Separação de ampicilina produzida enzimaticamente por reação entre éster metílico de fenilglicina e ácido 6-aminopenicilânico.

• Autor(a) principal: Vieira, Marcelo Fernandes
• Data de Publicação: 2003
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UFSCAR
• Texto Completo: https://repositorio.ufscar.br/handle/ufscar/3898
• topic: Engenharia bioquímica; Adsorção; Ampicilina; Cristalização; Síntese enzimática; Purificação de antibióticos; ENGENHARIAS::ENGENHARIA QUIMICA

The separation/concentration process of the products obtained from the enzymatic synthesis of ampicillin (AMP) catalyzed by immobilized penicillin G acylase was the focus of this Thesis. Hydrophobic resin adsorption and isoelectric precipitation were the processes herein evaluated. The antibiotic was produced in the solid phase, from 6-aminopenicillanic acid (6-APA) and phenylglycine methyl ester (PGME), and phenylglice (PG) was an undesired product. The immobilized enzyme was retained in the reactor by a sieve and the precipitated products formed (AMP and PG) were withdrawn and further dissolved at alkaline pH. After a filtration step, PG crystals were retrieved. In a second step, AMP crystals could be obtained by isoelectric precipitation. A combined process for separation/concentration of AMP, using isoelectric precipitation and hydrophobic adsorption, was put forth. The isolectric points of the molecules involved in the enzymatic synthesis were determined. In addition, the pH influence (range 5.5-7.5) on the solubility of the compounds was assessed. The results, in accordance with the literature, have shown that a pH increase leaded to an increase of the solubility of AMP and 6-APA. On the other hand, PGME solubility decreased with pH, while PG solubility was practically constant in this pH range. The adsorption efficiency and selectivity of three hydrophobic commercial resins XAD-4, XAD-7 and XAD-761 (Rohm and Haas) were evaluated at different pHs for the separation of dissolved products and unconverted reactants. Batch experiments were carried out to determine which resin had higher AMP adsorption capacity and higher selectivity (AMP/PG), at different pHs. XAD-4 was chosen: it presented higher selectivity (maximum AMP/PG = 7,0 at pH 8.5) and adsorption capacity (maximum 455 mg of AMP/g of resin, at pH 6.5). Equilibrium adsorption models were fitted: linear and Langmuir models represented PG and AMP adsorption on XAD-4 resin at pH 6.5, respectively. In a parallel work, our research group has decided to integrate synthesis and separation (precipitation of the products) in the enzymatic reactor: AMP and PG were in solid phase, and the immobilized enzyme was separated by sieving. Consequently, a combined process using isolectric precipitation and hydrophobic adsorption to separate/concentrate the AMP coming from enzymatic synthesis, with PG as impurity, was proposed. Based on the physical-chemical properties of the components, the separation process started by the elevation of pH, aiming at dissolving all AMP crystals, together with a minimum of PG crystals. Better results were obtained using pH 8.5. After filtration, un-dissolved PG was obtained. In a second step, the pH it was brought to the AMP isolectric point, causing its precipitation. Two acids and two temperatures were evaluated in this step: chloridric and sulfuric acid, 4 and 250C. Better results were obtained at 40C. No significant difference observed between the two acids. AMP crystals had purity above 97% at 40C. The following step was the separation/concentration of the mother solution, composed by a mixture of AMP and PG, using in a fixed bed of XAD-4 resin. Previously, adsorption constants, effective diffusion coefficients, and axial dispersion coefficients for AMP (at low concentrations, where a linear adsorption model fits the data) and PG were estimated by analysis of moments. The influence of temperature, flow rate, resin average particle diameter and the presence of ethanol on the performance of the system was assessed. The results obtained have shown that lower resin average particle diameters, as expected, leaded to a significant improve in the resolution of the elution curve of PG, and higher temperatures decreased retention times. The fixed bed was capable to separate, in an efficient way, a mixture of AMP and PG, reaching a separation resolution of 1.60 using a flow rate of 0.5 mL/min. Increasing ethanol concentration in the mobile phase influenced significantly the elution curves for both components. AMP de-sorption was facilitated and consequently, 5-fold and 2-fold reduction of the bed volume was achieved for AMP and PG, respectively. These results allowed the proposal of the following protocol for AMP purification, when it is reach in the solid form, containing PG as impurity: solubilization at pH 8,5 of the crystal mixture (AMP and PG), obtained from the antibiotic synthesis. Filtration of the solution, obtaining PG crystals. Reduction of the pH of the solution (containing PG, together with high concentrations of AMP), to precipitate the antibiotic at its isoelectric point. Concentration of the mother solution through adsorption on XAD-4 resin, using a water as mobile phase for adsorption and 15% (v/v) of ethanol for AMP elution. The concentrated solution should be recycled to the synthesis reactor.