A new air-stable and reusable tetraphosphine ligand for rhodium-catalyzed hydroformylation of terminal olefins at low temperature

Article abstract of DOI:10.1002/aoc.4646

Tetraphosphine and bisphosphine ligands were synthesized, characterized and employed in Rh-catalyzed hydroformylation of 1-octene and 1-hexene. Conversion of over 97.7% and aldehyde yield of 94.1% were achieved at 60°C, 20?bar. This remarkable performance could also be retained at lower temperature (i.e. 40°C) by prolonging the reaction time. The tetraphosphine ligand-modified Rh catalyst could be reused for at least seven successive runs with catalytic activity and selectivity almost unchanged; the catalyst was separated from the products and recycled directly in homogeneous hydroformylation, indicating that the catalyst might have good stability. ³¹P NMR and high-resolution mass spectral characterization hinted that the reason for the reusability of the catalyst might be that the tetraphosphine ligand is relatively air-stable and is probably slowly oxidized during the recycling runs. The tetraphosphine ligand has four phosphorus atoms to be partially oxidized and could still coordinate with the Rh center via the unoxidized phosphorus atoms to stabilize the catalyst, based on the multiple chelating modes of the tetraphosphine ligand. Hence, the catalytic activity and selectivity could be retained for a certain number of runs.

Full text of DOI:10.1002/aoc.4646

Received: 31 July 2018
DOI: 10.1002/aoc.4646
Revised: 17 September 2018
Accepted: 23 September 2018
F U L L P A P E R
A new air‐stable and reusable tetraphosphine ligand for
rhodium‐catalyzed hydroformylation of terminal olefins at
low temperature
Fanding Zhou1 | Lin Zhang2 | Qianhui Wu1 | Fuya Guo1 | Songbai Tang1 | Bin Xu3 |
Maolin Yuan1 | Haiyan Fu1 | Ruixiang Li1
| Xueli Zheng1 | Hua Chen1
1 Key Laboratory of Green Chemistry and
Tetraphosphine and bisphosphine ligands were synthesized, characterized and
employed in Rh‐catalyzed hydroformylation of 1‐octene and 1‐hexene. Conver-
sion of over 97.7% and aldehyde yield of 94.1% were achieved at 60°C, 20 bar.
This remarkable performance could also be retained at lower temperature (i.e.
40°C) by prolonging the reaction time. The tetraphosphine ligand‐modified Rh
catalyst could be reused for at least seven successive runs with catalytic activity
and selectivity almost unchanged; the catalyst was separated from the products
and recycled directly in homogeneous hydroformylation, indicating that the
catalyst might have good stability. 31P NMR and high‐resolution mass spectral
characterization hinted that the reason for the reusability of the catalyst might
be that the tetraphosphine ligand is relatively air‐stable and is probably slowly
oxidized during the recycling runs. The tetraphosphine ligand has four phos-
phorus atoms to be partially oxidized and could still coordinate with the Rh
center via the unoxidized phosphorus atoms to stabilize the catalyst, based
on the multiple chelating modes of the tetraphosphine ligand. Hence, the cat-
alytic activity and selectivity could be retained for a certain number of runs.
Technology, Ministry of Education,
College of Chemistry, Sichuan University,
Chengdu 610064, China
2 Department of Pharmacy, Guizhou
Medical University, Guiyang 550004,
China
3 Key Laboratory of Green Chemistry,
Sichuan Institutes of Higher Education,
Zigong 643000, China
Correspondence
Xueli Zheng and Hua Chen, Key
Laboratory of Green Chemistry and
Technology, Ministry of Education,
College of Chemistry, Sichuan University,
Chengdu 610064, China.
Email: scuhchen@163.com;
zhengxueli@scu.edu.cn
Funding information
Key Program of Sichuan Science and
Technology Project, Grant/Award Num-
ber: No. 2018GZ0312; National Natural
Science Foundation of China, Grant/
Award Number: No. 201202108; Opening
Project of Key Laboratory of Green
Chemistry of Sichuan Institutes of Higher
Education, Grant/Award Number: No.
LZJ1402; Sichuan university outstanding
scholar research fund, Grant/Award
Number: No. 2015SCU04A05
KEYWORDS
activity, hydroformylation, low temperature, recycling of catalysts, tetraphosphine ligand
1 | INTRODUCTION
recycling of the metal catalyst are always technological
challenges of homogeneous catalysis. Hydroformylation
Rhodium‐catalyzed hydroformylation is an easy and effi-
cient industrial catalytic process that transforms olefins to
their corresponding aldehydes, which could increase up
to 10 million tons per year worldwide.[1–3] Since catalyst
and product are in the same phase after reaction, the effi-
cient separation of product from catalyst as well as the
of short‐chain alkenes (i.e., ethylene, propene) has been
successfully applied. With respect to mid‐chain alkenes,
owing to the higher boiling points of the products, the
thermal separation under reduced pressure would put
thermal stress on the temperature‐sensitive catalyst, and
thereby the catalyst would slowly decompose under these
Appl Organometal Chem. 2018;e4646.
wileyonlinelibrary.com/journal/aoc
© 2018 John Wiley & Sons, Ltd.
1 of 10
https://doi.org/10.1002/aoc.4646

2 of 10
ZHOU ET AL.
conditions. Consequently, seeking an effective catalyst
which is insensitive to air or moisture and temperature
could also be more stable in the presence of moisture
compared to the P─O bond.[17] It seems that these
phosphines could have the advantages of activity and
stability.
is
highly
desirable
for
the
homogeneous
hydroformylation of mid‐chain alkenes.
The catalysts employed in hydroformylation are
always phosphorus ligand‐modified Rh complexes, and
the phosphorus ligand always plays a vital role in the cat-
alytic performance. With the development of
hydroformylation, a variety of phosphorus ligands have
been synthesized and extensively studied. According to
the atoms connected to the P atom, the ligands can be
approximately classified as follows[3] (Scheme 1). (1)
Phosphanes (or phosphines)[4–7] contain all P─C bonds;
P can be bonded to phenyl, such as in triphenylphosphine
(TPP)[8] and 2,2′‐bis(diphenylphosphinomethyl)‐1,1′‐
biphenyl.[9] These kinds of phosphane ligands seem to
be relatively stable and TPP is usually used in technical
homogeneous hydroformylation. Rh catalysts modified
with these ligands always give moderate activity and
selectivity, and relatively high temperature (80–120°C) is
usually desired to accelerate reaction rates. (2) Phos-
phites, phosphonites and phosphinites[10–12] contain at
least one P─O bond with other P─C bonds. (3)
Phosphoramidites[13,14] contain P─O and P─N bonds.
Compared to phosphanes with all P─C bonds, these
kinds of phosphorus ligands are found to be stronger π‐
acceptors and weaker σ‐donors, which could promote
CO dissociation from the Rh center to form four‐
coordinated catalytic species, and facilitate olefin coordi-
nation with Rh even at relatively low temperature.[15]
However, these ligands can easily decompose under ther-
mal stress, and the P─O bond is sensitive to moisture,
which might cause the ligand to undergo hydrolysis.[16]
(4) Phosphines[17–19] containing at least one P─N bond
with the rest being P─C bonds not only exhibit high
activity and selectivity due to the electron‐withdrawing
property at relatively low temperature, but the P─N bond
As mentioned above, most phosphorus ligands are sen-
sitive to air or moisture; these properties would accelerate
the deactivation of catalysts, shorten the service life or even
hinder their successful application on technical scale. And
to achieve good performance, high temperature is required
for general phosphine‐modified systems.[20–22] However,
high temperature might accelerate the degradation of the
ligands and increase hydrogenation or isomerization. Thus,
this contradiction prompted us to design and synthesize an
ideal phosphorus ligand bearing P─N and P─C bonds,
which is not sensitive to air and moisture and could exhibit
good activity even at low temperature, or even could with-
stand the thermal separation of products.
The work reported here dealt with the question of the
oxidation of catalyst/ligand during hydroformylation and
thermal separation. Two phosphine ligands bearing
indole‐based backbones were easily synthesized through
two steps. Both of them were air‐stable ligands and could
retain high activity in Rh‐catalyzed hydroformylation of 1‐
hexene and 1‐octene at low temperature (40–60°C). More
importantly, the catalytic activity and selectivity to alde-
hyde were maintained during seven successive runs of
the catalyst system of tetraphosphine‐modified Rh(acac)
(CO)2.
2 | EXPERIMENTAL
2.1 | Materials
The solvent used for hydroformylation was pure toluene
which had been desulfurized by concentrated sulfuric
acid and dried beforehand. Toluene (AR) was acquired
from Chengdu Chang Lian Chemical Reagent Co. Ltd
(Chengdu, China). Syngas was premixed at a pressure
ratio of 1:1 with hydrogen (99.99%) and carbon monoxide
(99%). The catalyst precursor used in this work was
Rh(acac)(CO)2 (acac = acetylacetonate) which was pre-
pared according to the literature.[23] 1‐Hexene (99%) was
purchased from J&K (Beijing, China) and 1‐octene
(>99%) was acquired from Adamas (Shanghai, China).
2.2 | Synthesis of ligands L1 and L2
The recommended new phosphine ligands L1 and L2 can
be easily prepared through two steps. Firstly, the back-
bones A and B were synthesized with high yield accord-
ing to our previous work.[24] The backbone was
SCHEME 1 Classification of phosphorus ligands according to
the P‐X (X = C, O, N) bonds

ZHOU ET AL.
3 of 10
constructed based on indole rather than pyrrole, both of
them being electron‐withdrawing groups but pyrrole
being more active than indole which made the pyrrolyl‐
based backbone more complicated. Then, the phosphine
ligands L1 and L2 were prepared through the nucleo-
philic substitution reaction of chlorodiphenylphosphine
with the corresponding backbone in the presence of
triethylamine (NEt3) (Scheme 2).
GC‐A91 gas chromatograph fitted with capillary columns
(30 m × 0.25 mm × 0.50 μm, KB‐1).
2.4 | General procedure for recycling of
catalyst Rh/L2 in hydroformylation of
1‐hexene
A 60 ml steel autoclave with an inner glass tube served
for the recycling experiments of catalyst Rh/L2. The
hydroformalytion process was performed as described in
Section 2.3. After the hydroformylation reaction was
completed, the glass tube was pulled out, and subse-
quently the solvent, the products and trace amount of
unconverted olefin were completely removed from the
catalyst via heating at 90–100°C under vacuum. Then
another identical charge of 1‐hexene and toluene was
added to the glass tube containing the catalyst residue,
and then the tube was loaded into the autoclave again.
The catalytic hydroformylation procedure was then
repeated.
2.3 | General procedure for
hydroformylation of l‐octene with L1 and L2
A 60 ml steel autoclave served as the vessel for the
hydroformylation reaction, equipped with a magnetic
stirrer, a thermocouple and a digital pressure gauge
which could record the pressure change inside the auto-
clave. The substrate 1‐octene (1 ml) was mixed with a tol-
uene solution, which contained a certain amount of
Rh(acac)(CO)2 and ligand. Then, the autoclave was
sealed and purged with syngas (CO/H2 at a molar ratio
of 1:1) three times. When the reactor was heated to the
preset temperature, the autoclave was pressurized to a
preset value and stirring was started, and, simulta-
neously, the reaction time was recorded. After the reac-
tion was finished, the products were analyzed using a
3 | RESULTS AND DISCUSSION
In this study, two new phosphine ligands L1 and L2 were
firstly synthesized and characterized, and then each
applied in the hydroformylation of 1‐octene. The catalyst
SCHEME 2 Synthesis of ligands L1
and L2

4 of 10
ZHOU ET AL.
for hydroformylation reaction was prepared in situ by
mixing each ligand with Rh(acac)(CO)2 in toluene.
Firstly, the P/Rh ratio was investigated as the basis of
the following experiments (Table 1). The conversion and
aldehyde yield firstly increased and next decreased with
increasing P/Rh ratio, which revealed the optimum
P/Rh ratio for the two ligands. Then the effects of temper-
ature and pressure on the activity and the regioselectivity
were evaluated. Temperature did not markedly impact
the selectivity to aldehyde when L1 was employed as
the ligand, with the exception of the relatively high value
of 120°C (Table 2, entries 1–4, 7, 10–12). And high alde-
hyde selectivity was preserved within a broad range of
temperature (40–100°C). Lowering the temperature from
100 to 60°C caused the aldehyde selectivity to slightly
increase (90.8 to 97.7%), but the selectivity to hydrogena-
tion (5.1 to 0.2%) and isomerization (4.1 to 2.1%)
decreased. On further decreasing the temperature to 50
or 40°C, good conversion could also be realized just by
prolonging the reaction time to 6 or 12 h, which
TABLE 1 Optimization of molar ratio of P to Rha
Entry
L
P/Rh
n/ib
Conv. (%)c
Aldehyde (%)d
Octane (%)e
Isomer (%)f
TONg
1
—
L1
L1
L1
L1
L2
L2
L2
L2
0
5
0.9
2.4
2.4
2.5
2.5
1.9
2.8
2.5
2.6
98.1
98.1
97.8
98.4
98.0
98.1
97.4
97.6
97.9
26.4
81.2
90.8
90.9
84.6
54.1
71.3
80.7
83.7
21.4
10.3
5.1
52.3
8.5
259
797
888
894
829
530
694
788
819
2
3
10
15
20
5
4.1
4
4.4
4.7
5
8.7
6.8
7
11.9
14.4
9.4
33.9
14.3
9.8
8
10
15
20
9
10
8.1
8.3
aS/C: 1000, [Rh]: 1.5 mM, CO/H2: 20 bar, toluene as solvent, 100°C, 2 h.
bLinear/branched aldehyde ratio, determined by GC.
cConversion of 1‐octene.
dSelectivity to aldehyde in all products.
eSelectivity to octane in all products.
fSelectivity to 2‐octene in all products.
gTurnover number, determined by GC.
TABLE 2 Effect of temperature and pressure for hydroformylation of 1‐octene with Rh/L1 at P/Rh molar ratio of 10a
Entry
T (°C)
P (bar)
t (h)
n/ib
Conv. (%)c
Aldehyde (%)d
Octane (%)e
Isomer (%)f
TONg
1
120
100
80
60
60
60
50
50
50
40
40
40
20
20
20
20
30
10
20
10
5
2
2
2.6
2.4
2.2
2.2
2.1
2.5
2.4
2.5
2.9
2.5
2.4
2.3
95.1
97.8
98.5
98.2
98.7
97.4
53.3
51.9
22.1
13.5
51.6
90.4
65.4
90.8
93.3
97.7
98.7
94.1
97.9
95.2
91.0
93.3
95.2
97.5
21.3
5.1
3.5
0.2
0.2
3.4
0
13.2
4.1
3.2
2.1
1.1
2.2
2.1
3.3
3.2
3.7
4.5
1.3
622
888
919
959
974
917
522
494
201
126
491
881
2
3
2
4
2
5
2
6
2
7
2
8
2
1.7
5.9
3.7
2.2
1.2
9
2
10
11
12
20
20
20
2
6
12
aS/C: 1000, [Rh]: 1.5 mM, toluene as solvent.
b–gSee Table 1.

ZHOU ET AL.
5 of 10
demonstrated good activity and good selectivity of our
catalytic system at low temperature. The syngas (CO/H2
1:1) pressure also influenced the reaction. Lower pressure
seemed to impact the formation of aldehyde slightly,
while benefitting the regioselectivity a little (Table 2,
entries 4–9). This may be explained by the competition
of CO and ligands, as with the concentration of CO
decreasing, the linear aldehyde active species II would
be generated easily to yield linear aldehyde[13]
(Scheme S1, supporting information). A similar phenom-
enon was observed when L2 was employed as the ligand,
where lower temperature also suppressed the hydrogena-
tion and isomerization, but facilitated the formation of
aldehyde (Table 3). And under milder reaction condi-
tions, Rh/L2 always exhibited better activity than Rh/L1
(Table 2, entries 9, 12 versus Table 3, entries 10, 11).
The n/i ratio also increased as the pressure was lowered.
But Rh/L2 afforded higher regioselectivity and aldehyde
selectivity, compared with Rh/L1.
To our delight, in the hydroformation of 1‐octene,
isomerization was largely suppressed by the presence of
L1 or L2 (Table 1, entries 1, 3, 9), and the isomerization
products were less compared to the results achieved by
our previously reported ligands L3 and L4 (the corre-
sponding values are 10.2 and 47.2%, respectively)
(Figure S1 and Table S1, supporting information),[24]
resulting high aldehyde yield. This phenomenon might
arise from their structural features. L1 and L2 have simi-
lar structures to L3 and L4, with the same backbones and
two phenyl groups replacing the two pyrrolyl groups. The
former bear more electron‐rich phosphorus atoms
because of the more electron‐withdrawing property of
pyrrolyl groups than phenyl, and thus the isomerization
of double bonds could be suppressed due to this character
during the hydroformylation process, just as elucidated
by Zhang.[14,25] And the electron‐withdrawing ability of
L1 and L2 could benefit the dissociation of CO from the
five‐coordinating catalytic species and facilitate olefin
coordination with Rh, resulting in the good activity of
the catalysts. Indeed, few phosphine ligands to the Rh‐
catalyzed hydroformylation of mid‐chain olefins, such as
1‐octene, presented both high conversion and high alde-
hyde yield when conducted at lower temperature (80°C
or below 80°C). In contrast, in this study, the aldehyde
yield could reach 90.4% (98.3%) at 60°C and to 97.5%
(98.0%) after heating at 40°C for 12 h in the presence of
ligand L1 (L2), with high conversation (>90.4%).
3.1 | Catalyst activity and kinetic
investigations
The catalytic activity was further investigated through
increasing the substrate‐to‐catalyst (S/C) ratio. The
hydroformylation results with various S/C ratios are pre-
sented in Table 4. When L1 was employed as the ligand, a
clear decrease in the aldehyde yield and the conversion
was observed when S/C ratio increased from 1000 to
4000 (entries 1–4). However, when L2 was employed as
the ligand, the reaction activity and aldehyde selectivity
remained high when the S/C ratio increased from 1000
to 4000 (entries 5–8). This might be explained in that
the local phosphorus concentration around the Rh center
is higher in L2 than in L1, due to the enhanced chelating
ability of tetraphosphine ligand L2 through the unique
multiple chelating modes[14] (see Scheme S2, supporting
information). Hence, the performances of L1 and L2 at
lower S/C (i.e. 1000) seem to be close, but L2 exhibited
TABLE 3 Effect of temperature and pressure for hydroformylation of 1‐octene with Rh/L2 at P/Rh molar ratio of 15a
Entry
T (°C)
P (bar)
t (h)
n/ib
Conv. (%)c
Aldehyde (%)d
Octane (%)e
Isomer (%)f
TONg
1
120
100
80
60
60
60
50
50
50
50
40
20
20
20
20
10
5
2
2
2.8
2.5
2.5
2.5
3.1
3.8
2.5
2.7
3.1
3.4
2.4
97.7
97.6
98.0
97.7
95.9
80.4
53.1
51.5
57.9
43.4
98.3
80.0
80.7
85.7
94.1
88.5
83.6
97.2
97.9
94.0
92.6
98.0
8.7
9.4
7.8
3.8
6.5
8.5
1.1
1.2
1.8
3.2
0.9
11.2
9.8
6.5
2.1
5.0
7.8
1.7
1.0
4.3
4.1
1.1
782
788
840
919
849
672
516
504
544
402
963
2
3
2
4
2
5
2
6
2
7
30
20
10
5
2
8
2
9
2
10
11
2
20
12
aS/C: 1000, [Rh]: 1.5 mM, toluene as solvent.
b–gSee Table 1.

6 of 10
ZHOU ET AL.
TABLE 4 Effect of substrate/catalyst ratio for hydroformylation of 1‐octene using ligands L1 and L2a
Entry
L
P/Rh
S/C
n/ic
Conv. (%)d
Aldehyde (%)e
Octane (%)f
Isomer (%)g
TONh
1
2
L1
L1
L1
L1
L2
L2
L2
L2
L2
L2
10
10
10
10
15
15
15
15
15
15
1000b
2000b
3000
4000
1000b
2000b
3000
4000
5000
6000
2.2
2.3
2.6
2.8
2.5
2.3
2.5
2.5
2.7
2.7
98.2
75.7
86.8
38.6
97.7
98.7
97.5
82.7
33.7
21.0
97.7
94.8
68.7
40.1
94.1
95.4
95.2
93.6
37.2
42.1
0.2
3.0
2.1
2.1
959
1435
1789
619
3
18.6
43.7
3.8
12.7
16.1
2.1
4
5
919
6
2.7
1.8
1883
2785
3096
627
7
3.1
1.7
8
4.3
2.1
9
45.0
42.0
17.9
15.7
10
530
aCO/H2: 20 bar, toluene as solvent, 60°C, 4 h, b2 h.
cLinear/branched aldehyde ratio, determined by GC.
dConversion of 1‐octene.
eSelectivity for aldehyde in all products.
fSelectivity for octane in all products.
gSelectivity for 2‐octene in all products.
hTurnover number determined by GC.
superior performance with increasing S/C and could
work well even at S/C ratio of 4000. Also, the increase
of S/C might somehow lower the concentration of ligand.
Van Leeuwen and co‐workers suggested that the equilib-
rium of reactive intermediates much depended on the
concentration of ligands and CO[13] (see Scheme S1,
supporting information), which revealed that a less active
intermediate was readily formed when the concentration
of ligands decreased, and thus suppressed the activity of
hydroformylation. We also found that by means of adding
a little more ligand, Rh/L1 might be capable of achieving
comparable aldehyde selectivity to the case at lower S/C
molar ratios (see Table S2, entries 3, 5, supporting
information).
A kinetic investigation of the hydroformalation of 1‐
octene was performed at 50, 60 and 80°C in the presence
of each of Rh(acac)(CO)2/L1 and Rh(acac)(CO)2/L2,
which was estimated by the pressure drop versus time.
As expected, the pressure drop was faster at a higher tem-
perature, since faster pressure drop ideally demonstrated
faster aldehyde formation. Figure 1 shows the pressure
drop as a function of time, which demonstrates that the
hydroformylation of 1‐octene displays an exponential
growth behavior. Determined from the half‐life calcula-
tion, the reaction order is close to 1. From the slopes
(Ea/R) and intercepts (ln A) of plots of ln k versus 1/T
(Figure 2), the activation energy and the pre‐exponential
factor were determined. The apparent activation energies
for hydroformylation of 1‐octene catalyzed by Rh/L1 and
Rh/L2 are 104.5 and 91.3 kJ mol−1, respectively. Compar-
atively, hydroformylation catalyzed by Rh/L1 has to sur-
pass a slightly higher energy barrier to yield aldehyde
than that by Rh/L2, which is in good agreement with
the superior activity of Rh/L2 over Rh/L1 discussed above
(Table 4, entry 4 versus 8). We attempted to derive the
corresponding barrier for the hydroformylation without
L1 or L2 but failed due to the lower aldehyde yield
(Table 1, entry 1), yet it is reasonable that the barrier
might be higher than those for Rh/L1 and Rh/L2. And
a high turnover frequency value for L2 at 80°C (Table 2,
entry 3) could be calculated as 4940 h−1 (see Scheme S3,
supporting information), for the reaction was nearly fin-
ished in about 10 min according to Figure 1b.
3.2 | Catalyst stability
Since Rh/L2 exhibited higher activity at higher S/C and
under milder conditions, the recycling experiments were
performed with the Rh/L2 system to investigate the sta-
bility of the catalyst. In consideration of easier separation
of aldehyde products in homogeneous system, 1‐hexene
was chosen as the olefin substrate in the recycling study
due to the lower boiling point of its aldehyde products
compared with those for the corresponding products of
1‐octene. After every hydroformylation run had finished,

ZHOU ET AL.
7 of 10
the solvent, the products and the trace amount of uncon-
verted olefin were completely removed from the catalyst
via evaporation under vacuum. In case an incomplete
separation might cause the accumulation of
hydroformylation products and thus contribute to the
recycling data, after new portions of olefin and toluene
were added into the glass tube with catalyst residue, the
mixture was analyzed using GC, and the data (1‐hexene:
aldehyde > 99.9%:0.1%) confirmed that the product alde-
hydes were almost completely removed and could hardly
contribute to the recycling data. The recycling perfor-
mance is shown in Figure 3, and the maintenance of
the activity and the selectivity to aldehyde for seven
reuses of the Rh/L2 system demonstrated its good stabil-
ity, especially the catalyst being separated from products
and recycled in homogeneous hydroformylation. In addi-
tion, fivefold substrate, the corresponding amount of cat-
alyst and the solvent were loaded for recycling
experiments. To our delight, the recycling results were
nearly the same as before (Table S4, supporting informa-
tion), which also evidenced that few accumulations of
hydroformylation products contributed to the recycling
performance.
In the seven times of reusing the L2/Rh system, the
conversion of 1‐hexene was almost unchanged (>99%).
The aldehyde yield only slightly decreased from 97.5 to
94.7% (Table S3, entries 1–7, supporting information).
Alsalahi and Trzeciak investigated the stability of the
Rh(acac)(CO2)/PPh3 catalytic system, which indicated
that the catalyst system possessed good stability during
five successive runs[26] (conversion from 97 to 81%). How-
ever, after every run, a new portion of PPh3 was added for
FIGURE
1
Time dependence of pressure drop in the
hydroformylation of 1‐octene catalyzed by (a) Rh(acac)(CO)2/L1 and
(b) Rh(acac)(CO)2/L2 at various temperatures. S/C = 1000, (H2/CO;
1:1) = 20 bar
FIGURE 3 Plot of the conversion of 1‐hexene, yield of aldehydes
(linear and branched) and turnover number (TON) versus the
number of cycles, for the hydroformylation of 1‐hexene employing
the Rh/L2 system. Conditions: S/C = 1000; [Rh]: 1.5 mM,
P/Rh = 15, toluene, 20 bar, 60°C, 2 h
FIGURE 2 Arrhenius plots of rate constants derived from Figure
1. The linear fit yields activation energy and pre‐exponential factor

8 of 10
ZHOU ET AL.
another run, in order to stabilize the Rh catalyst and
retain its activity. The L2/Rh system can retain a high
activity without excess phosphine ligand being added
during the seven successive runs. And the turnover num-
ber could accumulate up to 7225 (Table S3, supporting
information). This result hinted that the L2/Rh system
might also have good stability and good activity.
multiple peaks around 33.5 ppm firstly increased and
later decreased (Figure 4d–f). The rhodium–phosphorus
coupling constant (JRhP) was relatively large, generally
about 100–200 Hz[27] and even up to 580 Hz,[19] and the
coupling constant around 33.5 and 56.0 ppm was quite
small (10–25 Hz), which suggested that those resonances
were not attributed to Rh–phosphane complex. Mean-
while, the resonance around 56.0 ppm was not observed
after hydroformylation until distillation, and once gener-
ated it was accumulated during hydroformylation runs
(see Figure S2a,b, supporting information). As the oxida-
tion of L2 led to a resonance at 25.2 ppm, it is suspected
that the 56.0 ppm signal might be owing to the slow deg-
radation of L2 after distillation. As mentioned above,
Rh/L2 could maintain its activity and selectivity in seven
sequential recycling tests, which hinted that L2 was not
oxidized immediately. These phenomena encouraged us
to infer that the multiple peaks around 33.5 ppm might
be the chemical shift of partially unoxidized phosphorus
on L2. As demonstrated in Scheme 2, partially oxidized
P1 made P2/P3/P4 electron deficient and resulted in a
slightly higher chemical shift than L2 (33.5 versus
31.7 ppm). To confirm this, high‐resolution MS character-
ization was conducted on the sample collected after the
first hydroformylation run. To our delight, L2 and its part
oxidation products, and the degradation products of L2
were detected (see Figure S3, supporting information),
which may indicate a slow oxidation of L2 during the
To further clarify the reason for the good stability and
the good activity of the L2/Rh system, the residues col-
lected after the first, the third and the last runs of
hydroformylation were characterized using 31P NMR
spectroscopy. For reference, 31P NMR spectra of ligand
L2 before and after oxidization by O2 and H2O2 were
included (Figure 4a–c). It is not difficult to see that L2
is air‐stable and be oxidized only under more harsh con-
ditions, i.e. H2O2, and the corresponding resonance was
a sharp peak at 25.2 ppm. This might be due to two
aspects: one is that the indolyl group has greater stability
than pyrrolyl; the other is that the phenyl and the indolyl
groups might cause P atoms to be less electron‐rich, and
thus the ligand is air‐stable and oxidized only by H2O2.
The sample collected after the first hydroformylation run
(Figure 4d) exhibited resonances of low intensity at 25.2,
33.5 and 56.0 ppm, which were quite different from those
of the NMR spectra of L2 (Figure 4a). In addition, the
intensity of the resonances at 25.2 and 56.0 ppm was
enhanced along with the hydroformylation runs, while
the chemical shift for L2 (31.7 ppm) decreased, and the
FIGURE 4 31P NMR spectra of (a) ligand L2; L2 after exposure to (b) O2 for 2 h at 60°C and (c) H2O2 for 2 h at 25 °C; and the residue
collected after (d) first, (e) third and (f) last run of hydroformylation

ZHOU ET AL.
9 of 10
(no. 2015SCU04A05) and by the Key Program of Sichuan
Science and Technology Project (no. 2018GZ0312).
ORCID
Ruixiang Li http://orcid.org/0000-0003-2358-1226
Hua Chen http://orcid.org/0000-0003-1159-3780
REFERENCES
[1] A. Börner, R. Franke, Hydroformylation: Fundamentals, Pro-
cesses and Applications in Organic Synthesis, Wiley‐VCH,
Weinheim 2016.
[2] A. M. Trzeciak, J. J. Zijłkowski, Coord. Chem. Rev. 1999, 883, 190.
[3] R. Franke, D. Selent, A. Börner, Chem. Rev. 2012, 112, 5675.
SCHEME 3 Progressive oxidization and multiple chelating
modes of tetraphosphine ligand L2
[4] P. W. N. M. Van Leeuwen, P. C. J. Kamer, P. Dierkes, Chem.
Rev. 2000, 100, 2741.
[5] C. P. Casey, G. T. Whiteker, L. M. Petrovich, J. Am. Chem. Soc.
1992, 114, 5535.
whole hydroformylation run, one P atom at a time
(Scheme 3), and the unoxidized phosphorus atoms could
still chelate with Rh due to the multiple chelating modes
of the tetraphosphine ligand. Thus the L2‐modified Rh
system could possess good stability and good activity in
seven runs.
[6] F. Shibahara, K. Nozaki, T. Hiyama, J. Am. Chem. Soc. 2003,
125, 8555.
[7] M. Kranenburg, Y. E. M. Van der Burgt, P. W. N. M. Van
Leeuwen, Organometallics 1995, 14, 3081.
[8] (a) D. Evans, G. Yagupsky, J. Chem. Soc. A 1968, 2660. (b) C. K.
Brown, G. Wilkinson, J. Chem. Soc. A 1970, 2753.
[9] T. J. Devon, G. W. Phillips, T. A. Puckette, J. L. Stavinoha, J. J.
Vanderbilt, U.S. Patent 1987, 4,694,109; Chem. Abstr. 1988,
108, 7890.
4 | CONCLUSIONS
New phosphine ligands L1 and L2 have been easily syn-
thesized through two steps and successfully applied in
Rh‐catalyzed hydroformylation of 1‐octene and 1‐hexene.
Both of them exhibited high activity and aldehyde selec-
tivity at low temperature (40–60°C). The Rh/L2 system
exhibited superior performance over Rh/L1 in terms of
activity and aldehyde selectivity, not only under milder
conditions but also for high S/C ratio. A kinetic study also
revealed that the Rh/L2 system has lower activation
energy than Rh/L1. Significantly, L2 displayed excellent
performance in the catalyst stability tests. This is probably
because L2 could guarantee high phosphorus concentra-
tion around Rh on account of its multiple chelating
modes. Hence, Rh/L2 could be reused in seven runs, with
the phosphorus atoms being oxidized one by one, indicat-
ing its potential application in industry.
[10] J. I. Van der Vlugt, R. Sablong, D. Vogt, Organometallics 2004,
23, 3177.
[11] C. Vogl, E. Paetzold, U. Kragl, J. Mol. Catal. A. 2005, 232, 41.
[12] D. Selent, B. Kreidler, A. Börner, Organometallics 2011, 30, 4509.
[13] S. C. Van der Slot, J. Duran, J. Luten, P. C. J. Kamer, P. W. N.
M. Van Leeuwen, Organometallics 2002, 21, 3873.
[14] Y. J. Yan, X. W. Zhang, X. M. Zhang, J. Am. Chem. Soc. 2006,
128, 16058.
[15] Z. P. Zhang, C. Y. Chen, Q. Wang, Z. Y. Han, X. Q. Dong, X. M.
Zhang, RSC Adv. 2016, 6, 14559.
[16] S. Fischer, J. Hoyano, I. Johnson, L. K. Peterson, Can. J. Chem.
1976, 54, 2706.
[17] (a) E. Mieczynska, R. Grzybek, A. M. Trzeciak, ChemCatChem
2018, 10, 305. (b) A. M. Trzeciak, T. Głowiak, R. Grzybek, J. J.
Ziółkowski, J. Chem. Soc. Dalton Trans. 1997, 1831. (c) K. G.
Moloy, J. L. Petersen, J. Am. Chem. Soc. 1995, 117, 7696.
[18] B. Hamers, E. Kosciusko‐Morizet, C. Muller, D. Vogt,
ChemCatChem 2009, 1, 103.
ACKNOWLEDGMENTS
[19] O. Diebolt, H. Tricas, Z. Freixa, P. W. N. M. van Leeuwen, ACS
Catal. 2013, 3, 128.
This work was supported by the National Natural Science
Foundation of China (no. 201202108), by the Opening
Project of Key Laboratory of Green Chemistry of Sichuan
Institutes of Higher Education (no. LZJ1402), by the
Sichuan University Outstanding Scholar Research Fund
[20] X. F. Jia, Z. Wang, C. G. Xia, K. L. Ding, Chem. Eur. J. 2012, 18,
15288.
[21] P. J. B. Raymond, P. C. J. Kamer, P. W. N. M. van Leeuwen,
Organometallics 2003, 22, 5358.

10 of 10
ZHOU ET AL.
[22] S. C. Yu, X. W. Zhang, Y. J. Yan, C. X. Cai, L. Y. Dai, X. M.
SUPPORTING INFORMATION
Zhang, Chem. Eur. J. 2010, 16, 4938.
Additional supporting information may be found online
in the Supporting Information section at the end of the
article.
[23] F. Bonati, G. Wilkinson, J. Chem. Soc. 1964, 3156.
[24] L. Zhang, C. Li, Catal. Lett. 2014, 144, 1074.
[25] (a) Y. J. Yan, X. W. Zhang, X. M. Zhang, Adv. Synth. Catal.
2007, 349, 1582. (b) S. C. Yu, X. W. Zhang, Y. J. Yan, C. X.
Cai, L. Y. Dai, X. M. Zhang, Chem. Eur. J. 2010, 16, 4938.
How to cite this article: Zhou F, Zhang L, Wu
Q, et al. A new air‐stable and reusable
tetraphosphine ligand for rhodium‐catalyzed
hydroformylation of terminal olefins at low
temperature. Appl Organometal Chem. 2018;e4646.
https://doi.org/10.1002/aoc.4646
[26] W. Alsalahi, A. M. Trzeciak, J. Mol. Catal. A 2015, 408, 147.
[27] (a) L. A. van der Veen, P. C. J. Kamer, P. W. N. M. van
Leeuwen, Organometallics 1999, 18, 4765. (b) Y. Z. Jiao, M. S.
Torne, J. Gracia, J. W. (Hans) Niemantsverdrietbc, P. W. N.
M. van Leeuwen, Catal. Sci. Technol. 2017, 7, 1404.