Understanding the mechanism of the competitive adsorption in 8-methylquinoline hydrogenation over a Ru catalyst

Article abstract of DOI:10.1039/d0ra01277g

The competitive adsorption of 8-methylquinoline (8-MQL) and partially hydrogenated product, 4H-8-MQL, was studied by performing a combination of experiments and first-principles calculations over a selected Ru catalyst. A series of hydrogenation reactions were conducted with 8-MQL and 4H-8-MQL as initial reactants, respectively. 8-MQL exhibits stronger adsorption on catalyst surface active sites compared with 4H-8-MQL and the massive adsorption of 8-MQL hampers the further adsorption of 4H-8-MQL. The effects of temperature, pressure and solvent on the selectivity in 8-MQL hydrogenation were investigated as well. Full hydrogenation of 8-MQL to 10H-8-MQL was achieved within 120 min when the catalyst dosage increased from 5 wt% to 7 wt% under 160 °C and a hydrogen pressure of 7 MPa. The electronic charge of the N-heteroatom in 8-MQL and 4H-8-MQL was analyzed and the adsorption geometries of 8-MQL and 4H-8-MQL on the Ru(001) surface were optimized by DFT calculations to explain the competitive adsorption behaviors of 8-MQL and 4H-8-MQL.

Full text of DOI:10.1039/d0ra01277g

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Understanding the mechanism of the competitive
adsorption in 8-methylquinoline hydrogenation
over a Ru catalyst†
Cite this: RSC Adv., 2020, 10, 11039
Yuan Dong, Haoming Zhao, Zhenjie Liu, Ming Yang, *ab Zhenlin Zhang,
a
a
a
a
a
a
Ting Zhu and Hansong Cheng*
The competitive adsorption of 8-methylquinoline (8-MQL) and partially hydrogenated product, 4H-8-MQL,
was studied by performing a combination of experiments and first-principles calculations over a selected Ru
catalyst. A series of hydrogenation reactions were conducted with 8-MQL and 4H-8-MQL as initial
reactants, respectively. 8-MQL exhibits stronger adsorption on catalyst surface active sites compared
with 4H-8-MQL and the massive adsorption of 8-MQL hampers the further adsorption of 4H-8-MQL.
The effects of temperature, pressure and solvent on the selectivity in 8-MQL hydrogenation were
investigated as well. Full hydrogenation of 8-MQL to 10H-8-MQL was achieved within 120 min when the
Received 10th February 2020
Accepted 6th March 2020
ꢀ
catalyst dosage increased from 5 wt% to 7 wt% under 160 C and a hydrogen pressure of 7 MPa. The
electronic charge of the N-heteroatom in 8-MQL and 4H-8-MQL was analyzed and the adsorption
geometries of 8-MQL and 4H-8-MQL on the Ru(001) surface were optimized by DFT calculations to
explain the competitive adsorption behaviors of 8-MQL and 4H-8-MQL.
DOI: 10.1039/d0ra01277g
rsc.li/rsc-advances
2
3
basic CeO
2
.
98% 4H-QL was achieved over cobalt-based cata-
Introduction
ꢀ
lyst under 120 C and 2 MPa H with toluene as the solvent
The direct hydrogenation of quinoline (QL) and its derivatives is reported by Chen et al. Higher hydrogenation rates of 4H-QLs
2
24
considerably interesting for the industrial synthesis of a variety were observed in aqueous solvent compared to that in an
1–3
of petrochemicals, pharmaceuticals and ne chemicals. Both aprotic and apolar solvent using supported Pd and Ru cata-
10,18
partially hydrogenated products and fully hydrogenated prod- lyst.
Ionic liquid stabilized Ni and Pt nanoparticles were also
ucts are valuable chemicals as well as liquid organic hydrogen selected as highly efficient catalysts in chemoselective hydro-
4
–8
25–27
storage materials. Therefore, the role of reaction parameters genation of QLs.
In addition, homogeneous catalysts based
28
29,30
31
32,33
in the selectivity of QLs hydrogenation is widely investigated for on Ir, Ru,
obtaining the required product.
Mo, Co
Two hydrogenated inter- quinolines selectively as well.
mediates, 1,2,3,4-4H-QL and 5,6,7,8-4H-QL, were generated
The further hydrogenation, 4H-QL to 10H-QL, was proved to
along with the completely hydrogenated product, 10H-QL. It is be more difficult, which requires longer reaction time and
were proposed to hydrogenate
9–13
ꢀ
34,35
found that the selectivity to 4H-QL or 10H-QL is affected harsher conditions (T > 100 C, P > 5 MPa).
Only 0.6% 10H-
dramatically by the catalysts and solvents used in QL hydroge- QL was converted from 4H-QL aer QL hydrogenation for 14 h
1
4–16
3,17
18,19
ꢀ
36
nation.
Ru
Heterogeneous catalysts based on Rh,
Pd,
and Pt were successfully used in selective hydrogena- mostly by the irreversible adsorption of QL and/or 4H-QL on the
tion of QLs to 4H-QLs. Yu et al. reported that 99.8% 4H-QL can catalyst surface, also called catalyst poisoning, leading to the
over supported Ru at 150 C and 5 MPa. This was explained
20,21
22
37,38
be achieved over prepared Ru/OMNC (ordered mesoporous N- hydrogenation stops aer 4H-QL production.
doped carbon) catalyst with EtOH as solvent under 40 C and effective methods have been developed to realize selective
Actually, some
ꢀ
2
1
1
MPa H
2 2
. Pt/CeO
was found to be highly efficient and che- hydrogenation by adjusting the adsorption behaviors of QL and
39
ꢀ
moselective hydrogenation of QL into 4H-QL at 60 C, which 4H-QL. Addition of a Lewis base or introduction of aprotic
4
0,41
was attributed to the metal-support interaction between Pt and solvent such as n-hexane
to the reaction mixture was in favor
of 10H-QL formation. Fan et al. reported an excellent catalyst for
40
complete hydrogenation of QL, Rh/AlO(OH). They proposed
that the hydrogen bond formed between hydroxyl group of Rh/
AlO(OH) and the aromatic cycle of 4H-QL promotes 10H-QL
production. However, the particular research on the competi-
tive adsorption behaviors of QLs and 4H-QLs have been rarely
reported.
a
Sustainable Energy Laboratory, Faculty of Materials Science and Chemistry, China
University of Geosciences, Wuhan 430074, P. R. China. E-mail: yangming8180@
gmail.com; chenghs@cug.edu.cn
b
Zhejiang Institute, China University of Geosciences, Hangzhou 311305, P. R. China
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/d0ra01277g
1
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 11039–11045 | 11039

View Article Online
RSC Advances
Paper
In the present work, 8-methylquinoline (8-MQL) hydroge- experiment were: 0.2 g Ru/Al
2
O
3
, 4 g 4H-8-MQL, 40 ml dioxane,
ꢀ
nation was investigated over commercial Ru/Al O under 7 MPa hydrogen pressure and 160 C reaction temperature.
2
3
designed conditions to elaborate the competitive adsorption Fig. 2 demonstrates that the fully hydrogenated product, 10H-8-
between 8-MQL and 4H-8-MQL. A series of hydrogenation MQL was obtained as soon as the reaction started. Moreover,
reactions were conducted with 8-MQL and 4H-8-MQL as initial the same catalyst was reused 3 times for 4H-8-MQL hydroge-
reactants, respectively. It is found that the adsorption of 4H-8- nation. No signicant decrease in hydrogenation rate was
MQL on Ru/Al O is weaker than that of 8-MQL and 4H-8-MQL observed, indicating that Ru/Al O is highly active to catalyze
2 3 2 3
is not responsible for the catalyst poisoning. The role of reac- 4H-8-MQL to form 10H-8-MQL and that 4H-8-MQL is not
tion parameters including temperature, pressure and solvent in responsible for poisoning the catalyst surface active sites.
8-MQL hydrogenation was studied in detail. In addition, the
The reaction pathway for 8-MQL and 4H-8-MQL hydrogena-
adsorption geometries of 8-MQL and 4H-8-MQL was optimized tion has been described in Scheme 1. Both of process (1) and (3)
by DFT calculations and the corresponding adsorption energies can be realized successfully. In contrast, the process (2) of 10H-
were also compared.
8-MQL formation in hydrogenation of 8-MQL was rarely
observed even when the reaction temperature was increased to
ꢀ
1
80 C. By comparing process (1) and process (3), the conversion
Results and discussion
Hydrogenation of 8-MQL to form 4H-8-MQL
rate of 4H-8-MQL was slower than that of 8-MQL under the
same reaction conditions. 85% of 8-MQL can be converted to
The standard hydrogenation conditions designed in the present 4H-8-MQLwithin 120 min at 160 C and 7 MPa, while only 30%
work were: 0.2 g Ru/Al O catalyst, 4 g 8-MQL, 40 ml dioxane of 4H-8-MQL was consumed to produce 10H-8-MQL.
ꢀ
2
3
with the hydrogen pressure of 7 MPa and reaction temperature
It is widely acknowledgement that adsorption on the catalyst
ꢀ
range of 160–180 C. As shown in Fig. 1, almost 100% of 8-MQL active sites was an essential prerequisite for hydrogenation
was converted to form 4H-8-MQL under designed conditions reaction to occur. Therefore, we speculated that 8-MQL were
detected by GCMS. The structure of the produced 4H-8-MQL more easily adsorbed on the active sites such as metal surface
1
was determined to be 1,2,3,4-4H-8-MQL by H NMR, implying and acid sites than that of 4H-8-MQL, leading to the desorption
that 8-MQL hydrogenation rstly occurred in N-heterocycle as of 4H-8-MQL and thus 10H-8-MQL was rarely produced. The re-
a result of the strong interaction between N atom and the exposure of active sites was in favor of the continuous hydro-
catalyst active site. The NMR spectrum was presented in genation of 8-MQL. This hypothesis was proved by adding a new
ꢀ
ꢀ
Fig. S1.† Increase temperature from 160 C to 180 C, the portion of 8-MQL aer the previous 8-MQL was almost totally
conversion rate of 8-MQL was accelerated obviously whereas converted to 4H-8-MQL. The results showed that a further 50%
only a trace of fully hydrogenated products (yield < 1%), 10H-8- conversion of the fresh 8-MQL to 4H-8-MQL can be obtained
MQL, was detected. Similarly, hydrogenation of QL to produce within 60 min. The weaker interaction of 4H-8-MQL with
1
0H-QL catalyzed by supported noble metal catalyst was also surface active sites compared with that of 8-MQL probably
found to be very difficult, which was roughly attributed to the results in the difficulty in effective adsorption on active sites
strong adsorption of QL and/or 4H-QL on the effective active and low conversion rate for its further hydrogenation.
23,40
sites.
Hydrogenation of 8-MQL to form 10H-8-MQL
Hydrogenation of 4H-8-MQL to form 10H-8-MQL
In the view of the previous studies, it is can be deduced that 4H-
The 4H-8-MQL produced from 8-MQL hydrogenation was used
as reactant aer distillation. The standard conditions in this
8-MQL could probably be further hydrogenated to form 10H-8-
Fig. 1 Hydrogenation of 8-MQL at different temperatures over Ru/
ꢀ
Al
O
2 3
with catalyst dosage proportion of 5 wt% and hydrogen pressure Fig. 2 Hydrogenation of 4H-8-MQL at 160 C and 7 MPa with the
of 7 MPa.
same Ru/Al
O
2 3
catalyst for 3 cycles.
11040 | RSC Adv., 2020, 10, 11039–11045
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
RSC Advances
Scheme 1 Hydrogenation pathway of 8-MQL and 4H-8-MQL at
ꢀ
Fig. 4 Influence of catalyst dosage proportions on the hydrogenation
1
60 C and 7 MPa over Ru/Al
2
O
3
catalyst with the dosage proportion of
of 8-MQL. General reaction conditions: 4 g 8-MQL, 40 ml dioxane as
5
wt%.
ꢀ
solvents, 160 C, 7 MPa H
2
.
MQL if there were enough effective active sites in 8-MQL
hydrogenation. Hence, another dose of fresh catalyst was
introduced to the reaction mixture aer complete 8-MQL was
converted to 4H-8-MQL. The result has been presented in Fig. 3.
attributed to the more effective active sites given by the catalyst
with higher mass ratio, which makes more opportunities for
successful adsorption of 4H-8-MQL on catalyst surface and
benets the further hydrogenation of 4H-8-MQL.
4H-8-MQL was consumed rapidly as soon as the reaction star-
Fig. 5a and b display the hydrogen uptake curves vs. time of
ted. Finally, 10H-8-MQL with a yield of 100% was obtained. The
results conrmed that the absence of 10H-8-MQL in 8-MQL
hydrogenation is attributed to the catalyst poisoning caused by
strong adsorption of 8-MQL occurred in the process of 8-MQL to
8-MQL hydrogenation at different temperatures and pressures
with the catalyst dosage proportion of 10 wt%, respectively. It is
found that complete 10H-8-MQL can be produced within
ꢀ
6
0 min at 160 C and 7 MPa. The apparent active energy of 8-
4
H-8-MQL. However, it is believed that Ru/Al O is essentially
2 3
ꢁ
1
MQL consumption was derived to be 75.2 kJ mol based on the
rst order reaction kinetics model. The products concentration
capable of hydrogenation of 8-MQL to form 10H-8-MQL. This
assumption was conrmed by adding different dosage propor-

distribution curves of 8-MQL hydrogenation at different
temperatures and pressures have been presented in Fig. S2 and
S3,† respectively. The consumption of 8-MQL gives rise to a fast
accumulation of 4H-8-MQL. Subsequently, the conversion from
tion of Ru/Al
2
O
3
in the range of 4 wt% to 20 wt% for 8-MQL
ꢀ
hydrogenation at 160 C and 7 MPa. The hydrogen uptake
curves vs. reaction time have been displayed in Fig. 4.
Table 1 summarizes the conversion and selectivity during 8-
MQL hydrogenation with catalysts introduced in different mass
ratios. At the catalyst dosage proportion of 4 wt% and 5 wt%, the
rapid consumed 8-MQL was totally converted to 4H-8-MQL in
the rst 120 min. Keeping the reactions for another 4 hours, still
almost no 10H-8-MQL was observed. Differently, 4H-8-MQL was
gradually converted to form 10H-8-MQL aer 30 min when the
4H-8-MQL to 10H-8-MQL starts. Particularly worth mentioning,
the production rates of 10H-8-MQL accelerate sharply when the
concentration of 4H-8-MQL reaches the maximum, indicating
that the 8-MQL hydrogenation undergoes two steps:
8-MQL / 4H-8-MQL and 4H-8-MQL / 10H-8-MQL, and the
second step is the rate-controlling step as stated previously.
In order to make an investigation on the catalyst degrada-
dosage proportion of Ru/Al
2 3
O was increased from 5 wt% to
tion, the same catalyst was reused for 7 times hydrogenation of
7
wt%. Further increase the dosage proportion to 20 wt%, 100%
ꢀ
8
-MQL at 160 C and 7 MPa. The results have been displayed in
of 10H-8-MQL can be produced within 15 min. This can be
Fig. 5c and d. Clearly, full hydrogenation of 8-MQL was observed
2 3
only in the rst hydrogenation catalyzed by the fresh Ru/Al O .
However, as expected, almost 100% of 4H-8-MQL was obtained
and very few 10H-8-MQL was observed in the remaining cycles.
In addition, the conversion rates of 8-MQL gradually decreased
with the increase of cycle times. This is probably because more
effective active sites were covered by strong adsorption of 8-
MQL. Besides, the carbon deposits may be also responsible for
the decrease of hydrogenation rate aer recycling of the cata-
42
2 3
lyst. For the rst cycle, the active sites of fresh Ru/Al O are
enough to catalyze 8-MQL to form 4H-8-MQL and 10H-8-MQL.
For the remaining cycles, the new 8-MQL competes for effective
active sites, some of which were probably gradually occupied by
ꢀ
Fig. 3 Hydrogenation of 8-MQL at 160 C and 7 MPa over Ru/Al
2
O
3
ꢀ
(
4 g 8-MQL, 0.2 g Ru/Al
2
O
3
, 40 ml dioxane, 7 MPa H
80 min of reaction time another 0.1 g of fresh Ru/Al
2
, 160 C). After
1
2
O
3
was intro-
duced. (a) The hydrogen uptake curves for the whole process of 8- 8-MQL with the increase of cycle numbers, thus hydrogenation
MQL hydrogenation, (b) evolutions of 8-MQL and the hydrogenated rates decreased and 4H-8-MQL was difficult to be further
products during the whole hydrogenation process.
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 11039–11045 | 11041

View Article Online
RSC Advances
Paper
ꢀ
Table 1 Conversion and selectivity of 8-MQL hydrogenation with different catalyst dosage proportions at 160 C and 7 MPa
Selectivity (%)
Catalyst dosage
proportion (wt%)
Reaction time
(h)
Conversion
(%)
Test
4H-8-MQL
10H-8-MQL
1
2
3
4
5
4
2
5
2
5
2
5
2
5
2
5
83.4
100
91.3
100
100
100
100
100
100
100
100
99.1
100
99.5
78.2
23.1
0
0
0
0
0
0.9
0
0.5
20.9
76.9
100
100
100
100
5
7
10
20
15
hydrogenated to form 10H-8-MQL. At the 7th cycle, the hydroxyl and N-heterocycle of 8-MQL, which weakens the
ꢀ
ꢀ
temperature was raised from 160 C to 200 C, which benets strong adsorption of 8-MQL on the active sites and benets the
the hydrogenation rate but still 10H-8-MQL was rarely obtained. adsorption of 4H-8-MQL. For the base solvent, N,N-iso-
Competitive adsorption shows a huge impact on the reaction propylethylamine, the steric hindrance produced by the amine
rates in 8-MQL hydrogenation. For quinoline substitutes, the and its electronic interaction with the catalyst active sites
43
adsorption of the reactant and its intermediate(s) was mainly probably avoided the strong adsorption of 8-MQL, thus 100%
through aromatic ring and nitrogen electron double. The 10H-8-MQL can be obtained in 240 min over Ru/Al in the
solvents used in hydrogenation of quinoline derivatives play an presence of N,N-diisopropylethylamine (10 ml) and dioxane (30
2 3
O
39,43
important role in adsorption equilibrium.
Fig. 6 shows the ml) as solvent.
concentration distributions of intermediates and products in 8-
MQL hydrogenation with different solvents. The reactions were
Adsorption of 8-MQL and 4H-8-MQL on Ru(0 0 1) surface
ꢀ
conducted at 160 C and 7 MPa with Ru/Al O dosage propor-
2
3
The geometries of 8-MQL and 4H-8-MQL were optimized by DFT
tion of 5 wt%. In the non-protic solvents, dioxane and toluene,
almost complete conversion of 8-MQL was found to form 4H-8-
MQL with only a trace of 10H-8-MQL produced aer 240 min.
However, the conversion from 4H-8-MQL to 10H-8-MQL was
observed in 120 min with ethanol as solvent in our experiment.
A reason might be due to the hydrogen bonds formed between
3
calculations using Material Studio Dmol. The electronic
charges of N-heteroatom in 8-MQL and 4H-8-MQL were ꢁ0.158
and ꢁ0.124 calculated by Hirshfeld method as displayed in
Fig. 7a. The adsorption geometries of 8-MQL and 4H-8MQL
were also optimized as shown in Fig. 7b. Kinds of possible
adsorption geometries of 8-MQL were investigated. It is found
that the optimized adsorption geometry of 8-MQL kept the two
rings aligned parallel to the surface at a Ru–N distance of 2.150
Fig. 5 Time-dependent hydrogen uptake curves of 8-MQL hydro-
genation (a) at different temperatures (conditions: 4 g 8-MQL, 0.4 g
Ru/Al
conditions: 4 g 8-MQL, 0.4 g Ru/Al
with the reused Ru/Al catalyst (conditions: 4 g 8-MQL, 0.4 g Ru/ Al
Al , 40 ml dioxane, 160 C, 7 MPa H
2 3 2 2
O , 40 ml dioxane, 7 MPa H ) and (b) at different H pressures
ꢀ
(
2 3
O , 40 ml dioxane, 160 C) and (c) Fig. 6 The effects of solvents on hydrogenation of 8-MQL over Ru/
: (a) dioxane (40 ml), (b) toluene (40 ml), (c) ethanol (40 ml), (d)
2
O
3
2 3
O
ꢀ
2
O
3
2
); (d) conversion and selectivity N,N-diisopropylethylamine (10 ml) + dioxane (30 ml). General reaction
ꢀ
of 8-MQL hydrogenation in 60 min for the 7 times cycles.
conditions: 4 g 8-MQL, 0.2 g Ru/Al
2
O , 7 MPa H
3
2
, 160 C.
11042 | RSC Adv., 2020, 10, 11039–11045
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
RSC Advances
Experimental
Materials
8-MQL was provided from Alligator Reagent, which was further
puried by vacuum distillation. Commercial 5 wt% Ru/Al O
2 3
catalyst was supplied by Shanxi Kaida Chemical Engineering
Company Limited. Sinopharm Chemical Reagent Company
Limited supplied 1,4-dioxane, ethanol and toluene. Energy
Chemical Company provided N,N-diisopropylethylamine. All
solvents and catalysts were used as received. Ultra-high purity of
2
H and Ar were provided by Sichuan Ally High-Tech Company.
All liquid samples taken from reactors during experiments were
regularly detected by Gas Chromatography and Mass Spec-
trometry (Agilent 7890/5975C, GCMS) to determine the
Fig. 7 (a) The atomic charges distributions of 8-MQL and 4H-8-MQL;
(
b) the optimized adsorption geometries of 8-MQL and 4H-8-MQL
over Ru(0 0 1) surface with showing the distance between the N atom composition. The samples diluted by hexane were carried by
ꢁ
1
and the nearest Ru surface atom. Green: Ru atoms, blue: N, gray: C, ultra-high purity He at a ow rate of 1.5 ml min into the
white: H.
chromatographic column (HP-5, 30 m ꢂ 320 mm ꢂ 0.25 mm).
ꢀ
The injection temperature was 300 C and split ratio was 1 : 100.
ꢀ
ꢀ
The oven temperature started at 90 C, then increased to 220 C
ꢀ
ꢀ
ꢁ1
ꢀ
A with the adsorption energy of 1.8 eV. Preferential hydroge-
by 12 C min subsequently and kept 220 C for 1 min. Nuclear
magnetic resonance (Bruker 400, NMR) was applied to deter-
mine the structure of 4H-8-MQL produced in 8-MQL
hydrogenation.
nation of 8-MQL occurs in N-heterocycle in the view of large
amount of 1,2,3,4-4H-8-MQL formed in hydrogenation. It is can
be explained by the strong coordination interaction of N atom
to Ru surface. Two types of geometries of 4H-8-MQL adsorbed
on Ru(001) were optimized. Its aromatic ring aligned parallel to
the surface was recorded as 4H-8-MQL-p and the aromatic ring Hydrogenation of 8-MQL and 4H-8-MQL
was positioned inclined to the Ru surface recorded as 4H-8-
Hydrogenation process was carried out at a 600 ml stainless
MQL-i. The distance between N atom and the nearest Ru atom
steel batch reactor (Parr 4568). Aer the reactants were loaded,
the reactor was sealed and ushed with hydrogen to remove air
for 4 times. Then the reactor was heated to the designed
temperature and pressurized to the designed pressure. The
pressure kept unchanged over the entire experiment by adding
hydrogen continuously. The reaction time was recorded as zero
when the magnetic stirring was started with a speed of 600 rpm.
The liquid samples were taken periodically and analyzed by
GCMS.
ꢀ
for 4H-8-MQL-p and 4H-8-MQL-i were 2.605 A with the
ꢀ
adsorption energy of 0.6 eV and 2.657 A with the adsorption
energy of 0.4 eV, respectively. Obviously, 4H-8-MQL-p exhibited
a stronger adsorption to 4H-8-MQL-i. This suggests that the
adsorption of 4H-8-MQL was mainly via conguration 4H-8-
MQL-p, which can be attributed to the p-coordination of phenyl
44
at Ru surface.
Scheme 2 summarized the conversion and selectivity in
hydrogenation of 4H-8-MQL and 8-MQL with reused catalysts.
With more electronegativity of the N atom, basic 8-MQL is easier
to adsorb on the metal surface by coordination and occupy the DFT calculations
acid sites on the support with higher adsorption energy. 4H-8-
The geometry congurations of 8-MQL and 4H-8MQL were
optimized by the soware of Material Studio DMol. The double
numerical plus polarization (DNP) was applied to analyze
MQL can be hydrogenated to form 10H-8-MQL with reused Ru/
Al O in 3 cycles, indicating that 4H-8-MQL was not responsible
3
2
3
for poisoning the catalyst. For the process of 8-MQL hydroge-
nation, the occupied acid sites on the supported by 8-MQL
prevent the 4H-8-MQL from adsorbing and being active.
Consequently, 10H-8-MQL was rarely produced with limited
catalyst active sites. Increase of the catalyst dosage proportion
was benet of the formation of 10H-8-MQL. However, the new
added 8-MQL competes for effective active sites, some of which
were probably gradually occupied by 8-MQL with the increase of
cycle numbers, thus very little 10H-8-MQL was obtained in the
remaining cycles. It is noted that the adsorption energies of the
aromatic molecules are involved with the adsorption geome-
tries, activated sites and adsorption sites. Nevertheless, the
adsorption of QL substitutes benets from the electronic
charges of N atom, which leads to the strong interaction with
the catalyst surface.
4
5,46
valence electron orbitals.
to calculate nonlocal exchange and interrelation energies based
Perdew–Wang-91 (PW91) was used
47
on the generalized gradient approximation (GGA). The charges
of N-heteroatom in 8-MQL and 4H-8-MQL were calculated by
48
Hirshfeld method. The Ru(0 0 1) surface with a 5 ꢂ 5 supercell
ꢀ
and three Ru layers with 20 A vacuum region was built. The
adsorption geometries of 8-MQL and 4H-8-MQL was optimized
by stabilizing the last layer of Ru atoms in the supercell and
relaxing the molecule on the top two layers of Ru(0 0 1). The
adsorption energy was calculated by the Vienna Ab initio
Simulation Package (VASP) according to the equation below:
Eadsorption ¼ Esurface + Emolecule ꢁ Esurface–molecule
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 11039–11045 | 11043

View Article Online
RSC Advances
Paper
Scheme 2 Summary of catalyst cycles in hydrogenation of 4H-8-MQL and 8-MQL under designed conditions.
The Esurface, Emolecule and Esurface–molecule were the energy of catalyst active sites by the interaction of N atom and Ru atoms
the surface, the energy of absorbed molecule and the adsorp- compared with 4H-8-MQL. Hydrogenation of 8-MQL in mild
tion energy of the system. A plane wave basis set with a cutoff conditions mainly generates 4H-8-MQL. It is probably because
energy of 400 eV was used to solve the Kohn–Sham equation and of the strong adsorption of 8-MQL, which occupies most of the
the electron–ion interactions were described by the projector active sites on the support and metal active sites. Fully hydro-
49
augmented wave (PAW) method. The Brillouin zone was genated product can be detected by increasing the catalyst
. In addition, the production of
sampled within a 2 ꢂ 2 ꢂ 1 Monkhorst–Pack k-point mesh. dosage proportion of Ru/Al
2 3
O
Electron smearing was employed by using the Methfessel–Pax- 10H-8-MQL can be promoted by introducing a protic solvent or
ton technique with a width of 0.2 eV to minimize the errors in Lewis base. The selectivity to 4H-8-MQL or 10H-8-MQL is related
the Hellmann–Feynman forces. Structural optimizations were with the competitive adsorption between 8-MQL and 4H-8-
performed until the total energy of the system was converged to MQL. Increasing the metal dispersion and the amount of acid
ꢁ3
less than 10 eV.
sites on the supported is probably another way to benet the
formation of 10H-8-MQL in 8-MQL hydrogenation.
Conclusions
Conflicts of interest
4H-QLs are formed as the main products in hydrogenation of
There are no conicts to declare.
QLs under mild conditions in general. This was explained
mostly by the irreversible adsorption of QL and/or 4H-QL on the
catalyst surface. The particular research on the adsorption
behaviors of QLs and 4H-QLs have been rarely reported. In our
Acknowledgements
work, it is demonstrated that 4H-8-MQL is not responsible for The authors gratefully acknowledge the nancial support from
the catalyst poisoning. The competitive adsorption behaviors of the National Key Research and Development Program of China
8
-MQL and 4H-8-MQL on Ru/Al O was compared by a series of (No. 2018YFB1502903), the National Natural Science Founda-
2 3
hydrogenation reactions and DFT calculations. The adsorption tion of China (No. 21603195 and No. 21875225), the Zhejiang
geometries and adsorption energies of 8-MQL and 4H-8-MQL Provincial Natural Science Foundation of China (No.
were conrmed. It is found 8-MQL is easier to adsorb on the LGG20B030001), the Fundamental Research Funds for Central
11044 | RSC Adv., 2020, 10, 11039–11045
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
RSC Advances
Universities (CUGL170405 and CUG180604), Major Project of 24 F. Chen, A. E. Surkus, L. He, M. M. Pohl, J. Radnik, C. Topf,
Technical Innovation of Hubei Province (No. 2019AAA163).
K. Junge and M. Beller, J. Am. Chem. Soc., 2015, 137, 11718–
1724.
5 H. Y. Jiang, S. S. Zhang and B. Sun, Catal. Lett., 2018, 148,
336–1344.
1
2
Notes and references
1
1
2
3
F. Sotoodeh, B. J. M. Huber and K. J. Smith, Int. J. Hydrogen 26 X. Yuan, N. Yan, C. Xiao, C. Li, Z. Fei, Z. Cai, Y. Kou and
Energy, 2012, 37, 2715–2722.
P. J. Dyson, Green Chem., 2010, 12, 228–233.
A. M. Maj, I. Suisse, C. Hardouin and F. Agbossou- 27 Y. Hu, Y. Yu, Z. Hou, H. Yang, B. Feng, H. Li, Y. Qiao,
Niedercorn, Tetrahedron, 2013, 69, 9322–9328.
X. Wang, L. Hua, Z. Pan and X. Zhao, Chem.–Asian J., 2010,
5, 1178–1184.
M. Campanati, M. Casagrande, I. Fagiolino, M. Lenarda,
L. Storaro, M. Battagliarin and A. Vaccari, J. Mol. Catal. A: 28 A´ . Vivancos, M. Beller and M. Albrecht, ACS Catal., 2017, 8,
Chem., 2002, 184, 267–272.
17–21.
T. Ben Issa, C. Ben Ali Hassine, H. Ghalla, H. Barhoumi and 29 R. A. S ´a nchez-Delgado and E. Gonz ´a lez, Polyhedron, 1989, 8,
L. Benhamada, J. Mol. Struct., 2018, 1162, 71–80.
1431–1436.
V. Sridharan, P. A. Suryavanshi and J. C. Menendez, Chem. 30 R. H. Fish, A. D. Thormodsen and G. A. Cremer, J. Am. Chem.
Rev., 2011, 111, 7157–7259.
Soc., 1982, 104, 5234–5237.
D. Ventura-Espinosa, A. Marza-Beltran and J. A. Mata, 31 G. Zhu, K. Pang and G. Parkin, J. Am. Chem. Soc., 2008, 130,
Chem.–Eur. J., 2016, 22, 17758–17766.
1564–1565.
P. T. Aakko-Saksa, C. Cook, J. Kiviaho and T. Repo, J. Power 32 R. B. Xu, S. Chakraborty, H. M. Yuan and W. D. Jones, ACS
Sources, 2018, 396, 803–823.
Catal., 2015, 5, 6350–6354.
Y. Wu, Z. Chen, W. C. Cheong, C. Zhang, L. Zheng, W. Yan, 33 I. Sorribes, L. Liu, A. Dom ´e nech-Carb ´o and A. Corma, ACS
R. Yu, C. Chen and Y. Li, Chem. Sci., 2019, 10, 5345–5352.
Catal., 2018, 8, 4545–4557.
C. H. Yang, X. Chen, H. Li, W. Wei, Z. Yang and J. Chang, 34 Z. Wei, Y. Chen, J. Wang, D. Su, M. Tang, S. Mao and
Chem. Commun., 2018, 54, 8622–8625.
Y. Wang, ACS Catal., 2016, 6, 5816–5822.
0 D. M. Zhu, H. B. Jiang, L. Zhang, X. L. Zheng, H. Y. Fu, 35 F. Martinez-Espinar, P. Blondeau, P. Nolis, B. Chaudret,
4
5
6
7
8
9
1
M. L. Yuan, H. Chen and R. X. Li, ChemCatChem, 2014, 6,
C. Claver, S. Castill ´o n and C. Godard, J. Catal., 2017, 354,
2954–2960.
113–127.
11 X. Qiao, M. El-Shahat, B. Ullah, Z. Bao, H. Xing, L. Xiao, 36 Y. P. Sun, H. Y. Fu, D. l. Zhang, R.-X. Li, H. Chen and X.-J. Li,
Q. Ren and Z. Zhang, Tetrahedron Lett., 2017, 58, 2050–2053.
Catal. Commun., 2010, 12, 188–192.
12 M. G. Manas, L. S. Sharninghausen, E. Lin and 37 L. M. Kustov, A. L. Tarasov and B. P. Tarasov, Int. J. Hydrogen
R. H. Crabtree, J. Organomet. Chem., 2015, 792, 184–189.
Energy, 2013, 38, 5713–5716.
13 P. Ryabchuk, A. Agapova, C. Kreyenschulte, H. Lund, 38 R. M. Zhang, G. Y. Fan, C. Li, Y. Y. Wang, R. X. Li, H. Chen
H. Junge, K. Junge and M. Beller, Chem. Commun., 2019,
and X. J. Li, Acta Phys.-Chim. Sin., 2008, 24, 965–970.
39 M. C. M. Campanati, I. Fagiolino, M. Lenarda, L. Storaro,
M. Battagliarin and A. Vaccari, J. Mol. Catal. A: Chem.,
2002, 184, 6.
55, 4969–4972.
1
1
1
1
4 F. Fache and O. Piva, Synlett, 2004, 1294–1296, DOI: 10.1055/
s-2004-825603.
5 H. Mao, C. Chen, X. Liao and B. Shi, J. Mol. Catal. A: Chem., 40 G. Y. Fan and J. Wu, Catal. Commun., 2013, 31, 81–85.
2
011, 341, 51–56.
41 F. A. K. Bing Sun, A. Vallat and G. S u¨ ss-Fink, Appl. Catal., A,
2013, 467, 5.
42 B. F. M. F. Williams, A. Jentys, C. Breitkopf, J. A. R. van Veen
and J. A. Lercher, J. Phys. Chem. C, 2010, 114, 10.
6 C. Wan, Y. An, F. Chen, D. Cheng, F. Wu and G. Xu, Int. J.
Hydrogen Energy, 2013, 38, 7065–7069.
7 M. Campanati, S. Franceschini, O. Piccolo and A. Vaccari,
New ways for the full hydrogenation of quinoline in mild 43 A. V. M. Campanati and O. Piccolo, J. Mol. Catal. A: Chem.,
conditions, CRC Press-Taylor & Francis Group, Boca Raton,
2002, 179, 6.
2
005.
44 I. Favier, S. Massou, E. Teuma, K. Philippot, B. Chaudret and
M. Gomez, Chem. Commun., 2008, 3296–3298, DOI: 10.1039/
b804402c.
1
8 M. M. Dell'Anna, V. F. Capodiferro, M. Mali, D. Manno,
P. Cotugno, A. Monopoli and P. Mastrorilli, Appl. Catal., A,
2
014, 481, 89–95.
9 Y. Gong, P. Zhang, X. Xu, Y. Li, H. Li and Y. Wang, J. Catal.,
013, 297, 272–280.
0 B. Sun, D. Carnevale and G. S u¨ ss-Fink, J. Organomet. Chem.,
016, 821, 197–205.
45 F. W. Averill and D. E. Ellis, J. Chem. Phys., 1973, 59, 6412–
6418.
46 A. D. Becke and R. M. Dickson, J. Chem. Phys., 1988, 89,
2993–2997.
1
2
2
2
2
2
2
47 A. D. Becke, Int. J. Quantum Chem., 1985, 27, 10.
1 X. Yu, R. Nie, H. Zhang, X. Lu, D. Zhou and Q. Xia, 48 X. Y. Zhou, C. Y. Rong, T. Lu and S. B. Liu, Acta Phys.-Chim.
Microporous Mesoporous Mater., 2018, 256, 10–17.
Sin., 2014, 30, 2055–2062.
2 H. Bernas, N. Kumar, A. Aho, R. Leino and D. Y. Murzin, 49 G. Kresse and D. Joubert, Phys. Rev. B: Condens. Matter Mater.
Catal. Commun., 2014, 56, 41–44.
Phys., 1999, 59, 1758–1775.
3 S. Zhang, Z. Xia, T. Ni, Z. Zhang, Y. Ma and Y. Qu, J. Catal.,
2018, 359, 101–111.
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 11039–11045 | 11045