Application of rapidly generated bidentate ligand libraries to zinc catalyzed reductions

Article abstract of DOI:10.1016/j.jorganchem.2012.06.013

A methodology for the combinatorial synthesis of bidentate ligands - allowing direct screening of reaction products without the need for isolation or purification - has been employed in a zinc catalyzed hydrosilylation. This reaction allowed the robustness of the methodology to be examined, by employing it in a challenging case where the metal complex is not pre-formed prior to catalysis. Four different ligand families have been examined: imines, aminals, bis-imines and oxazolines and related compounds, with a small library of each type produced and directly screened in the reaction. Three ligands providing enantioselectivities of 50% or more in this very challenging reaction were identified, and ees and conversions were equivalent whether the ligand was obtained as a crude mixture from a library synthesis or as an isolated, purified compound.

Full text of DOI:10.1016/j.jorganchem.2012.06.013

Journal of Organometallic Chemistry 716 (2012) 159e166
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Application of rapidly generated bidentate ligand libraries to zinc catalyzed
reductions
,
,
Tariq Zaman a b, Robin Frauenlob a, Robert McCarthy a, Carolyn M. Walsh a, Enda Bergin a
*
a School of Chemistry, Trinity Biomedical Sciences Institute, University of Dublin, Trinity College, Dublin 2, Ireland
b School of Chemical & Materials Engineering, National University of Sciences & Technology, 44000 Islamabad, Pakistan
a r t i c l e i n f o
a b s t r a c t
Article history:
A methodology for the combinatorial synthesis of bidentate ligands e allowing direct screening of
reaction products without the need for isolation or purification e has been employed in a zinc catalyzed
hydrosilylation. This reaction allowed the robustness of the methodology to be examined, by employing
it in a challenging case where the metal complex is not pre-formed prior to catalysis. Four different
ligand families have been examined: imines, aminals, bis-imines and oxazolines and related compounds,
with a small library of each type produced and directly screened in the reaction. Three ligands providing
enantioselectivities of 50% or more in this very challenging reaction were identified, and ees and
conversions were equivalent whether the ligand was obtained as a crude mixture from a library
synthesis or as an isolated, purified compound.
Received 6 March 2012
Received in revised form
14 June 2012
Accepted 20 June 2012
Keywords:
Combinatorial catalysis
Asymmetric catalysis
Hydrosilylation
Ó 2012 Elsevier B.V. All rights reserved.
Ligand libraries
1. Introduction
chiral ketones [5]: not only is this an extremely attractive reaction
(as has been pointed out [5a] due to the inexpensive nature of
The development of combinatorial catalysis is driven by the
need to rapidly identify suitable ligands for a given reaction,
coupled with the lack of theoretical methods for predicating what
will constitute a successful ligand in advance. As such by necessity
a great deal of methodology is driven by trial and error. The inef-
ficient nature of such an approach has led to a number of groups
developing ways to accelerate and simplify this process, such as
supramolecular-guided ligand formation [1], solid phase synthesis
[2] and direct screening of “instant ligand libraries” [3].
We recently communicated initial results for a methodology for
the synthesis of chiral bidentate ligands in a highly modular
fashion, without the need for column chromatography or other
purification methods [4]. The ligands were coupled together by the
addition of two fragments, allowing the generation of imines,
aminals, oxazolidines and others in the reaction vessel at the
appropriate loading without the need for isolation.
PMHS and Zn(II) the cost of this asymmetric process is comparable
to a simple racemic reduction of ketones) it is also highly chal-
lenging, with very few ligands capable of delivering high
enantioselectivities.
2. Results and discussion
Initially we compared the results obtained with our in-situ
system to that obtained with the traditionally purified and iso-
lated ligand. Our procedure consisted of combining the amine and
aldehyde together and heating overnight, before adding the
remaining reagents. Due to the fact that no prior complex forma-
tion was needed, and the fact that the ligand formation could be
carried out in the same solvent as the reaction (toluene), the
amount of manipulation was minimal.
Even under these more direct conditions we found that the
results of in-situ vs. pre-prepared and purified ligands were
comparable. We set up conditions to test our in-situ conditions
against literature values (Table 1). Ligand 1 has previously been
reported in the literature, giving moderate ee with excellent
conversion (entry 1). We synthesised this ligand and purified by
column chromatography, and achieved similar results to those
reported (entry 2). We then repeated the reaction with our in-situ
system, by simply pre-mixing equimolar, microlitre quantities of
the amine and aldehyde and adding directly to the reaction,
without isolation or purification (entry 3). We were pleased to see
While the ligands could be made directly, the procedure we
were following required the isolation of the metal complex prior to
catalysis. In order to demonstrate the robustness of the method-
ology we applied it to a reaction that could be carried out in one
step, without the need for intermediate isolation or removal of
solvent. We focused on the Zn(II) catalyzed hydrosilylation of pro-
* Corresponding author. Tel.: þ353 (0) 1 8964449; fax: þ353 (0) 1 671 2826.
E-mail address: ebergin@tcd.ie (E. Bergin).
0022-328X/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jorganchem.2012.06.013

160
T. Zaman et al. / Journal of Organometallic Chemistry 716 (2012) 159e166
Table 1
that the results of this crude mixture coincided remarkably well
with the results obtained using isolated, chromatography-purified,
ligands.
Comparison of in-situ ligands with literature values.
We proceeded to examine different ligand classes, starting with
imines (Table 2). From a rapid screen we were able to find a ligand
giving an ee of 49% (entry 4). In addition we could begin to observe
what criteria seemed to be beneficial for higher selectivities. In
particular a substituent in the 6-position of the pyridine ring ten-
ded to improve selectivity (entries 3 vs. 4 and 5 vs. 6), although it
had a more variable affect on conversions.
Using a simple dipeptide as a chiral amine would be attractive
due to the degree of variability it would introduce to ligand
structure; as such we tested H-Phe-Val-OMe. Imine formation was
observed as before and the hydrosilylation reaction proceeded,
albeit in low yield (entry 7). Finally phosphine ligands were also
screened e surprisingly they had not been examined in this reac-
tion before. Again they allowed the reaction to proceed, although
they appeared inferior to analogues pyridine compounds (entry 8
and 9).
Entry
1a
Ligand system
ee (%) Conversion (%) Ref.
32
100
[5a]
2a
30
33
99
99
this work
this work
Having identified the most selective ligand, we synthesised,
isolated and purified compound 13. We then re-ran the reaction
with the isolated, purified ligand, and were pleased to find that the
results corresponded to those observed in our screening method
(entry 10 vs. entry 4), confirming the reliability demonstrated in
Table 1, and showing that the rapid screening is a valuable
3
a
Ligand not made in-situ.
Table 2
Screening of imine ligands.
Entry
1
Ligand system
Substrate
ee (%)
20
Conversion (%)
Config.
4a
4
S
2
4a
20
62
S
3
4
4a
4a
20
49
86
48
R
R

T. Zaman et al. / Journal of Organometallic Chemistry 716 (2012) 159e166
161
Table 2 (continued )
Entry
Ligand system
Substrate
ee (%)
Conversion (%)
Config.
5
6
4a
15
99
R
R
4a
4a
38
27
20
7
5
S
8
9
4a
4a
17
67
19
S
S
5
10a
4a
4b
50
56
51
62
R
R
11a
a
Ligand not made in-situ.
methodology that allows accurate identification of ligand efficiency
with minimal time, effort and materials required compared to
traditional methods.
to imines. However we did notice a similar trend to Table 2 in that
substituents on the ortho position of the pyridyl ring increased
enantioselectivity in the order Me > MeO > H (entry 1 vs. 2 vs. 3,
entry 4 vs. 5 vs. 6 and entry 7 vs. 8 vs. 9). As such this appears to be
a quite general trend in this reaction. Related compounds such as
oxazolidines and thiazolines performed worse in the reaction,
giving low ee and conversion (entry 10) or racemic product (entry
11). Of those screened one ligand system gave an ee of 50%,
matching that achieved with imine ligands (entry 6).
We subsequently examined a third class of ligands; bisimines
(see Table 4). As a background test we first used unsubstituted
cyclohexyldiamine in the reaction, it produced the product in 24%
ee and 99% conversion (entry 1, similar to that previously reported
by Mimoun et al. [5a]). Forming imines with pyridine and quinoline
aldehydes resulted in a greatly reduced enantioselectivity (entries 2
and 3), however selectivity improved when alkyl substituents were
In addition we examined the generality of ligand 13 by exam-
ining selectivity with a separate substrate. Rather than ketone 4a
we synthesised the N-phosphoryl imine 4b, and ran the reaction
under identical conditions (entry 11). Using this substrate the ee
increased to 56%, with an increase in conversion also observed.
We next examined aminal based ligands e these somewhat
under-examined ligands have proven useful in such reactions as
transfer hydrogenation [6] and palladium catalyzed asymmetric
allylation [7] but had not previously been employed in this reaction.
As we reported these could be made in high yields and purities by
the combination of secondary diamines with aldehydes, allowing
easy synthesis and screening to be undertaken (see Table 3) [4].
Firstly we observed that yields of product were quite low compared

162
T. Zaman et al. / Journal of Organometallic Chemistry 716 (2012) 159e166
Table 3
Screening of aminals and related ligands.
Entry
Ligand system
ee (%)
Conversion (%)
Config.
1
5
20
S
2
3
4
20
26
35
23
28
47
S
S
S
5
41
27
S
6
50
42
S
7
22
15
S

T. Zaman et al. / Journal of Organometallic Chemistry 716 (2012) 159e166
163
Table 3 (continued )
Entry
Ligand system
ee (%)
Conversion (%)
Config.
8
29
29
S
9
35
30
S
10
11
13
7
S
0
71
e
used e cyclohexyl provided the product in 80% conversion with an
ee of 51% (entry 4). Increasing the steric bulk of the alkyl group
decreased the conversion without any improvement in selectivity
e tert-butyl groups provided the same selectivity as isopropyl, but
with a reduced conversion (entry 5 vs. entry 6).
speeding up the catalysts optimization process. Four different
ligand classes were explored; with three ligands providing an ee of
50% or more being rapidly identified (a maximum ee of 56% was
obtained). Furthermore some insight into the factors affected
selectivity e such as the suitability of certain ligand classes and the
Finally we looked at the use of more complex ligand structures,
using imidate 22 as a starting point. This allowed the screening of
imidazolines (Table 5, entries 1 and 2), oxazolines (entries 3 and 4)
and thiazolines (entry 5). In general it was found that these
compounds were inferior to imine and aminal based ligands, giving
maximum enantioselectivities of 15% (although conversions were
generally much higher than those observed with aminals). While
this was disappointing, the fact that these ligand types could be
screened so easily means that their utility could be determined
without wasting time or materials on full-scale syntheses.
presence of ortho-substituents on the pyridine ring
obtained.
e was
4. Experimental
Toluene was dried over calcium hydride prior to use. Chemicals
were commercially available and used without further purification
unless otherwise stated. Analytical CSP-HPLC was performed using
a Chiralpak IB (4.6 mm ꢀ 25 cm) column. Compound 10 was made
as the N-Boc derivative as previously described [8], and subse-
quently deprotected by stirring in TFA/CH2Cl2 (1:1). Compounds 5
[9], 11 [10], 12 [10], 14 [11], 15 [11] and 16 [12] were synthesised
using established procedures.
3. Conclusion
In conclusion we have applied our methodology to the hydro-
silylation of ketones with zinc and PMHS. The robustness of the
process was highlighted by the comparability of results obtained by
purified ligands versus those obtained by a combinatorial approach
of mixing ligand fragments together prior to the reaction. This
means that compared to a traditional methodology of synthesizing
and purifying ligands individually the amount of time, chemicals
(including expensive chiral compounds) and organic solvents (for
purification) are greatly reduced. The entire catalytic process could
be carried out in the same flask and in the same solvent, greatly
4.1. General procedure for ligand formation
A vial was charged with aldehyde or imidate (0.085 mmol),
amine, amino alcohol or diamine or (0.085 mmol) and anhydrous
toluene (0.4 mL). To this was added 4Å molecular sieves and the vial
was sealed. The reaction was then stirred at 70 ꢁC for 12 h. Mixtures
were then added directly to the catalytic reaction. In cases where
ligands were synthesised prior to the reaction then the same

164
T. Zaman et al. / Journal of Organometallic Chemistry 716 (2012) 159e166
Table 4
Screening of bisimine ligands.
Entry
1
Ligand system
ee (%)
24
Conversion (%)
Config.
99
S
2
3
3
2
99
88
S
S
4
51
80
S
5
6
38
39
31
67
S
S
procedure was followed, and ligands were purified by column
(100 Hz, CDCl3) d 160.6, 157.5, 153.9, 140.0, 136.3, 133.6, 130.3,
chromatography prior to use.
128.5, 127.1, 125.6, 125.2, 124.9, 123.9, 123.6, 123.2, 117.9, 64.4,
23.9, 23.6. HRMS calcd for C19H18N2Na [M þ Na]þ 297.1374, found
297.1368.
4.1.1. Imine 1
1H NMR
d
(400 MHz, CDCl3) 8.48 (d, 1H, J ¼ 4.5 Hz), 8.37 (s, 1H),
7.95 (d, 1H, J ¼ 7.2 Hz), 7.49 (ap t, 1H), 7.39e7.29 (m, 2H), 7.27e7.17
4.2. General procedure for asymmetric hydrosilylation
(m, 2H), 7.17e6.98 (m, 2H), 4.51 (q, 1H, J ¼ 6.6 Hz), 1.49 (d, 3H,
J ¼ 6.6 Hz). 13C NMR
d
(100 MHz, CDCl3) 160.0, 154.3, 148.9, 144.1,
To a flask containing the crude ligand mixture ZnEt2 (0.085 mL,
1 M in toluene, 0.085 mmol) was added via syringe under N2.
Acetophenone (0.5 mL, 4.3 mmol) and PMHS (0.33 mL, 5.2 mmol)
were added to the mixture at 0 ꢁC, and the reaction was allowed to
warm to rt. After stirring the solution at rt for 24 h, NaOH (0.9 mL,
45%) was added carefully. After 15 min,10 mL of water was added to
the mixture. The product was extracted with diethyl ether
(2 ꢀ 10 mL). After evaporating the solvents the product was
distilled under reduced pressure.
136.1, 128.1, 126.6, 126.3, 124.3, 121.0, 69.2, 24.2. Consistent with
literature values [13].
4.1.2. Imine 8
1H NMR (400 Hz, CDCl3)
d
8.53 (s, 1H), 8.27 (d, 1H, J ¼ 8.3 Hz),
8.00 (d, 1H, J ¼ 7.7 Hz), 7.90 (d, 1H, J ¼ 8.0 Hz), 7.81 (apt t, 2H),
7.66 (apt t, 1H), 7.61e7.46 (m, 3H), 7.20 (d, 1H, J ¼ 7.6 Hz), 5.49 (q,
1H, J ¼ 6.7 Hz), 2.61 (s, 3H), 1.79 (d, 3H, J ¼ 6.7 Hz). 13C NMR

T. Zaman et al. / Journal of Organometallic Chemistry 716 (2012) 159e166
165
Table 5
Screening oxazolines, imidazolines and thiazolines.
Entry
Ligand system
ee (%)
Conversion (%)
Config.
1
15
98
(R)
2
3
0
0
30
88
e
e
4
5
7
5
82
21
(R)
(R)
(b) J.F. Jensen, K. Worm-Leonhard, M. Meldal, Eur. J. Org. Chem. (2008) 3785;
(c) T. Arai, N. Yokoyama, A. Yanagisawa, Chem. Eur. J. 14 (2008) 2052;
(d) C.A. Christensen, M. Meldal, J. Comb. Chem. 9 (2007) 79;
(e) C.M. Sprout, M.L. Richmond, C.T. Seto, J. Org. Chem. 69 (2004) 6666;
(f) M.B. Francis, E.N. Jacobsen, Angew. Chem. Int. Ed. 38 (1999) 937.
[3] (a) A.J. Minnaard, B.L. Feringa, L. Lefort, J.G. de Vries, Acc. Chem. Res. 40 (2007)
1267 (and references cited therin);
(b) J.F. Teichert, B.L. Feringa, Angew. Chem. Int. Ed. 49 (2010) 2486.
[4] R. Frauenlob, M.M. McCormack, C.M. Walsh, E. Bergin, Org. Biomol. Chem. 9
(2011) 6934.
Acknowledgement
We thank Trinity College Dublin and the Pakistan Higher
Education Commission (TZ) for funding.
References
[1] (a) B. Breit, Angew. Chem. Int. Ed. 44 (2005) 6816;
(b) X.-B. Jiang, L. Lefort, P.E. Goudriaan, A.H.M. de Vries, P.W.N.M. van Leeu-
wen, J.G. de Vries, J.N.H. Reek, Angew. Chem. Int. Ed. 45 (2006) 1223;
(c) A.C. Laungani, J.M. Slattery, I. Krossing, B. Breit, Chem. Eur. J. 14 (2008)
4488;
[5] (a) H. Mimoun, J.Y. de Saint Laumer, L. Giannini, R. Scopelliti, C. Floriani, J. Am.
Chem. Soc. 121 (1999) 6158e6166;
(b) V.M. Mastranzo, L. Quintero, C.A. de Parrodi, E. Juaristi, P.J. Walsh, Tetra-
hedron 60 (2004) 1781e1789;
(c) V. Bette, A. Mortreux, D. Savoia, J.-F. Carpentier, Tetrahedron 60 (2004)
2837e2842;
(d) J. Meeuwissen, M. Kuil, A.M. van der Burg, A.J. Sandee, J.N.H. Reek, Chem.
Eur. J. 15 (2009) 10272.
(d) V. Bette, A. Mortreux, F. Ferioli, G. Martelli, D. Savoia, J.-F. Carpentier, Eur. J.
Org. Chem. (2004) 3040e3045;
[2] (a) R. den Heeten, B.H.G. Swennenhuis, P.W.N.M. van Leeuwen, J.G. de Vries,
P.C.J. Kamer, Angew. Chem. Int. Ed. 47 (2008) 6602;

166
T. Zaman et al. / Journal of Organometallic Chemistry 716 (2012) 159e166
(e) V. Bette, A. Mortreux, D. Savoia, J.-F. Carpentier, Adv. Synth. Catal. 347 (b) E.-K. Lee, S.-H. Kim, B.-H. Jung, W.-S. Ahn, G.-J. Kim, Tetrahedron Lett. 44
(2005) 289e302;
(2003) 1971e1974.
(f) E. Santacruz, G. Huelgas, S.K. Angulo, V.M. Mastranzo, S. Hernandez-Ortega,
J.A. Avina, E. Juaristi, C.A. de Parrodi, P.J. Walsh, Tetrahedron: Asymmetry 20
(2009) 2788e2794.
[8] S. Ray, M.G.B. Drew, A.K. Das, A. Banerjee, Tetrahedron 62 (2006) 7274e7283.
[9] R. Rasappan, O. Reiser, Eur. J. Org. Chem. (2009) 1305e1308.
[10] L.B. Schenkel, J.A. Ellman, Org. Lett. 5 (2003) 545e548.
[11] O. Equey, A. Alexakis, Tetrahedron: Asymmetry 15 (2004) 1069e1072.
[12] T. Kylmala, N. Kuuloja, Y. Xu, K. Rissanen, R. Franzen, Eur. J. Org. Chem. (2008)
4019e4024.
[6] G.-J. Kim, S.-H. Kim, P.-H. Chong, M.-A. Kwon, Tetrahedron Lett. 43 (2002)
8059e8062.
[7] (a) M.-J. Jin, S.-H. Kim, S.-J. Lee, Y.-M. Kim, Tetrahedron Lett. 43 (2002)
7409e7411;
[13] E.V. Anslyn, S. Nieto, J.M. Dragma, Chem. Eur. J. 16 (2010) 227e232.