Efficient utilization of potash alum as a green catalyst for production of furfural, 5-hydroxymethylfurfural and levulinic acid from mono-sugars

Article abstract of DOI:10.1039/c7ra07147g

In present work, we have explored potash alum (PA) as an efficient and green catalyst for production of high value platform chemicals such as 5-hydroxymethylfurfural (HMF) furfural from bio-renewable feedstocks in a biphasic reaction medium. Maximum 64% and 49% HMF yields were obtained from the dehydration reactions of fructose and glucose, respectively, at 140 °C over a 6 h reaction time. Similarly, 55% furfural yield was obtained from xylose at 190 °C in 6 h period. Products obtained have been analyzed by ¹H-NMR spectra and a detailed mechanistic understanding of the dehydration reactions has been developed. This is first report highlighting a strategy for the utilization of cost effective and non-hazardous potash alum as catalyst for the conversion of glucose, fructose, and xylose to respective furans.

Full text of DOI:10.1039/c7ra07147g

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Efficient utilization of potash alum as
a green catalyst for production of furfural,
-hydroxymethylfurfural and levulinic
acid from mono-sugars†
Cite this: RSC Adv., 2017, 7, 41973
5
ab
a
a
bc
Dinesh Gupta,
Ejaz Ahmad,
Kamal K. Pant* and Basudeb Saha*
In present work, we have explored potash alum (PA) as an efficient and green catalyst for production of high
value platform chemicals such as 5-hydroxymethylfurfural (HMF) furfural from bio-renewable feedstocks in
a biphasic reaction medium. Maximum 64% and 49% HMF yields were obtained from the dehydration
ꢀ
reactions of fructose and glucose, respectively, at 140 C over a 6 h reaction time. Similarly, 55% furfural
Received 28th June 2017
Accepted 19th August 2017
ꢀ
yield was obtained from xylose at 190 C in 6 h period. Products obtained have been analyzed by
1
H-NMR spectra and a detailed mechanistic understanding of the dehydration reactions has been
DOI: 10.1039/c7ra07147g
developed. This is first report highlighting a strategy for the utilization of cost effective and non-
rsc.li/rsc-advances
hazardous potash alum as catalyst for the conversion of glucose, fructose, and xylose to respective furans.
15
gasoline, diesel, polymers, solvents and other value added
Introduction
16,17
productss.
Similarly, LA is one of the top ten biomass-
Forthcoming energy crisis due to expeditious depletion of derived platform chemicals that can be used to produce alter-
conventional fossil fuel resources has been a major cause for native of petroleum-derived chemicals and transportation
1
18–20
recent socio-economic changes across the world. Excessive fuels.
In addition to HMF, LA, and furfural (FL) which are
exploitation of conventional fuels have escalated the emission produced from cellulosic and hemi cellulosic component of the
of toxic and greenhouse gases, thereby causing an irrevocable lignocellulosic biomass, have sought attention as a major
2
17,19
threat to the environment as well as to human health. Thus, precursor for the production of biofuel.
attempts have been made to produce alternatives to conven-
In general, fructose is the preferred C
carbohydrate
6
tional fossil fuels from renewable sources. In this regard, substrate by choice for HMF and LA production due to simple
biomass has emerged as a likely bio-renewable source capable dehydration step. On the contrary, glucose as a substrate is
of producing fuel and value added chemicals in a closed carbon relatively cheaper and produced in bulk as compared to fruc-
3
loop, thus eliminating greenhouse gas problems. Biomass tose. Moreover, fructose itself is produced from glucose via
1
,4,5–8
derived high octane liquid fuels
DMF), 5-ethoxymethylfurfural (EMF), ethyl levulinate (EL), jet/ from glucose holds a better sustainability and industrial value.
diesel ranged alkanes and chemicals such as 2,5-bis- Since, HMF and LA are well known platform chemicals acting as
such as 2,5-dimethylfuran isomerization reaction, thus producing HMF and LA directly
9
21
(
1
0,11
ethoxymethylfurfural (BEMF),
2,5-furandicarboxylic acid precursor for an array of products including biodiesel, surfac-
FDCA), 2,5-diformylfuran (DFF), gamma-valerolactone (GVL), tants, solvents and lubricants, a more sustainable approach to
tetrahydrofuran (THF), 2-methyltetrahydrofuran (2MeTHF), produce these platform chemicals by using a readily available,
12
(
13,14
levulinic acid (LA), 5-hydroxymethylfurfural (HMF) route
some of the contenders for being alternatives to fossil based
are abundant, cost-effective and green acid catalyst is required.
Recently, efforts have been made to produce HMF and LA via
products. Moreover, HMF itself is a promising platform chem- isomerization, dehydration and rehydration of glucose at
ical employed to produce an array of chemicals and fuel such as elevated temperatures in the presence of homogeneous Lewis
acidic salts and Brønsted acidic materials such as LaCl ,CrCl ,
3
2
CrCl , FeCl , FeCl , CuCl, CuCl , VCl , MoCl , PdCl , PtCl ,
3
3
2
2
3
3
2
1,22
2
a
Department of Chemical Engineering, Indian Institute of Technology Delhi, New Delhi
2
PtCl , RuCl , RhCl , SnCl , SnCl AlCl , IrCl etc.,
ionic
4
3
3
2
4,
3
3
24
1
b
10 016, India. E-mail: kkpant@chemical.iitd.ac.in
23
14
liquids, solid acids and mineral acids. Interestingly,
glucose conversion into HMF requires isomerization of glucose
Department of Chemistry, University of Delhi, 110007, India. E-mail: basudeb_s@
hotmail.com
c
Catalysis Centre for Energy Innovation, University of Delaware, Newark, DE 19716, to fructose followed by dehydration reaction. The isomerization
USA
step essentially requires Lewis acidity whereas dehydration is
favored in the presence of Brønsted acids. Thus, neither Lewis
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c7ra07147g
1
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 41973–41979 | 41973

View Article Online
RSC Advances
Paper
acid nor Brønsted acidity alone may be helpful in complete with MIBK to extract remaining organic products. Collected
20,25
conversion of glucose into HMF and LA.
Indeed, a combi- organic phase was evaporated under vacuum in a rotary evap-
nation of Lewis and Brønsted acid sites may be more favorable orator for MIBK separation and crude product was weighed. The
1
than individual acid sites. In this regard, An et al. explored purity and yield of products were measured by H-NMR, using
metal sulfates such as (Al
Fe (SO , CuSO , CoSO , MgSO
ising catalysts for the synthesis of n-butyl levulinate from CH), 6.53(d, J ¼ 3.68 Hz, CH), 4.75 (s, CH
2
(SO
4
)
3
$18H
2
O, Ti(SO
4
)
2
, Co(SO
4
)
3
,
known amount of mesitylene as an internal standard.
1
2
4
)
3
4
4
4
, ZnSO
4
and K SO
2
4
) as prom-
H-NMR (CDCl
3
) HMF: d 9.55 (s, CHO), 7.24 (d, J ¼ 3.99 Hz,
13
2
–) ppm. C NMR
): d 177.68, 161.08, 123.23, 109.79, 57.073 ppm. LA:
reactions, the separation of metal salts and mineral acids from d 2.712 (t, J ¼ 12.43 Hz, CH ), 2.56 (t, J ¼ 10.96 Hz, CH ), 2.13
the product stream is a challenging task, whereas metal sup- (s, CH ) ppm. C NMR (CDCl ) LA: d 207.04 (C]O) 177.59 (C]
26
carbohydrates. However, in case of homogeneous catalyzed (CDCl
3
2
2
1
3
3
3
ported catalysts are relatively costlier which limits their appli- O, COOH), 37.8796 (CH ), 30.49 (CH ), and 29.77 (CH ) ppm.
2
2
3
1
cation. On the contrary, potash-alum (PA) is an inexpensive and
H-NMR of FL in CDCl : d 9.6447 (s, 1H), 7.68 (d, 1H), 7.24 (d,
3
readily available material that has been used for a long time in 1H), 6.59 (q, 1H).
2
7
water purication, due to its excellent internal pores. PA is
Determination of total yield and individual yield. The qual-
1
reported to have mild Lewis acidity and relatively non-corrosive itative and quantitative analysis of products were done by H
nature which makes it a promising catalyst for dehydration of NMR spectroscopic method. A known amount of mesitylene as
and C sugars into value added chemicals such as HMF and internal standard was added into the crude product mixture in
28
C
5
6
LA. Interestingly, reaction pathway for production of furfural CDCl . The yields of HMF, LA and FL were calculated from
3
from hemicellulose derived xylose and arabinose is analogous integrated values of –CHO and –CH protons of HMF, LA and
2
to reaction pathway followed in the production of HMF from three aromatic ring protons of mesitylene. The NMR yields were
glucose. Along these lines, catalysts which are effective for HMF further validated by HPLC analysis for some experiments. Agi-
production from glucose should also be effective in FL lent HPLC equipped with a C-18 reverse phase and Hi-Plex H
production from xylose. Moreover, application of PA as column, a refractive index (RI) detector and a photodiode array
a potential catalyst is important because of its low cost and (PDA) detector, was used. The PDA detector was used for anal-
environment friendliness, which essentially allows it to qualify ysis of HMF, LA and FL using water–acetonitrile as a mobile
in the category of green catalysts. Therefore, here we envisaged phase. Post this, characteristic peaks of products were identi-
the application of economical and commercially available PA as ed from their retention times. Each peak was integrated, and
a green and novel catalyst for the production of HMF, LA and FL the actual concentrations of products were calculated from their
from bio-renewable carbohydrate substrates such as glucose respective pre-calibrated plots of peak areas vs. concentrations.
and xylose, respectively. To the best of our knowledge, use of PA All the experiments were performed in triplicate and averaged to
for the synthesis of HMF, LA and FL from glucose and xylose is ensure the repeatability of data. There was less than 5% error in
reported for the rst time.
overall yield of the products.
Determination of remaining sugar. Unreacted sugar in
product mixture was calculated by phenol–sulphonic acid
method. For typical measurement, 100 mL aqueous solution in
Materials and methods
Materials
10 mL distilled water, mix. 1 mL this diluted solution in 5 mL
Glucose, fructose, xylose, and sucrose were purchased from measuring conical ask, added 0.5 mL, 5% phenol, stirred
Thomas-Baker Chemicals (India) and used as received. Methyl 30 min, then added 2.5 mL conc. H
isobutyl ketone (MIBK) was purchased from Spectrochem at 100 C, cool and analyzed by UV, spectra shown in Fig. S5.†
2
SO
4
, mix and heated 5 min
ꢀ
Chemicals, India, sodium chloride and potash-alum (PA) were
6 3
procured from local shop in Delhi. DMSO-d and CDCl were
purchased from Sigma-Aldrich, India. All the chemicals used
were of high purity (>99.9%).
Results and discussions
Preliminary experiments were carried out for dehydration of
Experimental. Effectiveness of PA as catalyst for production glucose/fructose using 1.2 mmol of PA as catalyst in a glucose
of HMF, LA and FL from bio-renewable feedstocks was tested in to PA molar ratio of 8.3, which led to 61% and 40% total known
a biphasic system (water and MIBK) by adding desired amount products yield from fructose and glucose, respectively, in 6 h at
ꢀ
of substrate, solvent and catalyst in a 100 mL hydrothermal 140 C reaction temperature. HMF was the major product (58%
reactor. In a typical experimental procedure, 10 mmol of sugar and 40% yield) from both the reactants fructose and glucose
and 1.2 mmol of PA were dispensed into the reactor containing (Table 1, entries 1 & 2) whereas levulinic acid was measured
10 mL of water + MIBK mixture (1 : 4 v/v). Prior to reaction, the minor product. It is evident that PA as a catalyst is very active
solution was stirred for 5 minutes with 600 rpm at room which can be attributed to the acidity arising from incom-
temperature to completely dissolve the reactants into the pletely coordinated aluminum atoms (three rather than four or
solvent. The reactor was then placed inside a temperature six). Moreover, it is well known that [Al(OH)
2 2
(aq) ] is an active
controlled furnace pre-heated to desired temperature range of species for the isomerization and dehydration of mono-sugar
ꢀ
29–31
1
40–190 C. Aer completion of reaction, the reactor was cooled including glucose and xylose.
down to room temperature and the organic phase was separated $12H O) in aqueous medium may have undergone a trans-
by decantation. The aqueous phase was washed three times formation to yield [Al(OH) (aq) ] and sulphate group causing
Thus, PA (KAl (SO ) -
2 4 3
2
2
2
41974 | RSC Adv., 2017, 7, 41973–41979
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
a
Table 1 Dehydration of hexose sugar (glucose/fructose) and product distribution
Yield (%)
S. No.
Substrate
Catalyst
Solvent
t (h)
Total yield (%)
HMF
LA
1
2
3
4
5
6
7
8
Fructose
Glucose
Fructose
Glucose
Glucose
Glucose
Glucose
Glucose
Glucose
Glucose
Glucose
Glucose
PA
PA
PA
PA
PA
PA
PA
PA
W + M
W + M
W* + M
W* + M
W* + M
W + T
W* + T
W* + T
W + M
W*+ M
W + M
W* + M
6
6
6
6
10
6
6
15
6
6
6
61
40
66
49
57
22
25
63
34
42
27
37
58
40
64
49
39
22
25
21
34
42
27
37
3
—
2
—
18
—
—
42
—
—
—
—
b
3+
+
9
Al + K
b
3+
+
+
10
11
12
Al + K
c
c
3+
2ꢂ
2ꢂ
Al + K + SO
4
3+
+
Al + K + SO
4
6
a
b
c
W ¼ water, M ¼ MIBK, W* ¼ water saturated with NaCl, T ¼ THF. AlCl
3
+ KCl ¼ 1.2 mmol. AlCl
3
+ KCl + Na
2
SO
4
¼ 1.2 mmol, substrate
¼
10 mmol, PA ¼ 1.2 mmol, solvent ¼ 10 mL (water + MIBK, 1 : 4).
signicant activity in PA. Other possible reason may be the extracted from the reactive phase, thereby leading to possibility
trapped water molecules inside the interlayer structure of of further reactions due to prolonged interaction between
3
2
potash alum which may have caused acidity. The results ob- products and the catalyst. On the contrary, LA selectivity
tained are in line with earlier studies by various prominent increased to 42% when experiments were performed in the
ꢀ
research groups which suggests that PA contains acidic prop- presence of water + THF solvent at 140 C temperature for 15 h
erties possibly due to presence of alumina which makes it an (Table 1, entry 8). The system consisting of water + THF as
interesting and effective catalyst for one pot conversion of solvent comparatively took longer time to desired yield (22%
ꢀ
mono-sugar conversion as compared to mineral acids, metals total yield in 6 h at 140 C) as compared to system with water +
33,34
ꢀ
salts and zeolites.
In this regard, Yao-Bing Huang et al., MIBK media (40% total yield in 6 h at 140 C). Again, formation
have recently reported 100% conversion of LA and 99.4% of undesired and byproducts could be attributed to prolonged
ꢀ
methyl levulinate yield from Al (SO ) at 110 C under micro- contact time between intermediates, reactants and the catalysts
2
4 3
35
wave-irradiation.
in water + THF phase. In contrast, use of MIBK leads to
Similarly, experiments were performed in the presence of formation of a biphasic system which immediately extracts
saturated solution of NaCl, (30 wt%) resulting an improvement products when formed; thus, minimizes the possibility of
yield of HMF (64% and 49%, from glucose and fructose further reaction or reverse mechanism. In addition, dehydra-
respectively) (Table 1, entries 3 and 4). The HMF yield improved tion of glucose or fructose in water phase for prolonged time in
from 58% to 64% and 40–49% from fructose and glucose, the presence of an acid catalyst is non-selective due to the
36
respectively by loading the sodium chloride in aqueous phase rehydration of HMF to LA and formic acid. The result obtained
Table 1, entries 1, 3 and 2, 4). Enhancement in yield could be for a longer reaction time (15 h) was found to be in line with this
attributed to the “salting-out effect” of metal salts which causes hypothesis (Table 1, entry 8). A further detailed study on other
(
36
improved partitioning of HMF into the extracting phase.
possible reasons for LA formation is undergoing which may
Eventually, the result indicates that yield of HMF can be provide a detailed insight into it in future. Nevertheless, overall
37
enhanced without using a high boiling solvent like DMSO.
Moreover, total product yield obtained from glucose further
role of PA remained instrumental in determining the total yield.
Further experiments were performed under comparable
improved from 49% to 57% when the reaction time was reaction conditions using homogeneous mixture of AlCl
3
+ KCl
increased to 10 h (Table 1, entry 5). In contrast, a swi decline in and AlCl + KCl + Na SO salts in order to develop insights into
3
2
4
HMF yield (from 49% to 39%) was measured on further increase PA catalytic activity, consequently 34% total yield was measured
in reaction time from 6 h to 10 h (Table 1, entry 5). It is possible in the presence of AlCl + KCl and 27% in the presence of AlCl +
3
3
that partial amount of HMF may have remained in the aqueous KCl + Na
phase which underwent rehydration reaction to yield LA with catalytic performances for the Al salts with other salts (KCl,
2 4
SO (Table 1, entries 9 and 11). The difference in
3
+
passage of time. Interestingly, 18% LA yield measured in pro- Na
2
4
SO ) may be due to different Brønsted acidities in the water.
longed reaction time of 10 h further supports this hypothesis.
Moreover, when experiments were repeated with NaCl saturated
Similarly, HMF yield from glucose and fructose declined water in the presence of other salts and sulphates, overall total
signicantly when experiments were performed using THF yield remained low as compared to pure PA catalyst as shown in
ꢀ
instead of MIBK as solvent at 140 C temperature for 6 h (Table Table 1 (entries 9–12). Thus, PA offered higher catalytic activity
1, entries 6 & 7). THF and water are miscible and thereby formed based on HMF yield as compared to the activity of the individual
a monophasic solution, which may be a possible cause for components mixed separately. Similarly, a blank run (without
ꢀ
reduced total yield. Moreover, HMF cannot be concurrently PA) yielded only ꢁ2% of HMF at 140 C for 6 h. Eventually,
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 41973–41979 | 41975

View Article Online
RSC Advances
Paper
glucose converted to blackish mass (in blank reaction) in 6 h at the reaction time was increased to 3 h (Table 2, entry 2). A
ꢀ
1
80 C temperature, suggesting hydrothermal conversion of further increase in reaction time to 6 h resulted in 49% total
sugar to undesired carbonaceous material. Consequently, it is yield with 100% HMF selectivity. Further experiments for longer
difficult to make a complete carbon balance of products due to reaction time (10–15 h) showed a marginal difference in total
limitations of NMR technique to identify all undesired yield (Table 2, entries 4 and 5). However, HMF selectivity
products.
decreased probably due to its rehydration to LA.
1
Fig. 1 shows a stack plot of H-NMR spectra of the crude
products obtained from reaction time variation experiments. It
shows HMF rehydration to LA and formic acid (FA) starts aer
Effect of reaction time and temperature
Table 2 shows total yield, HMF and LA yields. The trend of the 6 h. Self-polymerization or cross-polymerization of HMF with
results reveals that formation of key products depends on the glucose or fructose could be another possible reason for lower
reaction time and temperature. Dehydration of glucose for 1 h HMF yield. The signal intensity of the –COOH proton of formic
(
Table 2, entry 1) resulted in a merely 8% total yield with 100% acid (d ¼ 8.10 ppm) is very low as compared to the methylene
HMF selectivity. The total yield improved from 8% to 19% when protons of LA (2CH , d 2.7 and 2.5 ppm), although equimolar
2
amounts of LA and formic acid are expected from HMF rehy-
dration. This may be due to fact that the –COOH proton signal
Table 2 Dehydration of glucose to HMF, ring opening and formation of formic acid did not resolve well in a reaction mixture or
a
of LA at high temperature
disappeared due to its consumption and appeared in undened
side products like humins. In comparison to LA, low intensity of
FA peaks was also conrmed by the H-NMR spectroscopy NMR
%
Yield
1
ꢀ
S. No.
T ( C)
Time
% Total yield
HMF
LA (Fig. S3†). NMR spectra clearly shows that FA in the presence of
PA acts as formylating agent and formed 5-(formyloxymethyl)
furfural (FMF). Moreover, to explore the effect of reaction
1
2
3
4
5
6
7
140
140
140
140
140
160
180
1 h
3 h
6 h
10 h
15 h
6
8
19
49
57
63
62
53
8
19
49
39
21
28
15
—
—
—
18
38
temperature on total yield of the crude product and HMF yield,
ꢀ
we performed dehydration reaction at 160 C for 6 h. Where
42 signicant improvement in total yield was observed (62%).
34 However; HMF yield decreased to 28% (Table 2, entry 6),
6
38
possibly due to subsequent conversion of HMF into LA. Further
a
ꢀ
Other reaction condition: glucose ¼ 10 mmol (1.8 g), PA ¼ 1.2 mmol, increase in the reaction temperature to 180 C resulted in
solvent ¼ 10 mL (water + MIBK, 1 : 4).
a decrease in HMF yield (15%), but an enhanced yield of LA
1
ꢀ
Fig. 1 H NMR in CDCl
3
, aqueous phase PA catalyzed glucose dehydration in a stirred, oil-bath reflection reaction at 140 C. Other reaction
condition: glucose ¼ 10 mmol (1.8 g), PA ¼ 0.6 g, solvent ¼ 10 mL (1 : 4, water + MIBK). In this plot, X-axis of 3 h, move 0.2 ppm right, 10 h
1
0
.2 ppm left, 15 h, 0.4 ppm left, with respect to 1 h as standard parameter, for clarity of peak H-NMR analysis.
41976 | RSC Adv., 2017, 7, 41973–41979
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
38%) (Table 2, entry 7). One possible reason for the decrease in
RSC Advances
(
HMF yield is likely due to rehydration of HMF into LA. In
addition, loss of HMF yield can also be attributed to formation
of undesired products and humins at higher temperature.
Thus, optimum condition for higher HMF production (49%
yield) from glucose in the presence of PA catalysts is 6 h at
ꢀ
1
40 C reaction temperature whereas optimum conditions for
ꢀ
higher levulinic acid yield (42% yield) is 15 h at 140 C reaction
temperature.
The effect of PA loading. Fig. 2 represents effect of catalyst
ꢀ
loading on overall yield of HMF and LA at 140 C reaction
temperature for 6 h. It was observed that catalyst concentration
has played a key role in determining HMF and LA yield
respectively. PA concentration played a key role in the conver-
sion of glucose. At a PA/glucose weight ratio of 1 : 6, 28% HMF
yield was obtained which further enhanced to 49% at PA to
glucose weight ratio of 1 : 3. However, as the catalyst to glucose
ratio increased (1 : 1), the HMF yield and selectivity decreased Fig. 3 Reuse of PA catalyst, for the transformation of glucose to HMF,
reaction condition: glucose ¼ 10 mmol, PA ¼ 1.2 mmol, solvent ¼
in the form LA. However, total combined yield (HMF, 40% + LA,
ꢀ
1
0 mL (1 : 4, water + MIBK), at 140 C.
1
3%) increased to 63%. At high catalyst to substrate loading
ꢀ
(1 : 1) at 140 C for 6 h, rehydration of HMF occurred in the form
of LA, results shown in Fig. 3. This behavior can be attributed

ve catalytic cycles to ensure long term sustainability of the
that 1 : 3 catalyst to glucose ratio is sufficient for selective
ꢀ
process. The total conversion of glucose from each cycle is
shown in Fig. 3. The results indicate that total conversions of
glucose slightly decrease. This suggests that the catalyst activity
decreases slowly over cycles due to an interaction of side
products humins with active metal species Al . The color of the
solution turned from colorless (in the 1 cycle) to sticky-
conversion of HMF at the lower temperature (140 C) and
time (6 h).
3
+
Catalyst reusability
st
In order to study the catalyst's reusability, experiments were
th
ꢀ
brownish (in the 5 cycle) due to accumulation of humins in
the aqueous phase.
performed at optimal reaction conditions (140 C and 6 h). The
result of the rst cycle is discussed in earlier section and
tabulated in Table 1, entry 4. Post this, spent aqueous phase was
st
Dehydration mechanism
collected aer separation of the organic phase from the 1
nd
cycle. 2 reaction cycle was started aer addition of 10 mmol of
The dehydration of glucose and fructose into HMF precedes
either via cyclic or acyclic pathways. Overall, direct trans-
glucose and 8 mL fresh MIBK, in the spent aqueous phase.
Moreover, no fresh PA catalyst was added to compensate any
loss of the catalyst in the rst run. This process was repeated till
22
formation of glucose into HMF involves two steps: (1) isomeri-
zation of glucose (glucopyranose) to fructose (fructofuranose),
and (2) dehydration of fructofuranose (loss of 3 mol of water) to
HMF. It is well known that Gibbs free energy of fructose inter-
action with DMSO causes release of rst water molecule from b-
D-fructofuranoses, whereas the interaction through H atom of
-
1OH and -4OH with DMSO causes release of second and third
39
water molecules, respectively. In our earlier study, we have
proposed that glucose conversion proceeds via 1,6-anhydro-b-D-
glucopyranose (AGP), 1,6-anhydro-b-D-glucofuranose (AGF) and
1
,6-anhydro-a-D-fructofuranose (AFF) intermediates in water
20
3+
(Scheme 1). However, due to the presence of Al and Brønsted
acidity in the investigated reaction, an aluminate-mediated
rearrangement of glucose to fructose occurs through forma-
3
+
tion of a furanoid fructose–aluminate complex, where Al
catalyze the mutarotation of a-and b-glucopyranose isomers.
40
A
similar mechanism has been suggested by other prominent
research groups working in the area of biomass conversion and
41
related catalysis. Thus, we propose a similar mechanism for
experiments performed in the presence of Al containing PA
catalyst. Lewis acidity of PA is responsible for the isomerization
of glucose to fructose, whereas Brønsted acidity facilitates for
Fig. 2 Effect of catalyst loading on the dehydration of glucose at
ꢀ
1
40 C for 6 h.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 41973–41979 | 41977

View Article Online
RSC Advances
Paper
Scheme 1 Reaction pathway for catalytic conversion of glucose into HMF & LA.
42
the dehydration isomerized intermediates to HMF followed by temperatures can be attributed to the entropy effect.
HMF rehydration into LA and FA at a prolonged reaction time. Furthermore, effect of reaction time on FL yield was investi-
ꢀ
gated by performing the experiments at 180 C for longer time
Dehydration of xylose to furfural (FL)
(10 h), however, very less improvement in furfural yield (51%)
was observed.
Since xylose to furfural (FL) conversion follows a similar reac-
tion pathway to that of glucose to HMF, experiments were per-
formed using PA for conversion of xylose into FL. The
preliminary experiments for xylose conversion were performed
in biphasic solvent (water + MIBK). 10 mmol of xylose in the
Techno-economic assessment of carbohydrate conversion to
value added chemicals
Carbohydrate based chemicals synthesis and use have got
signicant attention in recent years due to its carbon neutral
ability and environmental benets. However, common imple-
mentation of carbohydrate based chemicals mainly depends on
the cost competitiveness with fossil based chemicals and fuels
and due to high cost of catalyst, carbohydrate based chemicals
are not attractive choice for industry. PA has great potential to
replace the high cost heterogeneous materials and best alter-
native of mineral acid based catalyst. PA alum is cheapest
materials (>1$ per kg) can be utilized in the large scale
production of furfural (5–9$ per kg), HMF (ꢁ300 $ per kg) and
LA (40–70$ per kg), and may be able to lower-down the price and
competitive with fossil fuel based chemicals.
ꢀ
presence of 3.1 mmol PA catalyst at 140 C yielded 5% furfural
in 6 h. On increasing the reaction temperature from 140 to
ꢀ
1
60 C, a signicant improvement in FL yield (27%) was
observed (as shown in Fig. 4) in 6 h. The increase in FL yield at
ꢀ ꢀ
1
80 C and 190 C resulting into 47% and 55% FL yield
respectively, in 6 h reaction time. Increase in FL yield at higher
Conclusion
The use of potash alum as catalyst for the conversion of carbon
dioxide neutral biomass fraction for the production of furfural
and HMF is a vital alternative to hazardous mineral acid and
costly nanoparticles solid acid. The use of potash alum is the
most inexpensive and green catalyst for dehydration of carbo-
hydrate fraction. The results showed that potash alum (PA) is
a promising catalyst for catalytic conversion of carbohydrates
and xylose into HMF and FL, respectively. Under optimum
reaction conditions, glucose and fructose yielded 40% and 58%
Fig. 4 Influence of the reaction temperature in the dehydration of D-
xylose in the presence of PA as catalyst in biphasic solvents. Other
reaction condition: xylose ¼ 10 mmol, PA as catalyst ¼ 1.2 mmol, time
ꢀ
¼
6 h, solvent (water + MIBK) ¼ 10 mL, 1 : 4, *yield ¼ molar yield.
HMF, respectively, in 6 h at 140 C temperature in the presence
41978 | RSC Adv., 2017, 7, 41973–41979
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
of 1.2 mmol PA. In biphasic solvent containing NaCl saturated 17 A. Bohre, S. Dutta, B. Saha and M. M. Abu-Omar, ACS
water as a reactive phase, HMF yield improved to 49% and 64% Sustainable Chem. Eng., 2015, 3, 1263–1277.
from glucose and fructose, respectively, under identical reac- 18 J. P. Lange, E. van der Heide, J. van Buijtenen and R. Price,
tion parameters. Similarly, conversion of xylose in the presence
of 1.2 mmol PA in 6 h at 190 C resulted in 55% FL. Overall, PA 19 G. M. G. l. Maldonado, R. S. Assary, J. A. Dumesic and
ChemSusChem, 2012, 5, 150–166.
ꢀ
shows signicant promises for catalytic conversion of carbo-
hydrates and xylose. The basic strategy, thereby, to use cheap 20 D. Gupta, S. Kundu and B. Saha, RSC Adv., 2016, 6, 100417–
and inexpensive PA catalyst as a replacement for mineral acids, 100426.
metal chlorides, ionic liquids, strong organic acids, strong 21 V. Choudhary, R. I. Burnett, D. G. Vlachos and S. I. Sandler,
L. A. Curtiss, Energy Environ. Sci., 2012, 5, 8990–8997.
Lewis and Brønsted acids which are comparatively expensive
and pose hazardous effect to the environment.
J. Phys. Chem. B, 2012, 116, 5116–5120.
22 S. Dutta, S. De, M. I. Alam, M. M. Abu-Omar and B. Saha,
J. Catal., 2012, 288, 8–15.
2
3 H. Zhao, J. E. Holladay, H. Brown and Z. C. Zhang, Science,
Conflicts of interest
2007, 316, 1597–1600.
There are no conicts to declare.
24 J. N. Chheda, Y. Roman-Leshkov and J. A. Dumesic, Green
Chem., 2007, 9, 342–350.
2
2
2
5 H. Yan, Y. Yang, D. Tong, X. Xiang and C. Hu, Catal.
Commun., 2009, 10, 1558–1563.
6 R. An, G. Xu, C. Chang, J. Bai and S. Fang, J. Energy Chem.,
Acknowledgements
Dinesh Gupta is grateful to Science and Engineering Research
Board (SERB), Government of India for National Post Doctoral
Fellowship (PDF/2016/000458).
2017, 26, 556–563.
7 K.-N. Oo, K.-S. Aung, M. Thida, W.-W. Khine, M.-M. Soe and
T. Aye, J. Diarrhoeal Dis. Res., 1993, 172–174.
References
28 G. S. S. Kumar and S. Kumaresan, J. Chem. Sci., 2012, 124,
57–863.
9 Y. Yang, C.-w. Hu and M. M. Abu-Omar, Green Chem., 2012,
4, 509–513.
0 Y. Yang, C.-W. Hu and M. M. Abu-Omar, ChemSusChem,
012, 5, 405–410.
1 Y. Yang, C. Hu and M. M. Abu-Omar, Bioresour. Technol.,
012, 116, 190–194.
2 H. Sachdeva, D. Dwivedi and R. Saroj, Sci. World J., 2013, 22,
953–4963.
8
1
2
3
4
F. W. Lichtenthaler, Acc. Chem. Res., 2002, 35, 728–737.
C. J. Campbell and J. H. Laherrare, Sci. Am., 1998, 278, 60–65.
R. A. Sheldon, Green Chem., 2016, 18, 3180–3183.
H. Olcay, A. V. Subrahmanyam, R. Xing, J. Lajoie,
J. A. Dumesic and G. W. Huber, Energy Environ. Sci., 2013,
2
3
3
3
3
3
3
1
2
6
, 205–216.
M. J. Climent, A. Corma and S. Iborra, Green Chem., 2014, 16,
16–547.
A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411–
502.
O. Casanova, S. Iborra and A. Corma, ChemSusChem, 2009, 2,
138–1144.
2
5
6
7
5
4
3 J. Azizian, A. A. Mohammadi, A. R. Karimi and
M. R. Mohammadizadeh, J. Org. Chem., 2005, 70, 350–352.
4 L. Zhou, H. Zou, J. Nan, L. Wu, X. Yang, Y. Su, T. Lu and J. Xu,
Catal. Commun., 2014, 50, 13–16.
2
1
8
9
M. Mascal and S. Dutta, Green Chem., 2011, 13, 40–41.
E. Ahmad, M. I. Alam, K. K. Pant and M. A. Haider, Green
Chem., 2016, 18, 4804–4823.
5 Y.-B. Huang, T. Yang, B. Cai, X. Chang and H. Pan, RSC Adv.,
2016, 6, 2106–2111.
3
3
6 B. Saha and M. M. Abu-Omar, Green Chem., 2014, 16, 24–38.
10 S. Van de Vyver, J. Thomas, J. Geboers, S. Keyzer, M. Smet,
˜
7 Y. RomA¡n-Leshkov and J. A. Dumesic, Top. Catal., 2009, 52,
W. Dehaen, P. A. Jacobs and B. F. Sels, Energy Environ. Sci.,
2
97–303.
8 S. De, S. Dutta and B. Saha, ChemSusChem, 2012, 5, 1826–
833.
9 M.-F. He, H.-Q. Fu, B.-F. Su, H.-Q. Yang, J.-Q. Tang and
C.-W. Hu, J. Phys. Chem. B, 2014, 118, 13890–13902.
2
011, 4, 3601–3610.
1 M. Balakrishnan, E. R. Sacia and A. T. Bell, Green Chem.,
012, 14, 1626–1634.
2 B. Saha, D. Gupta, M. M. Abu-Omar, A. Modak and
A. Bhaumik, J. Catal., 2013, 299, 316–320.
3
3
4
4
4
1
1
1
1
1
2
0 J. Tang, L. Zhu, X. Fu, J. Dai, X. Guo and C. Hu, ACS Catal.,
3 A. Dutta, D. Gupta, A. K. Patra, B. Saha and A. Bhaumik,
ChemSusChem, 2014, 7, 925–933.
2
016, 7, 256–266.
1 J. Tang, X. Guo, L. Zhu and C. Hu, ACS Catal., 2015, 5, 5097–
103.
2 A. S. Dias, M. Pillinger and A. A. Valente, J. Catal., 2005, 229,
14–423.
4 M. G. Mazzotta, D. Gupta, B. Saha, A. K. Patra, A. Bhaumik
and M. M. Abu-Omar, ChemSusChem, 2014, 7, 2342–2350.
5 D. Gupta and B. Saha, Catal. Commun., 2017, 100, 206–209.
6 A. Bohre, B. Saha and M. M. Abu-Omar, ChemSusChem, 2015,
5
1
1
4
8
, 4022–4029.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 41973–41979 | 41979