Catalytic Hydroprocessing of p-Cresol: Metal, Solvent and Mass-Transfer Effects

Abstract

A systematic study of the comparative performances of supported Pt, Pd, Ru and conventional CoMo/Al₂O₃, NiMo/Al₂O₃, NiW/Al₂O₃ catalysts as well as the effects of solvent, H₂ pressure and temperature on the hydroprocessing activity of a representative model bio-oil compound (e.g., p-cresol) is presented. With water as solvent, Pt/C catalyst shows the highest activity and selectivity towards hydrocarbons (toluene and methylcyclohexane), followed by Pt/Al₂O₃, Pd and Ru catalysts. Calculations indicate that the reactions in aqueous phase are hindered by mass-transfer limitations at the investigated conditions. In contrast, with supercritical n-heptane as solvent at identical pressure and temperature, the reactant and H₂ are completely miscible and calculations indicate that mass-transfer limitations are eliminated. All the noble metal catalysts (Pt, Pd and Ru) show nearly total conversion but low selectivity to toluene in supercritical n-heptane. Further, conventional CoMo/Al₂O₃, NiMo/Al₂O₃ and NiW/Al₂O₃ catalysts do not show any hydrodeoxygenation activity in water, but in supercritical n-heptane, CoMo/Al₂O₃ shows the highest activity among the tested conventional catalysts with 97 % selectivity to toluene. Systematic parametric investigations with Pt/C and Pt/Al₂O₃ catalysts indicate that with water as the solvent, the reaction occurs in a liquid phase with low H₂ availability (i.e., low H₂ surface coverage) and toluene formation is favored. In supercritical n-heptane with high H₂ availability (i.e., high H₂ surface coverage), the ring hydrogenation pathway is favored leading to the high selectivity to 4-methylcyclohexanol. In addition to differences in H₂ surface coverage, the starkly different selectivities between the two solvents may also be due to the influence of solvent polarity on p-cresol adsorption characteristics.

Appendix: Criteria for Evaluating Significance of Mass-Transfer Limitations

(a) Gas–liquid mass-transfer resistance is considered insignificant if

$$ \alpha_{ 1} = \frac{{R_{{{\text{H}}_{ 2} }} }}{{K_{\text{l}} a_{\text{b}} C_{\text{A}}^{ *} }} < 0.1 $$

(1)

where $K_{\text{l}} a_{\text{b}}$ is the gas–liquid mass-transfer coefficient and is calculated according to the correlation proposed by Gholap and co-workers [52].

$$ K_{\text{l}} a_{\text{b}} = (1.48 \times 10^{ - 3} )(N)^{2.18} (V_{\text{g}} /V_{\text{l}} )^{1.88} (d_{\text{i}} /d_{\text{t}} ){}^{2.16}(h_{1} /h_{2} )^{1.16} $$

(2)

(b) Liquid–solid mass-transfer limitation is considered unimportant if

$$ \alpha_{2} = \frac{{R_{{{\text{H}}_{ 2} }} }}{{K_{\text{s}} a_{\text{p}} C_{\text{A}}^{ *} }} < 0.1 $$

(3)

where $a_{\text{p}}$, the external surface area of the catalyst per unit volume for spherical particles is given by

$$ a_{\text{p}} = 6w /\rho_{\text{p}} d_{\text{p}} $$

(4)

and $K_{\text{s}}$ is the liquid–solid mass-transfer coefficient and is estimated by the correlation proposed by Sano and co-workers [53].

$$ \frac{{K_{\text{s}} d_{\text{p}} }}{{D_{\text{M}} F_{c} }} = 2 + 0.4\left[ {\frac{{e(d_{\text{p}} )^{4} \rho_{\text{l}}^{ 3} }}{{\mu_{\text{l}}^{ 3} }}} \right]^{0.25} \left[ {\frac{{\mu_{\text{l}} }}{{\rho_{\text{l}} D_{\text{M}} }}} \right]^{0.333} $$

(5)

where $F_{\text{c}}$ is the shape factor (assumed to be unity for spherical particles) and $D_{\text{M}}$ is the molecular diffusivity calculated by using the correlation proposed by Wilke and Chang [54].

$$ D_{\text{M}} = \frac{{(7.4 \times 10^{ - 8} )T(\chi M_{\text{w}} )^{{{1 \mathord{\left/ {\vphantom {1 2}} \right. \kern-\nulldelimiterspace} 2}}} }}{{\mu_{\text{l}} {{\upupsilon}}_{\text{M}}^{ 0. 6} }} $$

(6)

and e is the energy supplied calculated by using the correlation proposed by Calderbank [55].

$$ e = \frac{{N_{\text{p}} N^{3} d_{\text{i}}^{5} \psi }}{{\rho_{\text{l}} V_{\text{l}} }} $$

(7)

where $\psi$ is the correction factor for the presence of gas bubbles calculated by using the correlation proposed by Calderbank [55].

$$ \psi = 1. 0- 1. 2 6\left[ {\frac{{Q_{\text{g}} }}{{Nd_{\text{i}}^{3} }}} \right]\begin{array}{*{20}c} {} \\ \end{array} {\text{for }}\left[ {\frac{{Q_{\text{g}} }}{{Nd_{\text{i}}^{3} }}} \right] < 3.5 \times 10^{ - 2} $$

(8)

$$ \psi = 0.62 - 1. 8 5\left[ {\frac{{Q_{\text{g}} }}{{Nd_{\text{i}}^{3} }}} \right]\begin{array}{*{20}c} {} \\ \end{array} {\text{for }}\left[ {\frac{{Q_{\text{g}} }}{{Nd_{\text{i}}^{3} }}} \right] > 3.5 \times 10^{ - 2} $$

(9)

where $Q_{\text{g}}$ is the volumetric flow rate (m³/s) of gas calculated by using the formula

$$ Q_{\text{g}} = r_{ \max } V_{\text{L}} V_{\text{M}} $$

(10)

where $V_{\text{M}}$ is the molar gas volume (m³/kmol) calculated by using the formula

$$ V_{\text{M}} = V/n = RT/P_{\text{H2}} $$

(11)

(c) Pore diffusion resistance can be considered to be insignificant if

$$ \phi_{ \exp } = \frac{{d_{\text{p}} }}{6}\left[ {\frac{{(m + 1)\rho_{\text{p}} R_{\text{H2}} }}{{ 2D_{\text{e}} wC_{\text{A}}^{ *} }}} \right]^{ 1 / 2} < 0.2 $$

(12)

where $D_{\text{e}}$ is the effective diffusivity and is calculated by using the formula

$$ D_{\text{e}} = D_{\text{M}} ( \in / {{\uptau}}) $$

(13)

If gas–liquid mass-transfer limitation is significant, $C_{\text{A}}^{ *}$ is replaced by $C_{\text{AS}}$ which is calculated by using the formula

$$ C_{\text{AS}} = C_{\text{A}}^{ *} - \frac{{R_{\text{H2}} }}{{\left[ {\frac{ 1}{{K{}_{\text{l}}a_{\text{b}} }} + \frac{1}{{K_{\text{S}} a_{\text{p}} }}} \right]^{ - 1} }} $$

(14)