Unique physicochemical and catalytic properties dictated by the B3NO2 ring system

Article abstract of DOI:10.1038/nchem.2708

The expansion of molecular diversity beyond what nature can produce is a fundamental objective in chemical sciences. Despite the rich chemistry of boron-containing heterocycles, the 1,3-dioxa-5-aza-2,4,6-triborinane (DATB) ring system, which is characterized by a six-membered B₃NO₂ core, remains elusive. Here, we report the synthesis of m-terphenyl-templated DATB derivatives, displaying high stability and peculiar Lewis acidity arising from the three suitably arranged boron atoms. We identify a particular utility for DATB in the dehydrative amidation of carboxylic acids and amines, a reaction of high academic and industrial importance. The three boron sites are proposed to engage in substrate assembly, lowering the entropic cost of the transition state, in contrast with the operative mechanism of previously reported catalysts and amide coupling reagents. The distinct mechanistic pathway dictated by the DATB core will advance not only such amidations, but also other reactions driven by multisite activation.

Full text of DOI:10.1038/nchem.2708

ARTICLES
PUBLISHED ONLINE: 30 JANUARY 2017 | DOI: 10.1038/NCHEM.2708
Unique physicochemical and catalytic properties
dictated by the B3NO2 ring system
Hidetoshi Noda, Makoto Furutachi, Yasuko Asada, Masakatsu Shibasaki and Naoya Kumagai
*
*
The expansion of molecular diversity beyond what nature can produce is a fundamental objective in chemical sciences.
Despite the rich chemistry of boron-containing heterocycles, the 1,3-dioxa-5-aza-2,4,6-triborinane (DATB) ring system,
which is characterized by a six-membered B3NO2 core, remains elusive. Here, we report the synthesis of m-terphenyl-
templated DATB derivatives, displaying high stability and peculiar Lewis acidity arising from the three suitably arranged
boron atoms. We identify a particular utility for DATB in the dehydrative amidation of carboxylic acids and amines, a
reaction of high academic and industrial importance. The three boron sites are proposed to engage in substrate assembly,
lowering the entropic cost of the transition state, in contrast with the operative mechanism of previously reported
catalysts and amide coupling reagents. The distinct mechanistic pathway dictated by the DATB core will advance not only
such amidations, but also other reactions driven by multisite activation.
ollowing the rapid synthetic advances for polycyclic aromatic the requisite B–N–B array. The synthesis of DATB commenced
hydrocarbons (PAHs) such as graphene for prospective use as with Pd-catalysed Suzuki–Miyaura cross-coupling of 2,6-dibromo-
F
optoelectronic materials, heteroatom-doped PAHs have also aniline 2a and phenylboronic acid11. Treatment of thus obtained 3a
garnered extensive research interest on account of their potential with BBr3 introduced a boron atom, affording B-hydroxy azaborine
superior and/or unprecedented physicochemical properties1. The 4a. A subsequent coupling with phenylboronic acid derivative 5,
isoelectronic relationship between the CC and BN units renders which bears an o-B(dan) group (dan = 1,8-diaminonaphthalene),
BN-containing PAHs a particularly significant class of π-conjugated furnished m-terphenyl derivative 6a (ref. 12). Acidic removal of the
scaffolds. It is therefore not surprising that graphitic materials that
consist of B, C, N and O have been the target of intensive research,
1,3-dioxa-5-aza-2,4,6-triborinane
yielding a number of derivatives with unique features2–5. Although
B
DATB
decades of research development have explored the chemical
space of graphitic materials based on B, C, N and O, a six-membered
multiboron core containing both N and O atoms remains elusive,
N
B
N
B
O
B
and thus represents a missing link in this well-explored field
(Fig. 1). Accordingly, we embarked on the design, synthesis
and characterization of the six-membered B3NO2 heterocycle
1,3-dioxa-5-aza-2,4,6-triborinane (DATB) in order to elucidate
its physicochemical and catalytic properties. The coexistence of N
and O atoms that alternate with three B atoms in a six-membered
ring engenders distinct physicochemical properties relative to the
structurally related borazine (B3N3) and boroxine (B3O3) deriva-
tives. DATB is thermally stable with sufficient hydrolytic tolerance,
and the unsymmetrically distributed boron atoms differ in their
Lewis acidity. The suitably aligned boron atoms could be exploited
in the catalytic dehydrative amidation of carboxylic acids and
amines, an important chemical transformation6–8. DATB outper-
forms previously reported catalysts, refining the reaction to substan-
tially expand the reaction scope, most probably via a distinct
activation mechanism (see below). To the best of our knowledge,
the DATB ring system has so far been accessed only as a mixture
of boron-containing heterocycles, and the properties and potential
B
N
B
N
B
B
B
B
B3
O
O
O
O
Borazine
Boroxine
Missing link
N
C
O
C
B
C
B
C
B
C
B
C
B2
N
O
N
C
O
C
C
C
B
C
B
C
C
C
B1
C
C
C
C
C
B0
C
applications have not yet been examined9,10
.
C
At the outset of our study we exploited a rigid m-terphenyl unit to
reinforce the DATB skeleton of 1, anchoring the upper B–N–B frag-
ment to impart thermal stability (Fig. 2a). Because we encountered an
unexpected difficulty in simultaneously installing two boron atoms
into 2,6-diphenylanilines, we considered the stepwise construction
of two biaryl linkages to access the m-terphenyl unit concatenating
Figure 1 | Chemical space for six-membered heterocycles composed of B,
C, N and O atoms and the structure of DATBs. In contrast to the extensive
research on six-membered heterocycles comprising BC4N, BC4O, B2C3N,
B2C3O, B3N3 and B3O3, the DATB ring system of B3NO2 has largely been
neglected and attracted little attention.
Institute of Microbial Chemistry (Bikaken), Tokyo, 3-14-23 Kamiosaki, Shinagawa-ku, Tokyo 141-0021, Japan. e-mail: mshibasa@bikaken.or.jp;
*
nkumagai@bikaken.or.jp
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
1

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2708
ARTICLES
1,3-dioxa-5-aza-2,4,6-triborinane
PhB(OH)2 (1 equiv.)
Pd(PPh3)4 (5 mol%)
Na2CO3 (6 equiv.)
X
a
X
DATB
From6a
m-terphenyl
B(OH)2
template
Toluene/EtOH/H2O
reflux
Br
Br
Br
NH2
NH2
(10 equiv.)
Y
2a: X = H
2b: X = F
2c: X = Me
3a:
3b:
3c:
N
X = H 56%
X = F 56%
X = Me 61%
B
B
1a: Y = H
1b: Y = F
1c: Y = OMe 76%
72%
70%
BBr3 (2.5 equiv.)
O
O
B
CH2Cl2, RT
X
(pin)B
X
X
5
B'
Y
(1 equiv.)
5N HCl aq.
THF, 50 °C
DATB 1
Br
Pd(PtBu3)2 (20 mol%)
K3PO4 (3 equiv.)
1,4-dioxane/H2O
60 °C
N
X
NH
NH
B
B
B
B
B'
HO
OH
OH
OH
7a: X = H
7b: X = F
7c: X = Me
6a: X = H 57%
6b: X = F 40%
6c: X = Me 48%
4a: X = H 97%
4b:X = F >99%
4c: X = Me 98%
N
B
B
X
O
O
B
6a–c
From
H
N
B
without boronic acids
—dimerization—
(dan)B
:
NH
NH
B'
B
8a: X = H 76%
8b: X = F 70%
8c: X = Me 68%
OH
DATB 8
b
c
ΔGcalc = –13.2 kcal mol–1
O
O
+
H
B
Ph
H
+
2H2O
N
N
THF
B
B
B
B
O
O
O
O
H
H
7a
B
Ph
DATB 1a
Top view
Side view
LUMO+1
d
1a
9 (1a pyridine)
•
NICS(1)
1.445(7)
–1.9 –1.9
1.4424(15)
1.4427(15)
1.633(8)
B[21]
B[2] B[2]
B[2] B[2]
[2]
B
1a
+
N
[2]N
B[2]
B[2]
–0.4
B
O
1.3805(14)
1.3810(14)
1.449(7)
B[12]
B[12]
1.3712(14)
[1]O
1.3770(14)
B
Ph
N
9
Figure 2 | Synthesis and properties of DATBs. a, Synthesis of DATBs 1 and 8. b, Calculations on the dehydrative condensation of 7a and phenylboronic acid to
afford DATB 1a were carried out at the M06-2X-D3/6-311++G(d,p)/IEFPCM(THF)//B3LYP/6-31+G(d) level of theory. c, ORTEP drawing of the X-ray structure
of DATB 1a and selected bond lengths. Selected NICS(1) values and the LUMO+1 of DATB 1a are also shown. d, X-ray structure of DATB 1a · pyridine adduct 9
including selected bond lengths. Thermal ellipsoids are set at 50% probability. Colour code: white, hydrogen; orange, boron; grey, carbon; blue, nitrogen; red,
oxygen. ORTEP, Oak Ridge thermal ellipsoid plot. NICS, nucleus-independent chemical shift. LUMO, lowest unoccupied molecular orbital. pin, pinacolyl.
dan group provided the requisite B–N–B motif to afford intermediate m-terphenyl skeleton and the DATB core. The bond lengths of the
7, which captured excess phenylboronic acid via dehydrative conden- four B–O bonds in the DATB core fall in the range of typical B
sation to generate DATB 1a. 1a displayed excellent thermal stability (sp2)–O single bonds (1.31–1.41 Å)15, implying equally low levels of
(mp = 284 °C, <2% weight loss <300 °C, Supplementary Fig. 1) aromaticity. Remarkably, 1a was stable in 2 M HCl/THF aq. at
and its structure, bearing a B3NO2 core, was unequivocally deter- 27 °C, which demonstrates substantial resistance towards hydrolytic
mined by multinuclear NMR spectroscopy, mass spectrometry and cleavage. This behaviour distinguishes 1a (B3NO2) from borazines
single-crystal X-ray diffraction analysis. The solid-state structure of (B3N3)16,17 and boroxines (B3O3)18, as the former readily undergo
1a revealed almost perfect planarity, exhibiting an angle sum of hydrolysis (Supplementary Fig. 3) and the latter exist in a rapid
360 0.1° around the nitrogen atom and an offset face-to-face stack- equilibrium between dehydrative formation and hydrolysis. The
ing array with a π–π distance of ∼3.42 Å (Fig. 2c and Supplementary calculated free energy for the dehydrative condensation of 7a and
Fig. 15). Although the B–N bond lengths (1.4424(15)–1.4427(15) Å) phenylboronic acid to afford 1a (ΔG = –13.2 kcal mol−1) is consistent
are comparable to typical aromatic B–N bonds (1.45–1.47 Å)13, a with the unusually high hydrolytic tolerance of 1a (Fig. 2b).
nucleus-independent chemical shift (NICS(1)) analysis14 indicated Visualization of the LUMO + 1 of 1a suggested Lewis acidity for
little aromaticity for both 1,2-azaborine moieties embedded in the boron atom B[1], which was experimentally supported by the isolation
2
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2708
ARTICLES
3–
O
P
a
H
N
O
[2]
B
O
O
R2
H
B[1]
+
N
R1
O
R1
C
B[2]
OH
Ph
H
N
R
O
R
R2
N
O
[2]
B
N
B[1]
R
B[1]
O
[2]
O
B
[2]O
[2]
O
O
Ph
Ph
N
B
N
B
R2
O
R1
N
H
II
+ H2O
DATB 1a • amine
[DATB 1a • P(O)(O–)2(NH–)]
I
—Catalyst—
• B(OH)3
• RB(OH)2
• R2BOH
b
c
H
O
O
B
O
O
R2
H2N R2
+
R1
R1
R1
• Group IV,VI
metal
complexes
O
R
OH
N
H
III
O
O
Catalyst (5 mol%)
Toluene, 80 °C, 4 h
Ph
H2N
+
X
N
Ph
e
N
OH
H
F
F
MS4Å
10a
1 equiv.
11a
1 equiv.
12
N
B
Catalyst
B
B
B
B
X
1a
40%
DATB catalysts
O
O
O
O
B
1b 9%
1c
33%
1a: Y = H
1b:Y = F
1c: Y = OMe
8a: X = H
8b:X = F
8c: X = Me
1d
21%
17%
NH
B
1e
Y
OH
8a
95%
(10 mol%, RT, 24 h)
8a
55%
8b
57%
8c
85%
4a 0% (24 h)
13 0% (24 h)
14 0% (48 h)
15 7% (48 h)
N
N
B
B
B
B
O
O
O
O
B
B
0
20
40
60
80
100
Ph
Yield (%) of 12
NH
B
1d
1e
Ph
OH
O
d
O
Catalyst (10 mol%)
+
N
H2N
Ph
5
OH
Toluene, RT, 8 h
MS4Å
H
5
OH
B
Ph
O
Ph
B
B
Br
10b
1 equiv.
11b
1 equiv.
16
HO
O
O
Cp2HfCl2
B
NH
B
I
OMe
Ph
OH
8a
14
15
95%
95%
95%
4a
13
14
15
0
20
40
60
80
100
Yield (%) of 16
Figure 3 | Utility of DATB derivatives in catalytic direct amidations inspired by a calculated structure, where all three B atoms are coordinated.
a, Computational study to explore the Lewis acidity of B[2] in a DATB 1a · amine adduct, leading to phosphoramide adduct I (optimized at the B3LYP/6-31+G(d,p)
level of theory). Replacement of the P=O with a C–R1 moiety affords plausible transition state II for direct amidation. Colour code for the optimized molecular
model: white, hydrogen; orange, boron; grey, carbon; blue, nitrogen; red, oxygen; yellow, phosphorus. b, Current state of the art of catalytic direct amidations.
Various boronic and borinic acids as well as boric acid together with group IV and VI metal catalysts have been developed to promote direct amidation with
5–20 mol% catalyst loading. For boronic acid catalysis, activated intermediate III is proposed to be involved. Although some catalysts allow the reaction to
proceed at room temperature, they are largely less applicable to bulky substrate sets, which can be addressed by using DATB catalysts. c, Evaluation of DATB
catalysts in the direct amidation of a challenging substrate set, phenylisobutyric acid 10a and 11a. At room temperature, only DATB 8a is able to convert 10a.
d, Evaluation in a more reactive substrate set, heptanoic acid 10b and 11b. e, Structures of catalysts used in these studies. Cp, cyclopentadienyl.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
3
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2708
ARTICLES
Table 1 | Substrate generality of the catalytic direct amidations promoted by DATB 8a.
N
B
B
O
O
B
NH
B
O
R2
DATB 8a
O
OH
H
R2
0.5–5 mol%
N
+
R1
N
R1
OH
R3
Toluene, MS4Å
R3
10
11
0.3 mmol
1.0 equiv.
O
O
H
O
Ph
N
OMe
O
O
N
H
[gram
Ph
O
N
Ph
Ph
Ph
Ph
H
Ph
N
H
O
Ph
N
Ph
N
OTBS
N
F
H
H
H
scale]
17
18
19
20
21
22
[Acid 10a + amine 11c]
89% yield
[Acid 10c + amine 11d]
86% yield
[Acid 10d + amine 11e]
89% yield
[Acid 10e + amine 11b]
88% yield
[Acid 10f + amine 11a]
90% yield
[Acid 10g + amine 11b]
93% yield
(5 mol%, 80 °C, 14 h)
(5 mol%, 80 °C, 14 h)
(5 mol%, 80 °C, 6 h)
(5 mol%, reflux, 14 h)
(5 mol%, reflux, 14 h)
(2 mol%, reflux, 18 h)
O
OMe O
O
O
N
NH
O
O
Cl
N
H
S
N
H
Ph
N
H
N
N
H
H
H
F
NH
N
OMe
23
N
Cl
24
25
26
27
28
[Acid 10h + amine 11b]
87% yield
[Acid 10i + amine 11f]
85% yield
[Acid 10j + amine 11a]
85% yield
[Acid 10k + amine 11g]
74% yield
[Acid 10l + amine 11h]
81% yield
[Acid 10m + amine 11i]
84% yield
(5 mol%, 80 °C, 14 h)
(5 mol%, 80 °C, 14 h)
(5 mol%, reflux, 14 h)
(5 mol%, 80 °C, 14 h)
(5 mol%, 80 °C, 14 h)
(5 mol%, 80 °C, 14 h)
O
N
NH2
O
O
O
O
O
O
O
Ph
F
Ph
Ph
N
H
Ph
N
Ph
Ph
O
Ph
N
H
Ph
N
H
N
H
Ph
N
Ph
N
F
H
H
O
29
30
31
32
33
34
35
36
[Amine 11d]
[Amine 11j]
[Amine 11k]
[Amine 11l]
[Amine 11m]
[Amine 11n]
[Amine 11o]
[Amine 11p]
70% yield
95% yield
91% yield
70% yield
89% yield
90% yield
69% yield
86% yield
(0.5 mol%, reflux, 24 h) (5 mol%, 80 °C, 14 h) (5 mol%, 80 °C, 14 h) (5 mol%, 80 °C, 14 h) (5 mol%, 80 °C, 14 h) (5 mol%, reflux, 14 h) (5 mol%, reflux, 14 h) (5 mol%, 80 °C, 14 h)
Amisulpride
O
Et
N
Vorinostat
Sunitinib
F
O
O
O
O
Et
O
O
H
O
H
N
N
Et
Ph
N
H
S
N
OH
Ph
H
N
Ph
Et
H2N
N
Ph
N
NH2
N
N
H
Ph
5
HN
H
N
H
H
O
H
O
OH
OMe
42
O
O
O
N
H
37
38
39
40
41
O
[Acid 10o + amine 11q] [Acid 10p + amine 11q]
83% yield 80% yield
(5 mol%, 80 °C, 14 h) (5 mol%, 50 °C, 14 h)
[Acid 10p + amine 11r]
75% yield
[Acid 10q + hydroxylamine 11s]
82% yield
[Acid 10r + amine 11t]
52% yield for 2 steps
[Acid 10s + amine 11u]
76% yield
O
OH
H
HN
10r
(5 mol%, 50 °C, 14 h)
5 mol%, 80 °C
(5 mol%, 80 °C, 14 h)
subsequent condensation
with 5-fluoro-2-oxindole
(5 mol%, reflux, 14 h)
1.0 equiv. Et3N, 14 h
Amides 17–25 represent the reactions of bulky carboxylic acids. Amides 26–28 represent the reactions of substrates bearing Lewis basic heteroaromatic units. Amides 29–36 represent the reactions of benzoic
acid 10n and various amines 11d–11p. Amides 37–39 represent the reactions of homochiral substrates and were obtained in a stereochemically pure form. Amides 40–42 represent the applications of DATB
catalysis to the synthesis of APIs.
of a single crystal of pyridine adduct 9 (Fig. 2d). Intriguingly, the pharmaceuticals. For example, 25% of small-molecule active
coordination of oxygen-based Lewis bases such as alcohols or pharmaceutical ingredients (APIs) contain at least one amide
carboxylic acids to B[1] was not observed.
bond, and 16% of all reactions in medicinal chemistry are categor-
The thermal stability and hydrolytic tolerance of DATB in ized as N-acylations using stoichiometric activating reagents7,8,19
.
addition to the high affinity of B[1] to amines prompted us to use Indeed, a medicinal chemistry consortium selected practical cataly-
DATB as a catalyst. To examine the Lewis acidity of the remaining tic direct amidations as key chemical transformations to streamline
two B[2] atoms, an in silico screening of Lewis bases was conducted, the synthesis of APIs in a more sustainable manner, and their broad
which eventually located phosphoramide adduct I (Fig. 3a). utility in organic synthesis is highly prospected20. In sharp contrast
Adduct I possesses an adamantyl skeleton, where the boat-like to the high and constant demand for them, such catalytic direct
conformation of DATB is anchored by the phosphoramide amidations have made relatively slow progress6,21–25
. The
through two B[2]–O and B[1]–N bonds. The structure of I is thus pioneering work by Yamamoto showed that certain boronic acids
reminiscent of that of plausible transition state II in an amide catalytically promote direct dehydrative amidations at elevated
bond-forming reaction, given that the P=O moiety is replaced temperatures (Fig. 3b)26,27. These results triggered subsequent
with C–R1. This reasoning suggests that 1a might catalyse the explorations into improved catalysts including group IV and VI
direct coupling of carboxylic acids and amines. Amide bonds are metal complexes28–42, and some boron-based catalysts are found
ubiquitous in nature and found in various biomacromolecules in successful industrial applications for the production of APIs8,43–45
.
(such as proteins), functional materials (such as nylons) and However, intrinsic issues related to catalytic performance, including
4
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2708
ARTICLES
[2]
[2]
B
[1]O
[2]N
B
[1]
B
B
*
1) 1a
1) 1a
b
a
*
O
B
Ph
H H
H2N C6H4-p-F
2)
11a
[2]N
[2]
B
[2]
B
[2]N
B
O
B
[1]
B
[1] O
[2]
B
[1]
B
H H
3) 1a + 11a
O
O
2) 1a + 11a
B
Ph
N
C6H4-p-F
Ph
N
C6H4-p-F
H2
H2
4) 10b + 11a
5) 1a + 10b + 11a
[2]
B
[1] O
[2]N
B
O
[1]
B
3) 1a + 10b + 11a
[2]
B
B
Ph
N
C6H4-p-F
H2
70
60
50
40
30
20
ppm
10
0
–10 –20 –30
8.5
8.0
7.5
7.0
4.0
ppm
3.5
2.0
1.5
1.0
c
B[2]
O
N
B[1] Ar
R2
[2] O
H2O
B
H2N
R2
H2N
B[1]
H
OH
[2]
Ar
DATB
1 or 8
O
[2]
[1]
B
Ar
B
[2]
B
B
O
[2]O
N
B
N
O
VII
IV
O
O
R2
R1
OH
R1
N
H
R1
H
R1
H
R2
H
R2
H
O
N
O
B[2]
N
OH
B[2]
B[2]O
OH
B[1]
B[1]
[2]O
Ar
Ar
B
N
N
O
O
V
VI
Figure 4 | NMR study and plausible catalytic cycle for the catalytic direct amidations promoted by DATB. a, 1H NMR spectra (THF-d8, 50 °C) of solutions
of (1) catalyst 1a, (2) amine 11a, (3) 1a + 11a, (4) acid 10b + 11a, and (5) 1a + 10b + 11a. *Peaks derived from the solvent. b, 11B NMR spectra (THF-d8, 50 °C)
of (1) catalyst 1a, (2) 1a + 11a, (3) 1a + 10b + 11a. c, Plausible catalytic cycle for the DATB-catalysed direct amidation.
chemical yield and catalyst loading, have only been partially resolved
Table 2 | Comparative study in the amidation of α-amino acids
so far. Another bottleneck is the limited substrate generality,
10p and 11r using DATB 8a.
especially with respect to sterically demanding carboxylic acids.
Recent studies by Bode, as well Lear and Hayashi on the synthesis
O
O
of amides bearing bulky substituents in non-catalytic processes,
illustrate the high demand for the development of new catalytic
protocols to access such amides46,47. We reasoned that the observed
lack of reactivity of such bulky carboxylic acids might originate from
the activation mode of boronic acid III, which solely activates car-
boxylic acids. A mechanistically distinct catalysis via II should
offer an opportunity to overcome this drawback and pave the way
for a more general catalytic amidation (Fig. 3a,b). Initially, we exam-
ined the amidation of the representative sterically congested sub-
strate 2-phenylisobutyric acid (10a) with 4-fluorobenzylamine
(11a). At 80 °C in the presence of powdered MS4Å (molecular
sieves 4 Å), 5 mol% of DATB 1a–c proved effective, albeit with
moderate conversions (Fig. 3c). DATB derivatives 8a–c, obtained
from the dimerization of intermediates 6a–c by the removal of the
B-dan group in the absence of external boronic acids (Fig. 2a),
generally provided higher conversions, presumably due to hydro-
gen-bonding interactions exerted by the pendant B–OH group
that direct the incoming carboxylic acid and stabilize the transition
H
H
Conditions
Ph
N
Ph
N
NH2
H
NH2
OH
N
+
N
H
H
O
O
O
O
11r
10p
39
Conditions
Reaction outcome
75% yield, stereochemically pure†
89% yield, diastereomixture (53/47)†
8a: 5 mol%, toluene, 50 °C, 14 h*
EDCI: 1.0 equiv, CH2Cl2, RT, 4 h
Catalyst 14 or 15 (10 mol%)
Toluene or CH2Cl2
50 °C (bath temperature), 24 h*
No conversion
Ph
O
EtN
Oxazolone 43
EDCI:
•
N
O
N
NMe2
*MS4Å was used for the reactions with 8a and 14. †HPLC traces are provided in Supplementary
Section 5–1.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
5
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2708
ARTICLES
state. This notion was supported by the diminished catalytic activity the presence of both acid 10b and amine 11a afforded an almost
of DATB derivatives 1d and 1e, which lack the B-hydroxy azaborine identical complex to the one observed above (Fig. 4a,b). Based on
moiety (Fig. 3c,e). Among the DATBs evaluated, 8a exhibited the the findings that the B3NO2 heterocycle is indispensable for the cat-
best catalytic performance to furnish 12 in 95% yield, and the reac- alysis and that it is constitutionally stable as evident from the cross-
tion proceeded even at room temperature. Azaborine 4a and borox- over experiment (Supplementary Fig. 9), we propose a plausible
ine 13 did not catalyse this reaction, thus confirming that the DATB reaction mechanism, as outlined in Fig. 4c. The formation of
core is responsible for the catalysis. Commercially available boronic amine complex IV should be kinetically and thermodynamically
acid catalyst 14 and metal catalyst Cp2HfCl2 15 were not effective for favoured. This would enable the incorporation of a carboxylic acid
a reaction involving bulky acid 10a (refs 34,38). For the amidation of to give VI by harnessing two B[2]–O bonds. Although carboxylic
less bulky acid 10b, DATB 8a, 14 and 15 exhibited comparable acid complexes were not observed in the NMR experiments, VI is
activity at room temperature, which illustrates that 8a not only cat- most probably formed as a short-lived transient species with an ada-
alyses the amidation of bulky acids, but is also as effective as the best mantyl skeleton that is analogous to that of calculated phosphora-
previously known catalysts for less bulky substrates (Fig. 3d).
mide I (Fig. 3a). Subsequently, VI should rapidly collapse under
Subsequently, we tested a range of carboxylic acids with equi- concomitant extrusion of the amide product and H2O, thus regen-
molar amounts of amines for this DATB-catalysed amidation. erating DATB. The DATB catalyst probably plays a pivotal role in
Under these conditions, the reactions reached completion within compensating the entropic cost on assembling the carboxylic acid
24 h (Table 1; for the full scope and limitations, see and the amine by harnessing the three Lewis acidic sites. Such a
Supplementary Figs 7,8). Sterically demanding aliphatic and aro- mechanism is distinct from previously reported dehydrative amida-
matic acids (10a,c–j) were successfully converted even with tions, where both the stoichiometric promoters and catalysts only
electron-rich aniline 11c. In case of 10g, the refluxing temperature activate carboxylic acids. This notion is partly supported by the dis-
permitted the use of a dropping funnel packed with MS4Å pellets, parity of the reaction outcomes in the amidation of 10p and 11r
allowing us to isolate amide 22 as a white solid on a gram scale (Table 2). While product 39 obtained by using DATB 8a proved
by simple filtration with 2 mol% catalyst (Supplementary Fig. 4). stereochemically pure in high performance liquid chromatography
Indole-containing substrates could be converted without a protect- (HPLC) analysis, the use of conventional coupling reagent EDCI
ing group. The amidation of benzoic acid 10n was examined with (1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide) caused exten-
various amines: 0.5 mol% of catalyst was sufficient for simple sive epimerization via oxazolone 43 to produce 39 as a diastereo-
primary amine 11d, and the reaction of electronically deactivated mixture (53/47). No conversion was observed for catalysts 14 and
primary amines, α-branched amines, and cyclic secondary amines 15, presumably due to the steric hindrance of the substrates34,38
11j–m proceeded smoothly. For α-tertiary amine 11n and acyclic These studies demonstrate that m-terphenyl-templated DATB
.
secondary amine 11o, elevated temperatures were required for the derivatives featuring a B3NO2 heterocyclic core are stable chemical
reaction to reach full conversion within 14 h. Chemoselective entities and outperform previously reported catalysts in direct
amide formation could be achieved without protection of the amidations. The broadened substrate scope appears to be driven
aniline moiety (amide 36). This catalytic amidation is tolerant by a distinct mechanism, and it should be possible to refine it
towards a variety of Lewis basic functionalities, while acid-labile even further through structural and electronic manipulations of
protecting groups remained unaffected (amides 19, 26–28 and DATB, which will significantly advance the sustainable and
33). The aforementioned results not only demonstrate the broad economical catalytic synthesis of amides.
substrate scope of this catalytic method, but also its high level of
functional group compatibility, and the mild reaction conditions. Methods
To a test tube equipped with a magnetically stirred chip was added powdered MS4Å
Furthermore, the stereochemical integrity of α-chiral carboxylic
acids and α-chiral amines was maintained during the reaction.
The amide formation of Bz-Val-OH 10p (Bz, benzoyl; Val,
(240 mg, 800 mg mmol−1), and the tube was flame-dried under reduced pressure. To
this, acid 10a (49.3 mg, 0.30 mmol, 1.0 equiv.), amine 11c (34.5 μl, 0.30 mmol, 1.0
equiv.), catalyst 8a (8.6 mg, 0.015 mmol, 5 mol%), and toluene (3.0 ml, 0.1 M) were
L-valine) with 11q or H-Leu-NH2 11r (Leu, L-leucine) without
epimerization underlines the potential of DATB catalysis for
peptide synthesis. The present protocol was also applied to the
synthesis of several biologically active amides. Hydroxylamine
hydrochloride 11s was treated with 1.0 equiv of Et3N to afford vor-
inostat (Zolinza)48. The amide formation between fully substituted
pyrrole-3-carboxylic acid 10r and amine 11t, followed by conden-
sation with 5-fluoro-2-oxindole, furnished sunitinib 41 (Sutent)49.
Functionalized benzoic acid 10s, bearing a free aromatic amine,
could be chemoselectively coupled with aliphatic amine 11u to
give amisulpride 42 (Solian)50. It is noteworthy that the amide
bond in 41 and 42 was constructed from identical substrate sets
using stoichiometric activating reagents. DATB-promoted catalysis
provided a catalytic alternative to access these important therapeutics.
Subsequently, NMR studies were conducted to gain deeper
insights into the DATB catalysis. To obtain simpler spectral pat-
terns, C2-symmetric 1a was selected and its spectral changes in
THF-d8 were monitored with sequential additions of 4-fluorobenzyl-
amine (11a) and heptanoic acid (10b) at various temperatures. On
adding 11a to 1a, clear peak shifts were observed in the 1H, 19F and
11B NMR spectra (Fig. 4a,b and Supplementary Fig. 11), revealing
that 11a coordinates to the B[1] atom of 1a, which is consistent
with the crystallography results (Fig. 2d). In sharp contrast,
adding acid 10b to 1a did not produce any detectable changes in
added successively. The suspension was stirred for 14 h at 80 °C. After the addition
of H2O, the aqueous phase was extracted with EtOAc (3×). The combined organic
phases were washed with sat. aq. NaHCO3 and brine, dried over Na2SO4, filtered,
and removed under reduced pressure. The obtained material was purified by flash
column chromatography to give 17 (71.9 mg, 89%).
Data availability. Experimental data are available from a public repository. The
crystallographic data have been deposited at the Cambridge Crystallographic Data
Centre (CCDC) as CCDC 1476512 (1a), 1453971 (8a) and 1487400 (9) and can be
obtained free of charge from the CCDC via www.ccdc.cam.ac.uk/getstructures.
Received 6 July 2016; accepted 23 November 2016;
published online 30 January 2017
References
1. Jiang, W., Li, Y. & Wang, Z. Heteroarenes as high performance organic
semiconductors. Chem. Soc. Rev. 42, 6113–6127 (2013).
2. Campbell, P. G., Marwitz, A. J. & Liu, S. Y. Recent advances in azaborine
chemistry. Angew. Chem. Int. Ed. 51, 6074–6092 (2012).
3. Baranac-Stojanović, M. Aromaticity and stability of azaborines. Chem. Eur. J. 20,
16558–16565 (2014).
4. Wang, X. Y., Wang, J. Y. & Pei, J. BN heterosuperbenzenes synthesis and
properties. Chem. Eur. J. 21, 3528–3539 (2015).
5. Bonifazi, D. et al. Boron-nitrogen doped carbon scaffolding: organic chemistry,
self-assembly and materials applications of borazine and its derivatives. Chem.
Commun. 51, 15222–15236 (2015).
6. Pattabiraman, V. R. & Bode, J. W. Rethinking amide bond synthesis. Nature 480,
471–479 (2011).
1
the H and 11B NMR spectra even at elevated temperatures, and
6
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2708
ARTICLES
7. Roughley, S. D. & Jordan, A. M. The medicinal chemist’s toolbox: an analysis
of reactions used in the pursuit of drug candidates. J. Med. Chem. 54,
3451–3479 (2011).
8. Dunetz, J. R., Magano, J. & Weisenburger, G. A. Large-scale applications of
amide coupling reagents for the synthesis of pharmaceuticals. Org. Process Res.
Dev. 20, 140–177 (2016).
9. Meller, A., Habben, C., Noltemeyer, M. & Sheldrick, G. M. Crystal and
molecular structure of 5-p-fluorophenyl-2,4.6-trimethylcyclo-l,3-dioxa-2.4,6-
triborane. Z. Naturforsch. B. 37, 1504–1506 (1982).
10. Oesterle, R., Maringgele, W. & Meller, A. Gezielte synthesen für boroxazine.
J. Organomet. Chem. 284, 281–289 (1985).
34. Gernigon, N., Al-Zoubi, R. M. & Hall, D. G. Direct amidation of carboxylic acids
catalyzed by ortho-iodo arylboronic acids: catalyst optimization, scope, and
preliminary mechanistic study supporting a peculiar halogen acceleration effect.
J. Org. Chem. 77, 8386–8400 (2012).
35. Liu, S. et al. Direct amidation of amino acid derivatives catalyzed by
arylboronic acids: applications in dipeptide synthesis. Eur. J. Org. Chem.
2013, 5692–5700 (2013).
36. Yamashita, R., Sakakura, A. & Ishihara, K. Primary alkylboronic acids as highly
active catalysts for the dehydrative amide condensation of α-hydroxycarboxylic
acids. Org. Lett. 15, 3654–3657 (2013).
37. Mohy El Dine, T., Erb, W., Berhault, Y., Rouden, J. & Blanchet, J. Catalytic
chemical amide synthesis at room temperature: one more step toward peptide
synthesis. J. Org. Chem. 80, 4532–4544 (2015).
38. Lundberg, H. & Adolfsson, H. Hafnium-catalyzed direct amide formation at
room temperature. ACS Catal. 5, 3271–3277 (2015).
39. Mohy El Dine, T., Rouden, J. & Blanchet, J. Borinic acid catalysed peptide
synthesis. Chem. Commun. 51, 16084–16087 (2015).
40. Tam, E. K. W., Rita, Liu, L. Y. & Chen, A. 2-Furanylboronic acid as an effective
catalyst for the direct amidation of carboxylic acids at room temperature. Eur. J.
Org. Chem. 2015, 1100–1107 (2015).
41. Ishihara, K. & Lu, Y. Boronic acid–DMAPO cooperative catalysis for
dehydrative condensation between carboxylic acids and amines. Chem. Sci.
7, 1276–1280 (2016).
11. Suzuki, A. Cross-coupling reactions of organoboranes: an easy way to
construct C–C bonds (Nobel lecture). Angew. Chem. Int. Ed. 50,
6722–6737 (2011).
12. Iwadate, N. & Suginome, M. Differentially protected diboron for regioselective
diboration of alkynes: internal-selective cross-coupling of 1-alkene-1,2-
diboronic acid derivatives. J. Am. Chem. Soc. 132, 2548–2549 (2010).
13. Dewar, M. J. S., Kubba, V. P. & Pettit, R. New heteroaromatic compounds. part I.
9-Aza-10-bora-phenanthrene. J. Chem. Soc. 3073–3076 (1958).
14. Chen, Z., Wannere, C. S., Corminboeuf, C., Puchta, R. & Schleyer, P. V. R.
Nucleus-independent chemical shifts (NICS) as an aromaticity criterion. Chem.
Rev. 105, 3842–3888 (2005).
15. Hall, D. G. Boronic Acids: Preparation and Applications in Organic Synthesis,
Medicine and Materials 2nd edn (Wiley-VCH, 2011).
42. Chaudhari, P. S., Salim, S. D., Sawant, R. V. & Akamanchi, K. G. Sulfated
tungstate: a new solid heterogeneous catalyst for amide synthesis. Green Chem.
12, 1707–1710 (2010).
43. Bannister, R. M., Brookes, M. H., Evans, G. R., Katz, R. B. & Tyrrell, N. D. A
scaleable route to the pure enantiomers of verapamil. Org. Process Res. Dev. 4,
467–472 (2000).
16. Imamura, G. et al. Electronic structure and graphenization of
hexaphenylborazine. J. Phys. Chem. C 116, 16305–16310 (2012).
17. Müller, M., Behnle, S., Maichle-Mössmer, C. & Bettinger, H. F. Boron–nitrogen
substituted perylene obtained through photocyclisation. Chem. Commun. 50,
7821–7823 (2014).
18. Tokunaga, Y. Boroxine chemistry: from fundamental studies to applications
in supramolecular and synthetic organic chemistry. Heterocycles
87, 991–1021 (2013).
19. El-Faham, A. & Albericio, F. Peptide coupling reagents, more than a letter soup.
Chem. Rev. 111, 6557–6602 (2011).
20. Constable, D. J. C. et al. Key green chemistry research areas—a perspective from
pharmaceutical manufacturers. Green Chem. 9, 411–420 (2007).
21. Charville, H., Jackson, D., Hodges, G. & Whiting, A. The thermal and boron-
catalysed direct amide formation reactions: mechanistically understudied yet
important processes. Chem. Commun. 46, 1813–1823 (2010).
22. Lanigan, R. M. & Sheppard, T. D. Recent developments in amide synthesis:
direct amidation of carboxylic acids and transamidation reactions. Eur. J. Org.
Chem. 2013, 7453–7465 (2013).
23. Lundberg, H., Tinnis, F., Selander, N. & Adolfsson, H. Catalytic amide
formation from non-activated carboxylic acids and amines. Chem. Soc. Rev.
43, 2714–2742 (2014).
24. Zheng, H. & Hall, D. G. Boronic acid catalysis: an atom-economical platform for
direct activation and functionalization of carboxylic acids and alcohols.
Aldrichmica Acta 47, 41–51 (2014).
44. Mylavarapu, R. K. et al. Boric acid catalyzed amidation in the synthesis of active
pharmaceutical ingredients. Org. Process Res. Dev. 11, 1065–1068 (2007).
45. Shinde, G. B., Niphade, N. C., Deshmukh, S. P., Toche, R. B. & Mathad, V. T.
Industrial application of the Forster reaction: novel one-pot synthesis of
cinacalcet hydrochloride, a calcimimetic agent. Org. Process Res. Dev. 15,
455–461 (2011).
46. Schafer, G., Matthey, C. & Bode, J. W. Facile synthesis of sterically hindered and
electron-deficient secondary amides from isocyanates. Angew. Chem. Int. Ed. 51,
9173–9175 (2012).
47. Li, J., Lear, M. J. & Hayashi, Y. Sterically demanding oxidative amidation of α-
substituted malononitriles with amines using O2. Angew. Chem. Int. Ed. 55,
9060–9064 (2016).
48. Marks, P. A. & Breslow, R. Dimethyl sulfoxide to vorinostat: development
of this histone deacetylase inhibitor as an anticancer drug. Nat. Biotechnol.
25, 84–90 (2007).
49. Faivre, S., Demetri, G., Sargent, W. & Raymond, E. Molecular basis for sunitinib
efficacy and future clinical development. Nat. Rev. Drug. Discov. 6, 734–745 (2007).
50. McKeage, K. & Plosker, G. L. Amisulpride. CNS Drugs 18, 933–956 (2004).
25. Ishihara, K. & Yamamoto, H. Arylboron compounds as acid catalysts in organic
synthetic transformations. Eur. J. Org. Chem. 1999, 527–538 (1999).
26. Ishihara, K., Ohara, S. & Yamamoto, H. 3,4,5-Trifluorobenzeneboronic acid as
an extremely active amidation catalyst. J. Org. Chem. 61, 4196–4197 (1996).
27. Ishihara, K., Kondo, S. & Yamamoto, H. 3,5-Bis(perfluorodecyl)phenylboronic
acid as an easily recyclable direct amide condensation catalyst. Synlett 2001,
1371–1374 (2001).
28. Latta, R., Springsteen, G. & Wang, B. Development and synthesis of an arylboronic
acid-based solid-phase amidation catalyst. Synthesis 2001, 1611–1613 (2001).
29. Arnold, K. et al. To catalyze or not to catalyze? Insight into direct amide bond
formation from amines and carboxylic acids under thermal and catalyzed
conditions. Adv. Synth. Catal. 348, 813–820 (2006).
30. Arnold, K., Davies, B., Hérault, D. & Whiting, A. Asymmetric direct amide
synthesis by kinetic amine resolution: a chiral bifunctional aminoboronic acid
catalyzed reaction between a racemic amine and an achiral carboxylic acid.
Angew. Chem. Int. Ed. 47, 2673–2676 (2008).
31. Al-Zoubi, R. M., Marion, O. & Hall, D. G. Direct and waste-free amidations and
cycloadditions by organocatalytic activation of carboxylic acids at room
temperature. Angew. Chem. Int. Ed. 47, 2876–2879 (2008).
Acknowledgements
This work is dedicated to Prof. Stuart L. Schreiber on the occasion of his 60th birthday. This
work was supported by a Grant-in-Aid for Young Scientists A (KAKENHI no. 25713002)
from the JSPS. N.K. thanks the JSPS for financial support via KAKENHI grant no.
JP16H01043 ‘Precisely Designed Catalysts with Customized Scaffolding’. H.N. is a JSPS
research fellow. The authors thank T. Kimura for the X-ray crystallographic analysis of 1a,
8a and 9, and they would also like to thank R. Sawa, Y. Kubota and K. Iijima for technical
assistance with the analysis of 11B NMR spectra.
Author contributions
H.N., M.S. and N.K. conceived and designed the experiments. H.N., M.F. and Y.A.
performed the experiments and analysed the data. H.N. conducted the density
functional theory calculations. H.N., M.S. and N.K. co-wrote the paper.
Additional information
Supplementary information and chemical compound information are available in the
online version of the paper. Reprints and permissions information is available online at
www.nature.com/reprints. Correspondence and requests for materials should be addressed to
M.S. and N.K.
32. Adolfsson, H., Lundberg, H. & Tinnis, F. Titanium(IV) isopropoxide as an
efficient catalyst for direct amidation of nonactivated carboxylic acids. Synlett 23,
2201–2204 (2012).
33. Lundberg, H., Tinnis, F. & Adolfsson, H. Direct amide coupling of non-activated Competing financial interests
carboxylic acids and amines catalysed by zirconium(IV) chloride. Chem. Eur. J.
18, 3822–3826 (2012).
The Institute of Microbial Chemistry (BIKAKEN) has filed a patent application on DATB
catalysts for general direct amide-forming reactions.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
7
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.