Pharmaceutical Chemistry Journal
Vol. 29, No. 8, 1995
LIQUID-PHASE CATALYTIC OXIDATION OF CARBOHYDRATES BY OXYGEN: GLUCOSE OXIDATION IN ALKALINE SOLUTIONS IN THE PRESENCE OF COPPER(II) COMPLEXES G. G. A n d r e e v a I and S. R. T r u s o v I
Translated from Khimlko-Farmatsevticheskii Zhurnal, Vol. 29, No. 8, pp. 41 - 4 3 , August, 1995. Original article submitted February 7. 1994.
The carbon chain of monosaccharides is linked with a large number of electronegative hydroxyl groups, which produce a significant induction effect on the chain. The build-up of positive charge on carbon atoms facilitates nucleophilic attack on the molecule and rupture of the carbon-carbon bonds. For this reason, monosaccharides are unstable with respect to alkalis and oxidizers. It should be emphasized that any desired chemical transformations of monosaccharides in the presence of bases will be complicated by the reaction of enol formation, occurring in the alkaline medium, followed by enol reactions proceeding by various pathways [1]. This leads to the formation of a large variety of products, via both the alkaline hydrolysis of glucose and its analogs and the oxidation of these compounds in the alkaline medium [2]. However, data have been published [3, 4] showing that glucose can be also catalytically oxidized to gluconic acid by oxygen in an alkaline medium ( pH 8 - 11) in the presence of well-known carbon-supported catalysis containing Pt, Pd, Bi, etc. Also known are highly selective homogeneous catalysts used for the oxidation of hydroxymethyl groups to carboxy groups. It was established that Cu(II) complex compounds in the presence of bases catalyze the oxidation of primary alcohols to acids. Here, acids are the primary products formed when anions coordinated on the Cu(II) centers are attacked by molecular oxygen [5]. We have studied the oxidation of glucose by oxygen in an aqueous alkaline medium in the 60 - 80~ temperature interval in the presence of copper-phenanthroline complexes. It was suggested [5] that these complexes may exhibit rather high selective catalytic activity for the oxidation of hydroxymethyl groups to carboxy groups. An important condition for the glucose oxidation reaction to proceed with sufficiently high selectivity in an alkaline medium is that the rate of intrinsic oxidation must significantly exceed that of other possible conversions.
Both the first (oxidation) and the second (enolization plus isomerization) processes lead to the formation of acid products that can be identified by gas-liquid chromatography (GLC). The appearance of acid reaction products results in decreasing pH of the solution, which can be used for comparative evaluation of the rates of the two competing reactions. Figure 1 shows the plots of pH variation with time for the reactions of glucose oxidation under the experimental conditions described below (curve 1 ) and for glucose hydrolysis under the same reaction conditions without oxygen access (curve 2 ). As is seen, the rate of oxidation markedly exceeds that of hydrolysis. However, the chromatographic data show that glucose exhibits destructive changes at high pH values (pH 10 and
13-
12-
11
10
9-
8-
Fig. 1, Plots ofpH variation with time (h) for the reactions of glucose 1 ) oxidation and 2) hydrolysis,
z Riga Technical University, Riga, Latvia.
557 0091-150X/95/2908-0557512.500 1996 Plenum Publishing Corporanon
558
G.G. Andreeva and S. IL Trusov
160
140 120
100
80
60
40
20
5
15
25
35
45
55
65
Fig. 2. Kinetics of oxygen uptake during glucose oxidation in the presence of various amounts of NaOH: 1 ) 30 g/liter; 2) 40 g/liter; 3 ) 70 g/liter. Abscissa) Time, min; ordinate) Volume of absorbed oxygen ml.
Vo2,
above) as indicated by the presence of glycolic, glyceric, and erythronic acids among the reaction products. Figure 2 shows the kinetics of oxygen uptake during oxidation of I M aqueous glucose solution in the presence of copper complexes with o-phenanthroline ([o-phen] = 3[CUC12] = 0.015 M) and various amounts of NaOH. The shape of the curves indicates that oxygen is only absorbed until the alkali is neutralized by acid products formed during the reaction, after which the process ceases. This result suggests that glucose must be oxidized under pH-static conditions, which allow a relatively soft process regime to be maintained for an arbitrarily long time (including possibly low pH reducing the contribution due to the enol formation), and sufficiently high pH values to be provided for maintaining high catalytic activity. According to [6], the catalyst is active at pH > 11.
TABLE 1. Composition of the Oxidation Products Obtained upon Glucose Oxidation in an Alkaline Medium No. 1 2 3 4 5 6 7 8
Product Arabonic acid lactone Arabonic acid Glucose Erythronic acid lactone Erythronic acid Threonic acid Glyceric acid Glycolic acid
Oxidation in the pH-static regime was carried out in a gasbubble reactor vessel with special holes for electrodes and a capillary for the alkali introduction. The pH values were measured by a glass electrode (ESL-63-07) against silver chloride electrode (EVL-1M3) using a pH-121 model instrument connected to a BAT-15 automated titration unit controlling the supply of alkali to the reactor. The rate of glucose oxidation under the pH-static regime was monitored by the rate of alkali consumption that was necessary to maintain a given pH value. Figure 3 shows typical kinetic curves of alkali consumption under various process conditions. Despite various reaction rates, all curves are indicative of the nondecaying character of the process. Accordhag to the chromatographic data, oxidation leads to the formation of a combination of high- and low-molecular-weight acids. Table 1 shows the composition of oxidized products obtained in experiment 3 (Fig. 3). Removal of oxygen from the reactor system (by bubbling an inert gas through the solution) led to accumulation of the same acid products, albeit at a different ratio and much lower rate. Certain difficulties were encountered in the process, which were caused by using glass electrode in an alkaline medium at elevated temperatures. Under these conditions, electrodes of the type specified above exhibited very short working life. For this reason, the course of the glucose oxidation reaction was monitored by the redoximetric technique [7]. Apparently, interaction of glucose with the catalytic complex leads to modification of the redox properties of the catalyst, which is caused by the Cu II ~ Cu I transition. This circumstance must be reflected by changes of the redox potential of the system. Potentiometric measurements showed that each value of the redox potential (Fig. 4) corresponds to a certain pH value, as was previously established. Therefore, glass electrodes can be successfully replaced by platinum redox electrodes. In addition, the use of the redoximetric technique enable obtaining valuable information concerning the structure of catalytic complexes. The entire body of data obtained in studying the kinetic laws of the reaction of glucose oxidation by oxygen in an aqueous alkaline medium in the presence of copper-phenanthroline complexes as catalysts allows the rate of glucose oxidation to be markedly increased, so as to make this process dominating in the system studied. EXPERIMENTAL PART
Content, % 40 27 15 1.3 5.4 0.7 9.7 0.9
Experimental study of the kinetics of liquid-phase catalytic oxidation of glucose by oxygen was performed in a gasometric system comprising a 200-ml "duck" type reactor and a set of gas burettes with 0.1 ml readings, filled with ethylene glycol. The reactor was connected to gas burettes by rubber tubes via an electronic transducer. The total pressure in the system was maintained at atmospheric with the aid of a peristaltic pump connected to an electronic relay switch. The reac-
Liquid-phase Catalytic Oxidation of Carbohydrates by Oxygen tl
559
+ 500 + 400 + 300
j 2
+ 200
20
/
,........r 3
pH--8 1
+ 100 0 - 100 - 200 - 300
/
- 400 -500 - 600 0 0
i
i
i
15
30
45
i
i
t
i
i
60
75
90
105
120
i 135
i 150
i '" 165
Fig. 3. Typicalkinetic curves of alkali consumptionduring the oxidationof an 0.6 M glucose solution under various pH-static conditions: 1 ) [CuCI2]= 4.5 x 10-3 M, [o-phen]= 1.6 x 10 -2 M, TO= 70"C, PHconst 9.4; 2) [CuCI2] =5.4x10-3M, [o-phen]=1.6x10 -2M, T0-70*C, pHcons t 9.0; 3) [CuCI2] ~-5.4 x 10-3 M, [o-phen]- 1.6 x 10-2 M, TO-- 60"C, pHo~st 9.4 Abscissa)Time, min; ordinate) NaOH volume,ml.
tion mixture was agitated by a mechanical shaker of the Tabor type operated at a frequency selected so as to ensure that the process occurs in the kinetic region. The experimental procedure was as follows. The reactor maintained at a given temperature was purged with oxygen for 15 rain. The burettes of the gasometric system were filled with oxygen. The reactor was charged with components of the catalytic system (CuCI2 and o-phenanthroline), oxidized substrate, and solvent (H20). Then NaOH was introduced into the system to begin the reaction, and the shaker was switched on. Sixty seconds after starting the shaker, the excess pressure was released using a valve, and the gasometric burettes were connected to the system. From this moment on, the volume of oxygen absorbed was measured by the burette readings. The reduced volume of oxygen consumed in the system was calculated by the formula ~,0
O2
=
ET 273.2 Po Oz 760 (T+ 273.2) '
where V~2 is the volume of absorbed oxygen (cm 3 at the experimental temperature), T is the experimental temperature (~ and P0 is the atmospheric pressure (Torr).
Fig. 4. Time variation of the redox potential during glucose oxidation. Abscissa) Time, rain; ordinate) redox potential E, mV.
The temperature of the reaction mixture was maintained at a constant level using a TsT-15 thermostat. The volume of liquid phase was 10 ml in all experiments. The reaction products were analyzed by GLC using probes treated with a silylation mixture 0aexamethyldisilazanetrichlorosilane, 2 : 1). the analysis was performed on a Tsvet 2-65 type chromatograph with a plasma-ionization detector and a steel column (0.002 x 2 m) (f'med phase: 5% E-30 on N-AW chromaton; carrier gas, N2; thermostat temperature, 190~ evaporator temperature, 300~
REFERENCES 1. N. K. Kochctkov, A. F. Bochkov, B. A. Dmitriev, et al., Chemistry of Carbohydrates [in Russian], Moscow (1967). 2. L. A. Verhaar and H. G. Wilt,~ Chromatogr., 41, 168-179 (1969). 3. Japan Pat. Application No. 59-2053436 (1983). 4. US PaL No. 4943173. 5. A. M. Sakharov and I. P. Skibida, Kinet. Katal., 29(1), 118 - 123 (1988). 6. V. V. Chudaev, V. P. Tret'yakov, and E. S. Rudakov, Kinet. Katal., 27(2), 324 - 327 (1986). 7. A. L. Madelis, S. R. Trusov, and O. Y. Neiland, 1~. Akad. Nauk LatvSSR, Set. Khim., No. 5, 559 - 563 (1979).