Bromide promoted hydrogenation of CO2 to higher alcohols using Ru-Co homogeneous catalyst

Article abstract of DOI:10.1039/c6sc01314g

Iodides are commonly used promoters in C₂₊OH synthesis from CO₂/CO hydrogenation. Here we report the highly efficient synthesis of C₂₊OH from CO₂ hydrogenation over a Ru₃(CO)₁₂-Co₄(CO)₁₂ bimetallic catalyst with bis(triphenylphosphoranylidene)ammonium chloride (PPNCl) as the cocatalyst and LiBr as the promoter. Methanol, ethanol, propanol and isobutanol were formed at milder conditions. The catalytic system had a much better overall performance than those of reported iodide promoted systems because PPNCl and LiBr cooperated very well in accelerating the reaction. LiBr enhanced the activity and PPNCl improved the selectivity, and thus both the activity and selectivity were very high when both of them were used simultaneously. In addition, the catalyst could be reused for at least five cycles without an obvious change of catalytic performance.

Full text of DOI:10.1039/c6sc01314g

Chemical
Science
View Article Online
View Journal
EDGE ARTICLE
Bromide promoted hydrogenation of CO2 to higher
alcohols using Ru–Co homogeneous catalyst†
Cite this: DOI: 10.1039/c6sc01314g
Meng Cui,ab Qingli Qian, Zhenhong He,a Zhaofu Zhang,a Jun Ma,a Tianbin Wu,a
a
*
a
a
*
Guanying Yang and Buxing Han
Iodides are commonly used promoters in C2+OH synthesis from CO2/CO hydrogenation. Here we report
the highly efficient synthesis of C2+OH from CO2 hydrogenation over a Ru3(CO)12–Co4(CO)12 bimetallic
catalyst with bis(triphenylphosphoranylidene)ammonium chloride (PPNCl) as the cocatalyst and LiBr as
the promoter. Methanol, ethanol, propanol and isobutanol were formed at milder conditions. The
catalytic system had a much better overall performance than those of reported iodide promoted systems
because PPNCl and LiBr cooperated very well in accelerating the reaction. LiBr enhanced the activity and
PPNCl improved the selectivity, and thus both the activity and selectivity were very high when both of
them were used simultaneously. In addition, the catalyst could be reused for at least five cycles without
an obvious change of catalytic performance.
Received 23rd March 2016
Accepted 17th April 2016
DOI: 10.1039/c6sc01314g
www.rsc.org/chemicalscience
K/Cu–Zn–Fe catalysts were used at 300 ꢀC.8 It was found that
water could promote C2+OH generation when a Pt/Co3O4 cata-
lyst was used in a mixed solvent of water and DMI.9 In general,
Introduction
Carbon dioxide (CO2) is a greenhouse gas. On the other hand, it
is an abundant, nontoxic, easily available, and renewable C1
resource.1 Transformation of CO2 into value-added chemicals is
of great importance for the sustainable development of our
society. Currently, utilization of CO2 as a feedstock to synthesize
various chemicals,2 such as cyclic carbonates, carboxylic acids,
methanol, formic acid, methyl formate and dimethylforma-
mide, is investigated extensively.
the activity and selectivity of C2+OH over heterogeneous cata-
lysts were low and harsh reaction conditions were required.
Homogeneous catalysis is known for its higher catalytic
efficiency compared to heterogeneous catalysis. But it has rarely
been reported in CO2 hydrogenation to C2+OH, because the
metal complexes are usually unstable in the reaction condi-
tions. In the limited cases of homogeneous catalysis, iodides
were used as a promoter and played a key role in the formation
of C2+OH.10 But the catalytic systems suffer from low selectivity
and/or low activity of C2+OH formation. For example, in the Ru–
Co–KI system, only methanol and ethanol were generated and
the ethanol selectivity was low (26.4%).10a When noble Rh was
used to replace Co, the selectivity of C2+OH was improved but
the activity was still low.10b Moreover, it is well known that
iodides are the most commonly used promoters in C2+OH
synthesis from CO2/CO hydrogenation because of their stronger
nucleophilicity, which is favourable for generating larger alco-
hols.10–12 Bromides are much more stable and cheaper than
iodides, but poor performances for generating higher alcohols
limit their application in the reaction. Obviously, exploring
more efficient and cheaper catalytic systems for the reaction is
an interesting topic.
Alcohols are important bulk chemicals. The synthesis of
alcohols from CO2 hydrogenation has received much attention,
but research progress was mainly focused on the synthesis of
methanol.3 Higher alcohols (C2+OH) are more desirable in many
cases, especially as fuel and fuel additives. However, the
synthesis of C2+OH via CO2 hydrogenation is obviously a chal-
lenge. The acquired results for this topic are mostly focused on
heterogeneous catalysis. For example, a CoMoS based catalyst
produced 35.6% of C2+OH in alcohol products at 340 ꢀC.4 An
alkali-promoted Mo/SiO2 catalyst could generate alcohols at
250 ꢀC with a C2+OH selectivity of 75.6%.5 A [Rh10Se]/TiO2
catalyst could catalyze the reaction at 350 ꢀC with ethanol
selectivity of 83%.6 A combined Rh–Fe–Cu based catalyst could
produce ethanol with ethanol selectivity of about 70%.7
2+OH selectivity of 87.1% could be reached when modied
A
C
Herein we report the highly efficient synthesis of C2+OH from
CO2 hydrogenation promoted by bromide using a Ru–Co
aBeijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Colloid,
Interface and Chemical Thermodynamics, Institute of Chemistry, Chinese Academy
of Sciences, Beijing 100190, China. E-mail: qianql@iccas.ac.cn; hanbx@iccas.ac.cn;
Tel: +86-10-62562821
bUniversity of Chinese Academy of Sciences, Beijing 100049, China
† Electronic supplementary information (ESI) available: GC-MS and GC data. See
DOI: 10.1039/c6sc01314g
Scheme 1 Synthesis of C2+OH from CO2 hydrogenation.
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
bimetallic catalyst with PPNCl as the cocatalyst (Scheme 1). negligible (Fig. S1†). Only two homogeneous catalytic systems
Methanol, ethanol, propanol and isobutanol were generated at have been reported for this reaction.10a,b The products in this
milder conditions. The catalytic system had both a high activity work were different from those of CO2 hydrogenation by the Ru–
and selectivity of C2+OH compared to those of iodide promoted Co–KI system, in which ethanol was the only C2+OH product.10a
reactions. In addition, the catalytic system could be recycled In previous work,10b we found that Ru–Rh–LiI was a very effec-
and reused at least ve times without an obvious change of tive catalyst for producing C2+OH (12.9 C mmol Lꢁ1 hꢁ1).
catalytic performance. As far as we know, this is the rst work to Interestingly, the Ru–Co–PPNCl–LiBr catalyst designed in this
use PPNCl as a cocatalyst in C2+OH synthesis via CO2 hydroge- work had much higher activity (33.7 C mmol Lꢁ1 hꢁ1) with high
nation and we found that LiBr is a better promoter than LiI selectivity, although cheap Co was used to replace Rh, indi-
because PPNCl and LiBr cooperate effectively in enhancing the cating that LiBr and PPNCl played an important role for the very
activity and selectivity.
high activity and selectivity of the reaction, which will be dis-
cussed further in the following sections.
The LiBr promoter played a crucial role in accelerating the
reaction. Without LiBr, both the activity and selectivity of the
Results and discussion
C
2+OH synthesis were very low (entry 2). The LiBr also enhanced
Different catalytic systems were tested in the CO2 hydrogena-
tion, and the results are shown in Table 1. In this work, space
time yield (STY, in C mmol Lꢁ1 hꢁ1) is used to show the activity
of the catalytic systems, which is one of the commonly used
units, especially when multi-metals are utilized. Using LiBr as
the promoter, the reaction could proceed efficiently over the
Ru3(CO)12/Co4(CO)12 bimetallic catalyst with PPNCl as the
cocatalyst in 1,3-dimethyl-2-imidazolidinone (DMI) solvent
(entry 1). The alcohols in the reaction solution were methanol,
ethanol, propanol and isobutanol, and other products were
the stability of the catalyst. When LiBr was used without PPNCl,
the catalyst was also very active, but the selectivity to C2+OH was
much lower (entry 3). This indicates that LiBr and PPNCl
cooperated very well for producing C2+OH. LiBr enhanced the
activity and PPNCl improved the selectivity. Thus, both the
activity and selectivity were very high when both of them were
present. A black metal precipitate was observed if both LiBr and
PPNCl were absent (entry 4). We also tested the promoters with
other cations (Na+ and K+) and anions (Clꢁ, Iꢁ and BF4ꢁ), but
Table 1 Hydrogenation of CO2 to C2+OH using different catalytic systemsa
STYb [C mmol Lꢁ1 hꢁ1
]
C2+OH
Methanol Ethanol Propanol Isobutanol Total Sel. [%]
Entry Catalyst precursors
Promoter Cocatalyst Solvent
1
2
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12
LiBr
—
LiBr
—
LiCl
LiI
LiBF4
NaBr
KBr
KI
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
LiBr
PPNCl
PPNCl
—
DMI
DMI
DMI
DMI
DMI
DMI
DMI
DMI
DMI
DMI
DMI
DMI
DMI
DMI
DMI
DMI
3.1
10.2
9.5
0.5
35.5
0.4
29.5
0.7
19.2
0
13.1
1.9
0.3
2.5
2.6
4.5
20.8
0
5.9
23.4
13.3
13.7
11.5
13.6
0
0.6
0.1
0.4
0
1.0
3.9
0
0
0
0
0
0.5
0
0.3
0
0
0
0
0
0
0
0
0
0.2
0.7
0
0
0
4.0
0
0
33.7
11.0
29.4
0.5
49.6
6.2
90.8
7.3
67.7
0
28.4
93.5
10.0
5.6
5.2
9.2
63.2
0
83.5
73.7
57.2
56.2
69.3
71.9
0
0
0
3.1
7.7
3
4c
—
5
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
LiCl
6
7c
2.7
3.0
8
9
42.4
47.5
44.4
12.1
0.3
1.3
8.9
10.1
10.9
5.1
8.7
8.2
1.1
0
71.4
1.2
2.1
10.0
4.2
3.1
44.9
50.1
48.9
32.9
0.3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24c
25
26c
27
Co4(CO)12
0
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co4(CO)12
Ru3(CO)12, Co2(CO)8
RuBr3, CoBr2
0.5
0.8
0.2
0.3
0
4.7
0
0
0
0
0
0.1
1.1
0
7.9
TBACl
TPPTS
PPh3
33.8
23.6
24.9
16.6
31.0
8.2
1.1
0
73.7
1.3
3.3
Imidazole DMI
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
NMP
DMF
[Bmim]NTf2
1-Methylpiperidine
THF
Cyclohexane
H2O
DMI
DMI
DMI
0
0
0
0
0
0
0
0
0
2.3
0.1
1.1
22.6
8.5
4.2
36.4
70.3
66.9
57.5
33.7
12.7
7.3
(PPh3)3RuCl2, (PPh3)3CoCl LiBr
0
a
Reaction conditions: 40 mmol Ru catalyst and 20 mmol Co catalyst (based on the metal), 4 mmol promoter, 0.15 mmol cocatalyst, 2 mL solvent, 3
MPa CO2 and 6 MPa H2 (at room temperature), 200 ꢀC, 12 h. b STY stands for space time yield (C mmol Lꢁ1 hꢁ1). The STY was determined by GC
analysis using toluene as the internal standard. Black precipitate was observed aer the reaction. Sel.: selectivity.
c
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Edge Article
Chemical Science
the activity and/or selectivity of the catalyst were poor (entries 5– and transforming it into C2+OH. In addition, Brꢁ would occupy
10). The contribution of the lithium halide to C2+OH selectivity the active sites of the catalyst and inhibit the reaction when its
followed the order of: LiI > LiBr > LiCl, while their contribution dosage was high enough. The effect of PPNCl dosage on the
to the activity (STY) follows the reverse order (entries 1, 5 and 6). reaction is depicted in Fig. 1b. With the increase of PPNCl
The data show that the selectivities of the catalytic systems with dosage, the STY of C2+OH improved gradually, but it dropped
LiBr and LiI were similar, but the activity of the catalytic system when the dosage exceeded 0.15 mmol, which may be due to
with LiBr was much higher (entries 1 and 6). The excellent occupation of the active sites. In contrast, the STY of methanol
performance of the catalytic system with LiBr can be attributed always decreased with increasing the dosage of PPNCl. Hence
mainly to the presence of PPNCl, which enhanced the selectivity the appropriate dosage of PPNCl was 0.15 mmol. The above
signicantly, whilst retaining the high activity (entries 1 and 3). results also support the conclusion that the PPNCl promotes the
Thus LiBr was the best promoter for the above Ru–Co–PPNCl transformation of methanol into C2+OH.
catalyst in this reaction. In the previous work, the single Ru
The impact of reaction temperature is demonstrated in
catalyst using an iodide promoter had very poor performance Fig. 1c. Methanol and ethanol began to emerge at 140 ꢀC. A
for producing C2+OH.10 While in this work, the ethanol selec- minor amount of propanol and isobutanol appeared at 160 ꢀC.
tivity could reach 63.2% when the Ru–PPNCl catalyst was With the increase of reaction temperature, the STY and selec-
promoted by LiBr (entry 11). The Co catalyst itself showed very tivity of C2+OH enhanced evidently, but it became insensitive
poor catalytic performance (entry 12). But, when it was when the temperature was above 200 ꢀC. So a suitable reaction
combined with Ru catalyst, propanol and isobutanol were temperature was 200 ꢀC. The time course of the reaction is
produced and the selectivity of C2+OH increased to 90.8% (entry shown in Fig. 1d. Methanol and ethanol were formed in 1 h. At 3
1). Hence a synergistic effect existed in the Ru–Co–PPNCl h and 6 h, propanol and isobutanol began to appear, respec-
catalysts. The PPNCl was important for the catalytic properties. tively. With time going on, the yield and selectivity of C2+OH
Without PPNCl, the catalytic performance, especially the C2+OH increased rapidly, while the methanol content kept decreasing.
selectivity, was much lower (entry 3). We also tried other Aer 12 h, the change of methanol content was not evident and
cocatalysts, but the results were not satisfactory (entries 13–17). growth of the ethanol yield became slower, and at the same
The X-ray photoelectron spectroscopy (XPS) study revealed the time, the yield of propanol and isobutanol increased. As
coordination between Ru3(CO)12 and Clꢁ in PPNCl (Fig. S2†). a whole, the variation of C2+OH selectivity aer 12 h was not
The coordination increased the electron density of the Ru atom obvious. Fig. 1d also demonstrates that the amount of C2+OH
and could promote the oxidative addition of alkyl halides to the showed nearly a linear increase with reaction time, suggesting
active center, which is usually a key step in chain growth reac- that the water generated in situ had no considerable inuence
tions.12 The superiority of PPNCl over other chlorides (entries 1, on the activity of the catalyst.
13 and 14) may be due to the big steric hindrance of the
We also studied the recyclability of the catalytic system. Aer
substituents around the N atom, which weakened the electro- the reaction, the alcohols generated in the reaction were
static attraction between PPN+ and Clꢁ and the nucleophilicity removed under vacuum, which was conrmed by gas chroma-
of the Clꢁ was enhanced accordingly.13,14 Aer screening the tography. Then the catalytic system was reused directly for the
solvents, we found that DMI was the best for the reaction next run. The results of the recycling test are shown in Fig. 1e.
(entries 18–24). We also tried other Ru–Co combinations, but The STY and selectivity of C2+OH did not decrease obviously
the efficiency was lower than that of Ru3(CO)12 and Co4(CO)12 aer ve cycles (12 h each cycle), indicating that the catalytic
(entries 1 and 25–27). The Co2(CO)8 adopted in the literature10a system had excellent stability and reusability.
was not suitable here (entry 25).
As is shown in Fig. 1d, methanol is rstly generated and
The impact of catalyst dosage and gas pressure on the gradually consumed. At the same time, the C2+OH increased
reaction was studied and the results are given in Table 2. When accordingly. These phenomena suggest that methanol was rst
the total dosage of Ru and Co catalysts was xed, the optimized formed from CO2 hydrogenation, and then acted as an inter-
ratio of Ru/Co was 2 : 1 (entries 1–4 of Table 2). As expected, the mediate to produce larger alcohols. To support this assump-
increase of the catalyst dosage enhanced the catalytic efficiency, tion, we conducted tracer experiments by adding a small
but it was less sensitive when the dosage was large enough amount of 13CH3OH into the reaction. The GC-MS data indi-
(entries 3, 5 and 6 of Table 2). At a xed ratio of CO2 and H2, the cated that 13C appeared in all of the target C2+OH (Fig. S3†),
STY of the C2+OH increased rapidly with elevating pressure supporting the above argument. We also tried 13C2H5OH and
(entries 3, 7 and 8 of Table 2). The optimal ratio of CO2 and H2 obtained a similar result (Fig. S4†). Thus it can be concluded
was 1 : 2 at a given total pressure (entries 3 and 9–11 of Table 2). that in CO2 hydrogenation to generate the alcohols, the small
Fig. 1a shows that the STY and selectivity of C2+OH were alcohols acted as building blocks for the larger ones.
enhanced signicantly by increasing the LiBr dosage from 0–4
The possible mechanism for the synthesis of C2+OH from
mmol. When LiBr usage was further increased the STY of C2+OH CO2 hydrogenation is depicted in Scheme 2. Methanol and CO
decreased notably. Thus, the suitable dosage of LiBr was were generated via Ru catalyzed CO2 hydrogenation (Step 1).
4 mmol. In contrast, the STY of methanol increased evidently The methanol generated in situ was further converted into
when the LiBr usage increased from 0–2 mmol, whereas it ethanol via a hydrocarbonylation reaction (Steps 2–5). The
decreased drastically with further increasing of the LiBr dosage. generation of propanol from ethanol should follow similar
It is obvious that LiBr played a key role in generating methanol steps. Minor isobutanol was formed via the Guerbet reaction
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Table 2 Effect of reaction parameters on the synthesis of C2+ alcohols by CO2 hydrogenationa
STY [C mmol Lꢁ1 hꢁ1
]
C2+OH
Entry
Ru/Co [mmol]
pCO2 [MPa]
pH2 [MPa]
Methanol
Ethanol
Propanol
Isobutanol
Total
Sel. [%]
1
2
3
4
5
6
7
8
20/40
30/30
40/20
45/15
20/10
60/30
40/20
40/20
40/20
40/20
40/20
3
3
3
3
3
3
1
2
6
6
6
6
6
6
2
4
7.5
7.3
3.1
9.5
9.9
2.9
0.4
1.6
13.2
3.4
1.7
18.8
22.3
29.5
23.5
17.9
30.5
0
12.8
25.9
17.8
6.1
0.2
0.5
0.6
0.6
0.4
0.8
0
0.6
0.7
1.0
0.3
0.1
0.3
0.5
0.4
0.2
0.7
0
0.6
0.4
0.4
0.2
26.6
30.4
33.7
34.0
28.4
34.9
0.4
15.6
40.2
22.6
8.3
71.8
76.0
90.8
72.1
65.1
91.7
0
89.7
67.2
85.0
79.5
9
10
11
2.25
4.5
6
6.75
4.5
3
a
Reaction conditions: Ru3(CO)12 and Co4(CO)12 were used as catalyst precursors and their dosage was based on the metal, 4 mmol LiBr, 0.15 mmol
PPNCl, 2 mL DMI, 200 ꢀC, 12 h. Sel.: selectivity.
Scheme
2 Proposed mechanism of C2+OH synthesis from CO2
hydrogenation.
hydrogenated the acetaldehyde into ethanol. However, in this
work, the bromide promoted Ru catalyst predominated the
production of ethanol from methanol (entry 11 of Table 1),
while the single Co complex could not effectively catalyze the
reaction (entry 12 of Table 1). The Co catalyst ([Co]) in this work
mainly accelerated the generation of C2+OH in the reaction. The
coordination between the active Ru center (Ru*) and Clꢁ from
PPNCl enhanced the electron density of the metal center, which
would expedite the oxidative addition step (Step 3).17 Mean-
while, the increase of the electron density on the Ru* could
promote the hydrogenation step (Step 5).12
Fig. 1 Effect of reaction conditions (a–d) and results of recycling tests
(e) over 40 mmol Ru3(CO)12 and 20 mmol Co4(CO)12 (based on the
metal) in DMI under 9 MPa of the initial pressure (CO2/H2 ¼ 1/2): (a)
effect of LiBr dosage, 0.15 mmol PPNCl, 200 ꢀC, 12 h; (b) effect of
PPNCl dosage, 4 mmol LiBr, 200 ꢀC, 12 h; (c) effect of reaction
temperature, 4 mmol LiBr, 0.15 mmol PPNCl, 12 h; (d) effect of
reaction time, 4 mmol LiBr, 0.15 mmol PNNCl, 200 ꢀC; (e) the reaction
condition is the same as that of entry 1 in Table 1.
Conclusions
In summary, we have investigated different catalytic systems for
hydrogenation of CO2 to C2+OH. It was discovered that LiBr
could promote the reaction very efficiently using a Ru–Co
bimetallic catalyst with PPNCl as the cocatalyst. The catalytic
system could be reused ve times without an obvious loss of
catalytic performance. A synergistic effect existed between the
Ru and Co catalyst. Moreover, PPNCl enhanced the selectivity of
the catalytic system with LiBr signicantly, while keeping its
very high activity. Therefore, both the selectivity and activity
were very high in the presence of PPNCl and LiBr. The
between methanol and propanol.15 The Ru–halide catalyzed
synthesis of methanol and CO from CO2 hydrogenation has
been reported elsewhere.16 The mechanism of methanol
hydrocarboxylation using Ru–Co–iodide systems has been
extensively investigated.12 The Co catalyst was mainly respon-
sible for the generation of ethanol and acetaldehyde from
methanol hydrocarbonylation, and the Ru catalyst further
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Edge Article
Chemical Science
outstanding performance of the catalytic system results from GC-MS (SHIMADZU-QP2010) as well as by comparing the
the cooperative effect of Ru, Co, PPNCl, and LiBr. Very inter- retention times with respective standards in the GC traces.
estingly, the results of this work demonstrate that bromide is
To test the reusability of the catalytic system, the alcohols
a much better promoter than iodide in C2+OH synthesis via CO2 formed in the reaction were removed under vacuum at 80 ꢀC for
hydrogenation because the bromide promoter worked cooper- 2 h, and then the catalytic system was reused directly for the
atively with PPNCl. We believe that other bromide promoted next run.
catalytic systems with excellent performance can be explored for
the synthesis of alcohols from CO2 hydrogenation. It is also
Acknowledgements
instructive for designing catalysts of CO hydrogenation.
The authors thank the National Natural Science Foundation of
China (21373234, 21533011, 21173239, 21321063) and the
Experimental section
Chinese Academy of Sciences (KJCX2.YW.H30). We are also
Chemicals
grateful for the help of Jinchao Wei in GC-MS analysis.
Dodecacarbonyltriruthenium (Ru3(CO)12, 99%), dichlorotris-
(triphenylphosphine)ruthenium(II) ((PPh3)3RuCl2, 97%), ruth-
enium(III) bromide hydrate (RuBr3$xH2O, Ru 25% min),
dodecacarbonyltetracobalt (Co4(CO)12, 98%), chlorotris-
(triphenyl phosphine)cobalt(I) ((PPh3)3CoCl, 97%), cobalt(II)
bromide (CoBr2, 97%), lithium bromide (LiBr, 99%), lithium
iodide (LiI, 99.95%), lithium tetrauoroborate (LiBF4, 98%),
sodium bromide (NaBr, 99%), potassium bromide (KBr, 99%),
potassium iodide (KI, 99.9%), bis(triphenylphosphoranylidene)
ammonium chloride (PPNCl, 97%), triphenylphosphine (PPh3,
99%), imidazole (99%), 1-methyl-2-pyrrolidinone (NMP, 99%),
N,N-dimethylformamide (DMF, 99%), 1-methylpiperidine
(99%), tetrahydrofuran (THF, 99%), methanol, and cyclohexane
(99%) were purchased from Alfa Aesar China Co, Ltd. Dicobalt
octacarbonyl (Co2(CO)8), lithium chloride (LiCl, 98%), tetrabu-
tylammonium chloride (TBACl, 98%), and 1,3-dimethyl-2-imi-
dazolidinone (DMI, 98%) were provided by TCI Shanghai Co.,
Ltd. Tris(3-sulfonatophenyl)phosphine sodium salt hydrate
(TPPTS, 95%) was obtained from J&K Scientic Ltd. 1-Butyl-3-
methylimidazolium bis(triuoromethylsulfonyl)imide ([Bmim]-
NTf2, 99%) was purchased from Centre of Green Chemistry and
Catalysis, LICP, CAS. Toluene (99.8%) was obtained from Xilong
Chemical Co., Ltd. Methanol–13C (99 atom% 13C) and
ethanol–13C2 (99 atom% 13C) were provided by Sigma-Aldrich
Co. LLC. CO2 (99%) and H2 (99%) were supplied by Beijing
Analytical Instrument Company. All chemicals were used as
received.
References
1 (a) M. He, Y. Sun and B. Han, Angew. Chem., Int. Ed., 2013, 52,
9620–9633; (b) M. Aresta, A. Dibenedetto and A. Angelini,
Chem. Rev., 2014, 114, 1709–1742; (c) Q. Liu, L. Wu,
R. Jackstell and M. Beller, Nat. Commun., 2015, 6, 5933–
5947; (d) P. G. Jessop, F. Joo and C. C. Tai, Coord. Chem.
Rev., 2004, 248, 2425–2442; (e) W. Wang, S. Wang, X. Ma
and J. Gong, Chem. Soc. Rev., 2011, 40, 3703–3727.
2 (a) M. North and R. Pasquale, Angew. Chem., Int. Ed., 2009,
48, 2946–2948; (b) M. Cui, Q. Qian, Z. He, J. Ma, X. Kang,
J. Hu, Z. Liu and B. Han, Chem.–Eur. J., 2015, 21, 15924–
15928; (c) X.-H. Liu, J.-G. Ma, Z. Niu, G.-M. Yang and
P. Cheng, Angew. Chem., Int. Ed., 2015, 54, 988–991; (d)
O. Jacquet, X. Frogneux, C. D. N. Gomes and T. Cantat,
Chem. Sci., 2013, 4, 2127–2131; (e) Z. Han, L. Rong, J. Wu,
L. Zhang, Z. Wang and K. Ding, Angew. Chem., Int. Ed.,
2012, 51, 13041–13045; (f) R. Tanaka, M. Yamashita and
K. Nozaki, J. Am. Chem. Soc., 2009, 131, 14168–14169; (g)
C. Liu, J.-H. Xie, G.-L. Tian, W. Li and Q.-L. Zhou, Chem.
Sci., 2015, 6, 2928–2931; (h) C. Wu, Z. Zhang, Q. Zhu,
H. Han, Y. Yang and B. Han, Green Chem., 2015, 17, 1467–
1472; (i) L. Zhang, Z. Han, X. Zhao, Z. Wang and K. Ding,
Angew. Chem., Int. Ed., 2015, 54, 6186–6189; (j) L. Zhang
and Z. Hou, Chem. Sci., 2013, 4, 3395–3403.
3 (a) S. Wesselbaum, T. vom Stein, J. Klankermayer and
W. Leitner, Angew. Chem., Int. Ed., 2012, 51, 7499–7502; (b)
C. A. Huff and M. S. Sanford, J. Am. Chem. Soc., 2011, 133,
18122–18125; (c) N. M. Rezayee, C. A. Huff and
M. S. Sanford, J. Am. Chem. Soc., 2015, 137, 1028–1031; (d)
F. Liao, Y. Huang, J. Ge, W. Zheng, K. Tedsree, P. Collier,
X. Hong and S. C. Tsang, Angew. Chem., Int. Ed., 2011, 50,
2162–2165; (e) J. Ye, C.-j. Liu, D. Mei and Q. Ge, J. Catal.,
2014, 317, 44–53; (f) J. Graciani, K. Mudiyanselage, F. Xu,
A. E. Baber, J. Evans, S. D. Senanayake, D. J. Stacchiola,
P. Liu, J. Hrbek, J. Fernandez Sanz and J. A. Rodriguez,
Science, 2014, 345, 546–550; (g) S. Wesselbaum, V. Moha,
M. Meuresch, S. Brosinski, K. M. Thenert, J. Kothe,
T. v. Stein, U. Englert, M. Hoelscher, J. Klankermayer and
W. Leitner, Chem. Sci., 2015, 6, 693–704; (h) F. Studt,
I. Sharafutdinov, F. Abild-Pedersen, C. F. Elkjaær,
J. S. Hummelshøj, S. Dahl, I. Chorkendorff and
J. K. Nørskov, Nat. Chem., 2014, 6, 320–324.
Hydrogenation of CO2
The reactions were carried out in a 16 mL Teon-lined stainless
steel reactor equipped with a magnetic stirrer. In a typical
experiment, the desired amount of catalyst, cocatalyst,
promoter, tracer (if used) and 2 mL solvent were added into the
reactor. Aer the air in the reactor was removed under vacuum,
CO2 and H2 were charged into the reactor to the desired pres-
sure at room temperature. Then the reactor was placed in an air
bath of constant temperature, and the stirrer was started at 800
rpm. Aer reaction, the reactor was cooled in an ice-water bath
and the residual gas was released carefully in a hood. The
reaction solution was analyzed by a gas chromatograph (GC,
Agilent 7890B) equipped with a ame ionization detector (FID)
and a HP-5 capillary column. Toluene was used as the internal
standard. Identication of the liquid products was done using
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
4 D. L. S. Nieskens, D. Ferrari, Y. Liu and R. Kolonko Jr, Catal.
Commun., 2011, 14, 111–113.
M. Cui, Z. He, C. Wu, Q. Zhu, Z. Zhang, J. Ma, G. Yang,
J. Zhang and B. Han, Chem. Sci., 2015, 6, 5685–5689.
5 T. Tatsumi, A. Muramatsu and H. O. Tominaga, Chem. Lett., 11 B. D. Dombek, J. Organomet. Chem., 1983, 250, 467–483.
1985, 593–594. 12 M. E. Fakley and R. A. Head, Appl. Catal., 1983, 5, 3–18.
6 H. Kurakata, Y. Izumi and K. Aika, Chem. Commun., 1996, 13 H. Chen, L. N. Dawe and C. M. Kozak, Catal. Sci. Technol.,
389–390. 2014, 4, 1547–1555.
7 (a) T. Inui and T. Yamamoto, Catal. Today, 1998, 45, 209–214; 14 D. J. Darensbourg, R. M. Mackiewicz, J. L. Rodgers and
(b) T. Inui, T. Yamamoto, M. Inoue, H. Hara, T. Takeguchi
and J. B. Kim, Appl. Catal., A, 1999, 186, 395–406.
A. L. Phelps, Inorg. Chem., 2004, 43, 1831–1833.
15 S. Veibel and J. I. Nielsen, Tetrahedron, 1967, 23, 1723–1733.
8 S. Li, H. Guo, C. Luo, H. Zhang, L. Xiong, X. Chen and L. Ma, 16 K. I. Tominaga, Y. Sasaki, T. Watanabe and M. Saito, Bull.
Catal. Lett., 2013, 143, 345–355. Chem. Soc. Jpn., 1995, 68, 2837–2842.
9 Z. He, Q. Qian, J. Ma, Q. Meng, H. Zhou, J. Song, Z. Liu and 17 C. M. Thomas and G. Suss-Fink, Coord. Chem. Rev., 2003,
¨
B. Han, Angew. Chem., Int. Ed., 2016, 55, 737–741.
243, 125–142.
10 (a) K. Tominaga, Y. Sasaki, M. Saito, K. Hagihara and
T. Watanabe, J. Mol. Catal., 1994, 89, 51–56; (b) Q. Qian,
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016