Documents

Kinetic Study of Biological Ferrous Sulphate Oxidation By

Description
Journal Entry
Categories
Published
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Kinetic study of biological ferrous sulphate oxidation byiron-oxidising bacteria in continuous stirred tank and packedbed bioreactors Jose´ Manuel Go´mez*, Domingo Cantero Department of Chemical Engineering, Biological and Enzymatic Reactors Research Group, Food Technology and En v ironmental Technologies,Faculty of Sciences, Uni  v ersity of Cadiz (UCA), 11510 Puerto Real (Cadiz), Spain Recei v ed 26 November 2001; recei v ed in re v ised form 28 January 2002; accepted 19 February 2002 Abstract This paper describes kinetic study of biological ferrous sulphate oxidation by  Thiobacillus ferrooxidans  iron-oxidising bacteria insubmerged culture and immobilised in nickel alloy fibre as matrix support. In this way, two types of bioreactors has been used: acontinuous stirred tank reactor (CSTR) for free cells and a packed bed bioreactor for immobilised biomass. A mathematicalexpression has been de v eloped to explain kinetic beha v iour of micro-organism in bioreactors as function of the main processparameters. Model predictions of ferrous iron oxidation rate were found closely to experimental data and pro v ide high coefficient.Comparison between oxidation rates for two bioreactors showed that process with a biofilm reactor is more stable than thebioreactor with free immobilised biomass. #  2002 Elsevier Science Ltd. All rights reserved. Keywords:  Kinetic study; Ferrous oxidation;  Thiobacillus ferrooxidans ; Stirred tank; Packed bed 1. Introduction Thiobacillus ferrooxidans  is an acidophilic bacteriumthat has the ability to oxidise ferrous to ferric iron in thepresence of atmospheric oxygen and carbon dioxide.This bacterium is a dominant organism in the process of  v alue metal extraction by microbial leaching of pyriticores. The product of the oxidation is an acidic solutionof Fe(III), a potent chemical oxidant, that has beenexploited in the treatment of acidic mine effluentsprocesses for the remo v al of sulphide hydrogen fromsour gases and bioremediation of contaminated soils byhea v y metals.Most of the pre v ious research on this biologicaloxidation has been concerned with practical aspects of impro v ing the o v erall leaching rates of metals fromsulphide ores by studying the effect of such  v ariables astemperature, pH, nutrient concentration, particle sizeand mineral type. From an engineering point of   v iew,the principal factor affecting the cost effecti v eness of industrial processes is the rate of reaction; so, it isnecessary to impro v e it.Biological iron oxidation has been studied in se v eralexperimental systems with batch and continuous-flowmodes of operation [1]. Because of the interest in thekinetic aspects of the oxidation, attempts ha v e beenmade to impro v e the ferrous iron oxidation rate by theuse of   v arious reactor designs employing biologicalcontacting de v ices. These ha v e included bacteria insuspended and in fixed-film applications to pro v ide alarge surface area for their attachment [2]. Initial workwith  T. ferrooxidans  was primarily concerned with thede v elopment of rotating biological contactors [3,4].More recent efforts ha v e addressed other fixed-filmapproaches, which essentially in v ol v e  v arious config-urations of packed-bed [5,6] and fluidised-bed reactorswith inert carrier matrix materials [7].In the present work, we ha v e studied the beha v iour of biological oxidation of ferrous sulphate in a continuousstirred tank reactor (CSTR) with free suspended cells of  * Corresponding author. Tel.:   34-9560-16382; fax:   34-9560-16411. E-mail address:  josemanuel.montesdeoca@uca.es (J.M. Go´mez).Process Biochemistry 38 (2003) 867    / 875www.else v ier.com/locate/procbio0032-9592/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.PII: S0032-9592(02)00048-1  T. ferrooxidans  and in a packed-bed reactor withimmobilised cells in nickel alloy fibre as matrix support.With experimental data obtained in these studies, weha v e proposed a kinetic equation in order to calculatetheoretical ferrous oxidation rates as function of theinput substrate concentration and the substrate con v er-sion. Finally, it is possible to establish a comparisonbetween the reaction rates in packed-bed bioreactor withimmobilised cells and that in a bioreactor with freesuspended cells. 2. Materials and methods  2.1. Microorganism and growth conditions The strain of   T. ferrooxidans  used in this study wasisolated from the Rio Tinto mines of Huel v a (Spain) andkindly made a v ailable by the BiohydrometallurgyGroup of the Uni v ersity of Se v ille (Spain). This strainhas the same properties and characteristics of strainused by Nemati and Webb [8] obtained from NationalCollection of Industrial and Marine Bacteria (NCIMB9490).The bacteria were grown in a medium proposed bySil v erman and Lundgren [9]: (NH 4 ) 2 SO 4  3.0 g l  1 ;MgSO 4  0.5 g l  1 ; K 2 HPO 4  0.5 g l  1 ; KCl 0.1 g l  1 ;Ca(NO 3 ) 2  0.01 g l  1 and a  v ariable concentration of FeSO 4 , depending on the experiment to be performed. Thiobacillus ferrooxidans  was immobilised on nickelalloy fibre according to the procedure described inGo´mez et al. [10].  2.2. Analytical methods Ferrous sulphate oxidation was monitored by deter-mining the residual ferrous iron concentration at  v ariousinter v als. The 1,10 phenanthroline method of Vogel [11]was used. In order to determine the ferrous ironconcentration, a 10- m l sample was placed in a tube anddiluted with 1.0 ml of distilled water. The pH wasadjusted to between 3.0 and 6.0 with 2 mol l  1 sodiumacetate, 0.8 ml of 1,10-phenanthroline solution wasadded and, finally, an additional 10 ml distilled water.The absorbance at 515 nm was measured after 5    / 10min. In order to determine total iron concentration, 1.0ml hydroxylamine hydrochloride, as a reducing agent,was added to the sample instead of 1.0 ml distilled waterand the same procedure followed. A calibration cur v e of known FeSO 4  concentrations was used to calculate theiron concentrations. The concentration of iron (III) insolution was calculated by subtracting the a v erage iron(II) concentration from the total iron concentrationmeasured at each point in time.  2.3. Total cell number In order to measure the total biomass adhered to thematrix support, a known amount of nickel alloy fibrewas placed in a flask with 5 ml of oxalic acid 10% (w/ v ),at each ‘draw and fill’ cycle. After 10 min, the supportwas rinsed with 5 ml of distilled water for 10 min. Then,the rinsings were added to the pre v ious cell suspensionobtained.The biomass concentration was determined by directcounting using a Neubauer chamber counter of 0.02 mmdepth and 1/400 mm 2 area under an optical microscope.In some cases, it was necessary to dilute the sampleswith basal salt solutions because of the high biomassconcentrations.Each measurement was made in duplicate to minimisethe experimental errors inherent in working with micro-bial populations.  2.4. Bioreactors Ferrous sulphate oxidation by free  T. ferrooxidans cells was studied using an automatic continuous stirredtank reactor with an inlet for medium and air, and outletfor effluent at the top. A working  v olume of 5 l, aerationrate of 0.5  vv m and agitation rate of 200 rpm was used.The temperature was maintained at 30  8 C with anexternal heat exchanger and flow rates of both inlet andoutlet were regulated with peristaltic pumps controlledby automatic control equipment. Sampling of theeffluent was performed from the end of the effluenttubing at the reser v oir and analysed of ferrous, totaliron and biomass.In the case of immobilised  T. ferrooxidans,  ferroussulphate oxidation was studied using a glass column(1  / 0.05 m) with an inlet for medium and air at thebottom and outlet for effluent at the top. Nickel alloyfibre with immobilised  T. ferrooxidans  was placed in thecolumn and a layer of sintered glass was placed at thebottom of the matrix layer to keep this inside thecolumn. A working  v olume of 1350 ml and aeration rateof 0.675 ml min  1 were used. The temperature wasmaintained at 30  8 C and flow rates of both inlet andoutlet were regulated with automatically controlledperistaltic pumps. Sampling of the effluent was per-formed from the end of the effluent tubing at thereser v oir and analysed for ferrous iron and total iron.Both bioreactors were operated in batch mode untilcomplete oxidation of iron were achie v ed, then thereactor was switched to continuous operation. Steady-state operation was considered to be established whenthe ferrous iron concentration  v aried by less than 5%during a period of time equal to the theoretical retentiontime.Se v eral media with different initial concentration of ferrous iron (from 1000 to 8500 mg l  1 ) were tested. The J.M. Go´ mez, D. Cantero / Process Biochemistry 38 (2003) 867     /  875 868  initial pH of the media was adjusted to 1.8. For eachmedium,  v arious dilution rates were applied: 0.01    / 0.06h  1 for CSTR, and 0.08    / 0.25 h  1 for packed-bedreactor. 3. Results and discussion 3.1. Operation in continuous stirred tank reactor Se v eral experimental runs were carried out to deter-mine the influence of dilution rate and ferrous ironconcentration in the influent on rate of ferrous ironoxidation. As we mentioned abo v e, bioreactor wasoperated as an independent unit and was started witha 10% ( v ol/ v ol) inoculum of a spent, iron-grown cultureof   T. ferrooxidans . The e v olution of ferrous, ferric ironbiomass is shown in Fig. 1. Initially, ferrous irondecreased and ferric iron and total biomass increased;until minimum and maximum  v alues, respecti v ely, wereachie v ed. These  v alues are in dependence on theoperational conditions in the run. In this moment,steady-state conditions ha v e been achie v ed, a continu-ous-flow mode of operation was initiated. The time toachie v e steady-state conditions  v aried depending on theflow rate, but normally this state was reached within 150h before experimental run was established.After 300 h, the amount of precipitate deposited in thewalls of the bioreactor gradually increased. The pre-cipitate also caused problems in the outlet tubing and inthe distribution system of feed air, where it had to becleaned to pre v ent blocking. The extensi v e precipitationin the bioreactor had a detrimental effect on iron (II)oxidation rate and oxygen mass transfer rate to themedium.The e v olution of ferrous and ferric iron concentrationin steady-state conditions  v ersus dilution rate andferrous iron concentration in the influent are shown inFigs. 2    / 4, respecti v ely. In general, it can be obser v edthat Fe(II), Fe(III) and total biomass cur v es ha v esimilar progress.When an influent Fe(II) concentration is below 2.70 gl  1 , Fe(III) and total biomass were not affected becauseferrous iron acts as a limiting substrate for growth. Inthe same way, when the dilution rate is abo v e 0.03 h  1 ,there is no dependence between these  v ariables and feediron concentration.In Fig. 5 it is possible to see that ferrous ironoxidation rate was accelerated by increasing dilutionrates from 0.01 to 0.06 h  1 . These data confirm thatoperating conditions were below critical with respect tohydraulic loading. Ferrous sulphate oxidation rates arecomparable with those published by Karamane v  et al.[2], Torma [12] and Kelly et al. [13] for this process whenfeed ferrous iron is below 6 g l  1 .Results obtained can be expressed by a kineticequation that allows calculating theoretical ferrousoxidation rate as function of influent Fe(II) and Fe(II) Fig. 1. E v olution of total biomass, ferrous and ferric iron in acontinuous stirred tank bioreactor.Fig. 2. Steady-state ferrous iron concentration in a continuous stirredtank bioreactor as function of dilution rate.Fig. 3. Steady-state ferric iron concentration in a continuous stirredtank bioreactor as function of dilution rate. J.M. Go´ mez, D. Cantero / Process Biochemistry 38 (2003) 867     /  875  869  concentration in the reactor under steady-state condi-tions. As a pre v ious step, it is necessary to de v elop massbalance for this reactor. The model was de v elop, usingthe following main assumptions:1) The liquid medium in the bioreactor is perfectlymixed.2) Oxygen transfer rate in the bioreactor is high.3) Total  v olume is constant and equal to  V  R .4) Ferrous iron acts a limiting substrate.5) The process is in a continuous regime.6) Biomass growth rate is the same for free suspendedcells obtained pre v iously [14] in batch culture. m  m max    S K  S  S   K  I    P  where  m max ,  K  S  and  K  I  are parameters characteristics of the microorganism;  S   is the substrate concentration and P   is the product concentration.So, mass balance can be expressed in this mathema-tical form: V  R d S  d t  Q    S  0  Q    S   V  R    r S where  V  R  is the reactor  v olume;  Q  is the input andoutput flow rate;  S  0  is the input limiting substrateconcentration;  S   is the limiting substrate concentrationand  r S  is the substrate utilisation rate. In the case of perfectly mixed liquid: r S  D ( S  0  S  ) where  D  is dilution rate. If we consider there is nocellular death, it is possible to assume that dilution rate( D ) is equal to specific growth rate ( m ). So, the modelcan be rewritten as a function of   r S  /  f  ( S  ,  S  0 ) in thisform: r S  m max    S     ( S  0  S  ) K  S  S   K  I ( S  0  S  ) A non-linear regression procedure [15] was used toperform the mathematical fitting of the coefficients. Theapplication of this algorithm to set of experimental datapre v iously referred gi v es the following  v alues of theparameters:  m max  / 0.22 h  1 ;  K  S  / 0.92 g l  1 and  K  I  / 0.21. These data are in agreement with literature andhigh theoretical-experimental determination coefficient( r 2  / 0.92) pro v ides good application of this model. 3.2. Operation in packed-bed bioreactor with immobilised biomass Kinetics for continuous oxidation of ferrous iron byimmobilised cells of   T. ferrooxidans  were studied in apacked-bed bioreactor. Once inoculated with nickelalloy fibre particles, the bioreactor was started in batchculture until 95% of ferrous iron was oxidised. In thismoment, a continuous flow mode of operation wasinitiated. During the operation of the bioreactor,sampling from different parts of the bed was notpossible. Hence, the assessment of biomass hold-upwas performed at the end of each experimental run. Thiswas done by taking a know amount of nickel alloy fibrefrom  v arious parts of the bed. The particles were soakedin a flask with 5 ml of oxalic acid 10% (w/ v ) for 10 min.After this time, the support was rinsed with 5 ml of distilled water for 10 min. After that, the rinsing wasadded to pre v ious cell suspension obtained. The biomassconcentration was determined by direct counting using aNeubauer chamber counter with optical microscope.Occasionally, biomass concentration in the effluentwas measured and it showed that this concentration isdespicable with immobilised biomass. Fig. 4. Steady-state biomass concentration in a continuous stirredtank bioreactor as function of dilution rate.Fig. 5. Experimental data of ferrous iron oxidation rate in acontinuous stirred tank bioreactor. J.M. Go´ mez, D. Cantero / Process Biochemistry 38 (2003) 867     /  875 870
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks