Double Diastereoselective Approach to Chiral syn- and anti-1,3-Diol Analogues through Consecutive Catalytic Asymmetric Borylations

Article abstract of DOI:10.1021/acs.joc.7b00854

Homoallylic boronate carboxylate esters derived from unsaturated aldehydes via an imination, β-borylation, imine hydrolysis, and Wittig trapping sequence, were subjected to a second boryl addition to give 1,3-diborylated carboxylate esters. Control of the

Full text of DOI:10.1021/acs.joc.7b00854

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
A double diastereoeselective approach to chiral syn- and anti-1,3-
diol analogues through consecutive catalytic asymmetric borylations
Alba Pujol, and Andrew Whiting
J. Org. Chem., Just Accepted Manuscript • Publication Date (Web): 19 Jun 2017
Downloaded from http://pubs.acs.org on June 20, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 43
The Journal of Organic Chemistry
1
2
3
A double diastereoeselective approach to chiral synꢀ and antiꢀ1,3ꢀdiol analogues
4
5
6
through consecutive catalytic asymmetric borylations
7
8
Alba Pujol and Andrew Whiting*
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Centre for Sustainable Chemical Processes, Department of Chemistry, Durham University
University, Science Laboratories, South Road, Durham DH1 3LE, UK.
Eꢀmail: andy.whiting@durham.ac.uk
https://www.dur.ac.uk/chemistry/staff/profile/?id=204
Abstract
Homoallylic boronate carboxylate esters derived from unsaturated aldehydes via an imination, βꢀ
borylation, imine hydrolysis and Wittig trapping sequence, were subjected to a second boryl
addition to give 1,3ꢀdiborylated carboxylate esters. Control of the absolute and relative
stereochemistry of the two new 1,3ꢀstereogenic centres was achieved through: 1) direct chiral
catalyst controlled asymmetric borylation of the first stereocentre on the unsaturated imine with
high e.e.; and 2) a double diastereoselectively controlled borylation of an unsaturated ester
employing a chiral catalyst to largely overcome directing effects from the first chiral boryl centre to
give poor (mismatched) to good (matched) diastereocontrol. Subsequently, the two CꢀB functions
were transformed into CꢀO systems to allow unambiguous stereochemical assignment of the two
borylation reactions involving oxidation and acetal formation.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 43
1
Introduction
2
3
4
5
6
7
8
9
Organoborane compounds provide exceptional chemical features for transforming into a wide range
of functional groups, often with high degrees of stereochemical control.1 Amongst the range of
applications, their use as intermediates in asymmetric synthesis is perhaps the most valuable2 and
especially for preparing key chiral building blocks, often in an enantioselective manner by the
addition of boron reagents to CꢀC multiple bonds.3 In this respect, the βꢀborylation reaction has
emerged as an increasingly important and flexible approach for the preparation of chiral
organoboron compounds, leading to a wide range of borylated compounds.4 Allylic and homoꢀ
allylic boronates are particularly useful reagents in synthesis due to the potential for further
derivatisation to access key building blocks for constructing multiꢀfunctional chiral compounds.5
Indeed, in previous work, our group has developed effective enantioselective, oneꢀpot
methodologies for the synthesis of homoallylic boronate carboxylate esters 5 starting from α,βꢀ
unsaturated aldehydes 1 (Scheme 1). This nonꢀtrivial approach requires a fourꢀreaction synthetic
sequence that necessarily requires hindered imine formation to control the borylation process in
order to circumvent stability issues of the various intermediates involved and the chemoselectivity
of the boryl addition.6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1 Proposed route to access chiral 1,3ꢀdiols 7 via homoallylic boronates 5 and diboronates 6
from α,βꢀunsaturated aldehydes 1.
Having achieved the synthesis of homoallylic boronate esters 5 in previous work (see Scheme 1),
this system offered an opportunity for the introduction of a second boryl moiety, and hence,
creating a second CꢀB chiral centre with a 1,3ꢀstereochemical relationship. In turn, and in this work
(see Scheme 1), we report how we exploited the opportunity for developing further applications in
wider asymmetric synthesis through the synthesis of chiral 1,3ꢀdiborylated ester derivatives 6.
Indeed, the development of effective synthetic strategies towards 1,3ꢀdiols 7 with high
stereochemical control is a key area for the synthesis of natural and bioactive products, for example,
as found in important pharmaceuticals such as diospongin A7 or erythromycin,8 and the 3,5ꢀ
dihydroxy acid fragment widely present in statins9 (see Figure 1).
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1 Selected examples of compounds with biological interest containing a synꢀ or antiꢀ1,3ꢀdiol
moiety.
The control of both the relative and absolute stereochemistry in 1,3ꢀdiol analogues still represents a
substantial challenge, since only a few building blocks are available for the synthesis of a wide
range of structurally quite diverse compounds. In fact, it is generally found that small substrate
changes in wellꢀknown synthetic procedures for accessing such compounds can result in a decrease
in the yield and loss of stereoselectivity.10 Therefore, there is still a need for alternative, flexible and
highly stereoselective syntheses of key building blocks to add to the current list of aldol reactions,11
reductions,12 alkoxide additions13 and enzymatic reduction of 1,3ꢀdiketones.14 Hence, in this paper,
we report the use of chiral homoallylic boronate esters 5 as substrates for further asymmetric
copper(I)ꢀcatalysed βꢀborylation and examine the stereochemical control effects involved in
accessing the 1,3ꢀdiborylated compounds 6. Moreover, methods were developed to provide both
relative and absolute stereocontrol at each new asymmetric stereocentre using the copper(I)
phosphine ligands to effect stereocontrol. The two boryl units were then examined for
transformation into functionalities which would allow unambiguous stereochemical assignment of
the two borylation reactions.
4
ACS Paragon Plus Environment

Page 5 of 43
The Journal of Organic Chemistry
1
2
3
4
5
Results and discussion
6
7
8
Homoallylic boronate ester 5 synthesis and optimization
9
In previous studies (Scheme 2),6 the asymmetric formation of homoallyl boronates 5 was reported,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
i
carried out in PrOH as favoured solvent. Although this reaction worked well on smaller scales,
larger scale synthesis proved challenging, presumably due to mass transfer issues, i.e. it was not
possible to work scales larger than 2.0 mmol without significant loss of yield. In addition, when
i
examining this reaction (Scheme 2) on model substrate cinnamaldehyde, the use of PrOH as
reaction medium was also found to cause undesired transesterification during the subsequent Wittig
reaction, resulting in a mixture of homoallylic boronate esters 5a and 5ai (Scheme 2).
Scheme 2 Model process for the synthesis of unsaturated esters 5 and associated transesterification
due to the use of iPrOH as reaction medium.
Scheme 2 therefore graphically showed the need for optimization, both in terms of the βꢀboryl
aldimine addition and subsequent hydrolysis, as well as the in situ Wittig reaction. We therefore
examined the rate of addition of a preformed solution of the boryl imine 3a to a solution of
copper(II) sulfate and ylide; Tables 1 and 2 summarize the results.
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 43
1
2
Table 1 Effect of the rate of addition of boryl imine 3a to the hydrolysis/Wittig reaction.a
3
4
5
6
7
8
9
Entry
Addition rate
1 mL/ 30 minc
1 mL/ 15 minc
1 mL/ 10 mind
N/Ae
IYb (%)
32
54
61
1
2
3
4
22
aReaction conditions: see Experimental. Isolated yield
b
c
calculated on homoallylic boronate ester 5a. Addition
d
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
in batches of 1 mL. Dropwise addition via syringe
e
pump. Comparative batch reaction (direct addition of
Wittig reagent and copper salt directly to 3a).
Table 1 shows that compared to the standard batch reaction (Entry 4, Table 1), the slow addition of
the aldimine 3a to a stirred hydrolysis/Wittig reaction mixture (Entries 1ꢀ3, Table 1) resulted in
higher yields of the target homoallylboronate 5a, i.e. improved yields were obtained, from 22% up
to 61% mainly due to cleaner products being directly obtained.
To solve the issue of transesterification, the same reaction sequence outlined in Scheme 2 was
carried out in different solvents for the imine hydrolysisꢀWittig step, as summarised in Table 2.
Table 2 Solvent optimisation for the imine hydrolysisꢀWittig step converting imine 3a to homoallyl
boronate 5a (see Scheme 2).a
Entry
Reaction
solvent
βꢀBoryl
aldimine
solventb
iPrOH
Presence
Yield of
of 5aic
5ad (%)
1
2
3
4
iPrOH
MeOH
MeOH
THF
Yes
Yes
No
20
17
12
89
iPrOH
MeOH
THF
No
aReaction conditions: see experimental. Solid residue of the
b
crude βꢀboryl aldimine 3a was redisolved in the stated solvent.
cTransesterification product 5ai observed by H NMR (crude
1
product). dIsolated yield of pure homoallylboronate 5a.
From Table 2 it became apparent that the use of an alcohol solvent for the imine hydrolysisꢀWittig
step (either to solvate the copper saltꢀWittig reagent, or the aldimine) was not ideal either due to the
transesterification or diminished yields (Entries 1ꢀ3, Table 2). However, after boryl addition to the
unsaturated imine in THF (Entry 4, Table 2) the subsequent hydrolysis/Wittig sequence gave good
yields on the desired product 5a with complete suppression of the transesterification. Also, this set
of reaction conditions were readily scaled up for further use.
6
ACS Paragon Plus Environment

Page 7 of 43
The Journal of Organic Chemistry
1
2
Stereocontrolled βꢀborylation of homoallylic boronate carboxylate esters
3
4
5
Having examined solvent and scale up effects on the formation of homoallyl boronate 5, we then
6
7
turned to examine the second borylation step, i.e. the conversion of 5 to give diboryl system 6 (Eqn.
8
9
1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Initial attempts to perform a βꢀborylation on racemic homoallylic boronate ester 5a and under
racemic borylation conditions gave the corresponding diborylated ester 6a, as indicated by complete
1
consumption of substrate 5a (conversion >99% according to crude H NMR). However, the
presence of an additional product, compound 6ai (Fig. 2), was also detected due to the undesired
transesterification when using iPrOH as solvent and in the presence of NaOtBu. Separation of these
two esters 6a and 6ai for further use also proved impossible
Figure 2 Structures corresponding to the different diborylated esters 6a and 6ai.
In order to avoid the transesterification during the βꢀborylation reaction of homoallylic boronate 5a,
the solvent/additive system was changed to THF/MeOH (Eqn. 2) which successfully provided the
target diborylated ester 6a as a 1:1 mixture of diastereoisomers in a 42% isolated yield.
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 43
1
2
3
4
5
6
7
8
9
O
O
B
O
O
B2pin2 (1.0 equiv)
L1
NaOtBu (9 mol%), THF, MeOH
(2.5 equiv.), RT, 16 h
42%
O
O
B
B
(2)
CO2Me
CO2Me
CuCl (3 mol%), PPh3
(6 mol%)
Ph
Ph
6a
5a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For the purpose of following the second boryl addition outlined in Eqn. 2 to the α,βꢀunsaturated
ester of homoallylic boronate ester 5 to give the diborylated ester 6 on larger scales, we examined
the use of in situ IR spectroscopy (ReactIR), providing an opportunity to examine different reaction
parameters, i.e. time, temperature and ligand effects. Therefore, using the model racemic reaction
(Eqn. 2), the reaction was followed over time at room temperature, to give the results shown in Fig.
3.
Figure 3 ReactIR study of the βꢀborylation reaction on homoallylic boronate ester 5a at RT.
From Fig. 3 it was observed that the signal corresponding to the BꢀO stretch (1126 cmꢀ1) decreased
smoothly along with the consumption of the α,βꢀunsaturated ester 5a (1720 cmꢀ1) and formation of
the product 6a (1734 cmꢀ1) over 3 h. From this reaction, pure product 6a was obtained in a 47%
8
ACS Paragon Plus Environment

Page 9 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
yield; a result which was comparable to that previously reported for the first attempt in the case of
the 16 h reaction (42% IY), and showing that longer reaction times were not required for this
transformation.
8
9
When the reaction was examined at both higher (50 °C) and lower temperatures (0 °C),
respectively, an initial decrease in the B2pin2 concentration took place, however, there was no other
obvious influence of temperature on the reaction time (see Figs. 4 and 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4 ReactIR study of the βꢀborylation reaction on homoallylic boronate ester 5a at 50 °C.
Figure 5 ReactIR study of the βꢀborylation reaction on homoallylic boronate ester 5a at 0 °C.
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 43
1
2
3
4
5
6
7
Finally, the effect of the phosphine ligand was examined, and hence, a diphosphine [(R),(S)ꢀ
Josiphos L4] was compared with triphenylphosphine L1 in order to probe the chiral catalysed
reaction. The results are shown in Fig. 6.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6 ReactIR study of the βꢀborylation reaction on homoallylic boronate ester 5a using (R),(S)ꢀ
Josiphos L4 ligand, replacing PPh3 L1.
Comparing Figs. 3 and 6, it was clear that the presence of a more bulky, bidentate and chiral
(R),(S)ꢀJosiphos L4 ligand did not have a significant influence on the reaction time. However, the
initial consumption of B2pin2 was considerably faster than in the racemic reaction using
triphenylphosphine L1, suggesting that a rather more reactive phosphinylꢀcopperꢀboryl system was
generated. Nonetheless, the subsequent reaction proceeded similarly with high conversion over
approximately 3 h, confirming that the homoallylic boronate carboxylate esters 5 were suitable
substrates for the βꢀboryl addition reaction, providing access to diborylated esters 6.
Control of the relative and absolute 1,3ꢀdiboryl stereochemistry: double diastereoselectivity effects
Once the conjugate addition of boron to homoallylic boronate carboxylate ester 5a was exemplified,
giving a mixture of diastereoisomeric diboryl product 6a (using PPh3 L1 as ligand), we then
10
ACS Paragon Plus Environment

Page 11 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
examined the copperꢀcatalyzed borylation using chiral ligands to provide enantioselective
generation of the CꢀB bond.15 A set of reactions (Scheme 3) were, therefore, carried out using both
enantiomers of DMꢀBinap L2ꢀ3 as ligands, since previously this ligand had provided high e.e.s on
related borylation systems.16 Table 3 shows the results obtained.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3 Borylations of α,βꢀunsaturated aldimines 2a (1st borylation) and homoallylboronate esters
5a (2nd borylation).
Table 3 Evaluation of the diastereoisomeric ratio of 6a depending on the enantiomer of the chiral
ligand used.a
IYc
(%)
35
35
50
Conv.
(%)d
>99
>99
>99
Ligand
Entry
d.r.b
1st Borylation
2nd Borylation
(R)ꢀDMꢀBinap L2
(S)ꢀDMꢀBinap L3
(S)ꢀDMꢀBinap L3
(R)ꢀDMꢀBinap L2
1
2
3
4
(R)ꢀDMꢀBinap L2
(R)ꢀDMꢀBinap L2
(S)ꢀDMꢀBinap L3
(S)ꢀDMꢀBinap L3
4:1
1:1.7
4:1
1:1.7
30
>99
aReaction conditions: see experimental. bDiastereoisomeric ratio of 6a
c
determined by 1H NMR. bIsolated yield on 1,3ꢀdiborylated ester 6a. Determined
by 1H NMR on pure 1,3ꢀdiborylated ester 6a.
Table 3 confirmed that the conversion of imine 2a, generated in situ, followed by the first
asymmetric borylation, hydrolysis, Wittig olefination, and finally a second asymmetric borylation,
provided high conversions and stereocontrol in diboryl compound 6a that was clearly subject to
double diastereocontrol effects.17 Hence, use of the same enantiomer of DMꢀBinap for both
borylation step led to a 4:1 ratio of diastereoisomers 6a, with the first chiral centre matching the
chiral catalyst used for second borylation, whereas a mismatched stereoselection occurred when
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 43
1
2
3
4
5
6
7
8
using different enantiomers for the borylations. Hence, having ascertained that good matched
diastereocontrol was possible by this approach using DMꢀBinap on the cinnamaldehydeꢀderived
model, we were then able to examine the use of a wider range of chiral ligands, since (R),(S)ꢀ
Josiphos L4 and (R),(S)ꢀNMe2ꢀPPh2ꢀMandyphos L6, for example, have found efficient utility for
the asymmetric borylation of copperꢀmediated βꢀborylation on α,βꢀunsaturated esters and nitriles.15b
Hence, ligands outlined in Figure 7 were tested in the sequence shown in Scheme 4, by starting with
a sample 5a derived from a copper(I)ꢀ(R)ꢀDMꢀBinap catalysed first borylation of 2a, i.e. examining
the diastereocontrol for the second borylation step to access 6a, as summarised in Table 4.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4 Asymmetric induction in the synthesis of the diborylated ester 6a.
12
ACS Paragon Plus Environment

Page 13 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7 Phosphine ligands for the Cu(I)ꢀmediated βꢀborylation on homoallylic boronate esters 5.
Table 4 Diastereoisomeric ratio resulting from the borylation of 5a to give the diborylated ester 6a.a
Entry
Ligand
d.r.b
IYc (%)
Conv. (%)d
1
2
3
4
5
6
7
8
(R),(S)ꢀJosiphos L4
(S),(R)ꢀJosiphos L5
(R),(S)ꢀNMe2ꢀPPh2ꢀMandyphos L6
(S),(R)ꢀNMe2ꢀPPh2ꢀMandyphos L7
(R,R)ꢀMeꢀDuPhos L8
(S,S)ꢀMeꢀDuPhos L9
6:1
1:6
3:1
3:1
1.85:1
1.4:1
2.45:1
2:1
50
30
44
24
42
35
32
41
>99
>99
>99
>99
>99
>99
>99
>99
(S,S)ꢀDipamp L10
(R,R)ꢀDipamp L11
aReaction conditions: see experimental. bDetermined by 1H NMR on the
c
diborylated ester 6a pure sample. Determined on the pure diborylated ester 6a.
dDetermined by 1H NMR.
Table 4 shows that the ferroceneꢀbased ligands (Entries 1ꢀ4, Table 4) did indeed provide higher
levels of double diastereocontrol for the introduction of a second boryl unit into an α,βꢀunsaturated
ester 5a, with Josiphos (Entries 1 and 2, Table 4) standing out giving a 6:1 diastereoisomeric ratio,
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 43
1
2
3
4
5
6
7
and notably irrespective of the enantiomer used. Less bulky ligands (Entries 5ꢀ8, Table 4) showed
lower diastereocontrol in general, and some differences between the matched and mismatched
system.
8
9
In order to evaluate the effect of the Cβꢀsubstituent (R1) of the homoallylic boronate ester 5 on the
both the rate of the second borylation reaction (Scheme 4) and subsequent diastereocontrol,
different substrates 5 were subjected to the chiral copper catalysed borylation and initially
monitored by ReactIR to examine reaction progress (see ESI). Table 5 summarises the results
obtained examining different substrates 5, using diphosphine ligands L4 and L5.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 5 Substrate scope for the 2nd βꢀborylation reaction of unsaturated esters 5 (Scheme 5)a
Entry
R1 substituent
on 5
Reaction
d.r.c
(Ligand)
Conv. (%)d
(IY, %)
time (h)b
1
2
3
4
5
6
7
8
9
10
11
12
13
pClPh 5b
pClPh 5b
pClPh 5b
pMeOPh 5c
Me 5d
6
6
6
6
4
4
4
4
4
4
2
2
2
1:1 (L1)
8:1 (L4)
1:5 (L5)
1:1 (L1)
88 (20)
89 (19)
89 (18)
>90% (ꢀ)e
96 (51)
1:1f (L1)
1:1.38f (L4)
1:1.38 f (L5)
1:1f (L1)
1:2.5f (L4)
1.5:1f (L5)
1:1f (L1)
1:11f (L4)
7:1f (L5)
Me 5d
Me 5d
nPr 5e
nPr 5e
nPr 5e
iPr 5f
iPr 5f
iPr 5f
>99 (52)
>99 (42)
>99 (78)
>99 (73)
>99 (70)
>99 (60)
>99 (72)
>99 (69)
aReaction conditions: see experimental. bReaction followed by in situ
ReactIR. cDetermined by pure diborylated ester 1H NMR. dDetermined
by crude diborylated ester 1H NMR. eNot possible to obtain clean
product due to decomposition (see ESI for crude 1H NMR). fDetermined
by 700 MHz 1H NMR using D8ꢀtoluene as solvent.
As shown in Table 5, the linear alkyl substrates 5d and 5e (entries 5ꢀ10, Table 5) performed well in
the second borylation reaction, providing excellent conversions and moderateꢀgood yields, with the
pꢀchlorophenyl system being slower to react and resulting in lower yields (entries 1ꢀ3, Table 5).
Interestingly, use of pꢀmethoxyphenyl (entry 4, Table 5) had an even less beneficial effect on the
reaction, and in fact, it was not possible to obtain a clean sample of the corresponding diborylated
ester 6c. In contrast, asymmetric induction on the new CꢀB bond followed an inverse trend, i.e.
generally higher d.e.s were observed for the aryl substituted substrates (entries 2 and 3, Table 5).
14
ACS Paragon Plus Environment

Page 15 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
Since the alkyl substituent systems 5d and 5e gave faster and cleaner reactions, a more hindered
system was examined, i.e. 4ꢀmethylꢀ2ꢀpentenalꢀderived homoallylic boronate ester 5f, initially
under racemic conditions and monitored by ReactIR. Surprisingly, the reaction was complete in
only 2 h with the 1,3ꢀdiborylated ester 6f being obtained as a 1:1 mixture of diastereoisomers in
60% (entry 11, Table 5). When the second βꢀborylation reaction was carried out using both
enantiomers of Josiphos, i.e. (R),(S)ꢀJosiphos L4 and (S),(R)ꢀJosiphos L5 (entries 12 and 13, Table
5), the diborylated ester 6f was obtained in good yields (72% and 69%, respectively) and with
excellent diastereoisomeric ratios, i.e. 1:11 and 7:1, respectively. Hence, despite the sterically larger
iPr substituent, Josiphos was still able to largely overcome the impact of the existing chiral centre to
give high levels of ligandꢀcontrolled diastereocontrol for the introduction of the second boryl
moiety.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Stereochemical identification
Once it was confirmed that the conjugate addition of second boryl group to the homoallylic
boronate carboxylate esters 5 was possible and that moderate to excellent diastereocontrol was
possible, it was necessary to confirm the identity of the diastereoisomers, and therefore separation
through transformation to suitable derivatives was required. Diastereoisomers of 6a were examined
as the test substrate using two strategies: 1) transformation of 6a into the sixꢀmembered ring acetals
8a or 9a; and 2) formation of the diethanolamineꢀboron complex 10a (see Scheme 5).
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5 Synthetic pathways studied to obtain cyclic structures that allow the determination of the
relative stereochemistry.
Firstly, the strategy involving a sixꢀmembered ring acetal 8a formation was examined through a
double BꢀC bond oxidation, leading to the 1,3ꢀdiol intermediate 7a, followed by acetal
incorporation. Initially, we decided to examine nonꢀaqueous conditions to avoid water soluble
products being lost upon work up by using trialkylamine Nꢀoxides, which have been widely
demonstrated18 to oxidise the BꢀC bond in a range of systems (Eqn. 3). Application of 4ꢀ
methylmorpholineꢀNꢀoxide (4ꢀMMNO) to access the 1,3ꢀdiol intermediate 7a from 6a resulted in
the formation of the diol, however, after purification by silica gel chromatography, of the resulting
1,3ꢀdiol 7a was complicated by the high loadings of the oxidising agent (8.0 equiv.) used. Use of
16
ACS Paragon Plus Environment

Page 17 of 43
The Journal of Organic Chemistry
1
2
trimethylamine Nꢀoxide (TMANO)19 proved to be superior being readily used in lower loadings and
3
4
5
the products more readily purified, as summarised in Table 6.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 6 Oxidation conditions to obtain 1,3ꢀdiol 7a from the diborylated ester 6a.a
Entry NꢀOxide (equiv.) Reaction Temp. (°C) Isolated yieldb
time (h)
6
16
6
3
(%)
52 c
23c
26c
60d
1
2
3
4
4ꢀMMNO (2)
4ꢀMMNO (4)
TMANO (2)
TMANO (2)
RT
RT
RT
50
aReaction conditions: see experimental. bIsolated yield of pure diol
7a. cFor full conversion (>90%) longer reaction times (up to 32 h) as
well as addition of extra oxidising agent (2 or 4 equiv.) were
required. dReaction stirred during additional 3 h to ensure full
completion.
Table 6 shows the suitability of TMANO as oxidant, especially by running the reaction at 50 °C
(entry 4, Table 6), under which conditions, the racemic diboronate 6a could be successfully
transformed into the target diol 7a in a 60% yield. However, when this methodology was tested on
isopropyl substituted analogue diboryl ester 6f (Eqn. 4), the reaction was not as effective as it was
for the model substrate 6a and longer reaction times, as well as larger amounts of TMANO (up to
48 h and 8.0 equivalents, respectively) were required. Hence, a more general oxidative
transformation needed to be identified, and further oxidants were examined, as outlined in Table 7
using the more hindered substrate 6f.
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 43
1
2
Table 7 Conditions for the oxidation of diborylated ester 6f.a
3
4
Entry Oxidant reagent (equiv.)
Solvent
Conv.ꢀ6fb (%)
5
6
1
2
NaBO3.4H2O (6.0)
Oxone (6.0)
THF/H2O
MeOH
>99
0
7
8
9
3
MCPBA (6.0)
DCM
>99
aReaction conditions: For a 0.2 mmol reaction scale, stirred at RT
during 2h, monitoring by TLC. bDetermined by crude 1H NMR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For diboryl system 6f, NaBO3.4H2O (entry 1, Table 7) was found to be an ideal reagent for the
oxidation of the two CꢀB bonds.20 In contrast, Oxone surprisingly21 did not yield into the desired
1,3ꢀdiol 7f, whereas MCPBA22 gave complete conversion to the desired diol 7f, however, the
reaction was not as clean as with sodium perborate.
With the oxidation step optimised using sodium perborate, the acetal formation was then studied to
determine the optimal conditions for this transformation (Eqn. 5). Benzaldehyde23 was initially
examined, as outlined in Table 8 for the synthesis of acetal 8a from diol 7a.
Table 8 Conditions for the synthesis of acetal 8a from the diol 7a.a
Conv. 8ab
PhCHO
(equiv.)
1.1
Temp.
(°C)
0
Reaction
Time (h)
1.5
Entry
3 ÅꢀMS
(%)
1
2
3
4
5
6
7
8
No
No
ꢀ
24
20
25
ꢀc
2.5
RT
RT
50
1.5
1.1
Yes
No
1.5
1.1
1.5
1.1
No
RT
50
1.5
2.5
Yes
Yes
Yesd
1.5
50
ꢀc
2.5
1.5d
50
1.5
50
6
>99
aReaction conditions: see experimental. bDetermined by crude 1H NMR
c
d
on the acetal 8a. Toluene was used as solvent. PhCH(OMe)2 used
instead of PhCHO, using 4 Å M.S.
18
ACS Paragon Plus Environment

Page 19 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
Using benzaldehyde under a wide variety of reactions conditions (entries 1ꢀ7, Table 8), formation of
the acetal proved problematic, due to lack of clean reaction. However, use of benzaldehyde
dimethyl acetal readily provided the target acetal 8a (entry 8, Table 8) starting from a mixture of
diastereoisomers. Hence, we then examined the diborylated ester 6a, obtained from (R),(S)ꢀJosiphos
L4 and (S),(R)ꢀJosiphos L5 catalyzed borylation reactions.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Using (R),(S)ꢀJosiphos L4 as ligand, the acetal 8a was obtained in good conversion (i.e. >99%) as a
4:1 mixture of diastereoisomers (Scheme 6) which was purified by silica gel chromatography, to
give the major diastereoisomers with less than 5% of the minor (a mixed fraction was alos obtained,
with these fractions representing 15 and 55% yields, respectively). NMR analysis (see ESI) allowed
the determination of the relative stereochemistry of the major diastereosiomer which corresponded
to the antiꢀdiastereoisomer (Scheme 6).
O
O
B
O
B O
O
B
B2pin2 (1.0 equiv.),
O
CO2Me
Ph
CuCl (3 mol%), (R),(S)-Josiphos L4
CO2Me
5a
Ph
(3 mol%), NaOtBu (9 mol%),
6a
i
PrOH, RT, 16 h
TMANO (2.0 equiv.),
CHCl3, 50 °C, 3 h
62%
51%, 4:1 d.r.
Ph
O
OH OH
PhCH(OMe)2 (1.5 equiv.),
O
CO2Me
CO2Me
TsOH (10 mol%), 4 Å-MS, Ph
50 °C, 5 h
70% (mixture of diastereoisomers)
Ph
7a
8a
Major diastereoisomer analysis:
H2
Ph
H1
O
O
O
H2'
B
B
O
Ph
O
O
O
O
11.6 Hz
CO2Me
CO2Me
Ph
H3
Ph
Ph
6a
MeO2C
8a
Anti-diastereoisomer
Scheme 6 Coupling constants for the antiꢀconfiguration of the 6ꢀmembered ring acetal 8a.
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 43
1
2
3
4
5
6
7
8
Surprisingly, when NMR analysis was carried out on what was thought to be the minor
diastereoisomer, instead of the expected synꢀdiastereoisomer, a different acetal configuration of the
antiꢀdiastereoisomer was discovered (see ESI). These different diastereoisomeric acetals likely
interconvert under the reaction formation conditions, with the equilibrium ratio of diastereosiomers
I and II being biased by minimisation of 1,3ꢀdiaxial interactions of the phenyl groups (Scheme 7).
The interconversion between these diastereoisomers can take place due to ring opening/ring closing
of the acetal in the presence of catalytic amounts of acid during their formation. In this case the
benzylidene acetal position changes as results of the opening of this group and generation of the
oxonium ion which subsequently closes, as outlined in Scheme 8. Hence, a thermodynamic ratio of
acetal configurations is expected. Indeed, subsequent reꢀexamination of the crude 1H NMR showed
a 1:1 ratio of acetals I and II of 8a (see Scheme 7).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ph
Ph
O
O
O
O
ring opening/closing
(H+ cat.)
CO2Me
Ph
CO2Me
Ph
Ph
8a II
8a I
O
CO2Me
O
Ph
O
Ph
Ph
MeO2C
O
H+
H+
MeO2C
Ph
O
OH
Ph
Ph
O
Ph
OH
MeO2C
Scheme 7 Acetal configuration interconversion for the 6ꢀmembered ring phenylꢀsubstituted acetal
8a.
Similar derivatization studies on the isopropyl substituted system 6f was also examined due to its
excellent diastereoselectivity (up to 1:11 and 7:1, for (R),(S)ꢀJosiphos L4 and (S),(R)ꢀJosiphos L5,
respectively) as outlined in Scheme 8.
20
ACS Paragon Plus Environment

Page 21 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
O
O
B
O
O
B2pin2 (1.0 equiv.),
O
B
B O
CO2Me
i
Pr
CuCl (3 mol%), (R),(S)-Josiphos L4
(3 mol%), NaOtBu (9 mol%),
CO2Me
5f
i
7
Pr
6f
8
9
i
PrOH, RT, 16 h
72%, 1:11 d.r.
TMANO (2.0 equiv.),
CHCl3, 50 °C, 3 h
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ph
O
OH OH
PhCH(OMe)2 (1.5 equiv.),
CO2Me
O
CO2Me
i
TsOH (10 mol%), 4 Å-MS, Pr
iPr
7f
50 °C, 5 h
64% after two steps
8f
Major diastereoisomer analysis:
Ph
H1
CO2Me
9.1 Hz
O
O
O
i
Pr
Ph
Ph
O
H2
CO2Me
CO2Me
i
O
O
O
Pr
H3
H2'
8f I
B
B
O
O
+
Ph
CO2Me
i
Pr
6f
O
Anti-diastereoisomer
O
O
iPr
i
Pr
MeO2C
8f II
Scheme 8 Coupling constants for one of the antiꢀconfiguration 6ꢀmembered ring acetal analogues,
8f I.
It was interesting to note that the oxidation of 6f was found to be slower in comparison to the Phꢀ
substituted system, for both the TMANO and NaBO3.4H2O conditions. However, the target chiral
acetal 8f was successfully obtained in good yields from two samples, derived from different
reactions, i.e. 64% and 70% having used ligands L4 and L5, respectively (see Scheme 8). For these
acetal samples, preparative HPLC purification was required in order to obtain a diastereomerically
pure sample suitable for NMR analysis. Hence, using chiral ligand L4 during the borylation, giving
a 1:11 d.r. (see ESI for structural analysis) acetal 8f, HPLC separation resulted in a pure sample of
1
acetal 8f I (Scheme 7) and a mixed sample of both 8f I and II. By comparing the H NMR spectra
of both samples (see ESI), the relative stereochemistry and assigment of each, could be made, as
indicated by the key coupling constant shown in Scheme 7.
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 43
1
2
3
4
5
6
7
Due to the complications of acetal stereochemistry at the benzylidene centre, the use of an acetonide
acetal was considered to potentially simplify the stereochemical analysis protocol. Acetonides have
13
been widely reported for the protection of 1,2ꢀ and 1,3ꢀdiols24 and in fact, C NMR analysis has
8
9
proved an excellent tool for the determination of the relative stereochemistry in such systems.25
Rychnovsky et al. reported26 a method based on the different chemical shifts for the methyl groups
of the acetonide functionality, depending on their relative orientation and acetal ring conformation;
a method which has proved effective for the assignment of relative configurations, not only for
diols, but also for polyols and polyene macrolides. Based on this approach, the synthesis of the
acetonide acetal 9a was approached (Scheme 9).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 9 Synthesis of the chiral acetonideꢀsubstituted 6ꢀmembered ring acetal 9a.
Hence, 1,3ꢀdiol 7a was prepared as outlined above, using either (R),(S)ꢀJosiphos L4 or (S),(R)ꢀ
Josiphos L5 as chiral borylation ligand, to give the 6ꢀmembered ring acetals 9a in moderate isolated
yields (32% and 20%, respectively). In both cases, the acetal was obtained as a mixture of
diastereoisomers (4:1 and 2:1, respectively); a lower level of diastereoselectivity than expected
22
ACS Paragon Plus Environment

Page 23 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
suggesting loss of one of the diastereoisomers during the oxidation/acetal/purification sequence and
returning the low yields. Hence, in order to improve yields and minimise losses, a oneꢀpot method
avoiding the chromatography between each step was carried out. While yields did not improve, the
target acetals in isolated in a consistent ratio to those obtained previously for the precursor
diborylated esters 6a, i.e. 6:1 (23% IY) and 1:4 (12% IY), respectively. Moreover, separation by
preparative chiral HPLC provided clean samples of the major diastereoisomer in both cases to allow
1H NMR and 13C NMR analysis of the acetonides to be carried out (see ESI).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For the assignment of the relative stereochemistry for the case of using (R),(S)ꢀJosiphos L4 for the
synthesis of the diborylated ester 6a, the signals corresponding to Hꢀ2 and Hꢀ2’ (δ 2.10ꢀ1.95 ppm)
appeared as ddds in both cases, with two different 3J couplings in each case; i.e. 9.8 and 5.8 Hz for
Hꢀ2, and 9.3 and 6.3 Hz for Hꢀ2’, as well as the expected germinal coupling between these protons
(13.1 Hz). Figure 8 presents an analysis of the couplings for Hꢀ2 of acetal 9a, showing a large
coupling (9.8 Hz) to Hꢀ1 indicating that these protons are transꢀdiaxial, whilst the other 3J coupling
of Hꢀ2 corresponds to a cisꢀcoupling to Hꢀ3 (9.8 Hz). The analysis of these signals indicates that the
6ꢀmembered ring acetal 9a is in a twisted boat conformation which corresponds to the antiꢀ
diastereoisomer. Furthermore, in order to confirm this observation, the NOESY spectrum showed a
transꢀring coupling between Hꢀ1 and one of the methyl groups from the acetonide functionality
(signal at δ 1.42 ppm) as well as between Hꢀ3 and the other methyl group on the acetonide (signal at
δ 1.45 ppm) (see ESI).
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 43
1
2
3
4
5
6
5.8 Hz
Jcis
H
H2
9.8 Hz
Jtrans
H3
O
O
Ph
7
8
H1
CO2Me
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
O
O
O
CO2Me
CO2Me
Ph
Ph
H1 H3
H2
9a
Anti-diastereoisomer
Figure 8 Coupling constants analysis for the antiꢀdiastereoisomer of 6ꢀmembered ring acetal 9a.
In order to complete this study, the same analysis was carried out on another sample of chiral acetal
9a, however, in this case, the chiral diphosphine ligand employed for the 2nd βꢀborylation reaction
was (S),(R)ꢀJosiphos L5 (see ESI). The most relevant feature of the resulting NMR spectrum was
the multiplicity observed for Hꢀ2, which unlike for the case of using L4, this proton appeared as a dt
(instead of a ddd) indicating that in this case, the synꢀdiastereoisomer was obtained. In this case, the
6ꢀmembered ring acetal adopts a more chairꢀlike conformation meaning that Hꢀ2 couples with Hꢀ1
giving a doublet with a small coupling constant of 2.6 Hz, as well as coupling with Hꢀ4 and Hꢀ4’
giving a triplet with a large coupling constant of 15.4 Hz and overall resulting in the observed dt as
shown in Figure 9. In addition, a second double triplet was expected (which would correspond to Hꢀ
2’) which could not be observed in the 1H NMR spectrum (see ESI) because it appeared in the same
region where larger signals were observed (δ 1.6ꢀ1.4 ppm). However, these samples were also
analysed by the [13C]ꢀacetonide method confirming the relative stereochemistry; when using
13
(R),(S)ꢀJosiphos L4 as the ligand, the C NMR spectrum showed the two methyl signals at 24.9
and 24.6 ppm of the acetonide, meaning that in this case, these two groups were in a similar
chemical environment, indicating conformational flexibility due to interconversion between chairꢀ
like and twistꢀboatꢀlike conformations and due to the antiꢀdiastereoisomeric configuration. To
13
confirm this, the opposite enantiomeric ligand L5 product gave a C NMR spectrum showing two
24
ACS Paragon Plus Environment

Page 25 of 43
The Journal of Organic Chemistry
1
2
distinct methyl signals at δ 30.53 (equatorial) and δ 20.10 (axial) ppm (see ESI) clearly indicating
3
4
5
the presence of the synꢀdiastereoisomer (Figure 9).
6
7
H-2
H1
H3
8
9
J 15.4 Hz
(d)
Ph
H2
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
CO2Me
H2'
J 2.1 Hz
(t)
dt
Figure 9 Conformational analysis of the synꢀdiastereoisomer of acetal 9a.
Once the relative stereochemistry was elucidated for the phenyl systems outlined above, these
results needed to be corroborated by studying an additional substrate, i.e. the isopropyl substituted
analogue (Eqn. 6).
In this case, although the efficiency of the reaction differed depending upon the enantiomer of the
chiral ligand used for the βꢀborylation reaction on the homoallylic boronate ester 6f; i.e. from a low
yield in the final acetal for the case of L4 (10%) to a moderate yield for its opposite enantiomer
(37%), a matching/mismatching effect was observed (Scheme 10).
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 10 Synthesis of the chiral acetonideꢀsubstituted 6ꢀmembered ring acetal 9f.
The low yield could be associated to the poor effectivity presented by the CꢀB oxidation reaction for
these systems as previously reported. Nevertheless, the pure acetals (obtained as mixture of
diastereoisomers after column chromatography) were further purified by preparative HPLC, being
possible to develop the study of the J coupling constants and the [13C] acetonide method on samples
containing exclusively the major diastereoisomer for each case.
According to the [13C] acetonide method the use of (R),(S)ꢀJosiphos ligand L4 led to the antiꢀ
diastereoisomer; methyl groups at δ 24.95 ppm and 24.55 ppm, i.e. similar chemical environment,
whilst the use of (S),(R)ꢀJosiphos L5 lead into the synꢀdiastereoisomer; one methyl in equatorial
position (δ 30.02 ppm) and the other methyl in the axial position (δ 19.66 ppm) (see ESI).
Hence, it was concluded that the nucleophilic boryl unit adds into the homoallylic boronate ester
substrate either from the same face were the first boryl unit is already allocated or from its opposite
face upon the enantiomer of the diphosphine ligand used for the generation of this reactive boryl
26
ACS Paragon Plus Environment

Page 27 of 43
The Journal of Organic Chemistry
1
2
3
4
unit, as displayed in Figure 10. Hence, the stereochemical control being predomoninatly ligand
controlled, but with some double diastereocontrol tining the final diastereoisomeric ratios.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10 Stereoselectivity observed upon the addition of the 2nd boryl unit into chiral homoallylic
boronate esters based on the enantiomer of the chiral diphosphine ligand used.
Complimentarily, another strategy for determining the relative stereochemistry was approached; the
synthesis of the diethanolaminoꢀboron complex 10a, with the hope that Xꢀray crystallography
would be useful to confirm the stereochemistry. After years of research and with several studies,27
the facility for forming coordinating bonds with its neighbouring elements, e.g. N or O, exhibited
by acidic tricoordinated sp2 boron atoms has been demonstrated. Taking into account this
interesting property, it was envisioned that the reaction of the diborylated ester 6a with
diethanolamine28 could afford the formation of the bicyclic structure stabilised by the BꢀN
interaction (Eqn. 7).
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The diborylated ester 6a was treated with diethanolamine (6.0 equiv), which after azeotropic
removal of the pinacol with toluene and extraction of the remaining excess of diethanolamine to
give a semiꢀcrystalline material. Further attempts to crystallise from diethyl etherꢀDCM resulted in
an amorphous solid not suitable for Xꢀray diffraction. Alternative methodologies, such as
distillation in a Kugelrohr for the removal of pinacol as well as the diethanolamine in excess were
examined. It was not possible to obtain the desired complex due to possible decomposition
processes associated to the distillation. Moreover, Nꢀsubstituted diethanolamine (e.g. Nꢀ
methyldiethanolamine)29 was also evaluated for the BꢀN complex formation (Eqn. 8), being not
possible to obtain compound 11a.
Conclusions
In summary, unsaturated benzhydryl imines 2 are readily borylated efficiently and with high e.e. to
derive boryl imines 3 on >99% e.e. using DMꢀBinap L2. The intermediate
βꢀboryl imines are
inherently unstable, however, an in situ hydrolysisꢀWittig trapping protocol was developed that
provided the corresponding homoallylic boronate carboxylate esters 5, which were confirmed as
ideal substrates for the βꢀborylation reactions (see summary Table 9). The resulting diboryl esters 6
could be readily isolated and hence, we were able to examine the relative stereocontrol of the
28
ACS Paragon Plus Environment

Page 29 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
second boryl chiral stereocentre through a double diastereocontrolled approach. For the introduction
of the second boryl group, Josiphos ligands proved superior to other systems, showing interesting
double diasteroselective effects, which in broad terms showed that the original chiral centre
chirality could be overridden so that any particular diastereoisomeric combination is accessible with
reasonable to high diastereocontrol. There were clearly observable effects from the different
substituents on Cβ with alkyl substitued substrates (5d and e) providing lower diastereocontrol. The
more hindered isopropylꢀsubstituted system 5f, however, gave an 11:1 (matched) versus 1:7
(mismatched) diastereocontrol with Josiphos, to give the corresponding bisꢀboryl esters efficiently.
Interesting, the arylꢀsubstituted systems showed moderate to good diastereocntrol, though the final
bisꢀboryl compounds varied in terms of their chemical stability, with the more electron rich pꢀ
methoxy system 6b being particularly unstable; the origin of which is not clear. The relative
stereochemistry of the bisꢀboryl esters 6 was not easily examined by conversion to potentially
crystalline diethanolamine analogues 11, however, oxidative CꢀB cleavage and conversion to 1,3ꢀ
diolꢀderived acetals did provide relative stereochemical confirmation. Further application of this
stereochemical control is underway for the synthesis of bioactive, 1,3ꢀdihydroxylated derivatives
and will be reported in due course.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
29
ACS Paragon Plus Environment