A new VFA sensor technique for anaerobic reactor systems

A new VFA sensor technique for anaerobic reactor systems
of 8
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
  A New VFA Sensor Technique forAnaerobic Reactor Systems Peter F. Pind,* Irini Angelidaki,* Birgitte K. Ahring BioCentrum-DTU, Technical University of Denmark, Building 227, DK-2800 Lyngby, Denmark; telephone: +45 4525 6183 or +45 45 25 61 86; fax: +45 4588 3276; e-mail:  Received 24 April 2002; accepted 5 September 2002 DOI: 10.1002/bit.10537  Abstract:  A key parameter for understanding and con-troling the anaerobic biogas process is the concentrationof volatile fatty acids (VFA). However, this informationhas so far been limited to off-line measurements usinglabor-intensive methods. We have developed a newtechnique that has made it possible to monitor VFA on-line in one of the most difficult media: animal slurry ormanure. A novel in situ filtration technique has made itpossible to perform microfiltration inside a reactor sys-tem. This filter enables sampling from closed reactor sys-tems without large-scale pumping and filters. Further-more, due to its small size it can be placed in lab-scalereactors without disturbing the process. Using this filtra-tion technique together with commercially availablemembrane filters we have constructed a VFA sensor sys-tem that can perform automatic analysis of animal slurryat a frequency as high as every 15 minutes. Reproduci-bility and recovery factors of the entire system have beendetermined. The VFA sensor has been tested for a periodof more than 60 days with more than 1000 samples onboth a full-scale biogas plant and lab-scale reactors. Themeasuring range covers specific measurements of ac-etate, propionate, iso-/  n  -butyrate and iso-/  n  -valerateranging from 0.1 to 50 m M   (6–3000 mg). The measuringrange could readily be expanded to more componentsand both lower and higher concentrations if desired. Inaddition to the new VFA sensor system, test results fromdevelopment and testing of the in situ filtration tech-nique are being presented is this article.  © 2003 Wiley Pe-riodicals, Inc.  Biotechnol Bioeng   82:  54–61, 2003. Keywords:  anaerobic manure treatment; in situ filtration;on-line measurement; VFA INTRODUCTION Volatile fatty acids (VFA) are some of the most importantintermediates in the anaerobic biogas process. Conversionof the VFA through the acetogenic and acetoclastic step intomethane and carbon dioxide is the most important conver-sion in the biogas process. It is well recognized that moni-toring the specific concentration of VFA can give vital in-formation on process status (Ahring et al., 1995; Alonso,1992; Cobb and Hill, 1991; Hickey and Switzenbaum,1991; Hill, 1982; Hill and Bolte, 1989; Hill and Holmberg,1988; Mo¨sche and Jo¨rdening, 1999; O¨ztu¨rk, 1991; Pind et al.,1999; Rozzi et al., 1997; Sørensen et al., 1991). Severalauthors have shown that especially the isoforms of butyrateand valerate are fast and reliable indicators of changes in theprocess balance (Ahring et al., 1995; Cobb and Hill, 1991;Hill and Bolte, 1989; Hill and Holmberg, 1988), while pro-pionate and butyrate levels are known to increase if thehydrogen level increases to inhibiting levels (Ahring et al.,1995; Mo¨sche and Jo¨rdening, 1999; O¨ztu¨rk, 1991). Theactual effect of VFA levels on the different microorganismsinvolved in the biogas process is quite complex and depen-dent on the actual species involved, e.g., methanogens(Mladenovska and Ahring, 2000; Mo¨sche and Jo¨rdening,1998; Vavilin and Lokshina, 1996). Inhibition by increasedVFA concentration is also known to be dependent on pH(Meyer and Heinzle, 1998; Mo¨sche and Jo¨rdening, 1999).Sudden increases in VFA concentration will cause pH todecrease if the alkalinity is low. This very complex inter-action of inhibition, substrate affinity, and pH dependencyleads to the conclusion that no general assumptions on in-hibitory levels of VFA are possible. Instead, a more com-plex evaluation of the VFA concentration has to be con-ducted (Ahring et al., 1995; Bjo¨rnsson et al., 1997; Pind et al.,1999; Pullammanappallil et al., 1998; Renard et al., 1988;1991). Such a complex evaluation necessitates access tomuch and frequent information on VFA concentrations.VFA is easily measured using GC or HPLC, provided thatall particulate matter has been removed from the sample.When dealing with anaerobic waste treatment, the presenceof particulate matter is often high.Only a few reports exist on development of an on-lineVFA monitoring systems. Slater et al. (1990) placed a filteron the recirculation loop of a fluidized-bed reactor andtransferred the final permeate to a GC with a modified in- jection port allowing on-line analysis every 12 minutes.However, no data from these measurements and no valida-tion of the suitability of this procedure were ever published.Ryhiner et al. (1993) used a 0.45   m filter on a similarrecirculation loop acidifying the permeate with 1% formicacid. The sample was transferred to a specially designed Correspondence to:  Birgitte K. Ahring* Present address:  Environment & Resources DTU, Technical Univer-sity of Denmark, Building 115, DK-2800 Lyngby, Denmark.Contract grant sponsor: Danish Energy CouncilContract grant number: EFP no. 1383/98-0014 © 2003 Wiley Periodicals, Inc.  flow-through vial and then injected into the GC. Acetic,propionic, butyric and valeric acids, including their iso-forms, were analyzed and followed for a period of 5 hours.Again, no validation of the procedure was published andthis method has to our knowledge not been included in otherstudies.Zumbusch et al. (1994) used a membrane with a normalmolecular weight cut-off (NMWC) of 20,000 g (equal to20 kDa) for sampling in a UASB reactor. The permeate waspumped through a gas separator and then analyzed in aHPLC using an injection valve. The accuracy of the HPLCwas much lower than normally achieved by GC; however,the HPLC was not specifically optimized for this purpose.During the 40 hours the system was operated it showedproblems with fouling.To our knowledge, no on-line VFA measurement tech-nique has previously been developed for use on full-scaleanaerobic systems treating solid wastes. The published at-tempts all suffer from problems with membrane fouling anddifficulties with the transfer of a representative sample to aGC or HPLC. Fouling of the membrane reduces the perme-ate flow. A high permeate flow can be obtained by usingmembranes with a relatively large surface area (0.1–1 m 2 ),but large surface areas result in large hold-up volumes andlarge flows in the membrane cartridge, thus increasing thedemands for flow rates. Instead, small hollow fiber mem-brane cartridges can be used with a relatively high surface/ hold-up volume ratio. At present, commercial membranefilters are available that can perform ultrafiltration on smallsample volumes, provided that the maximum particle sizedoes not exceed 0.1–1 mm. However, solid waste oftencontains particulate matter of much larger sizes.In the present study we present a newly developed meth-odology for on-line VFA measurements in manure reactorsalong with a complete test and validation of the VFA sensor.The development of the sensor required the use of samplepurification by filtration, which introduces the possibility of concentration changes in the sample. Therefore, verificationof sample reproducibility for each part of the sampling tech-nique has been performed in both laboratory and full-scaleapplications. MATERIALS AND METHODSSensor Principle A schematic layout of the sensor system is shown in Figure1. The first part is a rotating filter for performing the pre-filtration that was placed directly in the biogas reactor. Theprefiltration technique was able to supply a sufficientamount of sample to a recirculation loop on a membranecartridge. Furthermore, the prefiltration technique resultedin a small retention time, less than 1 min. Second, an ultra-filtration unit was used which was able to remove most of the organic particles larger than approximately 100,000 Da.Third, a sample preparation unit mixes the sample withphosphoric acid and removes precipitates and gasses. Theflux/hold-up volume ratio was optimized resulting in a com-plete breakthrough-time of less than 10 min. Finally, a GCunit is used to perform the actual VFA analysis. Because of the presence of many organic compounds in the ultrafilteredmedium extra time is used for burn-out of the GC-column(5 min). This increases the analysis time from optimally3–5 min to 8–10 min, excluding time required to cool thecolumn. Detailed Sensor System A schematic layout of the sample preparation system isshown in Figure 2. The rotating prefilter (1) has a pore sizeof 60   m and an effective area of 25 cm 2 . The prefilter isplaced directly in the biogas reactor or in the recirculationloop in a full-scale reactor. Patent application has been filedfor both the microfiltration technique and the whole on-linesystem (Danish patent application numbers PA 2000 01013and 2000 01014). The hold-up volume between the cleaningvalves (6) is 55 mL and is initially flushed for 30 s withprefiltered medium prior to running a recirculation pump(5). The recirculation pump increases the flow and pressureto 1 m/s and 0.8–0.9 atm in an ultra-membrane (4): UFP- Figure 1.  Schematic layout used for the new VFA sensor. Figure 2.  Schematic illustration of the sample preparation system usedfor on-line VFA analysis in biogas systems. 1: rotating prefilter placed insitu. 2: sampling port for control extraction by syringe (prefilteredsamples). 3: peristaltic pump. 4: ultra-membrane. 5: recirculation pump.6: three-way valve for bypassing cleaning fluid for regeneration of theultra-membrane. 7: sampling port for control extraction by syringe (ultra-filtered samples). 8: nonreturn valve. 9: back-flushing pump. 10 and 11:linked peristaltic pumps pumping equal amounts of ultra-filtered sampleand 1% (w/v) phosphoric acid (15), respectively. 12: waste pump.13: overflow pump. 14: pump back-flushing with neutralizing liquid (16).17: Minifilter. 18: Flow-cell (vial). PIND ET AL.: A NEW VFA SENSOR TECHNIQUE FOR ANAEROBIC REACTORS 55  100-E-4A membrane (A/G Technology Corporation)100,000 NMWC, area 420 cm 2 . The hold-up volume (ap-proximately 20 mL) of the ultra-membrane cartridge isflushed with filtered permeate for 6 min prior to sampling.The permeate is returned through a nonreturn valve (8)while the flushing is performed, thus minimizing the sam-pling amount removed from the reactor system. Equalamounts of sample and 1% (w/v) phosphoric acid (15) arethen mixed by a peristaltic pump (10) and (11). This samplemixture is first removed by (12), flushing all previoussample hold-up volumes and afterwards passing it through amini-filter (17), capturing possible precipitate formed in thesample mixture. An open sample vial or container (18) isflushed with the sample mixture, allowing degassing of car-bonate. Overflow from the vial (18) is removed by an over-flow pump (13). The sample is then transferred by a auto-injector syringe to a GC. The sample is analyzed by usingthe auto-injector system on a Shimadzu GC-14A equippedwith a fused silica column (Nukol Poly(ethylene glycol),0.53 ID, 15 m, 0.5   m film). Injection: 150°C, detector250°C, column temperature 90°C (1.5 min) increased to110°C (during 2 min) increased to 195°C (during 2.12 min)and maintained for 5 min. After injection, the flow-cell (18)and mini-filter (17) are back-flushed with neutralizing liq-uid (16) to a waste container by (12) and (14). Finally, theflow-cell and tubes are emptied by (12) and the system isnow ready for the next sample. The complete set-up(pumps, valves, motors, and GC) is controlled by a speciallydesigned computer program allowing precise and reproduc-ible sample preparation. Validation and testing of the indi-vidual parts of the sensor system was performed in bothlaboratory and full-scale application as described below. Laboratory Reactors 4.5-L reactors with active volumes of 3.5 L were used forthe experiments. The reactor design was as previously de-scribed (Angelidaki and Ahring, 1993). Prefiltration testswith a rotating filter were performed on a mesophilic reac-tor, fed 3 times per day, with an average organic loading of 2.55 g VS/L . d (15 d HRT). Temperature was kept at 35°C.Complete VFA sensor tests were conducted on a similarreactor, fed every 6 h, with an average organic loading of 4.45 g VS/L . d (15 d HRT). The temperature was kept at55°C. Manure composition for both reactor experiments aswell as the full-scale reactor is shown in Table I.Gas was collected from the headspace of the effluentstorage. The gas flow was monitored using a modified liq-uid displacement technique similar to the one previouslydescribed (Angelidaki et al., 1992), using paraffin oil asliquid in a U-tube with a working volume of 10 mL. Analytical Methods Gas composition was measured as previously described(Angelidaki et al., 1990). Samples for VFA analysis wastaken either from the reactor content (5 mL) , the prefiltra-tion loop [2 mL, (D); Fig. 3] and/or the ultrafiltration loop[1 mL, (7); Fig. 2] by a syringe. Reactor samples wereextracted through a sample tube placed on top of the reactor[(F) in Fig. 3] using a 50-mL syringe to flush the tube priorto sampling. Reactor samples were diluted with tap water,and 1 mL was centrifuged after acidification with 30   L17% (w/v) phosphoric acid in an Eppendorf tube. 0.5 mLprefiltrate or ultrafiltrate sample were mixed with 0.5 mL1% (w/v) phosphoric acid in an Eppendorf tube and werecentrifuged afterwards. The supernatant was analyzed by aGC equipped with a flame-ionization detector. Manuallycollected samples were analysed on a HP 5890 GCequipped with a HP FFAP column (Bonded and modifiedcrosslinked poly ethylene glycol), 0.53 ID, 30 m, 1.0   mfilm). Injection: 175°C, detector 200°C, column tempera-ture 100°C (0.5 min) increased to 125°C (during 3 min)increased to 200°C (during 1.67 min) and maintained for7 min. 1   L sample was injected. All samples were com-pared to standards in the range from 1 to 50 m  M   for C 1 –C 4 and from 0.1 to 5 mM for C 5 .Total solids and volatile solids were determined accord-ing to standard methods (APHA-AWWA-WPCF, 1975). Prefiltration Tests A maximal flux capacity of the rotating filter (with a poresize of 60   m) was tested at varying temperature and total Table I.  Manure composition for biogas reactors used for testing pre-filter and sensor.ComponentMesophilic reactor experiments Thermophilic reactor experimentsFull-scalereactor manureRaw cowmanureInfluent60% manureReactormanureInfluent70% manureReactormanureTS g/L 87.4 52.7 37.1 76.6 55 48.5VS g/L 63.7 38.4 27.5 60.2 39.5 30Acetate g/L 6.34 3.82 0.07 5.3 0.57 1.14Propionate g/L 2.50 1.51 0.0 1.88 0.19 0.21Butyrate g/L 1.47 0.89 0.0 1.14 0.02 0.04Valerate g/L 0.59 0.36 0.0 0.37 0.06 0.01Total VFA g/L 10.91 6.57 0.07 8.72 0.84 1.40g Acetate/L 9.72 5.85 0.07 7.84 0.77 1.38Carbonate g CO 2  /L 3.75 2.26 7.75 5 12 12.10pH 7.09 7.09 7.41 7.41 7.6 8.01 56 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 82, NO. 1, APRIL 5, 2003  solids concentrations. The filter was mounted in a labora-tory reactor as previously described (Fig. 3). The reactorcontained digested manure (Table I) from a mesophilic re-actor with a TS of 35 g/L. The temperature of the reactorwas controlled using the reactor heating control. Rotatingspeed was controlled using a Heidolph RZR 2050 stirrer. Aperistaltic pump controlled flow from the filter. A liquidmanometer measured the suction vacuum on the filter. Dif-ferential pressure varied from 0.05 to 0.15 atm, with in-creasing differences at higher fluxes. Maximum flux wasestimated as the flux obtained just prior to filter constipa-tion, observed as a rapid and irreversible pressure drop.Total solids concentrations were increased by adding rawundiluted manure to the reactor while subtracting an equalamount of volume from the filtered stream, thereby increas-ing the TS content up to 106 g/L.Comparison of the VFA content in reactor samples andfiltered manure was made to validate VFA reproducibilityusing the 60-  m filter pore size. Samples were taken fromthe mesophilic reactor while conducting pulse additions of manure to obtain higher VFA variation. Ultrafiltration Tests Ultrafiltration tests were conducted on a set-up similar toFigure 2 [without instrument (1), (10)–(18) in Fig. 2]. Theultrafiltration unit consisted of an A/G-UFP-100-E-4Amembrane (A/G Technology Corporation), membrane area420 cm 2 , average pore size 100,000 NMWC. The flux wasmeasured by temporarily bypassing the permeate to a mea-sure beaker. Permeate samples for VFA analyses were takenfrom a specially designed septum (hold-up volume less than0.1 mL). The membrane cartridge volume exceeded themembrane flux that results in a non-instant recovery of VFAconcentrations in the ultrafiltered medium. The VFA con-centration recovered in the ultrafiltered medium will slowlyapproach the actual VFA concentration in the media recir-culating inside the membrane. Breakthrough of the mem-brane is defined as the percentage of the VFA concentrationrecovered in the ultrafiltered medium compared to the VFAconcentration in the prefiltered medium recirculation insidethe membrane. Breakthrough tests were conducted on abatch of prefiltered manure previously frozen to minimizepossible conversion of VFA during the test. Injecting apulse of acetate momentarily increased the acetate level inthe manure.The effect of fouling was monitored by using the com-plete set-up (Fig. 2) over a longer period of time on afull-scale biogas plant. The system ran continuously andsampling was conducted within 15-min intervals. Each in-terval included an active pump recycling on the membraneof 6 min resulting in a membrane “use-time” of 9 h and36 min every day. Fouling would only build up duringthe 6 min of pumping. If the flux decreased to less than15 L/(m 2   h), the membrane was washed with hot water(60°C) for 1 h, followed by back-flushing of the membrane. Full-Scale Test The complete system and the sample preparation systemwere tested on a full-scale biogas plant (Snertinge BiogasPlant, Denmark). Samples were taken from the recirculationloop servicing two reactors operating in the temperaturerange from 37–45°C with a reactor TS content ranging from44 to 48 g/L. Average manure characteristics are shown inTable I. The system was used and tested regardless of tem-perature and TS changes that occurred as part of the normaloperation. During the test, prefiltered and ultrafiltered me-dia were not returned to the sampling point (as opposed tothat shown in Fig. 2). Instead, the medium was sent to theinfluent tank of the plant.Membrane breakthrough was tested for each sample byback-flushing the outer membrane cartridge with 20 mL of distilled water [(9) in Fig. 2], ensuring that VFA from theprevious sample would not significantly influence the VFAmeasurement of the new sample. This enabled estimation of the recovery factor in the ultrafiltered sample, which willdecrease with decreasing flux. More than 1000 sampleswere taken and analyzed automatically during a period of 2 months. Control samples from reactor, the prefiltered andultrafiltered media were taken periodically during the test-ing of the sensor system. RESULTS AND DISCUSSIONPrefiltration Tests Preliminary test with a fixed filter (dead-end filtration)showed clogging within seconds and would therefore re- Figure 3.  Lab-scale reactor set-up including prefiltration unit. (A) Influ-ent of heating/cooling water to control temperature; (B) rotating filterplaced inside reactor; (C) liquid manometer; (D) sampling point for controlextraction by syringe; (E) sample outlet for large sample amounts,e.g., calibration of flow rate; (F) sample tube for reactor samples. PIND ET AL.: A NEW VFA SENSOR TECHNIQUE FOR ANAEROBIC REACTORS 57  quire a continuous cleaning of the filter. Backwashing wasevaluated to be unsuitable for this dead-end filtration. Forcrossflow filtration, both tubular and plate filtration wouldrequire crossflow velocities higher then 1 m/s. This wouldrequire unreasonably high recirculation rates in the range of m 3  /h. Therefore, it was decided to experiment with a reverseimplementation of crossflow: moving the filter surface in-stead of the liquid. Rotating a tubular filter inside the mediashowed high flux capabilities with very small dimensions.By having the filter freely submerged in the medium, par-ticles larger than the pore size employed could not clog thefilter. The pore size of 60   m, was chosen based on com-mercial availability of filters with a pore size lower then100   m still having a high pore area per filter area. Thisunique filter construction occupies only 12–20 cm 3 (length:4–6 cm and diameter: 2–3 cm) inside the reactor system andcan be placed in laboratory reactors and even standardpiping in full-scale plants. Pumping and pressure controlis applied on permeate (not on the medium), thus reduc-ing pumping and hold-up volumes considerably (less than20 mL).It was clear to us that a certain velocity barrier had to beovercome before a more or less linear relationship betweentangential speed and flux capacity was obtained (Fig. 4).This linear relationship can be modelled by a proportional-ity factor between the tangential speed of the filter surfacev 2  and the linear flow velocity through the filter surface v 3 .Adding the two vectored speeds and multiplying with afactor   , one should obtain the resulting cross flow velocityv 1  of the medium. The factor    is assumed to be dependenton the composition and viscosity of the medium.v 1  =     ( v 2  +  v 3 ) ( 1 ) Like the    factor, the linear proportional factor between v 2 and v 3  is assumed to be a function of particle compositionand viscosity of the medium.The 60-  m pore size filter could supply 2000 L/(m 3   h)at 1 m/s at normal reactor conditions: TS lower than 50 g/Land temperatures higher than 30°C, corresponding to ap-proximately 80 mL/min with an area of 25 cm 2 . It wasexpected that the hold-up volume of an ultrafiltration unitwould be 50–70 mL. Using a flow of 80–100 mL/min pre-filtered material a 99% breakthrough could be obtainedwithin 5 min, provided that this hold-up volume was com-pletely mixed, and lower than 1 min at plug-flow condi-tions. This was considered an acceptable time delay com-pared to the typical GC analysis time of 10–15 min.It was, therefore, decided to use a pumping rate of 80 mL/min and rotating rate of 1000 rpm during a 3-monthtest on the mesophilic reactor. An average run time of 2 hper day, 5 d per week was employed to see whether thereactor process was disturbed or showed a decreased biogasyield. The average biogas yield of 218 mL CH 4  /g VS wascompared to an average biogas yield of 215 mL CH 4  /g VSfrom a control reactor. The difference of 1.4% was compa-rable to the hold-up volume in the pumping loop of the filtercompared to the reactor volume of 3.5 L. However thedifference lies within the accuracy of ± 5% and it wasconcluded that the use of the rotating filter would not causeextra biological stress on the laboratory reactors. The pre-filtered manure had a TS content of 24–25 g/L when thereactor was running at normal conditions. The biogas po-tential of the prefiltered manure was one third of the reactorcontent (data not shown). An increase in the TS of theprefiltrate was observed when the reactor TS was artificiallyincreased up to 106 g TS/L, but it never exceeded 35 g/L.Testing of the consistency between the VFA concentra- Figure 4.  Maximal flux capacity of a 60-  m pore size rotating filter in digested manure with a TS concentration varying from 35 to 106 g/L as a functionof the tangential speed and temperature varying from 7 to 38°C. Differential pressure on the filter was in the range of 0.05 to 0.15 atm. 58 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 82, NO. 1, APRIL 5, 2003
Similar documents
View more...
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