA: Obviously, upon
aggregation of the lithioferrocene 3 to a more stable dimer,
a bidentate amine such as TMCDA
is unsuitable to coordinate the
Table 2: Catalytic asymmetric, TMCDA-mediated lithiation of 2 and consecutive quenching with
benzophenone to yield carbinol (Rp)-5.
lithium centers in the dimer owing
to its increased steric demand in
comparison to monodentate ether
donors. In contrast, as soon as the
formation of an etherate or dieth-
erate is considered, the dimeriza-
tion becomes highly favorable.
This releases the chiral catalyst
TMCDA from the lithiated inter-
mediate and its further activity in
the catalytic cycle (Scheme 4).
Entry
RLi
Equiv 4
Solvent
t [h]
T [8C]
% yield[a]
e.r.[b]
1
2
3
4
5
6
7
8
9
10
iPrLi
iPrLi
iPrLi
iPrLi
iPrLi
iPrLi
iPrLi
iPrLi
sBuLi
sBuLi
0.2
0.1
0.05
0.2
0.2
0.2
0.2
0.2
0.2
0.2
pentane/Et2O (1:1)
pentane/Et2O (1:1)
pentane/Et2O (1:1)
pentane
pentane/Et2O (12:1)
pentane/Et2O (12:1)
pentane/Et2O (1:1)
pentane/MTBE (1:1)
cyclohexane/Et2O (1:1)
cyclohexane/Et2O (1:1)
96
96
96
96
96
2
192
2
96
2
À78
À78
À78
À78
À78
À78[e]
À78
À78[e]
À78
À78[e]
86[c]
82[c]
83[c]
17[d]
93[c]
94[c]
68[f]
87[c]
41
81:19
78:22
71:29
82:18
86:14
85:15
>99:1
67:33
80:20
67:33
87
The solvent-mediated propen-
sity of the lithiated substrate for
dimerization constitutes the driv-
ing force behind the release of the
[a] Isolated product. [b] Determined by chiral-phase HPLC. [c] Yield in crude product was more than
95% (determined by NMR spectroscopy). [d] Yield of crude product was about 20%. [e] Intermediary
warming to RT. [f] Obtained upon separation of crystals from mother liquor.
4
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!