S. Srivastava et al.
total S/N ratio for factor (A) at level 2 can be obtained by
adding the individual S/N ratio for experiments 5–8 (see
Table 1). Figure 1 depicts the effect of each factor at each
level. Itisobserved that the low temperature and highpressure
favours the selective hydrogenation of furfural to FA. The
total S/N ratio of factor A was maximum for level 3 (130 °C)
and minimum for level 1 (170 °C). These results revealed that
increase in temperature results in decreased FA yield. This
substantial decrease may be attributed to further hydrogen-
olysis of FA to MF at higher temperature as increase in tem-
perature promotes the C–O hydrogenolysis for the cleavage of
the carbonyl group. It is noteworthy that hydrogenation of
furfural to FA is dominant at low temperature while C–O
hydrogenolysis dominated at high temperature is due to the
higher activation of the latter [19]. Further, there may be
polymerization of FA due to its adsorption to the catalyst
surface, which is favoured at high temperature [34].
increased solubility of hydrogen in the reaction mixture at
increased hydrogen pressure can account for the increased
conversion [2], the reaction pressure was varied in the
range of 10–40 bar. It was observed that the FA yield
increased with pressure and maximum yield was achieved
at 30 bars. Hence, 40 bar was selected as a maximum
pressure level. Considering the rate of reaction, which is
directly proportional towards active sites available in the
reduced catalyst, the catalyst loading was varied from 0.5
to 2 g. It was observed that FA yield was increased with
catalyst loading and reached maximum for 1.5 g, hence
catalyst loading of 2 g was selected as a maximum limit.
Solvent effect in hydrogenation reactions over heteroge-
neous catalysts have been rationalized by correlating
reaction rates and product distributions with solvent
polarity and dielectric constant [20]. The nature of solvent
can affect the kinetic of competitive hydrogenation reac-
tion of both polar and non-polar substrates. It is found that
polar solvent enhances adsorption of non-polar reactants
while a non-polar solvent enhances the adsorption of a
polar reactant [20]. In present work, the effect of 2-pro-
panol for the hydrogenation of furfural was investigated in
the range of molar ratio 1:3–1:9. It was observed that the
FA yield was increased with increasing solvent amount.
Hence, molar ratio of 1:3 was selected as maximum level.
It is necessary to evaluate the effects of time to understand
the kinetics of the reaction and to ensure the completion of
reaction; it was kept between 1 and 4 h. According to
Taguchi method, L16 orthogonal array was constructed and
experiments were performed accordingly. The contribution
of each parameter on FA yield was calculated by statistical
analysis and ANOVA using Minitab software.
The total S/N ratio of factor B increased with pressure
and reached maximum at 30 bar, thereafter decreased
sharply. This may be explained by the fact that high
pressure increases the solubility of hydrogen in the reaction
mixture. This increased hydrogen pressure can account for
increased conversion of furfural. Hence, the increased FA
yield can be observed. These results are in accordance with
the previous studies [2]. The hydrogenation of furfural to
FA is a reversible reaction; further increase in pressure may
reverse the equilibrium towards conversion of FA to fur-
fural which may result in decreased FA yield.
In order to find the effect of time on the FA yield, the
reactions were carried out at different time frame varying
from 1 to 4 h. The effect of reaction time on FA yield
suggested that longer reaction time does not favour FA
yield owing to further transformation of FA to other by
products such as MF or tetrahydrofurfuryl alcohol. The
maximum yield was observed for 3 h reaction time.
However, no significant increase in S/N ratio was observed
between 2 and 3 h duration, hence 2 h can be selected as
optimum time period. The effect of catalyst loading on S/N
ratio revealed that higher catalyst loading results in
decreased FA yield. Maximum yield was obtained for 1 g
catalyst. The increased activity of catalyst is ascribed to
availability of more number of active sites. However, the
decrease in FA yield for higher catalyst loading may be due
to further transformation of FA to other by products such as
methyl furan (MF) and cyclopentanol (CPL) due to the
availability of more number of active sites to polarise the
C–O bond present in FA molecule. These results are in
accordance with available reports [34, 35]. The S/N ratio
for factor E (molar ratio of furfural to 2-propanol) was
highest at level one (1:7) and decreased as the ratio
decreased from 1:7 to 1:3 because of availability of less
number of active sites to convert furfural to furfuryl alco-
hol. This revealed that higher yield can be obtained at a low
In the present study, contribution of selected parameters
on FA yield over Co–Cu/SBA-15 catalyst was investigated
and optimized. Identical procedure was followed to carry
out all the experiments. Since the aim of present study was
to maximize the FA yield, the signal to noise (S/N) ratio
was calculated considering the ‘‘larger is better’’ case using
following equation [27, 31–33];
ꢂ
ꢀ
ꢁꢃ
X
S
1
n
n
1
Yi2
¼ ꢁ10 log
ð1Þ
i¼1
N
where, n = number of repetitions carried out for each
experiment and Yi = the yield of ith experiment. Rose et al.
[27] has investigated three types of S/N ratios like, larger-the-
better, smaller-the-better, and nominal is the best. All the
experiments were carried out in duplicate and corresponding
S/N ratios (Table 1) were calculated accordingly. Table 2
depicts the values of the total S/N ratio. Symbols A, B, C, D,
and E were used for temperature (°C), pressure (bar), reaction
time (h), catalysts dose (g), and molar ratio of furfural to
2-propanol respectively. For each factor at each level, the
calculations were performed by using an individual ratio, i.e.,
123