Thermal deactivation of biogenic and fossil fuels: Experimental investigation and modeling approaches

Available online at www.sciencedirect.com Available online at www.sciencedirect.com

ScienceDirect ScienceDirect

Energyonline Procedia 00 (2018) 000–000 Available onlineatat www.sciencedirect.com Available www.sciencedirect.com Energy Procedia 00 (2018) 000–000

ScienceDirect ScienceDirect

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

Energy Procedia 158 Energy Procedia 00(2019) (2017)2025–2030 000–000 www.elsevier.com/locate/procedia

10th International Conference on Applied Energy (ICAE2018), 22-25 August 2018, Hong Kong, 10th International Conference on Applied Energy China(ICAE2018), 22-25 August 2018, Hong Kong, China

Thermal deactivation of biogenic and fossil fuels: Experimental Thermal deactivation of biogenic fossil fuels: The 15th International Symposiumand on District Heating andExperimental Cooling investigation and modeling approaches investigation and modeling approaches Assessing the feasibility of using the heat demand-outdoor T. Kreitzberga,a, *, C. Bormannaa, S. Pielsticker, O. Hatzfeldaa, R. Kneeraa T. Kreitzberg *, C. Bormann , S. Pielsticker, O. Hatzfeld , R. Kneerforecast temperature function for a long-term district heat demand Institute of Heat and Mass Transfer (WSA), RWTH Aachen University, Augustinerbach 6, D-52056 Aachen, Germany a a

Institute of Heat and Mass Transfer (WSA), RWTH Aachen University, Augustinerbach 6, D-52056 Aachen, Germany

I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc

Abstract a IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal Abstract b Veolia Recherche Avenue Dreyfous Daniel, annealing”, 78520 Limay,which Francedescribes the influence of In the presentc study, the phenomenon known&asInnovation, “thermal 291 deactivation” or “thermal Département Systèmes Énergétiques et Environnement IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes,gap France In thetreatment present study, phenomenon knownreactions, as “thermal deactivation”Aorliterature “thermalreview annealing”, which describes the heat on thethe kinetics of gas-solid is investigated. revealed a distinct in influence the field of heat treatment on the kinetics of gas-solid reactions, is investigated. A literature review revealed a distinct gap in the fieldthe of quantitative analysis of thermal deactivation influencing low rank fuels reacting under gasification conditions. Therefore, quantitative thermal influencing low rank fuels carbon reactingdioxide under was gasification Therefore, the reactivity of analysis Rhenish of lignite and deactivation torrefied poplar wood particles towards examinedconditions. after the process of heat reactivity of Rhenish lignite and torrefied poplar wood particles towards carbon dioxide was examined after the process of heat treatment. Abstract For both the heat treatment and the gasification step consecutively, a small-scale fluidized bed reactor was utilized. treatment. the heat treatmentwas and achieved the gasification step consecutively, reactor was utilized. AssessmentFor of both thermal deactivation by varying heat treatmenta small-scale temperaturefluidized between bed 1023–1173 K and heat Assessment of between thermal deactivation wasgases achieved by varyingusing heat atreatment temperature between 1023–1173 Kdeduce and heat treatment time 0–1800 s. Flue were analyzed Fourier-transform infrared spectrometer to District heating networks are commonly addressed in the literature as one of the most effective solutions for decreasingthe the treatment time between 0–1800 s. Flue gases were of analyzed usinggas a Fourier-transform infrared spectrometer to deduce the chemical reaction rate from the the temporal evolution the systems product species. The experimental results were through reproduced by greenhouse gas emissions from building sector. These require high investments which are returned the heat th chemical reactionn rate from the temporal of distributed the productactivation gas species. The to experimental results were reproduced by -order power law and aevolution modeland with energies for implementation application sales. Dueoftoanthe changed climate conditions building renovation policies, heatpropose demandparameters in the future could decrease, power law and a model types. with distributed activation energies to propose parameters for implementation application of an nth-order in gasification models and to compare these prolonging the investment return period. in gasification models and to compare these model types. The main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand Copyright © 2018 Elsevier Ltd. All rights reserved. forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a caseth study. The district is consisted of 665 © 2019 The Published by Elsevier Ltd. Copyright ©Authors. 2018 Elsevier Ltd. Allresponsibility rights reserved. International Applied Selection and peer-review under of scientificThree committee the 10 (low, buildings that access vary inarticle both under construction andthetypology. weatherofscenarios medium,Conference high) and on three district This is an open the CC period BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of the scientific committee of the 10th International Conference on Applied Energy (ICAE2018). Peer-review responsibility of the scientific of ICAE2018 The 10th the International Conference on Applied Energy. renovation under scenarios were developed (shallow,committee intermediate, deep). To– estimate error, obtained heat demand values were Energy (ICAE2018). compared with results from a dynamic heat demand model, previously developed and validated by the authors. Keywords: Thermal deactivation / annealing; CO2 gasification; gas-solid reaction kinetics; fluidized bed The results showed that when only weather change is considered, the margin of error could be acceptable for some applications Keywords: Thermal deactivation / annealing; CO2 gasification; gas-solid reaction kinetics; fluidized bed (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). 1.The Introduction value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the 1. Introduction decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and Efficientscenarios technicalconsidered). implementation any imaginable process increased is achieved by applying profound and detailed renovation On the of other hand, function intercept for 7.8-12.7% per decade (depending on the Efficient technical implementation ofcould anychemical process is achieved by applying profound and detailed knowledge about theThe governing fundamental physical mechanisms. When comes to theconsidered, utilization of coupled scenarios). values suggested beimaginable used toand modify the function parameters forit the scenarios and knowledge about the governing fundamental chemical and physical When it deep comesunderstanding to the utilization of coal and the biogenic fuels, which most commonly is a combustion or mechanisms. gasification process, of the improve accuracy of heat demand estimations.

coal and biogenic fuels, which most commonly is a combustion or gasification process, deep understanding of the © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. * Corresponding author. Tel.: +49-241-80-97530; fax: +49-241-80-92143. * E-mail Corresponding Tel.: +49-241-80-97530; fax: +49-241-80-92143. address:author. [email protected] Keywords: Heat demand; Forecast; Climate change E-mail address: [email protected]

1876-6102 Copyright © 2018 Elsevier Ltd. All rights reserved. 1876-6102 Copyright © 2018 Elsevier Ltd. All of rights reserved. committee of the 10th International Conference on Applied Energy (ICAE2018). Selection and peer-review under responsibility the scientific Selection and peer-review under responsibility of the scientific committee of the 10th International Conference on Applied Energy (ICAE2018). 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. 1876-6102 © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the scientific committee of ICAE2018 – The 10th International Conference on Applied Energy. 10.1016/j.egypro.2019.01.468

2026 2

T. Kreitzberg et al. / Energy Procedia 158 (2019) 2025–2030 T. Kreitzberg/ Energy Procedia 00 (2018) 000–000

heterogeneous gas-solid reactions plays a crucial role. One way to technically realize coal or biomass gasification is using an entrained flow gasifier. The dimensions of the reactor considerably influence residence times of the solid fuel particles and the reacting gases. For designing an entrained flow gasifier it is therefore essential to have an accurate estimation of the residence time needed for complete fuel conversion, which can only be provided by close investigation of gas-solid reactions. As for example Tremel and Spliethoff [1] point out, thermal deactivation is an important aspect of those gas-solid reactions in entrained flow gasification and should be considered for the design of such reactors. In literature, the term “thermal deactivation” as well as the synonymously used term “thermal annealing” subsume a combination of multiple effects occurring due to heat treatment of fuels, resulting in a significant reduction of reaction rates [2, 3]. This loss in reactivity substantially increases time spans needed for complete carbon conversion and thus influences dimensioning of industrial combustion and gasification apparatuses. While there are many studies addressing thermal deactivation in connection with oxidation reactions, investigations with focus on the effects of thermal deactivation on gasification reactions with CO2 or H2O are scarce [5, 8, 9]. Furthermore, the literature mentioned does not cover quantitative analysis of the impact of thermal deactivation on the gasification reaction rate of low rank coals and biogenic fuels. The present paper aims to fill the identified research gap by conducting a quantitative study, investigating the effect of thermal deactivation on the heterogeneous gasification reaction with CO2 in a lab scale fluidized bed reactor (FBR). The FBR system is optimized for fast response behavior and is used to investigate the char conversion process with Rhenish lignite and a biogenic fuel made of torrefied poplar wood. To evaluate thermal deactivation mathematically, an nth-order approach as well as an approach with distributed activation energies (DAE) for the deactivation step is applied to the measured data. 2. Experimental investigation The experimental setup used in this work was previously described in detail by the authors and was successfully applied for measurement of combustion and gasification reaction kinetics of different pulverized fuels [13-15]. Main characteristics of the system are presented in the following: The setup consists of three major components: A gas feeding system with thermal mass flow controllers, a small-scale fluidized bed reactor (FBR) and an Fouriertransform infrared (FTIR) spectrometer to analyze exhaust gas compositions. Small batches of pulverized solid fuel (coal or biomass) are fed to a fluidized bed consisting of aluminum oxide (Al2O3) particles. 2.1. Experimental setup The fluidized bed reactor shown in Fig. 1 consists of two coaxial ceramic pipes mounted in a stainless steel reactor head. All ceramic parts of the reactor are located in an electrically heated furnace which can be operated up to 1573 K. The fluidizing gas is fed in through the reactor head and directed downwards through the annular gap between the two pipes to the bottom of the reactor, while it heats up to the surrounding oven temperature. The gas fluidizes the bed within the inner reactor pipe (d = 55 mm) through a sintered silica glass distributor plate. This distributor plate with a typical pore size between 40 and 100 µm creates a homogeneous inflow and is impermeable for the Al2O3 bed particles, for which a mesh size fraction of 100–180 µm is used. The bed temperature is held constant for each individual experiment and measured by an S-type thermocouple placed in a ceramic shielding. Small amounts of fuel (5–10 mg) are introduced through a lock and a vertical ceramic pipe (d = 6 mm) into the reactor. The particle heating rate has been analytically approximated to be in the order of 104 K/s [13]. All gases, including the product gases of the reaction or devolatilization, flow through a sampling line and a filter into the measurement cell of the FTIR spectrometer. To reduce entrainment of fuel and bed particles, the opening of the sampling gas pipe is covered with a porous tip, consisting of the same material as the distributor plate. The employed Ansyco Gasmet DX-2000 spectrometer is analyzing in the mid-IR wave number region from 600 to 4200 cm-1 with a spectral resolution of 8 cm-1. The effective sampling frequency lies at 0.57 Hz.



T. Kreitzberg et al. / Energy Procedia 158 (2019) 2025–2030 T. Kreitzberg/ Energy Procedia 00 (2018) 000–000

2027 3

Fuel dosage Inert-gas injection

Fuel feed pipe

Exhaust gas to FTIR spectrometer

Thermocouple tube

Table 1: Properties of Rhenish lignite (RL) and torrefied biomass (TB) Unit

Sealings Feed-gas

L=750

Ceramic packing Fluidized bed Distributor

Lbed=70

Basis a

TB

RL

C H N S O

wt.-% wt.-% wt.-% wt.-% wt.-%

daf daf daf daf daf

54.2 5.78 0.21 0.02 29.8

68.9 4.81 1.28 0.27 24.8

ash water volatiles

wt.-% wt.-% wt.-%

dryb aac daf

0.59 5.44 78.15

5.78 5.13 52.1

a

daf: moisture and ash free dry: moisture free c aa: as analyzed

Electric heating

b

d=55

Fig. 1: Scheme of the fluidized bed reactor (dimensions in mm)

2.2. Fuel preparation and characterization The investigation is undertaken with torrefied biomass (TB) from poplar wood and Rhenish lignite (RL). The raw poplar wood is torrefied for 15 min at 553 K under inert gas atmosphere. After acquisition, both fuels are ground and sieved to achieve well defined particle size distributions. For all further steps only the mesh size fraction 125– 160 µm is used. Table 1 lists the results of proximate and ultimate analysis for the raw fuels. 2.3. Experimental procedure and data evaluation For investigating the influence of thermal pre-treatment on the fuels conversion rate with respect to CO2gasification reaction, the following experimental approach is adopted. It shall be noted that in this work all experiments and model predictions are assumed to be controlled by regime I conditions, i.e. mass-transfer effects such as pore or film diffusion have no impact on conversion. Therefore, temperatures have been kept moderately low and small particles are used. First, a gas flow of 100 slph consisting of pure nitrogen is fed into the reactor, providing an inert gas atmosphere in the fluidized bed. Second, batches of solid fuel particles (RL or TB) are introduced into the reactor via the fuel feed pipe (cf. Fig. 1). When reaching the fluidized bed, the first phase of the experiment begins, where particles are kept in an inert gas atmosphere at a defined temperature and hold time thold. During this phase, volatile matter of the particles, such as carbon monoxide, is released and exits the fluidized bed through the exhaust pipe. After expiration of thold, a reactive gas, in this case 20 Vol.-% carbon dioxide, is added to the feed-gas. The carbonaceous particles start to react heterogeneously with CO2 according to Eq. 1 and the second phase of the experiment begins. C + CO2 → 2 CO

(1)

By means of the measured temporal evolution of the produced CO in the exhaust gas the carbon conversion rate can be derived from a carbon balance of the fluidized bed by analyzing the exhaust gas stream via quantitative FTIR spectroscopy. The conversion rate is then calculated according to Eq. 2: ∂𝑋𝑋exp 𝑀𝑀c = ∙ 𝑛𝑛̇ tot (𝑡𝑡) ∙ 𝑐𝑐reac (𝑡𝑡) ∂𝑡𝑡 𝑚𝑚c,0

(2)

T. Kreitzberg/ Energy Procedia 00 (2018) 000–000 T. Kreitzberg et al. / Energy Procedia 158 (2019) 2025–2030

4 2028

With terms for the total molar flow leaving the reactor ṅtot and reaction product concentration difference between reactor inlet and outlet creac. In Fig. 2 the temporal evolution of carbon monoxide and carbon dioxide for one exemplified experiment of Rhenish lignite at 1123 K and a hold time thold = 600 s is shown. The two distinct experimental phases under inert gas atmosphere (phase 1) and reactive gas atmosphere with 20 Vol.-% (phase 2) are clearly visible. The experimental carbon conversion curve Xexp is calculated by integrating Eq. 2 with respect to time. The measured curve is then approximated by a model. In this case, the random pore model (RPM, Eq. 3, [16]) is applied, which can be interpreted using two parameters: the characteristic reaction rate r and a structural parameter Ψ. ∂𝑋𝑋mod = 𝑟𝑟 ∙ (1 − 𝑋𝑋mod ) ∙ �1 − 𝛹𝛹 ∙ ln(1 − 𝑋𝑋mod ) ∂𝑡𝑡

(3)

The characteristic rate r is determined iteratively by fitting the model prediction to the experimental data of one single experiment. The structural parameter Ψ is found individually for each fuel by minimizing the deviation of the model and the experimental data for all conducted experiments. Thus, values of Ψ = 0.68 (RL) and Ψ = 1.34 (TB) are computed. A detailed overview of this evaluation routine is given in [13]. The experimental and fitted carbon conversion rates in differential and integrated form are also shown in Fig. 2 (right). 5 40

3000

30

CO CO2

2000

20

1000

10 0

0 400

600 Time t [s]

1

800

1000

1200

4

0.8 ∂Xexp/∂t ∂Xmod/∂t Xexp

3

0.6 X [-]

4000

200

Phase 2, het. reaction

CO peak from reaction ∂X/∂t [s-1 × 1000]

CO [ppm]

CO peak from pyrolysis

0

Phase 1, hold time thold

Phase 2, het. reaction

CO2 [Vol.-%]

Phase 1, hold time thold

Xmod

2

0.4

1 0

0.2

0

200

400

600 Time t [s]

800

1000

0 1200

Fig. 2: Measured concentration profiles for CO2 and CO (left); Experimentally measured (exp) and fitted (mod) carbon conversion rates in differential and integral form (right); single deactivation experiment with 5 mg of RL at 1123K.

3. Results and Discussion By varying the two parameters hold time thold from 0 to 1800 s and bed temperature T from 1023 to 1173 K the results (exp) illustrated in Fig. 3 are obtained by following the evaluation procedure described under section 2.3. Each data point shown there depicts the arithmetic mean of at least five repeated experiments with error bars indicating the corresponding standard deviation (± σ). Both fuels clearly show an exponentially decreasing trend of reaction rates with increasing hold times thold. Comparing hold times of 180 and 1800 s at 1023 K, reaction rates are reduced by about 18 % for Rhenish lignite. This decrease is further intensified by a rise in temperature: 1123 K, rates are reduced by about 49 %. The relative change in rates of torrefied biomass lies at 21 % for 1023 K and at 30 % for 1123 K at the same interval of hold times. The two plots also reveal that the most significant drop in reactivity occurs for hold times thold < 600 s for the investigated range of temperatures. Overall, Rhenish lignite shows higher reaction rates than TB, comparing same reaction temperatures and hold times.



T. Kreitzberg/ Kreitzberg et al. / Energy Procedia 158 000–000 (2019) 2025–2030 T. Energy Procedia 00 (2018)

20295

𝐸𝐸A 𝑚𝑚 � ∙ 𝑝𝑝CO 2 𝑅𝑅𝑅𝑅 𝐸𝐸A 𝑚𝑚 𝑟𝑟d = (1 − 𝜉𝜉 ) ∙ 𝐴𝐴d ∙ exp �− � ∙ 𝑝𝑝CO 2 𝑅𝑅𝑅𝑅 𝐸𝐸A 𝑚𝑚 𝑟𝑟 = [𝐴𝐴𝑑𝑑 + 𝜉𝜉(𝐴𝐴f − 𝐴𝐴d )] ∙ 𝐴𝐴d ∙ exp �− � ∙ 𝑝𝑝CO 2 𝑅𝑅𝑅𝑅

(4)

𝑟𝑟f = 𝜉𝜉 ∙ 𝐴𝐴f ∙ exp �−

(5) (6)

For a mathematical description of the experimental results by means of an appropriate reaction model, two different mechanisms are considered in this work: A three-step mechanism proposed by Senneca et al. [3] as well as a two-step model from Hurt et al. [10]. The three-step mechanism allows active carbon sites (fresh) to be converted to less reactive sites (deactivated) by a deactivation reaction rD. Both types of sites can interact with a reactive gas (here CO2) forming the gaseous reaction products (here CO) with reaction rates rf (fresh) and rd (deactivated), respectively. Rates of these two reactions are expressed by mth-order Arrhenius equations with a value of m = 0.4 (Eqs. 4 and 5). Furthermore, both reactions are dependent on the dimensionless variable ξ, representing the share of remaining fresh sites. The overall carbon conversion rate r is determined by the summation of the reaction rates rf and rd (cf. Eq. 6) and is shifted from rf towards rd during the extent of the deactivation reaction. In contrast to this model, the two-step mechanism doesn't allow deactivated sites to react heterogeneously further on (rd = 0). In other words, the reduction of the total carbon conversion rate r is based on a reduced amount of carbon sites. The share of active sites ξ is given by two different equations, depending on the chosen mathematical expression for rD. Senneca et al. adopt an nth-order approach, resulting in Eq. 7. Hurt et al. chose a first order approach with distributed activation energies (DAE), which leads to Eq. 8: −1

(7)

�𝑛𝑛−1 𝐸𝐸D 𝜉𝜉nth = �(𝑛𝑛 − 1)𝐴𝐴D ∙ exp �− � ∙ 𝑡𝑡hold � 𝑅𝑅𝑅𝑅 ∞ 𝐸𝐸A,D 𝜉𝜉DAE = � �−𝐴𝐴D ∙ exp �− � ∙ 𝑡𝑡hold � ∙ 𝐷𝐷(𝐸𝐸A,D ) ∙ d𝐸𝐸A,D 𝑅𝑅𝑅𝑅 0

(8)

Here, D(EA,D) represents a log-normal distribution of activation energies with the parameters µ and σ. By fitting both approaches (Eqs. 7–8) for ξ in combination with the three-step mechanism (Eqs. 4–6) to the measured data, the parameters displayed in Tab. 2 are obtained. Both fits resemble the experimental results almost equally well, the calculated residuals are close together for both models and fuels. The difference between the two modelling approaches appears in the determined kinetic parameters: For example, the values of Af and Ad for TB are notably lower than those for RL in the nth-order power law, while they are in the same order of magnitude for both fuels in the DAE approach. Also, the frequency factors of deactivation AD differ considerably. As Senneca et al. point out, the differences between the two model types are likely to be reconciled by a compensation effect between different kinetic parameters [4, 11], which explains the distinct parameter sets. Due to its simplicity, using an nth-order

RL

Reaction rate r [s -1 ]

0.03 0.025 0.02 0.015

0.012 1023 K, exp 1073 K, exp 1123 K, exp 1023 K, nth-order 1073 K, nth-order 1123 K, nth-order 1023 K, DAE 1073 K, DAE 1123 K, DAE

0.01

1023 K, exp 1123 K, exp 1173 K, exp 1023 K, nth-order 1123 K, nth-order 1173 K, nth-order 1023 K, DAE 1123 K, DAE 1173 K, DAE

0.008 0.006 0.004 0.002

0.005 0

TB TB

0.01

Reaction rate r [s-1 ]

0.035

0

300

600

900 1200 1500 Hold time thold [s]

1800

2100

0

0

300

600

900 1200 1500 Hold time thold [s]

1800

2100

Fig. 3: Experimental (exp) and modeled (DAE and nth-order) reaction rates of Rhenish lignite (RL) and torrefied biomass (TB) as a function of thold.

T. Kreitzberg et al. / Energy Procedia 158 (2019) 2025–2030 T. Kreitzberg/ Energy Procedia 00 (2018) 000–000

2030 6

approach for modeling thermal deactivation might thus be a useful alternative to the DAE approach [9, 11]. In both models the calculated values of AD are remarkably higher for TB than for RL, which indicates a faster deactivation process of biomass. Equally, the ratio between Af and Ad of TB is higher than the same ratio of RL: For the nth-order approach a ratio of around 6.8 (RL) and 57.0 (TB) is calculated. This indicates a higher tendency of the biogenic fuel to be thermally deactivated, which corresponds to earlier findings in literature [6, 7]. Table 2: Fitted parameters for the nth-order and DAE model

nth-order RL TB DAE RL TB

Af [1/barms] 5.31·108 1.46·107 Af [1/barms] 3.10·108 1.87·108

Ad [1/barms] 7.78·107 2.57·105 Ad [1/barms] 5.35·107 3.28·106

EA [J/mol] 2.13·105 1.90·105 EA [J/mol] 2.08·105 2.04·105

AD [1/s] 6.12·1016 1.01·1021 AD [1/s] 4.52·1014 3.60·1022

EA,D [J/mol] 3.94·105 4.77·105 eµ [J/mol] 3.54·105 4.68·105

n [-] 2.60 5.25 eσ[J/mol] 1.05·103 1.13·103

4. Conclusion In this work, the thermally induced decrease of reactivity of Rhenish lignite and torrefied poplar wood towards CO2 was experimentally investigated and quantified. Measured reaction rates are well reproduced by a three-step mechanism. In this model, the deactivation reaction can be formulated by both an nth-order power law as well as a distributed activation energy rate expression without remarkable differences in the quality of the fit. Model parameters found are listed to close the gap in pertinent literature of deactivation kinetics of low rank fuels. Activation energies of the deactivation reaction appear to be distinctly higher than those of the gasification reaction, leading to a strong increase in deactivation rates with increasing temperature. Furthermore, it is shown that the biogenic fuel revealed a stronger tendency to deactivate than the fossil one (higher ratio of Af/Ad). Acknowledgements The authors would like to express their gratitude to the Hans Hermann Voss Foundation for funding Thobias Kreitzberg within the project “ReacMic” (OPSF445). The authors also thank the German Research Foundation (DFG) for funding Stefan Pielsticker and Oliver Hatzfeld within the SFB/Transregio 129 “Oxyflame”. References [1] A. Tremel, H. Spliethoff, Fuel 103 (2013) 663–671. [2] E. M. Suuberg, M. Wjtowicz, J. M. Calo, Carbon 27 (1989) 431– 440. [3] O. Senneca, P. Russo, P. Salatino, S. Masi, Carbon 35 (1997) 141–151. [4] O. Senneca, P. Salatino, S. Masi, Fuel 77 (1998) 1483–1493. [5] E. Bar-Ziv, A. Zaida, P. Salatino, O. Senneca, Proc. Combust. Inst. 28 (2000) 2369–2374. [6] H.-S. Shim, R. H. Hurt, Energy Fuels 14 (2000) 340–348. [7] A. Zolin, A. Jensen, K. Dam-Johansen, Proc. Combust. Inst. 28 (2000) 2181–2188. [8] O. Senneca, P. Salatino, Proc. Combust. Inst. 29 (2002) 485–493. [9] P. Salatino, O. Senneca, S. Masi, Energy Fuels 13 (1999) 1154–1159. [10] R. Hurt, J.-K. Sun, M. Lunden, Combust. Flame 113 (1998) 181–197. [11] O. Senneca, P. Salatino, Proc. Combust. Inst. 33 (2011) 1763–1770. [12] O. Senneca, F. Scala, R. Chirone, P. Salatino, Fuel 201 (2017) 65–80. [13] H. Haustein, T. Kreitzberg, B. Gövert, A. Massmeyer, R. Kneer, Fuel 158 (2015) 263–269. [14] T. Kreitzberg, H. Haustein, B. Gövert, R. Kneer, J. Energy Resour. Technol. 138 (2016) 042207-0–42207-7. [15] B. Gövert, S. Pielsticker, T. Kreitzberg, M. Habermehl, O. Hatzfeld, R. Kneer, Fuel 201 (2017) 81–92. [16] S. K. Bhatia, D. D. Perlmutter, AlChE J. 26 (1980) 379–386.