Alkali-modified heterogeneous Pd-catalyzed synthesis of acids, amides and esters from aryl halides using formic acid as the CO precursor

Article abstract of DOI:10.1039/d1ra05177f

To establish an environmentally friendly green chemical process, we minimized and resolved a significant proportion of waste and hazards associated with conventional organic acids and molecular gases, such as carbon monoxide (CO). Herein, we report a facile and milder reaction procedure, using low temperatures/pressures and shorter reaction time for the carboxyl- and carbonylation of diverse arrays of aryl halides over a newly developed cationic Lewis-acid promoted Pd/Co₃O₄catalyst. Furthermore, the reaction proceeded in the absence of acid co-catalysts, and anhydrides for CO release. Catalyst reusability was achievedviascalable, safer, and practical reactions that provided moderate to high yields, paving the way for developing a novel environmentally benign method for synthesizing carboxylic acids, amides, and esters.

Full text of DOI:10.1039/d1ra05177f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Alkali-modified heterogeneous Pd-catalyzed
synthesis of acids, amides and esters from aryl
Cite this: RSC Adv., 2021, 11, 26937
halides using formic acid as the CO precursor†
Charles O. Oseghale, Oluwatayo Racheal Onisuru, Dele Peter Fapojuwo,
Batsile M. Mogudi, Pule Petrus Molokoane, Nomathamsanqa Prudence Maqunga
and Reinout Meijboom
*
To establish an environmentally friendly green chemical process, we minimized and resolved a significant
proportion of waste and hazards associated with conventional organic acids and molecular gases, such as
carbon monoxide (CO). Herein, we report a facile and milder reaction procedure, using low temperatures/
pressures and shorter reaction time for the carboxyl- and carbonylation of diverse arrays of aryl halides over
Received 5th July 2021
Accepted 26th July 2021
3 4
a newly developed cationic Lewis-acid promoted Pd/Co O catalyst. Furthermore, the reaction proceeded
in the absence of acid co-catalysts, and anhydrides for CO release. Catalyst reusability was achieved via
scalable, safer, and practical reactions that provided moderate to high yields, paving the way for
developing a novel environmentally benign method for synthesizing carboxylic acids, amides, and esters.
DOI: 10.1039/d1ra05177f
rsc.li/rsc-advances
2
and H O remains a common issue with the effective use of
formic acid as a surrogate.
Introduction
Carbonylation reactions involving transition metals have
2 2
The decomposition of formic acid to H and CO has been
proven to be the most convenient and versatile method for successfully achieved using a variety of homogeneous cata-
22,23
synthesizing carbonyl-containing compounds such as anhy- lysts,
whereas the decomposition of formic acid to CO and
drides, esters, amides, aldehydes, alcohols, ketones, and
H
O is currently being investigated owing to the difficulty
2
1–6
acids, which have found applications in a variety of biologi- encountered in its transformation at lower temperatures.
cally active compounds, pharmaceuticals, and natural products. Although formic acid dehydration in the presence of strong
As a result, the carbonylation process has become an indis- mineral acids, such as sulfuric acid, may produce CO gas
pensable building block and a highly effective technique for preferentially. The acid is thought to act as an activator for the
synthesizing ne and bulk chemicals in academia and industry. CO release but has compatibility issues when applied to other
However, carbonylation reactions have not been extensively carbonylation reactions owing to the process's harsh reaction
7
19,24,25
used in a laboratory setting for chemical synthesis. This is due conditions. Similarly, acetic anhydride,
triethylamine,
to the handling of extremely ammable and toxic CO gas, which and others have been reportedly employed as activators for
works under higher pressure and requires sophisticated selective CO release. However, acetic anhydride requires a stoi-
stainless-steel autoclave reactors. To address this issue on chiometric amount to promote CO production, which generates
a laboratory scale, much emphasis has been directed towards waste and complicates the work. As a result, the quest for more
7
,8–10
the use of CO surrogates.
Numerous surrogates, including environmentally friendly methods to perform these reactions in
11
7,12–15
16,17
aldehydes, formic acids,
used in carbonylation processes catalyzed by transition metals.
Among these CO substitutes, formic acid (HCOOH) is the most perform carbonylation reactions under these conditions is
practical surrogate since it can be readily obtained via CO
highly desirable. Although homogeneous catalysis remains vital
However, the in academia and chemical industrial processes, it is estimated
ambiguous breakdown route of HCOOH to H and CO or CO to account for more than 20% of catalytic reactions in the
and formates,
have been situ in solution persists.
Developing a stable and reusable heterogeneous catalyst to
2
18,19
20,21
hydrogenation
and bio-waste fermentation.
2
2
26–28
chemical industry today.
Carbonylation reactions catalyzed
by transition metals have developed into a critical tool in
organic synthesis. These reactions have been carried out using
a variety of catalytically active metals. The palladium metal-
catalyzed addition of CO to alkynes and alkenes in the pres-
ence of a suitable nucleophile has garnered signicant interest
in recent years for its potential use in synthesizing a wide variety
Research Center for Synthesis and Catalysis, Department of Chemical Sciences,
University of Johannesburg, PO Box 524, Auckland Park 2006, Johannesburg, South
Africa. E-mail: rmeijboom@uj.ac.za; Fax: +27 11 559 2819; Tel: +27 72 894 0293
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/d1ra05177f
1
©
2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 26937–26948 | 26937

View Article Online
RSC Advances
Paper
of essential chemical products. Palladium-supported phos- palladium species on reducible Co
3
O
4
, thus minimizing active
phine functionalized porous polymers and various palladium– site leaching, acting as a potential “in-built” cationic Lewis acid
29–31
phosphine complexes have been extensively described.
However, these synthetic protocols oen require the use of non-
co-catalyst, and improving the catalyst's binding properties.
Furthermore, between the generated CO molecules, the
recyclable palladium precursors, a longer reaction time, and newly designed catalyst, and the substrates or products that
phosphine ligands, as well as harsh reaction conditions. In may limit the substrate's scope, reaction design must account
addition, these protocols sometimes intricate the carbonylation for the undesirable chemistry associated with these reactions.
of the unprotected nitrogen in the phosphine ligands. Moisture- As a result, performing carbonylation reactions on a laboratory
sensitive phosphine ligands and non-recyclable palladium scale with various substrates has thus far remained elusive. The
precursors pose signicant challenges in terms of catalyst study reports a highly reusable catalyst and an operationally
recovery, structural exibility, application, accessibility, leach- simple strategy for the stoichiometric carbonylation reaction
32,33
ing, and reusability.
This necessitates the development of under milder conditions (Scheme 1). Notably, this new system
highly reusable, economically viable, and environmentally exhibited superior activity for the reactions in the absence of
benign carbonylation catalysts. Recently, researchers have acid co-catalysts and anhydrides (activators). The critical
developed a method for immobilizing clusters or metallic success may be attributed to CO generated in situ in the solution
compounds (palladium complexes) on solid supports such as containing the ion-promoted Pd-catalyst. We anticipated that
34,35
,36
37,38
39
organic polymers,
Fe
3
O
4
SBA-15,
MOF-5, ZIF-8 (ref. this catalytic system would improve the rates of carboxy- and
40) that is sustainable, green, easy, and cost-effective. The het- carbonylation reactions by utilizing formic acid as a CO
erogenization of homogeneous catalysts has been extensively precursor. Additionally, we demonstrated that this green
researched recently to address sustainability concerns about method might be used to effectively synthesize many biologi-
recyclability, besides designing new ligands to enrich the cally active amide compounds. Notably, this facile protocol
homogeneous catalysts.
serves as a critical tool for drug development programs and
As part of our ongoing work on heterogeneous catalyzed discovery.
carbonylation reactions, we became motivated to ll the gap by
designing a new palladium heterogeneous catalyst for these
Experimental section
Chemicals
reactions. The catalytic system utilizes the synergistic combi-
nation of alkali metals as ion-promoters on the palladium
supported on Co O . The spinel mesoporous Co O is versatile
3
4
3
4
Apart from the calcium (98%) and potassium (>99%) nitrate
salts purchased from Associated Chemical Enterprises, all other
chemicals used in this work were of analytical grade and
purchased from Sigma-Aldrich. The reactions were performed
using an autoclave and a drilled aluminum block with securely
capped 20 mL dram-vials at different times. In addition, high
purity solvents from commercial suppliers were used without
the need for further purication.
and constitutes an interesting class of nanomaterials for noble
metal support in vast catalytic reactions due to the intrinsic
41–44
catalytic activity they exhibit.
Notably, incorporating
dopants such as alkali metals into catalysts has developed into
an interesting and active eld in catalysis. We recently reported
that alkali metals might induce a phenomenon and improve the
2,41,45
overall catalyst activity.
Various studies have documented
the increased catalytic activity of these alkali-promoted catalysts
due to their electronic and structural effects on these nano-
Synthesis of the catalysts
46–48
materials.
It was discovered in this study that doping with
The deposition–precipitation synthetic technique was used to
synthesize undoped Pd/Co O and alkali metal (x ¼ Cs, Na, Li,
alkali metals stabilized and induced the active dispersed
3
4
3 4
Scheme 1 Alkali-cation promoted Pd/Co O -catalyzed carboxy- and carbonylation of aryl halides.
26938 | RSC Adv., 2021, 11, 26937–26948
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
Ca and K) doped Pd/Co
3
O
4
catalysts with the same alkali metal/ sites on the catalyst. Approximately (0.20–0.30 g) was loaded
49
Pd mole ratio (2% mol/mol). Typically, 1 g of the as-prepared into a tube reactor containing quartz wool and further probed
ꢀ
Co O (ref. 2) was added to an aqueous Pd(OAc) solution and by passing 10% NH at a temperature range of 20–700 C using
3
4
2
3
stirred at room temperature (RT) for 2 h. Urea : Pd salt at helium as the probed gas. The ow and heating rates were
ꢁ
1
ꢀ
ꢁ1
a (2 : 1) molar ratio was added to the reaction mixture with maintained at 25 mL min and 3 C min for both experi-
ꢀ
heating and stirring for 1 h at 80 C. Aer cooling to room ments, respectively. The morphology of the catalyst was
4
temperature, freshly prepared NaBH was added to the reaction captured using a Transmission Electron Microscope (TEM)
mixture at a 10 : 1 molar excess and agitated for a further 4 h at (JEOL Jem-2100F) operated at a 200 kV accelerating voltage. A
RT. The alkali metal was then added to the mixture and stirred total of 0.2 mg of the catalyst was measured and ultrasonicated
for 1 h. The precipitate obtained was washed with 100 mL of in 1 mL ethanol for 30 min before being deposited on the
ꢀ
excess distilled water, dried, and calcined for 3 h at 350 C. Prior carbon-coated TEM-Cu grids.
to conducting the activity tests, the catalysts were reduced in H
2
ꢀ
for 20 min at 350 C and labeled as Co
3
O
4
(X1) and 3% Pd/Co
or Pd/
–Na (X4), 3% Pd/
–Cs (X6), 3% Pd/Co –Li (X*), 1%
–Li (X7) and 5% Pd/Co –Li (X8). The % mole of the
3 4
O
Catalytic test for carbonylation of the aryl halides
3 4
(X2). The doped equivalents were denoted as x–Pd/Co O
Typically, the reaction tube equipped with a stirring bar was
Co
Co
3
O
O
4
4
–x; 3% Pd/Co
–K (X5), 3% Pd/Co
3 4 3 4
O –Ca (X3), 3% Pd/Co O
ꢀ
charged with a 0.3 g catalyst (previously reduced at 350 C for
3
3
O
4
3 4
O
2
1
3
2
0 min in H ), 4 mmol substrate, 3 mL solvent, 1 eq. Et N B1,
2 3
mmol (200 L) decane (internal standard), 3 mL HCOOH, and
Pd/Co
3
O
4
3 4
O
alkali metal cations was maintained at 2 wt%.
mL nucleophile (for alkoxy- and aminocarbonylation). For 8–
ꢀ
4 hours, the mixture was heated to 90–130 C at a stirring
Catalyst characterization
speed of 550 rpm. Upon cooling to RT, 50 mL of water was
added to separate the organic phase using a separating funnel.
Ethyl acetate was used to extract the aqueous layer (3 ꢂ 25 mL).
The combined layer was washed with a 50 mL brine solution
and then dried over magnesium sulfate. The residual solvent
was evaporated using a rotary evaporator, and the product was
puried by ash column chromatography on silica gel with
hexane/ethyl acetate eluents at various intervals. The products
were quantied using a Shimadzu GC-2010 equipped with
Various characterization techniques, including ICP-OES, FT-IR,
p-XRD, TEM, SEM, TPR, and TPD, were used to characterize the
as-prepared nanomaterials. The metal loading (wt%) capacity of
the catalyst was determined using the Spectro Arcos ICP-OES
instrument. A thermogravimetric analyzer (PerkinElmer STA
6000) was used to perform the thermogravimetric analyses
ꢀ
(TGA) under air atmosphere (20–1000 C) at a ow and heating
ꢁ
1
ꢀ
ꢁ1
ramping rates of 20 mL min and 10 C min , respectively.
The surface morphologies of the synthesized nanomaterials
were studied using a Vega 3 Tescan LMH Scanning Electron
Microscope (SEM) equipped with EDX and operating at a high
2
a ame ionization detector (FID) and N as the carrier gas; the
FID and injection temperatures were adjusted to 370 and
ꢀ
2
00 C, respectively. The products were conrmed and
1
13
1
13
compared using GC-MS, H, and C NMR instruments. Simple
ltration, followed by washing and drying in a vacuum, was
voltage of 20 kV. The H NMR (500 MHz) and C NMR (125
MHz) spectra were acquired using a Bruker-500 MHz spec-
trometer, with values given relative to tetramethylsilane (0.0) as

used to test the catalyst's reusability.
2
the internal standard. The N -sorption studies were conducted
using a Micromeritics ASAP 2460 device. Prior to the experi-
ments, the samples were degassed at 90 C under N2 atmo-
Results and discussions
ꢀ
sphere and vacuum for 14 h and 2 h, respectively, to remove In this study, the newly designed alkali-promoted Pd catalyst
physisorbed moisture from the nanomaterials. The surface area was employed as an efficient catalyst for the carboxy- and
was calculated using the Brunauer–Emmett–Teller (BET) tech- carbonylation reactions of diverse arrays of aryl halides. Under
nique. Powder X-ray diffraction (p-XRD) analyses were per- milder reaction conditions and utilizing formic acid as a CO
formed using a Rigaku MiniFlex-600 instrument operating with surrogate source, the carbonylation processes were carried out
ꢀ
ꢀ
Cu K radiation (¼ 1.5406). Both low (0.5–4.0 ) and wide (20–70 ) with excellent conversions, selectivities, and yields to their
2
q angles diffraction patterns were measured at a step rate of respective desired products. Most importantly, eliminating the
.1 min . Fourier transform infrared (FT-IR) spectroscopy of use of acid co-catalyst and sacricial acetic anhydrides needed
ꢀ
ꢁ1
0
the materials was performed using a Bruker FT-IR Alpha spec- for the CO release. In general, this type of carbonylation process
ꢁ
1
trometer (KBr pellets in 4000 to 500 cm ). On a Micromeritics is difficult to perform without external additives (organic acids
AutoChem II chemisorption analyzer, the hydrogen and anhydrides) since these additives are oen required to
7,24
2
temperature-programmed reduction (H -TPR) was conducted. obtain reasonable reactivities. Similarly, homogeneous cata-
About 25 mg of the catalyst was loaded into a quartz tube lytic systems have been extensively used to carry out these
ꢀ
reactor and pretreated for 1 h at 200 C under helium/argon reactions, but with catalyst recovery and reusability problems.

ow. Aer cooling, the materials were probed by passing 10% As a result, the design of a more versatile heterogeneous catalyst
ꢁ
1
H (Ar) at a ow rate of 50 mL min from ambient temperature capable of performing these reactions with excellent product
2
ꢀ
ꢀ
ꢁ1
to 600
programmed desorption (NH
the Micromeritics AutoChem II equipment to identify the acidic (Cs, Na, K, Li, and Ca) for the carboxy- and carbonylation of aryl
C
(10
C
min ). The ammonia temperature- yields is highly desired. In this work, we report the rst
3
-TPD) tests were performed on heterogeneous Pd-based catalyst, promoted by alkali metal ions
©
2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 26937–26948 | 26939

View Article Online
RSC Advances
Paper
Fig. 1 (a) Wide-angle X-ray diffractograms of A (Co
O
3 4
3 4 3 4 3 4 3 4
), B (3% Pd/Co O ), C (1% Pd/Co O –Li), D (3% Pd/Co O –Li), and E (5% Pd/Co O –Li)
3 4
and (b) low-angle p-XRD patterns of the nanomaterials. A – Co O -phase (PDF no. 73-1701) and ) – PdO-phase (PDF no. 85-0713).
halides. Although the PdNPs were the active species for the Similarly, neither the alkali cations nor the PdNP loadings
reaction, however, we found out that the alkali cations in the distorted the spinel Co
catalysts had a remarkable promotional effect on the catalytic angle diffraction patterns (20–70 ) consistent with the spinel
3
O
4
phases. All catalysts exhibited wide-
ꢀ
3 4
system by acting as a promoter and an in built cationic Lewis Co O (PDF no. 73-1701) indexed in the JCPDS database; this
acid co-catalyst. Additionally, the activity was linked with the certies the absence of other bulk crystalline phases in the
NH -TPD analysis of each catalyst, showing that the lithium- nanomaterials. The nanomaterials exhibit low-angle diffraction
3
ꢀ
promoted catalyst with the highest amount of acidic sites was lines ranging from 0.5 to 4.0 , indicating the existence of
the most active for the reactions (Fig. 4). With the results, we a regular ordered mesoporous structure (Fig. 1b). As shown in
concluded that there was a strong bonding interaction between Table 1, the average size of the nanoparticles grew larger as the
the alkali cations, palladium nanoparticles, and the meso- Pd loading on the reducible cobalt oxide support was increased.
porous Co
3
O
4
support.
The positions of the low-angle lines and their corresponding
shi to smaller 2q values are well correlated with the particle's
calculated average size. Notably, a portion of the PdNP was
oxidized to PdO as observed from the spectra, which displayed
weak diffraction peaks at 33.7 , with the broadest peak assigned
to the PdO (101) facet; these peaks were detected for catalysts
Structural characterization of the catalysts
ꢀ
Powder X-ray diffraction (p-XRD). The XRD analysis was used
to conrm that the palladium nanoparticles (PdNPs) were
effectively immobilized on the support (Co O ), and the
3% Pd/Co
3
O
4
–Li (D) and 5% Pd/Co
3
O
4
–Li (E) but were not as
3
4
prominent for the other catalysts.
support's structure was not altered during the synthesis (Fig. 1).
Table 1 Structural characterization of the nanomaterials
a
2
ꢁ1
3
ꢁ1
b
Catalysts
Co
SABET (m g
)
Pore volume (cm g
)
Pore diameter (nm)
Crystallite sizes (nm)
3
O
4
63
58
51
43
29
0.357
0.392
0.378
0.371
0.244
12.8
19.2
16.8
15.7
10.6
7.4
11.6
14.2
16.8
22.4
3% Pd/Co
1% Pd/Co
3% Pd/Co
5% Pd/Co
3
O
3
O
3
O
3
O
4
4
4
4
–Li
–Li
–Li
a
b
2
Determined by N BET. Calculated using Scherrer's equation.
26940 | RSC Adv., 2021, 11, 26937–26948
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
Fig. 2 (a) and (b) SEM images of Co
O
3 4
and 5% Pd/Co
3
4
O –Li, respectively. TEM images and particle size distribution chart (inset) for (c) 1% Pd/
3 4 3 4 3 4
Co O –Li, (d) 5% Pd/Co O –Li and (e) HR-TEM image magnification of 5% Pd/Co O –Li (inset). The nanoparticles are indicated in green circles.
Additionally, the PdO particles grew in size with increased spectra. The XRD analysis revealed that the crystallinity of
intensities for the 3 and 5% Pd loadings. However, with a lower Co O was preserved, suggesting that the PdNPs and alkali
3
4
Pd loading (1 wt%), no PdO peaks were seen, indicating that the cations were successfully loaded onto its matrix.
particles were too tiny to be detected by the XRD machine. The Scanning/transmission electron microscopy (SEM/TEM). As
peaks corresponding to the Pd metal were not very visible in the revealed by the SEM images in Fig. 2a and b, the promoted
Fig. 3 (a) Nitrogen adsorption–desorption curves and (b) BJH pore size distributions for the catalysts.
©
2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 26937–26948 | 26941

View Article Online
RSC Advances
Paper
catalyst exhibits an almost identical ball-like nano-sized shape presented in Fig. 4. Compared to the other catalysts, the Li-
to Co O , suggesting that the alkali metal and PdNPs did not promoted catalyst showed a more prominent and robust
3
4
ꢁ1
ꢀ
alter the support's phases. The acquired TEM micrographs acidity site (0.4160 mmol g ) with a shoulder peak at 438 C.
show that the size of the PdNPs varies with the amount of Pd This phenomenon may be ascribed to the highly electropositive
loaded on the reducible support (Fig. 2c and d). Irrespective of alkali cations in the catalyst, which increased the acid strength
the Pd loading, the spinel Co
3 4
O particles display a similar size and enhanced the catalytic activity for the reactions. The
around 25–40 nm. A low population of the PdNPs on Co
3
O
4
was unpromoted Pd/Co
3
O
4
catalyst exhibited the least acidic sites of
ꢀ
detected for the 1% loading and grew larger for the 5% loading; all the catalysts analyzed, with a peak around 352 C. It is
this directly correlates with and is consistent with the XRD data. important to note that the increase in acidity was found to
The 0.215 nm lattice fringes found for 5% Pd/Co O –Li, can be correlate with the alkali metal ions' cationic Lewis-acid strength
3
4
50,51
indexed to the {110} planes of PdO (Fig. 2e).
Nitrogen sorption studies. To study how the Pd size varied
(Li > Ca > Na > K > Cs).
Following that, we conducted the H -TPR measurements to
2
with the loadings, the surface area, pore size, and pore volume determine the nanomaterials' reducibility (Fig. 5). Two reduc-
ꢀ
ꢀ
3+
2+
were determined using BET analysis. As shown in Fig. 3a, the tion peaks at 236 C and 264 C, assigned to Co / Co and
2
+
sorption curves of the nanomaterials exhibited a type IV Co / Co, respectively, were observed for the mesoporous
isotherm with minor distortions and a relative pressure range of Co . The addition of palladium and an alkali cation to Co
.76–0.92, conrming the material's mesoporous structure. It is facilitated the catalyst's reduction behavior, consistent with
3
O
4
3 4
O
0
5
8–60
noteworthy that increasing the Pd loading decreased the surface previous reports.
The Pd loadings substantially inuenced
area (SA), pore-volume, and pore diameter, with the least the shi to lower reduction temperatures; these peaks at lower
observed for the 5% Pd loading, suggesting that Pd loadings temperatures were attributed to PdO reductions, and the peak
had a signicant effect on the pore structures and catalyst's SA. positions are highly dependent on the PdO dispersion on the
61
The decrease in SA, pore volume/diameter, was most probably support. Although the low-temperature peaks may not be
caused by PdNPs occupying a portion of the Co mesopores. entirely due to PdO reduction, because catalysts have been
3 4
O
3
+
2+
In many catalytic reactions, catalytic performance is oen shown to display a reduction of Co / Co at temperatures as
ascribed to the physiochemical properties of the catalyst, such low as 180–200 C, apparently due to the H spillover effect.
2
ꢀ
61–63
52–57
ꢀ
as surface areas, pore sizes/volumes.
It is critical to highlight However, the lower reduction peak in the range of 151–196 C
that these features were not overt or considered a determining for the 3% X* and 5% X8 catalysts may be attributed to PdO
factor in the moderate to high catalytic activity reported for the reduction, as this was also evident from the XRD spectra. It is
nanomaterials in this study. The textural properties of nano- worth noting that the trend toward increasing catalytic activity
particles are summarized in Table 1.
Temperature-programmed measurements. The ammonia reduction temperatures.
temperature-programmed desorption (NH -TPD) measurement Thermogravimetric (TGA) and Fourier-transform infrared
of the catalyst may also be linked to the nanomaterial's low
3
was used to determine the acidic sites of promoted and (FT-IR) spectrometry. The thermographic analysis curves for the
unpromoted catalysts, and their representative spectra/total nanomaterials are depicted in Fig. S2 and S3.† This analysis was
acidity amount (mmol of NH3 per gram of catalyst) are used to validate the thermal stability of the as-prepared catalysts
ꢀ
over a temperature range of 0–1000 C. The spectra revealed
3
Fig. 4 NH -TPD profiles of the catalysts. All Pd loadings were set at
3
wt%.
2
Fig. 5 H -TPR profiles of the catalysts.
26942 | RSC Adv., 2021, 11, 26937–26948
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
promoted Pd/Co
3
O
4
. We focused our attention on bromo-
benzene 1a as a model substrate. The reactions were carried out
in the absence of toxic/ammable organic acids, and anhy-
drides (activators). To begin, we screened the catalysts, Co O4
3
(
(
X1), 3% Pd/Co
X4), 3% Pd/Co
3
O
4
(X2), 3% Pd/Co
–K (X5), 3% Pd/Co
3 4
–Li (X7) and 5% Pd/Co O
3
O
4
–Ca (X3), 3% Pd/Co
–Cs (X6), 3% Pd/Co
–Li (X8). As
3 4
O –Na
3
O
4
3
O
4
3 4
O –
Li (X*), 1% Pd/Co
3
O
4
shown in Fig. 6, the reaction did not occur in the absence of the
Pd and alkali metal ions for the X1 catalyst. When Pd was added
to X1 to give X2, an improvement in the 2a product yield of
12.4% was recorded, which rose signicantly to 36.1–60.3%
following doping with the alkali metal ions (x), suggesting that
the alkali metal ions are essential to achieve reasonable 1a
conversion. Gratifyingly, it was discovered that the lithium-
promoted catalyst X* was the most active to the others, with
a moderate selectivity and yield of 60.3% obtained. An increase
in the Pd loading up to 5% (X8) had no discernible effect on the
2a yield (57.8%), and likewise when the Pd loading was reduced
to 1% (X7). Meanwhile, the 2a* byproduct (phenyl formate)
formed was as a result of a possible nucleophilic attack of
HCOOH, which resulted in a lower yield for some of the reac-
tions. According to the activity trend, Pd is the active species in
these reactions, while the alkali-cations act as promoters/co-
catalysts and are critical for stabilizing the active Pd species
on the support. The increase in the catalytic activity of the Li-
promoted catalyst was proportional to the cationic Lewis-acid
strength, in the order Li > Ca > Na > K > Cs and may be attrib-
uted to the lithium's high charge density and low ionic radii,
which promoted and triggered the Pd species' charge site for the
reactions to proceed smoothly.
Fig. 6 Catalyst screening. Reaction conditions: catalyst (0.3 g), 1a (4
mmol), HCOOH : Me-THF 3 : 3 (mL), B1 (1 equiv.), 100 C, 12 h. Iso-
lated yield and selectivity (%) determined by GC-FID analysis.
ꢀ
ꢀ
a minor and signicant weight loss between 160 and 1000 C,
ascribed to absorbed moisture on the catalyst surface (160–200
ꢀ
C) and thermal degradation of the organic/inorganic matter
ꢀ
(206–800 C). All catalysts exhibited peaks associated with metal
bands in their FT-IR spectra. It is worth mentioning that the Pd
and alkali-cations did not alter the structure of the mesoporous
Co O (Fig. S4†).
3 4
Optimization of reaction conditions. With the best catalyst
X* in hand, we reported the optimization of reaction condi-
tions. First, we investigated several reaction parameters (cata-
Catalytic reactions
Catalyst screening. The carboxylation reaction was used to lyst loading, time, temperature, solvents, bases, and
establish the general concept for the newly developed alkali-
a
Table 2 Effect of reaction parameters
ꢀ
b
b
b
Entry
Cat. X* (g)
Time (h)
Temp. ( C)
Solvents
Nucleophiles
Conv. (%)
Sel. (%)
Yield (%)
1
2
3
4
5
6
7
8
9
0.3
0.1
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
12
12
12
8
100
100
100
100
100
110
120
70
110
110
110
110
110
110
110
110
110
110
Me-THF
Me-THF
Me-THF
Me-THF
Me-THF
Me-THF
Me-THF
Me-THF
Acetonitrile
1,4-Dioxane
Toluene
Me-THF
—
—
—
—
—
—
—
—
—
—
—
—
—
75.1
66.3
79.9
52.8
81.8
85.5
93.6
<20
70.7
90.5
85.5
80.6
85.3
75.2
68.4
75.3
60.9
69.3
80.4
71.3
76.8
70.4
79.6
90.2
74.7
—
65.3
67.1
70.3
38.2
57.2
51.9
52.4
44.2
38.5
50.5
60.3
47.4
61.7
36.8
66.2
76.2
71.1
<20
45.9
59.8
59.5
30.6
48.7
39.1
35.9
33.2
23.5
35.1
16
12
12
12
12
12
12
12
12
12
12
12
12
12
10
11
12
13
14
15
16
17
18
c
Methanol
Ethanol
Propanol
Butanol
tert-Butanol
Cl-butanol
—
—
—
—
—
a
b
Reaction conditions: 3% X* (0.1–0.5 g), 1a (4 mmol), HCOOH : nucleophiles : Me-THF 3 : 3 : 3 (mL), B1 (1 equiv.). Conversion, selectivity and
c
2 3
isolated yields determined by GC-FID. (X*, 0.5 g) with K CO (1 equiv.). Entries 1–12 (carboxylation). Entries 13–18 (alkoxycarbonylation).
©
2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 26937–26948 | 26943

View Article Online
RSC Advances
Paper
ꢀ
nucleophiles) under optimized reaction conditions. The results 110 C, the conversion and yield rose substantially to 85.5 and
ꢀ
are summarized in Table 2. When the catalyst amount was 76.2%, respectively (Table 2, entry 6). At 120 C, an increase of
reduced to 0.1 g, the yield 2a decreased signicantly to 47.4% 93.6% in 1a conversion was observed (Table 2, entry 7). The
ꢀ
(
Table 2, entry 2). At 8 h reaction time, the conversion and yield reaction was not favorable for temperatures <80 C, with the
decreased to 52.8% and 36.8%, respectively (Table 2, entry 4), conversion and yield recorded at <20% (Table 2, entry 8). Next,
and increased slightly to 66.2% aer 16 h (Table 2, entry 5). we focused on utilizing some selected conventional organic
ꢀ
Aer twelve hours, when the reaction temperature was raised to solvents with the reaction conditions: 110 C, 12 h. As expected,
Fig. 7 Reaction conditions: 3% X* (0.5 g), substrates (4 mmol), HCOOH : Me-THF 3 : 3 (mL), B1 (1 equiv.). Conversion and isolated yields
a
ꢀ
determined by GC-FID analysis. Reaction conditions: 5% X* (1.0 g), 1l (4 mmol), HCOOH : Me-THF 6 : 6 (mL), B1 (1 equiv.), 130 C, 24 h.
26944 | RSC Adv., 2021, 11, 26937–26948
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
the solvents (Table 2, entries 9–11) provided signicant yields
General substrate scope for the reaction. To further explore
and selectivities, with the highest yield still obtained for Me- the exibility of this new catalyst, we investigated a variety of
THF (76.2%) (Table 2, entry 6); this is presumably due to its aryl halides (Fig. 7). Interestingly, carboxylation of various
moderate polarity, which enables it to dissolve a broad range of substituted aryl halides (1a–1n) produced carboxylic acid
2 3
organic compounds. Between K CO and the triethylamine, products with moderate to excellent yields (2a–2n). Notably, the
signicant substrate conversions were obtained for both bases, halo group attached to the aryl bromides, such as 1b (Cl),
indicating that the triethylamine, a versatile base used in survived the reaction regardless of substitution mode. In terms
carbonylation reactions, can be conveniently replaced with of general applicability, aryl bromide substrates containing
a non-volatile base (K
2
CO
3
) (Table 2, entry 12).
electron-donating groups, such as methoxy 1c, para-methyl 1d,
Additionally, to demonstrate the versatility of this novel para-substituted dimethyl 1e, and amino 1f, and electron-
catalyst under the optimized reaction conditions, we validated withdrawing groups (cyano 1g, nitro 1h, and carbonyl 1i),
the possibility of using different nucleophiles to investigate were found to be compatible with the reaction, yielding
these alkyl group's effects on the alcohol for the alkox- moderate to excellent 2c–2i products. Furthermore, we
ycarbonylation reaction. Notably, methanol was the most effi- extended the protocol to the 2-iodonapthalene substrate 1j and
cient nucleophile compared to the others, though, no its sterically hindered counterpart 1k. Notably, the former
signicant difference in product yields and selectivities (Table provided a much better 2j yield than the latter (2k). Following
2, entries 13–18). The observation thus implies that an increase that, we screened the 1l chlorobenzene (aryl chloride) substrate;
in the alkyl chain length on the alcohols may substantially unfortunately, a yield of <10% was obtained. Despite signicant
64
reduce yields owing to the formation of additional side prod- progress made over the last decades, carbonylation of non-
ucts. The yield of 2a dropped to 23.5% when tert-butanol was activated chlorobenzene and its derivatives to their desired
used, presumably due to steric hindrance from the bulk methyl products remains a challenge compared to iodo- and bromo-
substituents attached to the alcohol molecule (Table 2, entry benzene. This is due to the higher CO pressure and temperature
1
7). Interestingly, the yield of the chloro-substituted butanol required to achieve reasonable conversion, owing to the mole-
3
signicantly increased compared to its unsubstituted counter- cule's robust C(sp )–Cl bonds. However, a reasonable 2l yield of
part, demonstrating the effect of the electron-withdrawing up to 31% was obtained by increasing the Pd loading/catalyst
group on the nucleophile (Table 2, entry 18).
amount, temperature, and time. Carboxylation of the aryl
iodide substrates (1m and 1n) was well-tolerated. The reaction,
1
Fig. 8 Reaction conditions: 3% X* (0.5 g), substrates (4 mmol), HCOOH : MeOH : R -NH
and isolated yields were determined by GC-FID analysis.
2
: Me-THF 3 : 3 : 3 : 3 (mL), B1 (1 equiv.). Conversion
RSC Adv., 2021, 11, 26937–26948 | 26945
©
2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
RSC Advances
Paper
however, was incompatible with aryl iodides containing pyri-
dine moieties (1o and 1p). Only trace amounts of 2o and 2p
yields were obtained.
Besides aryl iodides and bromides, benzyl halides (BnCl and
BnBr) can also be converted to carboxylic acids. Comparing the
1
reactivities of 1q (BnBr) and 1q (BnCl), the former yielded
1
much more 2q than the latter (2q ), at 66% and 54%, respec-
tively; this is directly related to and consistent with the relative
3
stabilities of the C(sp )–X bonds. Regardless of the substitution
position, moderate yields of 2r (2-Cl) and 2s (4-Cl) were obtained
using chloro-substituted benzyl chloride (1r and 1s). Similarly,
electron-donating groups (OMe and Me) attached to the benzyl
chloride substrates 1t and 1u were effectively carboxylated to 2t
(59%) and 2u (67%), respectively.
Additionally, we broadened the substrate scope for amino-
and alkoxycarbonylation of aryl halides (Fig. 8). Since these
reactions are relatively the same, we also utilized the optimized Fig. 9 Reaction kinetics for the 1a and 1l substrates: isolated yields
reaction conditions stated above in Fig. 7. The amino- determined by GC-FID analysis. Under the same reaction conditions
ꢀ
(
110 C, 8 h), the 1l substrate recorded a yield of 2l at <5%. At 5% X* (1.0
carbonylation of p-chlorobromobenzene 1b was catalyzed by X*
using benzylamine as the nucleophile. Electron-donating group
substrates, such as 1c (MeO) and 1d (Me), were successfully
utilized to obtain the desired amides (3c and 3d). Also, the
chlorosubstituted 1b substrate was tolerated. Next, we explored
the versatility of the catalyst for the alkoxycarbonylation reac-
tion. The chlorosubstituted bromobenzene 1b was successfully
alkoxycarbonylated to their corresponding ester 4b in moderate
yield. Interestingly, excellent yields of 4c–4h were obtained
using substrates substituted with electron-donating groups
ꢀ
g), 130 C, 24 h, the 2l yield reached >30%.
required to stabilize the catalytic system. Additionally, it is well
established that the same acid anhydrides are incompatible
with aminocarbonylation reactions due to the formylating
agents (acetic formic anhydride) formed in solution during the
CO release process. This necessitates the development of safer
procedures for this type of reaction. Typically, an autoclave
equipped with a stirring bar was charged with X* (0.5 g),
HCOOH (3 mL), Me-THF (3 mL), and B1 (1 eq.) and vigorously
stirred at 550 rpm for 16 h (110–130 C). The reaction was halted
and allowed to cool to ambient temperature before being
analyzed on the GC-TCD machine.
Interestingly, the formic acid decomposed into a mixture of
CO and CO , with the highest amount/area obtained for the CO,
2
67
such as 1c (OMe), 1d (Me), 1e (CH
electron-withdrawing groups such as 1g (NC) and 1h (NO
3
)
2
, and 1f (NH
2
), as well as
).
2
ꢀ
Lastly, it is worth noting that the alkoxycarbonylation of 1d was
completed in a shorter reaction time than the amino-
carbonylation reaction, demonstrating the substrate's high
reactivity for the alkoxycarbonylation reaction.
suggesting that this system can readily be used in lieu of acid
anhydrides for the in situ generation of CO from formic acid.
Kinetic and GC-TCD studies
Although, a minute amount of H gas was also detected in the
spectra aer 3.436 retention time. It is worth noting that the
2
The kinetic analysis was carried out on the substrates 1a (bro-
mobenzene) and 1l (chlorobenzene) (Fig. 9). Carboxylation of 1a
was much faster than 1l at a 8 h reaction time, with conversion
and yield to the target product 2a reaching 73% and 51%,
respectively. Conversion and yield of the 1l substrate were both
new catalyst serves a dual purpose: it facilitated the decompo-
3
sition of the HCOOH and triggers the substrates' C(sp )–X
bonds, allowing for clean carbonylative transformations.
Additionally, an open-system experiment was conducted to
prove the in situ generation of CO (Fig. S6†). The 2a yield
decreased substantially under this “open-system condition”
compared to the conventional procedure described before,
probably due to the reduced gas pressure released throughout
the process. As a result, the experiment veries that the gener-
ated gas is CO (see Fig. S6 in the ESI†).
<5% under identical reaction conditions. When the time was
extended to 20 h, the conversion of 1a reached almost 100%,
while the complete conversion of 1l took around 24–30 h. The
kinetic study implies that a shorter time (low pressure) is
required for carboxylation of the 1a, while a longer time (higher
pressure/temperature) is needed for carboxylation of the 1l due
3
to the stronger C(sp )–Cl bonds in the 1l molecule (Fig. 9).
To gain a better insight into the formic acid decomposition
of this new system, a GC-TCD analysis was performed (Fig. S5†).
Reusability and heterogeneity test
Several literature works have reported the use of acid anhy- The reusability of the Pd-based heterogeneous catalyst was
drides and other activators to facilitate the CO release from tested with the model reaction (carboxylation) using our
1
9,65,66
formic acid.
Despite the moderate to good yields obtained benchmark substrate 1a, under the optimized reaction condi-
ꢀ
with these additives, product contamination has been associ- tions of 110 C, 12 h (Fig. 10). Centrifugation at 450 rpm for
ated with this protocol due to the anhydride's slow decompo- 30 min was used to separate the catalyst, followed by simple
sition and severe reaction conditions, resulting in a longer time ltration and reactivation in an oven at 350 C for 30 min before
ꢀ
26946 | RSC Adv., 2021, 11, 26937–26948
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
protocol and therefore establish it as a critical tool for devel-
opment in academia, the chemical industry, and drug delivery
applications.
Conflicts of interest
There are no conicts of interest to declare.
References
1
2
3
T. Fukuyama, T. Totoki and I. Ryu, Green Chem., 2014, 16,
042–2050.
C. O. Oseghale, B. M. Mogudi, C. A. Akinnawo and
R. Meijboom, Appl. Catal., A, 2020, 117735.
2
X. Qi, R. Zhou, H.-J. Ai and X.-F. Wu, J. Catal., 2020, 381, 215–
2
21.
4 S. Zhang, H. Neumann and M. Beller, Chem. Soc. Rev., 2020,
9, 3187–3210.
ꢀ
Fig. 10 Catalyst reusability. Reaction conditions: 110 C, 12 h.
4
5
C. Zhu, J. Liu, M.-B. Li and J.-E. B ¨a ckvall, Chem. Soc. Rev.,
2020, 49, 341–353.
6 S. P. Chavan and B. M. Bhanage, Eur. J. Org Chem., 2015,
2015, 2405–2410.
7 C. Veryser, S. Van Mileghem, B. Egle, P. Gilles and W. M. De
Borggraeve, React. Chem. Eng., 2016, 1, 142–146.
8 R. Sang, C. Schneider, R. Razzaq, H. Neumann, R. Jackstell
and M. Beller, Org. Chem. Front., 2020, 7, 3681–3685.
9 J. H. Choi, H. Do Kim, J. Y. Kang, T. Jeong, P. Ghosh and
I. S. Kim, Bull. Korean Chem. Soc., 2021, 42, 542–547.
reuse. As shown in Fig. 10, the 2a yield decreased in the second,
third, and fourth cycles but signicantly rose in the h and
sixth cycles, reaching 82 and 85%, respectively, indicating its
stability. Additionally, our initial test reveals that the unpro-
moted catalyst exhibited signicant leaching of the active Pd in
the reaction mixture. The increased stability of the Li-promoted
catalyst over its unpromoted counterpart may be attributed to
the alkali metal in the catalyst, which minimized leaching of the
active Pd species, improved the catalyst's binding properties,
and enhanced the catalytic activity for the reactions (Fig. 10). It 10 R. Geitner, A. Gurinov, T. Huang, S. Kupfer, S. Graefe and
is worth noting that the reusability of thi novel catalyst signi-
B. M. Weckhuysen, Angew. Chem., 2021, 133, 3464–3469.
cantly improves the viability of our catalytic system in 11 J. Cao, Z.-J. Zheng, Z. Xu and L.-W. Xu, Coord. Chem. Rev.,
6
8–72
comparison to the homogeneous systems.
Next, the heterogeneity test was performed by rst ltering 12 L. Liu, X.-C. Chen, S.-Q. Yang, Y.-Q. Yao, Y. Lu and Y. Liu, J.
the composite catalyst from the reaction mixture, and aer that,
Catal., 2021, 394, 406–415.
the ltrate was reintroduced into the reaction tube under the 13 N. Hussain, A. K. Chhalodia, A. Ahmed and D. Mukherjee,
2017, 336, 43–53.
ꢀ
reaction conditions of 110 C and 8 h. Interestingly, no change
in conversion was observed aer 8 h of reaction time (Fig. S7†), 14 M.-C. Fu, in Studies on Green Synthetic Reactions Based on
suggesting that the active Pd species that catalyzed the reaction
Formic Acid from Biomass, Springer, 2020, pp. 1–26.
ChemistrySelect, 2020, 5, 11272–11290.
was not leached into the mixture. In addition, the ICP-OES 15 L. Wang, H. Neumann and M. Beller, Angew. Chem., 2018,
th
analysis of the ltrate was performed aer the 8 run, with
130, 7026–7030.
no possible bleeding of the Pd species.
16 P. Gautam and B. M. Bhanage, Catal. Sci. Technol., 2015, 5,
4
663–4702.
7 T. Ueda, H. Konishi and K. Manabe, Org. Lett., 2012, 14,
370–5373.
1
Conclusion
5
The carboxy- and carbonylation reactions were achieved with 18 K. Sordakis, C. Tang, L. K. Vogt, H. Junge, P. J. Dyson,
the newly designed heterogeneous Pd catalyst and synthetic M. Beller and G. Laurenczy, Chem. Rev., 2018, 118, 372–433.
protocol, which in situ generates CO from formic acid. Although 19 R. Sang, P. Kucmierczyk, K. Dong, R. Franke, H. Neumann,
the system was more favorable for carboxylation reaction,
acceptable product yields were also obtained for alkoxy- and
R. Jackstell and M. Beller, J. Am. Chem. Soc., 2018, 140,
5217–5223.
aminocarbonylation processes. Without the need for acid co- 20 F. Jin, J. Yun, G. Li, A. Kishita, K. Tohji and H. Enomoto,
catalysts and acetic anhydrides, the reactions proceeded Green Chem., 2008, 10, 612–615.
smoothly with this new catalyst under mild reaction conditions 21 R. W ¨o lfel, N. Taccardi, A. B ¨o smann and P. Wasserscheid,
and at a shorter reaction time. Besides the catalytic system's
Green Chem., 2011, 13, 2759–2763.
superior selectivity and unprecedented reactivity, another more 22 T. W. Myers and L. A. Berben, Chem. Sci., 2014, 5, 2771–2777.
exciting feature was observed in the catalyst's practical reus- 23 Y. Manaka, W.-H. Wang, Y. Suna, H. Kambayashi,
ability. We believe that the reported results adequately
demonstrate the synthetic utility of this green carbonylation
J. T. Muckerman, E. Fujita and Y. Himeda, Catal. Sci.
Technol., 2014, 4, 34–37.
©
2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 26937–26948 | 26947

View Article Online
RSC Advances
Paper
2
2
2
2
2
2
3
3
3
3
3
3
3
3
4 X.-D. Lang, F. You, X. He, Y.-C. Yu and L.-N. He, Green Chem., 48 T. Grewe and H. T u¨ ys u¨ z, J. Mater. Chem. A, 2016, 4, 3007–
019, 21, 509–514. 3017.
5 M. V Khedkar, S. R. Khan, T. L. Lambat, R. G. Chaudhary and 49 A. Aguilar-Tapia, L. Delannoy, C. Louis, C. W. Han,
A. A. Abdala, Curr. Org. Chem., 2020, 24, 2588–2600. V. Ortalan and R. Zanella, J. Catal., 2016, 344, 515–523.
6 J. Hagen, Industrial catalysis: a practical approach, John Wiley 50 W. Zhou, Y. Guan, D. Wang, X. Zhang, D. Liu, H. Jiang,
2
&
Sons, 2015.
7 L. Wang, W. Sun and C. Liu, Chin. J. Chem., 2018, 36, 353–
62.
J. Wang, X. Liu, H. Liu and S. Chen, Chem.–Asian J., 2014,
9, 1648–1654.
51 S. Li, Y. Zhang, J. Shi, G. Zhu, Y. Xie, Z. Li, R. Wang and
H. Zhu, Nanomaterials, 2020, 10, 31.
52 M. N. Pahalagedara, L. R. Pahalagedara, C.-H. Kuo,
S. Dharmarathna and S. L. Suib, Langmuir, 2014, 30, 8228–
8237.
3
8 J. Yang, J. Liu, H. Neumann, R. Franke, R. Jackstell and
M. Beller, Science, 2019, 366, 1514–1517.
9 B. Urb ´a n and R. Skoda-F ¨o ldes, Phosphorus, Sulfur Silicon
Relat. Elem., 2019, 194, 302–306.
0 B. Urb ´a n, P. Szab ´o , D. Srank ´o , G. S ´a fr ´a n, L. Koll ´a r and 53 Y. Liu, J.-F. Chen and Y. Zhang, RSC Adv., 2015, 5, 29002–
R. Skoda-F ¨o ldes, J. Mol. Catal., 2018, 445, 195–205.
29007.
1 Y. Lei, L. Wu, X. Zhang, H. Mei, Y. Gu and G. Li, J. Mol. Catal. 54 S. Zhuang, Y. Liu, S. Zeng, J. Lv, X. Chen and J. Zhang, J. Sol-
Chem., 2015, 398, 164–169. Gel Sci. Technol., 2016, 77, 109–118.
2 S. Mitchell, E. Vorobyeva and J. P ´e rez-Ram ´ı rez, Angew. Chem. 55 S. Sun, H. Li and Z. J. Xu, Joule, 2018, 2, 1024–1027.
Int. Ed., 2018, 57, 15316–15329.
3 F. Chen, X. Jiang, L. Zhang, R. Lang and B. Qiao, Chin. J.
Catal., 2018, 39, 893–898.
56 C. Gregor, M. Hermanek, D. Jancik, J. Pechousek, J. Filip,
J. Hrbac and R. Zboril, Eur. J. Inorg. Chem., 2010, 16, 2343–
2351.
4 A. Mansour and M. Portnoy, J. Mol. Catal. Chem., 2006, 250, 57 J. Dou, Z. Bao and F. Yu, ChemCatChem, 2018, 10, 250–258.
4
0–43.
58 Y. Wang, C. Zhang, F. Liu and H. He, Appl. Catal., B, 2013,
142, 72–79.
5 Z. S. Qureshi, S. A. Revankar, M.
V
Khedkar and
B. M. Bhanage, Catal. Today, 2012, 198, 148–153.
6 A. S. Prasad and B. Satyanarayana, J. Mol. Catal. Chem., 2013,
59 Z. Chen, S. Wang, Y. Ding, L. Zhang, L. Lv, M. Wang and
S. Wang, Appl. Catal., A, 2017, 532, 95–104.
60 N. Yang, S. Ni, Y. Sun and Y. Zhu, J. Mol. Catal., 2018, 452,
28–35.
370, 205–209.
7 M. Genelot, N. Villandier, A. Bendjeriou, P. Jaithong,
L. Djakovitch and V. Dufaud, Catal. Sci. Technol., 2012, 2, 61 J.-P. Dacquin, C. Dujardin and P. Granger, J. Catal., 2008,
886–1893. 253, 37–49.
8 M. Genelot, V. Dufaud and L. Djakovitch, Adv. Synth. Catal., 62 B. Liu, B. Ouyang, Y. Zhang, K. Lv, Q. Li, Y. Ding and J. Li, J.
013, 355, 2604–2616. Catal., 2018, 366, 91–97.
9 T. T. Dang, Y. Zhu, S. C. Ghosh, A. Chen, C. L. L. Chai and 63 B. Ouyang, S. Xiong, Y. Zhang, B. Liu and J. Li, Appl. Catal., A,
A. M. Seayad, Chem. Commun., 2012, 48, 1805–1807. 2017, 543, 189–195.
0 T. T. Dang, Y. Zhu, J. S. Y. Ngiam, S. C. Ghosh, A. Chen and 64 A. Brennf u¨ hrer, H. Neumann, S. Klaus, T. Riermeier,
A. M. Seayad, ACS Catal., 2013, 3, 1406–1410. J. Almena and M. Beller, Tetrahedron, 2007, 63, 6252–6258.
1
3
3
4
4
2
1 C. O. Oseghale, B. M. Mogudi, O. R. Onisuru, C. A. Akinnawo, 65 X. Qi, H. Li and X. Wu, Chem.–Asian J., 2016, 11, 2453–2457.
D. P. Fapojuwo and R. Meijboom, ChemCatChem, 2021, 13, 66 S. Cacchi, G. Fabrizi and A. Goggiamani, Org. Lett., 2003, 5,
1
311–1316.
2 Z. Jiang, X. Feng, J. Deng, C. He, M. Douthwaite, Y. Yu, J. Liu, 67 W. Stevens and A. Van Es, Recl. Trav. Chim. Pays-Bas, 1964,
Z. Hao and Z. Zhao, Adv. Funct. Mater., 2019, 29, 1902041. 83, 863–872.
3 Y. Wang, H. Arandiyan, J. Scott, H. Dai and R. Amal, Adv. 68 B. Bartal, G. Mikle, L. Koll ´a r and P. Pongr ´a cz, J. Mol. Catal.,
Sustainable Syst., 2018, 2, 1700119. 2019, 467, 143–149.
4 X. Hu, Z. Zhang, Y. Zhang, L. Sun, H. Tian and X. Yang, Eur. J. 69 C. Li, X. Qi and X. Wu, ChemistrySelect, 2016, 1, 1702–1704.
4269–4272.
4
4
4
4
4
4
Inorg. Chem., 2019, 2019, 2933–2939.
70 F. Wu, J. Peng, L. Meng, X. Qi and X. Wu, ChemCatChem,
2017, 9, 3121–3124.
71 A. Ahmed, N. Hussain, M. Bhardwaj, A. K. Chhalodia,
A. Kumar and D. Mukherjee, RSC Adv., 2019, 9, 22227–22231.
72 X. Qi, C. Li and X. Wu, Chem.–Eur. J., 2016, 22, 5835–5838.
5 B. M. Mogudi, P. Ncube, N. Bingwa, N. Mawila, S. Mathebula
and R. Meijboom, Appl. Catal., B, 2017, 218, 240–248.
6 Y. Li, C. Zhang, H. He, J. Zhang and M. Chen, Catal. Sci.
Technol., 2016, 6, 2289–2295.
7 S. Biswas, A. S. Poyraz, Y. Meng, C.-H. Kuo, C. Guild, H. Tripp
and S. L. Suib, Appl. Catal., B, 2015, 165, 731–741.
26948 | RSC Adv., 2021, 11, 26937–26948
© 2021 The Author(s). Published by the Royal Society of Chemistry