Asymmetric ketone hydroboration catalyzed by alkali metal complexes derived from BINOL ligands

Article abstract of DOI:10.1039/d0dt00232a

The ability of alkali metal complexes featuring functionalized BINOL-derived ligands to catalyze ketone hydroboration reactions was explored. The reduced products were formed in excellent yields and with variable enantioselectivities dependent upon the nature of the ligand and the alkali metal cation.

Full text of DOI:10.1039/d0dt00232a

Dalton
Transactions
View Article Online
View Journal
COMMUNICATION
Asymmetric ketone hydroboration catalyzed
by alkali metal complexes derived from BINOL
Cite this: DOI: 10.1039/d0dt00232a
ligands†
Received 22nd November 2019,
Accepted 22nd January 2020
Darren Willcox, a,b Jamie L. Carden,a Adam J. Ruddy,a Paul D. Newman *a and
Rebecca L. Melen
a
DOI: 10.1039/d0dt00232a
*
rsc.li/dalton
The ability of alkali metal complexes featuring functionalized available group I metal salts has been at the forefront of this
BINOL-derived ligands to catalyze ketone hydroboration reactions research area.
was explored. The reduced products were formed in excellent
Several groups have recently made major advancements
yields and with variable enantioselectivities dependent upon the demonstrating that simple sodium salts (NaOtBu, NaH and
nature of the ligand and the alkali metal cation.
NaOH)8–10 and lithium salts (nBuLi and LiHBEt3)11–13 are
highly active catalysts for carbonyl reductions. The simplicity
of these alkali metal species suggests that they could serve as
ideal pre-catalysts for the development of enantioselective
s-block catalyzed ketone reductions in the presence of a chiral
ligand. This in situ approach would bypass the need to syn-
thesize complex species from lithium intermediates and could
facilitate significant advancements in main group chemistry.
To this end, we sought to explore whether alkali metal cata-
lysts in the presence of chiral alcohols, may be utilized for
enantioselective ketone hydroboration. Asymmetric hydrobora-
tions are attractive as the products of such reactions furnish
optically active organoboron compounds which are valuable
Catalytic carbonyl hydroboration to give, ultimately, primary or
secondary alcohols has been realized utilizing a plethora of
different catalysts derived from transition metal or f-block
metal complexes.1,2 Many of these catalysts are expensive and/
or their preparation is synthetically challenging. This has
prompted a number of groups to explore the application of
main group compounds as alternative catalysts for this and
other reductions.3 While p-block elements have dominated
this research,4 the exploration of s-block catalysts is less preva-
lent with the alkaline Earth metals (mainly magnesium and
calcium) taking centre stage.5
Encouraging results demonstrating the effective catalytic
ability of group I metals in carbonyl hydroborations have been
reported recently (Scheme 1). Pioneering work by the Okuda
group revealed that a well-defined lithium hydridotriphenyl
borate, bearing a chelating ligand (tris{2-(dimethylamino)
ethyl}amine), was an extremely efficient catalyst for carbonyl
reductions with low catalyst loadings (0.001 mol%).6 The
Mulvey group demonstrated that carbonyl reduction was
building blocks for accessing
a
number of chiral
structures.14,15 The wide application of BINOL-derived frame-
works in asymmetric catalysis led us to choose ligands
achievable using
a
heterobimetallic lithium/aluminum
complex capable of participating in cooperative catalysis
leading to high yields of the desired alcohols.7 Despite these
elegant approaches, the applicability of these complexes is
limited, mainly due to ligand specificity and catalyst pre-prepa-
ration. As a result, the utilization of simple, commercially
aCardiff Catalysis Institute, School of Chemistry, Cardiff University, Main Building,
Park Place, Cardiff, CF10 3AT Cymru/Wales, UK. E-mail: MelenR@cardiff.ac.uk,
NewmanP1@cardiff.ac.uk
bDepartment of Chemistry, University of Manchester, Oxford Rd, Manchester,
M13 9PL, UK
†Electronic supplementary information (ESI) available: Detailed experimental
procedures, HPLC traces and compound data. See DOI: 10.1039/d0dt00232a
Scheme 1 Previously reported s-block ketone hydroboration catalysts;
Dipp = 2,6-diisopropylphenyl.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Communication
Dalton Transactions
(R)-L1–L7 for this study. Our initial investigations focussed on 3,5-xylyl) groups led to the product with significantly decreased
the reduction of acetophenone (1a). enantioselectivity (99% and 96% yield, and 58 : 42 and
Under optimized reaction conditions, 1.2 equivalents of 57.5 : 42.5 er respectively). Changing the electronic properties of
HBpin, mol% of lithium diisopropylamide (LDA) and the phosphine oxide from phenyl to isopropyl groups (compare
5
10 mol% (R)-L1 in 1,4-dioxane for 18 h (Table 1, entry 1) (see (R)-L1 and (R)-L4), delivered the product in high yield but again
ESI† for optimization tables), the scalemic alcohol product (2a) with low levels of enantioselectivity (53 : 47 er). (R)-BINOL
was formed in 94% yield and 79 : 21 enantiomeric ratio. Ligand ((R)-L5) and the simple monomethylated BINOL ((R)-L6) were
(R)-L1 was chosen initially as it contains a single alcoholic also tested with the products being observed in good yields but
proton, which would ideally lead to a single deprotonated low enantioselectivity. In the case of (R)-L5, the low er could be
species upon deprotonation by LDA. Furthermore, it was due to the presence of two alcoholic protons potentially produ-
thought that the presence of the closely tethered phosphine cing complex mixtures of active species upon deprotonation.
oxide group may be required for stabilizing the alkali metal Finally, (R)-1,1′-binaphthyl-2,2′-diyl-hydrogenphosphate ((R)-L7)
catalyst. Control experiments showed that, in the absence of was also screened as a ligand as chiral phosphoric acids have
alkali metal catalyst no reaction occurred (entries 2 and 3). been demonstrated to be privileged ligands for certain asym-
However, in the presence of LDA but absence of ligand, no metric transformations.16 Unfortunately, under our conditions
enantioselectivity was observed although the product was still (R)-L7 produced racemic product. Decreasing the catalytic
observed in a high yield (85%, entry 4). A change in the stereo- loading of (R)-L1 from 10 mol% to 5 mol% was deleterious to
electronic properties of the substituents on the phosphine oxide the enantioselectivity (entry 6). The combination of LDA and
moiety of the ligand ((R)-L1–L4) proved to be critical for enantio- (S)-2a was catalytically competent but gave 0% er of product
selectivity (entry 5). Indeed, changing the phenyl group for the proving that (R)-L1 is critical for enantioselective induction.
more sterically encumbered mesityl ((R)-L2) or mexyl ((R)-L3,
The influence of the base was next evaluated. Replacing
LDA for 5 mol% LiOtBu led to the desired product in an extre-
mely high yield with a moderate 70 : 30 er (entry 7). Given the
by-products from the pre-mixing of (R)-L1 with either LDA or
LiOtBu were diisopropylamine or tert-butanol respectively, it
was possible these were forming catalytically competent
racemic species in situ. With this in mind, we envisaged chan-
ging LDA for LiH would lead to higher enantioselectivity, as
the by-product from pre-mixing would be H2. Interestingly, an
er of 68 : 32, very similar to LiOtBu but lower than LDA (entry
8), was observed suggesting that either the by-products are
innocent and do not influence the catalyst or they are impor-
tant for enantioselectivity (mainly for diisopropylamine). The
reaction also proceeded in other ethereal solvents such as THF
in good yields albeit with a slightly reduced er (entry 9). The
use of other polar non-coordinating solvents, such as CH2Cl2
gave good yields but reduced er (56 : 44, entry 10). Changing
from 1,4-dioxane to toluene, a non-polar and non-coordinating
solvent led to only racemic products being observed (entry 11).
This result can be attributed to the low solubility of the
lithium phenolate salt in toluene (mixture remained hetero-
geneous). Replacing pinacol borane with catechol borane was
also effective however due to the higher reactivity of catechol
borane a decreased enantiomeric ratio was observed (entry 12).
Finally, lowering the reaction temperature to 10 °C provided
the desired product in low yield and enantioselectivity (entry
13). We attribute the lower er to insolubility of the lithium salt
in this solvent at this temperature.
Table 1 Selected optimization of reaction conditions
Entry
Deviation from standard conditionsa
Yieldb (%)
erc (%)
1
2
3
4
None
No LDA
No LDA and no ligand
No ligand
94
79 : 21
—
—
NRd
NRd
85
50 : 50
5
6
7
8
L2–L7 instead of L1
5 mol% L1
Listed below
95
99
70
98
98
96
98
27
58 : 42
70 : 30
68 : 32
70 : 30
56 : 44
50 : 50
56 : 44
56 : 44
LiOtBu instead of LDA
LiH instead of LDA
THF instead of 1,4-dioxane
CH2Cl2 instead of 1,4-dioxane
Toluene instead of 1,4-dioxane
HBCat instead of HBPin
10 °C instead of RT
9
10
11
12
13
With suitable conditions in hand, we next explored a small
substrate scope for this reaction (Scheme 2). A series of simple
acetophenone (1a–1l) derivatives exhibiting different steric and
electronic properties on the phenyl ring were evaluated. When
electron neutral acetophenone derivatives were employed, the
desired alcohols (2a and 2b) could be obtained in good yields
with moderate to good enantiomeric ratios (79 : 21 and 65 : 35,
respectively). Introduction of electron withdrawing groups
a All the reactions were run on a 0.25 mmol scale. b The yield was deter-
mined by 1H NMR spectroscopy using 1,1,2,2-tetrachloroethane as the
internal standard. c er determined by chiral HPLC analysis. Mexyl =
3,5-dimethylphenyl; mesityl = 2,4,6-trimethylphenyl. d Conversion ana-
lysed by 1H NMR spectroscopy before workup. NR denotes no reaction.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Communication
between L1 and LDA in 1,4-dioxane (with benzene-d6 lock) and
probed it using multinuclear NMR spectroscopy (Scheme 3,
top). As expected, deprotonation of the phenolic proton
occurred rapidly and cleanly (within 5 min) and the loss of
this proton was indicative by the disappearance of a resonance
at δ = 9.28 ppm in the 1H NMR spectrum.
The stoichiometric reaction between (R)-L1, LDA and pina-
colborane in 1,4-dioxane was subsequently explored. There was
no observable change in chemical shift in both the aromatic
region and for the tetramethyl protons of the pinacol group in
1
the H NMR spectrum. However, two new singlets appeared at
δ = 0.53 and 0.20 ppm. The 11B NMR spectrum identified the
presence of three boron containing species (δ = 28.4, 21.5, and
7.2 ppm). The doublet at δ = 28.4 (1JBH = 173.0 Hz) is attributed
to pinacolborane, indicating incomplete consumption of pina-
colborane. The second resonance at δ = 21.5 ppm can be attrib-
uted to the formation of the borate species (Scheme 3, bottom).
This species was also identifiable when (R)-L1 and pinacolbor-
ane were reacted in a stoichiometric fashion. The final 11B reso-
nance at δ = 7.2 ppm can be attributed to the formation of the
lithium trialkoxyborohydride species (Scheme 3, bottom). This
11B NMR resonance is consistent with trialkyloxyborohydrides
reported by Brown and Clark (δ = 0–7 ppm).8,17 The resonance
observed at δ = 7.2 ppm is significantly less intense than the
corresponding resonance at δ = 21.5 ppm and it was noted that,
Scheme 2 Substrate scope. All the reactions were run on a 0.25 mmol
scale. er determined by chiral HPLC analysis and given in parentheses.
NR denotes no reaction.
such as fluorine or nitrile onto the phenyl ring were tolerated, at the concentration these stoichiometric reactions were per-
resulting in good yields of the products (2c and 2d) with mod- formed (0.1 M), a large quantity of precipitate was observed and
erate enantiomeric ratios (up to 77 : 23).
that the borohydride species was only sparingly soluble at this
When the phenyl ring was substituted with mild inductive concentration. Evaluation of the 31P NMR spectrum showed
electron donating groups, such as methyl (2e–2g), good yields negligible changes in chemical shift upon both deprotonation
of the product could be observed for the para and meta-substi- and coordination with the pinacolborane. A repeat experiment
tuted acetophenones and similar enantiomeric ratios to aceto- using two equivalents of (R)-L1 to mimic the most successful
phenone itself were observed. Moving the methyl group into catalytic systems gave similar results to those detailed above
the ortho-position (2g) led to an observed decrease in yield and except complete consumption of the pinacolborane and full
enantiomeric ratio (47% and 62 : 38 respectively). The addition conversion to the borate species at δ = 21.5 ppm was observed.
of
a strong mesomeric electron-donating group, such as
p-methoxy (2h), led to a decreased yield compared to the para-
substituted methyl variant however similar enantiomeric ratios
were observed (75 : 25 vs. 72 : 28). This lower yield can be attribu-
ted to a decreased electrophilicity of the carbonyl group. Altering
the substitution on the alkyl side of the acetophenone was also
achievable with both ethyl-(2i) and cyclohexyl-(2j) groups being
tolerated in good to excellent yields. From this substrate scope,
it is evident that steric factors play an important role in both the
yield and enantioselectivity. Acetophenone derivatives bearing
either ortho substituents (2b and 2g) or bulky alkyl substituents
(2j) all resulted in the formation of the desired products albeit
with reduced yields and enantioselectivity. Whereas very steri-
cally hindered substrates such as mesityl or cyclopropyl (1k and
1l) resulted in recovery of the starting material. These obser-
vations suggest that there is a steric interaction between the
active catalytic species, bearing the bulky binaphthyl backbone
and the substrate, possibly favoring a faster uncatalyzed back-
ground reaction and resulting in diminished enantioselectivity.
In an effort to examine the nature of the species generated
Scheme 3 NMR experiments to try and elucidate the nature of the
in solution, we first performed a stoichiometric reaction catalytic species. All reactions were performed on a 0.1 mmol scale.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Communication
Dalton Transactions
There was no observable lithium trialkoxyborate species in the
NMR spectra under these conditions.
Catal., 2017, 7, 7257–7266; (f) D. Mukherjee and J. Okuda,
Angew. Chem., Int. Ed., 2017, 57, 1458–1473; (g) S. Dagorne
and R. Wehmschulte, ChemCatChem, 2018, 10, 2509–2520.
4 (a) M. Arrowsmith, T. J. Hadlington, M. S. Hill and G. Kociok-
In conclusion, we have developed an enantioselective s-block
catalyzed hydroboration of acetophenones. The chiral catalyst is
comprised of a BINOL derived ligand and LDA. Using multinuc-
lear NMR spectroscopy, we found that the phenolic proton in
the ligand is cleanly deprotonated with LDA and subsequent
addition of pinacol borane leads to the formation of a chiral
trialkyloxyborohydride species. This catalyst provides access to
scalemic secondary alcohols in good to excellent yields and is
operationally simple. This methodology opens the door for
other asymmetric s-block based catalysis.
Köhn,
Chem.
Commun.,
2012,
48,
4567–4569;
(b) T. J. Hadlington, M. Hermann, G. Frenking and C. Jones,
J. Am. Chem. Soc., 2014, 136, 3028–3031; (c) C. C. Chong,
H. Hirao and R. Kinjo, Angew. Chem., Int. Ed., 2014, 54, 190–
194; (d) Z. Yang, M. Zhong, X. Ma, S. De, C. Anusha,
P. Parameswaran and H. W. Roesky, Angew. Chem., Int. Ed.,
2015, 54, 10225–10229; (e) J. Schneider, C. P. Sindlinger,
S. M. Freitag, H. Schubert and L. Wesemann, Angew. Chem.,
Int. Ed., 2016, 56, 333–337; (f) Y. Wu, C. Shan, Y. Sun, P. Chen,
J. Ying, J. Zhu, L. (Leo) Liu and Y. Zhao, Chem. Commun.,
2016, 52, 13799–13802; (g) V. K. Jakhar, M. K. Barman and
S. Nembenna, Org. Lett., 2016, 18, 4710–4713; (h) M. K. Bisai,
S. Pahar, T. Das, K. Vanka and S. S. Sen, Dalton Trans., 2017,
46, 2420–2424; (i) J. R. Lawson, L. C. Wilkins and R. L. Melen,
Chem. – Eur. J., 2017, 23, 10997–11000.
5 (a) D. Mukherjee, A. Ellern and A. D. Sadow, Chem. Sci.,
2014, 5, 959–964; (b) L. Fohlmeister and A. Stasch, Chem. –
Eur. J., 2016, 22, 10235–10246; (c) D. Mukherjee, S. Shirase,
T. P. Spaniol, K. Mashima and J. Okuda, Chem. Commun.,
2016, 52, 13155–13158; (d) K. Manna, P. Ji, F. X. Greene and
W. Lin, J. Am. Chem. Soc., 2016, 138, 7488–7491; (e) S. Yadav,
S. Pahar and S. S. Sen, Chem. Commun., 2017, 53, 4562–4564.
6 (a) D. Mukherjee, H. Osseili, T. P. Spaniol and J. Okuda,
J. Am. Chem. Soc., 2016, 138, 10790–10793; (b) H. Osseili,
D. Mukherjee, K. Beckerle, T. P. Spaniol and J. Okuda,
Organometallics, 2017, 36, 3029–3034.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
We are grateful to the Leverhulme Trust for a research project
grant (D. W., A. J. R., P. D. N. and R. L. M. RPG-2016-020). We
also thank the EPSRC for funding: (J. L. C, EP/L016443/1;
R. L. M, EP/R026912/1).
Notes and references
1 (a) N. Eedugurala, Z. Wang, U. Chaudhary, N. Nelson,
K. Kandel, T. Kobayashi, I. I. Slowing, M. Pruski and
A. D. Sadow, ACS Catal., 2015, 5, 7399–7414; (b) G. Zhang,
H. Zeng, J. Wu, Z. Yin, S. Zheng and J. C. Fettinger, Angew.
Chem., Int. Ed., 2016, 55, 14369–14372; (c) R. Arévalo,
C. M. Vogels, G. A. MacNeil, L. Riera, J. Pérez and
S. A. Westcott, Dalton Trans., 2017, 7750–7757;
(d) V. L. Weidner, C. J. Barger, M. Delferro, T. L. Lohr and
T. J. Marks, ACS Catal., 2017, 7, 1244–1247; (e) A. Baishya,
S. Baruah and K. Geetharani, Dalton Trans., 2018, 9231–
7 (a) R. McLellan, A. R. Kennedy, R. E. Mulvey, S. A. Orr and
S. D. Robertson, Chem. – Eur. J., 2017, 23, 16853–16861;
(b) V. A. Pollard, S. A. Orr, R. McLellan, A. R. Kennedy, E. Hevia
and R. E. Mulvey, Chem. Commun., 2018, 54, 1233–1236.
8 I. P. Query, P. A. Squier, E. M. Larson, N. A. Isley and
T. B. Clark, J. Org. Chem., 2011, 76, 6452–6456.
9 S. J. Yang, A. K. Jaladi, J. H. Kim, S. Gundeti and D. K. An,
Bull. Korean Chem. Soc., 2019, 40, 34–38.
9236; (f) J. Wu, H. Zeng, J. Cheng, S. Zheng, J. A. Golen, 10 Y. Wu, C. Shan, J. Ying, J. Su, J. Zhu, L. L. Liu and Y. Zhao,
D. R. Manke and G. Zhang, J. Org. Chem., 2018, 83, 9442– Green Chem., 2017, 19, 4169–4175.
9448; (g) U. K. Das, C. S. Higman, B. Gabidullin, J. E. Hein 11 Z. Zhu, X. Wu, X. Xu, Z. Wu, M. Xue, Y. Yao, Q. Shen and
and R. Tom Baker, ACS Catal., 2018, 8, 1076–1081; X. Bao, J. Org. Chem., 2018, 83, 10677–10683.
(h) G. Zhang, J. Cheng, K. Davis, M. G. Bonifacio and 12 K. Kucinski and G. Hreczycho, Green Chem., 2019, 21,
C. Zajaczkowski, Green Chem., 2019, 21, 1114–1121. 1912–1915.
2 (a) S. Chen, D. Yan, M. Xue, Y. Hong, Y. Yao and Q. Shen, 13 M. K. Bisai, T. Das, K. Vanka and S. S. Sen, Chem.
Org. Lett., 2017, 19, 3382–3385; (b) Z. Huang, S. Wang, Commun., 2018, 54, 6843–6846.
X. Zhu, Q. Yuan, Y. Wei, S. Zhou and X. Mu, Inorg. Chem., 14 P. Kaur, G. Khatik and S. Nayak, A Review on Advances in
2018, 57, 15069–15078; (c) W. Wang, X. Shen, F. Zhao,
H. Jiang, W. Yao, S. A. Pullarkat, L. Xu and M. Ma, J. Org.
Organoborane-Chemistry: Versatile Tool in Asymmetric
Synthesis, 2016, vol. 14.
Chem., 2018, 83, 69–74; (d) H. Liu, K. Kulbitski, M. Tamm 15 D. G. Hall, in Boronic Acids: Preparation and Applications in
and M. S. Eisen, Chem. – Eur. J., 2018, 24, 5738–5742.
3 (a) S. Harder, Chem. Rev., 2010, 110, 3852–3876;
Organic Synthesis and Medicine, Wiley-VCH Verlag GmbH &
Co. KGaA, Weinheim, FRG, 2006, pp. 1–99.
(b) K. Revunova and G. I. Nikonov, Dalton Trans., 2015, 44, 16 R. Maji, S. C. Mallojjala and S. E. Wheeler, Chem. Soc. Rev.,
840–866; (c) C. C. Chong and R. Kinjo, ACS Catal., 2015, 5, 2018, 47, 1142–1158.
3238–3259; (d) M. S. Hill, D. J. Liptrot and C. Weetman, 17 H. C. Brown, J. S. Cha and B. Nazer, Inorg. Chem., 1984, 23,
Chem. Soc. Rev., 2016, 45, 972–988; (e) G. I. Nikonov, ACS 2929–2931.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020