Moderately flexible trans-chelating diphosphine ligands consisting of a triazine unit showing high ligand acceleration effect for heck reaction

Article abstract of DOI:10.1246/bcsj.20170392

Originally designed trans-chelating diphosphine ligands with moderate flexibility that allows trans-to cis-isomerization and bidentate to monodentate complex form were developed. The novel ligand shows high ligand acceleration effect (LAE) for Heck reacti

Full text of DOI:10.1246/bcsj.20170392

Advance Publication Cover Page
Moderately Flexible Trans-Chelating Diphosphine Ligands Consisting of
Triazine Unit Showing High Ligand Acceleration Effect for Heck Reaction
Toshinobu Korenaga,* Ryo Kowata, Tomohiro Kotaki, Yuji Shibasaki,
Yoshiyuki Oishi, and Kazuaki Shimada
Advance Publication on the web January 15, 2018
doi:10.1246/bcsj.20170392
© 2018 The Chemical Society of Japan

Moderately Flexible Trans-Chelating Diphosphine Ligands Consisting of Triazine
Unit Showing High Ligand Acceleration Effect for Heck Reaction
Toshinobu Korenaga,* Ryo Kowata, Tomohiro Kotaki, Yuji Shibasaki, Yoshiyuki Oishi, and Kazuaki Shimada
Department of Chemistry and Biological Sciences, Faculty of Science and Engineering, Iwate University, 4-3-5 Ueda, Morioka, Iwate 020-
8551
E-mail: korenaga@iwate-u.ac.jp
Toshinobu Korenaga
Toshinobu Korenaga received a Ph.D. degree (2000) from Tokyo Institute of Technology. He has been an associate
professor at Iwate University since 2012. His current research focuses on development of organometallic catalysts and
organocatalysts by actively utilizing a computational chemistry.
Abstract
Originally designed trans-chelating diphosphine ligands
construction of the ligand skeleton. In place of the N-amide
moiety, we have chosen a 1,3,5-triazinyl group because the
aromatic ring of triazine readily connects with anion atom
species via nucleophilic reaction. We designed novel
diphosphine ligand 1, which comprises two diphenylphosphino
triazinyl groups linked by dioxy-aromatic spacer (Figure 2,
below). The ligand 1 is expected to allow free conformational
change between cis- and trans-complexes according to
conditions and monodentate coordination in some cases. In this
study, we synthesized novel diphosphine ligand 1 and applied it
to the Heck reaction. The detailed catalytic mechanism was then
considered by DFT calculations.
with moderate flexibility that allows trans- to cis-isomerization
and bidentate to monodentate complex form were developed.
The novel ligand shows high ligand acceleration effect (LAE)
for Heck reaction. The catalyst loading was decreased to 0.0001
mol%, and the turnover number (TON) reached 970,000. The
catalytic cycle was considered by DFT calculations. The
calculations revealed the following: 1) the insertion step in
monodentate pathway was the rate-determining step, 2) the high
LAE effect of the novel ligand was owing to a facile change of
conformation from trans to cis form and flexibility to avoid
steric repulsion between substrate and catalyst in the insertion
step as well.
(A)
(B)
(C)
(D)
(E)
(F)
P
P
P
P
P
P
M
M
M
M
M
M
1. Introduction
Diphosphine ligands play
a
significant role in
P
P
P
P
P
P
homogeneous transition metal-catalyzed reactions. Previous
reports have focused on the cis-chelating diphosphine ligand;
only a few trans-chelating diphosphine ligands have been
reported.1,2 Certain trans-chelating ligands enable metal-
catalyzed reactions with high efficiency.1b,1d,1e,1h,3 The flexibility
of trans-chelating diphosphine ligands is important to enhance
the ligand acceleration effect (LAE);1d,3e therefore, different
types of flexible ligand backbones have been designed. A
schematic of different ligand backbones is described in Figure 1.
The most rigid ligands are type-A, and the most flexible ligands
are type-F. The flexibility of ligand backbone, which can be
mutual isomerization between a trans- and a cis-complex and
dissociation of one phosphorous atom, was very important for
Pd-catalyzed coupling. However, excessively flexible
diphosphine ligands such as type-F produce a dimetal species,
M2L1a,3b,4 or M2L21d,1e,1f,3e type complex, rather than the desired
bidentate complex ML. Therefore, appropriate rigidity is
required to avoid undesirable structural complexes.
We have chosen to investigate the type-D skeleton because
it has both flexibility and moderate rigidity. This type of ligand
has been reported by Süss-Fink et al. (dppbH)5 (Figure 2, above).
The dppbH has N-amide moieties to combine sp2 components
and can form both cis- and trans-chelate complexes with
transition metals. However, the cis complex of Pd/dppbH had
square-planar geometry comprising two P atoms and two N
atoms as well. To facilitate flexible coordination leading to cis-
and trans-complexes using diphosphine of type D, rigid N-amide
moiety should not be employed. However, the N-amide moiety
was used as a connecting unit between aromatics to facilitate the
more flexible
= aromatic ring or sp2 (sp) group
= hetero or carbon atom
Figure 1. Flexibility of trans-chelating diphosphine ligands
Süss-Fink et al. (2003)
NH PPh2
O
Cl Pd Cl trans
[PdCl2(cod)]
[PtCl2(cod)]
NH PPh2
NH PPh2
NH PPh2
N
N
PPh2
PPh2
cis
Pt
O
dppbH
this work
P
monodentate
M
PhO
N
N
PPh2
M
N
P
P
O
M
P
O
N
N
N
P
bidentate-trans
M
PPh2
P
PhO
1
bidentate-cis
Figure 2. Coordination mode of designed diphosphine ligand

4a (309 mg, 0.59 mmol), toluene (6.0 mL) and
diphenyl(trimethylsiliyl)phosphine (600 μL, 2.34 mmol) were
added to a dried 20 mL Schlenk flask with argon. The reaction
mixture was stirred at 100 °C until loss of
diphenyl(trimethylsiliyl)phosphine in situ (determined by 31P
NMR). After concentration of reaction mixture under reduced
pressure, the resulting solid was purified by silica gel
2. Experimental
1,2-Bis(2-chloro-4-phenoxy-1,3,5-triazinyloxy)benzene (4a)
2,4-Dichloro-6-phenoxy-1,3,5-triazine (7.26 g, 30.0 mmol),
prepared from cyanuric chloride with an equimolar amount of
phenol in chloroform / alkaline media (NaOHaq) in the presence
of tetrabutylammonium bromide (5 mol%), was dissolved in
THF (100 mL) in a flame-dried 300 mL round-bottomed flask
equipped with a dropping funnel. Then, the solution was cooled
to 0 °C. To the solution was added dropwise 100 mL THF
solution of catechol (1.65 g, 15 mmol) and N,N-
diisopropylethylamine (4.85 g, 37.5 mmol). And then, the
reaction mixture was stirred at 0 °C for 2 h, r.t. for 2 h, 40 °C for
3 h, and 60 °C for 12 h. The resulting mixture was extracted with
EtOAc, and treated in a usual manner to give a crude product.
The resulting solid was purified by silica gel column
chromatography (CHCl3), followed by recrystallization with
chloroform/hexane mixed solvent to give the target product as a
white solid (2.03 g, 3.90 mmol, 26% yield). M.p. 74-76 °C, 1H
NMR (400 MHz, CDCl3) δ 7.11 (d, J = 7.7 Hz, 4H), 7.25-7.33
(m, 6H), 7.38 (t, J = 7.7 Hz, 4H). 13C NMR (101 MHz, CDCl3)
δ 121.2, 123.5, 126.6, 127.6, 129.8, 142.6, 151.3, 171.9, 172.5,
173.8. IR [KBr (cm-1)]: 3051, 1553, 1490, 1274, 1164, 831, 767.
Elemental analysis (C24H14Cl2N6O4) Calcd. C: 55.29; H: 2.71;
N:16.12 (%) Found. C: 55.52; H: 2.80; N:16.13 (%).
chromatography
with
CHCl3
and
recrystallization
(CHCl3/hexane) to give 211 mg of 1a (43% yield) as a white
solid. M.p. 88 °C, 1H NMR (400 MHz, CDCl3) δ 6.98 (d, J = 7.5
Hz, 4H), 7.12-7.30 (m, 22H), 7.35-7.40 (m, 8H). 31P NMR (162
MHz, CDCl3) δ 2.99 (s). 13C NMR (101 MHz, CDCl3) δ 121.6,
123.5, 125.8, 126.7, 128.4 (d, J = 8.1 Hz), 129.3, 129.4, 132.84
(d, J = 4.7 Hz), 134.9 (d, J = 19.8 Hz), 143.3, 151.8, 170.4 (d, J
= 7.2 Hz), 170.9 (d, J = 6.8 Hz) 192.18. IR (KBr, cm-1):
3056,1544,1487,1348,1204. HRMS(FAB) m/z calcd for
C48H35N6O4P2: 821.2195. Found: 821.2182.
1,3-Bis(2-diphenylphosphino-4-phenoxy-1,3,5-
triazinyloxy)benzene (1b)
Ligand 1b was obtained from 4b (309 mg, 0.59 mmol) in a
similar manner to the case of 1a (404 mg, 0.49 mmol, 83% yield,
as a white solid). M.p. 97 °C, 1H NMR(400 MHz, CDCl3) δ 6.86
(t, J = 2.2 Hz, 1H), 6.92 (dd, J = 2.2, 8.2 Hz, 2H), 7.07 (d, J =
7.6 Hz, 4H), 7.18-7.37 (m, 19H), 7.41-7.46 (m, 8H). 31P
NMR(162 MHz, CDCl3) δ 3.13(s). 13C NMR(101 MHz, CDCl3)
δ 115.4, 119.0, 121.6, 126.0, 128.5 (d, J = 8.0 Hz), 129.5, 129.6,
132.7 (d, J = 4.8 Hz), 134.9 (d, J = 19.9 Hz), 151.8, 152.0, 170.7
(d, J = 40.8 Hz), 174.1, 192.4. IR (KBr, cm-1): 3062, 2361, 1538,
1342, 1210. HRMS(FAB) m/z calcd for C48H35N6O4P2:
821.2195. Found: 821.2187.
1,3-Bis(2-chloro-4-phenoxy-1,3,5-triazinyloxy)benzene (4b)
4b was obtained from resorcinol (1.65 g, 15 mmol) in a similar
manner (recrystallization with chloroform/hexane) to the case of
4a (2.19 g, 4.20 mmol, 28% yield, as a white solid). M.p. 86-
1
88 °C, H NMR (400 MHz, CDCl3) δ 7.05 (t, J = 2.2 Hz, 1H),
7.09 (dd, J = 8.2, 2.2 Hz, 2H), 7.14 (d, J = 8.2 Hz, 4H), 7.25-
7.29 (m, 3H), 7.41 (t, J = 8.2 Hz, 4H). 13C NMR (101 MHz,
CDCl3) δ 115.1, 119.5, 121.2, 126.7, 129.8, 130.3, 151.3, 151.7,
172.0, 172.5, 174.0. IR [KBr (cm-1)]: 3056, 1550, 1488, 1274,
1165, 834, 769. Elemental analysis (C24H14Cl2N6O4) Calcd. C:
55.29; H: 2.71; N:16.12 (%) Found. C: 55.33; H: 2.79; N:16.17
(%).
1,4-Bis(2-diphenylphosphino-4-phenoxy-1,3,5-
triazinyloxy)benzene (1c)
Ligand 1c was obtained from 4c (309 mg, 0.59 mmol) in a
similar manner to the case of 1a (359 mg, 0.44 mmol, 74% yield,
1
as a white solid). M.p. 227 °C, H NMR(500 MHz, CDCl3) δ
6.97 (br-s, 2H), 7.08 (d, J = 7.5 Hz, 2H), 7.21-7.37 (m, 22H),
7.42-7.45 (m, 8H). 31P NMR(202 MHz, CDCl3) δ 3.09 (s). 13C
NMR(126 MHz, CDCl3) δ 121.6, 122.4, 126.0, 128.5 (d, J = 7.9
Hz), 129.5, 129.6, 132.8 (d, J = 5.1 Hz) 134.9 (d, J = 19.5 Hz),
149.1, 151.8, 170.9, 171, 192.4. IR (KBr, cm-1): 3058, 1541,
1487, 1344, 1213, 1181. HRMS(FAB) m/z calcd for
C48H35N6O4P2: 821.2195. Found: 821.2195.
1,4-Bis(2-chloro-4-phenoxy-1,3,5-triazinyloxy)benzene (4c)
4c was obtained from hydroquinone (1.65 g, 15 mmol) in a
similar manner (recrystallization with chloroform/hexane) to the
case of 4a (3.52 g, 6.75 mmol, 45% yield, as a white solid). M.p.
86-88 °C, 1H NMR (400 MHz, CDCl3) δ 7.14-7.20 (m, 8H), 7.29
(t, J = 7.7 Hz, 2H), 7.42 (t, J =7.7 Hz, 4H). 13C NMR (101 MHz,
CDCl3) δ 121.3, 122.6, 126.8, 129.9, 149.1, 151.4, 172.3, 172.5,
174.0 IR [KBr (cm-1)]: 3055, 1576, 1504, 1250, 1093, 858, 761.
Elemental analysis (C24H14Cl2N6O4) Calcd. C: 55.29; H: 2.71;
N:16.12 (%) Found. C: 55.07; H: 2.88; N:16.09 (%).
2,7-bis(2-diphenylphosphino-4-phenoxy-1,3,5-
triazinyloxy)Naphthalene (1d)
Ligand 1d was obtained from 4d (341 mg, 0.60 mmol) in a
similar manner to the case of 1a (265 mg, 0.30 mmol, 51% yield,
as a white solid). M.p. 205 °C, 1H NMR(400 MHz, DMSO-d6) δ
7.19-7.45 (m, 32H), 7.64-7.70 (m, 2H), 7.94 (d, J = 9.0 Hz, 2H).
31P NMR(162 MHz, DMSO-d6) δ 3.12(s). 13C NMR(101 MHz,
DMSO-d6) δ 117.8, 121.3, 125.8, 128.4 (d, J = 8.3 Hz), 128.9,
129.2, 129.4 (d, J = 7.9 Hz), 131.7 (d, J = 9.7 Hz), 132.7 (d, J =
4.4 Hz), 133.7, 134.5, 134.7, 149.7, 151.4, 170.4 (d, J = 7.3 Hz),
170.5 (d, J = 7.2 Hz), 190.5 (d, J = 6.2 Hz). IR (KBr, cm-1): 3056,
2482, 1530, 1425, 1338, 1204. HRMS(FAB) m/z calcd for
C52H37N6O4P2: 871.2352. Found: 871.2384.
2,7-Bis(2-chloro-4-phenoxy-1,3,5-triazinyloxy)naphthalene
(4d)
4d was obtained from 2,7-dihydroxynaphthalene (2.40 g, 15
mmol) in a similar manner to the case of 4a (3.68 g, 6.45 mmol,
1
43% yield, as a white solid). M.p. 131-133 °C, H NMR (400
MHz, CDCl3) δ 7.14 (dd, J = 1.0, 8.5 Hz, 4H), 7.24 (t, 2H), 7.26-
7.31 (m, 2H), 7.38 (t, J = 8.5 Hz, 4H), 7.56 (d, J = 2.2 Hz, 2H),
7.89 (d, J = 9.0 Hz, 2H). 13C NMR (101 MHz, CDCl3) δ 118.4,
121.0, 121.3, 126.7, 129.8, 129.9, 130.1, 134.3, 149.8, 151.3,
172.5, 172.6, 173.9. IR [KBr (cm-1)]: 3057, 1531, 1491, 1295,
1164, 837, 767. Elemental analysis (C28H16Cl2N6O4) Calcd. C:
58.86; H: 2.82; N:14.71 (%) Found. C: 58.76; H: 2.99; N:14.85
(%).
Reaction of [{RhCl(CO)2}2] with 1a
Complex of [RhCl(CO)2]2 (7.18 mg, 0.019 mmol), 1a (30.0 mg,
0.037 mmol) and CH2Cl2 (2.0 mL) were added to a dried 20 mL
Schlenk flask under argon.. The reaction mixture was stirred at
room temperature for 1 h. After concentration under reduced
pressure, the yellow solid was obtained. The product was
measured by 31P NMR without purification. 31P NMR (162 MHz,
1,2-Bis(2-diphenylphosphino-4-phenoxy-1,3,5-
triazinyloxy)benzene (1a)

C6D6) δ 33.9(d, J = 133 Hz).
Cl
N
Cl
HO
OH
+
1/2
Reaction of [{RhCl(CO)2}2] with 1b
N
N
iPr2NEt
THF
The reaction and NMR analysis using 1b (30.0 mg, 0.037
mmol) were performed in a similar manner to the case of 1a. 31
NMR (162 MHz, C6D6) δ 35.7(d, J = 134 Hz).
OPh
P
2
3
N
O
O
N
PhO
PhO
OPh
Reaction of [{RhCl(CO)2}2] with 1c
Ph2PSiMe3
N
N
N
N
N
N
N
The reaction and NMR analysis using 1c (30.0 mg, 0.037
toluene
100 oC
mmol) were performed in a similar manner to the case of 1a. 31
P
Cl
Cl
N
4
NMR (162 MHz, C6D6) δ 34.9 (d, J = 129 Hz), 40.2(d, J = 138
Hz).
O
O
N
N
OPh
Reaction of [{RhCl(CO)2}2] with 1d
The reaction and NMR analysis using 1d (38.4 mg, 0.037
PPh2
PPh2
1
mmol) were performed in a similar manner to the case of 1a. 31
P
NMR (162 MHz, C6D6) δ 34.8(d, J = 131 Hz), 35.8 (d, J = 133
: aromatic spacer
(yields from 4 to 1)
Hz).
Heck reaction
1a:
1c:
(43%)
1b:
(83%)
A dried 100 mL Schlenk flask was flushed with argon and
charged with 1a (1.62 mg, 1.97 × 10-3 mmol), PdCl2(CH3CN)2
(0.51 mg, 1.97 × 10-3 mmol) and CH2Cl2 (0.4 ml). The mixture
was stirred at room temperature for 10 min. The reaction mixture
was concentrated under reduced pressure to give a Pd/1a catalyst
a yellow solid. Then, Et3N (550 μL), 5 (440 μL, 3.93 mmol) and
6 (640 μL, 4.98 mmol) were added under argon atmosphere. The
reaction mixture was stirred at 100 °C for 1 h. After addition of
EtOAc, the mixture was washed by water. The organic layer
were dried over Na2SO4 and concentrated under reduced
pressure. The crude mixture was purified by silica gel column
chromatography with hexane/EtOAc = 20 : 1 to give 7. 1H NMR
(400MHz, CDCl3) δ 1.54 (s, 9H), 6.37 (d, J = 16.0 Hz, 1H), 7.35
- 7.39 (m, 3H), 7.50 - 7.52 (m, 2H), 7.59 (d, J = 16.0 Hz, 1H)
13C NMR(101 MHz, DMSO-d6) δ 28.3, 80.6, 120.3, 128.1 ,
128.9, 130.1, 134.8, 143.7, 166.5.
1d:
(74%)
(51%)
Figure 3. Syntheses of designed ligands 1
The formation of cis- or trans-chelating complex can be
readily detected by 31P NMR using [RhCl(CO)(diphosphine)]
with 1. If cis complex is formed, two types of signals are
observed as double doublet (e.g., in the case of cis-
[RhCl(CO)(binap)] δ 26.0, dd, JRhP = 128 Hz, JPP = 44 Hz, and
δ 46.5, dd, JRhP = 163 Hz, JPP = 44 Hz).8 If trans complex is
formed, one type of signal is observed as doublet (e.g., in the
case of trans-[RhCl(CO)(dppbH)] δ 37.1, d, JRhP = 130 Hz).3g
2:1 mixture of 1 and [RhCl(CO)2]2 (ligand:Rh = 1:1) in C6D6
were measured using 31P NMR (Figure 4). For 1a and 1b, one
type of signal was observed without free ligand signal. The
A
3. Results and Discussion
The ligands 1a-1d were synthesized from 2,4-dichloro-6-
phenoxy-1,3,5-triazine (2)6 (Figure 3). The reaction of 2 with 0.5
equiv. of pyrocatechol (3a), resorcinol (3b), hydroquinone (3c),
or naphthalene-2,7-diol (3d) in the presence of iPr2NEt gave
corresponding 4 in medium yields. The corresponding ligands 1
chemical shifts (δ 33.9 or 35.7) and coupling constants (JRh-P
,
133 or 134 Hz) indicated that trans-[RhCl(CO)(1)] complexes
were formed as expected. In contrast, for 1c and 1d, two types
of signals were observed without free ligand signal. However, it
was evident that their complexes were not cis-[RhCl(CO)(1)]
because the observed coupling constants were not 31P–31P
coupling (ca. 40 Hz) but instead Rh-P coupling (ca. 130 Hz).
Unfortunately, the two types of complexes could not be isolated.
Because the ligands 1c and 1d have a wide spacer as compared
with 1a and 1b, the unknown complexes might be a binuclear
complex.1d,1e,1f,3e
were obtained via reaction of
4
with diphenyl
trimethylsilylphosphine7 in moderate to high yield.

BINAP, DPPP, DPPPE (1,5-Bis(diphenylphosphino)pentane),
and Xantphos were not effective at any bite angle (β)11,12 (entries
9-12). Even the wider bite angle of Xantphos produced poor
yields (entry 12) because the coordination mode was forced to
cis form. The results indicate that flexible diphosphine 1b is
suitable for the Heck reaction and decreases the necessary
catalyst loading of Pd/1b. The reaction of 5 with 6 using 0.001
mol% of Pd/1b catalyst (S/C = 100,000) at 100 °C for 9 h
resulted in 89% yield of 7 (entry 3). Using 0.0001 mol% of Pd/1b
catalyst (S/C = 1,000,000) resulted in 97% yield of 7 (TON =
970,000), although long reaction time (96 h) was required (entry
4).
Cl
1a or 1b
P
Rh
P
OC
trans-chelating
[RhCl(CO) 2]2
Cl
unknown complex
+
P
Rh
P
1c or 1d
OC
trans-chelating
[RhCl(CO)2]2 + 1a
Table 1. Heck reaction using designed ligands 1
δ 33.9 (d, JRh-P = 133 Hz)
O
I
O
tBu
Pd cat.
+
tBu
O
O
Et3N
[RhCl(CO)2]2 + 1b
δ 35.7 (d, JRh-P = 134 Hz)
100 °C
5
6
7
β
Pd
Time Yield of
Entry
Ligand
(deg) (mol%)
(h)
1
1
9
96
1
1
7 (%)
70
96
89
97
9
1a)
2a)
3b)
4c)
5a)
6a)
7d)
1a
1b
1b
1b
1c
0.05
0.05
0.001
0.0001
0.05
[RhCl(CO)2]2 + 1c
δ 34.9 (d, JRh-P = 129 Hz)
δ 40.2 (d, JRh-P = 138 Hz)
1d
PPh3
0.05
0.05
40
54
1
δ 35.8 (d, JRh-P = 133 Hz)
[RhCl(CO)2]2 + 1d
MeO
N
N
N
8d)
0.05
1
45
PPh2
δ 34.8 (d, JRh-P = 131 Hz)
MeO
9a)
BINAP
Dppp
Dpppe
85
87
93
0.05
0.05
0.05
0.05
1
1
1
1
17
15
2
10a)
11a)
12a)
40
30
20
Xantphos
111
2
Figure 4. 31P NMR of Rh/1 complexes
a) The reactions were carried out using 0.05 mol% of
PdCl2(CH3CN)2, 0.05 mol% of ligand, 1.25 equiv. of tert-butyl
acrylate based on phenyl iodide at 100 °C for 1 h. b) 0.001 mol%
of PdCl2(CH3CN)2, 0.001 mol% of 1b were used. c) 0.0001
mol% of PdCl2(CH3CN)2, 0.0001 mol% of 1b were used. d) 0.05
mol% of PdCl2(CH3CN)2, 0.1 mol% of ligand was used.
Following the detection of cis- and trans-chelating
complexes, the trans-chelating ligand 1 was used for a
palladium-catalyzed reaction. Many trans-chelating ligands
have given good results for cross-coupling. For example,
Protasiewicz et al. reported the Heck reaction using 0.01 mol%
of the trans-chelating diphosphine (TON = 10,000).3f Kawano
and Ueda et al. also reported the Heck reaction using 0.001
mol% of the trans-bidentate pyridine ligand (TON = 79,630).9
To compare with these previous reports, we applied the novel
ligand 1 to the Heck reaction. The reactions of iodobenzene (5)
with tert-butyl acrylate (6) were performed in the presence of
0.05 mol% of PdCl2(CH3CN)2 with ligand 1 in Et3N at 100 °C
for 1 h (Table 1). The most effective ligand was 1b that gave
96% of coupling product 7 (entry 2). The second-best ligand was
1a (70% yield of 7) (entry 1). In contrast, the ligands 1c and 1d,
which resulted in unknown complexes in the reaction with
[RhCl(CO)2]2, afforded unsatisfactory results (9% or 40% yield)
(entries 5 and 6). The results showed that the spacer size of
ligand 1 plays a significant role in the catalytic activity. An
overly wide spacer appears to provide inappropriate structure for
the catalytic reaction. The catalytic activity of Pd/1b catalyst was
compared with other phosphine ligands. Triphenylphosphine
ligand produced 7 in medium yield (entry 7). Triazinyl
diphenylphosphine,10 which has approximately the same
electronic ability as 1b, gave 45% yield of 7 (entry 8), indicating
that the bidentate diphosphine plays a role in catalytic efficiency
by reason of stability enhancement by chelating effect of 1b on
Pd catalyst to prevent decomposition. However, the bidentate
diphosphine is not necessarily efficient. The bidentate ligands
The high TON achieved in this experiment could be the
result of 1b having the ability to stabilize the Pd complex by
bidentate form and the capacity for flexible conformation change
between cis and trans mode. To obtain intimate information of
Pd/1b catalyst, the catalytic processes, oxidative addition,
insertion, and β-elimination were considered by DFT
calculations using 5 and methyl acrylate as a model structure of
6. Although several calculations for Heck reaction by DFT have
been reported,13 we referred to the study for the Heck reaction
catalyzed by palladium with two monophosphine ligand. In the
previous reports, two pathways, i.e., one or two phosphine-
coordinated Pd complexes including neutral or cationic Pd(II),
were considered.13,14 When we analyzed the Heck reaction
catalyzed by Pd/1b by DFT calculations, all structures, including
transition states, were fully optimized and characterized using
frequency calculations at the B3LYP functional15 with the 6-
31G(d) basis set16 for the organic molecules and the LanL2DZ
basis set16 (with effective core potentials) for palladium and
iodine atoms using Gaussian 09, revision E.01.17 Before the DFT
calculations, conformational search of 1b ligand was performed
at the PM6 level using Spartan ʼ14 to narrow down candidates
for the conformation of highly flexible 1b.18 Gibbs free energies
(298.15 K, 1 atm) were initially computed for the gas phase.
Relative Gibbs free energies in solvent were obtained using

single-point energy calculations of the optimized structures at
the M06-2X functional with the 6-31G(d) basis set for the
organic molecules and the LanL2DZ basis set with the SCRF
method based on CPCM (Et3N)19 followed by the addition of
thermal corrections, which were calculated using the above
mentioned geometrical optimization. The calculations results are
shown in Figure 5.
Figure 5. Energy diagram of oxidative addition, insertion, β-elimination through both of monodentate and bidentate coordination
process using ligand 1b
Although the optimization of both cis and trans-1b ligated-
Pd (bi-8) as initial structures was conducted, both of the
optimized calculations gave only the trans-8 structure. Although
the bidentate structure of Ph2P-Pd-OPh complex (trans-bi-P,O in
Figure 5) could be found out, the energy was higher than that of
trans-bi-8 by 12.3 kcal/mol. In monodentate coordination, one
phosphorous group is eliminated from trans-bi-8 to give
monodentate coordinated complex (mono-8) with an increase of
energy by 9.9 kcal/mol. After the coordination of 5 to mono-8,
the oxidative addition proceeds via transition state mono-TSOxAd
to give mono-10. The activation energy of mono-TSOxAd from
trans-bi-8 was 14.3 kcal/mol. Conversely, in bidentate
coordination, coordination of 5 to trans-bi-8 and next transition
state bi-TSOxAd involved a decrease of bite angle to give cis
complex (177.1° for trans-bi-8 to 119.7° for bi-TSOxAd).
Activation energy of bi-TSOxAd from trans-bi-8 (10.1 kcal/mol)
was lower than that of mono-TSOxAd, indicating that the oxidative
addition was advantageous to the bidentate coordination process.
Both the monodentate and bidentate processes gave stable
Pd(II)-Ph complex cis-bi-10. The complex further stabilized by
isomerization to trans-bi-10, and the energy difference was only
1.4 kcal/mol (−14.0 kcal/mol for cis-bi-10 vs. −16.6 kcal/mol for
trans-bi-10). The more stable trans-bi-10 was at resting state
because it did not undergo insertion in its original form. The
four-atom coordination mode is unsuitable for the insertion step
because a vacant site is required for the coordination of methyl
acrylate. If the process without dissociation of iodine ion or P
atom of 1b is possible, methyl acrylate should be able to bind to
cis-bi-10 or trans-bi-10.20 However, the structural optimization
of transition states using cis-bi-10 and methyl acrylate as an
initial structure gave bi-TSIns+ with desorbed iodine ion.
Therefore, the monodentate coordination process (= neutral
pathway) or bidentate coordination process (= cationic pathway
without iodine ion) would be suitable for insertion. In the former
process, mono-10 is an unsuitable initial structure of insertion
because vacant site is located to trans-toward phenyl group. The
complexes of mono-11 derived from cis-bi-10 and mono-11-t
derived from trans-bi-10 are appropriate for insertion (Figure 6).
The mono-11-t is more stable than mono-11. However, the
transition state of mono-TSIns derived from mono-11 is more
stable than mono-TSIns-t derived from mono-11-t. Therefore, the
insertion in monodentate coordination process is via cis-bi-10
and mono-11.
MeO 2C
P
P Pd
Ph
P
P
I
P Pd
Ph
P
Pd
P
P
I
Pd
Ph
mono-11
I
I
Ph
Ins
trans-bi-10
c is -bi- 10
mono-TS
9.7
-16.6
[0.0]
-14.4
[1.6]
6.8
[23.4]
[26.3]
I
MeO 2C
I
Pd
Ph
P
P
Pd
P
P
Ph
mono-11 -t
mono-TSIns -t
-0.7
12.6
[15.9]
[29.2]
Figure 6. Comparison of monodentate coordination processes
The activation energy of bi-TSIns+ (39.6 kcal/mol) in the
cationic pathway was largely higher than that of mono-TSIns (9.7
kcal/mol), indicating that the insertion proceeded through the
monodentate coordination process.
The β-elimination process requires a vacant site at the cis
position of the reactant. Therefore, a monodentate model or
cationic bidentate model is suitable for the insertion (Figure 5).
Considered only from the point of view of comparison of both
processes in β-elimination, the favorable activation energy of
transition state from corresponding complex 13 was cationic
bidentate model (cis-bi-13+→bi-TSElm+:14.7 kcal/mol vs. mono-
13→mono-TSElm: 18.4 kcal/mol). However, the energy of cis-bi-
13+ was higher than that of mono-13 by 31.9 kcal/mol. Therefore,
the β-elimination step will proceed through monodentate
coordination process.
When overviewing these calculations, the oxidative
addition proceeds by bidentate coordination with conformational
change of trans to cis chelation mode. The insertion and β-

elimination steps proceed through cationic monodentate
coordination, and the rate-determining step is the insertion step.
To clarify the structural effect of 1b, the case of using
BINAP ligand was estimated by calculations in the same manner
as above. The conformationally rigid and cis-chelating BINAP
was relatively ineffective for the Heck reaction as compared with
1b (Table 1, entry 9). The calculation results are shown in Figure
7.
Figure 7. Energy diagram of oxidative addition, insertion, β-elimination thorough both of monodentate or bidentate coordination
process using BINAP ligand
Although negligible difference between the monodentate
and bidentate process was shown in the oxidative addition
reaction, the monodentate process is obviously superior in
insertion and β-elimination steps, similar to the case of 1b.
Unexpectedly, one phosphorous group of rigid BINAP ligand
was eliminated from Pd, i.e., the energy difference between
mono-20 and cis-bi-20 was merely 1.6 kcal/mol.21 However, the
activation energy of insertion (cis-bi-20→mono-TSIns,BP: 31.9
kcal/mol), which was the rate-determining step, was larger than
that of 1b by 5.6 kcal/mol, where the calculation results are in
agreement with the experimental results (entry 2 vs. 9 in Table
1). The results indicated that the ease of the elimination of one
phosphorous group, resulting in monodentate coordination
complex, is not a major reason for the high catalytic activity of
1b. For further consideration in detail, structures of the insertion
step in both Pd/1b and Pd/BINAP were compared (Figure 8). In
the case of Pd/1b, coordination of methyl acrylate to mono-11
stabilized by 4.1 kcal/mol to give mono-12 (Figure 8A).
Although the coordination of methyl acrylate involved rotation
of P–Pd bond, torsion angle of DI-Pd-P-C(TZ) = −52.7° for mono-11
→ −38.6° for mono-12, filling the vacant site by coordination of
methyl acrylate to Pd stabilized mono-12. In the following
transition state of insertion, to approach the reactive sites
between ipso-carbon of phenyl group and β-carbon of methyl
acrylate, methyl acrylate is required to turn in the direction of a
Ph2P- group. At that time, the Ph2P- group can rotate for escaping
from methyl acrylate and torsion angle DI-Pd-P-C(Ph) of = −157.8°
mono-12 → −150.3° for mono-TSIns because of the highly
flexible structure of 2b. In contrast, the coordination of methyl
acrylate to mono-21 destabilized the complex mono-22 by 8.8
kcal/mol in the case of Pd/BINAP (Figure 8B). The
destabilization is due to the steric hindrance between the reactive
phenyl group and the skeleton of BINAP after rotation of P–Pd
bond and the torsion angle of DI-Pd-P-C(Np) = 173.2° for mono-21
→ −92.7° for mono-22, unlike the case of Pd/1b. Likewise, the
energy difference between mono-22 and mono-TSIns,BP was
larger than that of the case of Pd/1b (12.4 kcal/mol for
Pd/BINAP vs. 7.0 kcal/mol for Pd/1b). Steric repulsion between
methyl acrylate and Ph2P- group was observed due to less
rotation of Ph2P- group by regulation of skeleton of BINAP and
torsion angle DI-Pd-P-C(Ph) of = 149.9° mono-22 → 150.7° for
mono-TSIns,BP. The calculation models indicate that the real
system, i.e., using tert-butyl acrylate (6), should generate larger
difference of the activation energy of insertion in the cases
between 1b and BINAP, because the bulky tert-butyl group
should have a significant steric effect on mono-TSIns,BP
.
Therefore, we calculated transition states of insertion using
substrate 6 (Figure 9). Although the activation energy from
mono-12 using 6 in the case of Pd/1b was almost the same with
that using methyl acrylate (7.2 vs. 7.0 kcal/mol), the activation
energy from mono-22 using 6 in the case of Pd/BINAP was
larger than that using methyl acrylate by 1.0 kcal/mol (13.4 vs.
12.4 kcal/mol). The results are in agreement with the large
difference of catalytic activity between Pd/1b and Pd/BINAP in
Table 1.

Figure 8. Comparison of insertion step in the cases of 1b and BINAP.
addition step. The calculation results indicated that the insertion
step was the rate-determining step. Furthermore, the calculated
structures of the insertion step between the cases of 1b and
BINAP were compared. The comparison clarified that the
flexibility of the ligand 1b assists the stabilization of the
transition state of insertion by avoiding a steric repulsion
between substrate 6 and pendant phenyl groups of the catalyst.
Acknowledgment
This work was supported in part by JSPS KAKENHI
(Grant-in Aid for Scientific Research (C), no. 16K05764 (T.K.))
and a special grant from Iwate University (Kyoten-keisei-
Project).
References
1.
Recent examples: a) N. Nasser, R. J. Puddephatt, J. Inorg.
Organomet. Polym. Mater. 2017, 27, S76. b) K. Hirasawa,
S. Tanaka, T. Horiuchi, T. Kobayashi, T. Sato, N.
Morohashi, T. Hattori, Organometallics 2016, 35, 420. c)
A. E. Pascui, K. van Rees, D. W. Zant, D. L. J. Broere, M.
A. Siegler, J. I. van der Vlugt, Eur. J. Inorg. Chem. 2015,
34, 5687. d) C. C. Comanescu, V. M. Iluc, Inorg. Chem.
2014, 53, 8517. e) P. Stepnicka, B. Schneiderova, J. Schulz,
I. Cisarova, Organometallics 2013, 32, 5754. f) N. Nasser,
A. Borecki, P. D. Boyle, R. J. Puddephatt, Inorg. Chem.
2013, 52, 7051. g) Jarvis, A. G.; Sehnal, P. E.; Bajwa, S.
E.; Whitwood, A. C.; Zhang, X.; Cheung, M. S.; Lin, Z.;
Fairlamb, I. J. S., Chem. -Eur. J. 2013, 19, 6034. h) M.
Guitet, F. Marcelo, S. A. de Beaumais, Y. Zhang, J.
Jimenez-Barbero, S. Tilloy, E. Monflier, M. Menand, M.
Sollogoub, Eur. J. Org. Chem. 2013, 2013, 3691. i) C.
Barthes, C. Lepetit, Y. Canac, C. Duhayon, D. Zargarian,
R. Chauvin, Inorg. Chem. 2013, 52, 48. j) Barrett, B. J.;
Iluc, V. M., Organometallics 2017, 36, 730. k) B. J. Barrett,
Figure 9. Structures of transition states of insertion using
substrate 6.
4. Conclusion
In this study, we developed trans-chelating ligands, which
comprise two diphenylphosphino triazinyl groups linked by
dioxy-aromatic spacer. The 1a and 1b, which have relative
narrow spacers, are complexed with Rh to give simplex trans-
complexes. The ligand 1b shows high LAE for Pd-catalyzed
Heck reaction of iodobenzene (5) with tert-butyl acrylate (6).
The catalytic loading was decreased to 0.0001 mol%, and the
TON reached 970,000. The catalytic mechanisms were
considered by DFT calculations. There are two candidates for
catalytic processes, monodentate and bidentate coordination
pathways. In the reaction using 1b ligand, monodentate
coordination processes were favored except for the oxidative

V. M. Iluc, Inorg. Chim. Acta 2017, 460, 35.
2.
3.
Reviews: a) Z. Freixa, P. W. N. M. van Leeuwen, Coord.
Chem. Rev. 2008, 252, 1755. b) C. A. Bessel, P. Aggarwal,
A. C. Marschilok, K. J. Takeuchi, Chem. Rev. 2001, 101,
1031, and references therein.
a) Y. Canac, N. Debono, C. Lepetit, C. Duhayon, R.
Chauvin, Inorg. Chem. 2011, 50, 10810. b) P. Stepnicka,
M. Krupa, M. Lamac, I. Cisarova, J. Organomet. Chem.
2009, 694, 2987. c) L. Kaganovsky, K.-B. Cho, D. Gelman,
Organometallics 2008, 27, 5139. d) R. Kuwano, M.
Kashiwabara, Org. Lett. 2006, 8, 2653. e) L. Chahen, B.
Therrien, G. Süss-Fink, J. Organomet. Chem. 2006, 691,
4257. f) R. C. Smith, C. R. Bodner, M. J. Earl, N. C. Sears,
N. E. Hill, L. M. Bishop, N. Sizemore, D. T. Hehemann, J.
J. Bohn, J. D. Protasiewicz, J. Organomet. Chem. 2005,
690, 477. g) S. Burger, B. Therrien, G. Süss-Fink, Helv.
Chim. Acta 2005, 88, 478. h) M. Sawamura, M. Sudoh, Y.
Ito, J. Am. Chem. Soc. 1996, 118, 3309.
4.
5.
B. Heuer, S. J. A. Pope, G. Reid, Polyhedron 2000, 19, 743.
S. Burger, B. Therrien, G. Süss-Fink, Eur. J. Inorg. Chem.
2003, 3099.
6.
7.
8.
9.
R. Hirt, H. Nidecker, R. Berchtold, Helv. Chim. Acta 1950,
33, 1365.
M. Hayashi, T. Yamasaki, Y. Kobayashi, Y. Imai, Y.
Watanabe, Euro. J. Org. Chem. 2009, 29, 4956.
A. K. Bunten, H. D. Farrar, J. A. Poe, A. Lough,
Organometallics 2002, 21, 3344.
T. Kawano, T. Shinomaru, I. Ueda, Org. Lett. 2002, 4, 2545.
10. 2-(Diphenylphosphanyl)-4,6-diphenoxy-1,3,5-triazine
was synthesized from 2-chloro-4,6-diphenoxy-1,3,5-
triazine with Me3SiPPh2.
11. β for BINAP, DPPP, Xantphos: P. C. J. Kamer, P. W. N. M.
van Leeuwen, J. N. H. Reek, Acc. Chem. Res. 2001, 34,
895.
12. β for DPPPE: J.-P. Barras, B. Bourdin, F. Schroeder,
Chimia 2006, 60, 574.
13. Recent reports: a) P. Veerakumar, P. Thanasekaran, K.-L.
Lu, K.-C. Lin, S. Rajagopal, ACS Sustainable Chem. Eng.
2017, 5, 8475. b) V. H. Menezes da Silva, A. A. C. Braga,
T. R., Cundari, Organometallics 2016, 35, 3170. c) Y.
Dang, S. Qu, Z.-X. Wang, X. Wang, J. Am. Chem. Soc.
2014, 136, 986. d) C. Allolio, T. Strassner, J. Org. Chem.
2014, 79, 12096.
14. P. Surawatanawong, Y. Fan, M. B. Hall, J. Organomet.
Chem. 2008, 693, 1552.
15. A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
16. For Gaussian basis sets: a) M. J. Frisch, J. A. Pople, J. S.
Binkley, J. Chem. Phys. 1984, 80, 3265. b) W. J. Hehre, L.
Radom, P. v. R. Schleyer, J. A. Pople in Ab initio Molecular
Orbital Theory; John Wiley: New York, USA, 1986, and
references cited therein.
17. Gaussian 09, Revision E.01; Gaussian, Inc., Wallingford,
CT, 2009.
18. See supporting information.
19. V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102, 1995.
20. M. Sumimoto, T. Kuroda, D. Yokogawa, H. Yamamoto, K.
Hori, J. Organomet. Chem. 2012, 710, 26.
21. Unfortunately, the transition state of the conformational
change between cis-bi-20 and mono-20 could not be found.
However, the rotational barriers were calculated. See
Supporting Information.

Graphical Abstract
Moderately Flexible Trans-Chelating Diphosphine Ligands Consisting of Triazine Unit Showing High Ligand Acceleration Effect for
Heck Reaction
Toshinobu Korenaga,* Ryo Kowata, Tomohiro Kotaki, Yuji Shibasaki, Yoshiyuki Oishi, and Kazuaki Shimada
Originally designed trans-chelating diphosphine ligands with moderate flexibility were developed. The catalyst loading could be
reduced up to 0.0001 mol%, and the turnover number (TON) reached 970,000 for Pd-catalyzed Heck reaction. The high ligand
acceleration effect was considered by DFT calculations.