Preparation of Long-Chain Alkyl Glucoside Surfactants by One-Step Direct Fischer Glucosidation, and by Transacetalation of Butyl Glucosides, on Beta Zeolite Catalysts

Article abstract of DOI:10.1006/jcat.1998.2272

Long-chain alkyl glucoside surfactants have been obtained with good conversions and selectivities by the transacetalation of butyl glucosides with different fatty alcohols as well as by one-step direct glycosidation, using H-Beta zeolites as catalyst. The influence of the different process variables such as temperature, molar ratio of reactants, and fatty alcohol chain length has been investigated. Moreover, the study of the transacetalation process in the presence of H-Beta zeolites with different hydrophobic-hydrophilic properties has shown the strong influence of this zeolite property on catalyst deactivation. The rate of decay is lower for the more hydrophobic Beta Samples.

Full text of DOI:10.1006/jcat.1998.2272

JOURNAL OF CATALYSIS 180, 218–224 (1998)
ARTICLE NO. CA982272
Preparation of Long-Chain Alkyl Glucoside Surfactants by One-Step
Direct Fischer Glucosidation, and by Transacetalation of Butyl
Glucosides, on Beta Zeolite Catalysts
A. Corma1, S. Iborra, S. Miquel, and J. Primo
Instituto de Tecnolog´ıa Qu´ımica, UPV-CSIC, Universidad Polite´cnica de Valencia, Avenida de los Naranjos s/n, 46022 Valencia, Spain
E-mail: acorma@itq.upv.es
Received April 15, 1998; revised August 14, 1998; accepted August 26, 1998
In a recent work (8) it was shown that it is possible to pre-
pare butyl glucosides in very good yields by means of the
Fischer reaction in the presence of large-pore acid zeolite
catalysts. The best results were obtained on a large-pore
H-Beta zeolite, which, in addition to having an adequate
topology, can easily be prepared with a large range of
Si/Al ratios and consequently with different hydrophilic–
hydrophobic characteristics (9).
In this paper we present the influence of the different
process variables on the preparation of alkyl glucosides by
transacetalation reaction of butyl glucosides with different
fatty alcohols using zeolite Beta. The two-step procedure
is compared with a single-step process, which involves the
direct reaction of a higher alcohol with glucose.
Long-chain alkyl glucoside surfactants have been obtained with
good conversions and selectivities by the transacetalation of butyl
glucosides with different fatty alcohols as well as by one-step direct
glycosidation, using H-Beta zeolites as catalyst. The influence of the
different process variables such as temperature, molar ratio of reac-
tants, and fatty alcohol chain length has been investigated. More-
over, the study of the transacetalation process in the presence of
H-Beta zeolites with different hydrophobic–hydrophilic properties
has shown the strong influence of this zeolite property on catalyst
deactivation. The rate of decay is lower for the more hydrophobic
c
Beta samples.
ꢀ 1998 Academic Press
INTRODUCTION
Alkyl glucosides are known to have several uses includ-
ing their incorporation into detergent products as nonionic
surfactants (1). In this case, they form micellar and vesicular
microaggregates whose properties depend on the compo-
sition and structure of the carbohydrate head groups and
on the length and shape of the hydrocarbon tail, as a result
of a very fine balance of hydrophobic and hydrophilic in-
teractions. As a consequence of this, alkyl glucosides with
short and long hydrocarbon chains have different applica-
tions. For instance, octyl glucosides are being used exten-
sively in biological membrane research (2) since they were
found to be amongthe most effective detergentsin releasing
proteins from membrane-bounded compartments without
denaturing them (3, 4). The alkyl glucosides derived from
monohydric alcohol containing 8 to 20 carbon atoms can
be prepared by the direct Fischer reaction (1, 5), but owing
to the low solubility of glucose in a long-chain fatty alcohol
the common approach involves a two-stage process (6, 7).
In a first step, glucosides containing a shorter alkyl chain
are prepared, and these, which are soluble in fatty alcohols,
are transformed into a long-chain alkyl glucoside by means
of a transacetalation reaction.
EXPERIMENTAL
Materials
Anhydrous ꢀ-D-glucose and n-butanol obtained from
Aldrich and Quimon, respectively, with a nominal purity
>99% , were used without further purification. 1-octanol
and 1-dodecanol were supplied by Aldrich with a nominal
purity >99 and 98% , respectively.
The H-Beta-1 sample was supplied by PQ Corporation,
in the acidic form and with a Si/Al of 13 (H-Beta-1). The
H-Beta-2 was synthesized according to the method pre-
sented in Ref. (10) with a Si/Al ratio of 29 and with a marked
hydrophobic character.
Reaction Procedure
Preparation of butyl glucosides. The zeolite catalyst was
activated in situ, in a 10-ml batch glass reactor, by heating
0.075 g of the Beta zeolite at 353 K under vacuum (1 Torr)
for 3 h. After this time, the system was cooled at room
temperature and then n-butanol (NB) (5 ml, 0.054 mol)
was introduced, followed by glucose (DG) addition (0.25 g,
0.0014 mol). The reaction mixture was heated at 393 K
1 To whom correspondence should be addressed.
218
0021-9517/98 $25.00
c
Copyright ꢀ 1998 by Academic Press
All rights of reproduction in any form reserved.

LONG-CHAIN ALKYL GLUCOSIDE SURFACTANTS
219
in a silicone oil bath and the catalyst was uniformly sus- periment and the recovered products accounted for >95%
pended as slurry in the butanol reaction solvent by stir- of the reactants.
ring at 600 rpm. After 4 h reaction time the mixture was
HPLC analyses were performed with a system formed
cooled at room temperature, dissolved with methanol, and by a Waters pump (model 510) and a Waters 410 dif-
filtered. Then, the filtered zeolite was submitted to solid– ferential refractometer using a HYPERSIL-APS-25 ꢁm
liquid extraction in a microsoxhlet using methanol and (250 ꢁ 0.46 mm) column. In order to deal with the less polar
water as first and second solvents, respectively. The or- products, the column was eluted with an acetonitrile/water
ganic solutions, freed from the solvents and the remain- mobile phase composed of 5% water, while 20% water was
ing alcohol by evaporation in vacuum, were combined, used for the more polar products. In both cases, the flow of
weighed, analyzed by HPLC, and used as starting material the mobile phase was1ml/min. Preparative scale HPLC was
for carrying out the second step, i.e., the transacetalation performed using a HYPERSIL-APS-25 ꢁm (250 ꢁ 1 mm),
reaction.
and in this case the flow of the mobile phases was 4 ml/min.
1H and 13C NMR of the products isolated from the re-
action mixture were done in a Varian 400WB instrument
and were identified by comparison with the NMR spectra
of butyl, octyl, and dodecyl glucosides reported in the liter-
ature (11–13).
Transacetalation process. For the transacetalation step,
40 wt.% zeolite according to the amount of butyl glucosides
was activated, and 1-octanol (NO) or 1-dodecanol (ND)
was added in the required proportion followed by the addi-
tion of the butyl glucoside. The reaction mixture was stirred
and heated at 393 K under a vacuum of 400 Torr so that the
n-butanol formed during the transacetalation was contin-
uously removed. In order to follow the evolution of the
reaction, samples were taken at regular time intervals. The
filtered zeolite remaining in the microsyringe was extracted
with methanol and water as first and second solvents, re-
spectively. The different products were analyzed by HPLC,
adding a known amount of methyl-ꢀ-D-glucopyranoside as
internal standard. At the end of the reaction, the reaction
mixture and the catalyst were treated as described above
for the preparation of butyl glucoside. Mass balances were
performed in each experiment and the recovered products
accounted for >95% of the reactants.
RESULTS AND DISCUSSION
The Transacetalation Process
As described above, the first reaction step was carried out
at 393 K in the presence of a H-Beta-1 zeolite with a Si/Al
ratio of 13 and a n-butanol/glucose ratio (NB/DG) of 40.
After 4 h reaction time, the yields of butyl furanoside and
butyl pyranoside were 10 and 90% , respectively, and this
mixture was the starting feed for carrying out the transac-
etalation reaction.
The reaction between the mixture of butyl glucosides
(BG) and NO in a BG/NO ratio of 12 was carried out at
393 K and 400 Torr, in the presence of H-Beta-1 zeolite. The
results obtained (Fig. 1) show that anomeric mixtures of
two octyl glucoside isomers, the (ꢀ,ꢂ)-octyl furanoside (3a)
and the (ꢀ,ꢂ)-octyl pyranoside (4a), were formed. Both glu-
cosides appear as primary products, and consequently the
global reaction can be written as presented in Scheme 1.
The one-step direct acetalization of glucose with 1-octanol
and 1-dodecanol. For the direct Fischer glucosidation,
0.25 g zeolite catalyst was activated in situ, in a 10-ml batch
glass reactor, by heating at 353 K under vacuum (1 Torr)
for 3 h. After this, the system was cooled to room temper-
ature and then the alcohol (3 ml, 0.019 mol of 1-octanol or
0.013 mol of 1-dodecanol) was introduced. The system was
heated at 393 K, and 0.05 g (0.27 mmol) glucose were added.
For kinetic purpose, this was considered the zero reac-
tion time. At 30-min intervals, constant amounts of glucose
were manually added to the system. With 1-octanol, glucose
was added in increments of 0.1 g (0.55 mmol), until 0.85 g
(4.72 mmol), while with 1-dodecanol, glucose was added in
increments of 0.07 g (0.38 mmol), until 0.6 g (3.33 mmol) in
order to reach in both cases a final alcohol/glucose ratio of 4.
With the aim of following the evolution of the reaction, be-
fore each glucose addition, a sample was taken, filtered, and
analyzed by HPLC, adding methyl-ꢀ-D-glucopyranoside as
the internalstandard. One-halfhour after the last incremen-
tal addition, the reaction was stopped by cooling at room
temperature. Then, the product mixture and the catalyst
were treated as previously described for the preparation of
FIG. 1. Time conversion plot of butyl glucosides (1 + 2, ᭹) to octyl
glucofuranoside (3a, ᭡) and octyl glucopyranoside (4a, ꢁ) at 393 K in the
butyl glucosides. Mass balances were performed in each ex- presence of H-Beta-1 and starting from an initial 1/2 ratio of 10/90.

220
CORMA ET AL.
SCHEME 1
In a first approximation it is reasonable to expect that used before. The rest of the reaction variables were kept the
the reaction leading to the formation of the octyl gluco- same. The kinetic results presented in Fig. 2a clearly show
furanoside (3a) will be the transacetalation of the butyl that, in this case, the amount of butyl glucopyranoside (2)
glucofuranoside (1) (step A, in Scheme 2). On the other first increases with time, goes though a maximum, and then
hand, the formation of the octyl glucopyranoside (4a) can decreases. On the other hand, the butyl glucofuranoside
occur by the transacetalation of the butyl glucopyranoside (1) continuously decreases to almost complete consump-
(2) (step D), the isomerization of the octyl glucofuranoside tion. In the case of the octyl glucofuranoside (3a), this is
(3a) (step B), or even the isomerization of the butyl gluco- rapidly formed and then the concentration diminishes while
furanoside (1) followed by the transacetalation of the inter- the octyl glucopyranoside (4a) continuously increases with
mediate molecule during the ring opening (step E) (14a).
In order to study in greater detail the mechanism of
time (Table 1).
Taking into account the evolution of reactants and prod-
the transacetalation, this reaction was again carried out at ucts with reaction time (Table 1 and Fig. 2a), we can say that
393 K and 400 Torr but starting now with a reaction mixture the butyl furanoside not only undergoes an isomerization
formed by 30 and 70% of butyl pyranoside and butyl fura- to the corresponding pyranoside but also a transacetalation
noside, respectively, instead of the 90 and 10% , respectively, to the primary and unstable octyl furanoside. Consequently,
the octyl pyranoside is formed by two different routes: the
isomerization of the octyl furanoside to the thermodynami-
cally more stable six member ring compound and the direct
TABLE 1
Initial Disappearance Rates of Butylglucosides and Distribution
of Products Obtained at Two Conversion Levels, Starting from Two
Different Initial Mixture Compositions in Butyl Glucosides
Mixture composi-
Initial disappear-
tion at different
ance rates of
conversions
Initial mixture
composition
butylglucosides
Compounds
(mol hꢂ1 gꢂ1) ꢁ 103 55%
70%
1
2
3a
4a
70
30
0
15
2
43
19
36
2
28
12
58
0
1
2
3a
4a
10
90
0
5.2
0
45
9
0
28
5
0
46
67
SCHEME 2

LONG-CHAIN ALKYL GLUCOSIDE SURFACTANTS
221
FIG. 2. Time conversion plot of butyl glucofuranoside (1, o) and butyl glucopyranoside (2, ᭿) to octyl glucofuranoside (3a, ᭡) and octyl glucopy-
ranoside (4a, ꢁ) at 393 K in the presence of H-Beta-1 and starting from an initial 1/2 ratio of 70/30 (a) and 10/90 (b).
transacetalation of the butyl pyranoside initially present in glucosides, which implies an apparent activation energy of
the starting mixture. However, we do not reject that the ꢄ27 Kcal molꢂ1 for the transacetalation process. This value
transacetalation of a possible open intermediate form, giv- is close to the activation energy obtained for the glucosi-
ing rise to the direct formation of the octyl glucopyranoside, dation of glucose with butanol (8) and also similar to the
can also occur. Nevertheless, as the formation of 3a is kinet- activation energy of the transacetalation process when the
ically favored (14, 15), we can suppose that, if this reaction reaction is carried out using liquid acids in a single-phase
takes place, its contribution to the final product will be small process. These results indicate not only that one could
(Scheme 2).
strongly influence the rate of the transacetalation reaction
As we can observe in Table 1, the initial rate of dis- by increasing the temperature but also that in the solid–
appearance of the butyl glucosides is strongly influenced liquid process studied here, the reaction is not controlled
by the feed composition. However, the final compositions by the diffusion of reactants or products in the solid. Nev-
achieved with the two different feeds are similar, and only ertheless, it should be taken into account that with temper-
the selectivities to octyl pyranoside and octyl furanoside atures significantly greater than 393 K, side reactions may
seem to be slightly affected by the initial composition of increase faster than the primary reaction, giving rise to a
the mixture; i.e., starting from a higher concentration of poor quality product (16).
butyl furanoside we obtain a richer final composition in
Influence of the chain length of the alcohol. The alkyl
octyl furanoside.
chain linked to the sugar residue of the alkyl glucoside
In any case, taking into consideration that the octyl fura-
is compatible with nonpolar environments, whereas their
noside and the octyl pyranoside present similar detergency
hydroxyl groups are responsible for the high solubility in
properties, the total conversion of the butyl glucosides be-
comes more important than the distribution of the two iso-
mers when one is interested in the application of the gluco-
sides as environmentally friendly detergents. Thus, from the
results presented in Table 1 and Figs. 2a and 2b, it becomes
clear that starting with a mixture rich in butyl furanoside
(70% ) the reaction rate is higher, and a larger total conver-
sion is reached in shorter reaction times.
Influence of the reaction temperature. To study the in-
fluence of the reaction temperature on the transacetalation
process, we have carried out the reaction in the presence
of H-Beta-1, at 383 and 393 K, starting from a mixture of
butyl glucopyranoside and butyl glucofuranoside of 90 and
10% , respectively, using 1-octanol as the second alcohol in
a NO/BG ratio of 12 and under 400 Torr. The kinetic results
obtained are given in Figs. 1 and 3. It is possible to see that
FIG. 3. Time conversion plot of butyl glucosides (1 + 2, ᭹) to octyl
a 10ꢃ increase in the reaction temperature produces a 2.5
glucofuranoside (3a, ᭡) and octyl glucopyranoside (4a, ꢁ) at 383 K in the
increase in the initial rate of the disappearance of the butyl presence of H-Beta-1 and starting from an initial 1/2 ratio of 10/90.

222
CORMA ET AL.
water. The length of the alkyl chain has a direct influ-
ence on the alkyl glucoside solubilizing abilities and has
the most pronounced effect on the critical micelle concen-
tration (CMC) (17). It is known that short-chain alkyl glu-
cosides do not provide any detergency benefits (1, 18), and
only for alkyl chains above C8 is a drastic decrease of the
CMC with increasing alkyl chain observed (6). It is then of
interest to have a process flexible enough to produce alkyl
glucosides with different alkyl chains. This has been stud-
ied here by also performing the transacetalation of butyl
glucosides with 1-dodecanol.
Starting from the same initial mixture (2/1 = 9), the reac-
tion was carried out using 1-dodecanol as secondary alcohol
at 393 K and in a molar ratio BG/ND = 12. The results ob-
tained show the formation of the two expected isomers, the
dodecyl glucofuranoside (3b) and the dodecyl glucopyra-
noside (4b) (Fig. 4), while the initial rate and the final con-
version are smaller for the transacetalation of the longer
alcohol, which would need a longer reaction time to reach
interesting levels of conversion.
TABLE 2
Results Obtained in the Two-Step Process for the Production
of Octyl Glucosides without and with Addition of Fresh Catalysts
during the Transacetalation Step and Starting from a Butyl Gluco-
side Mixture 2/1 of 9, in the Presence of the H-Beta-1 Zeolite at
393 K after 4 h Reaction Time
Initial rate of
butylglucosides
disappearance
(mol hꢂ1 gꢂ1) ꢁ 103
Yields (% )
Catalyst
1
2
3a
4a
Renovated
Same
5.2
4
0
0
19
33
5
7
76
60
added just before adding the 1-octanol and performing the
transacetalation. After 4 h reaction time, the results pre-
sented in Table 2 were obtained. It can be seen there that
when the same sample of catalyst was used to carry out the
two steps, the initial rate of the transacetalation reaction
was 25% lower than when fresh catalyst was introduced to
perform it. The results indicate that some deactivation of
the catalyst does indeed occur during the glucosidation pro-
cess. After 4 h reaction time, the fresh catalyst (H-Beta-1)
still gives a much higher conversion than the used catalyst
(Table 2). In a previous paper, we demonstrated the im-
portance of the hydrophobic–hydrophilic properties of the
catalyst on activity (9). It can then be expected that those
catalyst properties will also have a strong impact on cata-
lyst deactivation. In order to study this we have repeated
the above experiments with a hydrophobic Beta zeolite
(H-Beta-2 with Si/Al = 29). The results presented in Fig. 5
show that with zeolite H-Beta-2, which is more hydropho-
bic than H-Beta-1, there is also an initial deactivation.
However, after 4 h reaction time the same conversion was
achieved with the fresh (Fig. 5a) and with the used catalyst
(Fig. 5b). This can indicate that with this catalyst, there is
some deactivation that occurs in the first period of time and
then the activity levels off. In any case it appears that the de-
activation of the catalyst during the first step, i.e., the forma-
tion of the butyl glucoside, does not impede carrying out of
the second step in very reasonable reaction times.
Catalyst deactivation. From the engineering point of
view, it will be of interest to carry out the two reaction
steps, i.e., glucosidation with n-butanol and transacetala-
tion of the butyl glucosides with 1-octanol, in a single pot,
avoiding separation of the catalyst and its removal after the
first step has occurred. However, in order to be able to do
this economically, little or no deactivation of the catalyst
has to occur during the glucosidation with n-butanol. To
study this, we have carried out a new series of experiments
using in both steps an alcohol/glucoside ratio of 12 (mol
molꢂ1) starting from a mixture of butyl glucopyranoside
and butyl furanoside of 90 and 10% , respectively, and using
1-octanol as the second alcohol. In one case, after distilla-
tion of the butanol we have added it directly to the reac-
tor without changing the catalyst, while in the second case,
the used catalyst was separated and new fresh catalyst was
We have analyzed the products remaining adsorbed after
the reaction and the extractions with soxhlet, and they are
composed mainly of alkyl glucosides and of little amounts
of glucose. When these products are burned off by calci-
nation at 500ꢃC, the initial activity is completely restored.
The fact that some deactivation occurs in the first step
may imply in some cases the need to separate the catalyst
after this and to add new catalyst for carrying out the second
step of the global reaction, i.e., the transacetalation. Such an
inconvenience strongly encouraged us to perform the glu-
cosidation of glucose with long-chain alcohols in one step,
thereby avoiding the transacetalation reaction required in
the conventional process. We return later to this point.
FIG. 4. Time conversion plot of butyl glucosides (1 + 2, ᭹) to dodecyl
glucofuranoside (3b, ᭡) and dodecyl glucopyranoside (4b, ꢁ) at 393 K in
the presence of H-Beta-1 and starting from an initial 1/2 ratio of 10/90.

LONG-CHAIN ALKYL GLUCOSIDE SURFACTANTS
223
FIG. 5. Time conversion plot of butyl glucofuranoside (1, o) and butyl glucopyranoside (2, ᭿) to octyl glucofuranoside (3a, ᭡) and octyl glucopy-
ranoside (4a, x) at 393 K (a) in the presence of fresh H-Beta-2 and (b) without renovating the catalyst (H-Beta-2) and starting from an initial 1/2 ratio
of 40/60.
Influence of the NO/BG ratio. From an economical The Direct Reaction of a Long-Chain
point of view it is important to minimize the alcohol-to-
glucose ratio while keeping the rate of the reaction and the
selectivity to alkyl monoglucosides at reasonable levels. In
order to study the former variable, the second step of the re-
action was carried out using a NO/BG of 12, 7, 4, and 2 (mol
molꢂ1) and starting from a mixture composition of 2/1 of 9.
The results from Fig. 6 clearly show the same behavior
we observed (8) for the first step of the reaction; i.e., the
formation of octyl glucosides increases to a NO/BG of 7
and then remains practically unchanged. However, most of
the gain in reaction rate has already been achieved at a
NO/BG ratio of 4.
In conclusion, it is possible to obtain good conversions
to glucosides containing hydrophobic chains by transaceta-
lation reaction of butyl glucosides with fatty alcohols us-
ing H-Beta zeolite as catalyst. The transacetalation is best
carried out starting with a feed composition rich in butyl
furanoside. In the case of transacetalation with 1-octanol,
ratios of NO/BG of 4 and above are required.
Alcohol with Glucose
There is no doubt that it will be very interesting to man-
ufacture higher alkylglucosides in a single step. The main
problem one may encounter with this is the low solubility
of glucose in long-chain alcohols with the corresponding
negative effect on reaction rate and on the quality of the
final product. For instance, the solubility of D-glucose in
1-octanol at 90ꢃC is quite low (ꢄ1.6 g lꢂ1) (1). Nevertheless
it should be possible to overcome this limitation using an ex-
cess of alcohol and adding incremental amounts of glucose
(19). Then, when a certain amount of the higher alkylgluco-
sides are formed, the reaction will be favored since the alkyl
glucoside acts as an emulsifier stabilizing the monophasic
system.
In order to verify this hypothesis, two reactions were car-
ried out, as indicated above, using 1-octanol (Fig. 7) and
1-dodecanol (Fig. 8) as the alcohol, and in the presence of
FIG. 7. Amount (mmol) versus time of the products obtained in the
direct acetalization of D-glucose (ᮀ) with 1-octanol at 393 K: octyl gluco-
furanoside (3a, ᭡) and octyl glucopyranoside (4a, ꢁ). Amount of octyl
glucosides that would be obtained if all the added D-glucose were con-
verted (4).
FIG. 6. Influence of the 1-octanol/butyl glucoside molar ratio on the
initial rate (mol hꢂ1 gꢂ1 103) of formation of octyl glucosides at 393 K in
the presence of H-Beta-1.

224
CORMA ET AL.
high conversions and selectivities by the transacetalation
of butyl glucosides with fatty alcohols in the presence of
H-Beta zeolite as the catalyst. Furthermore, we have found
that the initial rate of disappearance of butyl glucosides is
strongly influenced by the feed composition, and a higher
amount of butyl glucofuranosides gave rise to a higher reac-
tion rate and a larger total conversion over shorter reaction
time. The hydrophobic H-Beta zeolite proved to present
a better resistance to deactivation during the transaceta-
lation process. Moreover, for this two-step reaction, ra-
tios of fatty alcohol to butyl glucosides of 4 and above
were found to give a good initial reaction rate minimiz-
ing side reactions. Finally, it is possible to prepare alkyl
FIG. 8. Amount (mmol) versus time of the products obtained in the glucosides with an alkyl chain above C8, starting directly
direct acetalization of D-glucose (ᮀ) with 1-dodecanol at 393 K: octyl glu-
from glucose and fatty alcohol, obtaining good conver-
sions and excellent selectivities to the long alkyl glucoside
surfactant.
cofuranoside (3b, ᭡) and dodecyl glucopyranoside (4b, ꢁ). Amount of
octyl glucosides that would be obtained if all the added D-glucose were
converted (4).
H-Beta-1. The results presented in these figures indicate
(dashed line) the amount of alkyl glucoside (expressed in
mmol) that would be obtained if all the added glucose were
converted in each step. The continuous lines represent the
experimental values obtained. It can be seen that, in both
cases, the glucose readily reacts, giving as the only products
the two alkyl glucoside isomers expected. In the presence of
1-octanol and for each increment, the conversion is almost
complete, indicating first that the addition of the glucose
to the reaction mixture is carried out over an adequate pe-
riodic time interval and second that the quantity added is
an acceptable amount. On the other hand, in the presence
of 1-dodecanol the conversion reached after each stage is
lower. After 4 h reaction, no further glucose was added and
the conversion was still monitored. It was found that allow-
ing more time led to an increase in the conversion, showing
that the catalyst is still presenting activity. We can therefore
easily suppose that a better conversion can be reached by
adding in each step a lower amount of glucose or using a
longer interval time between each addition. These results
prove that it is possible, using H-Beta zeolite as the cata-
lyst, to prepare, in one step and with good conversions and
excellent selectivities, alkyl glucosides with an alkyl chain
above C8, starting directly from glucose and the high al-
cohol. These results suggest that there is a possibility for
the commercial application of zeolites in the preparation
of long-chain alkylglucosides (20).
REFERENCES
1. Straathof, A. J. J., van Bekkum, H., and Kieboom, A. P. G., Starch
Sta¨rke 40, 229 (1988).
2. Jeffrey, G. A., Acc. Chem. Res. 19, 168 (1986).
3. Womack, M. D., Kendall, D. A., and MacDonald, R. C., Biochim.
Biophys. Acta 733, 210 (1983).
4. Baron, C., and Thompson, T. E., Biochim. Biophys. Acta 382, 276
(1975).
5. McCurry, P. M., Jr., and Pickens, C. E., U.S. Patent 4,950,743 (1990).
6. Thiem, J., and Bo¨cker, T., Spec. Publ. R. Soc. Chem. 107, 123
(1992).
7. Boettner, F. E., U.S. Patent 3,219,656 (1965).
8. Corma, A., Iborra, S., Miquel, S., and Primo, J., J. Catal. 161, 713
(1996).
9. Camblor, M. A., Corma, A., Iborra, S., Miquel, S., Primo, J., and
Valencia, S., J. Catal. 172, 76 (1997).
10. Valencia, S., Camblor, M. A., and Corma, A., Sp. Patent P9600625
(1996).
11. Straathof, A. J. J., Romein, J., van Rantwijk, F., Kieboom, A. P. G., and
van Bekkum, H., Starch Sta¨rke 39, 362 (1987).
12. Raaijmakers, H. W. C., Zwanenburg, B., and Chittenden, J. F., Recl.
Trav. Chim. Pays Bas 113, 79 (1994).
13. Focher, B. G., and Torri, G., Chem. Phys. Lipids 53, 141 (1990).
14. (a) Capon, B., Chem. Rev. 68, 407 (1969). (b) Binkley, R. W., “Modern
Carbohydrate Chemistry,” p. 87, Food Science and Technology, Vol. 27.
Dekker, New York, 1988.
15. de Goede, A. T. J. W., van Deurzen, M. P. J., van der Leij, I. G., van
der Heijden, A. M., Baas, J. M. A., van Rantwijk, F., and van Bekkum,
H., S. Carbohydr. Chem. 15, 331 (1996).
16. McCurry, P. M., and Pickens, C., WO 90/07516 (1990).
17. de Grip, W. J., and Bovee-Geurts, P. H. M., Chem. Phys. Lipids 23, 32
(1979).
18. Mao, H. K. M., Miller, L. E., and Weeman, J. M., U.S. Patent 4,393,203
(1983).
CONCLUSIONS
19. Farris, D. D., Eu. Patent 0096917 (1983).
20. Corma, A., Iborra, S., Miquel, S., and Primo, J., Sp. Patent 9502187
(1995).
In the present work, we have elucidated the possibil-
ity of obtaining long-chain alkyl glucoside surfactants with