htc review biofuels 2011

8/11/2019 HTC Review Biofuels 2011

1/36

Biomass has been assigned many roles to play in strate-gies for sustainable consumption. In addition to being afood source and renewable raw material[1] , it can be usedfor energy production[2,3], carbon sequestration [46] and, nally, as an essential element to increase soil fer-tility[7]. Estimates of how much biomass is available forthe various applications vary widely, depending on thefocus of the investigators and how issues of soil man-agement and biodiversity, among others, are addressed;for example, the energy and raw material substitutionpotential of biomass in the USA has been estimated tocomprise more than a third of the current US petroleumconsumption for power, transportation and chemicals by2030[1] , while the worldwide potential for a competinguse in sequestering photosynthetically bound carbon as

biochar to increase soil fertility has been estimated to be1 GtC yr-1 [8] approximately one eighth of the globalCO

2 emissions from fossil fuels in 2006[301] .

A major disadvantage for almost all applications is thehigh degree of heterogeneity in the form, compositionand water content of biomass. Therefore, drying and/orconversion processes are usually required to improvematerial properties for easier handling, transport andstorage of such materials. A variety of thermochemicalor biological processes can be used to convert biomass inthe absence of oxygen to products with higher degrees ofcarbon content than the original biomass. Gas or liquidproducts (biogas or alcohol) predominate in biochemicaltransformations, while solids (charcoal) are the majorcommercial products of the thermochemical conversion

Biofuels (2011) 2(1), 89124

Hydrothermal carbonization of biomass residuals: acomparative review of the chemistry, processes andapplications of wet and dry pyrolysis

Judy A Libra 1, Kyoung S Ro 2, Claudia Kammann 3, Axel Funke 4, Nicole D Berge 5, York Neubauer 6,Maria-Magdalena Titirici 7, Christoph Fhner 8, Oliver Bens 9, Jrgen Kern 10 & Karl-Heinz Emmerich 11

The carbonization of biomass residuals to char has strong potential to become an environmentally soundconversion process for the production of a wide variety of products. In addition to its traditional use for theproduction of charcoal and other energy vectors, pyrolysis can produce products for environmental, catalytic,electronic and agricultural applications. As an alternative to dry pyrolysis, the wet pyrolysis process, also knownas hydrothermal carbonization, opens up the eld of potential feedstocks for char production to a range ofnontraditional renewable and plentiful wet agricultural residues and municipal wastes. Its chemistry offershuge potential to inuence product characteristics on demand, and produce designer carbon materials. Futureuses of these hydrochars may range from innovative materials to soil amelioration, nutrient conservation viaintelligent waste stream management and the increase of carbon stock in degraded soils.

R EVIEW

1acatech-German Academy of Science & Engineering, c/o GFZ German Research Centre for Geosciences, Telegrafenberg, C4, 14773 Potsdam2USDA-ARS Coastal Plains Soil, Water & Plant Research Center, 2611 West Lucas Street, Florence, SC 295013Justus-Liebig University Gieen Department of Plant Ecology, Heinrich-Buff-Ring 26-32 (IFZ), 35392 Giessen4 Technische Universitt Berlin, Institute of Energy Technology, Chair ETA , Marchstrasse 18, 10587 Berlin5University of South Carolina, Department of Civil & Environmental Engineering, 300 Main Street, Columbia, SC 292086 Technische Universitt Berlin, Ins titute of Energy Technology, Chair EVUR, Fasanenstr. 89, 10623 Berlin7Max-Planck-Institute of Colloids & Interfaces, Am Muehlenberg 1, 14476 Golm8Helmholtz Centre for Environmental Research UFZ; Environmental & Biotechnology Centre UBZ; Permoserstr. 15, 04318 Leipzig9Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences, Telegrafenberg, Haus G, 14473 Potsdam10Leibniz-Institut fr Agrar technik Potsdam-Bornim e.V. (ATB), Max-Eyth-Allee 100, 14469 Potsdam11Hessian Agency for the Environment & Geology, Department of soil conservation & protection, Rheingaustrae 186, 65203 WiesbadenAuthor for correspondence: Tel. +49 331 288 2831; Fax: +49 331 288 1570; E-mail: [email protected]

future science group 89ISSN 1759-726910.4155/BFS.10.81 2011 Future Science Ltd


2/36

processpyrolysis . Several milliontons of charcoal are produced everyyear[9].

During pyrolysis, the organic mat-

ter in the biomass is thermochemi-cally decomposed by heating in theabsence of oxygen. If it is carried outin the presence of subcritical, liq-uid water, it is often called hydrouspyrolysis orhydrothermal carboniza-tion (HTC). Dry or wet pyrolysis isused here to carbonize the biomass,making products with higher car-bon contents. The product charac-teristics, their relative proportions inthe gas/liquid/solid phases and theprocess energy requirements depend

upon the input material and theprocess conditions. The advantageof HTC is that it can convert wetinput material into carbonaceoussolids at relatively high yields with-out the need for an energy-intensive

drying before or during the process. This opens up theeld of potential feedstocks to a variety of nontraditionalsources: wet animal manures, human waste, sewage slud-ges, municipal solid waste (MSW), as well as aquacul-ture and algal residues. These feedstocks represent large,continuously generated, renewable residual streams thatrequire some degree of management, treatment and/orprocessing to ensure protection to the environment, andare discussed in more detail in this review.

Currently, researchers in many disciplines are par-ticipating in the search to nd environmentally soundconversion processes and applications for biomass. Thishas resulted in a variety of terms to describe the solidproduct from dry or wet pyrolysis. Chemically, the solidis achar a solid decomposition product of a natural orsynthetic organic material [10,11] . Traditionally, such sol-ids are called charcoal if obtained from wood, peat, coalor some related natural organic materials. In the eldsof soil and agricultural sciences, the term biochar hasbeen propagated to mean charred organic matter, whichis applied to soil in a deliberate manner, with the intentto improve soil properties , distinguishing it from char-coal, which is usually used for cooking purposes[12] . Amore restrictive concept of biochar, requiring that thesolid full positive environmental criteria, has also beensuggested[13] . In this review, we adopt this naming con-vention, using biochar only for the product of the drypyrolysis process when it is used for soil applications,and charcoal for other purposes[302] . The wet pyrolysisprocess is referred to as HTC with the solid product con-sistently called hydrochar , regardless of its application,

in order to distinguish it from biochar produced fromdry pyrolysis. Char will be used to include solids fromboth processes.

Extensive reviews and books have been published in

recent years on charcoal[9], biochar[7,8,1315] , hydrochar[16,17] and their production processes. The renaissanceof research on conversion processes and their productshas been initiated by current strategies to reduce globalwarming using CO2-neutral energy technologies andcarbon sequestration in organic matter[18] . The growthin the number of publications on biochar has beenalmost exponential[13] , stimulated by the discovery ofits role in sustained fertility in Amazonian soils knownas Terra preta and its stability[1921] . HTC had falleninto relative obscurity after the initial discovery, and theresearch activity in the early 20th Century to under-stand natural coal formation[16] , until recent studies

on hydrochar chemistry and applications in innovativematerials[16,17, 22,23] and in soil-quality improvement[24,25] revived interest. Therefore, literature on the wetprocess and its product hydrochar is limited in compari-son to that on char from dry pyrolysis.

This review focuses on contrasting the informationavailable for the two types of char in regards to the useof biomass residues and waste materials as feedstocks,the conversion processes and chemistry involved in theirproduction, as well as current and potential applications(Figure 1) , with the intent to highlight the areas requiringmore research. The applications in focus are those thatexploit the material properties of the chars (e.g., biochar,adsorbents and catalysts), rather than those based onthe thermal properties such as carbon-neutral fuels. Theopen questions, especially on hydrochar s suitability as asoil amendment, are discussed in the following sectionsand summarized in a section focusing on research needs.

Char productionConversion processes

The production of charred matter always involves athermochemical conversion process. The decompositionof organic material under the inuence of heat in a gas-eous or liquid environment, without involvement of fur-ther reactants, is called pyrolysis from the Greek words pyr for re and lysis for dissolution. It is an essentialreaction step in any combustion or gasication process.The various pyrolysis processes differ in how fast heat istransferred to fresh feedstock particles, the maximumtemperature that is reached (Tmax), residence time ofthe input materials under these conditions, and theproduct distribution between the three phases. Theyare typically classied according to the reaction con-ditions and the product yields (mass ratio of productformed to initial feedstock based on dry weight). Theseare compared inTable 1 and discussed later.

Key terms

Soil carbon sequestration: Addition ofdegradation-resistant carbonaceoussubstrates to soil.

Biochar: Distinguished from charcoaland similar materials by the fact thatbiochar is produced with the intent tobe applied to soil as a means to improvesoil health, to lter and retain nutrientsfrom percolating soil water, and toprovide carbon storage [301] .

Pyrolysis: Thermal decomposition ofbiomass under anaerobic conditions.

Hydrothermal carbonization: Carbonization of biomass in waterunder autogenous pressure andtemperatures at the lower region ofliquefaction process [32] , also called wetpyrolysis.

Char: A solid decomposition productof a natural or synthetic organicmaterial [10,11] .

Hydrochar: Char produced fromhydrothermal carbonization.

Biofuels (2011) 2(1) future science group90

Review Libra, Ro, Kammann et al.


3/36

Dry pyrolysisThe main process for char production with signi-cant yields is the dry pyrolysis process. It has beenused by mankind for millennia to produce charcoaland tar-like substances, a lthough it can be operated toproduce multiple products (e.g., oil and gas) besideschar. The so called dry distillation of wood[10,11,26,27] also yields methanol, acetic acid, acetone and manymore base chemicals. Moderate heating rates withlong residence times (slow or intermediate pyrolysis)yield high amounts of gases and vapors (3035%)[28] and approximately 2040% as char[9] . In industrialapplications, these processes are operated in closedkilns where the non condensable gases are used to rethe reactors.

If the desired product is a liquid to be used as aprimer for fuels, fast or ash-pyrolysis is used, whichinvolves rapid heating of the feed and fast coolingof the generated vapors. The yield of liquid prod-ucts increases from a few percent to up to 75%. Fastpyrolysis processes are operated in special reactorsallowing for high heating rates and good mixingconditions[29] .

GasicationIf an oxidizing atmosphere is applied by addition of air,oxygen, steam or carbon dioxide or mixtures thereof,such that only partial combustion takes place, theprocess is called gasication[30] . The product gases,a mixture of mainly H2, CO, CO2 and CH4, can beused directly as a fuel or as a synthesis gas (syngas) indownstream catalytic conversion processes to gener-ate synthetic natural gas (SNG), methanol, Fischer-Tropsch fuels and many more products. Gasicationis generally operated in a continuous mode and maxi-mized to produce gas, with yields of approximately85% [28] ; only a small amount of char is produced.However, since large gasiers have a large throughputof biomass and are optimized for an economic opera-tion (even with a small solid yield), large amounts ofchar could be recovered at acceptable costs. It shouldbe noted that some tar is also produced in this process. As in all dry pyrolysis processes, condensation of thistar on the char should be avoided to prevent contami-nation with polycyclic aromatic hydrocarbons (PAHs).Depending on the feed, heavy metals in the char alsocould be an issue.

Feedstock priceAvailabilityTransport cost

Quality of product(s)Flexibility of product(s) application

Process waste treatmentCompetition

Production of char

Feedstock Agricultural residuesEnergy cropsWoodAlgaeManure

Sewage sludgeOther waste materials

Process conditionsMaximum temperatureHeating rateResidence timePressureSurrounding medium(gases, water, steam)Cooling rate

Post treatment (if necessary)Thermal/chemical activationMixing with compostAnaerobic digestionActivation in soilSeparationDrying

Char propertiesPorosityParticle size (+ distribution)Water-holding capacityWater repellencyIon exchange capacitySorption capacityNutrient capacityNutrient contentpHContaminants (PAH, heavy metals)Salt content

ProductsSolids (char, coke)Liquids (water-soluble/insoluble)Gases (condensable,incondensable)

Application

Plant size/throughputInvestment costScale-up possible?Automatization/labor cost

Economics

ProcessesSlow pyrolysisFast (flash) pyrolysisGasificationHydrothermal carbonizationPartial combustion

Figure 1. Factors inuencing the production and applications of char.

Hydrothermal carbonization of biomass residuals Review

future science group www.future-science.com 91


4/36

Hydrothermal processesIn hydrothermal processes, the solid material is sur-rounded by water during the reaction, which is kept ina liquid state by allowing the pressure to rise with thesteam pressure in (high)-pressure reactors. As in drypyrolysis, reaction temperature (and pressure) deter-mines the product distribution. With process tempera-tures of up to 220C and corresponding pressures upto approximately 20 bar, very little gas (15%) is gener-ated, and most organics remain as or are transformedinto solids. At higher temperatures, up to approximately400C, and with the use of catalysts, more liquid hydro-carbons are formed and more gas is produced. This socalled hydrothermal liquefaction has drawn someinterest, although most liquefaction work is performedusing organic solvents instead of water[31] .

If the temperature and pressure are increased fur-ther, the supercritical state for water is reached and theprimary product is gaseous (hydrothermal gasica-tion)[32] . Depending on the process conditions, eithermore methane or more hydrogen is generated; char isnot yielded in noticeable amounts[33] .

FeedstocksConversion to char via dry pyrolysis has been tradi-tionally restricted to biomass with low water content,such as wood and crop residues, because of the highenergy requirements associated with the inevitable dry-ing prior and/or during the reaction by the evaporation

of water. Potential feedstocks for wethydrolysis span a variety of non-traditional, continuously generatedand renewable biomass streams:wet animal manures, human waste(i.e., excrements and faecal sludges),sewage sludges, MSW, as well asaquaculture residues and algae.

These feedstocks usually require some degree ofmanagement, treatment and/or processing to ensureprotection of the environment. Many of these streams(e.g., human waste and MSW) already have substantial

collection and treatment costs associated with them. Forsome applications, it may be benecial to have a mobilecarbonization facility.

Carbonization of biomass has a number of advantageswhen compared with common biological treatment. Itgenerally takes only hours, instead of the days or monthsrequired for biological processes, permitting more com-pact reactor design. In addition, some feedstocks aretoxic and cannot be converted biochemically. The highprocess temperatures can destroy pathogens and poten-tially organic contaminants such as pharmaceuticallyactive compounds[34,35] . Furthermore, useful liquid, gas-eous and solid end-products can be produced[36] , and

at the same time contribute to GHG mitigation, odorreduction[9] and additional socio-economic benets.In addition, use as biochar may contribute to climatechange mitigation andsoil amelioration [37] .

Biomass in general is dened in this review in termsof source:the biodegradable fraction of products, wasteand residues from biological origin from agriculture(including vegetal and animal substances), forestry andrelated industries including sheries and aquaculture,as well as the biodegradable fraction of industrial andmunicipal waste [38] . Using a similar denition, the USDepartments of Energy and Agriculture estimated thetotal sustainable biomass feedstock that can be harvestedfrom US forest and agricultural land to be 1.18 billiondry tons[1] . Estimates for the more densely populatedGermany show a potential of only 0.24 billion tons[2,39] .How much biomass is actually available for variousapplications is a matter for discussion and research[39,40] .

Feedstock characteristics such as chemical compo-sition, volatile and noncombustible fraction, moisturecontent, particle size and energy content signicantlyaffect conversion efciencies and char characteristics inboth processes. Typical values for these parameters forthe feedstocks targeted in this review, as well as for othertypical biomasses (e.g., woods and grass) are presentedin Table 2 [4149,303,304] . Further research on quantifyingtheir effect on process energetics, product distributionbetween phases and product quality is needed, especiallyfor HTC.

Agricultural residual feedstocksThe current total sustainable biomass feedstock that canbe harvested from US agricultural land is approximately176 million dry tons annually[1] . It includes crop resi-dues, grains for ethanol production, corn ber, MSWand animal manures. Although animal manures makeup a large portion of this biomass (18%), they have not

Table 1. Comparison of reaction conditions and typical product yieldsfor thermochemical conversion processes with char as a product.

Process Reaction conditions(temperature [C];vapor residence time)

Product distribution (weight%)

Char Liquid GasPyrolysis: slow ~ 400; hweek 35 30 35Pyrolysis:intermediate

~ 500; ~ 1020 s 20 50 30

Pyrolysis: fast ~ 500; ~ 1 s 12 75 13Gasication ~ 800; ~ 1020 s 10 5 85HTC ~ 180250; no vapor

residence time, ~ 112 hprocessing time

5080 520%(dissolvedin processwater, TOC)

25

HTC: Hydrothermal carbonization; TOC: Total organic carbon.Adapted from [16,28] .



Key terms

Carbonization: a process by which solidresidues with increasing content of theelement carbon are formed fromorganic material usually by pyrolysisin an inert atmosphere [10] .

Soil amelioration: soil quality build upor improvement of soil fertility.


5/36

been as extensively studied as other plant-based feed-stocks. The change in animal agriculture toward concen-trated animal feeding operations (CAFOs) over the lastdecade makes the discussion of feedstock from animalmanures timely and well justied. Manure productionfrom CAFOs is often greater than local crop and proxi-mal pastureland nutrient demands. Over-application ofthe animal manure can spread pathogens, emit ammo-nia, GHGs and odorous compounds, and enrich sur-face and ground waters with nitrogen and phosphorouscompounds, leading to eutrophication[50,51].

Carbonizing surplus animal manures from CAFOsis a viable manure management alternative, which notonly provides environmentally acceptable manure treat-ment, but may also bring potential income revenue tofarmers from producing value-added biochar. The pro-cess can also use a variety of blended on-farm seasonalcrop residues with animal manures, and concentrate theanimal and plant nutrients (e.g., P, K and possibly N)into a nutrient-dense char. This is not only potentiallyuseful for nutrient recovery and soil fertilization, it mayalso offer a sound route to sustainable nutrient cycling.

Animal manure feedstock characteristics(Table 2) can vary widely. Poultry and feedlot operations collecta mixture containing manure, bedding, waste feed,and, in some instances, underlying soil. These mixtures

generally have low moisture contents (typically 90%) comprisedprimarily of discharged wash water, but also manure,urine and undigested feed.

Human waste & sewage sludgeThe amount of human waste (untreated excrement andfaecal sludges) and sewage sludges (primary and sec-ondary sludges from wastewater treatment processes) iscontinuously increasing, not only through the install-ation of new sanitation facilities in developing countries,but also through intensied wastewater treatment indeveloped countries[52] . Currently, it is estimated thatat least 10 million tons on a dry weight basis of sewagesludge per year will accrue in the EU[53,54] . In 2005,the annual US biosolids production was estimated to beapproximately 8 million tons[305] .

The direct conversion of human waste and sludgesvia pyrolysis meshes well with the holistic approachof ecological sanitation[55] . It can be used to supporta more systematic closure of material ow cycles andresource conservation, since secondary wastewater treat-ment by biological processes and the generation of stabi-lized sewage sludges is accompanied by the removal of

Table 2. Feedstock properties relevant to thermal conversion processes.

Feedstock Woods Grasses Manures Sewage sludges Municipal solid waste

Primary Activated Digested Total ## Organic

Elementalanalysis (%, daf) Carbon 5055 4651 5260

53.3#

- 54.4#

2755 4752Hydrogen 56 67 68 7.2# - 7.7# 39 0.63Oxygen 3944 4146 2636 32.0 # - 29. # 2244 4042Nitrogen 0.10.2 0.41.0 36 5.3 # - 5.6 # 0.41.8 0.160.25Sulfur 00.1


6/36

valuable organic carbon, nitrogen and energy(Table 2) .New sanitation concepts with separated waste streamsand reduced water use can produce concentrated, rela-tively homogeneous waste streams requiring only minor

pretreatment before pyrolysis. This can be combinedwith nutrient-recovery processes.The present pyrolysis-based reuse concepts for sew-

age sludge apply to the energy recovery from feed-stocks[5659] and the material reuse of the solid con-version products as technical adsorbents[6063] or soilameliorants[6466] . By contrast, no experimental workhas been published to date on the HTC of sewage sludgeor on the conversion of human waste by either wet ordry pyrolysis.

Municipal solid wasteMunicipal waste-generation rates vary with location

and have been shown to correlate with average income,ranging from 0.1 tons/person-yr (low income countries)to over 0.8 tons/person-yr (high income countries)[67] .MSW is broadly dened as wastes originating from resi-dential (i.e., product packaging, newspapers, magazines,food waste, grass clippings, yard waste and recyclables),institutional (i.e., schools and prisons), and commercialsources (i.e., restaurants). Construction and demolitiondebris and combustion ash are not generally character-ized as MSW. For the purposes of this review, industrialand municipal wastewaters are not classied as MSW.

Thermochemical processing of MSW has the poten-tial to reduce GHG emissions associated with currentwaste management techniques (i.e., landlling andcomposting), while producing value-added products,such as activated carbon. The heterogeneous nature ofMSW (in terms of composition, chemical properties,and particle size, see Table 2 ) complicates its use as a feed-stock for pyrolysis, potentially requiring the waste to beprocessed (i.e., shredded and sorted) prior to introduc-tion, in order to minimize operation and maintenanceissues[43,44,6871] .

ChemistryMany similar reaction pathways occur during both drypyrolysis and HTC of biomass. Biomacromoleculesdegrade to form liquid and gaseous (by-)products,while solidsolid interactions led to a rearrangementof the original structure[72,73] . However, the differ-ence in the reaction media plays a dening role in thechemistry and characteristics of the products justifyingseparate consideration.

ReactionsDry pyrolysis of biomass at temperatures between 200and 500C in a largely inert atmosphere leads to thethermal degradation of biomacromolecules with no

oxidation except by the oxygen contained in the biomassfeed. Despite centuries of research, reaction mechanismsare only partly understood due to the high degree offeedstock complexity and number of possible reaction

mechanisms. Bond cleavage, manifold intramolecularreactions, decarboxylation, decarbonylation, dehydra-tion, demethoxylation, condensation and aromatizationare some of the characteristic mechanisms[73] . The reac-tion temperature largely governs which reaction domi-nates. However, owing to the nonuniform temperatureproles within pyrolysis reactors, it is common for manyof the aforementioned reaction mechanisms to occur inparallel. The highest (peak) temperature reached dur-ing the process has a critical inuence on the pyrolyticreactions and the properties of the char product[72] .Decomposition of specic compounds can also becharacterized by temperature. Hemicelluloses mainly

decompose between 200 and 400C, while cellulosedecomposes at higher temperatures (300400C). Bycontrast, lignin is the most stable component, graduallydecomposing between 180 and 600C[74] .

In comparison, during HTC the biomass is heatedin subcritical water to between 150 and 250C at auto-genic pressures for time frames typically greater than1 h. Feedstock decomposition is dominated by reactionmechanisms similar to those in dry pyrolysis, whichinclude hydrolysis, dehydration, decarboxylation, aro-matization and recondensation[16,32] . However, thehydrothermal degradation of biomass is initiated byhydrolysis, which exhibits a lower activation energythan most of the pyrolytic decomposition reactions.This has been shown by calorimetric measurements[75] .Therefore, the principle biomass components are lessstable under hydrothermal conditions, which leads tolower decomposition temperatures. Hemicellulosesdecompose between 180 and 200C, most of the ligninsbetween 180 and 220C, and cellulose above approx-imately 220C[76] . Although it has been observed thatboth time and temperature inuence product charac-teristics[77] , temperature remains the decisive processparameter[16,78] . It should be noted that a manipulationof the water pH has a signicant impact on the reactionmechanism of cellulose in water[76] . Alkaline conditionsare often used for the liquefaction of biomass (i.e., a shiftto products with a high H/C ratio)[79] .

The similarity of some of the reaction pathways dur-ing dry and wet pyrolysis processes are illustrated inFigure 2 [9,16,32,80]; for example, a substantial amount ofliquid oil and water is produced from ash pyrolysis ofbiomass[81] , so that hydrolytic reactions are likely totake place during conventional pyrolysis, especially atelevated pressures[9]. In addition, (dry) pyrolytic deg-radation pathways are likely to occur, to some extent,during hydrothermal conditions[82,83] . However, owing




7/36

to the different reaction media, a shift in the reactionnetwork takes place, which leads to distinct productsin quality and quantity; this will be discussed in thefollowing sections[16] .

ProductsProducts from pyrolysis are solids, liquids and gases(Table 1) . Compared with dry pyrolysis, HTC produceshigher solid yields, more water soluble organic com-pounds and fewer gases, comprised mainly of CO2 [84] .In addition, the composition and structure of the solidproduct (hydrochar) from HTC differs substantiallyfrom dry pyrolysis chars[85] . The chemical structureof hydrochar more closely resembles natural coal thancharcoal, with respect to the type of chemical bondsand their relative quantity, as well as its elemental com-position[86,87] . Both chars exhibit lower H/C and O/Cratios than the initial product, owing to the evolution ofH2O and CO2 in the dehydration and decarboxylationreactions. However, hydrochar generally has higherH/C and O/C ratios similar to natural coal[88] than

the solids resulting from dry pyrolysis(Figure 3) [9,86,89,90, R , U D ] . This implies that the ratioof the reaction rates of decarboxylation to dehydrationis higher in HTC than in dry pyrolysis[78,91] .

It can be observed that the elemental ratios forthe hydrochar from animal-derived biomass arecomparable to those from plant material, althoughtheir feed compositions may differ substantially(seeFigure 3) . Naturally, this results in different productcharacteristics.

Although chars from both processes contain exten-sive aromatic structures, they are arranged differently.The structure of char from dry pyrolysis consists ofturbostratically arranged sheets of conjugated aromaticcarbon that grow above 400C[72] . By contrast, carbonspheres with a distinctive size distribution have beenobserved for the case of HTC of glucose[22,9295] . Thesespheres were hypothesized to exhibit an aromatic coreof cross-linked furanic rings with mainly aldehydic andcarboxy functional end groups[96] . Recent publicationshave revealed that other carbonaceous structures of

Pyrolysis

Solidliquid

Liquidliquid

SolidgaseousSolidsolid

Tar

Char

Tar

HTC coal

H2O

TOC

Coke

Condensationreactions


Conversionreactions

Coke

HTC

CO 2

Liquidgaseous Gaseousgaseous

Pyrolysis

Condensation

Hydrolysis

fragments

Hydrolysis

Product GasVapors

Biomass

Pyrolysis


Figure 2. Comparison of simplied reaction schemes of hydrothermal carbonization and dry pyrolysis withregard to the reaction phases governing the processes and typical product classes. TOC: Total organic carbon in the liquid phase of hydrothermal carbonization (organic acids, furfurals and phenolsamong others).Based on work from [9,16,32,80].




8/36

high technical importance can besystematically created using hydro-thermal conditions (see the `Otherapplications section).

The distinctive differences inthe solid product composition andstructure result from a shift in thereaction mechanisms governing the

processes. Radical mechanism pathways, which takeplace especially at low temperature dry pyrolysis[97] ,are suppressed in subcritical hot water in favor of ionicreactions[98,99] . However, the major difference is thatHTC primarily starts with the hydrolysis of biomac-romolecules, yielding oligosaccharides, hexoses, pen-toses and fragments of the lignin[76] . The resultingfragments in the aqueous phase may allow initiationof completely different chemical pathways and pos-

sible products(Figure 2) [100,101,32]

. For example, dur-ing the dry decomposition of cellulose, one principleintermediate of dry pyrolysis is proposed to be anhy-droglucose[9,97] . In HTC its formation is comparablylow [101] . Instead, 5-hydroxymethylfurfural (HMF)is proposed as a crucial intermediate[102] . HMF is aninteresting platform chemical between carbohydrateand hydrocarbon chemistry because it can be used toderive an impressive array of highly useful organicintermediate chemicals and marketable products[100] .Research interest in the production of HMF has a longhistory[103,104] .

Hydrolysis can lead to a complete disintegration ofthe physical structure, which is not the case during drypyrolysis. However, the fraction of biomass that can behydrolyzed signicantly depends on the temperature

prole and process design[83,105] . Increasing reactionseverity usually results in an increasing amount of col-loidal carbon particles, while less structural features ofthe original feedstock remain[106] .

Many (thermal and hydrolytic) decomposition frag-ments of biomass are highly reactive, especially thoseassociated with lignin[83,107] . In dry pyrolysis, theseintermediate products recombine to form a solid prod-uct, so-called coke [9,108]. Containment of the gases canbe used to increase the chance of these recondensationreactions, therefore increasing solid yield and coke con-tent [109,110] . As a consequence, the composition of theproduct changes[88] . Under hydrothermal conditions,

nearly all fragments remain in the liquid phase wherethey have low mobility; thus, a similar effect to contain-ment is achieved(Figure 2) . The formation of solids ispredominated by recondensation reactions, (i.e., hydro-char is composed of a signicant fraction of the abovementioned coke)[9496] .

Results from HTC experiments highlight theseeffects. For biomass without a structural crystalline cellu-lose scaffold, carbonaceous nanoparticles were obtained,with particle size depending mainly on the carboniza-tion time and concentration[95,106] . For biomass witha structural crystalline cellulose scaffold, an inverted

structure was found, with the carbonbeing the continuous phase, pen-etrated by a sponge-like continuoussystem of nanopores (representingthe majority of the volume). Theseproducts exhibit highly functional-ized surfaces with the potential fora variety of applications (see theOther applications section).

In conclusion, HTC differs fromdry pyrolysis in that hydrolysis isthe determining rst step. The solidproduct hydrochar is largely formedby recondensation reactions andexhibits distinct characteristics fromdry pyrolysis char. The availability ofthe fragments from hydrolysis in theliquid phase offers a huge potentialto inuence product characteristicson demand.

Energetics A crucial element in determining thefeasibility of a biomass conversionprocess is the energy balance. This is

Initial biomass

0

0 .5

1 .0

1 .5

2 .0

0 0.25 0.5 0.75 1.0

Softlignite

Bituminouscoal

A t o m

i c r a

t i o

H / C [ - ]

Atomic ratio O/C [-]

H T C

D r y p y

r o l y s i s

ManureCellulose

Wood

HydrocharHydrochar

HydrocharCharcoal

Figure 3. Development of the elemental composition (molar ratios) of cellulose, wood [86] and manure [R , U D ] . Regions for typical compositions of biomass [89] , lignite andbituminous coal [90] as well as charcoal [9] are illustrated for comparison.HTC: Hydrothermal carbonization.



Key term

Coke: a solid high in content of theelement carbon and structurally in thenongraphitic state. It is produced bypyrolysis of organic material that haspassed, at least in part, through a liquidor liquid-crystalline state duringthe carbonization process [10] .


9/36

the case especially if the char is to be used as a fuel, butis also required for determining energy requirements to judge ecological and economical feasibility when usingsuch processes for other purposes (i.e., soil amendment

and value-added products). Preliminary results havebeen published for biochar from dry pyrolysis[111,112] .However, owing to the fact that research on the techni-cal development of hydrothermal production systems isstill in its embryonic stage, a detailed energetic analysisfor this technology cannot be given here. Instead, a com-parison of the energetics of dry pyrolysis and HTC willbe discussed on the basis of reaction enthalpy, followedby a qualitative comparison of the decisive energeticdifferences in their production processes.

Reaction enthalpyBoth dry pyrolysis and HTC of biomass can be exo-

thermic reactions. The amount of heat released is depen-dent on the feedstock used and the reaction parameters,mainly temperature and residence time. A rough esti-mate of the heat of reaction for each process can bemade from the approximate stoichiometric equations(Equation 1 : dry pyrolysis[113] ; Equation 2 : HTC [85]),which were deduced from experimental results usingcellulose (C6H10O5) as a model substance:

5 C H O C H O 0.5CO 0.25CO

2.88H O C H O

6 12 3.75 2.25 0.5 2

2 1.5 2 0.38

+ +

+ +

"

Equation 1

5 C H O C H O 0.75CO 3H O 6 12 5.25 4 0.5 2 2 + +"

Equation 2

The higher heating values fromEquations 1 & 2 aresummarized inTable 3 . These approximations shouldbe treated with care, since the chemistry of neither pro-cess is fully understood.Equation 2 does not considerany liquid organic reaction by-products that representan important fraction[91,114,115] . In addition, biomass

cannot be regarded as a well-dened reactant becauseof its high degree of chemical complexity and hetero-geneity. Nevertheless, these theoretical considerationsoffer insight to what can be expected.

Experimental results from calorimetric measure-ments with cellulose support these theoretical estima-tions. Values between 0.4 and 0.7 MJ/kg Cellulose havebeen reported for the (slow) pyrolysis of cellulose[75] ,while similar measurements for HTC produced valuesof approximately 1 MJ/kg Cellulose [F , ] .Thus, the caloric nature of dry pyrolysis and HTC reac-tions is comparable. However, it is important to keep inmind that complex reaction mechanisms are involved inthe processes and highly dependent on reaction condi-tions; for example, although the overall reaction is exo-thermic, the initial phases of both dry pyrolysis and HTCare endothermic. As a consequence, a mild pyrolysis (tor-

refaction) is slightly endothermic[116,117]

, and the samecan be expected for mild HTC, owing to the endothermicnature of the hydrolysis of cellulose[118] .

Process comparisonThese theoretical energetic considerations can be usedto provide some guidance in the choice between wet anddry processes. From a thermodynamic point of view,there is a certain water content that makes the use of dryprocesses uneconomical and/or senseless. A thresholdexists where wet processes become energetically moreefcient than dry processes. This has been illustrated bya theoretical comparison of wood combustion with andwithout HTC as a pretreatment process. Pretreatmentwith HTC is more efcient for feedstocks with a watercontent of more than 50%[119] .

Comparison of the heat demand for the evapora-tion of water to the heat of cellulose pyrolysis reaction(Equation 1) shows that above a water content of 30%,more energy is required to evaporate the water than issupplied by the heat released during pyrolysis. In someprocess designs, product gas is burned to preheat thefeed[112,120] . A process using this design could convert

Table 3. Comparison of the caloric nature of slow pyrolysis and hydrothermal carbonization of cellulose.

Slow pyrolysis (Equation 1) Hydrothermal carbonization (Equation 2)

Temperature range (C) 300500 180250HHV of feed (MJ/kg) -17.6 -17.6Heat of reaction (MJ/kg cellulose) -0.8 -1.6HHV of solid product (MJ/kg cellulose) -11.3 -16.0HHV of gaseous by-products (MJ/kg cellulose) -0.4 0

HHV of liquid by-products (MJ/kg cellulose) -5.1 0 Data from [113] .Data from [85] . This equation is not complete because liquid products, which represent an important product group, have not been considered.Values are according to Equation 2. Published experimental results from carbonization of different feed range from 12 MJ/kg feed for gaseous and 0.54.1 MJ/kg feed for liquidby-products [F , U D ]. HHV: Higher heating value.




10/36

cellulose with a water content of up to 70% withoutexternal energy supply. To pyrolyze cellulose with aneven higher water content, part of the char has to beburned, reducing the yield. At a water content of approx-

imately 90%, a ll of the char would need to be burned(i.e. the break-even point for a complete combustionwithout the release of usable heat is approached). Forbiomass, these values will be lower because less energyis released during carbonization. Roet al. reported thatthe moisture content of swine solids mixed with ryegrass needs to be less than 56% in order to be carbon-ized without requiring external energy[46] .

For the case of HTC, heat for water evaporationis avoided, but the reaction water has to be heated. Additional energetic aspects related to the transporta-tion of water-rich feedstocks and post-processing of thehydrochar, such as dewatering, also have to be consid-

ered. An energetic advantage of hydrothermal processescan be expected when the conversion products can eitherbe used without predrying, or when their dewaterabil-ity is improved compared with that of the feedstock.Carbonization reactions and disruption of colloidal struc-tures have been shown to improve dewaterability[121] .

To conclude, it appears unlikely that a dry pyroly-sis process can be driven economically for the conver-sion of feed with a water content above approximately5070%. Such a process would only be capable of pro-ducing charred material and an insignicant amountof additional energy and/or product gas. The lower thewater content of the feed, the more heat can be producedand used for other purposes[112] . By contrast, HTCtypically runs at a water content of 7590% or evenhigher. The amount of external heat necessary for HTCdepends on the process design, but it is expected to besubstantially lower than for the dry pyrolysis of sucha slurry. Hydrothermal carbonization of dry organicmatter (water content


11/36

maximum attainable yields approach one another asthe residence time increases, probably a result of thecontinued reaction of intermediates to char.

However, these observations cannot be generalizedyet, owing to the high variation in reported char yieldsfrom wet pyrolysis(Tables 4 & 5) . Another inuencingfactor is the concentration of solids. In general, the morewater added to the reaction, the higher the absolute car-bon loss per unit mass of feedstock to the liquid phase.This carbon loss directly lowers the yield of hydrocharand produces a concentrated wastewater that requiresfurther treatment (typically with several g/l chemicaloxygen demand).

In dry pyrolysis, the theoretical solid yields for char-coal (e.g., 35% for cellulose fromEquation 1 ) can beincreased by condensation of the liquids (tars) to coke,achieving theoretical yields in the range of 5071%[109] ,although these are normally not achieved. Experimentalyields for hydrochar have been observed to be higherand thus closer to the theoretical yield.

For meaningful process comparisons, process con-version efciencies for the various elements of interestshould be reported and calculated from the solid yieldand elemental composition. Carbon conversion ef-ciencies are of interest, especially for assessing carbon

storage strategies. Commonly in dry pyrolysis, 50% ofthe biomass carbon is converted to solids[111,126] , whilein HTC, 6084% of the biomass carbon may remain

in the hydrochar[R , U D ]

. A further rene-ment is the xed carbon yield[9], which describes themass ratio of the xed (nonvolatile) carbon in the charto the carbon in the initial biomass.

While it is important to increase the informationmeasured and reported on chars, it must be kept inmind that any attempt to merge data generated via dif-ferent methods must consider the different, operation-ally dened analytical windows detected by each tech-nique, as well as the limitations and potential biases ofeach technique, as highlighted by Hammeset al. [127] .

Char compositionThe nature of feedstock, process temperature andreaction time are the main factors inuencing thechar composition. A comparison of the elemen-tal composition of chars from wet and dry pyrolysis(Tables 5 & 6) [9,46,68,86,128130 R , U D ] demonstrates that chars from the dry pyrolysis of tradi-tional feedstocks usually have a higher carbon and lowerhydrogen content than those from HTC. However,as the temperature in HTC is increased, the carbon

Table 4. Comparison of experimental and theoretical solid yields from hydrothermal carbonization of cellulose and wood.

Temperature (C) Time (h) H/C O/C Solid yield (%, db)

Experimental Maximum theoretical

Cellulose - - 1.67 0.83 - -275 3 0.75 0.19 47 56340 72 0.67 0.07 35 45

Wood - - 1.45 0.65 - -225 504 0.8 0.19 62 63360 4 0.63 0.08 33 54

Dimensionless atomic (or molar) ratios: H/C Hydrogen to carbon ratio, O/C Oxygen to carbon ratio.Yield estimated from stoichiometry assuming only water and carbon dioxide as by-products.db: Dry weight basis.Data from [86].

Table 5. Product composition of hydrochar from different feedstock and process conditions.

Feedstock Tmax (C) Time (h) Solid yield(%, db)

C (%, daf) H (%, daf) O (%, daf) N (%, daf) S (%, daf) Ash (%, db) Ref.

Cellulose 225 3 63 51.9 5.6 42.5 0 0 0 [86]

Cellulose 275 3 41 76.4 4.7 18.9 0 0 0 [86] Cellulose 340 3 39 81.8 4.8 13.4 0 0 0 [86]

Wood 200 72 66.2 70.8 5.7 23.4 NR NR NR [128]

Wood 250 72 50.7 77.1 5.3 17.6 NR NR NR [128]

Wood 280 72 39.9 82.7 4.8 12.5 NR NR NR [128]

Swinemanure

250 20 60 70.2 7.9 16.8 3.6 1.5 27.6 [R , UD ]

Chickenlitter

250 20 60 58 5.9 30.5 4.3 1.4 43.6 [R , UD ]

daf : Dry ash-free weight; db: Dry weight; NR: Not reported; Tmax: Maximum temperature.




12/36

content increases. Char from waste feedstocks has acomparatively lower carbon content in both processes.

It is striking that both dry pyrolysis and hydrothermalchars are signicantly homogenized compared with thewidely varying CHONS compositions of the originalfeedstocks(Tables 2, 5 & 6) . However, this is only validfor the volatile fraction of the char(Tables 5 & 6) ; the ashcontent remains highly variable.

Retention of nutrientsRetention of nutrients in chars from dry pyrolysis hasbeen found to be highly variable[131] . In general, charsfrom the wet and dry pyrolysis of waste feedstocksretain high levels of calcium, potassium and phospho-rous. This was seen in animal manures[24 ,132] and withsewage sludge[133] . However, the question arises: howavailable are these nutrients for plants?

Shinogi demonstrated that by increasing the pyrolysistemperature from 250800 C, the P content of sew-age sludge becomes concentrated within the dry char.However, the plant-available phosphorus, as character-ized by the citrate extractable fraction, is decreased bymore than 90%[134] . Acidic conditions facilitate themobilization of nutrients and their uptake by plantroots. However, during wet pyrolysis, a low pH mayalso counteract the sorption of nutrients. Hydrocharsfrom plant residues such as corn stover may resultin low pH values of under 5[135], and may affect thesorption capacity of nutrients, primarily in the caseof phosphorus. Processing the same source materialby dry pyrolysis resulted in a biochar with a pH valueof 9.9, a high cation exchange capacity (CEC) and ahigh liming value. The phosphorus fertilizing effectof dry chars may also increase owing to thermolyti-cal deliberation of organic bound phosphates from thefeedstock[136] . Chars produced from sewage sludgeat 550C in dry pyrolysis signicantly stimulated thegrowth of cherry tomato[133] . This growth stimula-tion was attributed mainly to improvements of plant

nitrogen and phosphorus nutrition. In contrast to phos-phorus, though, almost half of the nitrogen of raw sew-

age sludge was volatized at 450C[65]

. Nitrogen lossescan be limited by low dry pyrolysis temperatures[9,137] .In HTC, dissolution of water-soluble minerals

can be signicant[138] ; however, the nutrient contentwill also depend on the technique for dewatering thesolid conversion product. The ratio between evapora-tion and mechanical dewatering governs the amountof plant nutrients that will be adsorbed/retained tothe HTC chars surface. In both pyrolysis processes,nutrient concentrations in the feedstock and in theresulting solid and liquid phases need to be taken intoaccount in the process design. The nutrient content inthe solid will have to be adjusted for the application,while additional research on process combinations torecover nutrients from the liquid phase is needed.

Heavy metalsHeavy metals cannot be destroyed during pyrolysis, incontrast to organic compounds. Since they may havea toxic risk potential, the fate of heavy metals has tobe followed. Their possible accumulation in the solidphase, especially if they can affect the food chain,has to be taken into account. There are some reas-suring studies on the fate of metals in chars resultingfrom the dry pyrolysis of sewage sludge[64,34] , animalwastes[129,139] , and MSW[43,140,141] . Although metalconcentrations associated with chars from carboniza-tion of animal wastes and sewage sludge were detected,concentrations were relatively low when comparedwith chars resulting from the pyrolysis of MSW. Inash carbonization (FC) of sewage sludges, heavy met-als with low boiling points (e.g., Hg, Cd, and Se) wereeluted from the FC reactor, whereas those with highboiling points (e.g., Pb, Cd, Ni, Cu, Zn and Sr) wereincorporated in the chars[9] . Metal concentrations indry chars from biosolids were all signicantly lowerthan the limits set for the exceptional quality biosolids,

Table 6. Product composition of char from dry pyrolysis for different feedstock and process conditions.

Feedstock Tmax (C) Time (h) Solid yield(%, db)

C (%, daf) H (%, daf) O (%, daf) N (%, daf) S (%, daf) Ash (%, db) Ref.

Pine wood NR NR NR 95.2 1.1 3.1 0.1 0.04 0.69 [9]

Corncob NR NR NR 90.3 1.3 5.6 0.6 0.1 4.31 [9] Rice hulls NR NR NR 89.7 1.4 6.6 1.0 0.1 41.35 [9]

Poultry litter 450550 NR NR 4174 2.33.7 5.948.7 3.15.2 1.53.6 43.854.5 [129]

Starter turkey litter 450550 NR NR 72.1 3.6 20.2 2.4 0.5 24.6 [129]

Poultry litter 620 2 4349 86.8 2.5 1.5 5.8 3.4 53.2 [46]

Swine manure 620 2 4349 89.7 3.4 0 5.8 1.2 44.7 [46]

Wastepaper 750950 NR NR 46.7 6.2 46.9 0.1 0.1 15.4 [130]

Refuse derived fuel 400700 1 NR 76.587.3 1.36.1 1.41.7 0.30.8 2642 [68]Municipal solid waste, minus recyclables.daf : Dry ash-free weight; db: Dry weight; NR: Not reported; Tmax: Maximum temperature.




13/36

suggesting unrestricted land application of carbonizedsludge may be possible[64] . Again, research on the fateof heavy metals in HTC is rather limited.

Recent experimental results show that several types

of char from wood met most of the limits set by theGerman Federal Soil Protection Act(Table 7) [142] . Onlyzinc exceeded the limit in all chars. However, for riskassessments, the loading rate is more important thanthe concentration of a pollutant. This means that anagricultural application is possible, as long as the load-ing of heavy metals is considered during periodicalfertilization. In addition to the measurement of heavymetals concentrations, sequential extraction proceduresshould be performed in order to gain comprehensiveknowledge about the mobility of heavy metal speciesin the soil[143] .

Organic compoundsSince the chemistry of pyrolysis involves not only thedecomposition of organic compounds, but also theformation of highly condensed aromatic structures,the discussion of organic compounds has two aspects :pyrolysis can destroy compounds present in the feed-stock or produce them in the process. This is greatlyinuenced by the process and its conditions. Bridleet al. [34] reported that over 75% of polychlorinatedbiphenyls (PCBs) and over 85% of hexachlorobenzenes(HCBs) present in a sewage sludge were destroyedin slow pyrolysis at 450C. Removal of wastewater-relevant organic pollutants by slow pyrolysis of sew-age sludges has also been demonstrated by otherauthors[35,144] . While PAHs are known to form as theresult of secondary thermochemical reactions at tem-peratures over 700C, few evaluations of char for PAHor other organic compounds are available[13,15,145].Investigation of ve chars produced via slow pyrolysisshowed background level dioxin concentrations andlittle PAH formation[306] .

More systematic investigation of the destruction andformation of toxic organic substances in both processesis indispensable for designing pyrolysis processes andin evaluating potential applications for the solid, liquidand gaseous products.

Knowledge gaps in char characterizationIn general, more comprehensive product characteriza-tion and reporting is needed to advance our understand-ing of processes, products and applications. This is an

essential step in the search to relate char properties toeffects in applications[146] , and requires a concertedeffort of key players across disciplines, producers andusers, to choose the relevant characteristics and developtesting methods. Adequate characterization of the charbefore it is used in soil and plant experiments is neces-sary, especially for biochar applications, so that resultscan be generalized and predictors for expected effectsin biochar applications developed. Seeing this needfor a classication system, the International BiocharInitiative has started the process of developing guide-lines for biochars[307] . Various authors have proposedclassication systems for reporting and building on test-

ing in other elds (e.g., charcoal, compost and biowaste/biosolids); a selection of important parameters is shownin Table 8 [147,148,307].

This list encompasses parameters relevant to manyof the stages in the product chain from biomass tosoil application: feedstock, production conditions,char composition and physical, chemical and biologi-cal characteristics. Owing to the heterogeneity of theinput and the char itself, most of the parameters aresum parameters. The International Biochar InitiativeDraft Guidelines for biochar[307] cover most of theseparameters. Other applications will probably not requirethe soil-relevant parameters. However, it is criticallyimportant that the various groups working in this eldare aware of the process and product variations, andrequirements in the product chain. Feedstock, processconditions and efciencies dictate char chemical com-position. Communication between char producers andusers must be developed in order to exploit the abilityto inuence char properties in the production processand ensure the quality of products. It is important tonote that this is an iterative process, especially in biocharapplications: while users try to understand how the charinteracts with the soil and which properties are respon-sible for the interactions, the producers will continueto develop the processes, changing material properties.

Table 7. Heavy metal content of chars after thermochemical conversion of wood.

Process Feedstock Temperature (C) Cd (mg/kg) Cr (mg/kg) Cu (mg/kg) Pb (mg/kg) Zn (mg/kg)

Wet pyrolysis poplar 200 0.35 8.4 23.3 6.7 618Wet pyrolysis pine 210 0.16 1.4 15.8 5.7 603Dry pyrolysis pine 430 0.66 12.7 19.7 8.4 684Gasication poplar 800 0.23 16.5 190.3 15.5 742Limits 1.00 100.0 60.0 100.0 150Limits are from the German Soil Protection Act [142] .Data from [K , U D ]




14/36

Application of char in soilsMuch research effort has been expended in recent yearsto show that returning carbon to the soil in the formof char can sequester carbon and increase soil fertil-ity. This application as biochar could quickly capturea large proportion of biomass production capacity ifit were to be included in the carbon trading schemes,such as of the Clean Development Mechanism (CDM)in the Kyoto Protocol[149] . Its suitability as a carbonsequestration strategy will depend on the overall

carbon balance of the production process and the long-term stability of char in soils[20] . Reliable and repro-ducible methods are needed to assess the sequesteredCO2 equivalents over the product life cycle with duediligence. In addition to the variability in productionand biochar properties, the soil, climatic and man-agement conditions may vary widely from locationto location and will inuence char recalcitrance sig-nicantly. In addition, biochar application may pos-sess additional carbon mitigation potential owing to

Table 8. Overview of proposed classication systems: feedstock, process and product characteristics to be reported.

Parameter Joseph et al. Okimori et al. IBI draftguidelines

Feedstock Source, type and composition Type of pre-processing

Process conditions T max (C) Time (h) Rate of heating Reactor pressure Solid yield (%, g char/g feed, db) Type of post-processing

Chemical composition Elemental composition (%) C, N, P, K Molar H/C ratio Volatile content Ash content Mineral N (mg/kg)

Available P (mg/kg) Heavy metals Extractable (in water and/or solvents), DOC,phenols, chlorinated hydrocarbons, PAH, dioxins

#

Physical characteristics Surface area Bulk density Particle size distributionPore volume(or pore size distribution: ratiomacro/micropore volumes)

Chemical characteristics pH Liming value (% CaCO 3) Electrical conductivity (s/cm) Water-holding capacity Water drop penetration timeCation exchange capacity Caloric value

Biological tests BiodegradabilityEarthworm avoidance/attraction Germination inhibition

Data from [147] .Data from [148] . Data from [307] .Water-soluble fraction.#Relevant local standards should be used.db: Dry weight; DOC: Dissolved organic carbon; H/C: Hydrogen to carbon ratio; IBI: International Biochar Initiative; PAH: Polycyclic aromatic hydrocarbon; T max: Maximum temperature.




15/36

indirect effects; for example, increases in soil organiccarbon (SOC), and decreases in GHG emissions andfertilizer use should be considered in addition to thedirect benets of carbon sequestration(Figure 4) .

From these considerations some fundamental issuesarise that need to be resolved before the true carbonmitigation potential of biochar can be quantied. Theseissues are especially relevant for evaluating hydrocharssuitability in soil applications.

First, the time frame over which carbon must remainsequestered has to be decided upon; for example, allbiochar that is stable for a century could be regardedas sequestered in terms of climate mitigation. Thistime frame must be related to other pathways thatclaim to sequester carbon; for example in the form ofwood (e.g., furniture and reforestation) or SOC (e.g.,restoration of bogs).

Second, a reliable method to determine the truecarbon sequestration potential has to be decided upon.Two pathways could theoretically be considered:

Direct accounting: based on the amount of carbonthat was applied to a soil. The fraction of the appliedcarbon that will remain after an agreed time periodis calculated from the absolute amount of carbonapplied to the soil and a minimum-C-sequesteredfactor. These factors need to be obtained from a broadrange of experimental results (various chars, soils,crops and climates);

Comparative accounting: based on measurements ofthe SOC in char-amended and reference plots. Suchan approach will account for possible carbon lossesassociated with char use. However, this change has tobe assured over the complete sequestration timeframe.

True carbon sequestration potential

Reference system

Referencesystem

Decompositionmodel

Carbon creditsavoided due to

less useof fertilizer

Changes in soilorganic carbon

Carbon creditsdue to

mitigatedN2O/ CH 4

Carbon loss by chardecomposition(estimated via factor )

Direct effects

Indirect effects

Carbon creditsassociated with

a change inproductivity

Minimum ofsequesteredcarbon

Measurementof the soil

organic carbon

Carbon amountapplied with

char

True carbon mitigation potential

Figure 4. Factors inuencing the true carbon mitigation potential of char application. Determination of thistrue carbon mitigation potential requires a well-dened timeframe, over which the carbon is regarded to besequestered and it crucially depends on suitable reference systems.Direct accounting.Comparative accounting.Any carbon-sequestration factor must be assessed conservatively (i.e., the fraction of char that is decomposedshould be overestimated rather than underestimated).




16/36

In addition to the assessment of the true carbonsequestration potential, several indirect effects of charapplication have to be taken into account, such as fer-tilizer use, N2O and CH4 emissions, change in SOC

and increased productivity. These changes have to bedetermined in comparison to a nonamended referencesystem. Naturally, the choice of a proper reference sys-tem is decisive for the accuracy of quantifying theseindirect effects. The sum of the true carbon sequestra-tion potential and the indirect effects represents the truecarbon mitigation potential(see Figure 4) . This sectionsummarizes the research on these direct and indirecteffects for biochar from dry pyrolysis and highlightsthe questions still open for hydrochar.

Stability of biochar & hydrochar in soilsChar from natural fires is usually the oldest car-

bon pool present in ecosystems prone to re[150,151]

.Lehmannet al. conservatively assessed mean resi-dence times (MRT) for naturally generated char in Australian woodlands to be 13002600 years (range:7189259 years[151] . The long-term stability of bio-char has been shown in Terra preta soils, which con-tain considerable amounts of carbon; most of which is5002000 years old, sometimes up to 7000 years (14Cdating,[19,152]). Considerable biochar stocks of 50 t Cha -1 have been found down to 1-m depth in such soils,despite the climatic conditions strongly favoring decom-position and the lack of additional biochar additions inthe last 450 years[19,153] . However, MRT assessmentsfrom natural ecosystems or Terra preta soils can onlydeliver orders of magnitude in accuracy, since there isno way to quantify the initial (repeated) biochar inputto obtain straightforward mass balances[20] .

Despite the fact that biochar contributes the longest-living organic carbon pool in soils[21,154,155] , biocharcannot be considered inert. It will ultimately be decom-posed and mineralized over sufciently long time scales;otherwise the worlds carbon stocks would nally endup in biochar[20,156] .

Hydrochar, with its less aromatic structure andhigher percentage of labile carbon species, will prob-ably decompose faster than char from dry pyrolysis butless quickly than uncarbonized material[157] . WhileKuzakovet al.calculated mean MRTs of approximately2000 years from their 3.2-year incubation study inthe laboratory with14C-labeled char from dry pyroly-sis[21] , Steinbeisset al. reported MRTs between 4 and29 years for two13C-labeled hydrochars made fromglucose (without nitrogen) and yeast (5% nitrogen)in a 4-month incubation study[158] . However, the lat-ter study probably underestimated the true MRT ofhydrochar. Chars usually show two decay phases thatcan be approximated with a double-exponential decay

curve[20] . First, labile carbon substances on the surfacesof the chars are decomposed and the outer surfaces areoxidized (which increases the cation exchange capacityin the case of biochar)[159,160] . Thereafter, the decay

of the more recalcitrant fraction continues much moreslowly (see Figure 11.9 in[20]).However, the true long-term MRT is hard to cap-

ture with a short-term study: the shorter the incuba-tion time and data set, the larger the underestimationof char stability will be. Lehmannet al. showed thatthe MRTs of the recalcitrant fraction, calculated fromthe same data set, can range from 57 years (using onlythe rst 2 years of a 100-year data set) to 2307 years(using the entire 100-year data set)[20] . Thus, long-termstudies are still required for a variety of different charsbefore MRTs can be reliably estimated and tied to charproperties, another reason why a systematic analytical

characterization of the properties of biochar/hydrocharis urgently needed(Table 8) [147] .Physical or biological forces such as freezethaw or

swellingshrinking of clay minerals, or in-growth ofplant roots and fungal hyphae into char particles, mayshatter larger particles into smaller ones. This exposessurfaces and increases the total surface area so that thechar can be further oxidized or degraded. In old bio-char-containing soils, biochar particles are very small(e.g., Terra preta or chernozems); most of the biocharis included within micro-aggregates where they appearto be protected from further decomposition[150,161] .Ploughing and priming by addition of labile carbonsubstrates (cometabilzation) increased biochar decom-position slightly[20,21,156]. White-rot fungi, whose pre-ferred substrate is lignin, and other basidiomycetes wereable to slowly decompose lignite, sub-bituminous coaland biochar via excretion of exoenzymes, such as lac-case, mangane-peroxidase or phenol-oxidase[162,163] . Anitrogen-rich hydrochar was preferentially decomposedby fungi[158] , and considerably stimulated root colo-nization of arbuscular mycorrhizal fungi in mixturesof up to 20% (by volume) beet-root hydrochar chipswas reported[25] . Therefore, fungi will probably be thedominant char decomposers; however, it is unknown towhat extent their activity will impact char stability insoils in the long-term.

Other means of char loss from soils include surfaceerosion or transport to the subsoil, either as small particleswith the rain water, or through dissolution as (highlyaromatic) dissolved organic carbon (DOC). Furthermechanisms of char movement from the surface to deeperlayers include bioturbation, kryoturbation or anthropo-genic management[156,164,165] . The nonmineralizationcarbon losses due to erosion or relocation to deeper soillayers can be considerable and quick: Majoret al. reporteda migration rate of 379 kg C ha -1 year-1 from biochar




17/36

application of 116 t ha -1 to the top 10 cm of a grasslandsoil downward to 1530-cm depth during a 2-year studyin Columbia[165] . Conversely, respiratory biochar-Closses or losses via DOC were rather small (2.2 and 1%,

respectively)[165] . The movement of char (components)to other ecosystems may have signicant effects on itsstability; the rate of mineralization in deeper layers maybe insignicant compared with that of top soils.

Carbon sequestration potential: soil carbonpriming or buildup? An intriguing fact of the Terra preta soils is the sig-nicantly increased SOC stocks besides the black car-bon, compared with adjacent Ferralsols[19] . By contrast Wardleet al. reported an increased mass loss of organicmatter over 10 years in mesh bags with a char mix-ture when compared with those without[166] ; the pH

increase associated with charcoal may have promotedmicrobial decomposition of the acid litter. Therefore,addition of biochar or hydrochar to soils or litter layersmust be carefully investigated with regard to possiblepriming effects that endanger the existing old soil car-bon pool. However, in Terra preta soils, the increasedSOC contents do not indicate a long-term SOC loss dueto biochar presence[19,150,161] .

Biochar and hydrochar seem to promote fungalgrowth, such as arbuscular mycorrhiza[25,167] . Thesefungi produce the protein glomalin which, as a bindingagent, signicantly promotes soil aggregation[168,169] .Thus, both chars may, in the long-term, increase theproduction of nonbiochar SOC by fungal promo-tion [25] , or by formation of organo-mineral complexesand aggregates[20] .

The net outcome of the two opposite mechanisms labilecarbon fraction induces SOC priming and fungal stimu-lation and soil aggregation protect SOC is unknown; itmay vary with ecosystem or hydrochar properties. Hence,eld studies are urgently required(Figure 4) .

Inuence of char on soil fertility and crop yieldsChange of soil characteristics with biochar & hydrocharBiochar application or the presence of charcoal insoils has demonstrated several benecial effects on soilphysicochemical properties. Its presence could:

Enhance the water-holding capacity (WHC)(seeFigure 5C) , aeration and hydraulic conductivity ofsoils[153,170] ;

Reduce the tensile strength of hard-settingsoils[171,172] ;

Increase the CEC of soils[153,159,173] , resulting in animproved nutrient retention or higher nutrient useefciency[174] ;

Stimulate growth, activity and the metabolic efciencyof the microbial biomass[174,175] , including arbuscularmycorrhiza[167,176,25] and N2-xing rhizobiota[177] ;

Attract earthworm activity[172,178,179] .

However, biochar application to soils is not a straight-forward road to happiness; results of no (signicant)change have also been obtained.

Hydrochar will affect soil properties based on thesame basic principles, soil physics and chemistry,water, nutrients and microbial activity. It willvery likely reduce the tensile strength, increase thehydraulic conductivity and enhance the soil WHC.Hydrochars will not have the same large internal sur-faces as biochars, owing to the lower production tem-perature[37,72] . Therefore, water retention curves ofhydrocharsoil mixtures may be different to biochar

soil mixtures and resemble that of peat- or compost-additions to soils.The WHC of hydrochar is usually greater than

that of mineral soils; for example fresh wet hydro-char, pressed wet hydrochar and oven-dried hydro-char (105C) produced from the same feedstock (sugarbeet reminder) showed WHCs of 6.6 0.2 g H2O g -1,5.9 0,4 g H2O g -1 and 1.6 0.1 g H2O per g -1 dryhydrochar, respectively (n = 4/char)[180] . The WHCwas considerably reduced after the hydrochar had beenfully dried, but not after water had been removed bypressing. Some hydrochars do become hydrophobicwhen oven- or completely air-dried. Keeping hydrocharat suitable moisture for use in soil without inducingfungal degradation may be a challenge.

The majority of hydrochars are more acidic thanmany biochars, which are often alkaline, owing to theirash content. Hence, the liming value that alkaline bio-chars can have, reducing, for example, the Al toxicity inacidic soils[19,181,182] , may not be associated with acidichydrochars. Thus, the effect of the two char productson soil biology may vary greatly.

It is likely that hydrochars will undergo ageing pro-cesses similar to biochars, where the number of functionalgroups on the biochar surface increases over time[159,160] ;however, large numbers of carboxylic groups on thehydrochar surfaces already exist that can theoreticallyincrease the CEC of soils, improving nutrients. To ourknowledge, this has not yet been investigated.

Soil fertility & crop yieldFor various biochars, it is well documented that theirapplication can improve crop yields[183] . However, ayield increase is not guaranteed[184,185] . The follow-ing pattern emerges from the recent body of biocharliterature. Biochar application can increase yields:




18/36

In degraded or low-fertility soils rather than atalready-fertile sites[186,171,184] ;In tropical soils[181,184] rather than in temperatesoils[185,178] ;

In combination with NPK fertilizers or nutrient-releasing substances rather than without extra nutrientsupply[171,179,181,186] ; When the chars themselves were sources of nutrients(e.g., biochar from poultry litter[172] ). However, nutrient supply, pH and other soil param-

eter changes alone are not always sufcient to fully

explain the observed positive or negative biochar effectson yields[184,185] . At low application rates (where lowis relative) the effect of biochar on yields was even some-times negative, with nitrogen immobilization by the

high-carbon additive being an often-cited explana-tion[171,183] . However, patterns may change when morestudies become available.

Hydrochars often exhibit higher labile carbon frac-tions such as carbohydrates and carboxylates thanbiochars[157] . Labile carbon in hydrochars could thusinitially induce nitrogen deciency by nitrogen immo-bilization, particularly in chars from nitrogen-poor

g biochar added to 100 g soil

0 2 4 8 160

200

400

600

800

1000

1200

1400

1600

g biochar added to 100 g soil

0 2 4 8 160

50

100

150

200

250

300

350a

b b

c

d

a

a

b

c

b

Mw

Biochar added (g per 100 g soil)0 5 10 15

M a x

i m u m

w a t e r - h o

l d i n g c a p

a c i t y

( g H

2 O / g d w

t )

0.32

0.34

0.36

0.38

0.40

0.42

0.44

0.46

0.48

0.50

0.52

% Biochar addedExponential rise to maximuma

a,b

b,cc,d

d

+6.3

+13.3

+18.5+27.8

CO 2 effluxBC: p < 0.001

N2O emissionBC: p < 0.001

N 2

O e m

i s s

i o n

( n g

N 2 O

- N k g

- 1 h

- 1 )

C O

2 e

f f l u x

( g

C O 2

k g

- 1 h

- 1 )

f(x) = y0 + a* (1-exp (-b*x))R2 = 0.993p < 0.001

A B

C

Figure 5. Reduction of (A) N 2O emissions and (B) CO 2 efflux (mean standard deviation, n = 4) from a sandyloam brown earth mixed with increasing amounts of biochar (peanut hull from Eprida, USA) and set to 65%of the maximum water-holding capacity of the respective mixture. Since the water-holding capacity increasedsignicantly with increasing amounts of biochar ( C ; in purple: percentage increase compared with the control), theabsolute amount of water added to the biochar mixtures was increasingly higher than that added to the control.Letters indicate signicant differences between treatments ( A , B and C : one-way analysis of variance and Student-NewmanKeuls all pairwise test procedure; statistics and curve tting: SigmaPlot 11.0). The ux measurement wasperformed 4 weeks after mixing soil and biochar and incubating it at 22 1C in the laboratory, and 1 day afteraddition of 50 g N g -1 soil of a NH4

+NO3- solution to stimulate denitrication. Jar incubations and GC analyses of the

gas samples for N 2O and CO 2 were carried out as described in [194] .




19/36

feedstock. Conversely, the large labile carbon fractionmay initiate microbial growth and stimulate soil-charnutrient cycling after a lag phase, as observed for somebiochars[175,182,187] .

With freshly produced hydrochars mixed with soil,we observed a preferential growth of fungi (basidiomy-cetes); Rilliget al. recently described a strong stimula-tion of arbuscular mycorrhizal root colonization[25] .Hence, depending on the type of fungus it may thus bepossible to create purpose-optimized designer hydro-chars that, for example, stimulate symbiotic fungi tohelp establish young trees after planting. However,experimental data are currently lacking.

The behavior of nutrients in the production processand in soil application is still an open question. Charfrom dry pyrolysis has been found to lose some of theinitial feedstock nutrients (e.g., via volatilization), or the

nutrients may become unavailable to plants by inclusionin aromatic stable structures[185,188] , while hydrochar,with its lower production temperature, may retain morenutrients in a plant-available form, either in the hydro-char itself, or in the aqueous phase. We have previouslydiscussed some of these results. In the face of declin-ing phosphorus deposits worldwide, the conservation ofplant nutrients from residual materials for agriculturaluse may become one of the most interesting features ofchar in soil applications. In the current quest for themost benecial SOC- and fertility-increasing man-agement practices[189] , it may be highly promising toconsider SOC-increasing soil additives such as chars[18] .

GHG emissions from soils containing biochar When char is applied to soils, it is crucial to quantify thesubsequent uxes of all GHGs (CO2, N2O and CH4)because any positive carbon sequestration effect couldbe diminished, or even reversed, if the emissions of otherpotent GHGs increase after char application. For CO2 emissions from soils, the possibility of priming of oldSOC was discussed earlier in this review. Hence, the fol-lowing section mainly deals with nitrous oxide (N2O)and methane (CH4) uxes.

Sources & sinks of N 2ONitrous oxide is a potent GHG and absorbs, integratedover 100 years, 298-times more infrared radiation thanCO2 [149] . It is predominantly produced during hetero-trophic denitrication of NO3- to N2 as a gaseous inter-mediate, and usually to a lesser extent by nitrication ofNH4+ to NO3- as a by-product (see reviews[190,191]). Ratesof N2O emission from agricultural soils are particularlyhigh after:

Nitrogen-rich fertilizers have been applied, in particu-lar in the presence of labile carbon (e.g., manuresand slurries);

When aerobic soils experience anaerobicity (e.g., aftera heavy rainfall or irrigation);

During freezethaw cycles in spring.

Neglecting N2O emissions may cause consid-erable misinterpretation of the real carbon- (i.e.CO2-equivalent-)sink capacity of an ecosystem, be itagricultural[192] or semi-natural[193] . Conversely, if N2Oemissions are signicantlyreduced by biochar applica-tion, this would considerably improve the GHG balanceof biochar-grown agricultural products.

Effect of biochar soil application on N 2O emissionsRondonet al. [205] reported reduced N2O emissionsafter biochar application. Since then, more reportsof reduced N2O emissions in the presence of biocharhave followed(Table 9) . We also observed a signicant

reduction of soil N2O emissions with biochar, bothwithout plants(Figure 5) [194] , and in the presence ofplants[K , U D ] . The effects of hydro-char on N2O formation in soils are even less investi-gated, let alone understood. In the short term, weobserved signicant reductions of N2O emissions inunfertilized loamy and sandy soils when hydrochar hasbeen added, compared with pure-soil controls[K& R , U D ] .

Mechanisms of N 2O suppression Although biochar application led mostly to reduced N2Oemissions[182] , the mechanisms of N2O suppression arenot well understood, nor have they been specically stud-ied. The following mechanisms might be involved:

Decrease in anaerobic microsites in soil. Char addi-tion can improve the soil aeration and lower the soilbulk density[8, 72, 153,171] . Hence, the presence of charscan probably reduce anaerobic microsites in soils thatare involved in N2O formation via denitrication[191] ;

Change in soil pH. Biochar may reduce N2O uxesvia pH increases because many biochars from drypyrolysis are alkaline. When soils become less acidic

Table 9. Overview of the effect on greenhouse gas ux associated withbiochar or charcoal applications.

Greenhouse gas ux Effect Ref.

N2O emission + [182,208]

- [37,179,182,197,205,209,210]

CH4 uptake + [182]

- [206,209]

CH4 emission - [182]

CO2 efflux (soil respiration) + [181,175,165,211]

- [182,209,211]+: Indicates an increase in the GHG ux; -: Indicates a reduction in the GHG ux.




20/36

(e.g., by liming) more NO3- is completely reduced toN2 (i.e., the N2O/N2 ratio declines)[195,196] . Yanaiet al. reported that charcoal amendments reduced theN2O emissions by 89%[197] ; however, simply adding

the alkaline ashes or the dominant ions associatedwith the biochar (K +, Cl-, SO42-) did not reduce N2Oemissions. Here, biochars and hydrochars may haveopposing effects owing to their pH, with the hydro-chars being often acidic and the biochar often alka-line. However, the acidic nature of the hydrocharseems to be lessened when it is combined with soil[25] ;

N immobilization in soil. Labile carbon compoundsin fresh biochar, or in particular in fresh hydrocharmay lead to a temporary nitrogen immobilization insoils, thereby reducing NO3- or NH4+ available fordenitrication and nitrication and hence for N2Oformation, respectively. However, a reduction of N2O

emissions due to nitrogen immobilization will probablynot be permanent, demanding long-term studies; Stimulation of plant growth. It has long been knownthat plant growth can reduce the nitrogen availabilityto soil (de)nitriers[198200] , which is one of the rea-sons for the use of catch crops. If plant growth isindeed stimulated by biochar, then a larger amountof nitrogen might be xed within the plant biomass,allowing less mineral nitrogen for N2O formation;

Change in nitrogen transformation pathways in soil.Biochar may have the potential to change N trans-formation pathways in soils. Van Zwietenet al.

reported that at the start of incubations of four dif-ferent biochars, NH4+ concentrations were reduced;at the end of the incubations, however, NO3- con-centrations were signicantly increased in all caseswhere biochar reduced N2O emissions[182] . There-fore, soil N transformations clearly changed owingto biochar addition. However, for hydrochar, moreor less nothing is known;

Chemical reduction of N2O. Biochars with their largeporous surfaces may even chemically lead to N2Oreduction, for example, via metallic or metal oxidecatalysers on biochar surfaces such as TiO2 [182] . Ifthis is one of the main mechanisms, it may be morestrongly associated with biochars than hydrocharsbecause of the larger internal surfaces per g of biochar.

For the effect of chars on N2O emissions, manyopen questions remain: do all chars have this reduc-ing effect on N2O emissions? Why does it occur andis the mechanism behind it universal? Could one orboth chars also lead to a stimulation of N2O emissionsunder certain circumstances? Most importantly, doesthe reduction prevail even after years in the eld? To

answer these questions, long-term eld studies with afocus on process-oriented measurements (e.g., involving15N-labeling-tracing techniques) are urgently required.

Sources & sinks of CH 4Methane production by methanogenic archaea willoccur when organic material starts to decay anaerobi-cally; for example, in swamps, bogs, rice paddies, landlls or within ruminant animals. By contrast, meth-ane oxidation by aerobic methanotrophic bacteria(i.e., CH4 uptake and hence CH4 sink capacity) takesplace in almost every soil under oxic conditions[201] .Interestingly, it is well-established that anthropogenicdisturbances such as forest clear-cutting, agricul-tural management (e.g. ploughing) or, in particular,N-(NH4+-) fertilization diminish soil CH4 sinks globally[202204] , for reasons that are not yet fully understood.

Effect of biochar soil application on CH 4 uxesThe effect that biochar or hydrochar additions to soilsmay have on CH4 production or CH4 oxidation isalmost completely unknown. Van Zwietenet al. [182] mention that CH4 production declined to zero in thepresence of biochar in a grass stand and in a soybeaneld, and that CH4 uptake in a poor acidic tropical soilincreased by 200 mg CH4 m-2 per year compared withthe controls[205] . Priem and Christensen[206] reportedthat CH4 uptake rates declined in a savannah that hadrecently been burned. Terra preta studies do not provideclues since, to our knowledge, no GHG uxes have beenmeasured so far.

Mechanisms of CH 4 ux changes after char applicationChanges in CH4 uxes due to char amendment will bethe same as outlined for N2O: reduced soil compactionand improved soil aeration may stimulate CH4 con-sumption, since O2 and CH4 diffusion are regulated bythe key factor, soil water content[207] . Increases in thesoil pH (via biochar) in acidic soils may enhance CH4 uptake rates. In highly nitrogen-loaded ecosystems,inhibition of CH4 oxidation may occur[204] . In suchecosystems, immobilization of nitrogen by biochar orhydrochar application may lead to a faster recovery ofthe CH4 sink capacity of those soils. Although it hasbeen cited and re-cited in many reviews, benecialchanges of CH4 uxes between biochar-amended soilsand the atmosphere clearly leave much room for furtherresearch the issue is far from being resolved.

In summary, the results of the various inves-tigations present a mixed picture(Table 9)[165,175,179,181.182,197,205,20 6,208 211]. As mentioned in thesection on char characteristics, research targeted atrelating char properties to soil interactions is needed(Table 8) .




21/36

Best practice considerations forbiochar/hydrochar soil applicationTo date, standardized best practice guidelines for theapplication of both char types have not been estab-

lished. Challenges that have to be considered for drychar include its dustiness and its very low packingdensity[183] . Major and Husk[178] estimated that, in acommercial larger biochar eld trial in Canada, approx-imately 30% was lost during transport and incorpora-tion. Since black carbon ne particles and soot haveconsiderable greenhouse potential[149] , dust loss must beavoided. In addition, the risk of spontaneous combus-tion or dust explosions in the presence of open re hasto be considered[183] .

To ease the application of biochar to agricultural sitesand minimize losses when working with machinery,several strategies have been suggested[183] , including

a preliminary mixing of the bio- or hydrochar withcompost, liquid manures or slurries; or wetting a drychar with water, followed by management practicessuch as ploughing, discing or deep-banded applica-tion. Strategies of mixing or composting the biocharor hydrochar with a nutrient-carrier substance such asgreen waste (composting), slurry or manure will have thepositive side effect of loading the char with nutrients.

In contrast to biochar, hydrochar is wet when it leavesits production process

htc review biofuels 2011

Documents