From biomass to chemicals: Synthesis of precursors of biodegradable surfactants from 5-hydroxymethylfurfural

Article abstract of DOI:10.1002/cssc.201200513

The selective acetalization of 5-hydroxymethylfurfural (HMF) with long-chain alkyl alcohols has been performed to obtain precursors of molecules with surfactant properties. If direct acetalization of HMF with n-octanol is performed in the presence of strong acids (homogeneous and heterogeneous catalysts), an increase in etherification versus acetalization occurs. Beta zeolite catalyzes both reactions. However, if the acidity of a zeolite (Beta) was controlled by partial exchange of H⁺ with Na⁺, the dioctyl acetal of HMF can be achieved in 95 % yield by transacetalization. It is possible to achieve a high yield in a very short reaction time through a two-step one-pot process, which includes the synthesis of the dimethyl acetal of HMF followed by transacetalization with n-octanol. The one-pot process could be extended to other alcohols that contain 6-12 carbon atoms to afford 87-98 % yield of the corresponding dialkyl acetal with a selectivity higher than 96 %. The optimized catalyst with an adequate Na content (1.5NaBeta) could be recycled without loss of activity or selectivity. From HMF to surfactants: Precursors of surfactant molecules are obtained by a two-step one-pot selective acetalization and transacetalization of 5-hydroxymethylfurfural with long-chain alkyl alcohols by controlling the zeolite acidity and polarity. Copyright

Full text of DOI:10.1002/cssc.201200513

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201200513
From Biomass to Chemicals: Synthesis of Precursors of
Biodegradable Surfactants from 5-Hydroxymethylfurfural
K. S. Arias,[a] Saud I. Al-Resayes,[b] Maria J. Climent,*[a] Avelino Corma,*[a] and Sara Iborra[a]
The selective acetalization of 5-hydroxymethylfurfural (HMF)
with long-chain alkyl alcohols has been performed to obtain
precursors of molecules with surfactant properties. If direct
acetalization of HMF with n-octanol is performed in the pres-
ence of strong acids (homogeneous and heterogeneous cata-
lysts), an increase in etherification versus acetalization occurs.
Beta zeolite catalyzes both reactions. However, if the acidity of
a zeolite (Beta) was controlled by partial exchange of H+ with
Na+, the dioctyl acetal of HMF can be achieved in 95% yield
by transacetalization. It is possible to achieve a high yield in
a very short reaction time through a two-step one-pot process,
which includes the synthesis of the dimethyl acetal of HMF fol-
lowed by transacetalization with n-octanol. The one-pot pro-
cess could be extended to other alcohols that contain 6–12
carbon atoms to afford 87–98% yield of the corresponding di-
alkyl acetal with a selectivity higher than 96%. The optimized
catalyst with an adequate Na content (1.5NaBeta) could be re-
cycled without loss of activity or selectivity.
Introduction
5-Hydroxymethylfurfural (HMF) is considered as an important
platform molecule for biorenewable fuel[1] and chemical pro-
duction[2] as it can be obtained by the dehydration of fructose,
glucose, and cellulose.[2–4] HMF is a versatile molecule that can
be converted into several derivatives with multiple applica-
tions, for example, pharmaceuticals, antifungals, and polymer
precursors. Recently, some transformations of HMF into mono-
mers able to replace terephthalic, isophthalic, and adipic acids
in the manufacture of polyamides, polyesters, and polyur-
ethanes[5] have been performed by our group by using hetero-
geneous catalysts.[6] With the aim to find other applications for
HMF derivatives, we have investigated the conversion of this
molecule into a new class of compounds with potential surfac-
tant applications.
hydroxyl group to form a carboxylic group or its salts
(Scheme 1). Although HMF derivatives have been mentioned
as precursors of surfactants, we did not find any reports of
such in the literature.
The acetalization of carbonyl groups can be performed by
the direct reaction of carbonyl compounds with alcohols or by
the transacetalization of dimethyl or diethyl acetals with other
alcohols in the presence of acid catalysts. A wide variety of
heterogeneous acid catalysts, such as montmorillonite and ion-
exchanger resins,[7] titania silicalite,[8] mesoporous materials,[9]
and zeolites[10] have been reported as solid catalysts that per-
form acetalization reactions with excellent results. Previously
we have shown that zeolites and mesoporous aluminosili-
cates[11] are suitable acid catalysts to obtain nonionic surfac-
tants, such as alkylglucosides, by the selective acetalization of
glucose with fatty alcohols. Therefore, for the selective acetali-
zation of HMF, acid zeolites appear to be promising catalysts
as, in addition to the general advantages associated with the
use of solid catalysts, they can be synthesized with different
crystalline structures, pore sizes, acid strength distributions,
and hydrophobicity. Notably, HMF is a bifunctional molecule
that bears alcohol and aldehyde functionalities, which are both
able to react with alcohols in the presence of acid catalysts to
give ethers and acetals, respectively. Therefore, it would be of
interest to be able to tailor the solid catalyst selectively to
direct the reaction to ethers or acetals, and we will present
how this can be achieved by using zeolite catalysts.
Surfactants are commonly used in detergents, pharmaceuti-
cals, cosmetics, and food. They form micelles and vesicular mi-
croaggregates, the properties of which depend on the compo-
sition and structure of the polar head groups and the length
and shape of the hydrophobic hydrocarbon component. With
this in mind, we have studied the selective acetalization of the
carboxaldehyde group of HMF with fatty alcohols by using
zeolites as acid catalysts. This type of molecule can be trans-
formed into surfactants by subsequent oxidation of the free
[a] K. S. Arias, M. J. Climent, A. Corma, S. Iborra
Instituto de Tecnologꢀa Quꢀmica (UPV-CSIC)
Universitat Politꢁcnica de Valꢁncia
Avda dels Tarongers s/n, 46022, Valencia (Spain)
Fax: (+34)963877809
Results and Discussion
E-mail: acorma@itq.upv.es
[b] S. I. Al-Resayes
Taking into account the dimensions of the desired product 5-
(dioctoxymethyl)-2-hydroxymethylfurane (3) (Scheme 2), we
started our study by performing the acetalization of HMF with
n-octanol in the presence of structured micro- and mesopo-
Chemistry Department, College of Science
King Saud University, B.O. BOX.2455 Riyadh 11451 (Saudi Arabia)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201200513.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 123 – 131 123

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Scheme 1. Synthesis of surfactant molecules from HMF.
rous acid catalysts, such as a large pore tridirectional (Beta)
and a monodirectional (Mordenite) zeolite, as well as a mesopo-
rous aluminosilicate (MCM-41) (Table 1). Thus, when the acetali-
zation reaction of HMF was performed with n-octanol in an n-
octanol/HMF molar ratio of 40 in the presence of these cata-
lysts at 658C, it was observed that although the Beta zeolite
and the mesoporous MCM-41
acid sites in the homogeneous catalyst and the stronger ones
that exist in the aluminosilicate catalysts mainly favor competi-
tive etherification reactions of the hydroxymethyl group (to
give 4) as well as the dimerization of HMF (to give 6) and its
subsequent reactions with n-octanol (to give 7 and 8).
gave rise to high conversions of
HMF, mordenite gave rise to
a lower conversion, which could
be attributed to fast poisoning
by pore blocking. However,
a complex mixture of products
was always formed (Scheme 2),
and the yield of 3 was very low
in all cases (Table 2). The main
byproducts observed were the
ether 5-(octyloxymethyl)furfural
(4) and its corresponding dioctyl
acetal 5. However, compounds
that came from the dimerization
Scheme 2. Possible routes for the formation of 3 and byproducts through the direct acetalization of HMF with n-
of HMF such as 5,5’-[oxy-bis(me-
thylene)]bis-(2-furfural) (6) and
the corresponding mono- and
bis(dioctyl) acetals 7 and 8 were
also obtained (Table 2). As
shown in Figure S1, 3, 4, and 6
have a primary character, where-
as 5, 7, and 8 are secondary
products. Octyl levulinate, ob-
tained from ring opening by the
hydrolysis of HMF followed by
esterification in acidic media,
was not detected. Scheme 2
presents the different possible
routes for the formation of the
byproducts detected. Additional-
ly, if a solvent such as trifluoroto-
luene (TFT) was used, no im-
provement in selectivity to 3
was observed (Table 2).
octanol.
Table 1. Main catalyst characteristics.
À1
Catalyst
Si/Al
BET
Pore volume
Acidity[a] [mmolpyridine gcat.
2508C
]
[m2 gÀ1
]
[cm3 gÀ1
]
1508C
3508C
B[b]
L[c]
B[b]
L[c]
B[b]
L[c]
HBeta
12
12
12
12
12
12
12
15
10
602
550
556
572
556
527
536
1100
550
0.36
0.35
0.34
0.36
0.34
0.35
0.36
0.94
0.40
65
51
41
40
33
22
0
69
35
28
29
27
14
0
58
41
35
27
23
15
0
56
35
27
24
21
10
0
25
25
23
20
12
3
29
28
23
20
16
9
0
34
28
0.26NaBeta
0.49NaBeta
0.58NaBeta
1NaBeta
1.5NaBeta
3.1NaBeta
MCM-41
0
4
19
67
62
25
5
54
46
25
mordenite
29
[a] Acidity at different temperatures calculated from the extinction coefficients given in Ref. [15]. [b] Brønsted
acidity. [c] Lewis acidity.
Table 2. Results of the direct acetalization of HMF with n-octanol in the presence of acid catalysts.[a]
If the reaction was performed
using p-toluenesulfonic acid
Catalyst
Initial rate
[molminÀ1 gÀ1
Conversion
[%]
Yield [%]
Selectivity
to 3 [%]
]
3
4
5
6
7
8
(PTSA) as
a
homogeneous
HBeta
0.40
0.12
0.02
0.35
158
93
97
46
100
93
14
13
5
10
6
20
17
10
24
14
13
16
7
16
24
15
12
7
14
11
26
33
13
26
31
5
6
4
10
7
15
HBeta[b]
Mordenite
MCM-41
PTSA[c]
13
11
10
6
Brønsted acid catalyst, the distri-
bution of products obtained was
similar to that achieved with
Beta zeolite and MCM-41. The
results indicate that, under these
reaction conditions, the strong
[a] Reaction conditions: 658C, OctOH/HMF=40, catalyst (51 mg, 40 wt% with respect to HMF), 18 h. [b] TFT
(5 mL) was used as a solvent. [c] PTSA (0.1 wt% with respect to HMF) after 3 h.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 123 – 131 124

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Figure S2 presents the selectivity to 3 versus conversion for
the different catalysts tested.
As etherification is a more demanding reaction than acetali-
zation from the point of view of the acid strength, one way to
increase the selectivity to 3 would be to decrease the number
of strong Brønsted acid sites, which could be achieved by
a partial exchange of protons with Na+ in the zeolite frame-
work. To investigate this, a series of Na-exchanged Beta zeo-
lites with Na contents from 0.26–3 wt% (calculated as Na2O)
were prepared and tested in the reaction. Figure 1 displays the
Figure 2. Acetalization of HMF with n-octanol in the presence of 1.0NaBeta
&
*
^
.
zeolite. HMF , 3 , 4
Figure 1. IR spectra of Beta zeolites with different Na contents. The B and L
bands arise from Brønsted and Lewis acid sites determined by the adsorp-
tion–desorption of pyridine at 1508C. Additionally, adsorption bands at 1592
and 1442 cmÀ1 correspond to the interaction of Na+ with pyridine.
IR spectra of pyridine adsorbed on the Beta samples with dif-
ferent Na contents. B and L correspond to pyridine protonated
by Brønsted acids and pyridine coordinated with Lewis acids,
respectively. Two absorption bands at 1592 and 1442 cmÀ1
were observed that correspond to the interactions of Na+ with
pyridine, which increase in intensity with increasing Na con-
tent.[12] Table 3 and Figures 2 and 3 show the catalytic results
obtained with these materials. These results show that an in-
creased Na content leads to an increased selectivity to 3. More
specifically, with Beta zeolite samples with a Na content higher
than 1 wt%, the formation of ethers, mainly 6, was drastically
suppressed. Moreover, if all of the acid sites of the zeolite were
exchanged by Na+ (3.1NaBeta, see Table 1), the catalytic activi-
ty was totally suppressed.
Figure 3. Acetalization of HMF with n-octanol in the presence of a) fresh
~
.
&
*
1.5NaBeta zeolite and b) 1.5NaBeta after Soxhlet extraction. HMF , 3 , 5
The results presented in Table 3 and Figure S3 show that
a Beta zeolite with a Na content of 1.5 wt% is a very selective
catalyst to perform the acetalization. However, the reaction
stops after 8 h and the final yield obtained was 63%. The Beta
zeolites with Na contents of 1 and 1.5 wt% show a similar ki-
netic behavior (initial reaction rate) and both catalysts deacti-
vate quickly (Figures 2 and 3a). This deactivation could be as-
sociated with a strong adsorption of the products onto the
catalyst surface at lower reaction temperatures. To confirm
this, the IR spectrum of 1.5NaBeta after the acetalization reac-
tion was obtained, and a wide absorption band at 3500 cmÀ1
(associated with hydroxyl groups) and bands at 2900 cmÀ1
(corresponding to CÀH vibrations associated with the n-octa-
nol alkyl chain) were observed. If the used 1.5NaBeta) was
Soxhlet extracted by using acetonitrile as the solvent, n-octa-
Table 3. Results of the direct acetalization of HMF with n-octanol in the
presence of Beta zeolites with different Na contents.[a]
Catalyst
Conversion
[%]
Yield [%]
Selectivity
to 3 [%]
3
4
5
6
7
8
0.26NaBeta
0.49NaBeta
1.0NaBeta
1.5NaBeta
3.1NaBeta
96
92
88
63
0
25
27
76
66
0
24
25
12
0
35
28
0
0
0
11
12
0
0
0
0
0
0
0
0
0
0
0
0
0
27
29
85
100
0
0
[a] Reaction conditions: HMF (1 mmol, 126 mg), n-octanol (40 mmol),
Beta zeolite (51 mg, 40 wt% catalyst with respect to HMF), 24 h, 658C.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 123 – 131 125

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
nol (11 wt%) was the only compound detected by GC–MS in
the organic phase. However, after the extraction process,
4 wt% of organic material remained on the solid catalyst (ther-
mogravimetric analysis). The acetalization reaction of HMF with
n-octanol was performed in the presence of the catalyst after
Soxhlet extraction, and the results show a decrease of the ini-
tial rate of reaction (0.20 molminÀ1 gÀ1) compared with that of
the fresh catalyst (0.43 molminÀ1 gÀ1), although a similar yield
(60%) of 3 was obtained if the reaction time was prolonged
(Figure 3b). These results confirm that an important part of the
catalyst deactivation is reversible and is mainly a result of the
adsorption of n-octanol onto the surface of the catalyst. The
activity and selectivity to 3 were totally recovered when the
used catalyst was calcined at 5808C for 3 h.
From an operational point of view, the amounts of reactants
and products adsorbed could be diminished by performing
the reaction at a higher temperature. Indeed, practically full
conversion was achieved at temperatures >1008C (Table 4), al-
though under these reaction conditions, the selectivity to 3 de-
creases and competitive etherification reactions are promoted.
Notably, 5 (52%) is the major product obtained at 1208C
(Scheme 2).
Table 5. Main characteristics of Beta zeolites prepared in hydroxyl (B) and
fluoride (F) media.
À1
Catalyst
Si/Al BETarea Water
Brønsted acidity [mmolpyridine gcat.
]
[m2 gÀ1
]
[%]
1508C
2508C
3508C
Beta15(B) 15
Beta25(B) 25
Beta12(F) 12
Beta25(F) 25
Beta50(F) 50
480
465
400
470
475
12
90
44
71
46
33
69
20
58
41
25
45
19
40
27
21
10.9
11.5
6.8
5.0
thesized in fluoride media contain a smaller amount of silanols
(see the band at 3745 cmÀ1 in Figure 4).
The resultant samples were tested for the acetalization of
HMF with n-octanol, and the results are given in Table 6.
By analysis of these results, it can be seen that, regardless of
the nature of the sample, selectivity to 3 is low for the Na-free
HBeta sample.
When the acetalization was performed by using Beta zeolite
synthesized in OHÀ media, such as Beta15(B) and Beta25(B),
which have large amounts of silanol groups, a higher activity
was obtained for the zeolite with the highest Si/Al ratio
(Beta25(B)), which indicates that some hydrophobicity is re-
quired to obtain a good yield and selectivity to 3 (Figure 5a
and b). However, if the more hydrophobic Beta zeolite samples
(i.e., those free of connectivity defects, Beta(F)), were used, the
results in Table 6 show that the
These results indicate that with zeolites, or at least with Beta
zeolite, it is difficult to obtain high yields of the acetal of HMF
with longer alkyl chain alcohols, even if zeolite acidity is opti-
mized. It appears that n-octanol is strongly adsorbed at lower
most active catalyst has a Si/Al
Table 4. Influence of temperature in the direct acetalization of HMF with n-octanol using 1.5NaBeta zeolite as
ratio of 12, which represents
a compromise between a large
amount of acid sites and polarity
(Figure 5c and d). Additionally,
the reuse of Beta15(B) and
Beta12(F) zeolites showed that
some deactivation of the cata-
lysts occurs during the reaction,
however, the selectivity to 3 is
maintained. For instance, if
Beta12(F) is used in a second
cycle, the conversion decreases
a catalyst.[a]
T
[8C]
Initial rate 3
[molminÀ1 gÀ1
Conversion
[%]
Yield [%]
5
Selectivity
to 3[b] [%]
]
3
4
6
7
8
65
80
100
120
0.51
0.59
0.74
0.82
63
65
94
99
63
60
49
21
0
0
2
25
52
0
0
0
0
0
2
2
2
0
0
0
0
100
92
98
3
17
24
72
[a] Reaction conditions: HMF (126 mg, 1 mmol), n-octanol (40 mmol), 1.5NaBeta zeolite (51 mg, 40 wt% catalyst
with respect to HMF), 5 h. [b] Selectivity to 3 at 65% conversion of HMF.
temperatures and competes with HMF for the zeolite adsorp-
tion sites, which leads to a reversible deactivation of the
catalyst.
At this point we thought that, in principle, it could be possi-
ble to modify the relative adsorption of HMF and n-octanol by
modification of the surface polarity of the zeolite. To do this,
we have prepared two series of zeolite catalysts. In the first
series, Beta zeolite samples were synthesized in fluoride media
with different framework Si/Al ratios (Table 5). It is well known
that synthesis in fluoride, instead of OHÀ media, produces sam-
ples with a smaller amount of silanols, which leads to less
polar Beta samples.[13] Moreover, for a given synthetic method,
the higher the framework Si/Al ratio the less polar the zeolite
will be. The second series of zeolite catalysts was prepared in
Figure 4. The OH stretching region of the IR spectra of Beta samples after
heating at 4008C under vacuum. The vibration band at 3745 cmÀ1 is as-
signed to the OH stretch of the silanol group.
OHÀ media with different framework Si/Al ratios. IR spectra of
two representative samples confirm that the Beta samples syn-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 123 – 131 126

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
possible to maximize the yield of either 3 or 5. However, de-
spite the large ratios of n-octanol to HMF used and the optimi-
zation of acid sites and polarity, we could not obtain, by direct
acetalization, yields of 3 greater than 70%.
Table 6. Results of the acetalization of HMF with n-octanol by using Beta
zeolites with different hydrophobic/hydrophilic characters.[a]
Catalyst
t
Conversion
[%]
Yield [%]
Selectivity
[h]
3
4
5
to 3 [%]
HBeta
0.5
8
0.5
8
0.5
8
0.5
8
0.5
8
0.5
8
66
93
69
80
70
70
70
75
68
72
57
68
11
13
67
40
70
67
70
70
66
48
55
35
18
23
–
18
–
–
–
1
1
5
11
2
20
–
3
–
4
1
13
1
17
14
97
50
100
96
100
93
97
67
96
Transacetalization
Beta15(B)
Beta25(B)
Beta12(F)
Beta25(F)
Beta50(F)
According to the results presented above and taking into ac-
count that transacetalization reactions are thermodynamically
and kinetically favored with respect to direct acetalizations, we
have explored an alternative route to obtain 3, which involves
the preparation of 5-(dimethoxymethyl)-2-hydroxymethylfur-
ane (9) as the first step followed by transacetalization with n-
octanol.
12
1
16
17
52
Compound 9 was prepared from HMF and MeOH by using
1.5NaBeta zeolite as the catalyst. The reaction was performed
at 658C, and after 2.5 h, 99% selectivity to 9 with 99% conver-
sion of HMF was obtained. Only traces of 9 were detected in
the reaction media. Then, starting with 9, we performed the
transacetalization reaction with n-octanol in the presence of
Beta zeolites with different Na contents (Table 7). The reactions
were performed with the previously optimized molar ratio of 9
to n-octanol of 1:3.7, and 15 wt% of catalyst (with respect to
9) at 658C, and MeOH was removed by distillation by using
a Dean–Stark apparatus.
[a] Reaction conditions: HMF (1 mmol, 126 mg), n-octanol (40 mmol),
Beta zeolite (51 mg, 40 wt% catalyst with respect to HMF), 8 h, 658C.
from 75 to 50%. However, when the sample is calcined at
5808C the initial activity is restored (Figure S4).
In conclusion, it appears that by controlling the acidity, po-
larity, and reaction temperature, it is possible to direct the re-
action between HMF and n-octanol to 3 or the ethers 4 and 5.
All of the products formed can act as surfactants though their
properties will be different. Therefore, by using Beta zeolite
synthesized with an adequate acidity and polarity it should be
Analysis of the results in Table 7 shows that HBeta zeolite
gives a low selectivity to 3 and promotes mainly etherification
Figure 5. Results of acetalization reaction between HMF and n-octanol that used Beta zeolite samples prepared in hydroxyl (B) and fluoride (F) media:
~
&
*
^
a) Beta15(B), b) Beta25(B), c) Beta12(F), d) Beta50(F). HMF , 3 , 4 , 5 , othersꢁ.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 123 – 131 127

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Table 7. Results of the transacetalization of 9 with n-octanol in the pres-
ence of Beta zeolites with different Na contents.[a]
Catalyst
Conversion
of 9 [%]
Yield [%]
Selectivity
others[b] to 3 [%]
4
5
1
3
HBeta
100
100
100
100
100
100
9
13 10 32 44
2
2
3
3
3
3
1
44
62
66
95
95
95
7
0.26NaBeta
0.58NaBeta
1.5NaBeta
1.5NaBeta[c]
1.5NaBeta-F
3NaBeta[d]
7
5
0
0
0
1
12 17 62
9
0
0
0
0
17 66
2
2
2
6
95
95
95
1
[a] Reaction conditions: Molar ratio of 9/n-octanol 1:3.7, catalyst (26 mg,
15 wt%), 658C, 30 min, using a Dean–Stark apparatus and N2 flow. [b] Oc-
tylether of 9 was observed. [c] n-Octanol was previously dried. [d] 24 h re-
action time.
reactions to give 4 and 5 and the hydrolysis of 9 to give HMF.
However, the selectivity to 3 increases considerably with a cata-
lyst with an increased Na content to reach 95% selectivity if
the Na content was 1.5 wt%. Dimerization of HMF to give 6
was totally suppressed.
Figure 6a and b display the kinetics curves for HBeta and
NaBeta. In the case of HBeta zeolite (Figure 6a), 3 reaches
a maximum concentration and then decreases with time as
etherification to 5 and possibly hydrolysis to HMF occur.
Indeed, the hydrolysis of 9 mainly gives HMF, which undergoes
subsequent etherification to give 4. The formation of HMF is
a result of the water released during the etherification reac-
tions as well as the water that remains on the zeolite. The
water content of the HBeta zeolite after activation was
12.5 wt%.
Figure 6. Yield vs. t plot for the transacetalization reaction of 9 with n-octa-
~
&
&
*
*
^
nol in the presence of a) HBeta, b) 1.5NaBeta. HMF , 5 , 3 , 4 , 9 , 11
,
othersꢁ.
medium was considerably lower (Table 7). These results indi-
cate that the presence of strong acid sites on the catalyst is
also detrimental for the transacetalization reaction as they cat-
alyze the hydrolysis of 9. To avoid the hydrolysis of 9, we per-
formed an additional experiment by using dry n-octanol in the
presence of 1.5NaBeta zeolite. In this case, the rate of forma-
tion of HMF was an order of magnitude lower than that with
undried n-octanol, although the concentration of HMF re-
mained the same over a long reaction time (Table 7).
At this point, it appeared to us that one way to reduce the
competitive hydrolysis of the acetals could be by using cata-
lysts with a higher hydrophobic character, which would adsorb
less water (Table 5). To test this hypothesis, a more hydropho-
bic Beta zeolite prepared in fluoride media (Si/Al=12) was ex-
changed with Na (1.5 wt%) and
A different kinetic behavior is observed with the 1.5NaBeta
catalyst (Figure 6b). 1.5NaBeta zeolite promotes the transace-
talization of 9 with n-octanol to give an intermediate methyl-
octyl acetal of HMF 11, which further reacts with another mol-
ecule of n-octanol to give 3 (Scheme 3). Meanwhile, secondary
reactions are minimized as a consequence of the lower acidity
of the 1.5NaBeta catalyst (Table 1). In this case, the formation
of HMF is also mainly attributed to the residual water present
in 1.5Beta zeolite (10 wt%), although the rate of the formation
of HMF was approximately twofold lower than that with
HBeta, and the final HMF concentration in the reaction
tested in the reaction. In this
case although the rate of forma-
tion of HMF was also decreased
by one order of magnitude a sim-
ilar final result was obtained for
long reaction times. Neverthe-
less, by properly controlling cat-
alyst and reaction conditions,
95% yield of dioctyl acetal can
be obtained through a transace-
talization reaction between the
dimethyl acetal and n-octanol.
Scheme 3. Routes for the formation of 3 and byproducts through the transacetalization of 9 with n-octanol.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 123 – 131 128

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
The results presented above show that 1.5NaBeta zeolite is
able to catalyze efficiently and selectively both the acetaliza-
tion of HMF with MeOH to give 9 and the subsequent transa-
cetalization of 9 with n-octanol to give 3. Taking into account
these results, it appeared to us that it would be of interest to
design a one-pot process in which both reactions occur in
a single vessel. To achieve this, HMF was reacted first with
MeOH in the presence of 1.5NaBeta zeolite at 658C for 2.5 h to
achieve the maximum yield of 9 (100% HMF conversion with
99% selectivity to 9). At this point, n-octanol was added to the
reaction mixture in a molar ratio of 9/octanol of 1:3.7, and
MeOH was removed by bubbling a stream of N2 through the
suspension. The results are compared with those obtained
with the most acidic HBeta zeolite in Table 8. 1.5NaBeta zeolite
with n-octanol occurs with a high reaction rate to achieve the
maximum yield and selectivity to 3 after 30 min.
Comparatively, the most acidic HBeta zeolite was much less
selective achieved only a maximum selectivity of 47% to 3
(Table 8).
Catalyst recycling in the one-pot process
To test the reusability of the 1.5NaBeta catalyst, it was reused
in various cycles. After each reaction, the zeolite was collected
by filtration, washed thoroughly with acetone, and activated
under vacuum before each reuse. Figure 8 shows that the ac-
tivity and selectivity were maintained at least for the four
consecutive cycles studied without a decrease in activity or
selectivity.
Table 8. Comparative one-pot formation of 3 using HBeta and 1.5NaBeta
catalysts.[a]
Catalyst
Step 2
others[b]
3
4
5
HMF
Selectivity[c] [%]
95
90
45
47
0
3
7
0
4
8
2
1
35
10
3
3
5
4
95 (30 min)
90 (60 min)
45 (30 min)
47 (60 min)
1.5NaBeta
HBeta[d]
11
28
[a] Reaction conditions: step 1: HMF(1 mmol), MeOH (5 mL, 123 mmol),
1.5NaBeta (26 mg), 658C, 2.5 h; step 2: n-octanol (3.7 mmol), 658C,
30 min. [b] Others is octylether of 9 in this case. [c] Selectivity to 3 is cal-
culated as the yield of 3 divided by the conversion of 9. [d] 15 min reac-
tion time in step 1.
Figure 8. Reuse of 1.5NaBeta zeolite catalyst in the one-pot formation of 3.
Finally the one-pot process was extended to other aliphatic
alcohols as it is known that an important property of surfac-
tant molecules is biodegradability, which is related to the
length and branching of the alkyl chain of the alcohol. We se-
lected unbranched alkyl alcohols (between C6 and C12) to
obtain the corresponding alkyl acetal in a one-pot reaction.
The results obtained show the general applicability of the cata-
lyst and procedure to obtain precursors of molecules with sur-
factant properties (Table 9).
is able to perform very efficiently in the one-pot process to
achieve a 95% yield of 3 in 30 min. If the reaction time is pro-
longed to 1 h, a decrease in selectivity is observed owing to
the consecutive transformation of 3 into 4 and 5. Figure 7
presents the evolution of the reactants and products with
time. At the beginning of the reaction, there is an induction
period in which 3 is not formed, the rate of disappearance of 9
is very low, and HMF is the only product formed. This behavior
can be explained by taking into account the experimental pro-
cedure. At the beginning of the reaction, the excess MeOH
used in the first step remains in the reaction media as a solvent
and reactant, which shifts the equilibrium towards the forma-
tion of 9. As the MeOH is removed, the transacetalization of 9
Table 9. Results of the synthesis of alkyl acetals of HMF by a one-pot pro-
cess that used 1.5NaBeta zeolite as a catalyst.[a]
Alcohol
t
[h]
HMF conversion
[%]
Formation of dialkyl acetal
yield [%]
selectivity [%]
hexanol
octanol
decanol
dodecanol
0.5
0.5
2
100
100
93
98
95
90
87
99
98
97
96
2
90
[a] Reaction conditions: step 1: HMF (1 mmol), MeOH (123 mmol, 5 mL),
1.5NaBeta (26 mg), 658C, 2.5 h; step 2: alcohol (3.7 mmol), 658C.
Conclusions
It is possible to obtain surfactant precursor molecules by react-
ing HMF and fatty alcohols in the presence of solid acid cata-
lysts. Beta zeolite appears to be a good catalyst for this reac-
Figure 7. One-pot formation of 3 that used 1.5NaBeta zeolite as catalyst.
&
&
*
*
HMF , 5 , 3 , 9 , othersꢁ.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 123 – 131 129

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
were determined by using GC with hexadecane as the external
standard.
tion, and the main products obtained are 5-(dioctoxymethyl)-
2-hydroxymethylfurane (3), the ether 5-(octyloxymethyl)furfural
(4), and its corresponding dioctyl acetal 5. By controlling the
zeolite acidity and polarity as well as the reaction temperature,
it is possible to modify the product selectivity. Control of the
zeolite acidity is best performed by introducing different
amounts of Na+. Beta zeolite partially exchanged with Na+ is
also able to perform transacetalization with excellent conver-
sions and selectivities within very reasonable reaction times.
Finally, a two-step one-pot process that involved the acetali-
zation of HMF with MeOH to form the dimethyl acetal 9 and
transacetalization with n-octanol gave rise to yields of 3 of
95%. The one-pot process catalyzed by partially Na-exchanged
Beta zeolite has been extended to alcohols from C6 to C12 with
excellent activities and selectivities.
At the end of the reaction, the catalyst was collected by filtration,
and the reaction mixture was analyzed by GC–MS. The catalyst was
submitted to continuous solid–liquid extraction with acetonitrile
by using a micro-Soxhlet apparatus. After solvent removal, the resi-
due was weighed and analyzed by using GC–MS. In all cases, the
recovered material accounted for more than 90% of the starting
HMF.
Synthesis of 9: A solution of HMF (500 mg) in MeOH (28 mL)
heated to 658C was added to activated 1.5NaBeta zeolite (100 mg),
and the mixture was stirred for 2.5 h. The catalyst was removed by
filtration and the MeOH was eliminated by heating under reduced
pressure. The yellow concentrate obtained corresponded to 9 (ꢀ
99%) and was used without any further purification in the transa-
cetalization reactions.
GC–MS for 9: m/z (%)=172 [M+, C8H12O4] (10), 141 (100), 125 (5),
109 (7), 97 (5), 81(14).
Experimental Section
Transacetalization of 9 with n-octanol: The transacetalization reac-
tion of 9 with n-octanol was carried out in a batch glass reactor
equipped with a magnetic stirrer, a reflux condenser, and a Dean–
Stark trap immersed in a thermostated silicone bath under a N2
flow. Before reaction, the catalyst (26 mg) was activated in situ as
described above. A preheated (658C) mixture of 9 (1 mmol) and n-
octanol (3.7 mmol) was added to the catalyst, and the mixture was
stirred at 658C for the required time. Samples were collected at
regular intervals and analyzed according to the procedure de-
scribed above.
HMF (ꢀ99%), n-hexanol (ꢀ98%), n-octanol (ꢀ99%), n-decanol
(98%), dodecanol (ꢀ99%), and hexadecane (99%) were purchased
from Aldrich and methanol (99.99%) was purchased from Scharlau.
Materials
HBeta (CP811) and 0.58NaBeta-zeolites (CP806) (Si/Al=12) were
purchased from PQ Zeolites B. V. and were calcined at 5808C for
3 h before use. MCM-41 (Si/Al=15) with a pore diameter of 3.5 nm
was prepared according to a literature procedure.[14]
GC–MS data for 3: m/z (%)= 368 [M+, C22H40O4] (2), 239 (100), 127
(96), 109 (18), 97 (5).
Na-exchanged Beta zeolites (0.26NaBeta, 0.49NaBeta, 1.0NaBeta,
1.5NaBeta, and 3.0NaBeta, in which the number indicates the per-
centage [wt%] of Na) were prepared by impregnating commercial
HBeta zeolite with different aqueous solutions of CH3COONa fol-
lowed by drying at 1008C overnight and then calcination at 5808C
for 3h. The Na contents of the samples were determined by chem-
ical analysis (Varian 715-ES inductively coupled plasma optical
emission spectrometer) after dissolution of the solids in a HNO3/HF
solution.
The acidity of the catalysts was measured by IR spectroscopy
(Nicolet 710 FTIR spectrophotometer) combined with the adsorp-
tion–desorption of pyridine[15] at 10À4 Torr at 150, 250, and 3508C
(Table 1) by using self-supported wafers of 10 mgcmÀ2 that were
degassed overnight under vacuum (10À4–10À5 Pa) at 4008C.
Specific surface areas and pore diameters were measured by N2 ad-
sorption at 77 K by using a Micrometrics ASAP 2000 apparatus.
Thermogravimetric analysis (TGA) was performed by using
a Netzsch STA 409 EP thermal analysis with approximately 20 mg
of sample and a heating rate of 108CminÀ1 in an air flow.
One-pot reaction: A solution of HMF (1 mmol) in MeOH (5 mL,
125 mmol) was added onto the previously activated catalyst
(26 mg) and heated at 658C for 2.5 h in a system equipped with
a magnetic stirrer, reflux condenser, and Dean–Stark trap in a sili-
cone oil bath. During this first step, 9 was obtained in a yield
ꢀ99% according to GC analysis. The second step of the reaction
(transacetalization) was carried out by adding the fatty alcohol
(3.7 mmol) while a stream of N2 was bubbled continuously through
the suspension (50 mLminÀ1) to facilitate the removal of MeOH.
Samples were collected at regular intervals and analyzed according
to the procedure described above.
GC analyses were performed by using a Varian 3900 chromato-
graph equipped with a flame ionization detector and HP-5 (5%
cross-linked phenyl methyl silicone 30 mꢁ0.25 mmꢁ0.25 mm) ca-
pillary column. Mass spectra were obtained by using a GC–MS (HP
Agilent 5973 with a 6980 mass selective detector).
Acknowledgements
Reaction procedure
The authors wish to acknowledge the Spanish Ministry of Educa-
tion and Science for financial support of the projects Consolider-
Ingenio 2010 and CTQ-2011-27550.
Direct acetalization of HMF with n-octanol: Micro- and mesoporous
catalysts (51 mg) were activated in situ in a 10 mL batch glass reac-
tor by heating the solid at 2008C under vacuum (1 Torr) for 2 h.
The system was cooled to room temperature, and a mixture of
HMF and n-octanol (1:40) previously heated to the desired reaction
temperature was added to the catalysts. The mixture was heated
by using a system equipped with a silicone bath, magnetic stirrer,
and condenser. Samples were taken at regular intervals, diluted
with acetone, and the catalyst was separated by centrifugation.
The products in the solution were analyzed by using GC–MS, and
the conversion of HMF and yields of the different compounds
Keywords: alcohols
· surfactants · zeolites · biomass ·
synthesis design
[1] a) Y. Romꢂn-Leshkov, C. J. Barrett, Z. Y. Liu, J. A. Dumesic, Nature 2007,
447, 982; b) A. Corma, O. de La Torre, M. Renz, N. Villandier Angew.
Chem. 2011, 123, 2423; Angew. Chem. Int. Ed. 2011, 50, 2375.
[2] M. J. Climent, A. Corma, S. Iborra, Green Chem. 2011, 13, 520.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 123 – 131 130

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
[3] A. A. Rosatella, S. P. Simeonov, R. F. M. Frade, C. A. M. Afonso, Green
Chem. 2011, 13, 754.
[10] M. J. Climent, A. Corma, A. Velty, M. Susarte, J. Catal. 2000, 196, 345.
[11] M. J. Climent, A. Corma, S. Iborra, S. Miquel, J. Primo, F. Rey, J. Catal.
1999, 183, 76.
[12] I. Kiricsi, C. Flego, G. Pazzuconi, W. O. Parker Jr. , R. Millini, C. Perego, G.
Bellussi, J. Phys. Chem. 1994, 98, 4627.
[13] M. J. Climent, A. Corma, S. Iborra, J. Catal. 2005, 233, 308.
[14] C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartulli, J. S. Beck, Nature
1992, 359, 710.
[4] M. E. Zakrzewska, E. Bogel-Lukasik, Chem. Rev. 2011, 111, 397.
[5] a) C. Moreau, M. N. Belgacem, A. Gandini, Top. Catal. 2004, 27, 11; b) A.
Gandini, M. N. Belgacem, Prog. Polym. Sci. 1997, 22, 1203.
[6] a) O. Casanova, S. Iborra, A. Corma, J. Catal. 2010, 275, 236; b) O. Casa-
nova, S. Iborra, A. Corma, ChemSusChem 2009, 2, 1138.
[7] a) J. Deutsch, A. Martin, H. Lieske, J. Catal. 2007, 245, 428; b) B. Thomas,
V. Ganga Ramu, S. Gopinath, J. George, M. Kurian, G. Laurent, G. L.
Drisko, S. Sugunan, Appl. Clay Sci. 2011, 53, 227.
[15] C. A. Emeis, J. Catal. 1993, 141, 347.
[8] M. Sasidharan, R. Kumar, J. Catal. 2003, 220, 326.
Received: July 18, 2012
Revised: October 1, 2012
[9] a) S. Ito, A. Hayashi, H. Komai, Y. Kubota, M. Asami, Tetrahedron Lett.
2010, 51, 4243; b) M. J. Climent, A. Corma, S. Iborra, M. C. Navarro, J.
Primo, J. Catal. 1996, 161, 783.
Published online on January 9, 2013
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 123 – 131 131