Photochemical reactor incidence models

Incidence Models. First attempts to describe the radiant power distribution in photochemical reactors can be summarized under the heading of the RI model (radial incident model, Figure 25a). This model is based on the hypothesis of a radial radiation field [2, 3], that is, that all the light striking the reactor wall will be directed radially inward. Corresponding profiles of radiant power or of irradiance are strongly dependent on the radius of the cylindrical reactor (Eq. 68). 

In the RI model, all incident rays intersect at the center axis of the reactor tube, and Eq. 68 produces an infinite value of irradiance as r - 0. The DI model, on the other hand, proposes parallel layers of rays which are wider than the diameter of the tubular reactor and which traverse the reactor perpendicularly to its axis from all directions with equal probability. The calculated results of both models are far from reality, as found in industrial size photochemical reactors. Matsuura and Smith [107] proposed an intermediate model (PDI model, partially diffuse model, Figure 25b) in which parallel layers of rays are assumed, and the width of each is smaller than the diameter of the tubular reactor. These two-dimensional bands form by themselves radial arrangements, the center ray of each band intersecting the... 

Stirred solutions of each polymer or copolymer at a concentration of 10 mg./ml. were Irradiated In air at 25°C In a quartz vessel with the light from a low-pressure mercury lamp at an absorbed Intensity of 2.95 x 10 quanta ml sec. . Polymer films, evaporated from DMM or methylene chloride solutions on quartz cuvettes or plates and vacutim dried, were Irradiated In air at about 35°C In a Rayonnet Model RPR-100 Photochemical Reactor containing 12 low-pressure mercury lamps the Incident radiation at the films was 1.3 x 10 quanta cm" sec"l. 

Photochemical reactor design involves simultaneous solution of the mass, energy, and momentum balance equations (as in normal reactors) along with equations for the radiation field and energy source (which are specific to photochemical reactors). Two approaches are possible (1) the intensity of the incident light, irrespective of the source, is used as the inlet boundary condition incidence models)-, (2) the emission from the source itself is part of the mathematical description emission models). The first approach has been extensively used but suffers from the weakness that the incident light is a function of scale, and hence a priori design from laboratory scale data tends to be uncertain. The second approach is formally correct, and involves no such uncertainty. 

---