A New Method for Producing Substituted Anthraquinones via Diene

ISSN 2070-0504, Catalysis in Industry, 2016, Vol. 8, No. 4, pp. 310–315. © Pleiades Publishing, Ltd., 2016. Original Russian Text © L.L. Gogin, E.G. Zhizhina, Z.P. Pai, 2016, published in Kataliz v Promyshlennosti.

CATALYSIS IN CHEMICAL AND PETROCHEMICAL INDUSTRY

A New Method for Producing Substituted Anthraquinones via Diene Synthesis in the Presence of Mo-V-P Heteropoly Acid Solution: Catalyst Regeneration L. L. Gogin*, E. G. Zhizhina**, and Z. P. Pai*** Boreskov Institute of Catalysis, Siberian Branch, Russian Academy of Sciences, Novosibirsk, 630090 Russia *e-mail: [email protected] **e-mail: [email protected] ***e-mail: [email protected] Received April 18, 2016

Abstract—The possibilities of developing catalytic processes for the synthesis of 2-methylanthraquinone from 1,4-naphthaquinone (NQ) and isoprene, and synthesizing 2,3-dimethylanthraquinone from NQ and 2,3-dimethylbutadiene in the presence of aqueous solutions of high-vanadium heteropoly acid (HPA) as a bifunctional catalyst are examined. Two methods for catalyst regeneration are discussed: oxidation with oxygen at PO 2 = 0.3–0.4 MPa and with concentrated nitric acid at atmosphere pressure. It is shown that regeneration with nitric acid ensures deeper oxidation of the reduced catalyst. Complete restoration of the properties of the catalyst following regeneration offers the possibility of its repeated use in the processes discussed in this work. Keywords: substituted anthraquinones, heteropoly acid, regeneration DOI: 10.1134/S2070050416040048

INTRODUCTION 9,10-Anthraquinone (AQ) and its derivatives are important products of organic synthesis [1–3]. They are used in the production of dyes [1], hydrogen peroxide [2], medicinal preparations [3], and as catalysts for wood delignification [3]. The industrial production of 2-substituted AQ is usually performed via synthesis from phthalic anhydride and respective monosubstituted benzenes producing о-benzoylbenzoic acids, followed by their cyclization in the presence of strong acids [2]. However, this method is accompanied by the release of large amounts of acidic waste. Non-substituted AQ is used as the initial raw material for the synthesis of 1- and polysubstituted AQ, which is subsequently transformed into sulfo- or nitro-derivatives followed by modification of the side chains [2]. The multiple stages of this process are a considerable drawback. The method of production used for the synthesis of unsubstituted AQ via the oxidation of anthracene obtained from coal tar is unsuitable, due to inaccessibility of the raw material (substituted anthracenes) [2]. We used Mo-V-phosphoric heteropoly acids (HPA) in seeking new ways of synthesizing substituted AQ because their unique properties allow us to conduct a variety of catalytic processes in their presence [4–8]. In [9], we developed one-pot processes for the production of 9,10-anthraquinone (AQ) and its substituted analogs from 1,4-naphtoquinone (NQ) and

1,3-dienes in the presence of high-vanadium modified (non-Keggin) heteropoly acid solutions (HPA-x, where x is the number of V atoms) with brutto-compositions H15P4Mo18V7O89 (HPA-7) and H17P3Mo16V10O89 (HPA-10) that allowed products to form under mild conditions and the complete conversion of NQ with yields of up to 90% and main substance contents of 97−99%. HPA-х solutions, being simultaneously strong Brønsted acids and rather strong oxidizers with redox potentials (Е) of approximately 1.0 V [10], play the role of bifunctional catalysts in these processes: an acid catalyst of diene synthesis and a catalyst of the oxidation of the resulting adducts. The reactions of substrate (Su) oxidation in the presence of HPA-х are usually conducted in separate technological stages (1) and (2) in different reactors. The sum of these reactions yields catalytic reaction (3) of substrate oxidation by oxygen. Reactions (1) + (2) comprise the catalytic cycle of reaction (3):

m 2 Su + m 2 H 2O + HPA- х → m 2 SuO + Н mHPA- х,

(1)

НmHPA-х + m/4O2 → HPA-х + m/2H2O,

(2)

Su + 1/2O2 → SuO.

(3) Conducting reactions (1) and (2) separately ensures high selectivity of target reaction (1). Considering that the reduced solutions of the catalyst (HmHPA-x, where m

310

A NEW METHOD FOR PRODUCING SUBSTITUTED ANTHRAQUINONES

is the number of electrons accepted by an HPA-x molecule)1 consists of a complex equilibrium mixture of anions [Hx + m − 1PV mIV V xV− mMo12 − xO40]4− (HPA-хm) with different integer values of m (including m = 0) and cations VO2+ and VO 2+ [9], the average degree of reduction of these solutions m = [VIV]/[HmHPA-x], where [VIV] is the total concentration of VIV and [HmHPA-x] is the total concentration of HP anions. The latter is equal to the initial concentration of HPA-х prior to reaction (1). During regeneration of the catalyst solution via reaction (2), vanadium (IV) in the composition of the HP-anions is oxidized to vanadium (V), and the degree of HPA-х reduction falls. Hence, HPA-х plays the role of reversibly acting oxidizer (essentially a catalyst) in catalytic process (3) of substrate oxidation by atmospheric oxygen, where the vanadium atoms are subjected to redox transformations: VV → VIV → VV. The productivity of such a catalyst or its oxidative capacity (the amount of substrate that can be oxidized during one cycle by one liter of the catalyst) depends on the concentration of V(V) in the solution prior to reaction (1). Hence, complete regeneration of the catalyst is required during multicycle testing according to scheme (1) + (2) to ensure the highest concentration of V(V) in the solution so that the amount of the substrate potentially oxidized by the HPA-х solution in the next cycle is at the maximum possible level. We showed in [11] that the physicochemical properties of Мо-V-Р HPA solutions (Е, рН, density ρ, viscosity η) were completely restored following regeneration. In addition, the possibility of using HPA-х solutions as catalysts of oxidation is due entirely to their thermal stability, since the oxidation of reduced HPA-х solutions using oxygen (air) is usually performed under pressure at temperatures above 140°C [11]. The problem of thermal stability was most severe for highvanadium HPA, which were most promising as catalysts. In recent years, we have developed ways of synthesizing high-vanadium HPA solutions with modified (non-Keggin) compositions [12] that exhibited enhanced thermal stability. These solutions retained their homogeneity in the course of multiple regeneration by oxygen that proceeds rapidly at temperatures 160−170°C and PO2 = 0.3–0.4 MPa [10]. It is these modified solutions of HPA that offer great promise for the development of two-step catalytic oxidation processes according to scheme (1) + (2). It should be noted that in addition to reaction (2), other approaches are possible for the regeneration of catalyst solutions based on Мо-V-Р HPA (see below). 1 The

reduction of m electrons in the form of НmHPA-x contains x atoms of vanadium, including m atoms of vanadium (IV); i.e. 0 ≤ m ≤ x. CATALYSIS IN INDUSTRY

Vol. 8

No. 4

2016

311

In this work, we present results obtained from investigating the process for preparing substituted AQ using the abovementioned solution of HPA-10 as a bifunctional catalyst. The possibility of its multicycle application is explored. In addition to the traditional regeneration of HPA-10 solution under oxygen pressure, the possibility of using additions of small amounts of concentrated nitric acid was also investigated. The experimental data were obtained while investigating two innovative processes: (1) preparing 2-methylanthraquinone (2-MAQ) from NQ and isoprene, (2) preparing 2,3-dimethylanthraquinone (2,3DMAQ) from NQ and 2,3-dimethylbutadiene. EXPERIMENTAL 1,4-Naphtoquinone (97% main substance), isoprene (99% main substance), and 2,3-dimethylbutadiene (98% main substance) produced by Alfa Aesar were used in the synthesis of anthraquinones. A 0.2 M solution of brutto-composition H17P3Mo16V10O89 (HPA-10, molecular mass 3580 g/mol, concentration 40 wt %), synthesized as described in [12], was used in our experiments. The resulting HPA-10 solution was characterized by E0 = 1.098 V and pH0 = −0.26. The reaction products were analyzed via HPLC using a Pro Star liquid chromatograph equipped with a Pro Star 410 AutoSampler, Pro Star 210 and Pro Star 218 solvent delivery modules, a Varian 500-LC column valve module, and a Photodiode Pro Star 335 UV detector (wavelength, 247 nm). The reaction products were separated on a Pursuit 3C18 column (247 × 4.6 mm) with an eluent flow rate of 1 mL/min. The solvents used for chromatography were methanol (from J.T. Baker, 99%, UV-IR-HPLC); trifluoroacetic acid (from Acros Organics, 99%, chemically pure grade), and deionized water. The eluent composition was 70% CH3OH + 30% CF3COOH (chloroform was used as the solvent for all samples). Commercially available samples of 2-MAQ and 2,3-DMAQ from Alfa Aesar and Sigma Aldrich with main substance contents of no less than 97% were used as reference samples; duroquinone (2,3,5,6-tetramethyl benzoquinoine) of the same purity from Alfa Aesar was used as our inner standard. The IR spectra of products were recorded in KBr on an IR Affinity-1 spectrometer (Shimadzu). Substituted AQ was synthesized in a thermostated glass reactor connected to a long reflux condenser. An aliquot (0.2 g) of NQ was introduced into the reactor, followed by the addition of 4 mL of 1,4-dioxane under stirring; after the substrate dissolved, 4 mL of 0.2 M solution of HPA-10 was added along with a 40% excess of diene. Next, the reactor jacket was connected to a thermostat preheated to 80 °C. All synthesis was conducted under intensive stirring of the reaction mixture

312

GOGIN et al.

with a magnetic stirrer (650 rpm) for 7 h. The HPA-10 solution was reduced during synthesis, changing color from dark red to green, and a precipitate of a weakly soluble product formed. The reaction mixture was doubly diluted with distilled water upon completion of the reaction for more complete product precipitation. The solid precipitate was filtered on a porous filter, washed with water to a neutral pH, and dried under vacuum over P2O5. The dry precipitate was weighed and analyzed via HPLC. The yield of substituted AQ (YАQ) was calculated using the equation YАQ = (M × CАQ × 100)/g, where М is the weight of the dry product precipitate, g; CАQ is the fraction of the substituted AQ in the precipitate according to our analysis; g is the theoretically possible yield of substituted AQ with the complete conversion of NQ, g. The catalyst was regenerated after reaction (1) either with oxygen (air) at temperatures of 150–160°C and PO2 = 0.4 MPa [9], or by boiling the solution in air with small amounts of concentrated nitric acid added (see tables). The HPA-10 solution was regenerated with HNO3: The solvent and excess water were evaporated from a solution of HPA-10 reduced to a volume of approximately 10 mL, and 0.15−0.2 mL of concentrated nitric acid (ρ = 1.405 g/cm3, 15.18 mol/L) was added dropwise. The solution was then boiled for at least 30 min (while maintaining the volume), followed by evaporation to the initial volume of 4 mL. The solution of the catalyst meanwhile changed color from green to red. The completeness of the HPA-10 solution’s oxidation was controlled using the value of its redox potential E, and its pH value. Universal dependencies E = f (m) and рН = f' (m) were used as calibration curves, with the former being more accurate [10]. The values of E and pH were measured at room temperature using an InoLab pH 730 pH meter (Wissenschaftlich Technische Werkstatten GmbH, WTW). A combination Sen Tix 41 pH electrode, calibrated at 25°С using buffer solutions with рН = 4.01 and 7.00, was used for pH measurements. In order to measure Е, this electrode was used in combination with a SenTix ORP Pt platinum electrode. The stability of Е was attained within 1 min, with an accuracy in the range of ±0.001 V. The stability of pH values was attained within 1–2 min with an accuracy in the range of ±0.01 pH units. The synthesis of substituted AQ was repeated using the regenerated solution of the HPA-10 in the next cycle. RESULTS AND DISCUSSION Our one-pot processes for the production of 2-MAQ and 2,3-DMAQ are presented in schemes А

(A)

O

O

СH3 +

СH3

HPA-10

O O

O O

(B)

+

СH3

СH3

СH3

СH3

HPA-10

O

O

Fig. 1. Schemes of one-pot processes for the synthesis of (A) 2-methylanthraquinone and (B) 2,3-dimethylanthraquinone in the presence of aqueous solutions of HPA-10.

and В (Fig. 1). The HPA-10 solutions in these processes display bifunctional properties in playing a role of acid catalysts at the stage of diene synthesis and immediately afterward oxidizing the produced adducts into substituted anthraquinones. Upon the completion of processes А and В, the solution of reduced HPA-10 was regenerated (oxidized) in a separate step after insoluble products were separated. Individual portions of the HPA-10 solution were used to investigate each of these processes. It was important to use the HPA-10 solutions correctly in order to evaluate the possibility of their multicycle application as bifunctional catalysts in the synthesis of substituted AQ. Since the HPA-10 solution is a concentrated high-vanadium solution, its evaporation above a concentration of 0.3 M cannot be accepted. In addition, this solution cannot be used at temperatures above 170°C. Regeneration of the catalyst at higher temperatures results in the gradual precipitation of vanadium oxide hydrates [11], reducing the oxidative capacity of the catalyst (see above) from cycle to cycle. As was shown [11, 12], however, efficient catalytic one-pot processes for the production of substituted AQ can be conducted in their presence if the solutions are treated with care. A strongly reduced solution of HPA-10 can ultimately be partially oxidized with a 30% hydrogen peroxide solution, an environmentally friendly oxidizer. We used this method of regeneration in our studies. However, it should be remembered in using this procedure that HPA-10 oxidation in the presence of hydrogen peroxide does not ensure deep levels of oxidation, since vanadium ions catalyze the decomposition of H2O2 into oxygen and water. The oxidation potential does not exceed 0.945 V following such regeneration, so the oxidative capacity of the catalyst is in this case greatly reduced starting from the second cycle. In addition, any organic admixtures remaining in the catalyst solution after reaction (1) will not be fully oxidized during regeneration and will gradually CATALYSIS IN INDUSTRY

Vol. 8

No. 4

2016


313

Table 1. Results from multicycle experiments on the production of 2-methylanthraquinone in the reaction of isoprene with NQ in the presence of HPA-10 solution Е value of HPA-10 solution Cycle Yield Main substance prior to reaction (1), Method of catalyst solution regeneration number of 2-MAQ, % (2-MAQ) content, % V (NHE) 1

1.098

74

97

160°C, PO 2 = 0.4 MPa, 15 min

2

1.066

70

97

150°C, PO 2 = 0.4 MPa, 20 min

3

1.064

69

97

Boiling in air with the addition of 0.15 mL of concentrated HNO3, 30 min

4

1.102

73

97


5

1.099

72.5

97

160°C, PO 2 = 0.4 MPa, 15 min

6

1.068

68

97


7

1.100

74

97

160°C, PO 2 = 0.4 MPa, 15 min

Conditions: Temperature, 80°C; 4 mL of 0.2 М solution of HPA-10; molar ratio, HPA-10 : NQ = 0.5 : 1; volume ratio, HPA-10 solution : 1,4-dioxane = 1 : 1; reaction time, 7 h. NQ conversion in all experiments was ≥99%. The Е value here and below is given relative to the normal hydrogen electrode.

(from cycle to cycle) accumulate in the HPA-10 solution, lowering the Е value prior to reaction (1) and thus reducing the catalyst’s productivity. Even though hydrogen peroxide does not ensure complete regeneration of the HPA-х, such periodic oxidation of the catalyst in developing the process technology can nevertheless be very advantageous for substantially enhancing catalyst productivity by increasing its oxidative capacity (the yield of the product from 1 L of solution per cycle). Concentrated nitric acid was also used in our studies on HPA-х regeneration. Even though this oxidizer is not environmentally friendly, it can be used in small amounts in oxidizing reduced HPA-х solutions with atmospheric oxygen under certain conditions. It is convenient that catalyst regeneration can be performed at atmospheric pressure using HNO3. When admixtures of organic compounds (side products) remain in the catalyst solution from the previous cycle, the addition of nitric acid at the end of regeneration allows their complete oxidation. Our data on the multicycle testing of the catalyst in the process of 2-MAQ production according to scheme A are presented in Table 1. The product was synthesized in seven cycles using freshly synthesized HPA-10 solution in the first cycle, and regenerated solutions in subsequent cycles. The yield of the product, the content of the main substance, and the identity of the samples produced in the experiments with the freshly prepared and regenerated HPA-10 solutions was determined upon the completion of reaction (1) in each cycle. Their identity was confirmed by comparing the IR spectra of the samples. Regeneration of the catalyst with redox potential CATALYSIS IN INDUSTRY

Vol. 8

No. 4

2016

after reaction (1), usually in the range 0.830−0.843 V, was performed either in the autoclave or by oxidizing the solution with atmospheric oxygen and additions of concentrated nitric acid. In earlier investigations of the physicochemical properties of modified (non-Keggin) solutions of МоV-Р HPA, we found that the state of these solutions in redox reactions (1) and (2) could be determined from the dependences of their redox potential Е and pH on the concentration of V(IV) or on the value of m. Dependences Е = f ([V(IV)]) and pH = f'([V(IV)]) are reproduced for HPA solutions of certain compositions and concentrations. As was noted above, the E = f ([V(IV)]) dependences are more accurate, which is why they were used in this work as calibration curves for the rapid determination of the V(IV) concentrations in the investigated HPA-10 solutions. Note that the HPA-10 solution is not consumed in this method of [V(IV)] determination, which is important for its multicycle use. We may conclude from an analysis of the data in Table 1 that the 0.2 M solution of HPA-10 works steadily as a bifunctional catalyst in the process of the 2-MAQ production over 7 cycles. The catalyst ensures 68–74% yields of the product with main substance contents of 97%. The catalyst was regenerated in cycles 1, 2, 5, and 7 under oxygen pressure; in cycles 3, 4, and 6, it was regenerated in air with the addition of concentrated HNO3. It can be seen that regeneration with the addition of HNO3 ensured deeper oxidation of the reduced HPA-10 solution (higher Е values before each cycle) than regeneration with oxygen in the autoclave. The oxidative capacity of the catalyst remained virtually the same as in the first cycle when

314

GOGIN et al.

Table 2. Results from multicycle experiments on the production of 2,3-dimethylanthraquinone via the reaction of NQ and 2,3-dimethylbutadiene in the presence of HPA-10 solution Е value of the HPA-10 Content of main Cycle Yield solution prior substance Method of catalyst solution regeneration number of 2,3-DMAQ, % to reaction (1), V (NHE) (2,3-DMAQ), % 1

1.098

81

98

150°C, PO 2 = 0.4 MPa, 20 min

2

1.063

77

98

160°C, PO 2 = 0.4 MPa, 15 min

3

1.066

77

97.5


4

1.101

80.5

98


5

1.097

80

98.5

160°C, PO 2 = 0.4 MPa, 15 min

6

1.063

77

98


7

1.104

81

98

160°C, PO 2 = 0.4 MPa, 15 min

Conditions: Temperature, 80°C; 4 mL of 0.2 М HPA-10 solution; molar ratio, HPA-10 : NQ = 0.5 : 1; volume ratio, HPA solution : 1,4-dioxane = 1 : 1; reaction time, 7 h. NQ conversion in all experiments was ≥99%.

this method of regeneration was used, since the Е values after regeneration were very close to that of the freshly synthesized HPA-10 solution used in the first cycle. Note that the HPA-10 solution remained homogenous throughout multicycle testing, the formation of vanadium-containing precipitates was not observed, and the productivity of the solution did not decline. However, such experiments must be conducted very carefully in order to prevent losses of the catalyst solution, especially during the filtration of 2-MAQ. Data on the multicycle testing of the catalyst in the production of 2,3-DMAQ according to scheme В are presented in Table 2. The product was synthesized over 7 cycles, as in the synthesis of 2-MAQ with the freshly synthesized HPA-10 solution used in the first cycle and the regenerated solutions in subsequent cycles. The catalyst was also regenerated in two ways (see Table 2). The data in Table 2 show that the HPA-10 solution worked steadily as a bifunctional catalyst over 7 cycles of the process for the production of 2,3-DMAQ, as in the production of 2-MAQ. The catalyst in this process ensured a slightly higher yield of the product (77−81%) and a slightly higher content of the main substance (98%). Based on the values of redox potential Е, regeneration of the HPA-10 solution with the addition of HNO3 ensures virtually the initial productivity of the catalyst in the next cycle. These results confirm that the physicochemical properties of Мо-V-P HPA solutions are restored completely in during redox transformations according to scheme (1) + (2). Isolated products 2-MAQ and 2,3-DMAQ have high contents of the main substance, which agrees with the data from HPLC and IR spectroscopy. Con-

sidering that the yield of anthraquinones was not quantitative (only 75–80%) with virtually complete conversion of the NQ, we may assume that the side products remaining in the catalyst solution accounted for the remaining 20–25%. Further studies are needed to determine the nature of these products in each particular case. It is important to determine through the development of one-pot processes that proceed according to schemes А and В how the yield and purity of the product change in the next cycle after adding nitric acid to the HPA-10 solution in the regeneration step. The IR spectra of the 2,3-DMAQ samples synthesized in the freshly prepared and regenerated solutions of HPA-10 were identical (Fig. 2). A comparison of the IR spectra shows that traces of nitric acid in the HPA-10 solution did not affect the course of 2,3-DMAQ synthesis. The residual concentration of HNO3 in the catalyst solution after regeneration was low (below 0.05 М), so no side nitration processes occurred during the next cycle. CONCLUSIONS The possibility of developing new catalytic one-pot processes for synthesizing 2-methylanthraquinone from 1,4-naphtoquinone and isoprene, along with 2,3-dimethyl anthraquinone from 1,4-naphtoquinone and 2,3-dimethylbutadiene in the presence of solutions of high-vanadium HPA-10 has been demonstrated. The HPA-10 solutions in these processes play the role of a bifunctional catalyst: an acid catalyst of the reaction of diene synthesis, and a catalyst of the oxidation reaction of the resulting adducts. Multicycle testing (seven cycles for each process) showed that the CATALYSIS IN INDUSTRY

Vol. 8

No. 4

2016


315

Wavenumber 4000

3500

3000

2500

2912.3

100

1500

1000

500 815.96 750.38

1781.7 1471.3

1099

2646.1 2922.4

3930.5

80

1809.9

1 3064

Transmission

2000 2

2001.6 1373.5

1866

2961.3 3051.1 3231.2

60

1102.6 1029 1169.2

1394.7

79916 900.9

1447

3479.5

1184.9

615.12

40 1251.5 1637.6 71543

1615.3

3656.2 3414.7

20

1229 1582.2

998.12

1295.4

1380.2

0

1570.6

Fig. 2. Infrared spectra of the products obtained from 2,3-dimethylbutadiene and NQ using (1) freshly prepared and (2) regenerated HPA-10.

catalyst was stable: It remained homogenous at all steps of the processes. HPA-10 solution can easily be regenerated with oxygen at temperatures of 150–160°C and PO2 = 0.4 MPa, and with atmospheric oxygen by boiling the reduced solution with additions of nitric acid. Regeneration with nitric acid results in deeper oxidation of reduced solutions of HPA-10, ensuring higher Е values before the next cycle. The complete oxidation of side products continues in the catalyst solution after the target reaction occurs during the process. Target products obtained with freshly prepared and regenerated samples of high-vanadium HPA-10 solution are completely identical. ACKNOWLEDGMENTS This work was supported by the Russian Academy of Sciences and the Federal Agency of Scientific Organizations (project no. V.44.2.8). REFERENCES 1. Anthraquinone dyes and intermediates, in Ullmann’s Encyclopedia of Industrial Chemistry, 2007, vol. A. 2. Anthraquinone, in Kirk-Othmer Encyclopedia of Chemical Technology, 2006, vol. 2, 5th ed.

CATALYSIS IN INDUSTRY

Vol. 8

No. 4

2016

3. Abdel-Hamid, A.M., Solbiati, J.O., and Cann, I.R.O., Adv. Appl. Microbiol., 2013, vol. 82, pp. 1–28. 4. Kozhevnikov, I.V., Catalysis for Fine Chemical Synthesis, Chichester (UK): John Wiley & Sons, 2002. 5. Hill, C.L. and Kholdeeva, O.A., in Liquid Phase Oxidation via Heterogeneous Catalysis, Clerici, M.G., Kholdeeva, O.A., Eds., Hoboken, NJ: John Wiley & Sons, 2013, ch. 6. 6. Misono, M., Stud. Surf. Sci. Catal., 2013, vol. 176, pp. 97–155. 7. Védrine, J.C. and Millet, J.-M.M., in Metal Oxide Catalysis, Jackson, S.D. and Hargreaves, J.S.J., Eds., Weinheim: Wiley-VCH, 2009, vol. 2, pp. 561–594. 8. López, X., Carbó, J.J., Bo, C., and Poblet, J.M., Chem. Soc. Rev., 2012, vol. 41, no. 22, pp. 7537–7571. 9. Zhizhina, E.G. and Odyakov, V.F., ChemCatChem, 2012, vol. 4, no. 9, pp. 1405–1410. 10. Odyakov, V.F., Zhizhina, E.G., and Matveev, K.I., Zh. Neorg. Khim., 2000, vol. 45, no. 8, p. 1258. 11. Zhizhina, E.G. and Odyakov, V.F., Appl. Catal., A, 2009, vol. 358, no. 2, pp. 254–258. 12. Odyakov, V.F., Zhizhina, E.G., and Maksimovskaya, R.I., Appl. Catal., A, 2008, vol. 342, nos. 1–2, pp. 126–130.

Translated by L. Brovko