activated carbon from co-pyrolysis of particle board and melamine (urea)

Chemical Engineering Journal 172 (2011) 835 846

Contents lists available at ScienceDirect

Chemical Engineering Journal

j ourna l ho mepage: www.elsev ier .c

Activated carbon from co-pyrolysis of particle boformal ati

K. Vanre pera NuTeC, Deparb Research Gro 3590 c Research Grou 3590 D

a r t i c l

Article history:Received 8 April 2011Received in revised form 22 June 2011Accepted 23 June 2011

Keywords:PyrolysisActivated carbEconomicsParticle boardMelamine form

The disposal and environmental problems associated with waste resin produced during the productionof melamine (urea) formaldehyde and wood waste (i.e. particle board) containing these aminoplastsrequires a processing technique which results in products of added value and which meets both ecologicaland economical needs. Several published results demonstrate that nitrogen incorporation in activatedcarbon can play a signicant role as a key parameter for the adsorption properties, as well as for the

1. Introdu

During t(both furthboard (PB) cannot be r

In additition) of wocause pollu

CorresponE-mail add

[email protected]@sonja.schreurs

1385-8947/$ doi:10.1016/j.on

aldehyde

catalytical activity and the dispersion of carbon supported catalysts.The production of high value nitrogenised activated carbon, after thermal treatment in an oxygen

decient environment and subsequent activation, is considered as a possible opportunity.This research paper investigates the feasibility of a process design for the production of a high added

value nitrogenised activated carbon by co-pyrolysing a mix of particle board and melamine (urea)formaldehyde waste. A process design and an economical model for estimating the total capital invest-ment, the production costs, the possible revenues, the net present value and the internal rate of return isdeveloped based on various literature sources. In addition, Monte Carlo sensitivity analysis has been car-ried out to determine the importance of the main input variables on the net present value. It is assumedthat the manufacturing facility obtains its waste from various sources and operates continuously during7000 h a year. The study investigates the plants protability in function of processing rate and mixingratio.

Even though the current assumptions rather start from a pessimistic scenario (e.g. a zero gate fee forthe melamine (urea) formaldehyde waste, a rst plant cost, etc.) encouraging results for a protable pro-duction of activated carbon are obtained. Moreover, the ability to reuse two waste streams and possibleproduction of a specialty carbon enhances the value or usefulness of the activated carbon manufacturingfacility.

2011 Elsevier B.V. All rights reserved.

ction

he production of melamine (urea) formaldehyde resinser abbreviated as MF) for the production of particlea considerable amount of waste resin is produced thate-used or recycled at this moment.on, classical thermo-chemical conversion (e.g. combus-od waste containing these aminoplasts resins mighttion because it results in the production of toxic gases

ding author. Tel.: +32 11 2 68 320.resses: [email protected] (K. Vanreppelen),uhasselt.be (T. Kuppens), [email protected] (T. Thewys),uhasselt.be (R. Carleer), [email protected] (J. Yperman),

@xios.be (S. Schreurs).

like ammonia, isocyanic and hydrocyanic acid and nitrous oxides[13].

A sustainable solution is more and more required to avoid envi-ronmental problems and landlling costs, and to turn this wastestream in a rather protable material resource. A possible oppor-tunity, is the production of high value activated carbon (AC) afterthermal treatment in an oxygen decient environment and subse-quent activation.

ACs are produced for a large number of dedicated applicationsboth as structural and functional materials. ACs are generally usedfor air, water and gas purication, chemical and pharmaceuticalprocessing, food processing, decolourization, solvent vapour recov-ery, llers in rubber production, refractory materials, catalysis andcatalyst support [46].

Marsh and Rodriguez-Reinso [5] estimated the world annualproduction capacity of AC to be around 400 kt in 2006, exclud-ing countries without accurately known gures like China and

see front matter 2011 Elsevier B.V. All rights reserved.cej.2011.06.071dehyde resin: A techno-economic evalu

ppelena,b, T. Kuppensc, T. Thewysc, R. Carleerb, J. Ytement TIW, XIOS, Agoralaan Gebouw H, 3590 Diepenbeek, Belgiumup of Applied and Analytical Chemistry, CMK, Hasselt University, Agoralaan Gebouw D,p Environmental Economics and Law, CMK, Hasselt University, Agoralaan Gebouw D,

e i n f o a b s t r a c tom/ locate /ce j

ard and melamine (urea)on

manb,, S. Schreursa

Diepenbeek, Belgiumiepenbeek, Belgium

836 K. Vanreppelen et al. / Chemical Engineering Journal 172 (2011) 835 846

some other Eastern countries. Furthermore the market is increasingconstantly, due to the environmental awareness and the growingindustrialization. Girods et al. [7] expect a growth of 5.2%/yearto 1.2 Mt by 2012. In Europe, Japan and the USA the growth is15%/year, countries. Tduction cosspecial carbet al. [7] staproducers w

The widof pores, hiACs. The nrial and thethat the phenced by thetc. In normligible [4,6]positive effeadsorption acid gases lpounds [2,3role for thecatalysts [9genised act(2.0 kEURbetween 0.8ing to Infomsaturated cato 6.0 kEURimpregnati

Becauserials result and activatis consideranitrogen comaterials hcost [1] for ter does noMF waste, bmaterial.

The objerendering tFor this purcarried outduction of diagram ofwith physicgenerated bhas been caAC has beenforming Moeconomic fethe protab

2. Process

The prelPB waste ccan be diviand packagtion facilitysize (a few cult to predet al. [2] de

Next, the grinded and dried waste will be transported to a rotarypyrolysis furnace (operating at 800 C). Here the waste is pyrol-ysed in an oxygen-free environment for a few minutes (25 min).The developed chars (solid fraction) are then transported to a sec-

tary mperThe t conrolysy mate hnd aclonC wit

harm mullled is is s a hefter c

screr wa. A pon can thient

ens e

nom

r capn orent

t or i ma, the

are uivatioestm

valuvest

calcu

Tn=1

h:

cashitiale lifescoun

sh o expulatecounns etulate

(1 tscouthe on pre spe MFwhereas this rate is much higher in the developinghe price of AC is a function of demand, quality, pro-t, etc. A typical price range is 1.46 kUSD/t, but for veryons the price can increase to 20 kUSD/t [5,8]. Girods

te that the average production cost of AC from the majoras on average 2.5 kUSD/t.

e range of applications exists thanks to the high volumegh surface area and the variety of surface chemistry ofal properties of the AC are related to the precursor mate-

activation process (physical or chemical). It is statedysicochemical properties of the ACs are strongly inu-e presence of heteroatoms like oxygen, nitrogen, sulfur,al conditions the amount of nitrogen in the AC is neg-. Several published results however, demonstrate thect of nitrogen incorporation as a key parameter for theproperties of the AC [9], especially for the removal ofike hydrogen sulde, sulfur dioxide and phenolic com-,7]. Nitrogen incorporation can also play a signicant

catalytic activity and dispersion of carbon supported]. According to Girods et al. [2] the value of such a nitro-ivated char from PB (in 2006) is on average 2.5 kUSD/t/t), whereas normal ACs are sold (in 2008) at prices

kEUR/t and 1.7 kEUR/t (1.22.5 kUSD/t) [10]. Accord-il [10], impregnated ACs (i.e. including pick-up of therbon) have a higher selling price (in 2008) of 4.0 kEUR/t/t (5.98.8 kUSD/t) due to higher costs incurred by theon step.

the chemical properties of the PB and MF waste mate-in in situ nitrogen incorporation during char formationion, the production cost of nitrogenised activated charbly reduced in comparison with post impregnation ofntaining components on AC. In addition, these wasteave the economic advantage of representing a negativea waste processing company, which means that the lat-t have to pay for obtaining resources such as PB andut instead receives a gate fee for processing the waste

ctive of this work is to identify the crucial variables forhe production of AC from PB and MF waste protable.pose, a preliminary economic feasibility study has been

for a process design especially developed for the pro-AC from PB and MF waste. After developing a process

an AC production technique (co-pyrolysis combinedal activation), the net present value of the cash owsy an investment in co-pyrolysis and char activationlculated. The minimum selling price of the produced

determined, taking into account uncertainties by per-nte Carlo sensitivity analysis. Finally, this preliminaryasibility study is used to identify the key variables forility of the production of AC from PB and MF waste.

design

iminary process design for the production of AC fromo-pyrolysed with MF is shown in Fig. 1. The processded in four parts: pretreatment, pyrolysis, activationing. After shipping the raw materials to the AC produc-, they are rst mixed and milled into a smaller particlemillimetre), dried and transported to a silo. It is dif-ict the moisture content of the incoming waste. Girodstermined the moisture in wood board to be about 7%.

ond roat a teagent. rate buthe pypied badequagases aby a cy1000

tion ofusing acontrobut thused aator. Abeforerecovecardedemissirated iequipmLemm

3. Eco

Pooity of ainvestmprojecgorizedcriteria(IRR), sis/actan invtodaysof an inNPV is

NPV =

Wit

- CFn =- I0 = in- T = th- i = di

The ca(R) andTo calcinto acKuppebe calc

CFn = The dirating Taxes The lifBecauskiln furnace where they are activated during 30 minature of 800 C in the presence of steam as activationpyrolysis and activation are carried out in two sepa-nected furnaces to achieve a continuous system. Both

is and activation kiln have a cross-sectional area occu-terial which is 10% of the cylinders length to ensure aneat transfer and mixing [11,12]. The produced pyrolysiserosols are conducted to a thermal combustor followede for complete combustion at a temperature of aroundh a residence time of at least 2.5 s. This reduces forma-ful compounds or promotes their breakdown [13]. By

tiple zone oxidizer the formation of NOx can be furtherby managing the oxygen inow in the different zones,not implemented at this stage. The hot ue gases areat source for pyrolysis/activation and the steam gener-ooling, the produced AC is transported to a storage siloening and packaging. The remaining gases are cooled toter from the steam generator. After cooling they are dis-elletisation device and an extra gas cleaning unit beforen also be installed, but are at the moment not incorpo-s analysis. The possible extra investment costs for this

can be found in recent literature e.g. Lima et al. [11] andt al. [14].

ical feasibility model

ital investment decisions can alter the future stabil-ganisation. Investors deal with this problem by using

decision rules which evaluate the protability of thenvestment. Biezma and San Cristbal [15] have cate-ny various investment criteria methods. Two of these

net present value (NPV) and the internal rate of returnsed to evaluate the economics of the MFPB pyroly-n. The NPV is the best criterion for selecting or rejectingent, either industrial or nancial [16,17]. The NPV ise of current and future cash ows, which are the resultment using a predetermined discount rate [17,18]. Thelated with Eq. (1) [1720].

CFn(1 + i)n I0 (1)

ows generated in year n; total capital investment (see Table 1, row 19) in year 0;

span of the investment;t rate.

w in a given year is the difference between revenuesenditure (E) after tax (t) generated by the investment.

the cash ow, depreciation (D) also needs to be takent because it lowers tax payments [19,20]. According to

al. [19] and Thewys and Kuppens [20] cash ows cand using the following equation:

) (R E) + t D (2)nt rate of the invested money is set at 9% incorpo-market interest rate and some risk premium [19,21].ots to be paid amount up to 33% in Belgium (t = 0.33).an of a reactor is described as 20 years [19,20,22].

is easy to coke, all the results are based on a rather

K. Vanreppelen et al. / Chemical Engineering Journal 172 (2011) 835 846 837

low averagthe year (ogeneral, whsion.

The IRR expected caexpected cacount rate tin nancialcan expect the requirerejected.

itionhis e NPVrderonte

is repdompose

Table 1Multiplying fa

Direct costs

Indirect cost

Fixed-capitaWorking captotal capital Total capitalFig. 1. Process ow.

e operating time of the reactor of 7000 h, the rest ofr 20% of the year) is used for maintenance etc. Inen the NPV is positive, the investment is a good deci-

is the discount rate (i) at which the present value ofsh inows from a project equals the present value ofsh outows of the project. In other words, it is the dis-hat makes the NPV equal to zero. It is frequently used

Addlated. Tthat th

In oNPV, Manalysous ranpresup markets because it gives the return that the investorfor a given level of risk [17]. If the IRR is lower thand return (discount rate) then the project should be

in a distribuncertaintyrespondingfor the tota

ctor for the delivered-equipment cost.

Cost component

Delivered equipment InstallationPiping (installed) Instrumentation and controls (installed) Electrical systems (installed) Buildings (including services) Yard improvements Service facilities (installed) Land Direct plant costs (DPC)

s Engineering, supervision Construction expenses Legal expensesContractors fee Contingency Indirect plant costs (IPC)

l investment ital (15% ofinvestment)

investment ally the minimal selling price of the AC has been calcu-is the minimal price at which the AC should be sold so

breaks even or in other words the NPV equals at least 0. to have an idea about the impact of uncertainties on the

Carlo sensitivity analysis is performed. The sensitivityeatedly calculates the NPV corresponding to numer-

draws for the value of uncertain variables following ad distribution. Monte Carlo simulations typically result

ution of NPVs that can be declared by the degree of

of each individual variable. Each variable with its cor- range of values and distribution is partly responsiblel uncertainty of the NPV. The variables with the high-

Percent of equipment cost

100393126102912556

308

32344

1937

126

43476

509


est inuence on NPV sensitivity should be identied and shouldbe the subject for further research so that they can be controlledwhen putting the project into practice. In our study 10 000 runs arecarried out using the @Risk software from Palisade Decision Tools.

Finally, when calcushould be sNPV.

4. Model a

The rsttial investmanalysis, thused, with a

Before asum of moequipment The total cabuilding itsing, electricalso indireccosts for enof these inthe direct pcapital inveraw materiabe added foing capital as a percendetermineddisplayed in

All costsexchange raMarshall an

Cost of Equi

4.1. Total ca

The totaby the Percement cost inery allowan

One of talways avaitions can befactor rule aeven total p

Eq (YCh

Cost of unit

The actual from less ththe capacitypyrolysis fa

used as suggested by Henrich et al. [28], Tock et al. [29] and Gassnerand Marchal [30].

The equipment cost of the pyrolysis and activation reactor(which is on its turn a pyrolysis reactor) is derived from the

apital investment (FCI) of a pyrolysis plant presented byater et al. [22] in Eq. (5) with Qfeed input pyrolysis the owf the feed (ton dry matter per hour):

olysis plant = 40.8 103 (Qfeed input pyrolysis 103)0.6194

(5)

calculates the cost of the fast pyrolysis reactor, the feed-tem and liquids recovery. Eq. (5) is a result of a regression4 cost data preformed by Bridgwater et al. [22], and the dataumed to be rst plant costs from a novel technology. This isant, because there can and probably will be a considerableduction from the learning effect. Henrich et al. [28] state

the same type of facility is designed, built and operatedl timlly w28], two-

usedent

ausete thent

ridgwdirecidgwnd inn, 315% fes foping

the

ent

so imr is re7):

put act

h YCin Tar in tm thsis an) to fo

ect pla + 10ect plathese uncertainties have been taken into accountlating the minimum selling price at which the ACold in order to guarantee a 95% chance on a positive

ssumptions

step in NPV calculation is the estimation of the ini-ent expenditure. For preliminary economic feasibility

e Percentage of Delivered-Equipment cost is commonlyn expected accuracy of 2030%.n industrial facility can be put into operation, a largeney needs to be spent on the necessary machinery,and their delivery: i.e. the cost of delivered equipment.pital investment however also consists of costs of theelf, the land on which the equipment is installed, pip-al systems, etc. These are all direct plant costs. Butt plant costs have to be taken into account: such asgineering, legal expenses, contingencies, etc. The sumvestment costs, i.e. the cost for delivered equipment,lant costs and the indirect plant costs are called xed-stment. The amount of money required for a stock ofls and cash kept on hand, i.e. the working capital, shouldr estimating the total capital investment. The work-and the direct and indirect plant costs are expressedtage of the delivered equipment cost. The percentages

by Peters et al. [23] are used in the calculation and are Table 1.

have been updated to 2009, based on the US Dollar/Eurotio provided by the European Central Bank [24] and thed Swift Index [25] (see Eq. (3)):

pment (2009) = cost of Equipment (year)

Cost index (2009)Cost index (year)

(3)

pital investment

l investment cost for production of the AC is determinedntage of Delivered-Equipment cost method. The equip-formation was provided by literature (Table 2). A deliv-ce of 10% on the purchased equipment cost is used [23].he problems in cost estimating is that cost data are notlable for the particular size or capacity involved. Predic-

made by using Eq. (4) which is known as the six-tenthsnd is widely used in approximations of equipment androcess costs [23].

uipment cost reactor = 40.8 103[(Qfeed input pyrolysis 103)

0.6194 +490.1%

(new) = cost of unit(ref.)

(

capacity (new)capacity (ref.)

)capacity exponent(4)

value of the cost capacity exponent in Eq. (4) can varyan 0.3 to greater than 1.0. Yassin et al. [27] stated that

exponent is in the range of 0.60.8 for gasication andcilities. In this calculation a capacity exponent of 0.7 is

xed-cBridgwratio o

FCI pyr

Eq. (5)ing sysfrom 1are assimportcost rethat, ifseveranentiaet al. [about cost isinvestm

Beccalculainvestmcost. Bof the that Brbe fouerectiotems, 5expensand picost of

Equipm

It is alreactoin Eq. (

Qfeed in

Witfound reacto

Fropyroly(5)(7

1

Dir31%

Dirar Qfeed input pyrolysis 103)0.6194

(8)

es the investment and operating costs decrease expo-ith the number of built plants. According to Henrichit is reasonable to set the total capital investment atthirds of the rst plant cost. In this paper the rst plant. Therefore, it may be inferred that a rather pessimistic

cost scenario is applied. Bridgwater et al. [22] used other and less factors toe indirect plant cost and direct costs, the xed capital

of the pyrolysis plant is recalculated to the equipmentater et al. [22] stated that the total plant cost is 169%t plant cost. The percentages of the direct cost factorsater et al. [22] used are not dened, but the factors can

Peters et al. [23]. So they are assumed to be 39% for% for piping, 26% for instruments, 10% for electrical sys-or civil works and 29% for structures and buildings. Ther lagging are included under the equipment installationcosts like Peters et al. [23] suggested. The equipmentpyrolysis reactor can thus be calculated by Eq. (6).1

cost pyrolysis reactor = Direct plant cost290%

= FCI pyrolysis plant490.1%

(6)

portant to note that the feed input of the activationlated to the feed input of the pyrolysis reactor as dened

ivation = YChar Qfeed input pyrolysis (7)

har the char yield from the pyrolysis step which can beble 4 and Qfeed input activation the input of the activation/h dry matter.e foregoing discussion the total equipment cost for thed activation reactor can be calculated by combining Eqs.rm Eq. (8).

nt cost = equipment cost pyrolysis reactor (100% + 39% + 26%+% + 29% + 55%) = 290% equipment cost pyrolysis reactor)nt cost = FCI pyrolysis plant

169%

.


Table 2Major equipment cost an their scaling factors used.

Item Sizing parameter Unit cost year Reference

Crusher 2 t/h 44.94 kUSD (2002) [13]Dryer 10 t/d 51.27 kUSD (2002) [13]Silo (raw material) 500 m3 380 kEUR (2009) [26]Nitrogen storage tank 1.92 m3 19.26 kUSD (2002) [13]Pyrolysis and activation reactor / Eq. (8)Steam boiler and condenser 3781 kg/h 116.84 kUSD (2002) [13]Thermal combustor 1000 N m3/h 40 kEUR (2004) [14]Cyclone 1000 N m3/h 1.5 kEUR (2004) [14]Silo (activated carbon) 500 m3 380 kEUR (2009) [26]Screening and Grading 139.9 kg/h 25.0 kUSD (2005) [11]Packing of activated carbon 139.9 kg/h 25.0 kUSD (2005) [11]

The formula is used in the assumption that condensable gases(pyrolysis liquids) will not be condensed (i.e. direct diversion to thecombustion system), but instead the AC needs to be cooled. There-fore it is assumed that the cost of the liquids recovery of Bridgwateret al. [22], is the same as for the AC cooling.

The produced gases can be considered as a mixture of ammable(toxic) comenvironmencombustiopounds. Thbe performa cyclone. Ttained at reduces thewill be destthrough a cLemmens eWith a max1000 N m3/respectively

4.2. Expend

The totacost and thassumed thyearly interBelgostat [3than 5 yearan amountof 4.0% is apcosts of pyinsurance, oare generalinvestment

A summary of literature percentages to calculate the annualoperating cost is displayed in Table 3.

In this model, the annual maintenance cost is accounted for 3%,the annual overhead and insurance cost for 2% of the total xed-capital investment. The cost for maintenance labour is incorporatedin Eq. (9) (Labour cost) [19,20].

labo2]. T

/h) Q

cost

onomyed R in uros

to bed thAC pr

delie MFaid, DispR/t (aste cilitys tha

the io.rovi

as a put (gwa

the

Table 3Summary of th

Annual oper

Maintenance

Insurance

Insurance an

Overheadspounds at enhanced temperature. From an energy andtal point of view there is a need to (re)use this heat orn energy and decompose or separate the toxic com-ermal treatment of the volatile combustible gases willed by a direct red thermal oxidizer combined withhe temperature of the combustion chamber is main-1000 C with a residence time of minimum 2.5 s. This

formation of NOx and harmful materials, like dioxins,royed [13]. For complete combustion, the ue gases passyclone where possible solid particles are separated out.t al. [14] calculated a cost estimate for these systems.imum cost of 40 kEUR and 1.5 kEUR for a gas stream ofh for the combustion chamber and the cyclone is used.

iture

l expenditure of the project consists of the operatinge yearly interest payments. Thewys and Kuppens [20]at an investment is nanced by means of a loan with aest of 4.60% in Belgium. The macro-economic database1] gives an average initial interest provision for mores of 3.9% on new credits (in 2009) for the euro area for

of more than 1 MEUR. In this model an interest rateplied as a realistic compromise. The annual operating

rolysis and activation consist of maintenance, labour,verhead, delivered feed, energy and water costs whichly expressed as a percentage of the total xed-capital

[20] except the last three items.

Theet al. [2feed (t

Labour

FPS Ecemplo48 kEUfrom Ematedassumof the

Theand th[7] is pplant. 220 EUthis wtion fapay lesmodelscenar

To pto act feed in

Bridcooling

e xed annual operating factors.

ating cost Expressed as

3% of xed-capital investment

5% of xed-capital investment (for a gasication sy3% of xed-capital investment (for a combustion p4% of xed-capital investment 6% of xed-capital investment 2.5% of xed-capital investment 2% of xed-capital investment

2% of xed-capital investment 1% of xed-capital investment

d general 1% of xed-capital investment

4% of xed-capital investment 2% of xed-capital investment ur costs are calculated with Eq. (9) based on Bridgwaterhe calculation is in function of the ow rate of the dryfeed input pyrolysis and will always be rounded up.

= 1.04 ([1 + YChar] Qfeed inputplrolysis)0.475 3shiftsannual salary (9)

y [36] states that the annual salary of one person,in the industrial sector in Belgium was on average2004. By using the annual nominal unit labour cost datatat [37] the average annual salary in industry is esti-e around 55 kEUR in the year 2009. In this model it isat 3 shifts are sufcient for a good and secure operationoduction facility.vered feed cost consists of the cost of the PB waste

waste. For processing PB waste a gate fee of 70 EUR/twhich is an incoming cash ow for the AC productionosing of MF waste to a landll site costs a MF factoryincluding transport) in Belgium. This could mean thatalso represents an income stream for the AC produc-, as the MF factory is already satised when it has ton 220 EUR/t for disposing its MF waste stream. In thiscost of the MF is set at 0 EUR/t to have a worst case

de an oxygen free environment, nitrogen gas is appliedpurging gas. In this study a rate of 8 kg nitrogen gas/tbased on [35]) with a cost of 2.5 EUR/kg is applied.ter et al. [22] used 18.5 m3 water/t input material forproduced pyrolysis liquid and Ko et al. [35] used 13.5 m3

Reference

[19,20,32,33]

stem) [13,27,30]rocess) [27]

[34][35][22][11]

[20,33][35]

[32]

[34][20,33]


Table 4Estimated costs and yields of the input feed.

Ratio (MF/wood) 4 MF1 PB 3 MF2 PB 2 MF3 PB 1 MF4 PB 0 MF5 PB

Yield Chara 22% 29% 36% 43% 50%Yield active carbona (vis--vis char) 44% 45% 47% 48% 50%

a Girods et al. [7] determined the carbon yield after pyrolysis (400 C) typically 50% and again 50% after pyrolysis combined with steam activation (800 C).Own laboratory experiments on MF give a yield of 15 and 42% respectively (800 C).

water/t input material to generate steam for the activation andcooling water for the produced pyrolysis liquid. The quantity ofcooling water (from surface water 20 C) needed to cool the pro-duced AC from 800 C to 20 C is 13 t/h. In this calculation it isassumed that the specic heat capacity of AC is equal to the specicheat capacity of graphite (709 kJ/(t C)) and the maximum tem-perature of

dened by are assume1.5 EUR/m3

Anothersplit in twomated thatelectricity. consumptioity is set atsolid wastebiomass a rture [3942and the actalso neededFor the dryIn most pyrrequired hethe char. Inonly AC anbe thermallsented modis assumed.

4.3. Revenu

It is expeand 4.5 kEUprocessing MF resin an(see Table 4incurred.

Table 4 shown thatin the mixtratio and th

In someical (gover. . ., recover

production of green electricity and heat. However, these are nottaken into account in this process because these are mostly meantas temporary regulations, which differ from country to coun-try.

5. Results and discussion

onom

NPV Fig

The ted in

us a lnd onsatghere (4tal oalys

ing c PB

minoduc

shaeeds

kEUR0 MF.y thee disF5

ccepsharePB) alear vbles ixturows en se

price NPVs a o bre

Table 5Summary of co

1 t/h F3 P

Total capital 263 kEOperating co 641 kEGate fee deliTotal operatProduced acMinimal sell cooling water that needs to be discharged is 30 C asthe Belgian legislation. Here, the water requirementsd to be 15 m3 water/t input material with a cost of.

utility required in the process is energy which can be parts, power and heat requirements. Ko et al. [35] esti-

a 1.25 t/h processing plant producing AC uses 200 kWSo it is assumed that for a 1 t/h facility the electricityn is 160 kW. In this estimation the price of electric-

0.0725 EUR/kWh. The heat of pyrolysis for municipal is calculated by Baggio et al. [38] as 1.8 MJ/kg. Forange of 2 MJ/kg to 3.47 MJ/kg can be found in litera-]. In our case a value of 2.5 MJ/kg for both the pyrolysisivation step is taken. In the activation step steam is. Heating water from 20 C to 800 C requires 5.5 MJ/kg.ing process 2.67 MJ/kg water in the wood is needed.olysis reactors (for the production of pyrolytic oil) theat is provided by the combustion of the gas and/or

this application, as explained before in Section 4.1,d gases (as by-product) are produced. The gases willy destroyed and provide the required heat. In the pre-el a higher heating value of 1617 MJ/kg for the gases

es

cted that the AC can be sold at a price between 1 kEUR/tR/t AC. Net present values have been calculated forcapacities of 1 t/h and 2 t/h waste in different ratios ofd PB waste. Different ratios result in different yields) and different qualities and hence different costs are

provides a guideline for the char and AC yields. It is the AC yield increases when the share of PB increasesure. So the highest yield is obtained by the 0 MF5 PBe lowest by the 4 MF1PB.

countries subsidies can be applied such as ecolog-nmental) premium, a discount for waste treatment,y and selling of other possible by-products, possible

5.1. Ec

Thelined inratios.displayand thment acompeand higate fethe toThe anoperatMF 1NPV.

Thethe pring thethat nto 1.7and a Table 6

Onlthan thand 0 Min an amajor MF2

A cand Tafeed mcash AC whsellinga 0 EURrequireorder t

sts for the production of active carbon by this model.

4 MF1 PB 3 MF2 PB 2 M

investment 10 221 kEUR 10 764 kEUR 11 st (without feed cost) 1547 kEUR 1595 kEUR 1

vered feed 98 kEUR 196 kEUR 294 kEing cost 1449 kEUR 1399 kEUR 1347 kEtivated carbon 678 t/year 914 t/year 1184 t/ying price activated carbon 4.2 kEUR/t 3.2 kEUR/t 2.5 EUic evaluation of the base case

s corresponding to a 1 t/h processing facility are out-. 2 as a function of the selling price of AC and the MFPBotal investment and operating costs for this facility are

Table 5. A higher AC yield (i.e. less MF in the feed mix)arger installation are the cause of slightly higher invest-perating (without feed cost) costs. This small increase ised by the income provided by the gate fee of the waste

yield (revenue) of AC: i.e. in the 0 MF5 PB ratio the90 kEUR/year) is responsible for a decrease of 30% ofperating cost (1726490 kEUR/year = 1236 kEUR/year).is of Table 5 and Fig. 2 illustrates that the lowerosts and higher AC yields in the successive range (4

0 MF 5 PB) of ratios are responsible for the higher

imal selling price (NPV = 0 EUR, break-even point) ofed AC can be found in Table 5 and Fig. 2. By increas-re of PB in the ratio the minimal selling price for AC

to be achieved gradually decreases from 4.2 kEUR/t/t which corresponds respectively to a 4 MF1 PB5 PB ratio. The accompanying IRRs are presented in

cases in the green box (full line) where the IRR is highercount rate can be accepted. The 2 MF3 PB, 1 MF4 PB

PB feed mixture appear to be the most likely to resulttable investment project (i.e. when IRR > 9%). When the

in the mix comes from MF waste (i.e. 4 MF1 PB or 3n investment is only acceptable a high AC prices.iew on the situation can be made by combining Fig. 25 and 6. For example: a 1 t/h processing facility with ae of 1 unit MF and 4 units PB would yield a NPV of theof 4.2 MEUR, an IRR of 14% and a yearly output of 1.4 ktlling the product at a price of 2.5 kEUR/t. The minimum

of a mixture of 1 unit MF and 4 units PB to yield at least is 2.0 kEUR/t. A feed mixture of 3 MF and 2 PB however

higher minimum selling price of at least 3.2 kEUR/t inak even.

B 1 MF4 PB 0 MF5 PB

UR 11 733 kEUR 12 180 kEURUR 1684 kEUR 1726 kEUR

UR 392 kEUR 490 kEURUR 1292 kEUR 1236 kEURear 1445 t/year 1750 t/yearR/t 2.0 kEUR/t 1.7 kEUR/t


5.2. Econom

When loone could aBecause thiand the higcompared tity of the reand RodriguFig. 2. Net present value for a 1 t/h processin

Fig. 3. Net present value for a 2 t/h processin

ic impact of the nitrogen content of the AC

oking at the previous analysis (and Fig. 2 and Tables 46)rgue that it is only usefull to study the 0 MF5 PB ratio.s mix has the lowest minimal selling price of 1.7 kEUR/thest output of AC (1.7 kt/year), it results in higher NPVso mixes with a higher share of MF. However, the qual-sulting AC needs to be considered. Bandosz [4], Marshez-Reinoso [5], Menndes-Dias and Martn-Gulln [6]

state that aformance ocan be seenin the succeet al. [7] ha1.52 wt% whigher nitrovation condand thus yieg facility.

g facility.

higher nitrogen content corresponds to a better per-f the AC resulting in higher attainable selling prices. It

that the nitrogen content of the resulting AC decreasesssive range (4 MF1 PB 0 MF5 PB) of ratios. Girodsve produced an AC from PB with a nitrogen content ofith an estimated value of 2.0 kEUR/t. They state that agen content could be obtained by optimizing the acti-itions, and hence probably better adsorption propertiesld a higher value. Therefore, if AC production from pure


Table 6IRR for the 1 t/h feed input factory.

PB is optimized, a somewhat higher selling price with an expectedmaximum of 2.5 kEUR/t corresponding to an NPV of 8.8 MEUR canbe achieved.

In our case of mixing the PB with MF, AC with an even highernitrogen content could easily be achieved. In recent literature, theestimated selling value of specialty (impregnated) carbons is inthe range of 4.06.0 kEUR/t (incl. pick-up of the saturated carbonand sellingAC is imprcase, providof the AC

tion, in some extra specialty cases even higher prices can beachieved.

Therefore, one should take the nitrogen content into account,which is the highest in the feed mixture of 4 MF1 PB and gradu-ally decreases as the share of MF in the ratio decreases. It meansthat the mixtures with a higher share of MF have a higher chanceof reaching a selling price of 4.0 kEUR/t to 6 kEUR/t. From Fig. 2

ble 5s a o bre

valu prices of 2008) [10]. An example of such a specialtyegnation with NaOH to trap acidic components. Oures the incorporated nitrogen for the basic properties

thus a similar sales value can be expected. In addi-

and Tarequireorder ta salesFig. 4. Importance of the distinct expenditu, it can be seen that, a feed mixture of 3 MF and 2 PBhigher minimum selling price of at least 3.2 kEUR/t inak even. However, this mixture is more likely to reache of 5.0 kEUR/t with a NPV of 10.2 MEUR (comparedre items.


to pure PB 18.4 MEUR so this pricthat the mixa NPV of 7.the situatiothe NPV of price of theaccount, it Fig. 5. Mean NPV output after Monte Carlo

Fig. 6. Average sensitivity of the crucial variables on the

NPV of 8.8 MEUR). In the case of 2 MF3 PB a NPV ofis achieved. However, its nitrogen content will be lower,e (of 5.5 kEUR/t) may not be achievable. A key point isture of 4 MF1 PB at a selling price of 6.0 kEUR obtains

3 MEUR. This is somewhat smaller than the pure PB inn of a selling price of 2.5 kEUR/t, at a price of 2.0 kEUR/tthe pure PB (3.5 MEUR) is lower. Taken the expected

0 MF5 PB (between 2.0 kEUR/t and 2.5 kEUR/t) intois possible to select the ratios in function of their sell-

ing prices wmixture. Thin Table 6 iand in the d(2.5 kEUR/tTable 6, thenor in the dshaded areand more. analysis.

NPV for a 1 t/h facility.

hich yield an equal or more positive result than thisese ratios with their accompanying IRRs are presentedn the shaded area in the case of a price of 2.0 kEUR/tashed box in the case that the maximum selling price

) for the AC of the pure PB is reached. As illustrated by 4 MF1 PB ratio neither appears in the shaded area,ashed box. This ratio would only be comprised in the

a if Table 6 would be expanded to prices of 5.5 kEUR/tThe 4 MF1 PB ratio would only appear in the dashed


Table 7Variables with their protability distribution.

Input variable Protability distribution Most likely value Variance of the distribution

Delivered feed cost Triangular Dependent of the ratio 10%Discount rat 9%Electricity co 0.0Cost of wate 1.5Char output SeActive carbo SeStaff cost/sh 55Total Capita DeLiquid nitrog 2.5

box if Tableimportant tinteresting which mean

5.3. Econom

Anothercapacity, i.epyrolysis). Dowaste/h to Fig. 2). Thisincorporatepower expobling the pinput pyrolysistotal equipmthe total casequently, athe height oto 52% (depsequence thwith 24%.

5.4. Share o

Fig. 4 pexpressed ae Triangularst Triangular r Triangular

(%) (pyrolysis) Triangular n output (%) (activation) Triangular ift Triangular l Investement Triangular en Triangular

Table 8Percentage of Monte Carlo simulation runs that gain a positive NPV. 6 is expanded to contain prices above 6.5 kEUR/t. It iso keep this in mind, as the 4 MF1 PB ratio is potentiallyif higher nitrogen content can be scientically proved,s that these high selling prices can be attained.

ies of scale

important factor affecting the NPV is the processing. the hourly ow ratio of the input material (Qfeed inputubling the processing rate of the AC plant from 1 t2 t waste/h results in higher NPVs (compare Fig. 3 to

is a consequence of the economies of scale that ared in the total equipment cost equation (Eq. (8)). As thenent in Eq. (8) is smaller than one (0.6194 < 1), dou-rocessing capacity (in other words multiplying Qfeedby 2) does not result in a proportional increase of theent cost. Doubling the processing rate thus augments

pital investment with only 57% instead of 100%. Con-lso the total operating costs which partly depend onf the total capital investment increase with only 39%ending on the mix ratio of MF and PB waste). As a con-e break-even selling price of AC decreases on average

f expenditure items in total expenditure

resents the share of the distinct expenditure itemss an average percentage for all ratios of the total dis-

counted exrepresents deviation dand maintetively a shaand a maxideviation dest paymenclosely folloelectricity. payments, lifetime of investmentuid nitrogeSection 5.5

5.5. Monte

As descr100% certaiare uncertathe NPV if texpected toplant are sechange folllikely, a minare performdraws a ran10%725 EUR/kWh 10%

EUR/m3 10%e Table 4 10%e Table 4 10%

kEUR 10%pendent of the ratio 10%

EUR/kg 10%penses (over 20 years). The total capital investmenton average the major share of 42.9% (with a maximumown of 0.9% and a maximum deviation up of 0.7%). Staffnance present the main operating costs with respec-re of 11% (with a maximum deviation down of 0.7%mum deviation up of 0.8%) and 10% (with a maximumown and up of 0.2%) of the total expenses. The inter-ts amount to 8% (0.2%; +0.1%) of the total expenditurewed by insurance, overhead, liquid nitrogen, water andThe xed operating costs (insurance, overhead, interestmaintenance) are considered as unchanging during thethe project (20 years). The impact of the total capital

and the cash ows generated by the staff cost, liq-n, water and electricity on the NPV will be analysed in

by means of Monte Carlo sensitivity analysis.

Carlo sensitivity analysis

ibed in Section 3, results are only valid in the case ofnty of the base case variables. Some variables howeverin by denition, other variables might strongly inuenceheir value changes slightly. Nine main variables that are

affect the economic attractiveness of the productionlected. The variables, listed in Table 7, are allowed to

owing a triangular distribution characterized by a mostimum and a maximum value. Monte Carlo simulationsed, in which each run of the Monte Carlo simulationdom value for each of these variables, between the min-


Table 9Minimal selling price at which the AC should be sold to guarantee a 95% chance on a positive NPV.

Ratio (MF/wood) 4 MF1 PB 3 MF2 PB 2 MF3 PB 1 MF4 PB 0 MF5 PB

Minimal selling price 5.0 kEUR/t 3.7 kEUR/t 2.9 kEUR/t 2.5 kEUR/t 2.5 kEUR/t

imum and distributionues drawn f(10 000 perresult in a N

Fig. 5 illuthe Monte Cfor the 1 t/hrates can be

By meanobtaining aTable 8.

For examsales value If the produmore, 100%summarizein 95% of thscenarios fothe selectedof the ratioMF1 PB at by the MonNPV is highminimal 95

These sdened in tsider uncertBased on thpoint of viewith an evethese resultprice of the

The shadan equal ofthe Monte the dashedfor this ratithe dashed for respectistarts at 5.5

Neverthresent a ratMF-waste awill probabpositive. Ana specialty described in

5.6. Identi

Finally, tis determinfor further iable is dendependent For each vaFig. 6 for a 1by the stand

inpuut vaectedacto

vity =

tive e in tive e cha

for this me in ons o

moointhe in

charhe Ansitivlearl

the w liqui

factpita

clus

asibias bon vant mlity oulatent

A senhe mestmlly en of isticobabof nit

a higfulneeusecans at ng th

the m sensnsiti

pricmaximum value and in accordance with the selected. Each run results in a NPV corresponding to the val-or each of the nine uncertain variables. Numerous runs

ratio in this research) of the Monte Carlo simulationPV distribution.strates this distribution, characterized by the mean ofarlo analysis with their respective standard deviations

processing plant. Similar results for the other input calculated.s of Monte Carlo simulations also the probability of

positive NPV is calculated. The results are listed in

ple, a 1 MF4 PB ratio that yields an AC quality with aof only 2.0 kEUR/t has a 42% chance on a positive NPV.ct is of better quality and can be sold at 2.5 kEUR/t or

of the cases have a positive NPV. The green full line boxs the ratios and selling prices that would yield at leaste cases a positive NPV. These are the most promisingr an AC production facility put into practice. Comparing

cases (full line green box) of Tables 6 and 8 the scenario 2 MF3 PB with a selling price of 2.5 kEUR/t and the 4a selling price of 4.5 kEUR/t are supplementary rejectedte Carlo simulation, because the chance on a negativeer than 5%. The minimal price corresponding to this% chance on a positive NPV is determined (see Table 9).elling prices are somewhat higher than the valueshe base case (see Table 5) because the latter did not con-ainties in the assumed values of the base case variables.ese results one can say that it is, from an economicalw, not interesting to study the 4 MF1 PB and ratiosn higher MF portion. Nevertheless, in order to analyses, one should follow the same consideration (expected

0 MF5 PB ratio) as in Section 5.1.ed area in Table 8 represents the scenarios which yield

more positive NPV than the 0 MF5 PB mixture (fromCarlo simulation, Fig. 5) at a price of 2.0 kEUR/t and in

box the cases where the maximum price (2.5 kEUR/t)o is applied. For the 3 MF2 PB and the 4 MF1 PB ratiobox starts at a sales value of 5.5 kEUR/t and at 6.5 kEUR/tvely. In addition, for the 4 MF1 PB ratio the shaded area

kEUR/t. The conclusions in Section 5.1 thus still hold.eless, we need to keep in mind that these results rep-her worst case scenario, with a zero income from thend rst plant costs. In normal conditions the MF wastely yield a gate fee and will make the projections moreother important fact is the likelihood of producingcarbon of very high added value 4.06.0 kEUR/t (as

Section 5.5).

cation of the key variables

he sensitivity of the NPV to the diverse input variablesed, in order to identify the crucial process parametersnvestigation. The sensitivity of the NPV for a given vari-ed as the extent to which the variability of the NPV isto the variability of the variable under consideration.

of the an inpthe selevery f

sensiti

A posiincreasa nega(1) Thfactors0.55. Tincreasdeviatithe 3rdA key pprice, tAC andyield, tage secosts cfee forwater,sitivitytotal ca

6. Con

A feof AC hbased differefeasibiby calcinvestmof AC. mine tthe inv

Reaductiopessimwill prration sold ator useity to ra signicarbondoubli24% of

Thevery sesellingriable of Table 7 an average NPV sensitivity is given in

t/h facility. The coefcients on the graph are normalizedard deviation of the output and the standard deviation

Future rcreate a mat and not in actual euros. The higher the coefcient ofriable (the longer the bar), the greater the impact that

variable has on the NPV. The variability of the NPV forr can be calculated by using Eq. (10).

NPV/NPVvariable/variable

(10)

value (the bar extending to the right) means that anthe variable leads to an increasing NPV. In the case ofsign the NPV decreases by an increase of the variable.r and AC yield are on average the most determininghe NPV variability, both have an average sensitivity ofeans that for every k fraction of a standard deviation

char/AC yield, the NPV will increase by 0.55k standardf the NPV. (2) The total capital investment is on average

st important variable in declaring the NPVs sensitivity. is that, depending on the mix ratio and the AC sellingvestment cost is sometimes more important than the

yield, but the three main variables are always the charC yield and investment expenditure. The negative aver-ity results of the discount rate and the delivered feedy indicate that a lower discount rate and a higher gateaste respectively result in a higher NPV. (3) The cost of

d nitrogen, electricity and staff have also negative sen-ors. However these are almost negligible relative to thel investment, discount rate, char and AC yield.

ion

lity study to process MF and PB waste for the productioneen performed. A preliminary process design has beenrious literature sources, for an input feed of 1 t/h andixing ratios of the two waste products. The economicf the preliminary process design has been investigated,ing the NPV and IRR of the cash ows incurred by an

in a pyrolysis and activation plant for the productionsitivity analysis has been performed in order to deter-ost crucial variables that inuence the protability ofent.ncouraging results are obtained for a protable pro-AC, as the current assumptions start from a rather

scenario: e.g. a zero gate fee for the MF waste (whichly be higher in practice). Besides that, the in situ incorpo-rogen can result in a high quality product which can beh price or even in a niche market. In addition, the valuess of the AC production plant is enhanced by its abil-

two waste streams. Also the processing capacity playst role. A larger manufacturing plant is able to producea lower cost despite the higher initial investment. Bye input rate to 2 t/h (dry matter) a reduction of averageinimal selling price is obtained.

itivity analysis reveals that the AC plant economies areve to the investment cost, the product yield and the ACe which is an indication for product quality.

esearch needs to focus on these prime properties torketable high value product.


References

[1] P. Girods, A. Dufour, Y. Rogaume, C. Rogaume, A. Zoulalian, Pyrolysis of woodwaste containing urea-formaldehyde and melamine-formaldehyde resins, J.Anal. Appl. Pyrolysis 81 (2008) 113120.

[2] P. Girods, A. Dufour, Y. Rogaume, C. Rogaume, A. Zoulalian, Thermal removalof nitrogen species from wood waste containing urea formaldehyde andmelamine formaldehyde resins, J. Hazard. Mater. 159 (2008) 210221.

[3] P. Girods, A. Dufour, Y. Rogaume, C. Rogaume, A. Zoulalian, Comparison of gasi-cation and pyrolysis of thermal pre-treated wood board waste, J. Anal. Appl.Pyrolysis 85 (2009) 171183.

[4] T.J. Bandosz, Surface chemistry of carbon materials, in: P. Serp, J.L. Figueiredo(Eds.), Carbon Materials for Catalysis, John Wiley & Sons Inc., New Jersey, 2009,pp. 4592.

[5] H. Marsh, F. Rodriguez-Reinoso, Activated Carbon, Elsevier Ltd., London, 2006.[6] J.A. Menndes-Dias, I. Martn-Gulln, Types of carbon adsorbents and their

production, in: T.J. Bandosz (Ed.), Activated Carbon Surfaces in EnvironmentalRemediation, Elsevier Ltd., London, 2006, pp. 148.

[7] P. Girods, A. Dufour, V. Fierro, Y. Rogaume, C. Rogaume, A. Zoulalian, A. Celzard,Activated carbons prepared from wood particleboard wastes: characterisationand phenol adsorption capacities, J. Hazard. Mater. 166 (2009) 491501.

[8] Ng. Chilton, W.E. Marshall, R.M. Roa, R.R. Bansode, J.N. Losso, Activated carbonfrom pecan shell: process description and economic analysis, Ind. Crops Prod.17 (2003) 209217.

[9] T.J. Bandosz, Surface chemistry of activated carbons and its characterization,in: T.J. Bandosz (Ed.), Activated Carbon Surfaces in Environmental Remediation,Elsevier Ltd., London, 2006, pp. 159230.

[10] Infomil, Luchtemissie beperkende technieken Adsorptie actiefkool/Actief kool ltratie/Koollter, Available from http://www.infomil.nl/onderwerpen/klimaat-lucht/luchtemissie/factsheets/overzicht-factsheets/adsorptie-actief/, 2011 (accessed on 6 January 2011).

[11] I.M. Lima, A. McAloon, A.A. Boateng, Activated carbon from boiler litter: processdescriptio

[12] A.A. Boattransfer w

[13] K.K.H. ChscaffoldinEng. J. 10

[14] B. LemmeGids luch

[15] M.V. Biezeration p

[16] J.H. Lorie229239.

[17] P. VernimTheory an

[18] C.T. Horning, third

[19] T. KuppeT. Thewycoppice 2736274

[20] T. ThewyInt. J. Phy

[21] S. Ocheleoden, Dep

[22] A.V. Bridpower prRenew. S

[23] M.S. Peters, K.D. Timmerhaus, R.E. West, Plant Design and Economics for Chem-ical Engineers, fth ed., McGraw-Hill, New York, 2004.

[24] European Central Bank, Statistical Data Warehouse, Available fromhttp://sdw.ecb.europa.eu/home.do?chart=t1.9, 2010 (accessed on 16 July2010).

[25] Chemical Engineering, 2010, Available from http://www.che.com/ (accessedon 20 April 2010).

[26] Dutch Association of Cost Engineers, DACE-prijzenboekje. twenty-seventh ed.Reed Business BV, Doetinchem, 2009.

[27] L. Yassin, P. Lettieri, S.J.R. Simons, A. German, Techno-economic performanceof energy-from-waste uidized bed combustion and gasication processes inthe UK context, Chem. Eng. J. 146 (2009) 315327.

[28] E. Henrich, N. Dahmen, E. Dinjus, Cost estimate for biosynfuel production viabiosyncrude gasication, Biofuels Bioprod. Bioref. 3 (2009) 2841.

[29] L. Tock, M. Gassner, F. Marchal, Thermochemical production of liquid fuelsfrom biomass: thermo-economic modeling, process design and process inte-gration analysis, Biomass Bioenergy 43 (2010) 18381854.

[30] M. Gassner, F. Marchal, Thermo-economic process model for thermochem-ical production of synthetic natural gas (SNG) from lignocellulosic biomass,Biomass Bioenergy 33 (2009) 15871604.

[31] Belgostat, Rentetarieven op nieuwe kredieten Eurogebied, 2010,Available from http://www.nbb.be/belgostat/PresentationLinker?TableId=248000024&Lang=N (accessed on 4 November 2010).

[32] A.C. Caputo, M. Palumbo, P.M. Pelagagge, F. Scacchia, Economics of biomassenergy utilization in combustion and gasication plants: effects of logisticvariables, Biomass Bioenergy 28 (2005) 3551.

[33] M.N. Islam, F.N. Ani, Techno-economics of rice husk pyrolysis, conversionwith catalytic treatment to produce liquid fuel, Bioresour. Technol. 73 (2000)6775.

[34] G.V.C. Peacocke, A.V. Bridgwater, J.G. Brammer, Techno-economic assessmentof power production from the Wellman Process Engineering Ltd and BTGfast pyrolysis processes, in: A.V. Bridgwater, D.G.B. Boocock (Eds.), Science

herm518.K. Koste tir888

Econilableen/acostat,p://ap

(acceaggiolysis os and c

Diebotion omaa, dboo

. ReeddgwatversieckhaA.V. Bckie A. Polagions u125n and cost of Production, Biomass Bioenergy 32 (2008) 568572.eng, P.V. Barr, A thermal model for the rotary kiln including heatithin the bed, Int. J. Heat Mass Transfer 39 (1996) 21312147.

oy, J.P. Barford, G. McKay, Production of activated carbon from bamboog wasteprocess design, evaluation, and sensitivity analysis, Chem.9 (2005) 147165.ns, H. Elslander, J. Ceulemans, K. Peys, H. Van Rompaey, D. Huybrechts,tzuiveringstechnieken, Academia press, Gent, 2004.ma, J.R. San Cristbal, Investment criteria for the selection of cogen-lants-a state of the art review, Appl. Therm. Eng. 26 (2006) 583588., L.J. Savage, Three problems in rationing capital, J. Bus. 28 (1955)

men, P. Quiry, Y. Le Fur, M. Dillocchio, A. Salvi, Corporate Finance:d Practice, John Wiley & Sons Ltd., West Sussex, 2005.

gren, A. Bhimani, S.M. Datar, G. Foster, Management and Cost Account- ed., Pearson Education limited, Essex, 2003.ns, T. Cornelissen, R. Carleer, J. Yperman, S. Schreurs, M. Jans,s, Economic assessment of ash co-pyrolysis of short rotationand biopolymer waste streams, J. Environ. Manage. 91 (2010)7.

s, T. Kuppens, Economics of willow pyrolysis after phytoextraction,toremediation 10 (2008) 561583.n, B. Putzeijs, Milieubeleidskosten begrippen en berekeningsmeth-artement Leefmilieu, Natuur en energie, Brussels, 2008.

gwater, A.J. Toft, J.G. Brammer, A techno-economic comparison ofoduction by biomass fast pyrolyses with gasicaton and combustion,ust. Energy Rev. 6 (2002) 181248.

in T178

[35] D.Cwa875

[36] FPSAva/lon

[37] Eurhtt=en

[38] P. Banaysi

[39] J.P.ducOasHan

[40] T.RBriCon

[41] D. Bin: Bla

[42] B.Lopt105al and Chemical Biomass Conversion, CPL Press, Newbury, 2004, pp.02., E.L.K. Mui, K.S.T. Lau, G. McKay, Production of activated carbons frome process design and economical analysis, Waste Manage. 24 (2004).omy, Arbeidskosten en gewerkte uren per sector (2004), 2010,

from http://statbel.fgov.be/nl/statistieken/cijfers/arbeid leventiviteit/index.jsp (accessed on 4 November 2010).

Unit labour cost Annual data, 2010, Available frompsso.eurostat.ec.europa.eu/nui/show.do?dataset=nama aux ulc&langssed on 4 November 2010)., M. Baratieri, A. Gasparella, G.A. Longo, Energy and environmentalf an innovative system based on municipal solid waste (MSW) pyrol-ombined cycle, Appl. Therm. Eng. 28 (2008) 136144.ld, A.V. Bridgwater, Overview of fast pyrolysis of biomass for the pro-f liquid fuels, in: A.V. Bridgwater, S. Czernik, J. Diebold, D. Meier, A.C. Peacocke, J. Piskorz, D. Radlein (Eds.), Fast Pyrolysis of Biomass: Ak, CLP Press, Newbury, 2003, pp. 1432., S. Gaur, The high heat of fast pyrolysis for large particles, in: A.V.er, D.G.B. Boocock (Eds.), Developments in Thermochemical Biomasson, Blackie Academic and Professional, London, 1997, pp. 97103.m, R.G. Graham, Economic assessment of a wood fast pyrolysis plant,ridgwater (Ed.), Advances in Thermochemical Biomass Conversion,cademic and Professional, London, 1994, pp. 13141324.ye, K.T. Hodgson, P.C. Malte, An economic analysis of bio-energysing thinning from overstocked forests, Biomass Bioenergy 31 (2007).

Activated carbon from co-pyrolysis of particle board and melamine (urea) formaldehyde resin: A techno-economic evaluation1 Introduction2 Process design3 Economical feasibility model4 Model assumptions4.1 Total capital investment4.2 Expenditure4.3 Revenues

5 Results and discussion5.1 Economic evaluation of the base case5.2 Economic impact of the nitrogen content of the AC5.3 Economies of scale5.4 Share of expenditure items in total expenditure5.5 Monte Carlo sensitivity analysis5.6 Identification of the key variables

6 ConclusionReferences

activated carbon from co-pyrolysis of particle board and melamine (urea)

Documents

waste resin

wood waste

waste streams

cactivated carbon

specialty carbon

dispersion of carbon

production costs

net present value