BusinessLaw

Proton transfer reaction mass spectrometry technique for the monitoring of volatile sulfur compounds in a fuel cell quality clean-up system

Description
Proton transfer reaction mass spectrometry technique for the monitoring of volatile sulfur compounds in a fuel cell quality clean-up system
Categories
Published
of 11
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
  Protontransferreactionmassspectrometrytechniqueforthemonitoringof volatile sulfur compounds in a fuel cell quality clean-up system Davide Papurello a,b, ⁎ , Lorenzo Tognana c , Andrea Lanzini a , Federico Smeacetto e , Massimo Santarelli a ,Ilaria Belcari b , Silvia Silvestri b , Franco Biasioli d a Department of Energy (DENERG), Politecnico di Torino, Corso Duca degli Abruzzi, 24, 10129, Turin, Italy b Fondazione Edmund Mach, Biomass and Renewable Energy Unit, Via E. Mach, 1, 38010 San Michele a/A, Italy c SOFCpower S.P.A., Viale Trento, 115/117, c/o BIC, I-38017 Mezzolombardo, TN, Italy d IASMA Research and Innovation Centre, Fondazione Edmund Mach, Food Quality and Nutrition Area, Via E. Mach 1, 38010 S Michele a/A, TN, Italy e  Applied Science and Technology Department (DISAT), Politecnico di Torino, Corso Duca degli Abruzzi, 24, 10129 Turin, Italy a b s t r a c ta r t i c l e i n f o  Article history: Received 13 June 2014Received in revised form 25 September 2014Accepted 28 September 2014Available online 20 October 2014 Keywords: Volatile sulfur compounds (VSCs)BiogasSolid oxide fuel cell (SOFC)Organic fraction of municipal solid waste(OFMSW)Proton transfer reaction-mass spectrometry(PTR-MS) BiogasfromthedryanaerobicdigestionofOFMSWfromapilotplant wasanalyzedintermsofsulfurcompoundremoval through a gas cleaning section based on activated carbons, from lab.scaletoreal plant. In general,eventhe presence of sub-ppm(v) of selected biogas contaminants can hamper the life-time of SOFC systems. For thisreason,stringent fuelcell quality requirementsapply. The challenge ofreal-time monitoring of theperformanceandqualityofthefuelfeedingtheSOFCcanbesolvedthroughtheuseofPTR-MS.Thistechnique – onceproperlyand preliminary calibrated as shown in this study  –  has the capability of rapidly resolving the wide spectrum of contaminants slipping from the clean-up section. A commercial sorbent material was adopted to remove sulfurcompounds and was tested for 80 h in a pilot gas cleaning system. H 2 S, the main sulfur compound detected(99.36% of total sulfurs) was removed to a satisfactory level. The sulfur compounds elute from the cleaning sec-tion in the following order: CH 3 SH, CH 3 SCH 3 , CH 3 CH 2 CH 2 SH, CH 3 (CH 2 ) 3 SH, CS 2  and H 2 S. The  󿬁 lter section wasabletoprovideacleanbiogas(1 ppm(v))throughoutthewholeexperimentaltrial(almost450h)withanaver-age H 2 S inlet concentration of 52 ppm(v).© 2014 Elsevier B.V. All rights reserved. 1. Introduction Amongst the various biogas production sources, the anaerobicdigestion of organic fraction of municipal solid waste (OFMSW) offersthe possibility to obtain a valuable bio-fuel thus recovering the energycontentofawastethatwouldbeotherwisedisposedinland 󿬁 llcontrib-uting to atmospheric, soil and water pollution. Several applications of biogas have been investigated, mostly for combined heat and powersystems (CHP) with internal combustion engines (ICEs) [1]. A growinginterest in the  󿬁 eld of energy production was more recently gained bymicro-turbines or by high temperature fuel cells systems, i.e., moltencarbonate technology fuel cells (MCFCs) or solid oxide fuel cells(SOFCs) [2].At present, fuel cell systems suffer from still very high investmentcosts. However the potential for higher electric power generationef  󿬁 ciency remains, which means increased fuel saving [3], while at thesame time reducing atmospheric emissions (nitrogen oxides (NOx),sulfur oxides (SOx), carbon monoxide (CO) and volatile organic com-pounds (VOCs)). In addition, the carbon capture storage or re-useoption in fuel cell systems can be achieved more readily than withconventional combustion power systems [4 – 7], mostly thanks to theabsenceofnitrogeninthefuelexhaust.Oneofthemainproblemsrelat-ed with the energy production with SOFC systems is the biofuelsreforming into hydrogen, e.g. biogas and the trace compounds impacton cell performance. Lanzini and Leone [6] investigated on the perfor-mances of SOFC fed by simulated biogas mixtures comparing differentdirect reforming options [8,9].SofarstudieswithrealbiogasfeedingonSOFCgeneratorshavebeenfew. The importance of investigating real biogas feeds is essentially re-lated to the presence of micro-contaminants and their removal tomeetthestringentfuelcellrequirements[10]fortheSOFCanode.Inad-dition to the main biogas constituents (methane and carbon dioxide),trace compounds can seriously affect SOFC systems including thereforming section [11]. Sulfur and chlorine compounds, as shown bySasaki et al. [8], affect the fuel cells and the reforming section throughnickel deactivation, being sulfur far more deleterious than chlorine.Ourstudies demonstrate how siloxanecompounds affectpreferentiallythe interconnectors, limiting the fuel  󿬂 ow to reach the nickel activesites. The subsequent detrimental action of siloxanes concerns the Fuel Processing Technology 130 (2015) 136 – 146 ⁎  Corresponding author at: Department of Energy (DENERG), Politecnico di Torino,Corso Duca degli Abruzzi, 24, 10129 Turin, Italy. Tel.: +39 3402351692. E-mail address:  davide.papurello@polito.it (D. Papurello).http://dx.doi.org/10.1016/j.fuproc.2014.09.0410378-3820/© 2014 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Fuel Processing Technology  journal homepage: www.elsevier.com/locate/fuproc  threephaseboundaryobstructionwiththeconsequentelectrochemicalactive site reduction [9]. As a conclusion, the most detrimental andabundant biogas trace compound for the nickel anode deactivation ishydrogensul 󿬁 deasreportedinliteraturestudies[7,8,11].Nevertheless,Papurello et al. [12,13] have shown that other biogas trace compoundsbesides hydrogen sul 󿬁 de could have a role. These are for instancemethanethiol, dimethylsul 󿬁 de, carbon disul 󿬁 de, propanethiol andbutanethiol, which are all detected in the biogas from OFMSW anaero-bic digestion.Studies from Hernandez et al. (2008 – 2011) reported that a propergas clean-up section is essential to feed SOFC energy generators [14,15]. The effectiveness of micro-contaminant removal widely dependsonthebiogastracecompounds'variability.Sotherearenostandard so-lutions; a rather speci 󿬁 c combination of impurity removal methodsmust be used to ensure a fuel gas of the quality that meets the fuel celltolerancethresholdsasgivenbythemanufacturer[16]. Severalauthors[17 – 19]indicate thatthefollowingstepsare required: a primaryclean-up step, in which a condenser and a  󿬁 rst sorbent bed are inserted,followed by a  󿬁 ne guard bed before delivery of the biogas to the fuelcell system.A condenser is useful in removing water because of its competitiveadsorption behavior towards other contaminants in the downstreamguardbeds.Siloxanesaregenerallyheavierthanotherbiogastracecom-pounds and thus easily removed from biogas by water condensation.Nevertheless, according to results by Hagmann et al. [20] on siloxanecompounds, the removal through a condenser section achieves only a26vol.%degreeofseparationat − 25 °C,whereasat − 70 °Ctheremovalratio increased up to 99vol.%. Schweigko 󿬂 er and Niessner [21] pointedout that, by cooling the gas to 5 °C, approximately 12vol.% of siloxanecompoundsareremoved.Therefore,theseresultsshowhowtheremov-alofVOCsbyacondensersectionisnotsigni 󿬁 cant,butonlystrictlynec-essary to remove water from the biogas stream in order to protect theactivatedcarbonguardbedlifetime[15].Activatedcommercialcarbons,impregnated with metal oxides such as iron, copper and silver, arewidely used mainly to remove sulfur compounds [15 – 17]. Other sor-bentmaterialssuchasZincOxide(ZnO)arealsousedtoremovevolatileorganic compounds as reported by Hernandez et al. [14]. Nano particle zincoxidehasahigheradsorptioncapacitywithrespecttothecommer-cialactivated carbonforH 2 Sremoval.InPapurello et al.[13]it isshownhow the breakthrough fraction and the breakthrough time are affectedby the type of the sulfur compounds that have to be removed, by thegashourly space velocity (GHSV) in thegascleaningsection and 󿬁 nallyby the presence of vapors of organic compounds besides sulfur ones. Itwas demonstrated that even only 1 ppm(v) of aromatic, carbonyl andCl compounds can reduce the removal  󿬁 lter ef  󿬁 ciency by 11% of theexpected value [13]. It is evident that further investigation of the ef  󿬁 -ciency obtainable by cleaning  󿬁 lter in fuel cell related applications isnecessaryespeciallyunderrealworkingconditions,asthoseinthecou-pling with an anaerobic digester pilot plant.High-sensitivityandrobustmethodsforthereal-timeanalysisofthevolatilecompoundsreleasedbyOFWSWdigestionareavaluabletooltobetter understand and support the development of better industrialprocedures for biogas exploitation. In this context, direct InjectionMass Spectrometry (DIMS) offers interesting performances in terms of rapidity, sensitivity and absence of pre-treatments [22]. One of themost promising DIMS techniques is certainly PTR-MS. It is based on anef  󿬁 cient implementation of chemical ionization based on proton trans-ferfromhydroniumions[22]andallowstherapidandon-linemonitor-ing of most volatile compounds. It has been applied in many situationsrangingfrom breath analysis toenvironmentalmonitoringand,recent-ly, also to issues related to waste management and odorant emissioncontrol [23 – 27].In this study, the monitoring of contaminants contained in a biogasproduced from a pilot plant loaded with OFMSW is assessed beforeand after a gas cleaning section. The as-produced biogas from the di-gester  󿬁 rst passed through a condenser and then through an activatedcarbon 󿬁 lterbed.Tracecompoundscontainedinsuchabiogaswerean-alyzed by PTR. A gas calibration unit was adopted in combination withthe PTR-MS instrument in order to con 󿬁 rm the quanti 󿬁 cation obtainedby PTR-MS. 2. Material and methods  2.1. Gas cleaning section  —  pilot plant  The dry anaerobic digestion of OFMSW was conducted in a pilotplant located at Fondazione Edmund Mach (FEM) (S. Michele a/A,Italy). A detailed description of the pilot plant digester can be found in[12]. A gas cleaning section was built according to the scheme inFig. 1. The biogas line was connected to a condenser  󿬁 lled by plasticrings with an overall capacity of 54 L. Two  󿬁 lter reactors are connectedin seriesafter thecondenser.The 󿬁 rst 󿬁 lter reactor was 󿬁 lled withplas-tic rings to entrain small water drops, while the second one with com-mercial Sulfatrap R8 activated carbons (TDA research Inc., USA). The Nomenclature AD anaerobic digesterACFs activated carbon  󿬁 ltersCAR carboxen  󿬁 berC/C 0  removal performance ef  󿬁 ciency, ratio between actualconcentration (C) and starting concentration (C 0 )CHP combined heat and power systemDIMS direct injection mass spectrometryDPM dew point mirrorDVB divinylbenzene  󿬁 berEDS energy-dispersive X-ray spectroscopyFC fuel cellGC-MS gas chromatography mass spectrometryGCU gas calibration unitGHSV gas hourly space velocityICE internal combustion engineMCFC molten carbonate fuel cellsMFC mass  󿬂 ow controllerm/z mass overOFMSW organic fraction of municipal solid wastePA proton af  󿬁 nityPDMS polymeric dimethylsiloxane membranePEG polyethylene glycol  󿬁 berPEEK polyether ether ketonePFA per 󿬂 uoroetherPM polymeric membraneppb(v) parts per billion (volume)ppm(v) parts per million (volume)ppq(v) parts per quadrillion (volume)ppt(v) parts per trillion (volume)PTR-MS proton transfer reaction-mass spectrometryQMS quadrupole mass analyzerSEM scanning electrode microscopySOFC solid oxide fuel cellSPME solid phase micro extractionTd Townsend, unit for measuring E/N, 1 Townsend (1Td) = 10 − 17 Vcm 2 mol − 1 THT tetrahydrothiopheneToF time-of- 󿬂 ight (analyzer type of mass spectrometers)VMRs volatile mixing ratiosVOCs volatile organic compoundsVSCs volatile sulfur compoundsZnO zinc oxide sorbent material 137 D. Papurello et al. / Fuel Processing Technology 130 (2015) 136  – 146   󿬁 lterreactor'svolumewasabout2L.TheGHSVvalueadoptedwas 󿬁 xedat300h − 1 whilethetemperatureandpressurewere25 °Cand1.3bargwith a biogas saturated with water (relative humidity 100%). The acti-vated carbon mass loaded into the second  󿬁 lter reactor was 863 g,while the  󿬁 rst one was  󿬁 lled with  󿬁 lling bodies. A pressure controller(Bürkert GmbH & Co. KG, Germany) was inserted into the gas line inorder to control an electro-valve (Bürkert GmbH & Co. KG, Germany)throughapressuresensor(WIKAInstrument,LP,USA).Thiscontrolsys-temkeepsthepressureatthedigesteroutletat3 mbarginordertopro-tect the microbiological activity. A 260 Wel vacuum pump (KNFNeuberger GmbH, Germany) was inserted at the end of the line inorder to ensure the required pressure difference. Two sampling pointswere used to  󿬁 ll Nalophan bags before and after the cleaning section.Biogassamples were collected, 8 L Nalophan bags, and sealed with Tef-lon stoppers. Samples were analyzed within 30 min from sampling tominimize VOC losses or artifacts [28]. To avoid or reduce saturation, the biogas was diluted with nitrogen (1:5) and incubated at 35 °C for30 min using a thermostatic bath (Techne Ltd., Cambridge, UK).Postmortem analysis were performed byscanningelectron micros-copy (SEM) (FEI Inspect, Philips 525 M) coupled with EDS (SW9100EDAX) analysis.  2.2. Gas cleaning section  —  󿬁 lter cartridges Filter cartridges section were built in a laboratory to compare thevolatile organic compound removal between natural gas and biogas.The  󿬁 lter section was  󿬁 lled by commercial activated carbons SulfatrapR8 (TDA research Inc., USA). Two electronic mass  󿬂 ow controllers(MKS instruments Inc., USA) were adopted to feed carbon cartridgeswith the biogas and natural gas pollutants. Single  󿬁 lter bed con 󿬁 gura-tions were tested according to the scheme reported in Fig. 2 using PFAtubes (1/4 in. diameter, length ~3 m) and  󿬁 ttings (Swagelok Ltd.,USA). Carbon cartridges were built with Te 󿬂 on tubes (4 cm length and6 mm diameter), by positioning 1 g of activated carbon in the middleof the cartridge, and placing gauze at both ends as physical support forthecarbon.Theblankcartridgewaspreparedinthesamewaybutwith-outthecarbonsampleinthemiddle.Eachcartridgehasbeenfedcontin-uously for 6 h with biogas or natural gas. The temperature was  󿬁 xed at25 °C with a GHSV value  󿬁 xed at 110 h − 1 .  2.3. PTR-MS for monitoring contaminants A PTR-MS instrument was used for monitoring the VSCs and theother organic compounds. PTR-MS has already been successfullyused to investigate the removal performance of adsorbent materialsemployed for gas  󿬁 ltration from gas cylinder [12,24]. PTR-MS is a highsensitivity mass analytical instrument particularly well suited for con-centration measurements of trace gases down to parts-per-quadrillionvolume(ppq(v))[29].ThebasicprincipleofPTR-MSisthechemicalion-izationwithhydroniumions,H 3 O + .Itisadirectinjectionmassspectro-metric technique and it does not require sample preparation [27].Measurements have been performed by a PTR-QMS 300 (IoniconAnalytik GmbH, Innsbruck, Austria) which description is reported in[27].Theexitgasisdirectlyinjectedintothedrifttubeoftheinstrument Fig. 1.  Biogas  󿬁 ltration pilot plant set-up. Fig. 2.  Biogas and natural gas  󿬁 lter cartridges set-up (PM — polymeric dimethylsiloxane membrane; MFC — mass  󿬂 ow controller).138  D. Papurello et al. / Fuel Processing Technology 130 (2015) 136  – 146   viaaheated(80 °C)PEEKtube.Thesamplingtimeiscloseto5sforeachmass to charge ratio upto m/z 200. The drift tubeconditionsare 600 V,2.01 mbar and 80 °C. For every sample, 10 spectra are acquired beforeproceeding to the next sample. Each sample was collected every hour.With given drift conditions, it is possible to estimate, in principle with-out any calibration, the VOC concentration in the gas phase via the fol-lowing equation: R ½  ¼  RH  þ   k τ   H 3 O ½  þ ð Þ  :  ð 1 Þ Eq. (1) relates the unknown number density of a neutral molecule,[ R ], to the measured intensity of the parent ion in its protonated formvia the rate coef  󿬁 cient ( k ), and the average time ( τ  ) spent by the ionsin the reaction region. Eq. (1) is useful to estimate gas concentrations,but there are several caveats [25], e.g., the necessity to consider the ef-fect of humidity and water clusters, the different mobilities of differentions, the discrimination factors of mass analyzers and the necessity toknowratecoef  󿬁 cients.Calibrationwithknownstandardsremainsabet-ter approach, especially considering compounds with proton af  󿬁 nityclose to that of water. One of them is hydrogen sul 󿬁 de, the most abun-dantsulfurcompounddetectedfromthebiogaspilotplantdigester.Cal-ibrationdatawereperformedusingagascalibrationunit(GCU,IonimedAnalytik, Austria) including a gas dilution system, in conjunction withexternal gas cylinders with known amount of VSCs (Rivoira Spa, Italy) —  Fig. 3 and Table 1. GCU provides a constant stream of dry or humidi- 󿬁 edVOC-freeair,inwhichasteady 󿬂 owofstandardgascanbeadded.Acarrier gas stream containing known volume mixing ratios (VMRs) of the compounds of interest for our investigation purposes was thusobtained (Table 1).Air, drawn from ambient through a pump, is regulated by a mass 󿬂 ow controller that can either by-pass a humidi 󿬁 cation chamber ordirectly feed the mixing chamber, in which the desired VMR isestablished. The desired relative humidity conditions for the streamgas, from RH ~19% to RH 100%, are achieved by entering the mixtureto the dew point mirror (DPM) at a speci 󿬁 c temperature conditions.InordertoprotectthePTR-QMSinstrumentagainstcarbonparticles,a 50  μ  m polymeric dimethylsiloxane membrane (PDMS) was inserted.  2.4. GC-MS VOCs preliminary identi  󿬁 cation Anagilentgaschromatograph(GC)coupledwithaClarus500PerkinElmer mass spectrometer (MS) was employed for preliminary com-pound identi 󿬁 cation. The column used was an HP INNOWAX 19091N-213V, with a length of 30 m, an internal diameter of 0.32 mm and athickness of the stationary  󿬁 lm phase (PEG) of 0.50  μ  m; wherebythe mobile phase is helium with a  󿬂 ow of 2 mL/min. The temperatureofthetransferlinewas220 °Cand150 °Cfortheionsource.Thebiogassample was  󿬁 rst pre-concentrated in a triphasic solid phase micro-extraction  󿬁 ber (SPME technique) and then eluted into the GC-MS in-strument. The selected  󿬁 bers were made from polydimethylsiloxanePDMS,divinylbenzeneDVB andcarboxenCAR;thesamplewasextract-ed at ambient temperature for 20 min, then the 󿬁 ber was inserted intothe instrument and desorbed for 4 min at 40 °C. The GC temperaturepro 󿬁 le used started from 40 °C for 4 min, then a 5 °C/min ramp wasemployed up to 250 °C where the temperature was kept for 5 min.  2.5. SOFC single cells Simulatedgasmixtureisconsideredfortheexperimentswithsulfurtrace contaminants.Experiments were performed withNi-based anodesupported solid oxide fuel cells (ASC). Planar circular type seal-lessanode supported cells, with a diameter of 80 mm and a screen printedcathode of 78 mm, were used:ASC4 (H.C. Starck, Germany) cell consists of a 465 – 555  μ  m porousNiO/YSZ anode support with a 5 – 10  μ  m NiO/YSZ porous active layer, a4 – 6  μ  m dense electrolyte YSZ and a 2 – 4  μ  m YDC barrier layer plus a30 – 60  μ  m porous LSCF cathode.Planar SOFCs (47 cm 2 surface area) were used for the experimentaltest session, fed with synthetic gases CH 4 , CO 2 , CO, N 2  and H 2  cylinders(Siad, Italy). A variable concentration of trace compounds was addedtothefuelstreambydilutingandthusmixingthepuregasmixturestothegas cylinders containing the trace compound of interest. Table 2 sum-marizes the test conditions adopted.The oxidant  󿬂 ow (air) at the cathode side was 0.5 Nl min − 1 duringthe start-up and shut-down procedures, otherwise it was  󿬁 xed at1.2 Nl min − 1 for all the performed experiments.  Table 1 Volatile sulfur compounds (VSCs) considered for the PTR-MS calibration.Compound ppb(v)rangeProtonated molecularweight (g/mol)Relative humidityrange (%)CH 3 (CH 2 ) 3 SH 5.98 – 1,495 91.058 89.0CH 3 CH 2 CH 2 SH 6.01 – 1,502.5 77.042CS 2  5.98 – 1,495 76.952CH 3 SCH 3  5.84 – 1,460 63.027 88.98CH 3 SH 4.75 – 1,187.5 49.011H 2 S 5.51 – 1,377.5 34.995 Fig.3. Experimentalset-upforPTR-QMSVSCcalibration(DPM — dewpointmirror;MFC — mass 󿬂 owcontroller,VOC — volatileorganiccompound) — (IoniconAnalytikGes.m.b.H.,Austria).139 D. Papurello et al. / Fuel Processing Technology 130 (2015) 136  – 146   3. Experimental results  3.1. Volatile sulfur compound removal The time evolution of the concentration of VSCs was monitoredalong the initial phase of the digestion process, lasting approximately75h.Table3showstheconcentrationresultsinppb(v).Themainsulfurcompound detected is hydrogen sul 󿬁 de with a variable concentrationrangingfrom3to31 ppm(v).TheH 2 Smaximumconcentrationwasob-served at the beginning of the anaerobic digestion process, as alreadyveri 󿬁 ed in our previous study [12]. Of the overall sulfur amount, H 2 Saccounts for more than 99% on a volumetric basis. Other organic sulfurcompounds are essentially thiols (from methanethiol to butanethiol).The concentration order of the volatile sulfur compound containedin the biogas stream produced from the biogas pilot plant is: H 2 S,CH 3 (CH 2 ) 3 SH, CH 3 SH, CH 3 CH 2 CH 2 SH, CH 3 SCH 3 , CS 2 .ConcerningH 2 SdetectionusingPTR-MS,thesensitivitytowardsH 2 Sis humidity-dependent due to the low proton af  󿬁 nity of H 2 S with re-spect to hydronium ion, (cf. Feilberg et al. [26]) proper calibrationwith a known standard is recommended in the case of humid gas.Reaction (2) is only possible if energetically allowed, i.e., if the protonaf  󿬁 nity of [ R ] is much higher than the proton af  󿬁 nity of H 2 O (that is165.2 kcal/mol), otherwise the preferred direction of the reaction isnot towards ionization of the contaminant molecule.H 3 O þ þ  R → RH  þ þ H 2 O :  ð 2 Þ Considering the volatile sulfur compounds under investigation, hy-drogen sul 󿬁 de (168.5 kcal/mol) and carbon disul 󿬁 de (163.2 kcal/mol)show a proton af  󿬁 nity lower or slightly higher than that of ion source,respectively. Hence these two compounds strictly require a calibrationsession in order to obtain calibration equations to convert raw sig-nals (count per seconds) into reliable quantitative measurements(ppm(v)).InacalibrationapproachforH 2 Swithstandardgas,asam-ple with varying humidity was approximated, based on the ratio of protonated water clusters (H 2 O) 2 H 3 O + /H 3 O and linearly correlatedwith the percent of reaction rate, de 󿬁 ned as ratio of the concentra-tion of calibration gas/concentration calculated from PTR-MS signal[with % of k-rate at relative humidity 100%]. Water humidity of 89%was  󿬁 xed at the GCU instrument in order to achieve real biogas con-ditions as observed during the digestion process. Fig. 4 shows thecalibration equationsobtainedfromtheGCU.All compoundsconsid-ered have a coef  󿬁 cient of determination above 99%, exceptbutanethiol (R  2 = 90%).Fig. 5 depicts how the gas cleaning section is able to withstand at avariable VSC concentration in the gas feed. Being the biogas pilot planta complex system with a variable production of VSCs along the testtime, the best parameter to describe the  󿬁 ltration process is the ratio(C/C 0 ) between the concentration in the  󿬁 ltered outlet stream (C) andthe concentration of the compound of interest before the puri 󿬁 cationstage (C 0 ). The main sulfur compound contained in the biogas, asshown previously (Table 3), is hydrogen sul 󿬁 de with a concentrationranging from 3 to 32 ppm(v). The mass 34.995 identi 󿬁 es H 2 SH + andafter 75 h of biogas  󿬁 ltration the concentration ratio is close to 1%. Inthe same way concentration outlet ratios for the remaining volatilesulfur compounds contained in the biogas are given. The concentrationofbutanethiolcontainedinthebiogasrangesfrom 45to65 ppb(v)andthe C/C 0  ratio after 75 h of test is close to 5%. Methanethiol concentra-tionrangesfrom27to45 ppb(v)andtheC/C 0 percentagevalueachieveafter 75 h of test is 16%. The remaining sulfur compounds showconcentration values below 30 ppb(v) and the  󿬁 ltering ratio standsfrom 2.5 to 20% in case of dimethylsul 󿬁 de.  Table 3 Volatile sulfur compounds (VSCs) monitored from the digester plant.VSCs digester concentration C (ppb(v))Elapsedtime (h)H 2 S CH 3 SH CH 3 SCH 3  CS 2  CH 3 CH 2 CH 2 SH CH 3 (CH 2 ) 3 SH1 31,451.7 44.8 23.1 4.8 25.3 65.32 26,098.9 40.3 21.7 0.7 24.9 55.23 26,415.2 38.3 22.0 3.6 24.5 56.34 13,415.2 32.3 18.9 18.6 23.4 47.55 10,256.9 30.4 17.7 23.1 23.1 42.819 14,352.4 36.9 24.1 6.9 25.1 65.221 22,267.8 38.8 19.9 8.5 25.4 61.923 19,062.6 36.3 22.1 7.2 24.9 58.825 25,383.9 40.1 22.4 3.3 24.7 59.427 22,622.8 34.8 20.5 7.8 24.2 53.229 10,860.4 34.4 18.1 5.6 24 51.843 16,804.5 36.5 25.9 15.6 25.7 75.645 20,195.5 37.7 21.1 9.6 25.0 64.147 14,097.8 33.3 19.9 7.2 25.2 58.449 23,940.1 37.9 22.6 3.6 25.7 60.651 23,631.7 36.8 21.4 15 24.6 65.667 13,225.8 31.6 22.6 12 25.1 65.469 13,607.8 34.2 22.7 5.4 24.8 68.071 17,132.5 32.3 22.5 21.6 24.3 55.473 18,967.1 30.8 20.7 13.6 24.1 54.375 3,156.2 27.7 16.5 25.1 23.4 45.8Max.concentrationover total (%)99.38 0.14 0.08 0.08 0.08 0.24  Table 2 SOFC test conditions.Conc. range(ppm(v))CelladoptedH 2 (ml min − 1 )CO(ml min − 1 )CO 2 (ml min − 1 )CH 4 (ml min − 1 )N 2 (ml min − 1 )H 2 O(ml min − 1 )H 2 O(g h − 1 )T(°C)i(A cm − 2 )FU I(A)0.84 – 6.4 ASC700 151.5 136.4 68.2 15.2 386.4 60.2 2.9 750 0.32 30.0 150.8 – 6.7 ASC4 250 0 41.7 62.5 0 124.5 6 750 0.32 20.9 15140  D. Papurello et al. / Fuel Processing Technology 130 (2015) 136  – 146 
Search
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