Theoretical and Experimental Chemistry, Vol. 44, No. 5, 2008

KINETICS OF THE DEHYDRATION OF 2-PROPANOL ON MODIFIED ACTIVATED CHARCOAL CONTAINING ACID SITES

V. E. Diyuk, L. N. Grishchenko, and V. K. Yatsimirskii

UDC 661.183.2+541.128.13

A study was carried out on the kinetics of the dehydration of 2-propanol on activated bone charcoal (ABC) with supported acid sites. The rate-limiting step is the loss of water from the alcohol molecule adsorbed on the acid site. The sites containing the minimum amount of water are the most active. The reaction on these sites presumably proceeds through a concerted mechanism, which is possible at sufficiently high temperatures (above 120 °C). Key words: activated charcoal, surface modification, acid sites, dehydration of 2-propanol.

The common use of activated charcoals (AC) as supports and adsorbents is a function of their highly developed surface and porous structure, the hydrophobic nature of the carbon matrix, and their capacity to undergo chemical modification [1]. In contrast to typical acid–base catalysts such as metal oxides and cation exchangers [2-6], a functional coating not only containing chemically-attached highly acid sites but also possessing considerable thermal and hydrolytic stability in a broad pH range may be produced on the AC surface. The insignificant use of acid AC presently in catalysis is a function of the paucity of approaches and methods for the preparation of materials derived from AC with a given structure of the surface layer. This is especially true for the preparation of systems with chemically-attached surface groups. As shown in our previous work [7], samples of activated bone charcoal (ABC) with supported acid sites are extremely active catalysts for the dehydration of 2-propanol in comparison with oxide systems [2-4]: the reaction onset temperature (To) and temperature for 100% conversion of 2-propanol to propylene (T100%) are 100-110 °C and 150-170 °C, respectively (Table 1). Samples of ABC oxidized with H2O2 or HNO3 as well as samples containing polymethylstyrene (PMS) on the surface were used as the matrices for the deposition of acid sites. The surface acid sites were obtained by treating the ABC with sulfuric acid, phosphotungstic acid, sulfur vapor with subsequent oxidation by hydrogen peroxide, and also sulfonation of the samples with supported PMS by oleum [3]. The systems obtained were designated ABC/H2SO4, ABC-HNO3/H2SO4, ABC/H7[P(W2O7)6], ABC-HNO3/H7[P(W2O7)6], ABC/S-H2O2, and ABC-PMS/SO3. These samples contained a large amount of acid groups (up to 1 mmol/g) and possessed high thermal stability; decomposition of the acid groups occurred at 200-450 °C [8, 9]. In the present work, we studied the kinetics of the dehydration of 2-propanol in the gas phase on modified ABC samples. The kinetic study was carried out at 110-250 °C in a flow system using a gradientless reactor [10]. The catalyst sample was 0.25 g and the gas mixture flow rate was 45 cm3/min. The reaction mixture components were analyzed by IR spectroscopy. The study of the adsorbed forms of the reagent and product was carried out by thermal-programmed desorption with mass spectrometric detection of the products (TPDMS). The ABC samples were placed in a quartz cell attached to an MKh 7304A mass spectrometer (resolution 40000), which was evacuated to 10–4 Pa at room temperature to remove physically adsorbed substances. Then, mass spectra of the desorption products were taken at 25-900 °C at a heating rate of 10 deg/min. The desorption activation energy of the various forms were found using the following formula [11]: ___________________________________________________________________________________________________ Taras Shevchenko Kyiv National University, Vul. Volodymyrs’ka, 60, Kyiv 01033, Ukraine. E-mail: [email protected]. Translated from Teoreticheskaya i Éksperimental’naya Khimiya, Vol. 44, No. 5, pp. 321-327, September-October, 2008. Original article submitted July 29, 2008; revision submitted October 28, 2008. 0040-5760/08/4405-0331 ©2008 Springer Science+Business Media, Inc.

331

TABLE 1. Specific Surface (Ssp), Concentration of Strong Acid Sites (cA), Dehydration Onset Temperature (To), and 100% 2-Propanol Conversion Temperature (T100%), and Sample Deactivation Onset Temperature (Tdeac) Ssp, m2/g

cA·103, mol/g

To, °C

T100%, °C

Tdeac, °C

1350

–

210

–

–

ABC/H2SO4

710

0.94

110

170

>150

ABC-HNO3/H2SO4

450

1.00

100

155

>150

ABC/S-H2O2

625

0.84

110

175

>150

ABC-PMS/SO3

695

0.90

105

160

>180

1220

0.99

100

150

>400

470

1.06

100

150

>400

Sample

ABC

ABC/H7[P(W2O7)6] ABC-HNO3/H7[P(W2O7)6]

Ed = 25RTm,

(1)

where Tm is the maximum desorption temperature and R is the universal gas constant. The specific surface of the samples was determined by low-temperature nitrogen adsorption. The strong acid site concentration (cA) determined by thermal programmed desorption with IR detection of the products (TPDIR) [9], reaction onset temperature, 100% 2-propanol conversion temperature, and sample deactivation onset temperature are given in Table 1. Prior treatment and the deposition of acid sites lead to a decrease in the specific surface, especially in the case of ABC-HNO3 and the samples derived from this matrix. Independently of the prior treatment and the nature and concentration of the acid sites, the reaction onset temperatures for all the samples studied were similar and fell in the range 100-110 °C, while the range for the 100% conversion temperature was much broader (150-180 °C). The typical kinetic curves for the dehydration of 2-propanol on ABC containing various acid sites show saturation (Fig. 1a). In order to describe the mechanism for the dehydration of alcohols in the case of sulfonated cation exchange resins, we used a three-step scheme (2a) [12], entailing intermediate formation of a carbocation. Different products may be formed depending on the structure of the carbocation and reaction conditions. Still another variant of the alcohol dehydration mechanism (scheme (2b)) entails steps involving formation of alkoxy compounds [13]. These mechanisms differ, in effect, due to the strength of the Z—O and O—C bonds in the Z—O—C group. Scheme (2a) is characterized by a weak O—C bond due to the strong acid properties of Z—OH and weakness of the conjugate base Z—O–, which provides sufficient mobility of the intermediate and formation of a carbocation. Scheme (2b), which is characteristic for catalysis on oxides such as Al2O3, TiO2, and ZrO2, is typified by surface diffusion of the alkoxy compounds and parallel dehydration [14] due to the weak acidity of the oxides. In our case, propylene is virtually the sole reaction product and, thus, scheme (2a) is preferred for analyzing the experimental data although we should note that schemes (2a) and (2b) are kinetically indistinguishable. k−1 , k1 1. ZOH + (CH3)2CH—OH ← → ZOd–L[HOH—CH(CH3)2]d+, k2 2. ZOd–L[HOH—CH(CH3)2]d+  → ZO–L[CH(CH3)2]+ + H2O, k3 3. ZO–L[CH(CH3)2]+  → ZOH + CH3CH=CH2,

332

(2a)

Fig. 1. Typical kinetic curves for the dehydration of 2-propanol on ABC modified by acid sites in plots a) r = f(cal) and b) cal/r = f(cal) for ABC/H7[P(W2O7)6] samples at 1) 115, 2) 130, 3) 140, and 4) 150 °C.

k−1 , k1 1. ZOH + (CH3)2CH—OH ← → ZOHOH—CH(CH3)2, k2 2. ZOHOH—CH(CH3)2  → Z—O—CH(CH3)2 + H2O,

(2b)

k3 3. Z—O—CH(CH3)2  → ZOH + CH3CH=CH2,

where ZOH is the acid site on the surface. The second step involving formation of the carbocation is considered the slow step in scheme (2a). The first and third steps are usually considered fast and reversible. In our opinion, the third step in the gas phase under flow conditions above 90-100 °C is virtually irreversible since propylene has a low boiling point (–47.8 °C) and its concentration on the surface is much lower than the concentration of the alcohol. The flow chart corresponding to scheme (2a) is as follows: ZOH CH3CH=CH2

k3

ZOd–···[CH(CH

3)2

]d+

k1c((CH3)2CHOH)

k–1 k2

ZOd–···[HOH—CH(CH

(3) 3)2

]d+

H2O

where ZOd–···[HOH—CH(CH3)2]d+ is the protonated alcohol molecule on the surface and ZO–···[CH(CH3)2]+ is the adsorbed carbocation. The kinetic equation obtained using chart (3) may be written as follows r=

c al  1 1 1 1 + + c al  +  k1 K 1 k 2  k2 k3 

(4)

where K1 is the equilibrium constant of the first step of scheme (2a) and cal is the alcohol concentration. In order to check the validity of Eq. (4) for the experimental data and find the constants in scheme (2a), the above equation was linearized (Eq. (5)), which permitted us to find k ef ′ and k ef ′′ :

333

TABLE 2. Kinetic Parameters of the Dehydration of 2-Propanol on ABC Modified by Acid Sites Sample

ABC/H7[P(W2O7)6]

ABC-HNO3/H7[P(W2O7)6]

ABC/H2SO4

ABC-HNO3/H2SO4

ABC/S-H2O2

ABC-PMS/SO3

c al r where k ′ef =

T, °C

K1·10–3, L/mol

k2·105, mol/min

Ea, kJ/mol

cN·108, mol/g

115

1.8

1.0

81 ± 8

1.2

130

1.4

1.9

140

1.1

5.0

150

1.1

7.1

110

5.2

1.0

76 ± 8

0.7

120

3.0

2.1

130

1.5

5.3

145

1.5

7.1

130

1.8

0.7

87 ± 9

3.3

140

2.0

1.3

145

2.4

1.7

120

2.1

0.4

114 ± 6

1.8

127

1.7

0.7

134

1.8

1.1

140

2.2

1.4

145

3.9

0.6

89 ± 7

4.5

155

3.5

1.1

160

3.2

1.9

172

3.0

2.9

128

2.7

1.2

85 ± 6

2.5

140

2.1

2.5

150

1.8

5.2

167

1.5

11.0

=

 1 1 1 1 + + c al  +  = k ef ′ + k ′′ef cal k1 K 1 k 2  k2 k3 

(5)

1 1 1 1 , while k ef + ′′ = + . k1 K 1 k 2 k2 k3

The data given in Fig. 1b show that the experimental curves are satisfactorily linearized in coordinates of Eq. (5). The slope of the lines (k ′ef ) and intercept (k ef ′′ ) decrease with increasing temperature (Fig. 1b). In comparing the terms comprising k ef ′′ , we should note that the contribution of 1/k2 to the effective constant is much greater since the second step of scheme (2a) is the slow step. Thus, we may assume that k2 » 1/k ef ′′ . Clearly, the ratio k–1/k2 is much greater than unity and the contribution of the second term to k ′ef is much greater. Thus, K1 » 1/(k ′ef k2). For all the samples studied, we used Eq. (5) and the above assumptions to find the rate constant and activation energy for conversion of the 334

Fig. 2. Typical thermal desorption spectra of the samples with supported acid sites before (a) and after catalysis (b): 1) H2O, 2) CO2, 3) SO2, 4) CO, 5) [CH3—CH2—O]+, 6) CH2=CH—C+H2. Sample: ABC-PMS/SO3.

protonated alcohol molecule into a carbocation with loss of a water molecule and also the equilibrium constant for the alcohol protonation step (Table 2). The values for k2 correlate with the kinetic activity data (Table 1). Thus, the most active samples have the highest rate constants for the second step. The equilibrium constants found for the first step show that the equilibrium is steadily shifted toward the formation of the protonated form of the alcohol with decreasing temperature and increasing thermal stability of the acid site. The activation energy for conversion of the protonated alcohol molecule to the carbocation on most of the modified ABC in the temperature range studied is in the range 80-90 kJ/mol. This value is somewhat lower when using the acidified matrices. The activation energy of this step may be found independently by the TPDMS method using Eq. (1) since this step is virtually only the desorption of water from the acid site containing the adsorbed reagent. Figure 2 shows parts of the thermal desorption spectra typical for samples with supported acid sites before and after catalysis. The TPD spectra of ABC samples modified by acid sites show the presence of water, CO2, and CO prior to the reaction (Fig. 2a, curves 1, 2, 4, respectively) and also the product of the decomposition of the SO2 acid sites for the sulfur-containing samples (curve 3). After use of the samples in the catalysis (Fig. 2b), the TPD spectra also show decomposition fragments of the reagent [CH3—CH2—O]+ and the product CH2=CH—C+H2. The desorption peaks of the fragments of both the alcohol (reagent) and propylene (product) are rather narrow, which indicates that the reaction proceeds on virtually energetically homogeneous acid sites and the impossibility of the reaction occurring on other sites such as the carboxyl groups of the matrix. Comparison of the curves for the release of SO2 shows that the conversion of unstable acid sites having decomposition maxima at 200-210 °C occurs as the result of the reaction due to the action of the reaction medium. The decomposition of these sites is very fast and the kinetic curves obtained correspond to catalysis on the stable acid sites with decomposition maximum at about 300 °C. The stability of a part of the sites is indicated by the invariance of the reaction rate under isothermal conditions over 4 h. The presence of two broad and overlapping water desorption maxima at about 80-90 and 150-165 °C is common for all the samples containing acid sites, which indicates the existence of several forms of adsorbed water. A characteristic feature of water desorption at 50-220 °C is that it is independent of the release of CO and CO2, indicating that the water species are not formed upon decomposition of the functional groups of the carbon matrix but rather are bound to the strong acid sites. The range of activation energies found using Eq. (1) and the maximum desorption temperatures (80-165 °C) required for the removal of water from the solvated acid site is 70-95 kJ/mol. This range includes the activation energies found using the rate constant of the second step in scheme (2a), which also supports our assumptions concerning the kinetic scheme and ratio of the rate constants of the steps in this scheme. Water as an active promoter of proton transfer [15] participates in the formation of reactive acid sites for the dehydration of 2-propanol on various catalysts including sulfonated cation exchangers [16]. However, the virtually complete removal of water in the case of sulfonated cation exchangers leads to a decrease in the catalytic activity of the systems 335

containing sulfonic acid groups due to hindrance to proton transfer. In the case of a slight decrease in the water content in the system, a concerted mechanism involving the concurrent reaction of the reagent molecule with several acid groups becomes more likely [17, 18]. In the case of acid sites supported on ABC, the maximum desorption temperatures of the low-temperature water species and 2-propanol dehydration onset temperature are similar (Table 1), indicating difficulty, in contrast to sulfonated cation exchangers, for the reaction on the strongly solvated acid site by a proton transfer mechanism. Indeed, the ABC surface below 90 °C contains a significant amount of physically adsorbed water, which virtually blocks the acid sites. The desorption maximum temperatures of the high-temperature water species at 150-165 °C are almost identical to or 5-10 °C higher than the 100% 2-propanol conversion temperature and probably correspond to the release of the last water molecule from the acid site. The rather broad water release maxima (half-width 150-200 °C) suggest the possibility of different solvation of the acid site depending on the reaction temperature and concentration the reaction participants, while the finding of a maximum on the temperature dependence curves for the dehydration of 2-propanol [8] may be attributed to a decrease in the activity of the site upon the complete removal of water. The reaction of water with the active site leads to significant hydration of the site and a decrease in the acidity as a consequence of the reaction below: —SO3H + H2O « —SO −3 + H3O+.

(6)

An increase in temperature clearly shifts the equilibrium of reaction (6) toward the left and decreases the water content in the solvation shell of the active site, which facilitates a concerted mechanism for the reaction involving the concurrent reaction of the alcohol molecule with at least two sites. Thus, the probable explanation for our results lies in the assumption of a concerted mechanism, which may be represented by the following general scheme for sulfonated ABC:

where X—H is a polar group capable of removing a proton. This group may be attached to the surface (—SO3H, —CO2H, —OH) or be an adsorbed molecule (H2O, H2SO4). Support for a concerted mechanism for this reaction and a positive effect of the interaction between the sulfonic acid groups on the reaction is found in the moderate decrease in activation energy for the step involving loss of water from the adsorbed reagent when using the acidified matrices. The decrease in the activation energy by about 10 kJ/mol in this case is related, in our opinion, to the interaction of the acid sites and polar groups of the matrix leading to a more facile loss of water from the protonated alcohol molecule through a concerted mechanism [5]. The concentration of the active sites (cN) may be found independently from the temperature dependence of k2 assuming that the entropy of the system hardly changes during the second step of scheme (2a). For all the systems studied, cN (see Table 2) is in the range from 1·10–8 to 5·10–8 mol/g, which is approximately four orders of magnitude less than the concentration of strong acid sites (cA) determined by TPDIR. This discrepancy is related to three major factors: 1) the formation of water rather strongly bound to the active site in significant amounts leads to temporary deactivation of the site, 2) in the case of the concerted mechanism, the concentration of active sites containing at least two acid groups is much lower than cA, and 3) as a consequence of the microporous structure of the starting ABC, far from all the acid sites are available for the reaction. These results suggest ways to enhance the catalytic activity of such AC systems, namely, by using mesoporous AC, increasing 336

the hydrophobicity of the AC surface, and the development of methods providing for an “insular” distribution of the supported acid sites. Thus, the dehydration of 2-propanol on modified ABC is described by scheme (2a) and takes place on sites with different environment depending on the temperature. The sites containing the minimal amount of water proved most active. The concerted mechanism, which is possible only at relatively high temperatures (>120 °C), takes place predominantly on such sites. The rate-limiting step of the reaction occurring on modified ABC is the loss of a water molecule from the adsorbed protonated reagent molecule. The active site is formed by at least two acid groups interacting through an adsorbed molecule or group capable of removing a proton form the active site.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

V. B. Fenelonov, Porous Carbon [in Russian], Institute of Catalysis, Novosibirsk (1995). F. J. Tzompantzi, M. E. Manríquez, J. M. Padilla, et al., Catal. Today, 133-135, 154-159 (2008). M. A. Abdel-Rahim, A. C. B. dos Santos, V. L. L. Comorim, and A. da Costa Faro, Jr., Appl. Catal. A, 305, 211-218 (2006). T. Onfroy, G. Clet, S. B. Bukallah, et al., Appl. Catal. A, 298, 80-87 (2006). T. S. Petkevich, G. K. Berezovik, T. L. Sen’ko, and Yu. G. Egiazarov, Vestsi Nats. Akad. Navuk Belarusi. Ser. Khim. Navuk, No. 1, 24-28 (2000). T. S. Petkevich, G. K. Berezovik, T. L. Sen’ko, and Yu. G. Egiazarov, Vestsi Nats. Akad. Navuk Belarusi, Ser. Khim. Navuk, No. 1, 56-61 (1999). V. E. Diyuk, T. M. Bezugla, V. L. Budarin, et al., Physical Chemistry of Condensed Systems and Phase Boundaries [in Ukrainian], Vol. 1, Vyd. Kiev University, Kiev (2003), pp. 42-46. V. K. Yatsimirskii, L. N. Gomonyuk, T. N. Bezugla, and V. E. Diyuk, Ukr. Khim. Zh., 73, Nos. 1/2, 25-31 (2007). V. E. Diyuk, L. M. Grishchenko, A. M. Savyts’ka, and V. K. Yatsymyrs’kii, Vopr. Khim. Khim. Tekhnol., No. 2, 96-101 (2008). G. P. Korneichuk, V. A. Ostapyuk, and N. A. Boldyreva, Catalysis and Catalysts [in Russian], Vol. 22, Naukova Dumka, Kiev (1984), pp. 77-79. M. U. Kislyuk and V. V. Rozanov, Kinet. Katal., No. 1, 89-98 (1995). P. A. Sykes, A Guide Book to Mechanism in Organic Chemistry, John Wiley & Sons, New York (1970). O. V. Krylov, Heterogeneous Catalysis [in Russian], Akademkniga, Moscow (2004). M. E. Manríquez, T. López, R. Gómez, and J. Navarrete, J. Mol. Catal. A, 220, 229-237 (2004). M. I. Vinnik and P. A. Obraztsov, Usp. Khim., 59, No. 1, 105 (1990). B. Kh. Cherches, G. N. Lysenko, and Yu. G. Egiazarov, Zh. Prikl. Spektr., 67, No. 5, 587-590 (2000). B. C. Gates, J. S. Wisnouskas, and H. W. Heath, J. Catal., 24, No. 2, 320 (1972). E. A. Bekturov and S. Kudaibergenov, Catalysis by Polymers [in Russian], Nauka, Alma Ata (1988).

337