T. Moreno et al. / Carbohydrate Research 349 (2012) 33–38
35
8.8 mL batch tube reactors. The glucose solution (0.1 mol/L) was
mixed with the appropriate amount of H2O2 (H2O2/glucose molar
ratio 1:2) to a total volume of 7 mL (80% of total capacity) and
the reactors were placed into an electric furnace (Akico Co. Ltd,
SHS-R4-500, Japan) and quickly heated to the desired reaction
temperature (120–200 °C). In most cases, there was no further
acidification/alkalinization, which provided a pH of 5.4. This value,
however, was modified to 2.0 or 9.0 in a few experiments, which
were carried out with the addition of phosphoric acid or KOH in or-
der to evaluate the influence of pH on the reaction. During the
reaction, the reactors were shaken using a mechanical device. After
a certain reaction time (15–60 min), the reactors were taken out of
the furnace and quickly cooled down to stop the reaction.
The second batch of experiments combined wet oxidation and
hydrothermal electrolysis (WO + HE), and were carried out in a
500 mL batch autoclave made of stainless steel. A cylindrical irid-
ium plate electrode (23 mm diametre, 30 mm length), coaxial to
the titanium beaker (74 mm internal diametre, 100 mm length),
was used as an anode, and the titanium beaker was used as a cath-
ode to promote cathodic protection of the autoclave from corro-
sion. Glucose aqueous solution (0.1 mol/L) was used as starting
material. Phosphoric acid was used as electrolyte, and the pH of
the solution was adjusted to 2.0. At the beginning of the experi-
ment, the aqueous solution (300 mL) was loaded in a titanium bea-
ker and placed in the autoclave. After its lid was closed tightly, the
autoclave was purged twice and then pressurized with argon gas to
an initial pressure of around 20 bar. At this point, the system was
heated up to the desired temperature (120–160 °C) with stirring
at 180 rpm. Temperature was measured with a thermocouple con-
nected to the reactor. Reaction pressure varied between 25 and
30 bar during the reaction. A constant electric current (1–3 A)
was applied to the autoclave when the whole system had reached
the desired temperature, and that was considered the starting
point of the reaction. Liquid samples were taken every 30 min.
After 180 min the electric current was cut off and the autoclave
was cooled with stirring until the temperature dropped below
40 °C. The volume of gas generated was measured using a cumula-
tive flow metre and the liquid product was collected.
smaller amounts. In the degradation experiments (i.e., without
any H2O2), the only products obtained were glyceraldehyde, 1,6-
AHG, 5-HMF and fructose. However, 1,6-AHG and 5-HMF were also
found in the experiments at the highest temperature (200 °C),
along with 2-furfural. Fructose was also obtained at the longest
reaction times (60 min).
Table 1 summarizes the results obtained, expressed as glucose
conversion and carbon-based yield for each product. The effect of
reaction temperature on the oxidation reaction at a constant
H2O2/glucose ratio of 1.5 (corresponding to 5100 ppm H2O2) can
be observed: conversion clearly increases with temperature, reach-
ing over 80% at 200 °C, twice as much as at 120 °C. However, at
200 °C the yield to gluconic acid is considerably lower, and dehy-
dration products such as 5-HMF and 2-furfural appear. An accept-
able balance is found at 160 °C, with 50–60% conversion and
around 10% yield to gluconic acid, so for the next round of exper-
iments the temperature was set to 160 °C and the effect of the
amount of peroxide was studied. In this case, conversion increases
with higher amounts of peroxide, but the effect is not as steep as
with temperature, which suggests that the reaction is kinetically
controlled. However, selectivity to gluconic acid increases consid-
erably for a ratio of 2 (corresponding to 6800 ppm H2O2), particu-
larly at shorter reaction times (15 min), reaching 30%. Yield of
gluconic and formic acids also increase notably under these condi-
tions, reaching 15% and 10%, respectively (note that this 10% yield
to formic acid is expressed on a carbon basis and corresponds to a
60% molar yield). The higher amount of oxidant also provides a sig-
nificant enhancement in the yield to glucaric acid, coming from the
oxidation of gluconic acid.
In the catalytic oxidation of glucose, a slightly alkaline medium
(ca. 9) is preferred in order to increase the reaction rate and to
avoid deactivation of the catalyst.16 Even though our reaction sys-
tem was not catalytic, some experiments were carried out under
alkaline conditions (pH 9). As observed in Table 1, significantly
lower glucose conversion and yields to the products of interest
were obtained when operating at high pH and the formation of
dehydration products such as 1,6-AHG and 5-HMF was especially
enhanced at the higher temperatures. This suggests that the alka-
line medium is not particularly favourable in the absence of a suit-
able catalyst. However, these were only preliminary experiments
that require further validation.
3. Results and discussion
Alternatively, the experiments carried out with an acidified
starting solution showed very similar conversions to those with
the unmodified reaction medium, but generally lower yields, par-
ticularly to the product of interest, that is gluconic acid. Nonethe-
less, the possibility of using an acidic medium would be extremely
interesting in order to develop a direct process from starch, as well
as to avoid the production of gluconate instead of the free acid.
3.1. Wet oxidation of D-glucose
Typical HPLC chromatograms obtained after the oxidation reac-
tion of glucose are shown in Figure 2. The main products obtained
were gluconic and formic acids, dihydroxyacetone and glyceralde-
hyde. Glycolic, acetic and glucaric acids were also obtained in
3.2. Wet oxidation + hydrothermal electrolysis of D-glucose
Since the hydrothermal electrolysis of glucose had already been
studied,15 we focused on the combination of electrolysis with a
small amount of H2O2, and therefore the main parameters analysed
were the intensity of the electric current and the amount of perox-
ide added to the reaction medium. Although conversion was gen-
erally lower than that in the previous wet oxidation experiments,
only carboxylic acids (i.e., oxidation products) were obtained:
mainly gluconic and formic acids, but also acetic, glycolic and gluc-
aric, suggesting that the oxidation reaction is predominant. In all
cases, we observed that both conversion and yield increased line-
arly with reaction time, unlike in some of our previous experi-
ments. In accordance with previous results obtained by this
research group, a reaction pathway was proposed (see Fig. 3).
Figure 4 shows the effect of the intensity of the current on the
yield to gluconic acid, formic acid and by-products (glucaric,
Figure 2. RI and UV chromatograms and identified peaks.