Elsevier

Catalysis Today

Volume 54, Issues 2–3, 3 December 1999, Pages 205-216
Catalysis Today

Kinetic analysis of photoinduced reactions at the water semiconductor interface

https://doi.org/10.1016/S0920-5861(99)00183-2Get rights and content

Abstract

A reduced kinetic model for the initial steps of the photocatalytic process is presented with the aim to analytically solve the resulting kinetic system. Several possible kinetic models have been explored. Attention was paid to obtain equations with physical meaning and reduced complexity. An analytical equation is obtained for the rate and the quantum yield, which retains the principal features of the photocatalytic process, namely the light induced charge separation and recombination, the oxidative and reductive electron transfers, the formation of a stable oxidized intermediate, and, unlike the previous kinetic models, also the back reaction of the oxidized substrate. Compared to the two-parameter Langmuir–Hinshelwood equation, all the previous features and the dependence on the light intensity are described with only three parameters, which collect all the kinetic constants, and account for experimental concentrations of substrate and electron scavenger, light intensity and catalytic system characteristics. The kinetic behavior of photocatalytic systems under all the possible values of experimental parameters can be graphically presented. The analysis of the obtained rate equation shows that the best utilization of photons is attainable at low light intensities, suggesting that preconcentration of solar light is unnecessary.

Introduction

The photocatalysed conversion of organic species under aerobic conditions has been deeply investigated from basic, mechanistic and practical points of view due to the great potential of the process for pollutant abatement and waste treatment [1]. The basic mechanism was treated in a number of papers [2]. Following the light absorption primary excited species are formed which can either recombine or migrate to the surface of the semiconductor, where several redox reactions may take place. The organic substrate reacts with formed active species (oxidant or reducing) depending on its initial oxidation state and the nature of substituents [3], forming radicals and other species that are further oxidized or reduced. Several complex networks of reaction have been reported on the basis of detailed chemical analyses of the time evolution of substrate and formed intermediates or by-products [4], [5], [6], [7].

In photocatalytic processes the degradation rate depends, no matter what the mechanism of degradation is, upon the photon absorption rate. In practical systems the distribution of radiation intensities inside the reactor space is highly non-uniform. Its dependence on reactor geometry, and, for slurries, on the relative contributions of absorption and scattering was deeply investigated [8]. The solution of the integro-differential equation that results is not a simple task even for a simple geometry. Equations more manageable may be forecast for thin films. Given the spectral volumetric absorption, the absorption and scattering coefficients, and the distribution function for photon scattering, the incident radiation at any point inside the reactor space can be obtained [8]. The last is directly related to the local photon absorption rate. The photocatalytic degradation rate could be calculated, provided that a chemical kinetic model relating primary chemical events with absorbed photon rate is available.

Owing to the complex network of reactions, even for a chosen substrate it is difficult to develop a model for the dependence of the degradation rate on the experimental parameters for the whole treatment time. All the possible formed species must be identified, their kinetic constants for reaction either with photogenerated species or with other active species (radicals, other intermediates) have to be measured (by independent degradation experiments) or estimated. Since this is a heavy (and perhaps useless) task, kinetic modeling of the photocatalytic process is usually restricted to the analysis of the initial rate of degradation. This is the product of the initial slope and the initial substrate concentration in an experiment in which the variation of the substrate concentration is measured as a function of time. The extrapolation of the rate to time = 0 avoids the possible interference from by-products, both for the kinetic aspect (avoiding the growing complexity of the system as the degradation time increases), and for the possibility that formed species are physisorbed or chemisorbed [9], eventually reducing the possibility of formation of active species, or, in extreme cases, blocking or poisoning surface sites.

However, even under the reduced complexity deriving from the initial rate simplification, the interpretation of the rates obtained in photocatalytic experiments is still under debate. An attempt to model the initial steps of the photocatalytic process and to analytically solve the resulting kinetic system is presented. Attention was paid to obtain equations with physical meaning and reduced complexity. Several possible kinetic models have been explored.

Section snippets

The kinetic of photocatalyzed transformation

The common shape observed for the initial rate of degradation ro as a function of the substrate or catalyst concentration is an increasing function tending to a limiting value (zero order rate), as if the rate would be determined by the adsorption properties of the substrate on the catalyst surface. This was historically, but even now interpreted in terms of the Langmuir–Hinshelwood two-parameter equation either invoking physical reasons or simplicity [10], [11].ro=kKadsCo1+KadsCo

LH equation (

Results

The differential kinetic system that can be derived from reactions 2 is not solvable, even under steady illumination, unless numerically with a huge number of parameters. The hypothesis of the steady state can be applied to transient species (e.g. differential equations = 0). Nevertheless, the resulting system has no analytical solution. Its numerical solutions give the dependence of the rate as a function of {Red1} and {Ox2} similar to that reported below (see Fig. 1). Because the number of

Conclusions

The present kinetic analysis lead to manageable analytical solutions for the dependence of the rate on operational parameters. From a mechanistic point of view, it has been concluded that, when a maximum is observed for the rate as a function of substrate concentration, the reaction of Ox2 with Ox1radical dot is not predominant. On the contrary, when a maximum is not observed, the reaction (3i) may be important, or the conditions of the photocatalytic systems are close to that reported in Fig. 1 top (i.e.

Acknowledgements

This work was supported by EEC program SOLARDETOX under contract BRPR-CT97-0424 (DG12-GZMM) and by MURST research program under contract with Tecnoparco ValBasento.

References (49)

  • D.F. Ollis et al.

    J. Catal.

    (1984)
  • C. Minero et al.

    Solar Energy

    (1996)
  • J. Cunningham et al.

    J. Photochem. Photobiol. A: Chem.

    (1991)
  • A.P. Davis et al.

    Chemosphere

    (1993)
  • C. Minero

    Solar En. Mat. Sol. Cells

    (1995)
  • C. Minero et al.

    Marine Chem.

    (1997)
  • E. Pelizzetti, C. Minero, E. Pramauro, Photocatalytic Processes for Destruction of Organic Water Contaminants, in: H.I....
  • E. Pelizzetti et al.

    Comments Inorg.Chem.

    (1994)
  • E. Pelizzetti et al.

    Colloid Surfaces A:

    (1999)
  • P. Piccinini et al.

    J. Chem. Soc. Faraday Trans.

    (1997)
  • C. Minero et al.

    Res. Chem. Interm.

    (1997)
  • C. Minero et al.

    Environ. Sci. Technol.

    (1995)
  • P. Calza et al.

    Environ. Sci. Technol.

    (1997)
  • M.I. Cabrera, O. M. Alfano, A.E. Cassano, J. Phys. Chem. 100 (1997) 20043, and references cited...
  • S. Tunesi et al.

    J. Phys. Chem.

    (1991)
  • T.N. Obee et al.

    Environ. Sci. Technol.

    (1997)
  • N.J. Peill et al.

    Environ. Sci. Technol.

    (1998)
  • D. Bahnemann, J. Cunningham, M.A. Fox, E. Pelizzetti, P. Pichat, N. Serpone, in: G.R. Helz, R.G. Zepp, D.G. Crosby...
  • C. Kormann et al.

    Environ. Sci. Technol.

    (1991)
  • V. Augugliaro et al.

    Toxicol. Environ. Chem.

    (1988)
  • C. Minero et al.

    Langmuir

    (1994)
  • R.F. Howe et al.

    J. Phys. Chem.

    (1987)
  • M.A. Fox, R.B. Draper, M. Dulay, K- O’Shea, in: E. Pelizzetti, M. Schiavello (Eds.), Photochemical Conversion and...
  • R.B. Draper et al.

    Langmuir

    (1990)
  • Cited by (155)

    • Novel ZnO photocatalysts for pollutants’ abatement under solar radiation at pilot plant scale

      2023, Catalysis Today
      Citation Excerpt :

      The values of the kr parameter (of the same order of magnitude) indicate a similar reactivity of the photoactivated catalyst surface with Ph in both testing conditions, as kr is proportional with the rate of light absorption able to generate photoreactive species [19] and not with the operational conditions in the photoreactor. The L-H model is widely used in the kinetic modelling of the photocatalytic processes, but it is generally agreed that the two model constants are only apparent [19,30]; they have no physical meaning and cannot be used to identify the surface process as they only allow the calculation/estimation and comparison of the reaction rates in different experimental conditions, being used for the reactor optimization and design. The Ph degradation efficiency increased while decreasing its initial concentration, as the amount of pollutant to be degraded close to the photocatalyst surface decreased (Fig. 5).

    • Kinetic models in photoelectrocatalysis

      2023, Photoelectrocatalysis: Fundamentals and Applications
    • Continuous flow photoreactor undergoing variable simulated irradiation conditions: Experimentations and modeling

      2022, Chemical Engineering Journal Advances
      Citation Excerpt :

      However, the literature on photo-oxidation includes very few studies on the influence and consideration of energy availability variables and in particular on the intermittency of sunlight on a daily scale [19,25,26]. Moreover, photodegradation capacities are defined from an average amount of energy received (cumulative UV per day) [15,22,24,25], ignoring the fact that photo-excitation rates are not linearly dependent on flux density [27–29]. Consequently, the influence of the nature of the irradiation (spectral range and intensity level) on the capacity and efficiency of the photoprocess has not been considered.

    • Continuous degradation of micropollutants in real world treated wastewaters by photooxidation in dynamic conditions

      2022, Water Research
      Citation Excerpt :

      To analyze the results in a more global way, Fig. 12 plots the daily photo-oxidation capacities as a function of the feed concentration, independently of the nature of the substance concerned. It is established in the literature that photo-oxidation rate depends mainly on two quantities, namely the applied flux density and the target concentration (Emeline et al., 2000; Minero, 1999; Plantard and Goetz, 2012). Although there is no universal law representing the photodegradation process, several formalisms expressing the dependency of the rate on these two quantities have been established (Emeline et al., 2000; Plantard and Goetz, 2012).

    View all citing articles on Scopus
    View full text