T.A. Gokhale et al.
Molecular Catalysis 510 (2021) 111667
method [11]. Zhang and his group showed conversion of nitriles to
primary amines by using nickel supported on doped carbon structure
[28]. His group further loaded cobalt on nitrogen-doped carbon to
achieve good yields using carbonyl compounds for reductive amination
reaction [29]. Jv et al. showed decent conversions with pH-controlled,
ligand stabilized Pd nanoparticles for the reductive amination of aryl
aldehydes [30]. Recently, Xie et al. used ruthenium loaded on titanium
phosphate exhibited good yields in the case of furfural substrate in 24 h
of reaction time. [31]
dissolved in 100 mL of Milli-Q water. This reducing solution was then
added dropwise to the suspended stirring solution in the three-neck flask
with continued N2 bubbling of nitrogen over a period of 30 min. The
resultant black material was separated by decantation and washed once
with Milli-Q water and twice with acetone. The material was dried at
room temperature in a desiccator for 24 h. The prepared and dried
nanocomposites with ruthenium and nickel were calcined at 170 ◦C and
500 ◦C respectively with 80:20 ratio of N2:H2 gaseous mixture for three
hours each in a tube furnace. The respective reduction temperatures
were determined from H2-TPR analysis for both the nanocomposites.
Mainly, the focus in this sub-branch of the field has been about
optimizing and testing the support material used for loading of the
nanoparticles as it holds significant standing in regards to the perfor-
mance of the catalyst. One of the environmentally benign, inexpensive,
and omnipresent support materials, i.e., Montmorillonite clay (MMT)
might be a good choice as 1) It has a lamellar based structure in its
magnesium aluminum silicate assembly, translating into higher than
usual surface area when exfoliated 2) High tolerance temperature and
pressure stability 3) Present electrostatic force due to the negatively
charged surface, which holds the nanoclusters in place reducing any
metal leaching and 4) Mixture of weak to moderate surface Brønsted and
Lewis acidic sites [32,33].
General procedure
The general procedure for the reductive amination reactions involves
the use of a 100 ml capacity autoclave at 400 rpm. The optimized re-
action conditions for Ru- based nanocomposite are as follows: 90 ◦C, 0.5
mmol of the aldehyde, 4 ml of 25% ammonia solution, 30 mg catalyst,
10 bar H2 pressure. While the Ni-based nanocomposite employed reac-
tion conditions as: 130 ◦C, 0.5 mmol of aldehyde, 4 mL of 25% ammonia
solution, 30 mg catalyst, 15 bar H2 pressure. After the reaction, the re-
action mixture was centrifuged and filtered to separate the catalyst. The
solvent from the reaction mixture was removed on rotary evaporator.
The dense residual liquid was then dissolved in ethyl acetate and passed
through sodium sulfate bed for removing residual moisture. Further
purification of the product was carried out by column chromatography
using neutral alumina with PET ether/ethyl acetate system.
Present work aids in establishing a nexus between the inherent cat-
alytic performance metric of the two metals (i.e., ruthenium and nickel)
in terms of the reductive amination reaction applied to bio-derived
molecules. At the same time, all controllable variables have been kept
identical at every stage, including the support material followed by
tuning of these parameters to maximize the reaction yields and con-
version. We report the catalytic performance comparison in terms of
reaction yield, reaction time, required metal composition in the catalyst,
reaction temperature, and pressures between ruthenium and nickel on
MMT-K10 clay in varying amounts towards the reductive amination of
biomass-derived as well as other aryl molecules to corresponding pri-
mary amines. Similarly, reductive amination of crude furfural extracted
from xylose and lignocellulosic biomass hydrolysis has also been
explored using both Ru and Ni-nanocomposites.
Results and discussion
Various characterization techniques were employed to analyze dis-
tinctions between the respective nanocomposites. The techniques uti-
lized includes P-XRD, XPS, SEM-EDX, BET surface area analysis, NH3-
TPD, H2-TPR, Py-IR, HR-TEM. X-ray diffraction studies were conducted
to identify the metal by characteristic peaks as well as verify the nano-
scale of metal crystallite in the nanocomposite. Diffraction spectra stack
has been shown in Fig. 1A and B. Representative peaks of montmoril-
lonite K10 clay appear at 2θ = 5.19◦, 20.81◦, 26.65◦, 34.98◦, 45.55◦, and
61.75◦. The representative ruthenium nanoparticle peaks appear at 2θ
= 43.98◦ and 78.6◦, which match the ICDD-JCPDS card number: 6-0663
for ruthenium metal, matching with the (101) and (103) mirror planes
respectively [35]. The low intensity of the characteristic ruthenium
nanoparticle peaks can be owed to the sub-4% loading on the support
material. Average crystallite size determined using Scherrer’s equation
in reference to the highest intensity peak at 43.98◦ was found to be in the
range of 10–12 nm.
Experimental
Materials
Furfural (>98%), 5-hydroxymethyl furfural (HMF) (>98%), 5-
methyl furfural (>97%), 4,5-dimethyl furfural (>97%), and other
phenyl aldehyde substrates were purchased from Sigma Aldrich, India,
and Alfa Aesar, India. MMT-K10 clay was purchased from Sigma
Aldrich, India. Nickel sulfate hexahydrate (NiSO4. 6H2O) (98%) was
acquired from Alfa Aesar, India. Ruthenium chloride (RuCl3. xH2O) and
Sodium borohydride (NaBH4) was purchased from Sigma Aldrich, India.
Ammonia aq. solution (25%) was purchased from Sigma Aldrich, India.
All solvents (AR grade) were obtained from Alfa Aesar, India.
Around 2θ = 5.14◦, the peak in Fig. 1B, which depletes in intensity
considerably, after subsequent loading of Ni metal on the support ma-
terial, signifies the exfoliation of the ordered lamellar structure of the
K10-clay resulting in increased surface area and porosity. The same ef-
fect is not observed in the 2% Ru/MMT as the percent loading of
ruthenium is significantly lower than the nickel nanocomposite. The
pre-calcined nanocomposite peaks do not demonstrate peaks relative to
the β-Ni, unlike the calcined material. This emphasizes the amorphous
nature of the material prior to the calcination step. The calcined mate-
rial, however, shows the definitive characteristic peaks of the Ni crys-
tallite structure at 2θ = 44.29◦, 51.77◦ and 76.40◦, [36] in good
agreement with the ICDD-JCPDS card number: 01-071-4655. The
average crystallite size of the Ni nanoparticles was calculated using
Scherrer’s equation with the highest intensity peak at 44.29◦, which was
found to be in the range of 18–20 nm. The remainder peaks of MMT clay
have been highlighted in Fig. 1A and B. There is a negligible presence of
NiO in the material post calcination with respect to the diffraction data.
XPS measurements were further employed to confirm and determine the
Ru and Ni oxidation states in the structured nanocomposite of 2%
Ru/MMT and 12.5% Ni/MMT in Fig. 1C and D. The peaks corresponding
to the 3d3/2 and 3d5/2 at 284.8 eV and 280.1 eV ascertain the presence of
Catalyst synthesis
The nanocomposites were synthesized using a known procedure with
minor modifications [33,34]. Ruthenium loaded nanocomposites were
synthesized with 1%, 2%, 3%, 4% wt. metal loading percentage, with
respect to the MMT clay support. Nickel nanocomposites utilized 10%,
12.5%, and 15% metal loading percentages for the reaction under focus.
The procedure for synthesis was as follows: 100 mL of Milli-Q water was
charged into a three-neck round bottom flask. 1 g Na-MMT was added to
the flask, and the flask was sonicated for 30 min. Corresponding metal
salt (RuCl3 .x H2O or NiSO4. 6H2O) was weighed according to desired
weight loading and transferred to a 250 mL beaker. Then the salt was
dissolved in 150 mL of Milli-Q water. Then the metal salt solution was
transferred to the three-neck flask. This solution was then subjected to 1
h of N2 gas bubbling to remove any dissolved oxygen with vigorous
stirring. 5 equivalents of NaBH4 were weighed separately and then
2