Optimization of the transfer hydrogenation reaction of acetophenone on Ni@MOF-5 nanoparticles using response surface methodology

Article abstract of DOI:10.1007/s11164-019-03959-1

In this study, the transfer hydrogenation reduction of acetophenone using supported nickel on MOF as catalyst was investigated. BBD and RSM were employed to investigate the effect of the experimental parameters such as Ni content, catalyst content, temper

Full text of DOI:10.1007/s11164-019-03959-1

Research on Chemical Intermediates
https://doi.org/10.1007/s11164-019-03959-1
Optimization of the transfer hydrogenation reaction
of acetophenone on Ni@MOF‑5 nanoparticles using
response surface methodology
Azadeh Nemati Chelavi1 · Vahid Zare‑Shahabadi1 · Soheil Sayyahi1 ·
Hossein Anaraki‑Ardakani1
Received: 27 June 2019 / Accepted: 23 August 2019
© Springer Nature B.V. 2019
Abstract
In this study, the transfer hydrogenation reduction of acetophenone using supported
nickel on MOF as catalyst was investigated. BBD and RSM were employed to inves-
tigate the efect of the experimental parameters such as Ni content, catalyst content,
temperature, and reaction time. Optimum reaction conditions for the formation of
1-phenylethanol were 45% mol of Ni content, 0.11 g of catalyst, reaction tempera-
ture at 83 °C, and reaction time 90 min. The catalyst was characterized by FESEM,
EDX, FT-IR, XRD, and H2-TPR techniques. The catalyst-free reaction phase was
analyzed by AAS, and fortunately, no leaching of nickel was detected.
Keywords Transfer hydrogenation · Ni@MOF-5 · Acetophenone · Experimental
design
Introduction
Metal–organic frameworks (MOFs) are formed by reticular synthesis [1–3], creating
strong bonds between inorganic and organic units [4]. These crystalline materials
have been obtained with signifcant properties such as ultrahigh porosity up to 90%
free volume, ultrahigh surface areas up to 10,400 m2/g, a large pore aperture (98 Å)
[5], high mechanical and thermal stability up to< 500 °C in an inert atmosphere and
>350 °C in oxidizing atmospheres, low densities down to 0.13 g/cm3, and chemical
stability in aqueous acid at pH<0 or in aqueous base at pH>14 [6]. These prop-
erties, along with the extraordinary degree of variability for both the organic and
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s1116
4-019-03959-1) contains supplementary material, which is available to authorized users.
* Soheil Sayyahi
sayyahi.soheil@gmail.com
1
Department of Chemistry, Mahshahr Branch, Islamic Azad University, Mahshahr, Iran
Vol.:(0123456789)
1 3

A. Nemati Chelavi et al.
inorganic components of their structures, have made academic and industrial sectors
to be interested in using MOFs for a variety of applications including gas adsorption
and storage (like CO2, H2, CH4), catalysis, chemical separations, biomedicine, light
harvesting and energy, as well as the degradation of toxic substances such as chemi-
cal warfare agents [7–10].
MOFs are considered as dream compounds for catalysis organic reactions and the
use of them allows for the combination of the advantages of heterogeneous catalysis
(easy post-reaction recovery separation, catalyst reusability, and high stability) and
homogeneous catalysis (high efciency, selectivity, controllability and mild reac-
tion conditions) [5, 11]. The tunability, versatility, high robustness under extreme
conditions, high surface areas, and high density of active sites of MOFs ofer many
advantages regarding their use in catalysis and placing them at the frontier between
zeolites and enzymes [12, 13].
In order to take advantage of catalytic behavior, that is, unique to MOFs, it might
be necessary to pay more attention to study the application of MOF catalysts in for-
mation reactions of more valuable organic molecules, which can take place under
relatively mild conditions [14].
Design and synthesis of MOF-5 (Zn4O(BDC)3, BDC2− =1,4-benzenedi-
carboxylate) was considered as a pioneering work in the chemistry of MOFs
[15]. The architectural robustness of MOF-5 allowed for gas adsorption with
Brunauer–Emmett–Teller (BET) surface area from 1010 to 4400 m2/g Langmuir
resulting from various preparations methods, with thermal stability of up to 400 °C
[16]. These values are substantially higher than those commonly found for zeolites
and activated carbon and make MOF-5 suitable candidate in the feld of catalysis
and gas storage [17].
As an attractive alternative to direct hydrogenation with H2 gas pressure, trans-
fer hydrogenation (TH) reaction referring to the addition of hydrogen to a molecule
from a non-H2 hydrogen source is a convenient and powerful method to access vari-
ous hydrogenated compounds. It is an attractive alternative to direct hydrogenation,
and it has recently been the interest of many researches in hydrogenation science
due to operational simplicity, mild reaction conditions, commercial availability, and
environmental friendliness [18]. In recent years, various transition-metal-based cata-
lysts (such as iron, ruthenium, osmium, cobalt, rhodium, iridium, nickel, palladium,
and gold) have been applied in both homogenous and heterogeneous systems of TH
reactions. In line with this and considering the development and replacement of the
more expensive and toxic noble metals, nickel in diferent forms is found active as
catalyst in the reduction of unsaturated compounds with TH methodology [19, 20].
Recently, Lu et al. and Chou et al., in their studies, investigated on the catalytic
activity of Ni@MOF-5 for CO2 methanation at low temperature and for hydro-
genation of crotonaldehyde, respectively [21, 22]. In the present chemistry study
that is in continuation of our efort to develop new transition-metal-based catalysts
[23–25], the results of the MOF-5 catalyzed TH reaction using supported nickel for
the reduction of acetophenone to 1-phenylethanol, as an industrial interest com-
pound was reported [19, 26, 27]. Also, experimental design was applied as a power-
ful tool to get maximum information with minimum number of experiments in order
to optimize reaction conditions [28–30].
1 3

Optimization of the transfer hydrogenation reaction of…
Results and discussion
In preliminary experiments, it has been shown that when the amounts of loaded Ni
(wt%) were more than 45%, the conversion of acetophenone decreased. This might
occur due to the surface segregation of Ni particles.
Experimental design
A Box–Behnken design RSM was employed to evaluate the efect of four experimental
factors including amount of nickel content (A), amount of catalyst (B), time of reaction
(C), and temperature (D). These factors were selected as efective experimental vari-
ables according to the preliminary studies. A set of 29 experiments was designed using
Design Expert software (version 11). Each factor was coded by (−1, 0, +1). The inde-
pendent factors, their notations, and levels are listed in Table 1.
Using the BBD–RSM experimental design, it was possible to consider a quadratic
polynomial regression model to predict the response based on the independent vari-
ables. Equation (1) shows the resulted quadratic model.
Yield = 13.07 + 3.35A + 4.87B + 1.91C + 1.24D
− 0.45AB + 2.17AC + 3.18AD + 1.20BC
(1)
− 3.34BD + 1.15CD − 1.43A2 − 4.68B2 − 6.67C2 − 2.38D2
R2 = 0.75; Ra2dj = 0.50; F = 2.98
where A, B, C, and D represent the amount of Ni content, amount of catalyst, reac-
tion time, and temperature, respectively. The product of the reaction was analyzed
by GC after each experiment, and the peak area of 1-phenylethanol was considered
as response in RSM. The regression sum of squares as a percentage of total sums of
squares was equal to 75%, revealing that most portion of variance in the response
was explained by the model. Analysis of variance of the quadratic model is pre-
sented in Table 2. Table 2 shows that there are some nonsignifcant terms in the
model that must be removed. Removing all of these terms resulted in the following
equation:
Yield = 10.99 + 3.35A + 4.87B − 4.08B2 − 6.06C2
(2)
R2 = 0.58; Ra2dj = 0.51; F = 8.21
The statistics of the model presented in Eq. (2) revealed that the model was not relia-
ble. In order to obtain an empirical equation with a better ability to predict the response,
higher order of interactions was considered. In the other words, the main efects along
with linear and nonlinear interaction terms of the efective factors were considered as
a pool of independent variables. To do so, stepwise method implemented in Design
Expert software was employed, and the following regression model was obtained.
1 3

A. Nemati Chelavi et al.
Table 1 Independent variables and their levels used in the Box–Behnken design
Factors
Level
Low
−1
Center
0
High
1
Ni content (mol%)
Catalyst content (g)
Temperature (°C)
Reaction time (min)
35
0
70
30
40
0.1
80
60
45
0.2
90
90
Run
A
B
C
D
Peak area of
1-phenylethanol
1
2
3
4
5
6
7
8
1
−1
0
−1
1
1
1
0
0
0
−1
1
0
−1
0
1
0
0
0
0
0
1
0
0
1
−1
0
1
0
−1
0
0
0
0
0
0
−1
0
0
−1
0
1
0
0
0
0
1
0
−1
0
−1
0
0
0
1
−1
0
0
0
0
0
0
0.04
11.54
8.53
3.70
7.94
0.07
6.91
0.00
19.20
11.45
0.05
11.87
5.88
2.63
0.04
14.84
12.65
6.06
9.74
14.77
0.04
14.55
0.03
0.02
2.38
0.14
0
−1
−1
1
0
0
0
−1
0
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
0
−1
−1
0
0
1
1
1
0
0
0
0
0
0
0
−1
0
−1
0
0
0
0
1
−1
1
1
0
0
−1
−1
1
0
−1
1 3

Optimization of the transfer hydrogenation reaction of…
Table 1 (continued)
Run
A
B
C
D
Peak area of
1-phenylethanol
27
28
29
0
1
0
0
0
−1
1
0
0
1
1
−1
7.11
24.95
0.01
Table 2 Analysis of variance of the quadratic model
Source
Sum of squares
df
Mean square
F value
p value
Model
A–A
B–B
C–C
D–D
AB
AC
AD
BC
BD
968.26
134.72
284.49
43.79
18.54
0.80
18.84
40.50
5.78
14
1
1
1
1
1
1
1
1
1
1
1
1
1
1
14
10
4
69.16
134.72
284.49
43.79
18.54
0.80
18.84
40.50
5.78
2.98
5.81
12.27
1.89
0.80
0.03
0.81
1.75
0.25
1.92
0.23
0.57
6.13
12.45
1.58
0.02
0.03
<1×10−4
0.19
Not signifcant
Not signifcant
Not signifcant
Not signifcant
Not signifcant
Not signifcant
Not signifcant
Not signifcant
Not signifcant
0.39
0.86
0.38
0.21
0.63
0.19
0.64
0.46
44.54
5.31
44.54
5.31
CD
A2
13.33
142.19
288.59
36.61
324.59
315.03
9.56
13.33
142.19
288.59
36.61
23.19
31.50
2.39
B2
0.03
C2
<1×10−4
0.23
D2
Not signifcant
Signifcant
Residual
Lack of ft
Pure error
13.18
0.01
Yield = +13.15 + 5.25A + 5.78B + 1.91C + 2.17AC + 3.18AD − 3.34BD − 3.15B2
− 7.47C2 − 3.17D2 + 5.84A2D − 5.70AB2 − 3.32B2D − 2.73BC2 − 4.65A2B2
R2 = 0.97; Ra2dj = 0.94; F = 33.96
(3)
The results of the squared correlation coefcient of the model and other statisti-
cal parameters presented in Eq. (3) showed that the resulting model was reliable
and can be used to quantitatively and accurately predict the response. Analysis of
variance of the model is also presented in Table 3. It was found that all terms were
signifcant with high F values and low P values. The F value of the model was equal
to 33.96 (compared to the critical value of 2.48 at the 0.05 level of signifcance)
with a P value of less than 3×10−8. In order to test whether the data set is normally
distributed, the normal probability plot of residuals is drawn and, and it is depicted
in Fig. 1. Figure 1 shows that the points fell close enough to the straight line and
1 3

A. Nemati Chelavi et al.
Table 3 Analysis of variance of the model shown in Eq. 3
Source
Sum of squares
df
Mean square
F value
P value
Model
A–A
B–B
C–C
AC
AD
1255.88
220.61
267.08
43.79
18.84
40.50
44.54
53.99
319.85
57.75
136.31
86.69
44.12
19.83
44.44
36.98
27.41
9.56
14
1
1
1
1
1
1
1
1
1
1
1
1
1
1
14
10
4
89.71
220.61
267.08
43.79
18.84
40.50
44.54
53.99
319.85
57.75
136.31
86.69
44.12
19.83
44.44
2.64
33.96
83.52
101.12
16.58
7.13
15.33
16.86
20.44
121.10
21.86
51.61
32.82
16.70
7.51
<1×10−4
<1×10−4
Signifcant
<1××10−4
0.001
0.018
0.002
0.001
BD
B2
0.001
C2
<1×10−4
0.001
D2
A2D
<1×10−4
<1×10−4
0.001
AB2
B2D
BC2
0.016
0.001
A2B2
Residual
Lack of ft
Pure error
16.82
2.74
2.39
1.15
0.49
Not signifcant
Fig. 1 Normal probability plot of residuals of the model presented in Eq. 3
1 3

Optimization of the transfer hydrogenation reaction of…
Fig. 2 Predicted values versus actual values for peak area of 1-phenylethanol of TH reaction of acetophe-
none
Fig. 3 Response surface plots a A–C; b A–D; c B–D; d A–B; and e B–C
passed through the center of the plot. As a result, the experimental points followed
a straight line suggesting normal distribution in the data. A plot related to the pre-
dicted response versus actual values is shown in Fig. 2. It is clear that the data points
1 3

A. Nemati Chelavi et al.
were well distributed close to a straight line with squared correlation coefcient of
0.971.
Figure 3 shows response surface plots, which are useful to evaluate interaction
between two independent variables while keeping the others’ value as constant.
Such three-dimensional plots provide valuable information about behavior of the
system. Figure 3a shows the interaction of Ni content of catalyst with reaction tem-
perature and their relation with peak area of 1-phenylethanol of the TH reaction. It is
well known that the temperature is regarded as a very important parameter in chemi-
cal reactions. Figure 3a shows that, in the reaction system, the maximum product
was formed at about 80 °C. In the other words, as temperature increases, the con-
version of the TH reaction increases, but at temperatures above 80 °C, the conver-
sion decreases due to the steady evaporation of acetophenone. The progress of the
TH reaction in the presence of the catalyst with diferent amounts of Ni content is
depicted in Fig. 3b. Figure 3b shows that the progress of the TH reaction in the pres-
ence of catalyst with lower amount of Ni content was not satisfactory. The yield of
the reaction increases as the reaction time increases, when the catalyst with higher
load of nickel is used. In the other words, in the presence of the catalyst with lower
Ni content than 35%, the reaction does not proceed considerably. Figure 3c shows
the interaction of amount of catalyst with reaction time and their relation with the
amount of 1-phenylethanol. It should be mentioned that the lower amount of catalyst
used in the experimental design was equal to 0.0 g, to check whether the reaction
proceeds in the absence of the catalyst. Figure 3c shows that the reaction did not
proceed considerably in the absence of the catalyst. The interaction of Ni content
with amount of catalyst and their relation with the response is shown in Fig. 3d. The
regression coefcients of two terms responsible to the interaction between factors
A and B (e.g., AB2, A2B2) were found to be signifcant and large compared to the
main efects of these factors. Some extent of curvature of the graph occurred due
to higher order of terms and interactions used in development of the model. Nev-
ertheless, Fig. 3d shows that the yield of the reaction was low in the absence of the
catalyst and/or in the presence of the catalyst with lower Ni contents. The interac-
tion between catalyst amount and reaction temperature is displayed in Fig. 3e. The
trend in this fgure was found to be similar to that in Fig. 3a. Optimal conditions to
get the highest amount of 1-phenylethanol were determined based on the investiga-
tion of the profle of the desirable response, and they are depicted in Fig. S1. The
desirability value corresponding to the maximum peak area of the product of the TH
reaction was equal to 0.997 with the peak area of 24.88. This high product yield was
obtained provided that experimental conditions were set as: 45% mol of Ni content,
0.11 g of catalyst, at the temperature of 83 °C, and at 90 min of reaction time. The
average peak area at this optimum condition was about 24.8%.
Characterization of catalyst
Characterization of the catalyst was done using various techniques including
FESEM, EDX, FT-IR, and XRD.
1 3

Optimization of the transfer hydrogenation reaction of…
Fig. 4 FESEM images and SEM–EDS elemental mapping images of C, Ni, O, and Zn in 45Ni@MOF-5
(up); EDX of 45Ni@MOF-5 (down)
1 3

A. Nemati Chelavi et al.
Fortunately, Fig. 4 shows the elemental mapping, indicating that nickel (Yellow)
has a good distribution over the surface of the catalyst, and the average size of the
particles is from about 22 to 33 nm with almost uniform shape. Besides, the EDS
spectrum shows signals clearly confrming the presence of Ni in addition to C, O,
and Zn.
The FT-IR spectra of guest-free MOF-5 and 45Ni@MOF-5 are shown in Fig. S2.
Figure S2 shows that almost all bonds shifted after the loading of MOF-5 with Ni.
For example, the strong band belonging to the C=O stretching vibration of carboxy-
late anions appeared at~1554 cm−1 and~1583 cm−1 for MOF-5 and 40Ni@MOF-5,
respectively. In the case of 40Ni@MOF-5, a strong band appeared at~1381 cm−1
was associated with the carboxyl group symmetric stretching, and bonds appeared
at~1016 cm−1,~825 cm−1, and~684 cm−1 were ascribed to the out of plane
vibrations of the BDC linker [31]. Also, the broadbands appeared at~3440 cm−1
to~3608 cm−1 were attributed to the presence of water in the metal coordination
sphere [32].
The analysis of XRD pattern revealed information about the crystallographic
structure, chemical composition, and physical properties of the catalyst [33]. The
XRD pattern of 45Ni@MOF-5 catalyst, MOF-5, and simulated pattern are pre-
sented in Fig. 5. Figure 5 shows the characteristic peaks in the 2θ region of 0–80°,
indicating that they were matched well compared to the corresponding standard and
similar reported pattern (JCPDS no. 87-0712) [21]. A very sharp peak at 2θ=8.99
indicated the high crystallinity of MOF-5 based on the catalyst. Additionally, peaks
Fig. 5 XRD pattern of 45Ni@MOF-5; MOF-5 (A) and simulated (B) patterns
1 3

Optimization of the transfer hydrogenation reaction of…
0
200
400
600
800
Temp.(oC)
Fig. 6 H2-TPR profle of 45Ni@MOF-5
at 2θ=44.52, 51.91, and 75.48 were assigned to metallic Ni with a face-centered
cubic structure.
The TPR technique is used to study the reducibility of supported nickel catalysts
[34–36]. A number of factors such as the amount of Ni loading, catalyst prepara-
tion protocol, and interaction between Ni ions and supports have proven to afect
the reducibility [37]. Figure 6 shows the H2-TPR profle for 45Ni@MOF-5 from
30 to 900 °C. It indicates that there are two hydrogen consumption peaks: a small
one at around 322 °C and a broad one at around 450 °C which can be attributed
to the reduction of Ni species at surface and within frameworks of the catalyst,
respectively.
Experimental section
General
All chemicals were purchased from commercial sources and were used right
after reception. Progress of reactions was followed by gas chromatographic (GC)
analyses and was performed using a DANI GC equipped with a fame ionization
detector (FID) and a column (length=150 m, inner diameter=0.25 mm, and flm
thickness=1.00 μm). The Fourier transform infrared (FT-IR) spectra (νmax) were
measured as a KBr disk using a BOMEM MB-Series 1998 FT-IR spectrometer and
at a scanning range of 4000–400 cm−1. Scanning electron microscopy (SEM) anal-
yses were performed on a Tescan Mira 3 LMU instrument coupled with energy-
dispersive X-ray (EDX) spectroscopy at a potential of 15 kV with 1-nm resolution
for microanalysis and mapping of the catalyst. X-ray powder difraction (XRD) was
applied for obtaining the XRD patterns of samples using Cu Kα radiation source
on a PANalytical X-ray difractometer Model X’Pert Pro. The H2 temperature-pro-
grammed reduction (H2-TPR) experiments were performed on a NanoSORD NS91
apparatus (Sensiran Co. Ltd., Tehran, Iran) equipped with a thermal conductivity
detector (TCD).
1 3

A. Nemati Chelavi et al.
Synthesis of the Ni@MOF‑5
MOF-5 was synthesized using an optimized procedure starting with the use of
terephthalic acid, zinc nitrate, and diethylformamide as organic solvents [9]. Ni@
MOF-5 catalysts were prepared by a modifed procedure of impregnation methods
according to the previous studies [21]. A mixture of Ni (acac) 2 (1.71 g, 6.6 mmol)
was magnetically stirred in ethanol (250 mL) and was heated at 60 °C. MOF-5
(1.8 g) was added to the obtained dark green solution, and the resulted mixture was
heated for 16 h. Then, hydrazine solution (5 mL, 80 wt%) was added to the mixture
and was heated for another 6 h. The ensuing mixture was fltered, was washed with
water (5×10 mL), then was dried at 110 °C for 12 h, and was calcined at 250 °C
under an air atmosphere for 5 h to obtain the fnal product.
Catalytic study
The activities of the catalysts with diferent amounts of Ni content were measured
based on the results obtained from the transfer hydrogenation of acetophenone. All
experiments were carried out in a Tefon-lined autoclave. Typically, a mixture of
acetophenone (0.12 g, 1 mmol), 2-propanol (4 mL), and specifed amount of the
catalyst were transferred into the Tefon-lined autoclave and were treated for pre-
defned reaction time at specifed temperature. Then, the catalyst was fltered of,
and the progress of the reactions was analyzed by gas chromatograph. The type and
amount of the catalyst, time, and temperature of reaction were considered as those
experimental factors infuencing the reaction conversion of acetophenone to 1-phe-
nylethanol, and, therefore, they were studied by experimental design in details.
The reusability and leaching of the catalyst is a great concern for most of the
supported catalyst. In this regard, the catalyst was recovered, was washed with etha-
nol for several times, then was dried at 110 °C in an oven, and then it was reused.
Fortunately, the catalyst did not show signifcant loss in activity after fve runs. The
catalyst-free reaction phase was also analyzed by atomic absorption spectroscopy,
and the amount of <2 ppm of nickel was found.
Experimental design and optimization
In order to investigate the efect of the experimental parameters on the transfer
hydrogenation reaction yield for acetophenone, Box–Behnken experimental design
(BBD) and response surface methodology (RSM) were employed. Compared to the
one-factor-at-a-time method, the use of RSM design reduces the number of experi-
ments required to determine the efect of understudied factors.
Box–Behnken design (BBD) is a spherical, rotatable, or nearly rotatable quad-
ratic RSM design. It is done based on a three-level fractional factorial design includ-
ing the central point and middle points of the edges. The number of experimental
runs (N) was calculated by the following formula: N = 2k(k − 1) + Cp, where k
represents the number of variables and Cp represents the number of center points
[38]. The factors selected for the optimization of reaction yield included amount of
loaded nickel, amount of catalyst, time of reaction, and temperature. Peak area of the
1 3

Optimization of the transfer hydrogenation reaction of…
product, e.g., 1-phenylethanol, was considered as the response or dependent variable
to be measured in relation to the conversion in the reaction. BBD experiments were
designed using Design Expert software (version 11), and experiments were per-
formed in randomized order. The central point was replicated fve times to obtain a
good estimation of random errors. Empirical equation relating response to the inde-
pendent variables was generated in Design Expert environment. The same software
was used to extract response surface plots and to predict optimum conditions. The
actual and coded values of each factor are presented in Table 2. A preliminary study
was carried out to select the levels of each factor. The lower amount of the catalyst
was equal to 0.0 g, which was used to monitor the progress of the reaction in the
absence of the catalyst. The upper limit of catalyst amount, e.g., 0.2 g, was deter-
mined according to the preliminary study. Results showed that higher amount of the
catalyst caused a signifcant decrease in the formation of the product in the TH reac-
tion of acetophenone, which may occur due to the catalyst agglomeration. Limits
of Ni contents were assigned to be between 35 and 45 wt%, respectively, since the
preliminary study indicated that insignifcant amount of conversion was obtained
outside of this range.
Conclusion
In this study, an extensive experimental design and surface response method were
used to study the efective parameters such as Ni content, catalyst amount, reaction
temperature, and time on the synthesis of 1-phenylethanol catalyzed by Ni-MOF
nanoparticles through TH reduction. Also, the interaction of variables was evaluated
with the least number of possible tests. According to the statistical analysis, the best
obtained results belonged to the catalyst loaded with 45% of Ni.
Acknowledgements We gratefully acknowledge the funding support received for this project from the
Mahshahr Branch, Islamic Azad University, Iran.
References
1. O.M. Yaghi, M. O’Keefe, N.W. Ockwig, H.K. Chae, M. Eddaoudi, J. Kim, Nature 423, 705 (2003)
2. J.L.C. Rowsell, O.M. Yaghi, Microporous Mesoporous Mater. 73, 3 (2004)
3. Y. Liu, M. O’Keefe, M.M.J. Treacy, O.M. Yaghi, Chem. Soc. Rev. 47, 4642 (2018)
4. H. Furukawa, K.E. Cordova, M. O’Keefe, O.M. Yaghi, Science 341, 1230444 (2013)
5. A.H. Chughtai, N. Ahmad, H.A. Younus, A. Laypkov, F. Verpoort, Chem. Soc. Rev. 44, 6804
(2015)
6. A.J. Howarth, Y. Liu, P. Li, Z. Li, T.C. Wang, J.T. Hupp, O.K. Farha, Nat. Rev. Mater. 1, 15018
(2016)
7. H.-C. Zhou, J.R. Long, O.M. Yaghi, Chem. Rev. 112, 673 (2012)
8. P. Silva, S.M.F. Vilela, J.P.C. Tomé, F.A.A. Paz, Chem. Soc. Rev. 44, 6774 (2015)
9. U. Mueller, M. Schubert, F. Teich, H. Puetter, K. Schierle-Arndt, J. Pastré, J. Mater. Chem. 16, 626
(2006)
10. D. Wang, Z. Li, Res. Chem. Intermed. 43, 5169 (2017)
1 3

A. Nemati Chelavi et al.
11. D. Burrows Andrew, K. Cadman Laura, J. Gee William, H. Amer Hamzah, V. Knichal Jane, and S.
Rochat, Tuning the Properties of Metal–Organic Frameworks by Post-synthetic Modifcation, ed. by
H. García and S. Navalón. Metal–organic frameworks. Wiley (2018)
12. F. David, A. Sonia, P. Catherine, Angew. Chem. Int. Ed. 48, 7502 (2009)
13. J. Lee, O.K. Farha, J. Roberts, K.A. Scheidt, S.T. Nguyen, J.T. Hupp, Chem. Soc. Rev. 38, 1450
(2009)
14. J. Liu, L. Chen, H. Cui, J. Zhang, L. Zhang, C.-Y. Su, Chem. Soc. Rev. 43, 6011 (2014)
15. H. Li, M. Eddaoudi, M. O’Keefe, O.M. Yaghi, Nature 402, 276 (1999)
16. S.S. Kaye, A. Dailly, O.M. Yaghi, J.R. Long, J. Am. Chem. Soc. 129, 14176 (2007)
17. T. Segakweng, N.M. Musyoka, J. Ren, P. Crouse, H.W. Langmi, Res. Chem. Intermed. 42, 4951
(2016)
18. D. Wang, D. Astruc, Chem. Rev. 115, 6621 (2015)
19. K. Vijayakrishna, K.T.P. Charan, K. Manojkumar, S. Venkatesh, N. Pothanagandhi, A. Sivaram-
akrishna, P. Mayuri, A.S. Kumar, B. Sreedhar, ChemCatChem 8, 1139 (2016)
20. F. Alonso, P. Riente, M. Yus, Acc. Chem. Res. 44, 379 (2011)
21. W. Zhen, B. Li, G. Lu, J. Ma, Chem. Commun. 51, 6556 (2015)
22. H. Zhao, H. Song, L. Chou, Inorg. Chem. Commun. 15, 261 (2012)
23. F.K. Olia, S. Sayyahi, N. Taheri, C. R. Chim. 20, 370 (2017)
24. S. Sayyahi, S. Mozafari, S.J. Saghanezhad, Res. Chem. Intermed. 42, 511 (2016)
25. A. Shouli, S. Menati, S. Sayyahi, C. R. Chim. 20, 765 (2017)
26. R. Molinari, C. Lavorato, P. Argurio, Chem. Eng. J. 274, 307 (2015)
27. B. Zhang, F. Xie, J. Yuan, L. Wang, B. Deng, Catal. Commun. 92, 46 (2017)
28. C.A. McNamara, F. King, M. Bradley, Tetrahedron Lett. 45, 8239 (2004)
29. H. Naeimi, V. Nejadshafee, S. Masoum, Appl. Organomet. Chem. 29, 314 (2015)
30. M. Ghiassee, M. Rezaei, F. Meshkani, S. Mobini, Res. Chem. Intermed. 45, 4501 (2019)
31. M. Zhang, J. Guan, B. Zhang, D. Su, C.T. Williams, C. Liang, Catal. Lett. 142, 313 (2012)
32. N.T.S. Phan, K.K.A. Le, T.D. Phan, Appl. Catal. A 382, 246 (2010)
33. R.K. Sharma, Y. Monga, A. Puri, J. Mol. Catal. A Chem. 393, 84 (2014)
34. Y. Jiang, T. Huang, Y. Xu, X. Li, Z. Qin, H. Ji, Chem. Eng. Technol. 41, 175 (2018)
35. T.A. Le, M.S. Kim, S.H. Lee, T.W. Kim, E.D. Park, Catal. Today 293–294, 89 (2017)
36. D.V. Peron, V.L. Zholobenko, M.R. de la Rocha, M. Oberson de Souza, L.A. Feris, N.R. Marcilio,
V.V. Ordomsky, A.Y. Khodakov, J. Mater. Sci. 54, 5399 (2019)
37. A. Parmaliana, F. Arena, F. Frusteri, N. Giordano, J. Chem. Soc. Faraday Trans. 86, 2663 (1990)
38. N. Chamkouri, A. Niazi, V. Zare-Shahabadi, Spectrochim Acta A Mol Biomol Spectrosc 156, 105
(2016)
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional afliations.
1 3