Speciation control during Suzuki-Miyaura cross-coupling of haloaryl and haloalkenyl MIDA boronic esters

Article abstract of DOI:10.1002/chem.201500970

Boronic acid solution speciation can be controlled during the Suzuki-Miyaura cross-coupling of haloaryl N-methyliminodiacetic acid (MIDA) boronic esters to enable the formal homologation of boronic acid derivatives. The reaction is contingent upon control of the basic biphase and is thermodynamically driven: temperature control provides highly chemoselective access to either BMIDA adducts at room temperature or boronic acid pinacol ester (BPin) products at elevated temperature. Control experiments and solubility analyses have provided some insight into the mechanistic operation of the formal homologation process.

Full text of DOI:10.1002/chem.201500970

DOI: 10.1002/chem.201500970
Full Paper
&
Cross-Coupling Reactions
Speciation Control During Suzuki–Miyaura Cross-Coupling of
Haloaryl and Haloalkenyl MIDA Boronic Esters**
James W. B. Fyfe,[a] Elena Valverde,[b] Ciaran P. Seath,[a] Alan R. Kennedy,[a]
Joanna M. Redmond,[c] Niall A. Anderson,[c] and Allan J. B. Watson*[a]
Abstract: Boronic acid solution speciation can be controlled
during the Suzuki–Miyaura cross-coupling of haloaryl N-
methyliminodiacetic acid (MIDA) boronic esters to enable
the formal homologation of boronic acid derivatives. The re-
action is contingent upon control of the basic biphase and
is thermodynamically driven: temperature control provides
highly chemoselective access to either BMIDA adducts at
room temperature or boronic acid pinacol ester (BPin) prod-
ucts at elevated temperature. Control experiments and solu-
bility analyses have provided some insight into the mecha-
nistic operation of the formal homologation process.
Introduction
The development of protected boronic acids has been pivotal
to the growth of iterative Suzuki–Miyaura cross-coupling pro-
cesses.[1,2] In particular, the boronic esters (N-coordinated boro-
nates) derived from N-methyliminodiacetic acid (BMI-
DA)[2a,b,d,3–5] and the aminoboranes derived from 1,8-diamino-
naphthalene (BDAN)[2c,d,6,7] are readily installed and removable
protecting groups that render iterative Suzuki–Miyaura cross-
coupling relatively facile: following a first cross-coupling event,
protecting group hydrolysis under basic (BMIDA) or acidic
(BDAN) conditions liberates the reactive parent boronic acid,
primed for further cross-coupling (Figure 1a).
Similar to the majority of methods for the preparation of re-
active boron species, these chemistries proceed via stoichio-
metric and step-wise manipulation of a single reactive boron
species,[8] such as a boronic acid. Conversion of a protected
boronic acid to an alternative reactive boron species, such as
a boronic acid pinacol ester (BPin) typically proceeds by the
same synthetic pathway; conversion of BMIDA to BPin requires
Figure 1. (a) Iterative cross-coupling using protected boronic acids. (b) Cross-
coupling of BMIDA followed by conversion to BPin. (c) Formal BPin homolo-
gation by controlled speciation. DAN=1,8-diaminonaphthalene; MIDA=N-
methyliminodiacetic acid; Pin=pinacol/pinacolato.
[a] J. W. B. Fyfe, C. P. Seath, Dr. A. R. Kennedy, Dr. A. J. B. Watson
WestCHEM, Department of Pure and Applied Chemistry
University of Strathclyde
hydrolysis and subsequent esterification (Figure 1b).[4a,d] We re-
cently demonstrated that it is possible to convert a BMIDA
ester to a BPin ester during the Suzuki–Miyaura cross-coupling
of a haloaryl BMIDA with an aryl BPin.[9] This is achieved
through pinacol recycling by control of multiboron solution
295 Cathedral Street, Glasgow G1 1XL (UK).
E-mail: allan.watson.100@strath.ac.uk
[b] E. Valverde
Laboratori de Química Farmac›utica
Facultat de Farmàcia and Institute of Biomedicine (IBUB)
Universitat de Barcelona
speciation, leading to
(Figure 1c).
a
formal sp2 BPin homologation
Av. Joan XXIII, Barcelona, 08028 (Spain).
Here, we provide the full details of this study, demonstrat-
ing: (i) the dependence of the reaction on pH as well as the
physical properties of the base; (ii) that the chemoselectivity of
boron speciation can be thermodynamically controlled to pro-
vide selective access to either BMIDA or BPin products; and (iii)
that the general concept of speciation control is transferrable
across boronic acids, BPin esters, and catechol esters. We also
[c] Dr. J. M. Redmond, Dr. N. A. Anderson
GlaxoSmithKline
Medicines Research Centre
Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2NY (UK).
[**] MIDA=N-methyliminodiacetic acid.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500970.
Chem. Eur. J. 2015, 21, 8951 – 8964
8951
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
provide an analysis of the parameters resulting in effective spe-
ciation control for this transformation and insight into the
issues surrounding anomalous reactions.
readily achieved with either NaOH or K3PO4.[4e,5d,g,q] Hydrolysis
of boric acid esters, such as 4, is similarly facile under aqueous
basic conditions.[12] The final esterification of 5 with pinacol is
also typically a high yielding and rapid process under a variety
of conditions ranging from acidic to basic.[12,13] Based on this,
steps (ii)–(iv) could all tentatively be controlled by using an ap-
propriate aqueous basic medium.
Results and Discussion
Boronic acids and esters are known to exhibit complex and dy-
namic solution speciation equilibria.[10] Chemoselective control
of boronic acid solution speciation comprising a mixture of
boron species may therefore be expected to be difficult based
on the requirement to simultaneously manipulate interlinked
equilibria. Accordingly, the preparation of synthetically useful
boron species, such as boronic acids and esters, is typically
performed by manipulation of a single boron component to
avoid possible difficulties arising from these equilibria, poten-
tially leading to mixtures of products.[10,11] However, exerting
control over the equilibria associated with multiboron systems
may provide useful and more efficient methods for the prepa-
ration of valuable boron reagents without resorting to the pos-
sibly more laborious single-molecule manipulations that are
common throughout this preparative area.
However, aqueous base is incompatible with the first reac-
tion event owing to the base lability of BMIDA esters.[4e,5d,g,q]
Cross-coupling of BMIDA-containing compounds is typically
performed under anhydrous conditions to avoid hydrolysis. In
the envisioned process in Scheme 1, premature hydrolysis of 2
or 3 would lead to 5 and/or 7, both of which may undergo un-
controlled oligomerization to 8 and/or 9 (Scheme 2).
We sought to explore this idea in the context of Suzuki–
Miyaura cross-coupling by using two different boronic esters,
specifically BPin esters (1) and haloaryl BMIDA esters (2), with
the goal of ascertaining whether the boron speciation may be
controlled during the reaction to produce a new BPin ester
and thereby establishing a formal homologation process that
would offer increased step efficiency over conventional
approaches.[9]
Scheme 2. Oligomerization of haloaryl BMIDA species during Suzuki–Miyaura
cross-coupling owing to premature in situ hydrolysis.
In addition, the reaction would need to be staged appropri-
ately to avoid cross-coupling conflict owing to the similarities
in reactivity profiles of starting material 1, intermediate boronic
acid 5, and product 6 towards cross-coupling.
The overall reaction was envisaged to take place via four ele-
mentary steps (Scheme 1): (i) CÀC bond formation resulting
Design Plan
To reconcile the requirement for anhydrous conditions during
cross-coupling and the aqueous basic conditions that would
facilitate control over the subsequent reaction events, we
sought to establish an internal water reservoir. This would be
achieved by exploiting the physical properties of the inorganic
bases typically associated with Suzuki–Miyaura cross-cou-
pling.[14] Many of these bases are hygroscopic and generate
stable hydrates. In contrast to the majority of Suzuki–Miyaura
reactions, which employ relatively large quantities of H2O
(commonly 4:1–7:1),[1e] addition of a controlled quantity of H2O
to a suitably hygroscopic inorganic base was proposed to se-
quester H2O and safeguard BMIDA integrity during cross-cou-
pling while simultaneously providing sufficient H2O and base
within the reaction mixture to facilitate the downstream hydro-
lytic and esterification events.
Scheme 1. Proposed formal homologation of aryl BPin by controlled boron
speciation during Suzuki–Miyaura cross-coupling of haloaryl BMIDA esters.
from conventional Suzuki–Miyaura cross-coupling to generate
an intermediate product BMIDA 3; (ii) hydrolysis of 3 to the
parent boronic acid 5; (iii) hydrolysis of the Suzuki–Miyaura by-
product HOÀBPin 4 to liberate pinacol; and (iv) esterification of
5 with the in situ generated pinacol to deliver the desired, for-
mally homologated, product 6.
Each of the elementary steps are theoretically straightfor-
ward and are supported by studies from other research
groups: cross-coupling of aryl BPin 1 with haloaryl BMIDA 2 to
deliver the BMIDA 3 is typically a high yielding process.[4f] The
subsequent hydrolysis of 3 to the parent boronic acid 5 is
Accordingly, we began by exploring a benchmark Suzuki–
Miyaura cross-coupling reaction between phenylboronic acid
pinacol ester (BPin) 10 and 4-bromophenyl BMIDA 11 a using
a common Pd catalyst ([PdCl2dppf], dppf=1,1’-bis(diphenyl-
phosphino)ferrocene) in THF using two typical inorganic bases,
K3PO4 and Cs2CO3, in conjunction with a comparatively restrict-
ed quantity of H2O (10:1) compared with typical Suzuki–
Miyaura reactions (Table 1).
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8952
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Systematic H2O Evaluation
Table 1. Initial reactions with K3PO4 and Cs2CO3 using 10:1 THF/H2O.
We first evaluated the quantity of H2O added to the reaction
using K3PO4 as the base. In terms of hygroscopicity, K3PO4 is
known to form a stable tetrahydrate[14] and so was expected to
support a specific quantity of H2O. However, the availability of
this ‘captured’ H2O was unknown. In addition, based on the
envisioned solution processes taking place (Scheme 1), as the
reaction progresses, boric acid will accumulate and may con-
dense to release additional H2O.[14] This may be promoted by
a desiccant, such as K3PO4. Therefore, the exact quantity of
H2O available within the reaction at any stage was uncertain.
As such, we undertook a comprehensive H2O evaluation
(Scheme 3 and Figure 2).
Entry
Base
T [8C]
6a/3a/5a/12 [%][a]
1
2
3
4
K3PO4
Cs2CO3
K3PO4
50
50
90
90
57:13:7:0
52:6:7:0
30:0:0:70
27:0:0:73
Cs2CO3
[a] Determined by HPLC analysis.
The results of the reactions at 508C and 908C were highly
positive for both bases employed. At the more moderate 508C
(entries 1 and 2), good conversion to product was observed
(approx. 50%) with some of the intermediate boron species
also detected—BMIDA 3a was observed in 6–13% yield, with
the parent boronic acid 5a at
Scheme 3. Evaluation of H2O and its effect on conversion to 6a.
7% yield in both cases. Pleasing-
ly, no oligomerization was ob-
served, but conversion was in-
complete at approximately 70%.
Increasing the temperature of
the reaction to 908C (entries 3
and 4) provided complete con-
version of starting material, rela-
tively low conversion (approx.
30%) to the desired product 6a,
with the mass balance consisting
of undefined oligomeric material
(12).[15]
We believed that the large
degree of oligomerization was
Figure 2. Experimental response surface: conversion to 6a versus H2O equivalents/time for the formal homologa-
tion. Yields determined by HPLC analysis.
due to a more rapid hydrolysis
of the BMIDA starting material
11 a and/or intermediate 3a at
this higher temperature: BMIDA
compounds are readily hydrolyzed in the presence of aqueous
base and this can proceed rapidly with strong bases (e.g., with
NaOH) or more slowly with weaker bases (e.g., with
K3PO4).[4e,5d,g,q] Burke and co-workers employed K3PO4-mediated
slow hydrolysis of BMIDA as a method to facilitate the cross-
coupling of notoriously sensitive boronic acids through a slow-
release protocol.[5h,o] For the reaction in Table 1, although the
conversion to product was greater at 508C, overall conversion
was greater at 908C. Based on this, we elected to pursue opti-
mization at 908C as we believed that an appropriately bal-
anced basic biphase[1d,e] would mitigate premature BMIDA hy-
drolysis and thereby eliminate the oligomerization issue.
As expected, the conversion to 6a was highly dependent on
the level of H2O added to the system. The response surface in
Figure 2 displayed four main regions in which the reaction
could be predicted to deliver specific outputs.
(1) Using 0 equivalents of H2O: cross-coupling was found to
be very inefficient with only modest levels of product ob-
served (<60%) and extended reaction times failing to pro-
vide any increase. No oligomerization was detected and
the mass balance was principally unreacted starting
material.
(2) Using 1–15 equivalents of H2O: when the reaction was al-
lowed to take place over 24 h, excellent levels of conver-
sion to 6a could be obtained (up to 92% at 5 equivalents
H2O) with no oligomerization and complete consumption
of starting material. Shorter reaction times resulted in
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8953
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
lower conversions to 6a with variable levels of intermedi-
ate boron species 3a and 5a detected.
and K3PO4 (entries 2–4), cross-coupling efficiency immediately
improved and speciation control was also possible, with con-
version reaching an optimum in the presence of K3PO4.
Starting material consumption was incomplete for KOAc and
K2CO3. Interestingly, for entries 2–4, 3a and 5a were not de-
tected—all the Suzuki–Miyaura product was converted to the
BPin adduct 6a. Therefore, under these conditions, the overall
reaction efficiency becomes entirely dependent on cross-cou-
pling efficiency. Use of KOH and KOtBu delivered incomplete
conversion of starting materials, poor conversion to product
(approx. 20%), and extensive oligomerization, presumably
owing to rapid hydrolysis of BMIDA (entries 5 and 6). The rela-
tionship between pKa and conversion to 6a is clearly demon-
strated by entries 1–6.
(3) Using 15–25 equivalents of H2O: conversion to products,
intermediates or byproducts was unpredictable and varia-
ble depending on the reaction time—shorter times ap-
peared to enable good levels of conversion to 6a, with oli-
gomerization increasing markedly as the reaction time in-
creased, potentially indicating that cross-coupling of 11 a
was inefficient allowing further reaction of 11 a with 6a or
3a.
(4) Using >25 equivalents of H2O: poor, but consistent levels
of conversion to 6a (approx. 20–30%) were observed
throughout with the mass balance composed of oligomer-
ic material, indicating poor control of the rate of BMIDA
hydrolysis.
However, the relationship between base and reaction effi-
ciency is not as straightforward as bases of similar pKa were
found to provide starkly different results. KH2PO4 (entry 7), of
similar pKa to KTFA, also provided no conversion. In contrast,
K2HPO4 (entry 8) provides no conversion although this has
a similar pKa to KOAc, which provides 37% yield of 6a. Conse-
quently, the reaction is not solely dependent upon pKa (or the
resultant solution pH) although this is clearly highly important.
This observation was confirmed when the effect of the
metal countercation was evaluated. Tribasic phosphate ap-
peared to be optimum for the reaction, but the effect of varia-
tion of the associated metal ion—alkali metals or alkaline
earths—was surprising (Table 3).
Based on this evaluation, we selected 5 equivalents of H2O
to move forward. This was chosen as it provided excellent
levels of conversion as well as providing a tolerance for any ad-
ditional H2O arising from a less stringent reaction set up.
Base Evaluation
With a functional knowledge of H2O influence, we next evalu-
ated the role of the base. Different bases were predicted to
have broadly different impacts on the reaction. Amatore and
Jutand have demonstrated the triple role of HOÀ in the
Suzuki–Miyaura reaction; affecting oxopalladium formation,
boron solution equilibria, as well as reductive elimination.[16]
These authors also demonstrated that different metal cations
also affect Suzuki–Miyaura cross-coupling.[17] For the reaction
under development, in addition to the expected effects de-
tailed by Amatore and Jutand, variation of physical properties
was expected to have a profound impact.
Table 3. Tribasic phosphate countercation survey.
Entry
Base
6a [%][a]
1
2
3
4
5
6
Li3PO4
Na3PO4
K3PO4
Cs3PO4
Mg3(PO4)2
Ca3(PO4)2
0
0
92
0
0
A first survey of potassium bases immediately revealed the
importance of pKa (throughout, pKa refers to the pKa of the
conjugate acid; Table 2). Using KTFA (TFA=trifluoroacetic acid,
entry 1), no cross-coupling took place and the starting materi-
als were returned. As the pKa increased through KOAc, K2CO3,
0
[a] Determined by HPLC analysis.
The pKa and solution pH ranges of these phosphate salts are
approximately equivalent. Accordingly, the widely different re-
action response must be due to other factors. As noted above,
Amatore and Jutand have shown that the countercation can
impact cross-coupling efficiency by influencing transmetalla-
tion.[17] H2O plays an important role in the transport of metal
ions from the aqueous phase to the organic phase.[19] Accord-
ingly, the quantity of H2O present in the system may directly
affect the availability of metal ions in the organic phase. This
could contribute to the results observed in Table 3.
Table 2. Potassium base survey.
[a]
Entry
Base
pKa
6a [%][b]
1
2
3
4
5
6
7
8
KTFA
0
6
0
KOAc
K2CO3
K3PO4
KOH
KOtBu
KH2PO4
K2HPO4
37
51
92
22
23
0
For the reaction under development, however, the physical
properties of the base appear to be one of the principal con-
tributors to reaction efficiency. Selected physical constants for
the evaluated bases are provided in Table 4.
10
12
16
18
2
From this available data, two principal relationships can be
established.
7
0
[a] Approximate values.[18] [b] Determined by HPLC analysis.
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8954
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
priate relative saturation (%RS) values have not been docu-
mented for all of these bases. Indeed, only KOAc, K2CO3, and
KOH have %RH values available—23.1%, 43.2%, 9.3% (at
208C), respectively.[20] Accordingly, establishing a relationship
between reaction efficiency and hygroscopicity was not possi-
ble. However, solubility data was informative. Specifically, as
the aqueous solubility of the base increases, conversion also
increases. For example, when comparing the alkali metal and
alkaline earth phosphates, moving from Ca2+ to Mg2+ to Li+
to Na+ to K+, both solubility and conversion increase (en-
tries 1–3, 6, and 7). Unfortunately, no solubility data was avail-
able for Cs3PO4. If solubility is removed as a factor, then pH
drives the reaction efficiency. For example, K2CO3 and Cs2CO3
both exhibit good solubility (>1 gmLÀ1) and equivalent pH
and deliver very similar levels of conversion (approx. 50%).
KOAc again demonstrates good solubility, but with a lower pH,
conversion decreases (entry 9). At the low quantity of H2O
used in this system (5 equiv), low base solubility appears to be
a key issue. We considered the possibility that this may be rec-
tified if the quantity of H2O was increased. Indeed, analysis of
the reactions of the alkali metal phosphates at 22 equivalents
of H2O (10:1 THF/H2O) and 50 equivalents of H2O shows that
bases of lower solubility can begin to deliver some improved
conversion in certain cases (Table 5). For example, Li3PO4 starts
Table 4. Selected physical constants for the bases used in Tables 2 and 3.
Entry Base
pKa[a] Approx. pH of
aqueous metal
ion[b]
Solubility at RT
[g per 100mL
H2O][b]
6a
[%][c]
1
2
3
4
5
6
7
8
Li3PO4
Na3PO4
K3PO4
Cs3PO4
Cs2CO3
Mg3(PO4)2 12.7 11.2
Ca3(PO4)2 12.7 12.7
12.7 13.6
12.7 13.9
12.7 14.0
0.027
14.25
106
–
261
0.0009[d]
0.00012
–
269
111
121
–
0
6
92
6
48
0
12.7
10.3
–
–
0
0
KTFA
À0.25 14.0
4.8 14.0
10.3 14.0
14.2 14.0
17.0 14.0
9
KOAc
K2CO3
KOH
KOtBu
KH2PO4
K2HPO4
37
51
22
23
0
10
11
12
13
14
2.1
7.2
14.0
14.0
25
168
0
[a] Approximate values.[18] [b] Approximate values.[14] [c] Determined by
HPLC analysis. [d] Value for the pentahydrate.
The relationship between pKa/pH and conversion
From the results in Table 2 as well as previous studies of
BMIDA cross-coupling and boronic ester esterification process-
es, the reaction is evidently dependent on pH control. An opti-
mum is clearly reached with K3PO4 with an approximate pKa
and pH range of 12.7 and 10–14, respectively. However, evalua-
tion of different metal phosphates, which exhibit approximate-
ly similar pKa and pH, shows that K3PO4 is exclusively effective
whereas the other phosphates result in no conversion to the
desired product—indeed, no cross-coupling at all is observed
under the same reaction conditions.
Table 5. Increasing the quantity of H2O with alkali metal phosphate
bases.
Entry
Base
H2O [equiv]
6a [%][a]
Solvation effects driven by the electrostatic parameter result
in aqueous solutions of metal ions varying markedly in their
pH, from 11.2–14.0 for the ions employed in Table 4.[14] The
more acidic cations, such as Mg2+ or Ca2+, may therefore
result in a buffering effect and thereby negatively modulate
pH; however, this is likely to be minor in contrast to the pH
contribution of the anion and does not account for the com-
plete absence of reactivity seen. For example, KOAc delivers
a considerably lower solution pH than Mg3(PO4)2; however,
KOAc does deliver some observable cross-coupling and specia-
tion control whereas this is completely absent for Mg3(PO4)2
(entry 6 vs. entry 9). Accordingly, other properties of the bases
must be considered in conjunction with pH to explain these
observations.
1
2
3
4
5
6
7
8
Li3PO4
Li3PO4
Na3PO4
Na3PO4
K3PO4
K3PO4
Cs3PO4
Cs3PO4
22
50
22
50
22
50
22
50
0
8
16
20
30
26
8
6
[a] Determined by HPLC analysis.
to show some CÀC bond formation as well as speciation con-
trol at 50 equivalents H2O (Table 5, entry 2) and Na3PO4 im-
proves from 6% (Table 4, entry 2) to 20% conversion to 6a
when the H2O quantity is increased 10-fold (Table 5, entry 4).
Conversely, control is rapidly lost in the reactions with K3PO4
using excesses of H2O (Table 5, entries 5 and 6 vs. Table 4,
entry 3), leading to extensive uncontrolled oligomerization,
while H2O loading had little effect on reactions using Cs3PO4
(entries 7 and 8).
The relationship between solubility of the base and
conversion
Information on the hygroscopicity of the bases in Table 4 is
generally only qualitative: these bases are typically designated
as either hygroscopic or deliquescent with little quantitative in-
formation available. Some salts have specific hydrate states,
such as K3PO4 and Mg3(PO4)2 existing as the stable tetrahydrate
and octahydrate, respectively.[14] In terms of saturated aqueous
solutions, relative humidity (%RH) as well as the more appro-
Overall, pH and solubility of the base are the primary factors
responsible for control over the formal homologation reaction.
When solubility is good, appropriate pH modulation then en-
sures effective control of the speciation events, with K3PO4 pro-
viding an ideal balance of both of these properties that allows
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8955
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
efficient CÀC bond formation and hydrolysis/esterification.
There may be a ‘threshold solubility’ for a specific base pKa in
order to ensure reaction efficiency; however, this could not be
established from the available data.
However, not all phosphine ligands were effective in promot-
ing CÀC bond formation under the conditions employed. For
more complex ligands, the reaction performance was generally
greater when the catalyst was preformed—addition of sepa-
rate PdII source and ligand was often less effective than use of
the same preformed catalyst. For example, addition of PPh3 to
PdCl2 delivered approximately the same conversion to 6a as
the preformed [PdCl2(PPh3)2] (entry 5 vs. entry 7); whereas, ad-
dition of dppf to PdCl2 was significantly less effective than use
of the preformed [PdCl2dppf] catalyst (entry 6 vs. entry 15). Use
of more active catalyst systems such as the biaryl monophos-
phines developed by Buchwald,[21] gave good results but were
less effective with bromophenyl BMIDA substrate 11 a than the
simpler [PdCl2dppf] (entry 6 vs. entries 21–28).
Catalyst and Electrophile Evaluation
Following optimization of H2O and base, we subsequently per-
formed a thorough analysis of reaction performance in relation
to the catalyst and electrophile. From the preceding optimiza-
tion phase, we were aware that, under specific conditions, the
overall reaction efficiency became dependent upon the cross-
coupling efficiency, that is, that speciation events could be
readily controlled and all of the available initial cross-coupling
product 3a could be smoothly funneled to 6a. To ensure
a robust CÀC bond formation, we analyzed a range of catalyst
systems under the emerging optimum base/H2O conditions
(Table 6). From these results, it was clear that use of PdII pre-
catalysts was preferred over Pd0 (for example, entry 4 vs.
entry 5). In addition, the reaction clearly requires a phosphine
ligand in order to be synthetically useful and, in the majority
of cases, Pd(OAc)2 was superior to PdCl2. In the absence of
a ligand (entries 1–3), very little cross-coupling was observed.
To ensure synthetic scope, an analysis of halide and pseudo-
halide derivatives of 11 a was conducted with the most suc-
cessful catalyst ([PdCl2dppf]) as well as a more activated
Pd(OAc)2/monophosphine-based catalyst system (Table 7). With
the exception of the less reactive chlorophenyl BMIDA sub-
strate (entries 7–9), [PdCl2dppf] provided superior levels of con-
version, with bromophenyl BMIDA being the optimum sub-
strate. Pleasingly, excellent conversion could be achieved with
chlorophenyl BMIDA by using Pd(OAc)2/SPhos (entry 9).
Table 7. Variation of the electrophile.
Table 6. Catalyst evaluation for the reaction of 10 and 11.
Entry
Catalyst
Ligand
X
6a [%][a]
1
2
3
4
5
6
7
8
9
[PdCl2dppf]
Pd(OAc)2
[PdCl2dppf]
Pd(OAc)2
[PdCl2dppf]
Pd(OAc)2
[PdCl2dppf]
Pd(OAc)2
–
I
I
60
34
90
77
61
48
0
Entry
Catalyst
Ligand[a]
6a [%][b]
SPhos
–
SPhos
–
SPhos
–
CyJohnPhos
SPhos
1
2
3
4
5
6
7
8
PdCl2
–
–
–
–
–
–
PPh3
PPh3
PtBu3
PtBu3
dppe
dppe
dppp
dppp
dppf
0
5
7
Br
Br
OTf
OTf
Cl
Cl
Cl
Pd(OAc)2
[Pd2(dba)3]
[Pd(PPh3)4]
[PdCl2(PPh3)2]
[PdCl2dppf]
PdCl2
Pd(OAc)2
PdCl2
Pd(OAc)2
PdCl2
Pd(OAc)2
PdCl2
Pd(OAc)2
PdCl2
Pd(OAc)2
PdCl2
Pd(OAc)2
PdCl2
Pd(OAc)2
PdCl2
Pd(OAc)2
PdCl2
Pd(OAc)2
PdCl2
Pd(OAc)2
PdCl2
36
63
92
56
70
41
55
4
68
82
Pd(OAc)2
[a] Determined by HPLC analysis.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
0
0
Substrate Scope
55
1
24
13
67
0
10
14
77
20
67
4
72
23
71
The scope of the optimized reaction conditions was explored
through the synthesis of a range of substrates (Figure 3).[9]
A broad range of common and synthetically useful function-
ality was tolerated, including amides (6b), esters (6e, 6n),
ethers (6h), and nitriles (6g), encompassing both electron-rich
and electron-poor BPin starting materials. Pleasingly, the reac-
tion also tolerated heterocyclic moieties, such as pyrazoles,
furans, pyrans, and thiophenes (6d, 6i, 6k, 6m). All three sub-
stitution patterns on the haloaryl MIDA were compatible, al-
though ortho-substitution was more effective with less sterical-
ly demanding BPins. For these reaction conditions, fluoro-sub-
stituted BMIDA esters were found to be amenable but other
functionalization of the BMIDA component was less successful.
dppf
BINAP
BINAP
XantPhos
XantPhos
SPhos
SPhos
XPhos
XPhos
CyJohnPhos
CyJohnPhos
DavePhos
DavePhos
Pd(OAc)2
[a] Added independently. [b] Determined by HPLC analysis.
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8956
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 4. Using alkenyl BMIDA boronic esters. Yields of isolated products
given. [a] As a mixture of olefin regioisomers and stereoisomers.
Figure 3. Formal homologation of Ar-BPin using haloaryl BMIDA esters.
Yields of isolated products given.
This could be overcome by the use of a more active catalyst
system (see below).
The homologation process was also found to be immediate-
ly transferable to haloalkenyl BMIDA reagents (Figure 4).[9] This
enabled the preparation of a set of elaborated alkenes that in-
cluded both aryl (13c, 13e, 13g) and heteroaryl (13a, 13b,
13d, 13 f) substituents. Although 1,2-disubstituted haloalkenyl
BMIDA components were broadly successful, the use of 1,1-dis-
ubstituted olefins led to isomerization, providing mixtures of
1,1- and 1,2-disubstituted olefinic BPin products (13g).[5h] Un-
fortunately, dienyl BPin products could not be prepared by
using this protocol (13h).
Figure 5. Homologation employing chloroaryl BMIDA and specific substitut-
ed aryl BMIDA components. Yields of isolated products given. [a] Using bro-
moaryl BMIDA. [b] Using chloroaryl BMIDA.
To further broaden the scope of the reaction, a set of func-
tionalized haloaryl BMIDAs was employed (Figure 5).[9] For
these substrates the standard catalyst system ([PdCl2dppf]) was
not sufficiently reactive to promote efficient CÀC bond forma-
the formal BPin homologation process, giving 6a in an isolated
yield of 88%. Changing the starting boron species to either
boronic acids or boronic acid catechol esters (BCat) was found
to be relatively well accommodated when using these condi-
tions to provide access to the expected formally homologated
adducts 14 and 15, respectively, without any further optimiza-
tion. It should be noted that the low conversion to 15 was due
to the stability of the catechol ester, which was found to readi-
ly hydrolyze to the boronic acid. These processes demonstrate
the promising generality of speciation control to facilitate
access to higher homologues of boron species in a one-pot
operation.
tion. However, use of
a more reactive catalyst system
(Pd(OAc)2/SPhos) easily circumvented this reactivity issue, al-
lowing these less reactive electrophiles to be effectively cross-
coupled as well as preserving the speciation control. This ena-
bled the use of haloaryl BMIDA esters with CF3 (6o, 6q) and
OMe (6r) functionality as well heterocyclic BMIDA esters (6p).
Certain functionality, however, in particular o-OMe (6s, 6t) and
o-CO2Me (6u, 6v), were not tolerated (see below).
The generality of the overall reaction with regards to specia-
tion control was also assessed by using three different boron
species (Scheme 4). As shown above, the model BPin system is
readily controlled under the optimized conditions to enable
To probe whether the Pd catalyst remained active after com-
pletion of the formal BPin homologation, a second aryl bro-
mide was added to the reaction mixture (Scheme 5a).[9] Pleas-
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8957
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 6. Evaluation of reaction temperature during Suzuki–Miyaura cross-
coupling of Ar-BPin and haloaryl BMIDA. See also Figure 11, below.
Scheme 4. Generality of speciation control using different boron species.
Cat=catecholate.
Figure 6. Temperature-dependent control of speciation during Suzuki–
Miyaura cross-coupling of Ar-BPin and haloaryl BMIDA. Black: 6a, grey: 3a.
Yields determined by HPLC analysis.
However, it was noted that although conversion to 6a was
decreased at lower temperature, the mass balance of the reac-
tion was the product of the initial cross-coupling, specifically
the biphenyl BMIDA species 3a. Indeed, at room temperature,
3a was found to be the sole product of the reaction. This
demonstrated that, in the absence of a thermal driving force,
the availability of aqueous base was sufficiently retarded under
the developed conditions to ensure the integrity of BMIDA
ester 3a. Upon heating, 3a is hydrolyzed to boronic acid 13,
allowing conversion to 6a. Unlike BMIDA esters, BPin esters
are not easily hydrolyzed under the prevailing hydrolytic condi-
tions.[22] Accordingly, 6a is thermodynamically more stable
under the basic reaction conditions.
Scheme 5. One-pot double Suzuki–Miyaura and double formal homologa-
tion reactions. Yields given are of the isolated products.
ingly, the catalyst was found to be sufficiently active to enable
a second Suzuki–Miyaura cross-coupling to take place between
the newly formed BPin species and the added aryl bromide.
This provided a method for one-pot double Suzuki–Miyaura
cross-coupling reactions, which proceeded in good yield for
products 16a and 16b. Moreover, if a second equivalent of
bromoaryl BMIDA was added, the formal homologation reac-
tion could be extended further (Scheme 5b).[9] Pinacol turnover
could be conducted once again in this one-pot reaction, now
enabling a method for controlled oligomerization of BPin spe-
cies, again in good yield for products 17a and 17b.
This was readily demonstrated in a control reaction where
carrying out the optimized reaction at room temperature led
to 97% of 3a which, upon heating to 908C, was smoothly con-
verted to 6a (Scheme 7).
Accordingly, it became possible to chemoselectively control
the outcome of the haloaryl BMIDA cross-coupling reaction in
terms of two possible boron species, BMIDA 3a or BPin 6a, en-
Speciation Control by Temperature Regulation
During the course of the optimization process, the effect of
temperature on both the cross-coupling and speciation turn-
over was investigated. Although 908C was found to be effi-
cient at enabling conversion to product 6a, lower tempera-
tures gave much lower conversion (Scheme 6, Figure 6).
Scheme 7. Temperature control of speciation. Yields determined by HPLC
analysis.
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8958
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tirely through temperature control. The ability to control the
product of this reaction by simply altering the temperature
opened up a potentially useful synthetic possibility. Owing to
their rapid hydrolysis with aqueous base, cross-coupling of
haloaryl BMIDA esters is normally carried out under strictly an-
hydrous conditions, often employing elevated temperatures or
alternate promoters such as FÀ to ensure synthetic efficiency.[4e]
However, overly harsh thermal promotion can limit the poten-
tial scope of these processes owing to conflicting decomposi-
tion pathways of sensitive substrates, including promoting pro-
todeboronation of the boron-derived coupling partners.[23] The
ability to carry out cross-couplings of haloaryl BMIDA species
at ambient temperature in the presence of aqueous base may
therefore be desirable. With no further optimization required,
we then sought to demonstrate the utility of this reaction by
generating a small library of functionalized BMIDA products
(Figure 7).
phy, is exceptionally difficult. Beyond the examples given in
Figure 7, many similar cross-couplings do proceed effectively
to deliver the product in good yield but in approximately 90%
purity. Alkenyl BPin compounds were also readily employed,
with the synthesis of a set of vinyl MIDAs including aryl (18a,
18b, 18d), heterocyclic (18c, 18e), and dienyl (18 f) functional-
ity (Figure 8).
Once again, a range of common functionality was compati-
ble with the developed process. In addition, this protocol read-
ily accommodated temperature-sensitive functional groups
such as heterocyclic BMIDA (3b, 3d, 3h, 3j) and protecting
groups (3e, 3h, 3n), which were found to protodeboronate or
hydrolyze, respectively, at more elevated temperatures.
Figure 8. Room-temperature cross-coupling of haloalkenyl BMIDA in the
presence of aqueous base. Yields of isolated products given.
It is worthwhile noting that this procedure has the added
benefit of requiring very little purification—no chromatogra-
phy was necessary, with the products being isolated by
a single aqueous wash and precipitation of the product using
Et2O. If reactions do not proceed to completion, separation of
two different BMIDAs, either by crystallization or chromatogra-
From the utility perspective, the developed method com-
pares favorably with existing methods. A comparison of reac-
tion performance with the developed room-temperature pro-
tocol versus previously described methods[4b] using five repre-
sentative substrates (aryl, heteroaryl, alkenyl and with variation
of regiochemistry) is provided in Table 8.
The mild room-temperature protocol provided consistently
useful yields of the desired BMIDA products (conditions A). In
some cases, the previously described protocol (conditions B)
was comparable (entries 1 and 2). In other cases, conditions B
provided low yields of the desired product (entry 5) or no
product at all (entries 3 and 4). Lack of product using the con-
ventional protocol could be attributed mainly to the stability
of either the starting materials (3c, 3j) or product (3j) for
which protodeboronation was a significant issue, even at the
very moderately elevated reaction temperature.
Lastly, the room-temperature procedure was also found to
be readily scalable and the product can be straightforwardly
isolated without resorting to chromatography (Scheme 8).
Rationalization of Anomalous Observations
1. Efficiency of Cross-coupling: Regioisomer Disparity
During the course of substrate application for the room-tem-
perature BMIDA cross-coupling studies above, we observed
a reactivity difference with the regioisomers of bromophenyl
BMIDA (11 a, 11 b, and 11 c, Figure 9). Specifically, in several
cases we observed the efficiency of the cross-coupling of the
meta-isomer 11 b to be noticeably lower than that of 11 a and
11 c, and that this was independent of the BPin coupling
partner.
Following NMR analysis, Burke and co-workers noted that
the BMIDA motif is neither a strongly electron-donating nor
electron-withdrawing functional group.[4b] Based on this analy-
Figure 7. Room-temperature cross-coupling of haloaryl BMIDA in the pres-
ence of aqueous base. Yields of isolated products given.
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8959
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(124.4 ppm for 11 a, 123.3 ppm for 11 b, and 128.7 ppm for
11 c).
Table 8. Comparison of similar procedures for retaining the BMIDA func-
tionality during Suzuki–Miyaura cross-coupling.
Based on this NMR analysis, it may be predicted that,
amongst these regioisomers, 11 c would have been most likely
to exhibit a different reactivity profile. Similarly, the crystal
structure data of 11 a, 11 b, and 11 c suggests that 11 c would
potentially experience the largest issue with reactivity owing
to the proximity of the bulky BMIDA, whereas 11 a and 11 b
would be relatively much more accessible (Figure 10).
Entry
1
Product
Procedure
Yield [%][a]
A
B[b]
84
64
A
80
87
2
3
B[b]
A
B[b]
86
—
[c]
A
80
—
4
5
[d]
B[b]
Figure 10. Selected views of the crystal structures of 11 a, 11 b, and 11 c. For
full details, see the Supporting Information.
A
B[b]
84
39[e]
[a] Yields of isolated products given. [b] See reference [4b]. [c] No cou-
pling observed, BMIDA starting material returned. Pyranyl BPin observed
to rapidly decompose at the temperature associated with conditions B.
[d] No coupling observed, BMIDA starting material returned. [e] Reaction
did not proceed to completion.
Interestingly, the C-C-B bond angle for 11 a and 11 b was
ꢀ1228 but ꢀ1288 for 11 c, highlighting the nature of the
steric environment of the CÀBr bond in 11 c.
Based on these overall stereoelectronic considerations, 11 c
would seem to have the greatest likelihood of diminished reac-
tivity. However, 11 b was the consistent outlier, with 11 a and
11 c remaining comparable throughout, providing the steric
demands of 11 c were met. Accordingly, we considered physi-
cal properties as the source of this anomaly. Empirical observa-
tions recorded during experimental set up suggested 11 b was
less soluble in the reaction mixture than 11 a or 11 c. BMIDA
substrates exhibit low solubility in many organic solvents—
a property that enables their facile purification.[5] Although
many BMIDA-based reactions are performed in solvents such
as DMF, presumably to aid solubility of these compounds,
other solvents have been used, such as 1,4-dioxane, THF, and
PhMe.[4d,f,5n]
Scheme 8. Room-temperature cross-coupling of haloalkenyl BMIDA in the
presence of aqueous base on gram scale. Yield of isolated product given.
Boc=tert-butoxycarbonyl.
To gauge whether solubility may be a factor, we analyzed
the solubility of 11 a, 11 b, and 11 c in THF at room tempera-
ture, and obtained the following values: 11 a, 56 mgmLÀ1, 11 b,
19 mgmLÀ1, 11 c, 27 mgmLÀ1. 11 b was found to be markedly
less soluble than 11 a and 11 c. We believe that this lower solu-
bility may contribute to the observed discrepancy in reaction
efficiency when using 11 b.
Figure 9. Regioisomeric bromophenyl BMIDA.
sis, the disparity in the efficiency of cross-coupling of 11 a–c an-
alogues was unlikely to be electronic in nature, that is, that the
dissimilarity was unlikely to be driven by large variation in the
rates of oxidative addition of the regioisomeric bromides.[24]
Analysis of the 13C NMR spectra of 11 a–c, as an indication of
relative electronic disposition of the bromide-bearing carbon
atom, revealed that the para- and meta-isomers, 11 a and 11 b,
were very similar but that the ortho-isomer 11 c was the elec-
tronic outlier based on the large downfield shift of this signal
2. Efficiency of Speciation Control with ortho-Substituted
BMIDA
The cross-couplings of substituted haloaryl BMIDA compounds
(6c, 6d, 6g, Figure 3 and 6o–v Figure 5) were typically reason-
ably effective, providing yields of BPin products in the region
50–70%. However, we noticed a particular disparity when cer-
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8960
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tain ortho-substituted BMIDA components were used. Specifi-
cally, when a methoxy or methyl ester substituent was located
ortho to the BMIDA group, we observed little to no conversion
to the desired BPin product (Figure 5, 6s–6v). In both cases,
the initial cross-coupling and hydrolysis to the boronic acid
were sufficiently effective; however, the turnover of pinacol to
form the desired BPin ester was found to be problematic. For
MeO-substituted products, 6s and 6t, the reaction tended to
produce only the biphenyl boronic acid intermediate 5b even
after extended periods of time, suggesting a sluggish esterifi-
cation process (Scheme 9a). The reasons for this are unclear;
although we suspect this could be due to an intramolecular
O–B Lewis pair interaction (as shown in 5b).[25] Such an interac-
tion may inhibit the esterification process. However, NMR anal-
ysis did not confirm any deviation of the 11B signal for this spe-
cies. Regioisomeric MeO-substitution did not present this issue
(for example, 6r, Figure 5).
Conversion to BPin was similarly poor for the ortho-ester
substituted products 6u and 6v. For these reactions, we ob-
served a large quantity of the protodeboronated biphenyl
product 19 (Scheme 9b). We believe this is due to the proximi-
ty of the electron-withdrawing ester functionality, which leads
to accelerated rates of protodeboronation.[23b] It should also be
noted that ortho-F was tolerated and did not provide any
issues with either the esterification process or protodeborona-
tion (see 6c, 6d, 6g, Figure 3).
Manipulation of Boron Speciation Equilibria: Control
Reactions
We believe the formal homologation reaction relies upon the
simultaneous control of a series of boron speciation equilibria
(Scheme 10). Cross-coupling of BPin 1 with conjunctive BMIDA
2 provides the expected adduct 3.[4f] A frequently overlooked
and generally discarded byproduct of this process is the boric
acid ester 4. Both of these intermediate boron species, 3 and
4, can then participate in independent equilibria that can be
modulated by pH control.[12,13]
Liberation of pinacol requires hydrolysis of 4 and control
over the formation of the 2:1 complex (20).[12] Hydrolysis of 4
under aqueous basic conditions delivers B(OH)3 (and the boro-
nate derivative 21), both of which will be sequestered to the
basic phase.[12,26] Hydrolysis of 3 under basic conditions liber-
ates the corresponding boronic acid 5,[4e,5d,g,q] which can estab-
lish a series of equilibria including formation of the boronate
Scheme 9. Inhibition of esterification (a) and protodeboronation (b) when
using ortho-substituted bromophenyl BMIDA reagents.
Scheme 10. Main solution-phase speciation equilibria associated with the
formal homologation process.
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8961
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
22 and boroxine 23.[27] Esterification of boronic acids (5) and
the corresponding boronate derivatives (22) with 1,2-diols is
accelerated at high pH, with the former being the kinetically
more competent species.[13] Following esterification, the newly
generated BPin 6 will exist as the thermodynamically favored
form of boronate 24, with 6 isolated upon completion of the
reaction.
The staging of the reaction is crucial. The initial cross-cou-
pling of 1 and 2 to produce 3 must be complete before hy-
drolysis of 3 takes place. If 3 hydrolyzes prematurely to boronic
acid 5 before consumption of 2, competing cross-coupling
may take place. Similarly, cross-coupling of 1 and 2 must be
complete before generation of product 6 in order to avoid
competing cross-coupling with 2 (see Scheme 2). Analysis of
these events by using independent reactions demonstrated
that, under the optimized reaction conditions, cross-coupling
is rapid and is complete in <1 h whereas hydrolysis of BMIDA
intermediate 3 requires approximately 4 h. Accordingly, oligo-
merization can be robustly avoided with this hydrolysis latency
period. For the benchmark reaction, production of the desired
BPin product 6a versus presence/consumption of the inter-
mediate BMIDA 3a could be followed by HPLC (Figure 11).
Scheme 11. Conversion of 5a to 6a under representative reaction condi-
tions. Yields determined by HPLC analysis.
Scheme 12. Esterification of 3a by using pinacol to deliver 6a in the pres-
ence and absence of base. Yields determined by HPLC analysis.
BMIDA 3a to boronic acid 5a is followed by subsequent
esterification (Scheme 13, pathway A). In the absence of base,
pinacol engages BMIDA to induce hydrolysis to the boronic
acid 5a. In doing so, a pinacol-MIDA ester (25 or 26) is gener-
ated, disabling boronic acid esterification (Scheme 13, pathway
B).
It may be reasoned that 6a could ultimately be generated
following formation of 25/26, through subsequent hydrolysis
to liberate pinacol, followed by the expected esterification of
5a (Scheme 13, pathway B+C). However, after completion of
the base-free reaction (Scheme 12), addition of K3PO4 did not
induce formation of 6a, lending further support to the basic
hydrolysis/esterification sequence of events.
Figure 11. Production of 6a and presence/consumption of 3a. Black: 6a,
grey: 3a. Yields determined by HPLC analysis.
The formation of boronic esters from boronic acids and diols
has been extensively researched and the requirements for ef-
fective esterification have been thoroughly established.[12,13]
Throughout, no boronic acid 5a was detected under the op-
timized conditions, an observation that is in agreement with
previous observations that the esterification process is rapid
and the efficiency of the reaction is directly linked to the effi-
ciency of cross-coupling. Indeed, independent treatment of 5a
with pinacol under the reaction conditions delivers quantita-
tive formation of 6a in <1 h (Scheme 11 a). Similarly, 5a is
quantitatively converted to 6a from the byproduct of the ini-
tial Suzuki–Miyaura cross-coupling 4 under the representative
reaction conditions (Scheme 11 b).
To ensure no other possible esterification pathways, we con-
ducted a series of control experiments. Treatment of 3a with
pinacol under the reaction conditions either in the presence or
absence of base was informative (Scheme 12).
In the absence of base, 3a is converted to the boronic acid
derivative 5a only. However, when base is included, 3a is con-
verted smoothly to the BPin adduct 6a. These results support
a mechanism in which, when base is present, the hydrolysis of
Scheme 13. Reaction of 3a with pinacol in the presence and absence of
base.
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8962
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Under basic conditions, formation of 6a from 5a and pinacol
is rapid. Accordingly, as might be expected based on the gen-
erally high efficiency of the reaction (Figure 3), the reverse pro-
cess is unfavorable. The BPin product 6a is rapidly converted
to the boronate derivative 24, which is the thermodynamic
end point for the boronic acid species in the reaction mixture.
Direct hydrolysis of BPin, under basic conditions, is exceedingly
difficult. Indeed, exposure of 6a to the reaction conditions,
even for prolonged reaction times, failed to deliver any of the
derivative biaryl boronic acid 5a (Scheme 14).
Hydrolysis of BPin is more readily achieved by exploiting
speciation equilibria with the addition of a second boron spe-
cies, such as a polymeric phenyl boronic acid, relying upon
equilibrium distortion to completely drive pinacol transfer.[22a,b]
In this regard, treatment of PhBPin 10 with boronic acid 5a
leads to equilibration to deliver mixtures of 5a and 6a
(Scheme 15).
haloaryl and haloalkenyl BMIDA esters. This internal aqueous
base control mechanism enables the cross-coupling to be
readily conducted with speciation control possible by tempera-
ture modulation to enable the production of BPin adducts
or BMIDA adducts. The requirements for effective speciation
control have been investigated and the proposed sequence
of events is supported by
transformations.
a
series of independent
Acknowledgements
This work was supported by the Engineering and Physical Sci-
ences Research Council (EPSRC). We thank the EPSRC UK Na-
tional Mass Spectrometry Facility at Swansea University for
analyses and GlaxoSmithKline for financial support.
Keywords: boron
·
chemoselectivity
·
cross-coupling
·
palladium · speciation
[1] For reviews of the Suzuki–Miyaura reactions, see: a) N. Miyaura, A.
Suzuki, Chem. Rev. 1995, 95, 2457–2483; b) C. Valente, M. C. Organ, The
Contemporary Suzuki–Miyaura Reaction In Boronic Acids: Preparation
and Applications in Organic Synthesis and Medicine (Ed.: D. G. Hall),
Wiley-VCH, Weinheim, 2005, vol. 2, pp. 213–262; c) Science of Synthesis
Cross Coupling and Heck-Type Reactions (Eds.: G. A. Molander, J. P. Wolfe,
M. Larhed) Thieme, Stuttgart, 2012; d) A. J. J. Lennox, G. C. Lloyd-Jones,
Angew. Chem. Int. Ed. 2013, 52, 7362–7370; Angew. Chem. 2013, 125,
7506–7515; e) A. J. J. Lennox, G. C. Lloyd-Jones, Chem. Soc. Rev. 2014,
43, 412–443.
Scheme 14. Attempted hydrolysis of 6a to 5a under the reaction condi-
tions.
[2] For reviews of iterative cross-coupling using MIDA- and DAN-derived
boronic esters, see: a) E. P. Gillis, M. D. Burke, Aldrichimica Acta 2009, 42,
17–27; b) C. Wang, F. Glorius, Angew. Chem. Int. Ed. 2009, 48, 5240–
5244; Angew. Chem. 2009, 121, 5342–5346; c) M. Tobisu, N. Chatani,
Angew. Chem. Int. Ed. 2009, 48, 3565–3568; Angew. Chem. 2009, 121,
3617–3620; d) J. W. B. Fyfe, A. J. B. Watson, Synlett DOI: ST-2015-P0028-
SP.
Scheme 15. Equilibration of 5a and 10. Yields determined by HPLC analysis.
[3] First development of BMIDA: a) R. Contreras, C. Garcia, T. Mancilla, B.
Wrackmeyer, J. Organomet. Chem. 1983, 246, 213–217; b) B. Garrigues,
M. Mulliez, A. Raharinirina, J. Organomet. Chem. 1986, 302, 153–158;
c) T. Mancilla, R. Contreras, B. Wrackmeyer, J. Organomet. Chem. 1986,
307, 1–6; d) B. Garrigues, M. Mulliez, J. Organomet. Chem. 1986, 314,
19–24.
[4] Use of BMIDA in iterative cross-coupling: a) E. M. Woerly, J. Roy, M. D.
Burke, Nature Chem. 2014, 6, 484–491; b) S. Fujii, S. Y. Chang, M. D.
Burke, Angew. Chem. Int. Ed. 2011, 50, 7862–7864; Angew. Chem. 2011,
123, 8008–8010; c) S. J. Lee, T. M. Anderson, M. D. Burke, Angew. Chem.
Int. Ed. 2010, 49, 8860–8863; Angew. Chem. 2010, 122, 9044–9047;
d) E. M. Woerly, A. H. Cherney, E. K. Davis, M. D. Burke, J. Am. Chem. Soc.
2010, 132, 6941–6943; e) S. J. Lee, K. C. Gray, J. S. Paek, M. D. Burke, J.
Am. Chem. Soc. 2008, 130, 466–468; f) E. P. Gillis, M. D. Burke, J. Am.
Chem. Soc. 2007, 129, 6716–6717.
[5] For other uses/preparations of BMIDA reagents, see: a) S. Adachi, A. B.
Cognetta III, M. J. Niphakis, Z. He, A. Zajdlik, J. D. St. Denis, C. C. G.
Scully, B. F. Cravatt, A. K. Yudin, Chem. Commun. 2015, 51, 3608–3611;
b) J. Cornil, P.-G. Echeverria, P. Phansavath, V. Ratovelomanana-Vidal, A.
GuØrinot, J. Cossy, Org. Lett. 2015, 17, 948–951; c) J. D. St. Denis, A. Zaj-
dlik, J. Tan, P. Trinchera, C. F. Lee, Z. He, S. Adachi, A. K. Yudin, J. Am.
Chem. Soc. 2014, 136, 17669–17673; d) J. D. St. Denis, C. C. G. Scully,
C. F. Lee, A. K. Yudin, Org. Lett. 2014, 16, 1338–1341; e) L. Xu, P. Li, Syn-
lett 2014, 1799–1802; f) L. Xu, S. Ding, P. Li, Angew. Chem. Int. Ed. 2014,
53, 1822–1826; Angew. Chem. 2014, 126, 1853–1857; g) N. A. Isley, F.
Gallou, B. H. Lipshutz, J. Am. Chem. Soc. 2013, 135, 17707–17710;
h) E. M. Woerly, J. E. Miller, M. D. Burke, Tetrahedron 2013, 69, 7732–
7740; i) Z. He, P. Trinchera, S. Adachi, J. D. St. Denis, A. K. Yudin, Angew.
Chem. Int. Ed. 2012, 51, 11092–11096; Angew. Chem. 2012, 124, 11254–
This observation supports the proposed sequence of events,
in particular a rapid Suzuki–Miyaura cross-coupling that is com-
plete before hydrolysis of 3a. If 3a were hydrolyzed before
consumption of 10, equilibration of 5a and 6a would lead to
problems with oligomerization from residual 11 a present in
the reaction mixture owing to the higher cross-coupling reac-
tivity of the boronic acid 5a than the BPin starting material 10.
All of the above observations support the following se-
quence of events for the formal homologation reaction: (1) a
rapid Suzuki–Miyaura cross-coupling; (2) a comparatively slow
BMIDA hydrolysis; and (3) a rapid esterification of the liberated
boronic acid. The reaction is also highly dependent upon:
(1) the nature of the base, which must possess good solubility
and guarantee a suitable pH to enable speciation control; and
(2) the thermodynamic stability of the BPin product.
Conclusion
The fundamental physical properties of inorganic bases enable
the formation of an in situ desiccant that controls the availabil-
ity of aqueous base during Suzuki–Miyaura cross-coupling of
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8963
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
11258; j) G. R. Dick, E. M. Woerly, M. D. Burke, Angew. Chem. Int. Ed.
2012, 51, 2667–2672; Angew. Chem. 2012, 124, 2721–2726; k) H. Wang,
C. Grohmann, C. Nimphius, F. Glorius, J. Am. Chem. Soc. 2012, 134,
19592–19595; l) J. Li, M. D. Burke, J. Am. Chem. Soc. 2011, 133, 13774–
13777; m) E. M. Woerly, J. R. Struble, N. Palyam, S. P. O’Hara, M. D. Burke,
Tetrahedron 2011, 67, 4333–4343; n) G. R. Dick, D. M. Knapp, E. P. Gillis,
M. D. Burke, Org. Lett. 2010, 12, 2314–2317; o) J. R. Struble, S. J. Lee,
M. D. Burke, Tetrahedron 2010, 66, 4710–4718; p) S. G. Ballmer, E. P.
Gillis, M. D. Burke, Org. Synth. 2009, 86, 344–359; q) D. M. Knapp, E. P.
Gillis, M. D. Burke, J. Am. Chem. Soc. 2009, 131, 6961–6963; r) B. E. Uno,
E. P. Gillis, M. D. Burke, Tetrahedron 2009, 65, 3130–3138; s) E. P. Gillis,
M. D. Burke, J. Am. Chem. Soc. 2008, 130, 14084–14085.
teen, S. Deeter, B. Wang, Tetrahedron 2004, 60, 11205–11209; h) L. I.
Bosch, T. M. Fyles, T. D. James, Tetrahedron 2004, 60, 11175–11190; i) G.
Springsteen, B. Wang, Tetrahedron 2002, 58, 5291–5300; j) R. Pizer, C. A.
Tihal, Polyhedron 1996, 15, 3411–3416; k) L. Babcock, R. Pizer, Inorg.
Chem. 1980, 19, 56–61.
[14] CRC Handbook of Chemistry and Physics, 94th Edition (Ed.: W. M.
Haynes), Taylor and Francis, Boca Raton, 2014.
[15] For the use of bromothiophenyl BMIDA reagents as a method for pre-
paring polymeric thiophenes, see: J. A. Carrillo, M. J. Ingleson, M. L.
Turner, Macromolecules 2015, 48, 979–986.
[16] C. Amatore, A. Jutand, G. Le Duc, Chem. Eur. J. 2011, 17, 2492–2503.
[17] C. Amatore, A. Jutand, G. Le Duc, Chem. Eur. J. 2012, 18, 6616–6625.
[18] M. B. Smith, March’s Advanced Organic Chemistry: Reactions, Mechanism,
and Structure, 7th Edition, Wiley, Hoboken, 2013. See also ref. [14].
[19] For example, see: a) H. Deng, P. Peljo, T. J. Stockmann, L. Qiao, T. Vainik-
ka, K. Kontturi, M. Opallo, H. H. Girault, Chem. Commun. 2014, 50,
5554–5557; b) W. Murakami, K. Eda, M. Yamamoto, T. Osakai, J. Electroa-
nal. Chem. 2013, 704, 38–43; c) D. Rose, I. Benjamin, J. Phys. Chem. B
2009, 113, 9296–9303; d) P. Sun, F. O. Laforge, M. V. Mirkin, J. Am. Chem.
Soc. 2007, 129, 12410–12411.
[6] Use of DAN in iterative cross-coupling: a) H. Noguchi, K. Hoj, M. Sugi-
nome, J. Am. Chem. Soc. 2007, 129, 758–759; b) H. Noguchi, T. Shioda,
C.-M. Chou, M. Suginome, Org. Lett. 2008, 10, 377–380; c) N. Iwadate,
M. Suginome, Org. Lett. 2009, 11, 1899–1902; d) N. Iwadate, M. Sugi-
nome, Chem. Lett. 2010, 39, 558–560; e) N. Iwadate, M. Suginome, J.
Am. Chem. Soc. 2010, 132, 2548–2549.
[7] For other uses/preparations of DAN reagents, see: a) H. Yoshida, Y. Take-
moto, K. Takaki, Chem. Commun. 2015, 51, 6297–6300; b) X. Feng, H.
Jeon, J. Yun, Angew. Chem. Int. Ed. 2013, 52, 3989–3992; Angew. Chem.
2013, 125, 4081–4084; c) H. Ihara, M. Koyanagi, M. Suginome, Org. Lett.
2011, 13, 2662–2665; d) J. C. H. Lee, R. McDonald, D. G. Hall, Nature
Chem. 2011, 3, 894–899.
[8] For selected examples of the preparation and chemoselective manipu-
lation of multiboron species, see: a) S. N. Mlynarski, C. H. Schuster, J. P.
Morken, Nature 2014, 505, 386–390; b) C. Sun, B. Potter, J. P. Morken, J.
Am. Chem. Soc. 2014, 136, 6534–6537; c) J. Jiao, K. Hyodo, H. Hu, K. Na-
kajima, Y. Nishihara, J. Org. Chem. 2014, 79, 285–295; d) K. Endo, T.
Ohkubo, M. Hirokami, T. Shibata, J. Am. Chem. Soc. 2010, 132, 11033–
11035. See also refs. [5e, 5f, 7b, 7d].
[9] J. W. B. Fyfe, C. P. Seath, A. J. B. Watson, Angew. Chem. Int. Ed. 2014, 53,
12077–12080; Angew. Chem. 2014, 126, 12273–12276. See also: J. J.
Molloy, R. P. Law, J. W. B. Fyfe, C. P. Seath, D. J. Hirst, A. J. B. Watson, Org.
Biomol. Chem. 2015, 13, 3093–3102.
[20] L. Greenspan, J. Res. Nat. Bur. Stand. 1977, 81A, 89–96.
[21] For selected reviews of biaryl phosphane ligands, see: a) D. S. Surry, S. L.
Buchwald, Angew. Chem. Int. Ed. 2008, 47, 6338–6361; Angew. Chem.
2008, 120, 6438–6461; b) R. Martin, S. L. Buchwald, Acc. Chem. Res.
2008, 41, 1461–1473.
[22] a) T. E. Pennington, C. Kardiman, C. A. Hutton, Tetrahedron Lett. 2004,
45, 6657–6660; b) S. J. Coutts, J. Adams, D. Krolikowski, R. J. Snow, Tet-
rahedron Lett. 1994, 35, 5109–5112; c) D. S. Matteson, R. Ray, R. R.
Rocks, D. J. S. Tsai, Organometallics 1983, 2, 1536–1543.
[23] For selected studies of protodeboronation, see: a) G. Noonan, A. G.
Leach, Org. Biomol. Chem. 2015, 13, 2555–2560; b) J. Lozada, Z. Liu,
D. M. Perrin, J. Org. Chem. 2014, 79, 5365–5368; c) H. G. Kuivila, J. F.
Reuwer Jr. , J. A. Mangravite, Can. J. Chem. 1963, 41, 3081–3090; d) H. G.
Kuivila, K. V. Nahabedian, J. Am. Chem. Soc. 1961, 83, 2167–2174;
e) H. G. Kuivila, K. V. Nahabedian, J. Am. Chem. Soc. 1961, 83, 2164–
2166; f) H. G. Kuivila, K. V. Nahabedian, J. Am. Chem. Soc. 1961, 83,
2159–2163.
[10] For general information see: Boronic Acids: Preparation and Applications
in Organic Synthesis and Medicine (Ed.: D. G. Hall), Wiley-VCH, Weinheim,
2005.
[24] I. J. S. Fairlamb, Chem. Soc. Rev. 2007, 36, 1036–1045.
[11] H. C. Brown, Organic Synthesis via Organoboranes, Wiley Interscience,
New York, 1975.
[25] For an example of a four-membered cyclic Lewis pair, see: P. Spies, G.
Erker, G. Kehr, K. Bergander, R. Frçhlich, S. Grimme, D. W. Stephan,
Chem. Commun. 2007, 5072–5074. For a recent review of frustrated
Lewis pairs, see: D. W. Stephan, Acc. Chem. Res. 2015, 48, 306–316.
[26] For discussion of the basic biphase, see: A. J. J. Lennox, G. C. Lloyd-
Jones, J. Am. Chem. Soc. 2012, 134, 7431–7441. See also refs. [1d, 1e].
[27] Other binary and ternary boronate species, such as those derived from
the inorganic base are not shown. For examples, see ref. [13h]. See
also: M. Sanjoh, D. Iizuka, A. Matsumoto, Y. Miyahara, Org. Lett. 2015,
17, 588–591.
[12] J. P. Lorand, J. O. Edwards, J. Org. Chem. 1959, 24, 769–774.
[13] For informative studies on the esterification of boronic acids, see: a) Y.
Furikado, T. Nagahata, T. Okamoto, T. Sugaya, S. Iwatsuki, M. Inamo,
H. D. Takagi, A. Odani, K. Ishihara, Chem. Eur. J. 2014, 20, 13194–13202;
b) T. Okamoto, A. Tanaka, E. Watanabe, T. Miyazaki, T. Sugaya, S. Iwatsu-
ki, M. Inamo, H. D. Takagi, A. Odani, K. Ishihara, Eur. J. Inorg. Chem.
2014, 2389–2395; c) E. Watanabe, C. Miyamoto, A. Tanaka, K. Iizuka, S.
Iwatsuki, M. Inamo, H. D. Takagi, K. Ishihara, Dalton Trans. 2013, 42,
8446–8453; d) M. A. Martínez-Aguirre, R. Villamil-Ramos, J. A. Guerrero-
Alvarez, A. K. Yatsmirsky, J. Org. Chem. 2013, 78, 4674–4684; e) C. Miya-
moto, K. Suzuki, S. Iwatsuki, M. Inamo, H. D. Takagi, K. Ishihara, Inorg.
Chem. 2008, 47, 1417–1419; f) S. Iwatsuki, S. Nakajima, M. Inamo, H. D.
Takagi, K. Ishihara, Inorg. Chem. 2007, 46, 354–356; g) J. Yan, G. Springs-
Received: March 11, 2015
Published online on May 8, 2015
Chem. Eur. J. 2015, 21, 8951 – 8964
www.chemeurj.org
8964
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim