Implications of feeding or cofeeding bio- oil in the fluid catalytic cracker (FCC) in terms
of regeneration kinetics and energy balance
Item Type Article
Authors Ochoa, Aitor;Vicente, Héctor;Sierra, Irene;Arandes, José M.;Castaño, Pedro
Citation Ochoa, A., Vicente, H., Sierra, I., Arandes, J. M., & Castaño, P. (2020). Implications of feeding or cofeeding bio-oil in the fluid catalytic cracker (FCC) in terms of regeneration kinetics and energy balance. Energy, 209, 118467. doi:10.1016/
j.energy.2020.118467 Eprint version Post-print
DOI 10.1016/j.energy.2020.118467
Publisher Elsevier BV
Journal Energy
Rights NOTICE: this is the author’s version of a work that was accepted for publication in Energy. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Energy, [209, , (2020-07-27)] DOI:
10.1016/j.energy.2020.118467 . © 2020. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://
creativecommons.org/licenses/by-nc-nd/4.0/
Download date 2024-01-23 21:20:31
Link to Item http://hdl.handle.net/10754/664564
Implications of feeding or cofeeding bio-oil in the fluid catalytic cracker (FCC) in terms of regeneration kinetics and energy balance
Aitor Ochoa,a Héctor Vicente,a Irene Sierra,b José M. Arandes,a Pedro Castañoa,c,*
a. Department of Chemical Engineering, University of the Basque Country (UPV/EHU), P.O. Box 644, 48080, Bilbao, Spain.
b. Department of Chemical Engineering, Faculty of Pharmacy, University of the Basque Country (UPV/EHU), Vitoria-Gasteiz, Spain.
c. Multiscale Reaction Engineering, KAUST Catalysis Center (KCC), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Saudi Arabia.
Abstract
Feeding or cofeeding bio-oil (biomass pyrolysis oil) into the fluid catalytic cracking (FCC) has a direct impact on product distribution, reaction kinetics and deactivation of this key catalytic valorisation strategy. In this work, we have analysed the impact in terms of the catalyst regeneration kinetics and energy balance of the unit. These factors are linked to the holistic viability of revamped refineries turned into biorefineries. Deactivated catalysts were obtained in FCC experiments using vacuum gasoil and raw bio-oil. The regeneration kinetics of coke combustion were analysed in a thermobalance, whereas the heats dissipated throughout the combustion (high heating values) were analysed in a calorimeter. Overall, the regenerator does not require major design amendments to treat bio-oil. We found a linear correlation between the higher heating value of the reactants and the coke produced, which enable to predict possible scenarios in the FCC unit. When incorporating higher amounts of bio- oil, the heat balance of the unit changes significantly: rising the temperature in the regenerator up to 36 K, requiring significant energy inputs in the heat the bio-oil but offering the chance to recover more (electrical) energy when the proportion of bio-oil is greater than about 50%.
1. Introduction
The growing demand of high added value hydrocarbons, such as gasoline and olefins, is driving intensification endeavors within the refinery scheme. In addition, the upcoming depletion of fossil resources announces the future use of renewables as a source of hydrocarbons [1], in what is known as bio-refinery concept. Within this framework, biomass pyrolysis keeps gaining visibility and viability due to its lower environmental impact and its technological feasibility [2–4]. The most attractive product obtained from biomass pyrolysis is the liquid, known as raw bio-oil, containing water (15-50 wt %) and oxygenates, such as acids, ketones, esters, alcohols, phenols and guaiacols, among others. The raw bio-oil has a low higher heating value (HHV = 16-19 MJ kg-1), low volatility, high acidity, high viscosity and it is relatively unstable due to the re-polymerization of some of its constituents.
There are several pathways in order to upgrade raw bio-oil and incorporate this feed within the refinery scheme, among which the fluid catalytic cracking (FCC) process is considered a very promising route [5–7]. Other routes involve hydrodeoxygenation (HDO) [8–10]. FCC units have been used for decades to treat heavy hydrocarbons like the ones present in the vacuum gasoil (VGO) or in the residues [11]. As a result, the co-feeding of bio-oil and VGO in the FCC unit has been studied before in detail by means of analyzing the product distribution [12–14]. Wu et al. [15] have performed a technological assessment of different process technologies
as more viable from the mass balance point of view. Some of these authors [16] have later performed a technoeconomic analysis of this valorization technology, pointing to the fact this valorization pathway could make sense economically depending on the price of gasoline and diesel. Besides, future legislations may force refineries to use their infrastructures to be more aligned with environmental concerns.
By incorporating bio-oil, the product distribution of the FCC changes due to a change in the reaction scheme, kinetics and deactivation [17]. In general, the unanimous observations are that adding bio-oil to the FCC causes an increase in the yields of gases (particularly CO2 and CO coming from decarboxylation and decarbonylation, respectively) and that of coke [6,18]. Wang et al. [19] have obtained similar results at pilot scale, so they indicated that the concentration of bio-oil should be kept below 10 wt%
in order to keep similar product yields. Any higher concentration should use hydrotreated bio-oil. In any case, incorporating bio-oil in the FCC affects many more aspects of the operation of the unit, e.g., Brady et al. [20]
have studied the higher corrosion of the stainless-steel pipes occurred by the bio-oil oxygenates and ways to cope with it. Other aspects that should be pointed are (1) the miscibility of bio-oil with gasoil which can be problematic in certain units such as heat exchangers [21] but not in the riser as all components will be in the gas phase, and (2) oxygen content in the products, which should be optimized for not affecting much the rest of the hydrocarbon/oxygenate energy chain [22].
Coke is formed in the FCC during the contact between
other parameters among which the feed composition is the most important one [24,25]. The main precursors of coke in bio-oil correspond to polyols and (methoxy)phenols [26–29], which should be removed by hydrotreating in a previous step in order to minimize coke formation in the FCC [5,13,30,31]. The presence of bio-oil in the feed has a great impact on the composition of the coke nature and location [32]. In general, bio-oil promotes the formation of coke in greater amounts, more oxygenated and deposited in the outer surface of the catalyst particle [33,34]. FCC is a very interconnected unit, and the fact that bio-oil promotes the formation of different type of coke can have an impact on different parts of the unit. That type of coke is an oxygenated one, which co-exist with the more “standard” polycondensed aromatic coke formed by VGO [33].
Coke has a pivotal role in the FCC unit: from one side is detrimental for the cracking reactions as it deactivates the catalyst, but at the same time its presence is beneficial to enhance the margin of the unit by generating power and therefore lowering the energetic input. Due to the high degree of interconnectivity within the parameters, FCC is a very complex unit to monitor, control and optimize [35].
Modelling has proved its value to solve this gap [36] in the two scenarios typically proposed: steady-state and dynamic models. Although the vast majority of the modelling work has been focused in developing kinetic models for the cracking stage, representative and impactful models have been developed for the control of the regenerator [37,38] or the overall FCC unit [39–43].
Coke burning kinetics is a function of its intrinsic nature and location across the catalyst surface [44], and the most used technique to calculate these combustion kinetics is the thermogravimetric analysis (TGA) [45]. In fact, advanced TGA methods can lead to intrinsic kinetic models [46] and advanced mechanisms [47] of carbon combustion-oxidation. However, TGA can also have hurdles to extract relevant kinetic information due to mass transfer limitations [48]. From the groundworks in shoot oxidation of the groups of Moulijn [49,50] and Stanmore- Gilot [51–53], mass transfer can be neglected if the proper conditions are used in the TGA: particle size and weight, flowrate of gas, temperature rate. In a previous work, we evaluated the best thermogravimetric methodologies in order to extract relevant kinetic information of coke combustion, avoiding mass transfer limitations [54].
The aim of this work is to quantify the impact of feeding or cofeeding raw bio-oil in benchmark FCC conditions on the kinetics and heat balance of coke combustion. The experimental work has been performed using a coked industrial FCC, obtained in a riser simulator under industrially relevant conditions (773 K and 3 s), feeding vacuum gasoil, raw bio-oil or a blend of the previous two. The kinetic and thermodynamic data of coke combustion has been obtained, using several thermogravimetric and calorimetric strategies. Then, the data has been used to evaluate the implications of feeding
and cofeeding bio-oil to FCC unit operating at benchmark conditions, using a simulation previously developed.
2. Experimental
Three feeds have been used: vacuum gasoil (VGO), raw bio-oil (RBO) and the mixture of VGO (80 wt %) with bio-oil (20 wt %), abbreviated as MIX. The vacuum gasoil (VGO) has been supplied by Petronor S.A. (Somorrostro, Spain). The RBO has been obtained by fast pyrolysis of black poplar sawdust at 723 K in a N2 stream, using a pilot plant provided with a conical spouted bed reactor [55].
Table 1 summarizes all the different analysis performed on the feed, catalyst and coke and the corresponding analysis methodology or equipment. The feeds have been characterized by simulated distillation (following the ASTM D2887 standard in an Agilent 6890 Series GC System with a column Simdis D2887 Fast/Ext.) and elemental analysis (LECO TruSpec CHN Macro). The concentrations of the component families in the RBO have been determined by GC-MS (Shimadzu GC-MS QP2010, BPX5 column of 50 m × 0.22 mm × 0.25 μm). The characterization results of the VGO and RBO feeds have been previously reported:[33] (i) RBO, C fraction (WC) = 29.4 wt %, H fraction (WH) = 8.5 wt %, O fraction (WO) = 61.7 wt %, all this fractions in wet basis, water content 46%, simulated distillation (D2887), initial boiling point
= 308 K, 5 wt% = 328 K, 25 wt% = 347 K, 50 wt% = 427 K, 90 wt% = 536 K, final boiling point = 608 K; (ii) VGO, WC = 85.4 wt %, WH = 11.8 wt %, WO = 0.0 wt %, simulated distillation (D2887), initial boiling point = 457 K, 5 wt% = 640 K, 25 wt% = 705 K, 50 wt% = 747 K, 90 wt% = 806 K, final boiling point = 887 K, density = 0.918 g cm–3, average molecular weight = 370.3 g mol–1, pH = 2.5, viscosity (323 K) = 7 cP, solids = 0.1–1 wt %.
Table 1.
Properties of the HY zeolite catalyst.
Analysis of the feed
Simulated distillation (ASTM D2887) Agilent 6890 GC
Elemental analysis LECO TruSpec CHN Macro
Concentrations of species (GC-MS) Shimadzu QP2010
HHV Eq. (1)
Characterization of the catalysts
SBET, Vpore, Vmicropore, Dpore and zeolite % Micromeritics ASAP 2010 Unit cell size (XRD) Philips PW 1710 Acidity and acid strength Setaram 111 DSC Brönsted/Lewis acid site ratio (FTIR) Nicolet 740 SX Characterization of coke
Coke combustion profiles (TGA) TA Instr. Q5000TA
HHVcoke Setaram 111 DSC
The HHVfeed have been obtained by means of the Dulong’s formula and the previous elemental analysis [56]:
𝐻𝐻𝑉 = [338.8𝑊𝐶+ 1442.8 (𝑊𝐻𝑊𝑂
8 )] 0.001 (1)
Where Wi stands for the fraction of each element (i).
An industrial catalyst based on HY zeolite has been employed, supplied by Petronor S.A. The properties of the catalyst are displayed in Table 2. The physical properties have been determined by means of N2 adsorption isotherms in a Micromeritics ASAP 2010. The unit cell size of the zeolite has been measured by X-ray diffraction (XRD) in a Philips PW 1710. Total acidity and acid strength distribution have been determined by differential scanning calorimetry on a Setaram 111 during the differential adsorption of NH3 at 423 K. Subsequently, temperature programmed desorption (TPD) of absorbed NH3 has been performed following a ramp of 5 K min-1 up to 823 K (TA Instruments SDT 2960, coupled with a MS Thermostar Balzers Instruments mass spectrometer). The Brönsted/Lewis site ratio has been determined from the intensity ratio of the bands at 1545 and 1455 cm-1 measured by pyridine adsorption and FTIR spectrophotometry (Nicolet 740 SX).
The catalytic cracking runs were carried out in a riser simulator reactor consisting in a basket with the catalyst, an internal impeller which moves the gases (containing the reactants and products) throughout the bed, fluidizing it.
This reactor operates in conditions similar to those of industrial FCC risers (short contact times and high bed porosity) [57–59]. The runs have been carried out at 773 K; contact time, 3 s; catalyst/feed ratio in dry base, 6 gcatalyst gfeed-1. A detailed characterization of the coke deposited in the catalytic cracking can be found elsewhere [33].
Table 2.
Properties of the HY zeolite catalyst.
SBET 122 m2 g-1
Vpore 0.15 cm3 g-1
Vmicropore 0.017 cm3 g-1
Dpore 117 Å
Zeolite 15 wt %
Unit cell size 24.3 Å
Acidity 30 µmolNH3· gcatalyst-1
Acid strength 100 kJ molNH3-1
Brönsted/Lewis acid site ratio 0.75 mol mol-1
Coke combustion runs were performed in a TA Instruments TGA Q5000TA thermobalance. Samples of about 50 mg were swept of impurities with a N2 stream (50 mL min-1), at a heating rate of 10 K min-1, reaching 623 K and keeping this temperature for 30 min. After cooling down to 523 K, a stream consisting in 50 mL min-
1 of air was fed (and 5 mL min-1 of N2 as balance gas), then temperature was raised up to 973 K. The modulated runs involved two ramps: (1) β = 5 K min-1, amplitude (H) = 8 K and frequency (ω)= 5·10-3 s-1, (2) and β = 15 K min-1, H
= 8 K and ω = 2·10-2 s-1.
The combustion conversion (α) has been calculated assuming that the amount of coke at the end of the
combustion experiment (973 K) is null. Then, the rate of combustion (-rα) is calculated as follows:
−𝑟𝛼=𝑑𝛼
𝑑𝑡 = 𝐷𝑇𝐺
∫0𝑡𝑓𝐷𝑇𝐺𝑑𝑡 (2)
where DTG stands for the differential thermo-gravimetric profile of the combustion. When a linear heating program is employed, -rα can be calculated integrating within the temperature range (dT), leading to different units of -rα as K-1. The units of -rα has been obtained assuming a linear heating rate in which the time and temperature are linearly dependent. of A total of 6 different methods were applied to the data obtained from the TGA: (m1) each linear heating program fitted to a first order kinetic model; (m2) all linear heating programs fitted to the Salvador and Ferrasse model [60], assuming θ1 = 302.5 K and θ2 = 4.1 s [54]; (m3) all linear heating programs fitted to the differential method; (m4) all linear heating programs fitted to the Li-Tang [61] integral method; (m5) a single modulated heating program with β = 5 K min-1, amplitude (H) = 8 K and frequency (ω)= 5·10-3 s-1; (m6) a single modulated heating program with β = 15 K min-1, H = 8 K and ω = 2·10-2 s-1. The calculation methods are explained elsewhere along with their corresponding equations [54].
A more detailed explanation of the mathematical expression of the modulated method and its limitations is given by Moukhina [62].
The most significant difference between the isoconversional methods is the fact that the Friedman one (m3) considers the dependence of activation energy on α when obtaining the equation, whereas the Li-Tang one (m4) does not consider this dependence, which inevitably leads to deviations in the results between both methods.
The HHVcoke values were obtained in a Setaram DSC 111 calorimeter by heating from 373 to 1073 K at a 5 K min-1 heating rate in an oxidizing atmosphere (air, 30 mL min-1) and an inert atmosphere (He, 30 mL min-1).
The model fitting of the combustions have been performed by minimizing the sum of square errors (SSE) using the fminsearch function of Matlab® R2014a. The confidence intervals have been calculated considering a 95% interval of confidence, using the statistical parameters provided by the re-calculated parameters using lsqcurvefit (which provides with the Jacobian) and nlparci function of Matlab® R2014a.
3. Scheme and simulation of the FCC unit
Fig. 1 shows the main features of the FCC unit, which can be divided in three main sections: (R1) cracking section encompassing the riser reactor, (R2) stripping section to remove volatile hydrocarbons from deactivated catalyst and separate hydrocarbons from catalyst, and (R3) regeneration section to burn off coke and rejuvenate the catalyst. The feed (A) consists in the regenerated catalyst and VGO, along with a small fraction of a heavy fraction
named decant oil (DO), coming from the recirculation of the bottoms of the downstream distillation columns. The regenerated catalyst (G) is very hot (TR3) and exchange a total heat equivalent to Qcat. The mixture of catalyst+DO+VGO reaches the initial temperature (TR1in) and rises throughout R1 performing the cracking reactions during the few seconds of contact time before leaving the riser (B). As the cracking of hydrocarbons is an endothermic reaction, the output of R1 is at lower temperature (TR1out < TR1in). The deactivated catalyst suffers a series of (1) stripping stages with steam (D) in R2 to separate it from the hydrocarbon products and (2) exothermic combustion with air or oxygen-enriched-air (F) in R3. The exhausts of combustion (E) are sent to a turbo-expander, to convert thermal- into electrical-energy.
As seen in Fig. 1, FCC is a very complex unit. For example, if the coke deposited on the catalyst changes its nature towards one with higher enthalpy of combustion or increase in yield, the temperatures of the exhausts and regenerated catalysts (TE and TR3) will also increase. The increase of TE will turn into a higher amount of power obtained in the expander, while an increase in TR3 will turn into higher Qcat. The latter will lead to a number of possible scenarios or actions to counteract this effect.
Typically, VGO comes straight from the vacuum distillation unit at an initial temperature (TVGO), whereas DO comes hot from a distillation unit at a certain temperature (TDO). The catalyst:oil ratio is controlled in order to allow the mixture to reach the required temperature (TR1in). In a standard unit, the Qcat is enough to close the energy balance, avoiding any addition heat input (TVGO = Tin).
Fig. 1. Main sections and streams in a standard FCC unit: R1, riser or cracking section; R2, stripping section; R3, regeneration section; A, feedstock inlet; B, riser outlet, C, product outlet to separation units; D, steam inlet; E, exhaust outlet to expander; F, air or enriched air inlet; G, regenerated catalyst recirculation.
Let us assume a set of values for the cracking of VGO as benchmark conditions, summarized in Table 3. These values are representative of a standard FCC unit of approximately 40,000 bpd, and these have been taken
from the following references [63–66]. In this table YC, F, T, P and Q stands for the coke yield, flowrates, temperatures, pressure and transferred heat of the system.
We have taken those values and ran a simulation of the FCC unit using a previously developed model and industrially validated [41]. In the conditions of the simulation displayed in Table 3, the regenerated catalyst provides 66.2 MJ s-1 of heat: 64.2 MJ s-1 to warm up the VGO and the remaining 1.97 MJ s-1 for the DO.
Table 3.
General parameters for the definition of the FCC unit and VGO properties.
FVGO† 59.8 kg s-1
FRBO† 0.00 kg s-1
TVGO 496 K
FDO 2.06 kg s-1
TDO 589 K
FR3 (catalyst)† 425 kg s-1
TR3† 970 K
TRBO 313 K
Qcat† 66.2 MJ s-1
QRBO† 0.00 MJ s-1
Catalyst/oil ratio 7.11 kg kg-1
TR1in† 821 K
TR1out 783 K
YC† 6.00 wt %
PR1 1.72 atmg
FD 1.27 kg s-1
TD 673 K
FF† 60.6 kg s-1
TF 430 K
† Parameter that can vary according to the processed feedstock.
In the scenario of feeding or cofeeding raw bio-oil in the unit, this mixture will be fed in the bottom of the riser at a certain temperature (TRBO) and an additional heat exchanger might be required to rise the temperature up to the required one for the cracking reactions (TR1in), after mixing with VGO, DO and catalyst. We have assumed that the bio-oil is being produced delocalized, condensed and transported to the refinery, so TRBO = 313 K.
We have programmed a model of the FCC unit using Fortran code [41]. The conditions employed here are different from the ones used in the simulations (Table 3).
The conditions used experimentally differ from those used in the simulation because the intention was to employ the most realistic set of conditions in each approach. Then, use the degree of change between the base case (VGO) and the situation mixing with bio-oil to correct the parameters in the simulation.
4. Results and discussion
4.1. Effects on regeneration kinetics
The profiles of coke combustion for catalyst used in the cracking of the three feeds are shown in Fig. 2, i.e. in terms of differential thermogram, DTG (Fig. 2a) and the rate of combustion, -rα (Fig. 2b), by means of Eq. 2. The
temperature at which the rate of combustion of coke is maximal is achieved at similar values for the three feeds (~815-825 K). The area under the DTG curve is linked with the total coke content deposited on the catalyst (Ccoke) for each feed: RBO, 22.9 mg gcatalyst-1; VGO, 17.1 mg gcatalyst-1; MIX, 14.4 mg gcatalyst-1. In principle, oxygenates of RBO lead to faster coke formation compared with the hydrocarbons of VGO, however the MIX blend leads to slower coke formation due to the compositional differences between the feeds [67,68]: (i) attenuation of the heavy hydrocarbon deposition pathway caused by the steam contained or originated from the bio-oil, and (ii) the hydride transfer from hydrocarbons to the precursors of the oxygenates pathway.
The obtained DTG profiles (Fig. 2a) prove the absence of oxygen mass transfer limitations due to the relatively low temperature of coke combustion [69] and maximum value of DTG (< 4.19 104 µg s g–1catalyst) [54]. Moreover, the results of other authors [70,71], also indicate that the conditions of combustion used here are in the absence of mass transfer limitations.
Fig. 2. (a) DTG and (b) normalized -rα profiles of the combustion of coke deposited on the catalyst used in the cracking of the three feeds, at β = 5 K min-1.
The differences in the mechanism and rate of coke deposition upon the three different feeds have been explained before [33]. The profiles of -rα in Fig. 2b do not show significant differences in the kinetics of combustion
for the different feeds. The relatively high temperature employed in the FCC unit leads to the homogenization of the coke composition regardless of the feed, by means of deoxygenation, dehydration or dehydrogenation reactions, among others [33,72]. Interestingly, although the blend feed (MIX) shows the lower coke formation, its corresponding -rα profile is approximately within the corresponding individual feeds (RBO and VGO).
Fig. 3 shows the apparent kinetic constant (k823 K, Fig.
3a) and activation energy (E, Fig. 3b) for the combustion of coke deposited on the catalyst used in the cracking of the three feeds, using the 6 different methods (m1-6). Some fundamental issues of the results presented in Fig. 3 are the following: k823 K can only be calculated through m1-2
methods, the values of k823 K and E of m1 correspond to the individual values calculated at different β, the values of k823 K and E of m2 correspond to single values and the standard error, m3-6 only displays values of E within the range 0.2 < α < 0.8, m3-4 correspond to isoconversional methods and m5-6 correspond to two modulated heating ramps.
Fig. 3. Values of apparent kinetic constant, k823 K (a), and activation energy, E (b), of the combustion of coke deposited on the catalyst used in the cracking of the three feeds as a function of the method used (mi).
The takeaway message from Fig. 3 is that models m1-2
lead to comparable k823 K values with very high standard error, and relatively lower values of E compared with model-free methods (m3-6), whereas modulated ramps (m5- 6) lead to highly scattered E values. This last observation can be linked to some concerns raised by this technique recently [73]. Choosing an approach among the 6 used is very hard and out of the scope of this work, but
irrespectively, two reproducible observations are that k823
K decreases and E increases when the bio-oil/VGO ratio increases. However, these differences are within the experimental error and are irrelevant. In other words, the combustion of coke is not significantly slowed down by incorporating oxygenates in the cracking section of the FCC unit.
The trend of the values of kinetic parameters obtained during coke combustion can be correlated with its composition and location. As previously explained [33], the coke deposited during the cracking of VGO is a broad distribution of highly condensed aromatic structures placed in the meso- and macropores of the catalyst, whereas the one deposited on the cracking of raw bio-oil (RBO) is more hydrogenated, oxygenated, lighter, with a narrower distribution and placed in the micro- and mesopores of the catalyst. Regardless these differences, the kinetic combustion constants (k823 K and E) of coke are similar and cannot be correlated with its location or its nature.
In the eventuality that the kinetics of coke combustion are slowed down in the FCC unit regenerator, no less than one of these approaches should be considered: (i) increase oxygen partial pressure by using enriched air in F (Fig. 1), (ii) increase residence time of the catalyst within the regenerator (R3), what can be performed revamping the regenerator or increasing the stock of catalyst within the unit (assuming that the amount of catalyst in the rest of sub-units at any time would be the same), (iii) increase regenerator temperature (TR3). However, there is not significant slowdown of coke combustion kinetics when co-feeding oxygen-containing hydrocarbons in the FCC unit. On top, the way FCC regenerators operate are under and excess of temperature (TR3) of about 970 K as shown in Table 3, and residence times of catalyst (dense phase) several order of magnitude overcalculated. That is to say, the subtle kinetic differences observed in the coke combustion profiles indicate that the FCC regenerator does not required amendments to work with bio-oil as cracking feedstock.
4.2. Effect on heats of combustion
Fig. 4 shows the higher heating values of the three feeds (HHVfeed, by means of Eq. (1)) and the corresponding to the coke deposited on the catalyst (HHVcoke, by means of DSC-TPO results). The values of HHVfeed and HHVcoke are linked with the oxygen concentration of the sample (feed or coke). Interestingly, the differences in HHVcoke are less significant than those of the HHVfeed, which agrees with the “homogenization” in the FCC conditions by the removal of oxygen. Besides and above all, a linear correlation, displayed in Fig. 4, was found between the values of HHVfeed and HHVcoke, so the net effect of feeding or cofeeding bio-oil on the heat obtained from the combustion of the deposited coke is similar, regardless the method used to obtain HHV.
Considering both the coke content, obtained in TGA, and its HHVcoke, the energy dissipated in the combustion referred per gram of catalyst in circulation are: RBO = 0.41 kJ gcatalyst-1; MIX = 0.28 kJ gcatalyst-1; VGO = 0.35 kJ gcatalyst-1. Hence, taking VGO as base case, the co-feeding of 20 wt % of bio-oil would implicate dropping the circulating heat -19%, whereas feeding pure bio-oil would increase this parameter +19%. In order to cope with these modifications, a rational approach would be balancing the estimated HHVfeed and the amount of coke formed in the riser.
Fig. 4. Relationship between the higher heating values of the three feeds (HHVfeed) and their respective values of the coke deposited (HHVcoke).
4.3. Effect on energy balances
For better assessing the impact of heat dissipated throughout the coke combustion in the entire unit energy balance, we have ran a simulation [41] of the FCC operation. Regardless the type of feed used, some simulation parameters shown in Table 3 have been fixed:
TVGO, FDO, TDO, TRBO, catalyst/oil ratio, TR1out, PR1, FD, TD
and TF. We have performed several simulations assuming that the values of FVGO, FRBO and Cp are these displayed in Table 4 while taking the values of YC and combustion factor () from the experimental results obtained before.
The has been calculated as follows:
𝜑𝑖= 𝐻𝐻𝑉𝑖
𝐻𝐻𝑉𝑉𝐺𝑂 (3)
The enthalpy of vaporization (∆Hvap) of bio-oil has been calculated from the individual ones of water and the organic phase as:
𝛥𝐻𝑣𝑎𝑝 = [2.257 · 𝑥𝑤𝑎𝑡𝑒𝑟+ (1 − 𝑥𝑤𝑎𝑡𝑒𝑟) · 0.610] MJ kg-1 (3) Where xwater is the weight fraction of water in the bio- oil. The values of the enthalpy of vaporization and heat capacity of bio-oil, in dry basis, which are 0.610 MJ kg-1 and 2.43 MJ kg-1 K-1 respectively, have been taken from
elsewhere [74]. The air flowrate (FF) is calculated to keep O2 composition in the regenerator outlet as close as possible to the one obtained in the VGO feeding case. The simulation provides certain parameters such as TR3 and FF
in order to keep the unit working steadily and controlled.
As seen in Table 4, the temperature in the regenerator (TR3) and the amount of air used in this section (FF) first decrease and then increase as the percentage of the bio-oil is increased. The rest of the parameters such as Qcat, QRBO
and TR1in depend on the enthalpy of reaction of bio-oil (∆HRBO). In a previous work [75], the values of ∆HRBO are slightly endothermic or even exothermic during the cracking of model compounds of bio-oil. Thus, we have studied four cases: ∆HRBO = 586, 527, 502 and 477 kJ kg-
1. We may well expect certain variation of the value of
∆HRBO depending on the biomass used to produce the bio- oil. The value of ∆HVGO has been fixed to 586 kJ kg-1.
Table 4.
Simulation parameters for each of the three feeds.
VGO MIX RBO
FVGO, kg s-1 59.8 47.8 0.00
FRBO, kg s-1 0.00 12.0 59.8
Cp mixture, kJ (kg K)-1 3.43 3.39 3.22
1.00 0.95 0.87
YC, wt % 6.00 5.05 8.04
TR3, K 970 935 1006
FF, kg s-1 60.6 50.6 82.6
Using the four values of ∆HRBO and the rest of the conditions explained before, Fig. 5 shows the evolution of the parameters that depend on the ∆HRBO (Qcat, QRBO and TR1in) as a function of the proportion of the bio-oil in the feed. The results of the simulation (Table 4 and Fig. 5) show that as the proportion of the bio-oil in the feed increases and regardless of the ∆HRBO assumed, the heat required to warm up the bio-oil (QRBO) is increased. Thus, if biomass pyrolysis is performed in delocalized units, a significant amount of energy is required to heat up the ambient temperature feedstock. Otherwise, a new pyrolysis unit located in the refinery and close to the FCC unit should be built with the corresponding cost of transporting to the refinery the biomass, which has very low energy density. Within this strategy, a thermal conversion unit in the refinery could be revamped for treating biomass, such as the visbraker or the coker.
The increase of bio-oil in the FCC unit lead to an initial decrease and later increase of the amount of heat that can be recovered by circulating the catalyst (Qcat). The initial decrease is caused by the drop in the yield of coke (YC) with the parallel increase of the oxygen content of coke and the drop of its HHV. The final increase is caused by the predominance of the massive YC increase when higher proportions of bio-oil is used in the feed. The Qcat obtained with the blend bio-oil:VGO equalize with that of the VGO when the proportion of bio-oil is about 50%.
Fig. 5. (a) Exchanged heat by the catalyst (Qcat) and the bio-oil (QRBO) as a function of the percentage of bio-oil in the feed for different values of ∆HVGO. (b) Temperature in the inlet of the riser (TR1in) as a function of the percentage of bio-oil in the feed for different values of ∆HVGO
When co-processing a 20 % of the feed as bio-oil, regenerator temperature drops by 35 K. This drop of temperature in the regeneration lead to a 23 % decrease in the energy transported by the regenerated catalyst. Along with this, the bio-oil requires about 31 MJ s-1 (QRBO) to finally obtain the TR1in (catalyst, DO, VGO and RBO).
However, this temperature lowers as the proportion of bio- oil increases and its enthalpy of reaction decreases (Fig.
5c). This correlation between TR1in and the proportion of bio-oil in the feed is almost linear, as observed in Fig. 5c, and it does not reach the final riser temperature (TR1out) in the range of ∆HRBO used.
When processing 100 % bio-oil, which is already challenging from the standpoint of keep a constant flow of bio-oil of about 5200 t day-1, more heat is required to warm up the feed (QRBO). The massive deposition of coke, despite its lower HHV, cause a rise in the temperature of the regenerator of 36 K. This will eventually cause an increase of the heat obtained by the catalyst (Qcat) of 24- 48% from the VGO case. However, considering that the temperature of the regenerator (TR3) is the same than the one of the regenerator exhaust outlet (TE): the amount of
energy obtained as electricity in the expander will be significantly higher.
5. Conclusions
In this work, we have compared the combustion kinetics and thermodynamics of the coke deposited under the same fluid catalytic cracking (FCC) conditions and using one of the three feeds: standard FCC feedstock such as vacuum gasoil (VGO) as base case, pure raw bio-oil (RBO) and a mixture of the two (MIX, 20 wt % of bio-oil). The combustion runs have been performed in a thermogravimetric apparatus and a calorimeter, following methodologies already optimized before. The results have been used in a simulation to predict the behavior of the industrial FCC operation.
Our results show, regardless of the methodology used, that the kinetics of coke combustion (based on the apparent kinetic constant at 823 K, and activation energy) are unaffected by incorporating oxygenates in the FCC feed. Besides, the harsh conditions used in the FCC regenerator (with high temperatures and catalyst residence times) ensures the complete combustion of coke in all cases.
The amount of heat obtained in the combustion decreases linearly when the fraction of oxygenates is higher. Our unit simulation has proven that as the proportion of bio-oil in the feed increases, so it does the heat required to warm up the bio-oil, whereas the temperature of the regenerator drops and then rises.
Meanwhile, the temperature in the regenerator will rise by the presence of bio-oil up to 36 K due to the greater coke deposition. Thus, using proportions of bio-oil higher than about 50% we can recover more energy by the catalyst and by the exhausts in the expander (due to the higher temperature of the regenerator). From this particular point of view, the feeding of bio-oil in the FCC unit seems energetically viable and attractive.
Abbreviations
A feedstock of the FCC unit bpd barrels per day
Ccoke coke content on the catalyst, mg gcatalyst-1
Cp specific heat, kJ (kg K)-1 D steam
DO decant oil Dpore pore diameter,Å
DTG differential thermo-gravimetry, µg s g–1catalyst
E exhausts of combustion E activation energy, kJ mol-1 F air or oxygen-enriched-air Fi mass flowrate for i, kg s-1 FCC fluid catalytic cracking G regenerated catalyst H amplitude, K
HDO hydrodeoxygenation HHV higher heating value, kJ g-1
k823 K kinetic constant at reference temperature, s-1 mi method i in the TGA
MIX mixture of 80 wt % VGO with 20 wt % bio-oil Pi pressure in i section (see Ri)
Qi heat exchanged in different sections of the unit. i = catalyst from the regenerator or RBO for the one to pre-heat the bio-oil
-rα rate of combustion, K-1
Ri reactor or section in the FCC unit. i = 1, cracking section; i = 2, stripping section; i = 3, regeneration section
RBO raw bio-oil
SBET surface area using BET equation, m2 g-1 SSE sum of square errors
TGA thermogravimetric analysis Ti temperature of i, K
VGO vacuum gasoil Vpore pore volume,cm3 g-1
Vmicropore micropore volume,cm3 g-1 Wi mass fraction of i, wt %
xwater weight fraction of water in the bio-oil, wt%
Yi yield of product i, wt %. i = C, coke Greek symbols
β heating ramp in the TGA, K min-1
∆Hi enthalpy of cracking for i, kJ kg-1
∆Hvap enthalpy of vaporization, MJ kg-1
θi parameters of for the TGA model [54,60]
combustion factor, dimensionless ω frequency, s-1
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
CRediT authorship contribution statement
Aitor Ochoa: Conceptualization, Methodology, Formal analysis, Investigation, Visualization, Writing original draft. Héctor Vicente: Methodology, Validation, Writing original draft. Irene Sierra: Visualization, Writing - review
& editing. José M. Arandes: Methodology, Software, Writing - review & editing. Pedro Castaño: Supervision, Conceptualization, Project administration, Resources, Writing - review & editing.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
This work was carried out with the support of the Ministry of Economy and Competitiveness of the Spanish Government, some cofounded with ERDF funds (CTQ2016-79646-P) and the Basque Government (IT748-13). A. Ochoa is grateful for his predoctoral grant from the Department of Education, Language Policy and Culture of the Basque Government (PRE_2016_2_0129).
References
[1] A. Demirbas, Competitive liquid biofuels from biomass, Appl. Energy. 88 (2011) 17–28.
doi:10.1016/j.apenergy.2010.07.016.
[2] A. V. Bridgwater, Review of fast pyrolysis of biomass and product upgrading, Biomass and Bioenergy. 38 (2012) 68–94.
doi:10.1016/j.biombioe.2011.01.048.
[3] W.N.R.W. Isahak, M.W.M. Hisham, M.A. Yarmo, T.Y. Yun Hin, A review on bio-oil production from biomass by using pyrolysis method, Renew. Sustain.
Energy Rev. 16 (2012) 5910–5923.
doi:10.1016/j.rser.2012.05.039.
[4] K. Onarheim, I. Hannula, Y. Solantausta, Hydrogen enhanced biofuels for transport via fast pyrolysis of biomass: A conceptual assessment, Energy. 199 (2020) 117337. doi:10.1016/j.energy.2020.117337.
[5] G. Fogassy, N. Thegarid, Y. Schuurman, C.
Mirodatos, From biomass to bio-gasoline by FCC co- processing: Effect of feed composition and catalyst structure on product quality, Energy Environ. Sci. 4 (2011) 5068–5076. doi:10.1039/c1ee02012a.
[6] T.M. Huynh, U. Armbruster, H. Atia, U. Bentrup, B.M.Q. Phan, R. Eckelt, L.H. Nguyen, D.A. Nguyen, A. Martin, Upgrading of bio-oil and subsequent co- processing under FCC conditions for fuel production, React. Chem. Eng. 1 (2016) 239–251.
doi:10.1039/c5re00068h.
[7] Y.K. Ong, S. Bhatia, The current status and perspectives of biofuel production via catalytic cracking of edible and non-edible oils, Energy. 35 (2010) 111–119. doi:10.1016/j.energy.2009.09.001.
[8] F.L.P. Resende, Recent advances on fast
hydropyrolysis of biomass, Catal. Today. 269 (2016) 148–155. doi:10.1016/j.cattod.2016.01.004.
[9] H. Jahromi, F.A. Agblevor, Upgrading of pinyon- juniper catalytic pyrolysis oil via
hydrodeoxygenation, Energy. 141 (2017) 2186–2195.
doi:10.1016/j.energy.2017.11.149.
[10] T.-S. Kim, S. Oh, J.-Y. Kim, I.-G. Choi, J.W. Choi, Study on the hydrodeoxygenative upgrading of crude bio-oil produced from woody biomass by fast pyrolysis, Energy. 68 (2014) 437–443.
doi:10.1016/j.energy.2014.03.004.
[11] J.R. García, M. Falco, U. Sedran, Intracrystalline mesoporosity over Y zeolites. Processing of VGO and resid-VGO mixtures in FCC, Catal. Today. 296 (2017) 247–253. doi:10.1016/j.cattod.2017.04.010.
[12] P.L. Cruz, E. Montero, J. Dufour, Modelling of co- processing of HDO-oil with VGO in a FCC unit, Fuel. 196 (2017) 362–370.
doi:10.1016/j.fuel.2017.01.112.
[13] I. Grac̀a, J.M. Lopes, H.S. Cerqueira, M.F. Ribeiro, Bio-oils upgrading for second generation biofuels, Ind. Eng. Chem. Res. 52 (2013) 275–287.
doi:10.1021/ie301714x.
[14] L. Gueudré, N. Thegarid, L. Burel, B. Jouguet, F.
Meunier, Y. Schuurman, C. Mirodatos, Coke chemistry under vacuum gasoil/bio-oil FCC co- processing conditions, Catal. Today. 257 (2015) 200–
212. doi:10.1016/j.cattod.2014.09.001.
[15] L. Wu, Y. Wang, L. Zheng, M. Shi, J. Li, Design and optimization of bio-oil co-processing with vacuum gas oil in a refinery, Energy Convers. Manag. 195 (2019) 620–629.
doi:10.1016/j.enconman.2019.05.041.
[16] L. Wu, Y. Wang, L. Zheng, P. Wang, X. Han, Techno-economic analysis of bio-oil co-processing with vacuum gas oil to transportation fuels in an existing fluid catalytic cracker, Energy Convers.
Manag. 197 (2019) 111901.
doi:10.1016/j.enconman.2019.111901.
[17] L. Gueudré, F. Chapon, C. Mirodatos, Y. Schuurman, R. Venderbosch, E. Jordan, S. Wellach, R.M.
Gutierrez, Optimizing the bio-gasoline quantity and quality in fluid catalytic cracking co-refining, Fuel.
192 (2017) 60–70. doi:10.1016/j.fuel.2016.12.021.
[18] Á. Ibarra, I. Hita, J.M. Arandes, J. Bilbao, Influence of the Composition of Raw Bio-Oils on Their Valorization in Fluid Catalytic Cracking Conditions, Energy and Fuels. 33 (2019) 7458–7465.
doi:10.1021/acs.energyfuels.9b01527.
[19] S. Wang, X. Li, F. Zhang, Q. Cai, Y. Wang, Z. Luo, Bio-oil catalytic reforming without steam addition:
Application to hydrogen production and studies on its mechanism, Int. J. Hydrogen Energy. 38 (2013) 16038–16047. doi:10.1016/j.ijhydene.2013.10.032.
[20] C. Wang, R. Venderbosch, Y. Fang, Co-processing of crude and hydrotreated pyrolysis liquids and VGO in a pilot scale FCC riser setup, Fuel Process. Technol.
181 (2018) 157–165.
doi:10.1016/j.fuproc.2018.09.023.
[21] A. Krutof, K. Hawboldt, Blends of pyrolysis oil, petroleum, and other bio-based fuels: A review, Renew. Sustain. Energy Rev. 59 (2016) 406–419.
doi:10.1016/j.rser.2015.12.304.
[22] G.M. Stokes, C.C. Wear, W. Suarez, G.W. Young, Reformulated gasoline will change FCC operations and catalysts, Oil Gas J. 88 (1990) 58–63.
[23] X. Dupain, M. Makkee, J.A. Moulijn, Optimal conditions in fluid catalytic cracking: A mechanistic approach, Appl. Catal. A Gen. 297 (2006) 198–219.
doi:10.1016/j.apcata.2005.09.009.
[24] H.S. Cerqueira, G. Caeiro, L. Costa, F. Ramôa Ribeiro, Deactivation of FCC catalysts, J. Mol. Catal.
A Chem. 292 (2008) 1–13.
doi:10.1016/j.molcata.2008.06.014.
[25] D. Padovan, S. Tolborg, L. Botti, E. Taarning, I.
Sádaba, C. Hammond, Overcoming catalyst deactivation during the continuous conversion of sugars to chemicals: maximising the performance of Sn-Beta with a little drop of water, React. Chem. Eng.
3 (2018) 155–163. doi:10.1039/C7RE00180K.
[26] A. Ochoa, B. Aramburu, B. Valle, D.E. Resasco, J.
Bilbao, A.G. Gayubo, P. Castaño, Role of oxygenates and effect of operating conditions in the deactivation of a Ni supported catalyst during the steam reforming of bio-oil, Green Chem. 19 (2017) 4315–4333.
doi:10.1039/c7gc01432e.
[27] R.T.J. Gerards, A. Fernandes, I. Graça, M.F. Ribeiro, Towards understanding of phenolic compounds impact on Ni- and V-USY zeolites during bio-oils co- processing in FCC units, Fuel. 260 (2020) 116372.
doi:10.1016/j.fuel.2019.116372.
[28] H.W. Lee, B.R. Jun, H. Kim, D.H. Kim, J.-K. Jeon, S.H. Park, C.H. Ko, T.-W. Kim, Y.-K. Park, Catalytic hydrodeoxygenation of 2-methoxy phenol and dibenzofuran over Pt/mesoporous zeolites, Energy. 81 (2015) 33–40. doi:10.1016/j.energy.2014.11.058.
[29] M. Bertero, U. Sedran, Immediate catalytic upgrading of soybean shell bio-oil, Energy. 94 (2016) 171–179.
doi:10.1016/j.energy.2015.10.114.
[30] G. Fogassy, N. Thegarid, G. Toussaint, A.C. van Veen, Y. Schuurman, C. Mirodatos, Biomass derived feedstock co-processing with vacuum gas oil for second-generation fuel production in FCC units, Appl. Catal. B Environ. 96 (2010) 476–485.
doi:10.1016/j.apcatb.2010.03.008.
[31] A. Eschenbacher, T. Myrstad, N. Bech, J.Ø. Duus, C.
Li, P.A. Jensen, U.B. Henriksen, J. Ahrenfeldt, U.V.
Mentzel, A.D. Jensen, Co-processing of wood and wheat straw derived pyrolysis oils with FCC feed—
Product distribution and effect of deoxygenation, Fuel. 260 (2020) 116312.
doi:10.1016/j.fuel.2019.116312.
[32] B. Valle, P. Castaño, M. Olazar, J. Bilbao, A.G.
Gayubo, Deactivating species in the transformation of crude bio-oil with methanol into hydrocarbons on a HZSM-5 catalyst, J. Catal. 285 (2012) 304–314.
doi:10.1016/j.jcat.2011.10.004.
[33] A. Ibarra, A. Veloso, J. Bilbao, J.M.M. Arandes, P.
Castaño, Dual coke deactivation pathways during the catalytic cracking of raw bio-oil and vacuum gasoil in FCC conditions, Appl. Catal. B Environ. 182 (2016) 336–346. doi:10.1016/j.apcatb.2015.09.044.
[34] A. Ibarra, I. Hita, M.J. Azkoiti, J.M. Arandes, J.
Bilbao, Catalytic cracking of raw bio-oil under FCC unit conditions over different zeolite-based catalysts, J. Ind. Eng. Chem. 78 (2019) 372–382.
doi:10.1016/j.jiec.2019.05.032.
[35] J. Long, T. Li, M. Yang, G. Hu, W. Zhong, Hybrid Strategy Integrating Variable Selection and a Neural
Network for Fluid Catalytic Cracking Modeling, Ind.
Eng. Chem. Res. 58 (2019) 247–258.
doi:10.1021/acs.iecr.8b04821.
[36] C.I.C. Pinheiro, J.L. Fernandes, L. Domingues, A.J.S.
Chambel, I. Graça, N.M.C. Oliveira, H.S. Cerqueira, F.R. Ribeiro, Fluid Catalytic Cracking (FCC) Process Modeling, Simulation, and Control, Ind. Eng. Chem.
Res. 51 (2012) 1–29. doi:10.1021/ie200743c.
[37] M.A.F. Martins, A.C. Zanin, D. Odloak, Robust model predictive control of an industrial partial combustion fluidized-bed catalytic cracking
converter, Chem. Eng. Res. Des. 92 (2014) 917–930.
doi:10.1016/j.cherd.2013.08.005.
[38] M. Garcia-Dopico, A. Garcia, Modeling and Simulation of the Fluidized Bed and Freeboard of an FCCU Regenerator, Chem. Technol. Fuels Oils. 52 (2017) 716–731. doi:10.1007/s10553-017-0765-y.
[39] Y. Du, L. Sun, A.S. Berrouk, C. Zhang, X. Chen, D.
Fang, W. Ren, Novel Integrated Reactor-Regenerator Model for the Fluidized Catalytic Cracking Unit Based on an Equivalent Reactor Network, Energy &
Fuels. 33 (2019) 7265–7275.
doi:10.1021/acs.energyfuels.9b01616.
[40] H. Sildir, Y. Arkun, U. Canan, S. Celebi, U. Karani, I.
Er, Dynamic modeling and optimization of an industrial fluid catalytic cracker, J. Process Control.
31 (2015) 30–44. doi:10.1016/j.jprocont.2015.04.002.
[41] J.M. Arandes, M.J. Azkoiti, J. Bilbao, H.I. De Lasa, Modelling FCC units under steady and unsteady state conditions, Can. J. Chem. Eng. 78 (2000) 111–123.
doi:10.1002/cjce.5450780116.
[42] H. Ali, S. Rohani, J.P. Corriou, Modelling and Control of a Riser Type Fluid Catalytic Cracking (FCC) Unit, Chem. Eng. Res. Des. 75 (1997) 401–
412. doi:10.1205/026387697523868.
[43] A.-F. Chang, K. Pashikanti, Y.A. Liu, Refinery Engineering, Wiley-VCH Verlag GmbH & Co.
KGaA, Weinheim, Germany, 2012.
doi:10.1002/9783527666836.
[44] R. Singh, E. Gbordzoe, Modeling FCC spent catalyst regeneration with computational fluid dynamics, Powder Technol. 316 (2017) 560–568.
doi:10.1016/j.powtec.2016.10.058.
[45] A.G. Gayubo, J.M. Arandes, A.T. Aguayo, M.
Olazar, J. Bilbao, Calculation of the kinetics of catalyst regeneration by burning coke following a temperature ramp, Chem. Eng. J. Biochem. Eng. J. 54 (1994) 35–40. doi:10.1016/0923-0467(93)02802-4.
[46] P.A. Carlsson, Detailed modeling of carbon
oxidation, J. Phys. Chem. C. 116 (2012) 9063–9071.
doi:10.1021/jp300738a.
[47] E. Picheau, F. Hof, A. Derré, B. Daffos, A. Pénicaud, Thermal Oxidation of Carbonaceous Nanomaterials Revisited: Evidence of Mechanism Changes, Angew.
Chemie - Int. Ed. 58 (2019) 16013–16017.
doi:10.1002/anie.201906026.
[48] S. Mahamulkar, K. Yin, P.K. Agrawal, R.J. Davis,
C.W. Jones, A. Malek, H. Shibata, Formation and Oxidation/Gasification of Carbonaceous Deposits: A Review, Ind. Eng. Chem. Res. 55 (2016) 9760–9818.
doi:10.1021/acs.iecr.6b02220.
[49] T. Wigmans, H. van Cranenburgh, R. Elfring, J.A.
Moulijn, Mass transfer phenomena during potassium carbonate catalysed carbon steam gasification reactions in a microbalance setup, Carbon N. Y. 21 (1983) 23–31. doi:10.1016/0008-6223(83)90152-5.
[50] J.P.A. Neeft, F. Hoornaert, M. Makkee, J.A. Moulijn, The effects of heat and mass transfer in
thermogravimetrical analysis. A case study towards the catalytic oxidation of soot, Thermochim. Acta.
287 (1996) 261–278. doi:10.1016/S0040- 6031(96)03002-X.
[51] P. Gilot, A. Brillard, B.R. Stanmore, Geometric effects on mass transfer during thermogravimetric analysis: Application to reactivity of diesel soot, Combust. Flame. 102 (1995) 471–480.
doi:10.1016/0010-2180(95)00049-C.
[52] B. Stanmore, P. Gilot, G. Prado, The influence of mass transfer in DTG combustion tests, Thermochim.
Acta. 240 (1994) 79–89. doi:10.1016/0040- 6031(94)87030-6.
[53] B.R. Stanmore, J.F. Brilhac, P. Gilot, The oxidation of soot: A review of experiments, mechanisms and models, Carbon N. Y. 39 (2001) 2247–2268.
doi:10.1016/S0008-6223(01)00109-9.
[54] A. Ochoa, A. Ibarra, J. Bilbao, J.M.J.M. Arandes, P.
Castaño, Assessment of thermogravimetric methods for calculating coke combustion-regeneration kinetics of deactivated catalyst, Chem. Eng. Sci. 171 (2017) 459–470. doi:10.1016/j.ces.2017.05.039.
[55] J. Makibar, A.R. Fernandez-Akarregi, M. Amutio, G.
Lopez, M. Olazar, Performance of a conical spouted bed pilot plant for bio-oil production by poplar flash pyrolysis, Fuel Process. Technol. 137 (2015) 283–
289. doi:10.1016/j.fuproc.2015.03.011.
[56] S. Hosokai, K. Matsuoka, K. Kuramoto, Y. Suzuki, Modification of Dulong’s formula to estimate heating value of gas, liquid and solid fuels, Fuel Process.
Technol. 152 (2016) 399–405.
doi:10.1016/j.fuproc.2016.06.040.
[57] A. Errekatxo, A. Ibarra, A. Gutierrez, J. Bilbao, J.M.
Arandes, P. Castaño, Catalytic deactivation pathways during the cracking of glycerol and glycerol/VGO blends under FCC unit conditions, Chem. Eng. J. 307 (2017) 955–965. doi:10.1016/j.cej.2016.08.100.
[58] J.M. Arandes, I. Torre, M.J. Azkoiti, P. Castaño, J.
Bilbao, H. de Lasa, Effect of catalyst properties on the cracking of polypropylene pyrolysis waxes under FCC conditions, Catal. Today. 133–135 (2008) 413–
419. doi:10.1016/j.cattod.2007.12.080.
[59] J.M. Arandes, M.J. Azkoiti, I. Torre, M. Olazar, P.
Castaño, Effect of HZSM-5 catalyst addition on the cracking of polyolefin pyrolysis waxes under FCC conditions, Chem. Eng. J. 132 (2007) 17–26.
doi:10.1016/j.cej.2007.01.012.
[60] S. Salvador, J.-H. Ferrasse, A Practical Method to Derive Sample Temperature during Nonisothermal Coupled Thermogravimetry Analysis and Differential Scanning Calorimetry Experiments, Chem. Eng.
Technol. 29 (2006) 696–702.
doi:10.1002/ceat.200600022.
[61] C.R. Li, T.B. Tang, A new method for analysing non- isothermal thermoanalytical data from solid-state reactions, Thermochim. Acta. 325 (1999) 43–46.
doi:10.1016/S0040-6031(98)00568-1.
[62] E. Moukhina, Direct analysis in modulated thermogravimetry, Thermochim. Acta. 576 (2014) 75–83. doi:10.1016/j.tca.2013.11.024.
[63] R. Sadeghbeigi, Fluid Catalytic Cracking Handbook, Elsevier, 2012. doi:10.1016/C2010-0-67291-9.
[64] J.G. Speight, Catalytic Cracking, Refin. Futur. (2011) 181–208. doi:10.1016/b978-0-8155-2041-2.10006-2.
[65] S.A. Treese, P.R. Pujadó, D.S.J. Jones, Handbook of petroleum processing, 2015. doi:10.1007/978-3-319- 14529-7.
[66] Multiphase Reactor Engineering for Clean and Low- Carbon Energy Applications, 2017.
doi:10.1002/9781119251101.
[67] M. Bertero, G.D. La Puente, U. Sedran, Products and coke from the conversion of bio-oil acids, esters, aldehydes and ketones over equilibrium FCC catalysts, Renew. Energy. 60 (2013) 349–354.
doi:10.1016/j.renene.2013.04.017 Technical note.
[68] M. Bertero, U. Sedran, Upgrading of bio-oils over equilibrium FCC catalysts. Contribution from alcohols, phenols and aromatic ethers, Catal. Today.
212 (2013) 10–15. doi:10.1016/j.cattod.2013.03.016.
[69] J. Song, C.H. Jeon, A.L. Boehman, Impacts of oxygen diffusion on the combustion rate of in-bed soot particles, Energy and Fuels. 24 (2010) 2418–
2428. doi:10.1021/ef900692m.
[70] C. Wang-Hansen, C.J. Kamp, M. Skoglundh, B.
Andersson, P.A. Carlsson, Experimental method for kinetic studies of gas-solid reactions: Oxidation of carbonaceous matter, J. Phys. Chem. C. 115 (2011) 16098–16108. doi:10.1021/jp204539g.
[71] T.J. Keskitalo, K.J.T. Lipiäinen, A.O.I. Krause, Kinetic modeling of coke oxidation of a ferrierite catalyst, Ind. Eng. Chem. Res. 45 (2006) 6458–6467.
doi:10.1021/ie060521g.
[72] A. Ibarra, E. Rodríguez, U. Sedran, J.M. Arandes, J.
Bilbao, Synergy in the Cracking of a Blend of Bio-oil and Vacuum Gasoil under Fluid Catalytic Cracking Conditions, Ind. Eng. Chem. Res. 55 (2016) 1872–
1880. doi:10.1021/acs.iecr.5b04502.
[73] P. Budrugeac, Critical study concerning the use of sinusoidal modulated thermogravimetric data for evaluation of activation energy of heterogeneous processes, Thermochim. Acta. 690 (2020) 178670.
doi:10.1016/j.tca.2020.178670.
[74] M.B. Shemfe, B. Fidalgo, S. Gu, Chemical
Engineering Research and Design Heat integration for bio-oil hydroprocessing coupled with aqueous phase steam reforming, Chem. Eng. Res. Des. 107 (2015) 73–80. doi:10.1016/j.cherd.2015.09.004.
[75] M.W. Jarvis, J. Olstad, Y. Parent, S. Deutch, K. Iisa, E. Christensen, H. Ben, S. Black, M. Nimlos, K.
Magrini, Catalytic Upgrading of Biomass Pyrolysis Oxygenates with Vacuum Gas Oil Using a Davison Circulating Riser Reactor, Energy and Fuels. 32 (2018) 1733–1743.
doi:10.1021/acs.energyfuels.7b02337.