An Analysis of the Energy Potential of Anaerobic Digestionof Agricultural By-Products and Organic WasteS. Menardo & P. BalsariPublished online: 22 March 2012#Springer Science+Business Media, LLC 2012Abstract Anaerobic digestion is a promising option for recy-cling agricultural by-products and some organic wastes. Whileboth agricultural by-products and wastes have no direct com-mercial value, their management is both complicated and cost-ly. One option to simplify by-product management and reducethe costs associated with biogas plant feedstock is to substitutededicated crops with vegetal by-products. Given that the chem-ical composition of some of these by-products can differ con-siderablyfrommoretypical biogasplant feedstock(suchasmaize silage), more complete knowledge of these alternativestoproduceenvironmentallyfriendlyenergyiswarranted.Tothisend, batchtrialsundermesophilicconditionswerecon-ducted to evaluate the potential biogas yield of many agricul-turalby-products:maizestalks,ricechaff,wheatstraw,kiwifruit, onions, andtwoexpiredorganicwasteproducts(dairyand dry bread) from the retail mass-market. Among the con-sidered biomasses, the highest methane producer was the ex-pireddairyproductmixture, whichyielded554lNCH4kg1volatile solids (VS). Maize stalks and wheat strawproduced thelowest yields of 214 and 285 lNCH4 kg1VS, respectively. Anassessment of the biogas and methane yields of each biomasswas also undertaken to account for the specific chemical com-position of each biomass as it can affect the anaerobic digestionoperating system. Finally, the total Italian green energy produc-tion that might be derived from feeding all these biomasses to abiogas digester was estimated, in order to understand its poten-tial impact.KeywordsAnaerobic digestion.Agricultural by-products.Organic waste.Biogas.Energy potentialIntroductionBiogas plants acrossEuropearemainlyfedwithanimaleffluentsanddedicatedenergycrops[1]. However, asthenumber of biogas plants has increased over the past decade,the quantity of energy crops dedicated to energy productionrather than animal feed has also increased. This shift to energycropsmaycauseseriousproblemsforboththeagriculturalsector and the world economy [2] by incentivizing producersin a way that upsets the delicate balance the world maintains tofeed all peoples. Consequently, research on new substrates foranaerobic digestion plants (ADPs) is necessary.Currently, largequantitiesof agro-industrial by-productswith no commercial value are destined for landfills, or, in thecase of cereal straws and maize stalks, abandoned in fields afterharvest. Energyproductionfromagricultural by-productsisgenerallycombustion-basedastheirlowwatercontent(1214%) makes them well suited for burning [3]. The combustionplants usually produce only thermal energy, are suited for usewithin only a few kilometer radius, and are not economicallyincentivizedbypolitics. Onthecontrary, electrical energyproducers receive economic incentives, and, as a consequence,thermal energy is often transformed to electrical. However, thisis a complicated process that makes combustion-based electricenergy production very inefficient [4]. In fact, to obtain elec-trical energyinacombustion-basedplant fromby-productsrequiresthattheyburninordertoheatwateruntil itformssteam so that it drives a turbine, which in turn connects to analternator that collects and stores the electric energy for lateruse. Not surprising, given this dynamic, combustion plants thatburn vegetal by-products in Italy are few and at distances farS. Menardo (*):P. BalsariDepartment of Agriculture, Forestry, Environmental Engineeringand Land Based Economics (DEIAFA) Mechanics Section,Turin University,via Leonardo da Vinci 44,10095 Grugliasco, TO, Italye-mail: simona.menardo@unito.itBioenerg. Res. (2012) 5:759767DOI 10.1007/s12155-012-9188-0from by-productsources,which further underminestheeco-nomicsoftheprocess. Theelectricalenergyefficiencyofamethane-poweredengine,normallyusedinbiogasplants,ishigher (el03540%) than that of an electrical-powered engine[5].Moreover, thevalorizationof theproducedbiogas isenergy-efficient and environmentally friendly because of thelow emission of hazardous pollutants. The emissions of vol-atileorganiccompoundsareverylimitedsince99%ofthevolatile compounds are completely oxidized during combus-tion in the co-generator. This is in contrast to incinerators thatsuffer fromthe emission of hazardous compounds like dioxinsand hence, require extensive flue gas purification [6]. Further-more, the digestate, main biogas plant by-product, is nitrogen-rich and can be utilized in agriculture as a nutrient fertilizer ororganic amendment [7].A major improvement to the economics of energy produc-tion from these biomasses in ADPs occurs if they are substitut-ed for energy crops [8]. The use of agricultural by-products toproduceelectrical energythroughanaerobicdigestion, en-hanced by simple biomass pre-treatment to improve bacterialdigestibility and optimize methane yields (such as mechanicalshredding), couldbeaneconomicallyandenvironmentallyattractive alternative [9]. Estimates suggest that agro-industrialby-productsandanimal slurriescouldcontributemorethan20% of the 2020 Italian renewable energy goal established bytheEuropeanSettlement [10] whichequatesto17%ofthecountrystotalenergyproduction. Dhleretal. [11]demon-strated another important environmental benefit when energycrops are replaced with by-products as the main feedstock of abiogas plantgreen house gas (GHG) emissions are reducedsignificantly (40%). This finding highlighted the positive influ-ence that by-product utilization also has on GHG atmosphericemissions.In principle, many different agricultural by-products can beused in biogas plants; however, knowledge of the anaerobicdegradationofthesubstrates, plustheirbiogas(mixtureofCH4, CO2, H2, O2, and H2S) and methane production poten-tials is still limited. Furthermore, the potential negative effectsof utilizing little known by-products in biogas plants are notfully understood. Clearly, a detailed evaluation of the techni-cal and economic merits of each is needed before their use canbe optimized.This study, performedas part of the EU-Agrobiogasproject (www.eu-agrobiogas.net), investigatedtheissuesoutlined above by analyzing the biogas and methane yieldsof fiveagricultural by-products inbatchtrialsunder meso-philic conditions. The products includedkiwi fruit andonions that did not meet market standards, maize stalks, ricechaff, andwheat straw, aswell astwosourcesoforganicwaste that originatedfrommass-market retail channels:expired bread and a mixture of expired dairy products (milk,cheese, yogurt).Thestudyshowedthat thechemical compositions ofseveral of the by-products considered influenced the results.Several haverather lowpHvalues (kiwi, onions, dairyproducts), and some contain high protein levels (dairy prod-ucts),both ofwhich cancause,inthecaseof overloading,reducedmethaneproductionsecondarytodigesteracidifi-cation[12]. Moreover, the high lignin content of some by-products, such as straw, stalks, and rice chaff can negativelyaffect anaerobic digestion [13], as its presence reduces hemi-cellulose andcellulosedegradability[14].ToavoidlowpHand high lignin content problems in ADPs, the biodegradabil-ityoftheseby-productswasinvestigatedinthelaboratory.TheprincipalparametersthatrequiredcontrolbecausetheyindicatepossibleinhibitionphenomenawerepHlevel andammonia concentration.Finally, since methane concentration in biogas is a usefulparametertounderstandthetrendoftheprocess,theycanassistintheevaluationoftheeconomicsofabiogasplantoverall. Theresults obtainedinthis studywereusedtoestimatethetotalItaliangreenenergyproductionpotentialthat feedingall thesebiomassestobiogasdigestersmighthave.Materials and MethodsEach agricultural by-product (Table 1) sample was collectedfromfarmsandareasthroughout thePiedmont Regionofnorthwest Italy as described specifically below. Maize stalkswere collected from a farm in Oleggio (Novara) at the startofOctober, 1weekaftermaizeharvest. Asampleofricechaff was also taken in October, but from a biogas plant inPiverone(Torino)whereitwasusedasdigesterfeed.Thissamefarmwasalsothesourceofthewheatstrawsample,but it was collected 4 days after the July harvest. In Novem-ber, following the main kiwi harvest at a farmin Cavaller-maggiore(Cuneo),thefruitthat remainedon thetreesthatfailed to meet market standards was collected. Onions, alsotoosmallformarket,werecollectedfroma farminAnde-zeno (Torino) in July. The expired wheat bread sample camefrom a Turin supermarket located where supermarket oper-atorsassemblealltheirorganic wastefordisposal.Finally,the dairy product mixture was sampled from a firm situatedin Alessandria where a machine built by Nordischer Maschi-nenbau (Rud.Baader GmbH of Lbeck, Germany) was usedtoseparatetheinorganicpackagingfromtheorganic dairyproductfraction. Thedairyproductsamplewascomposedof about 50% milk, 30% yogurt, and 20% fresh mozzarellacheese.The study inoculum was collected from a 500 kWel bio-gas plant fittedwithaprimaryandsecondaryanaerobicfermenter for a combined total volume of 5,000 m3. It wassituated in Bra (Cuneo) and was fed with 70% swine slurry,760 Bioenerg. Res. (2012) 5:75976718% maize silage, 5% whey, 4% grass silage, and 3% cattlemanure. The anaerobic fermenters worked at 40.0Cand had atotal hydraulic retention time of 40 days. Each day of opera-tion, 130 m3of digestate was produced. Following collection,theinoculumwas storedunder anaerobicconditions for20daysat 40.0Cinathermostatedroomtodepleteitsmethane production potential.All biomass samples werecollectedandpreparedforanalysis accordingtotheGermanguidelines for abatchfermentationprocess [12].Someofthebiomasses, suchasthe rice chaff and the dairy waste mixture, were suitable forbatch trials in the form in which they were sampled; the sizeof other samples required reduction to make them manage-ablefor batchtrials.Themaizestalks werecutinto smallerpieces with dimensions of 5 cm in length and 1 cm in width;wheatstrawwasreducedtoa12cmsize;kiwifruitsandonions were pieced to a 12 cm size, and the dry bread wasground in a knifed food mill.At trial start, a biomass and inoculum quantity were pre-pared according to a calculated 1:3 ratio (biomass to inocu-lum) based on their volatile solid content [12]. Mixed in a 2-liter reactor, 300ml of deionizedwater wassuccessivelyaddedtoimprovetheblendingandhomogeneityof eachblend. Glassreactorswerefinallysealedwithglass topsconnected to Tedlar gas bags (1-, 3-, and 5-liter capacities)bytygontubing(6.4mminner diameter). Reactorsweremanually stirred at least twice daily.Biogasandmethanewereproducedinthereactorsinathermostat-controlledroomunder mesophilic conditions(40.0C) according to VDI 4630 [12]. The batch trials wereconducted until the specific methane yield per day was lessthan1%of thecumulativespecificmethaneyield(about4070 days). Headspace volume accounted for in the calcu-lations of methane yields by a correction factor as reportedinVDI4630[12].Allproducedbiogaswascollectedintogas bags andregularlymeasuredusingavolumemeter(Drum-typeGas Meter, Ritter). At thebeginningof theanaerobicdigestion,thebiogasvolumereadingswereper-formed about every 2 days since the biogas production washigh. In the second part of the trial (after 10 days from thestart of anaerobicdigestion), whenthebiogasproductiondecreased,thereadingswereperformedlessfrequently.Atthesametime,themethaneconcentrationofeachbagwasdetermined by using a Drger XAM 7000 gas analyzer. Thedailybiogasdatavolumeswereconvertedtonormalliters(lN) (drygas, temperature00Candpressure01,013hPa)according to VDI 4630 [12].Acontrol sample, composedonlyof inoculumand300 ml of deionized water, was also prepared and analyzed.The net biogas and methane yields of the tested biomasseswere obtainedbysubtractingtheir biogas andmethanevolumesfromthat producedbythecontrol. Althoughthecontrol values were very low, it was necessary to eliminate Table1ChemicalcompositionofthebiomassesanalysedinbatchtrialsBiomassDM(%FM)VS(%TS)Ntot(%TS)Ctot(%TS)C/NFAT(%TS)NDFa(%TS)H-CEL(%TS)CEL(%TS)ADL(%TS)pHNH4+-N(%FM)Inoculum6.770.76.2n.d.n.d.n.d.n.d.n.d.n.d.n.d.7.90.17Maizestalks87.789.01.242.5351.176.326.744.05.67.3n.d.Ricechaff87.187.72.647.81819.247.832.29.56.16.7n.d.Wheatstraw86.695.70.545.8921.788.929.749.89.46.9n.d.Kiwifruit12.089.82.251.8235.839.511.015.712.83.4n.d.Onion9.193.61.736.6211.310.13.36.50.33.2n.d.Dairyproducts13.493.85.450.0918.2n.d.n.d.n.d.n.d.5.8n.d.Drybread87.596.42.243.9200.86.04.710.70.65.5n.d.aNDFisthesumofH-CEL,CEL,andADLobtainedbyTrioloetal.[14]Bioenerg. Res. (2012) 5:759767 761everypossibleeffect of theinoculumonthebiogas andmethaneyieldsofthebiomasses.Allbiomasses,includingthe control, were analyzed in triplicate.Thebiomass samples werecharacterizedaccordingtotheir moreimportant physical andchemical parameters.Thedrymatter (DM) content of eachmixturewasdeter-mined after drying in an oven at 105C for 24 h. The weightof theashesof eachbiomasssamplewascalculatedafterburning the DM in a muffle furnace at 550C for 5 h, whichallowed determination of the volatile solids (VS) calculatedas the difference between DM and ashes.According to accepted analytical methods [15], the pH inwater of each biomass was determined by a by a portable pHmeter (Hanna Instruments HI 9026) using a glass electrodecombined with a thermal automatic compensation system. ACHNanalyzer was usedtodeterminethetotal nitrogen(Ntot) andtotal carbon(Ctot) for theelemental analysis.ThepHinwater was determinedbypH-meter HI 9026(Hanna, Italia). Hemicellulose(H-CEL), cellulose(CEL),andlignin(ADL) wereanalyzedbyVanSoest methods[16]. Hemicellulosesandcelluloseswerecalculatedasthedifference between neutron detergent fiber (NDF) and aciddetergentfiber(ADF), andADFandADL(aciddetergentlignin), respectively. Digestates at the end of the batch trialswere analyzed for DM and VS content to determine the VSdegradation during anaerobic digestion. The pH, ammoniumnitrogen(NH4+-N) [17], andNtotwerealsomeasuredtobetter understand the process trend.Differences in the specific biogas and methane yields weretested with a pair-wise comparison of regression parametersby the TukeyHSD-test. The level of significance was set to0.05 [1].Results and DiscussionBiomasses Chemical AnalysisTable 1 reports the inoculum and biomass chemical analysisresults. The following were batch trial inoculum key values6.7%DM,70.7%VScontent,and7.9pH.TheDMofthebiomassesvariedhighlyandwasmeasuredaslowas9.1%(onions) and as high as 87.7% (maize stalks). Rice chaff andwheat straw also displayed high DM contents at 87.1% and86.6%, respectively. As expected according to the DM con-tent, theVScontentswerealsoverydifferent andrangedbetween 8.5% on FM (onion) and 84.6% on FM (dry bread).Ideally, biomassessuitablefor anaerobicdigestionshouldhave C/Nratios between 20 and 30 [18]. Calculated C/Nratiosinpresent studyrangedbetween9and92. Inthisregard,maize stalks and wheat straw, which carried high C/N valuesat 35and92, respectively, werenot veryfit for anaerobicdigestion.DairyproductsshowedverylowC/Nratios(11)duetotheirhighNtotvalue(5.4%), whichcanalsoinhibitanaerobic digestion. The remaining biomasses had C/N ratiosin the anaerobic suitability range.The NDF parameter represents the percentage sum of H-CEL, CEL, and ADL in a compound and is an indicator of itsdigestibility. Results showed onions and dry bread had verylow values of NDF at 10.1% and 6.0%, respectively. On theotherhand,alltheotherconsideredbiomassesweregreaterthan39.5%. Thebiomasswiththehighest NDFvaluewaswheat straw at 88.9%. Among the various NDF components,H-CEL, CEL, andADLpercentagesacrossthebiomasseswere non-homogeneous. Specifically,theH-CEL valuewasbetween 3.3% (onion) and 32.2% (rice chaff), the CEL valuebetween 0.7% (dry bread) and 49.8% (wheat straw), and theADLvaluerangedbetween0.6%(drybread) and12.8%(kiwi). Kiwi had the lowest NDF value, as well as high lignincontent, and 14% volatile solids content.BiomasspHvaluesvariedwidely; kiwi (3.2)andonion(3.4) were highly acidic. Prior to the start of anaerobic diges-tion trials, the pH of each inoculumbiomass blend was de-terminedtoidentifyeventual negativeeffectsbythesetwomore acidic biomasses. The other biomass had pH values thatranged from7.7 to 7.9, which are optimal values for AD[1] asdescribed previously.Specific Biogas and Methane YieldsTheorganicwasteanaerobicbatchtrialsdigestedquicklyduetotheir high, easy-to-digest organiccompoundcon-tent, andlower, not easy-to-digest ligno-cellulosiccom-poundcontent. Thiswasevident inthenumber of daysfor which organic waste (onions and dairy products)measurementsweretaken(4555days)comparedwiththelength of days that agricultural by-product (maize stalks, ricechaff, wheat straw, and kiwi) measurements were performed(5560 days).Specific Biogas and Methane Yields of AgriculturalBy-ProductsAgricultural by-product yields rangedbetween420and809 lNkg1VSfor biogas and between 214 and 381lNkg1VS for methane (Fig. 1). Maize stalks produced thelowestmethaneyieldsofallagriculturalby-products.Ricechaff, wheat straw, kiwi, andonionshowednosignificantdifferencesamongtheir methaneyieldsbasedonTukeystest (05%). Similar values have beenreportedintheliteratureforonionsandwheatstraw[6,19,20].Thespe-cific methane yields comparedwiththespecific biogasyieldsofthebiomasseswereinfluencedbytheirgascom-positions and methane content, in particular. Kiwi and onion762 Bioenerg. Res. (2012) 5:759767biogases contained 46% and 47% methane, respectively; ricechaff contained 55% methane. With their low ligno-cellulosiccontent compounds (NDF) of 10.1% and ADL of 0.3%, theonions produced the highest biogas-specific yield. Kiwi alsopresentedahigh biogasyield(800 lNkg1VS),even with alarge ADL percentage (nearly 13.0%). While these two bio-masses had acidic pHvalues, they did not appear to inhibit themethanogenic process. When employing acidic biomasses inADPs, it is mandatory to pay attention to the volume used toavoid digestate over-acidification. All biomass pHlevels weredetermined at the end of the experiment and highlighted nosignificant problems.Figure 2 displays the cumulative methane produced anddailyspecificmethaneproducedbytheagricultural by-products. Several biomasses stood out in particular. The ricechaff methane yield at 381 lN kg1VS did not vary signifi-cantlyfromotherbiomassyields,butitschemicalcompo-sitiondid, as it includedanorganic fractionwiththefollowing principal componentsNtot of 2.6%, fats of 19.3%,H-CELof32.2%,andanADLofonly6.1%.Theearlinesswithwhichricechaffcommencedmethaneproductioncom-pared with the other biomasses indicated that the ideal initialconditions for hydrolytic bacteria activity were optimized. Theverysmallsizeofthericechaffitselfalsopromotedspeedydegradation.Onionsandkiwi displayedcomparabletrendsintheircumulative methane production curves. Both curves showeda peak at about the fifth day, whereas, after the 25th day, themethanedailyproductiondroppedquicklyandapproachedzero. Their chemical compositions, C/Nratios between 20 and25, and low ligno-cellulosic compound content all promote afast start to methane production [21].The daily methane yield curves of wheat straw and maizestalksdifferedconsiderablyfromtheothersamples.Eithermethane yields obtainedinthis studyagreedtoresultsshowninliterature [6]. Theyfailedtoreallypeak, butinstead displayed a gradual increase in the methane produc-tion, whichextendedandthenslowedafterday35. Thesetwobiomasseshaveanorganicmattercompositionthatisrichinfibers andpoorincompoundsrapidlydigestiblebybacteria, which limited methane production. In general, theratio of fibre and lignin determined the digestibility of everybiomass andits requireddegradationtime [22]. Conse-quently, thosebiomasseswithhighcontent of thesetwocompounds lacked initial degradation peaks; instead, its meth-ane production increased slowly, was extended for a period,and then stabilized.On the contrary, kiwi and rice chaff, which also containedhigh ADL values (12.8% and 6.10%, respectively), but a lowNDF, showedhighermethaneyieldscomparedwithmaizestalksandwheat strawsamples. Thechemistryof ligno-cellulosic compounds varied considerably from one biomassto another [22, 23], which might explain why it is currentlyassumed that lignin concentration does not necessarily predictthe degree to which lignin inhibits cellulose bioavailability. Ifacompoundislignin-incrusted, thencellulaseswill limitcellulose fiber access. If the cellulose is mainly in a crystallineform, then cellulases can attach to the compound and hydro-lyze it relatively quickly [24]. The chemical composition of acompound, thefibrousfractioninparticular, isessential toestimatethebiogaspotential of abiomassasmanyotherstudies have reported [25, 26].Biogas and Methane Yields of Organic WastesAs expected, dairyproducts producedhighbiogas andmethaneyieldsduetotheir highprotein(33.9%) andfat(18.2%) contentscoupledwithfiber amountssolowthattheywereundeterminable.Alsotrueofthedairy products,themethanepercentageatopthebiogas(56%)washigherrelative to the other biomasses. During the first phase of theanaerobicprocess, dairyproduct methaneproductionwasvery slow and irregular, which probably relates to their largenutrient and protein concentrations available to the bacteria.A rapid hydrolytic phase might have then caused a tempo-rarypHdecreaseandvolatilefattyacidincrease, which506070Methane [%]80420 (d)674 (bcd)570 (d)800 (ab)809 (ab)966 (a)650 (cd)02004006008001.000MaizestalksRicechaffWheat strawKiwiOnionDairyproductsDrybreadBiogas specific yield [lNkg-1VS]49% 49%46% 47%56%47%55%Methane percentage in biogas570 (d)49% 49%46% 47%56%47%55%214 (d)381 (bc)285 (bc)371 (bc) 378 (bc)554 (a)313 (cd)02004006008001000MaizestalksRicechaffWheat strawKiwiOnionDairyproductsDrybreadMethane specific yield [lcc-1NkgSV]Means followed by the same letter are not statistically different using Tukeys test at 5% level.Fig. 1 Biomass specific biogas and methane yields (biogas yield on left, methane yield on right). The bars indicate the standard deviationBioenerg. Res. (2012) 5:759767 763causedaninhibitedandslackenedmethanogenic phase[2729]. Innormal conditions, after this first phase, thepHincreases againandcomes backtoinitial values orslightly higher [30]. Analysis of pH levels (7.7) and NH4+-N concentrations (0.23%) at the end of anaerobic digestion(Table 2) made clear that these inhibitory conditions were, infact, limitedintimeanddidnot compromisetheentireprocess. TheNH4+-Nconcentrationsincreases duringtheanaerobic process in batch, as reported in [30], and in somecases, can reach toxic values for the methanogenic bacteria,inhibitingthemethaneproduction. Thisvalueisevaluatedclose to 3.0 g/l.The dry bread specific biogas resulted in 650 lNkg1VS,which is very close to another dry bread biogas yield duringFig. 2 Specific cumulative methane yield and daily methane yield764 Bioenerg. Res. (2012) 5:759767anaerobicdigestioninbatchreportedintheliteratureofabout700750lNkg1VS[31].Thespecificmethanepro-duction was 313 lN kg1VS, and the methane percentage was47%. Thebiogasyieldof drybreadwasnot significantlydifferent thanthoseof maizestalks, ricechaff, andwheatstraw(00.05),butthedrybreadmethanepercentagewasunexpected. Other studies have indicated that the amount andquality of the starch contained in dry bread can affect specificmethane yield and that starch quality can influence its digest-ibility in anaerobic conditions [32].Organic Matter Degradation and Digestate ParameterEvaluationAt theendof batchtrials, thedigestedbiomasses wereanalyzed, and the results are shown in Table 2. The digestateDMpercentagesrangedbetween4.4%and6.4%, respec-tively, for wheat straw and dry bread. The biomass organiccomponents represented, onaverage, about 67%of drymatter content. Comparing the VS content in digestate andtheVScontent at thebeginningof thebatchtrials, theorganicmatter degradationwas estimated. Thesevalueswere verydifferent amongthe biomasses analyzedandindicatedthat between24%and85%of organicmatterwas digested during the batch anaerobic digestion. The largerangeof organicmatter degradationisduetothevariousbiomassorganiccompositions,andinparticular,thefibersand lignincontent(NDF),which affected little thedegrad-ability of organic substances. Figure 3 displays the correla-tion between the organic matter degradation of the analyzedbiomasses and their NDF percentages.PHvalueswere higherthan7.7 for each digestate;con-sequently, it was possible to exclude the problem of acidifi-cation, even for acid biomasses such as kiwi fruit and onions[1]. AlthoughthepHofbothbiomasseswaslow, respec-tively 3.4 and 3.2, the low amount of biomass used in batchtrial did not significantly affect the pH of blend inoculumbiomassinthebatchdigester. TheNH4+-Ndeterminationwas alsoperformedtocontrol theanaerobicprocess andhighlight inhibitory conditions for bacteria activity. AccordingtothebiomassdigestatepHvalues,itisreportedthatthesenegative conditions occur when the concentration of NH4+-Nis greater than 3.0 g/l [32]. The NH4+-N concentration in thedigestate of each biomass was lower than the inhibitory limit;this was also true for the dairy product digestates, which had avalue of 2.3 g/l.Estimation of the Electrical Energy Potential Obtainablefrom the Analyzed Biomasses in ItalyTo extend the study results, an estimate of the actual potentialof the biomasses analyzed to produce biogas and methane iny = -0,3871x + 58,583R2 = 0,758201020304050607080900 20 40 60 80 100NDF [% on DM] Organic matter degradation [%]Fig. 3 Correlation between organic matter degradation during anaerobicdigestionbatchtrial andNDF. Thevaluefordairyproductswasnotshown, since NDF was not determined for this biomass0100200300400500600Maize stalksRice chaffWheat strawKiwiOnionDairy productsDry breadSpecific yield (m3N *t -1fm)Biogas MethaneFig. 4 SpecificbiogasandmethaneyieldspertonoffreshbiomassmatterTable 2 Analysis of digestateat the end of anaerobicdigestion trialsBiomass DM (% FM) VS (% FM) pH NH4+-N (% FM) Ntot (% FM)Maize stalks 4.7 3.2 7.9 0.14 0.20Rice chaff 4.5 3.0 8.2 0.20 0.25Wheat straw 4.4 3.0 8.2 0.16 0.19Kiwi fruit 5.1 3.4 7.8 0.19 0.27Onion 6.2 4.1 7.8 0.17 0.25Dairy products 5.2 3.7 7.7 0.23 0.28Dry bread 6.4 4.2 8.4 0.17 0.20Bioenerg. Res. (2012) 5:759767 765Italy in a meaningful way was prepared. Therefore, calcula-tion would need to be based on actual volumes and quantitiesavailable within the country. It is also chosen to express thespecific production of the various biomasses in cubic metersper ton of fresh biomass (mN3t1fm) rather than in liters perkilogramof volatile solids (lNkg1VS), which made the watercontent of the fresh biomasses play an important role in theseestimates.The dry bread showed high biogas and methane productiondue to its very low (12.5%) water content (Fig. 4). Rice chaffand wheat straw produced 291 mN3and 236 mN3of methane(CH4) per tonof freshmatter. Inthesecases, their yieldsappearedhigher becauseof their largetotal solidscontent(87.1% and 86.6%, respectively). Maize stalks also had lowwater content (12.3%) and a very low specific yield, but theystill producedmorethan160mN3t1fm. Onthecontrary,onions and kiwi produced the lowest yields per ton of freshmatter added to the digester due to their low total and volatilesolids content. The dairy product mixture also produced lowyields when expressed on fresh matter due to its high watercontent (85%).Based on study results and the supplies of these variousbiomasses inItaly, it was possibletoestimatetheir totalenergypotential (Table3). Forthecalculation, datacamefromseveral different sources: (1)theannual Italianagri-culturecensus [33] providedanannual volumeof maizestalks, rice chaff, wheat straw, kiwi, and onions that could bedestined to biogas plants; (2) the available quantity ofdairy products was based on data obtained fromanItalianproject onbio-fuel andbiogas[34];and(3)thedrybread amount was sourced from the annual Italian Report ofurban wastes [35].The available amount of agricultural by-products is verylarge;inparticular,maizestalksaremorethan7.7milliontons of fresh matter per year, and wheat straw is more than2.5 million tons of fresh matter per year. The yearly produc-tion of rice chaff is also large at more than 64,000 tons peryear. Althoughtheseby-products producedlowmethaneyields, their national availability is so large that their energypotential represents about 99%of the electrical energypotential of the biomasses analyzed in this study. This valuerepresents about 2.2% of the total electric energy demand inItaly [36].While currently quite restrictive, Italian legislation govern-ing the use of organic wastes in biogas plants will likely makethese substrates members of a widening circle of by-productsthat contribute income to biogas plants. Considering the Ital-ian availability of just the maize stalks, rice chaff, and wheatstrawbiomassesstudiedhere, theyalonecouldreplace, atleast in part, for the energy crops currently used as feedstockin biogas plants. Their total energy potential would approach7,500 GWel per year.Thisstudyanalyzedthebiogasproductionofindividualbiomasses only; however, further studies should be performedto evaluate the possible synergies that can be obtained fromthe blend of by-products and organic wastes to feed anaerobicdigesters.Thestudyofspecificblendsmayalsoreducethecritical factorsof different biomasses, e.g., lowpH, highprotein content, and also solve the problems associated withby-product seasonality.ConclusionsThe results of this study highlighted the energy potential of anumberofagriculturalby-productsandorganicwastesthatcouldbeusedassubstitutestoenergycropsinanADPtoproduce electric energy.Biomass chemical composition, particularly true in ligno-cellulosiccompoundscontainedinmaizestalksandwheatstraw organic matter, affected methane yields, but the sheersizeof their productionannuallymakes thempotentiallyattractive as contributors to the total Italian energy demand.Their employment as anaerobic digester feedstock should behardlyconsidered. Similarly, thehighpotential methaneyield of rice chaff, combined with its high volume produc-tion per year in Italy, make it another suitable feedstock forbiogasplants. Kiwi andonionstooshowedattractivespe-cific methane yields, but their use in an ADP should be wellmanaged and combined with other biomasses having a basicTable 3 Energy potentialof analyzed biomassesBiomass Availability of biomass inItaly (tons of fresh matter/year)Electrical energy potential(EEP) GWel/yearEEP/EEP total %Maize stalks 7,713,000 5,057 67Rice chaff 64,384 66 1Wheat straw 2,514,271 2,374 31Kiwi 17,390 3