(+)-Catechin and other reagents described below were purchased from Sigma and Nacalai tesque, respectively. (+)-Catechin hydrate, ethanol (EtOH, 99.5%), methanol (MeOH, 99.8%), 1-propanol (1-PrOH, 99.5%), 2-pro- panol (2-PrOH, 99.7%) and tert-butyl alcohol (t-BuOH, 99.0%) were used without further purification. Monoe- thanol amine (MEA), diethanol amine (DEA), triethanol amine (TEA), l-arginine (Arg), sodium carbonate (Na₂CO₃), potassium carbonate (K₂CO₃), sodium phosphate (Na₃PO₄) and sodium hydroxide (NaOH) were used as base without further purification.

(+)-Catechin and a base were dissolved in 100 g of distilled water/alcohol mixed solvent. Oxygen gas (≥99.5 vol%) was introduced continuously into the reaction solution through porous glass ball filter (pore size: 40 - 50 μm) at 100 ml∙min⁻¹ of flow rate, and the solution was stirred for prescribed time at 20˚C - 50˚C.

The reaction solution was sampled at several time intervals and diluted by distilled water with a dilution ratio (f). The ultraviolet-visible absorption spectra of the sampled solution were measured by a Hitachi U-3900H spec- trophotometer at 25˚C. The dye formation behaviour was monitored by using the absorbance at 430 nm (A₄₃₀) at each reaction time (t). The obtained f·A₄₃₀ values correspond to the concentration of the formed dye [7] and were used the index of the amount of the dyestuff formed. The dyestuff formation rate (v) was calculated by v = [Δ(f·A₄₃₀)] _max/Δt, where Δt is the reaction time interval and [Δ(f·A₄₃₀)]_max the maximum increment of f·A₄₃₀ for the Δt. Constant Δt was adopted for the calculation of v for each the experiment.

The apparent basicity of the reaction mixed solution (pH^*) was determined through the measurement by a TOA-DKK MM-60R multi water quality meter with a TOA-DKK ELP-031 glass combination electrode at 25˚C.

The colour of reaction solution turns orange and reddish brown with reaction time showing the formation of catechinone dyestuff. Catechinone is hardly obtained in acidic and neutral reaction solution without base. The rate of catechinone formation increases with increasing the basicity of aqueous reaction solution [6] and it is in- creased by using a stronger organic base [7] . The previous results indicate the strong influence of basicity on the reaction. Then, the effects of basicity and sort of bases on the catechinone formation were studied first. The pH is not defined in the water/organic solvent mixture, of which organic composition is high, and the apparent ba- sicity was estimated as using pH^* in the study. The amount of the formed dyestuff at 200 min of reaction time ((f·A₄₃₀)_200min) and the dyestuff formation rate (v) value in the water/ethanol mixed solution for each of the ex- periment with a variety of bases changing their concentration (molarity: m_B) were measured and determined. The alcohol molar fraction of the mixed solvent (x_A) was 0.28.

Figure 2 shows the relationship between pH^* and (f·A₄₃₀)_200min for the water/ethanol solution with MEA. The resulting values are also summarised in Table 1. The (f·A₄₃₀)_200min and v increase with increasing pH^* and m_B of MEA as shown in the figure and for experimental number 2 - 5 in the table. The results show that the reaction is

^aPrevious work [7] ; ^bThe base dissociation constant in water at 25˚C [8] ; ^cApparent pH value in the water/ethanol mixed solvent; ^df·A₄₃₀ at 80 min.

promoted by higher basicity of the reaction solution. This relationship is similar to that between pH and the rate of catechinone formation in aqueous solution [6] . The MEA is also more effective for the formation than DEA and TEA. The deprotonated (+)-catechin in a basic medium works as an electron donor and reacts with O₂ eas- ily [6] [9] . Therefore, the higher basicity should contribute to promote the formation reaction also in the water/ ethanol solution.

Furthermore, the preparation of the dyestuff was made in the mixed solution with basic amino acid l-arginine or inorganic bases such as Na₂CO₃, K₂CO₃, Na₃PO₄ or NaOH (as listed in Table 1). The solubility of Na₃PO₄ in the water/ethanol mixed solution is low and its saturated concentration is 0.0068 mol∙kg⁻¹. The m_B of other in- organic bases was 0.10 mol∙kg⁻¹. The results in Table 1 show that f·A₄₃₀ for the reaction system with Na₂CO₃, K₂CO₃ or NaOH is larger than that with same concentration of MEA. However, it is lower than that with 0.50 mol∙kg⁻¹ of MEA, and the solubility of inorganic bases in the water/ethanol mixed solution is lower as compared with that of MEA. The solution pH^* decreases sharply at the earlier stage of the reaction for the inorganic base systems. This results in the decrease of the reactivity. In contrast, the pH^* decreases gradually in the MEA sys- tem. Consequently, using MEA as the base is of greater advantage for the production.

The rate of oxidation reactions are generally influenced by temperature. It is instructive to study the effect of re- action temperature on catechinone formation for obtaining better condition to produce the dyestuff efficiently. Figure 3 indicates the relationship between the reaction temperature (T ) and f·A₄₃₀ at 140 min ((f·A₄₃₀)_140min) for the catechinone preparation in water/ethanol solution (x_A = 0.28) with MEA. Here, the values of f·A₄₃₀ at 140 min are adopted, because f·A₄₃₀ increases and then decreases with reaction time at higher temperature and the downward turn begins before 200 min as described at the next section. The amount of the formed dye increases with increasing T up to 30˚C and it decreases with increasing T over 30˚C. The formation of colourless products and/or the precipitation of colourants may be caused by the further reactions of (+)-catechin and catechinone, and such the side reactions give rise to the decrease in f·A₄₃₀ at higher temperature [6] . In fact, the maximal value of f·A₄₃₀ decreases with T. This indicates that higher temperature promotes the reactions too much. It can be said in conclusion that 30˚C is the most preferable for producing the dyestuff in the water/alcohol mixed so- lution.

Figure 4 shows the time course of change in the amount of the formed dye (f·A₄₃₀) for catechinone preparation in water/ethanol mixed solution with MEA. The x_A was changed from 0.10 to 0.97. The dye formation behav- iour strongly depends on the x_A as shown in the figure. The f·A₄₃₀ increases monotonously with t at x_A = 0.10 - 0.20 and 0.69 - 0.97. On the other hand, f·A₄₃₀ increases and then decreases with t at x_A = 0.25 - 0.59. The maximal value of f·A₄₃₀ also changes with x_A of 0.25 - 0.59 and the maximum of f·A₄₃₀ is given at x_A = 0.25. The formation of colourless products and/or the precipitation of colourants as described above may be caused also here leading to the decrease in f·A₄₃₀ at x_A = 0.25 - 0.59.

Figure 5 depicts the relationship between x_A and f·A₄₃₀ obtained at 140 min ((f·A₄₃₀)_140min) in water/ethanol mixed solution with MEA. The reason why the value of f·A₄₃₀ obtained at 140 min is also adopted here, is that the same experiments were made in water and other alcohols as described below and the reactivities can be compared by using (f·A₄₃₀)_140min. The (f·A₄₃₀)_140min increases steeply from x_A = 0.10 to 0.25 and decreases from 0.28 to 0.97. The results show that the amount of formed dye depends strongly on the ethanol molar fraction and reaches its peak at x_A = 0.25. The results are interesting and it is thought that the effect of alcohol content for the reaction consists of two factors at least such as promotion and depression. Further, the same experiments were carried out using other alcohols being miscible with water such as methanol, 1-propanol, 2-propanol and tert-butyl alcohol. The results exhibit that the catechinone can be produced efficiently by adding all the alcohols used here and time change of f·A₄₃₀ for the alcohols is in a similar manner as in the water/ethanol solution. The f·A₄₃₀ for each reaction in the water/alcohol solutions increases monotonously with time, or reaches a peak and decreases. The f·A₄₃₀ value for water/1-propanol solution at x_A = 0.15 takes a downward turn after 140 min as an earliest beginning of the decrease. Therefore, the value of (f·A₄₃₀)_140min is adopted in order to make better com- parison of the whole results of f·A₄₃₀ under increasing. Figure 6 shows the relationships between the alcohol molar fraction for each of the mixed solvents (x_A) and (f·A₄₃₀)_140min for each of the reactions made in each water/ alcohol mixed solution. The (f·A₄₃₀)_140min for all the systems shows maximum against x_A. The (f·A₄₃₀)_140min represents the maximal value at 0.45, 0.25, 0.20, 0.15 or 0.10 of x_A for the MeOH, EtOH, 2-PrOH, 1-PrOH or t- BuOH-mixed system, respectively. It is, then, important to clarify the basis for the order of the x_A values. The change in (f·A₄₃₀)_140min as shown in Figure 6 might relate with the concentration of O₂ in the solution, the polar- ity of the mixed solvent, and/or a particular microstructure formed in the reaction solution. The authors study the mechanism and the reason for obtaining the results as shown in Figure 5 and Figure 6 at present.

Next, the dependence of the maximal amount of formed dye on the concentration of (+)-catechin in the reac- tion solution with the alcohols was investigated to elucidate the optimum condition for producing the dyestuff. MeOH, EtOH and 1-PrOH are convenient linear alcohols and they were employed in the experiments as at each x_A = 0.45, 0.25 or 0.15, respectively for obtaining maximum. Figure 7 exhibits the relationships between the concentration of (+)-catechin (molarity: m_S) and the maximal amount of formed dye ((f·A₄₃₀)_max) in the three

kinds of water/alcohol systems. The maximum value of f·A₄₃₀ is ca. 1600 in all the reaction systems. The m_S value to give the maximum f·A₄₃₀ for water/EtOH and water/1-PrOH systems is 0.35 mol∙kg⁻¹ and the values for water/MeOH system are 0.35 and 0.59 mol∙kg⁻¹.

The results show that the amount of formed catechinone depends on the sort of alcohols, the alcohol molar fraction and the concentration of (+)-catechin, and there are optimum x_A and m_S values to obtain the dyestuff ef- ficiently. It can be said that the EtOH is most preferable as the solvent added for the production of the dyestuff, because its safety is highest, its handling is easiest during and after the production and it is most economical.