Films were prepared by reaction of HEMA monomer with the crosslinking agent ethyleneglycol dimethacrylate (EGDMA) at crosslinking ratios, X, of 0.005, 0.01, 0.0128, 0.025, and 0.050 mol EGDMA/mol HEMA in the presence of 40 wt.% water at 60 degrees C for 12 h. These membranes were subsequently swollen in water at 37 degrees C and their structure analyzed using a modified Gaussian distribution equation of equilibrium swelling. The calculated values of Mc varied between 800 and 3700 daltons, which corresponded to a correlation length of the mesh size xi of 16.2 to 35.6 A. The structural analysis and diffusive studies of PHEMA membranes indicate that the recent determinations of Mc for PHEMA by Migliaresi et al. (C. Migliaresi, L. Nicodemo, L. Nicolais, and P. Passerini, "Physical characterization of PHEMA gels," J. Biomed. Mater. Res., 15, 307 (1981). and others are not accurate. The methods presented can also be used for analysis of any highly crosslinked polymer network produced by simultaneous polymerization and crosslinking reactions.

Urea hydrolysis by urease immobilized onto ion exchange resins in a fixed-bed reactor has been studied. A modified Michaelis-Menten rate expression is used to describe the pH-dependent, substrate- and product-inhibited kinetics. Ionic equilibria of product and buffer species are included to account for pH changes generated by reaction. An isothermal, heterogeneous plug-flow reactor model has been developed. An effectiveness factor is used to describe the reaction-diffusion process within the particle phase. The procedure for covalent immobilization of urease onto macroporous cation exchangers is described. Urea conversion data are used to estimate kinetic parameters by a simplex optimization method. The best-fitted parameters are then used to predict the outlet conversions and pH values for systems with various inlet pH values, inlet urea and ammonia concentrations, buffers, particle sizes, and spacetimes. Very good agreement is obtained between experimental data and model predictions. This immobilized urease system exhibits quite different kinetic behavior from soluble urease because the pH near the enzyme active sites is different from that of the pore fluid. This effect results in a shift of the optimal pH value of the V(max) (pH) curve from 6.6 (soluble urease) to ca. 7.6 in dialysate solution, and ca. pH 8.0 in 20mM phosphate buffer. The reactor model is especially useful for estimating intrinsic kinetic parameters of immobilized enzymes and for designing urea removal columns.

SynopsisA theoretical model was developed to describe solute transport through moderately swollen networks. The model is an extension of the previous analysis of Peppas and Reinhart (1983), and it describes the normalized effective diffusion coefficient of the solute through the network as a function of the equilibrium degree of swelling, Q, the hydrodynamic radius of the solute, r,, the number average molecular weight between crosslinks, M, and a function f ( [ ) of the mesh size [, which takes into consideration barriers due to cro&links, entanglements, etc. For the development of this model, the Cohen-Turnbull (1958) free volume theory was modified to incorporate topological and mobility characteristics from de Gennes' analysis (1979).

Studies with phenylalanine diffusing in well-characterized poly(2-hydroxyethyl methacrylate) (PHEMA) membranes swollen in water at 37°C were conducted. Films were prepared by reaction of HEMA monomer with the cross-linking agent ethylene glycol dimethacrylate (EGDMA) at cross-linkingratios, X, of 0.005,0.01,0.0128, 0.025, and 0.050 mole EGDMA/mole HEMA in the presence of 40 weight percent water at 60°C for 12 hours. These membranes were subsequently swollen in water at 37°C and their structure analyzed using a modified Gaussian distribution equation of swelling. The calculated values of Mc varied between 1,700 and 3,425 daltons, which corresponded to a correlation length of the mesh size, t , of 24 to 35 A. The phenylalanine solute diffusion coefficient varied from 0.17 X 1 0-6 to 0.97 X 1 0-6 cm2/s, and depended on the aforementioned structural parameters of the membranes.

Urea hydrolysis in an enzyme electrode that is sensitive to urea concentration has been analyzed with a theoretical model. The model incorporates nonlinear product-inhibited, pH-dependent kinetics, Nernst-Planck diffusion, and ionic equilibria of product and buffer species. Simulation results indicate that film resistance can significantly reduce the detection range of the electrode and a Biot number (the ratio of characteristic internal to external mass transfer resistance) greater than 75 is necessary to minimize the film resistance. The electrode response depends on a balance between the reaction-generated pH changes and the membrane diffusion resistances that sustain such pH shifts. The Damkohler number (the ratio of characteristic rate of reaction to that of diffusion) should be between one and five to maximize electrode sensitivity.The model has been used to demonstrate the effects of acetate, phosphate, and Tris buffers on electrode response. The carbonate concentration of dialysate solution is sufficient to shift the detection range of the electrode and reduce the sensitivity of the electrode by an order of magnitude. Tris buffer reduces the pH response of the electrode for buffer concentrations greater than 2 mM, because the pK of Tris falls within the range of the reaction-generated pH shijt.