Štefan Schlosser

Butyric acid production by Clostridium butyricum with integrated extraction and pertraction

Abstract The properties of an organic phase (Hostarex (20% w/w) in oleylalcohol) for extractive and pertractive fermentative production of butyric acid by Clostridium butyricum were investigated and the ability of this organic phase to be used in the fermentation process was established. The distribution coefficient of butyric acid between water and the organic phase could be increased up to 13 in diluted solutions but decreased to 3.0 with an increase of pH up to 5.2 (pH of the fermentation process). The kinetics of direct extraction of butyric acid during both extraction and pertraction were two and three times, respectively higher than the reverse processes. Reverse stripping was faster, but remained slower than direct stripping. These results demonstrated that both extraction and pertraction could be successfully joined with fermentation. An increase of total butyric acid concentration, from 7.3 g l −1 in the control to 10.0 g l −1 in extractive fermentation and to 20.0 g l −1 in pertractive fermentation and concurrent increases of yield of butyric acid from 0.24 g g −1 in the control up to 0.30 g g −1 in pertractive fermentation, showed the great possibility of successful integration of these separation processes with fermentative production of butyric acid by C . butyricum .

Phosphonium ionic liquids as new, reactive extractants of lactic acid

Solvent properties of ionic liquids with trihexyltetradecylphosphonium cation and bis(2,4,4-trimethylpentyl)phosphinate anion (Cyphos IL-104) or chloride anion (Cyphos IL-101) were studied. IL-104 effectively extracted lactic acid (LA) with distribution coefficients above 40 at low acid concentrations. IL-104 extracted only undissociated acid (LAH) what supported the coordination mechanism of lactic acid extraction via H-bonding. In the extraction of lactic acid by phosphonium chloride (IL-101) an ion-exchange mechanism contributed remarkably to the extraction especially at basic pH where anionic form of this acid predominated. A high solubility of water in hydrophobic IL-104 up to 14.4 mass % was connected with the formation of reverse micelles. A dual mechanism of water extraction to phosphonium ionic liquids was identified, which consisted of water incorporation into reverse micelles and the inclusion of water into the hydrated complex of lactic acid with ionic liquid (IL). The extraction of lactic acid caused splitting of reverse micelles with liberation of water from the solvent. In the saturated solvent only hydration water remained in the complex of lactic acid with phosphonium ionic liquid, with the suggested structure (LAH)p(IL)(H2O)2, where the value of p ranged from 1 to 3.

Evaluation of solvent and pH for extractive fermentation of butyric acid

The effects of nine organic extractive reagents on growth and acid production by Clostridium butyricum S21 were examined to evaluate their possible use in extractive fermentation of butyric acid. Oleylalcohol alone and in a mixture with Hostarex A327 (n-octyl(n-decyl)amine) and pure tri-hexylphosphate (THP) did not appear to be toxic. The presence of THP decreased the selectivity of butyrate production. High toxicity was shown by n-octanol and isodecanol alone and in mixtures with the extractant Hostarex A327 and diluent n-aleanes. An acceptable decrease of butyric acid concentration, productivity and production selectivity (7·26, 0·33 and 0·54 respectively) compared with optimal pH of 6·5 (11·75, 0·60 and 0·73, respectively) was obtained at pH 5·2. At this pH the distribution coefficient of butyric acid between water and Hostarex A327 (20% w/w) in oleylalcohol was about 3·0. It was four times lower than for the diluted solution (13·2) and two times lower than for the concentrated (6·5) solution of butyric acid at pH 4·0, but about three times higher than in the case of pure oleylalcohol (1·01) at pH 5·2. Hostarex A327 (20% w/w) in oleylalcohol as an extractive reagent at a pH of 5·2, can be used for the production and in situ extraction of butyric acid.

Liquid-liquid equilibria of butyric acid for solvents containing a phosphonium ionic liquid

L/L equilibrium data of butyric acid (BA) in aqueous solutions contacted with the solvents containing ionic liquid (IL), trihexyl-(tetradecyl)phosphonium bis 2,4,4-trimethylpentylphosphinate (Cyphos IL-104), and a related model are presented. IL-104 and its solutions in dodecane were found to be effective solvents of BA. The values of the distribution coefficients of BA were higher than those for solvents with the widely used extractant trioctylamine, especially at low acid concentrations and were also several-fold higher than those of lactic acid (LA). IL extracted BA only in its undissociated form (BAH) at pH well below pKa of the acid. The loading of IL was independent of IL concentration and it achieved a value higher than four at saturation. Complexes with 1–5 molecules of BA per one IL molecule were supposed in the mass action model in which the reactive formation of complexes (BAH)p(IL)(H2O)2 was supposed. Up to 10 % of the total extracted BA was extracted physically by dodecane as a monomer and dimer, in the solvent. The water content in the organic phase steeply decreased with the BA concentration, which was caused by splitting water-IL reverse micelles due to the formation of the BAH/IL complexes.

Factors influencing transport through liquid membranes and membrane based solvent extraction

Based on experimental data from pertraction through liquid membranes and membrane based solvent extraction (MBSE) or solvent stripping (MBSS) of butyric acid (BA), phenylalanine (Phe), and heterocyclic carboxylic acid (HCA), the influence of the formulation of the membrane phase, kinetics of the acid/carrier complex decomposition, composition of the donor phase, and hydrodynamic conditions on pertraction and MBSE is discussed. The analysis of mass-transfer resistance including the resistance based on reaction kinetics in pertraction through bulk and supported liquid membranes and in MBSE and MBSS in hollow fibre contactors is presented. It has been found that mass-transfer resistance based on the reaction kinetics on the stripping interface represents about 38% of the overall resistance in pertraction of BA and about 60% in MBSS of BA from the complex with trioctylamine extractant. Aggregation and/or formation of a microemulsion at the concentration of DEHPA in the membrane phase above 1 kmol m -3 substantially decrease the transport rate of Phe through the layered liquid membrane. Competitive transport of HCI decreases greatly the transport of HCA through liquid membrane with TOA as a carrier while the influence of the co-transport of H 2 SO 4 is very small. Transport of the mineral acid from chloride media is much higher than from sulphate media.

Hollow-fibre pertractor with bulk liquid membrane☆

Pertraction of phenol in a bulk liquid membrane pertractor of the “hollow-fibres in tube” type was studied. n-Alkanes and solutions of triisobutylphosphine sulfide (TIPS) in n-alkanes were used as membrane phase. Pulsation of the membrane phase increases the mass flux by more than 40% at relatively low pulse velocity. Wettability of the microporous walls plays an important role. It is preferable when the feed flows in hydrophobic fibres and the stripping solution is in contact with the hydrophilic wall. Pertraction through a membrane with TIPS was three times faster due to a higher distribution coefficient for phenol. Analysis of mass-transfer resistances of the different liquid layers in the pertractor is presented. The mass flux per unit volume of pertraction varied from 20 to 70 mol-m−3-hr−1, which is several times larger than the value for other types of bulk liquid membrane equipments.

Fermentation of lactic acid with Rhizopus arrhizus in a stirred tank reactor with a periodical bleed and feed operation

Semicontinuous fungal fermentation of lactic acid (LA) by Rhizopus arrhizus was performed using a periodical bleed and feed (PBF) operation in a stirred tank bioreactor with a volume of 4.5 dm3. LA productivity up to 2.91 kg m−3 h−1 and yield up to 75.3% were achieved. Using PBF cycles, fermentation was prolonged to 240 h with an average LA productivity of 2.31 kg m−3 h−1 and a total yield of 67.3% at a mean pH of 5.30. The most effective production of LA required filamentous morphology of the biomass. When such morphology changed to pellets biomass growth stopped, the LA productivity decreased to 1.63 kg m−3 h−1 and the yield decreased to 62.6%. The filamentous morphology was successfully maintained for 152 h when the inoculation was with spores. CaCO3 initially used for pH neutralisation did not dissolve well since it was entrapped by biomass. It was successfully replaced by 25% ammonia. Long-time operation below pH 5 decreased LA productivity. Accumulating biomass in the fermenter did not increase the LA productivity. With an increase in the biomass growth rate, the LA productivity increased. The on-line extraction or pertraction of LA by trialkylamines demands a pH value of feed below 4. This can be achieved for a short time by interruption neutralisation at the end of each PBF cycle without a decrease in LA productivity after the restoration of the optimal pH value.

Microfiltration of beer yeast suspensions through stamped ceramic membranes

Abstract The possibility of permeate flux enhancement based on the surface modification of a membrane was investigated. Crossflow microfiltration of a pure beer yeast suspension through smooth and stamped ceramic membranes was performed. The stamped membrane had a helical reversed thread to increase turbulence in the feed flow. Both membranes had an outer diameter of about 9 mm, the same zirkonia oxide active layer with a mean pore size of 0.23 μm. Membranes were produced by LMPM ENSCM. Fluxes of permeate were measured for different crossflow velocities and concentrations. A ceramic membrane with a helically stamped surface has several advantages compared with a smooth membrane: flux, as well as limiting flux, are higher at the same velocity of the feed, power consumption per unit volume of permeate is lower for the stamped membrane and increases with increasing crossflow velocity of the feed. Modelling of the flux decline indicates different mechanisms of fouling for smooth and stamped membranes.

RECOVERY OF PHENOL WITH CYANEX® 923 IN MEMBRANE EXTRACTION-STRIPPING SYSTEMS

The recovery of phenol from aqueous solutions with CYANEX® 923 was studied. Classical dispersive extraction and three membrane extraction-stripping systems (bulk liquid membranes, three-phase hollow fiber contactor and two hollow fiber modules set-up) were used. It was found that CYANEX® 923 was a convenient carrier for recovery of phenol from aqueous streams in extraction-stripping membrane processes. The problem of emulsion formation, so important in dispersive extraction, was avoided. Both mass transfer experiments in different membrane systems and measurement of the dynamic interfacial tension demonstrated importance of the interfacial phenomena occurring in the stripping stage. A blocking of this interface was observed that resulted in a decrease of phenol mass transfer.

Extraction Equilibria of Butyric Acid with Organic Solvents

Abstract Fourteen solvents (five with a tertiary amine and different diluents, four C8-C18 alcohols, dibutylether, two hydrocarbons, and two vegetable oils) have been tested for the extraction of butyric acid. The highest distribution coefficient for butyric acid is shown by solvents with tertiary amines. A ternary solvent with amine extractant, n-alkanes as diluent, and higher alcohol as modifier can be advantageous in this procedure. Amines enable the extraction of acid at a pH above the pK a value up to about pH 5.6. With an increase of the molecular weight of alcohol, the value of the distribution coefficient decreases. Its value for pure alcohols is independent of the concentration of acid in the aqueous phase. Equilibrium data suggest that the stoichiometry of the acid-alcohol complex is 2:1, and only undissociated acid is extracted.