Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed

Applied Energy xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed q Yuming Zhang a,b, Meiqin Yao b, Shiqiu Gao b,⇑, Guogang Sun a, Guangwen Xu b,⇑ a b

State Key Laboratory of Heavy Oil Processing, China University of Petroleum-Beijing, Beijing 102249, China State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China

h i g h l i g h t s  Steam gasification of petroleum coke with black liquor (BL) was conducted in a micro fluidized bed.  Using BL evidently enhanced gasification reaction and increased H2 content in the produced gas.  The work studied the effects of BL loading amount, reaction temperature and oxygen content in steam.  The shrinking core model well described the steam gasification of petroleum coke to estimate kinetic parameters.  Activation energy of petroleum coke gasification was decreased by BL catalysis and O2 addition into steam.

a r t i c l e

i n f o

Article history: Received 19 August 2014 Received in revised form 10 December 2014 Accepted 5 January 2015 Available online xxxx Keywords: Petroleum coke Black liquor Catalytic gasification Syngas Kinetics

a b s t r a c t Steam gasification of petroleum coke catalyzed by black liquor (BL) was conducted in a micro fluidized bed to investigate the reaction characteristics and kinetics, including the effects of temperature, particle size, BL loading amount and oxygen content in steam on product gas composition and reaction rate. The completion time of petroleum coke steam gasification at 900 °C decreased from 120 min for pure coke to about 40 min for the coke blended with 10 wt.% BL. The corresponding hydrogen fraction in the produced syngas increased by 9 vol.%. The gasification reaction was further enhanced by introducing a small amount of oxygen into the steam. The shrinking core model (SCM) and homogenous model (HM) were used to calculate the kinetics of petroleum coke gasification, finding that SCM enabled the better correlation with experimental data than HM did. Using SCM the activation energy was 77 kJmol1 for coke gasification with 10 wt.% BL as catalyst, which was much lower than 120 kJmol1 for the case without BL blended. The activation energy was further reduced to about 63 kJmol1 by adding 5% oxygen into the steam, showing a synergistic effects of BL and O2 on petroleum coke gasification. The study also justified the feasibility of syngas production from petroleum coke via fluidized bed gasification. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Petroleum coke is the final by-product from the delayed coking process in refinery, and its amount has increased rapidly with the development of deep conversion refining technology. There is also gradually more petroleum coke with high sulfur content in response to the quality deterioration of crude oil. High-sulfur petroleum coke is usually combusted as fuel for generation of

q This article is based on a four-page proceedings paper in Energy Procedia Volume 61 (2015). It has been substantially modified and extended, and has been subject to the normal peer review and revision process of the journal, Applied Energy. ⇑ Corresponding authors. Tel.: +86 10 82544885. E-mail addresses: [email protected] (S. Gao), [email protected] (G. Xu).

steam or power, which would cause serious environmental impacts. Petroleum coke gasification can produce syngas (CO + H2) for the production of hydrogen required by hydrogenation [1–4]. Meanwhile, the inherent sulfur in petroleum coke is mainly transformed into H2S in gasification to be easily recovered as S. However, the low gasification reactivity of petroleum coke greatly restricted its use in actual gasifiers, especially for fluidized bed gasification at temperatures of about 1000 °C [5,6]. It is well known that the gasification of carbonaceous materials, such as coal, biomass and their char can be catalyzed by various alkali and alkaline earth metal (AAEM) compounds [7–10]. However, the use of pure AAEM compounds as catalyst is expensive and economically unfeasible for practical applications because its recovery and treatment after gasification has still great difficulty. Cheap and effective catalyst is urgently needed for petroleum coke

http://dx.doi.org/10.1016/j.apenergy.2015.01.009 0306-2619/Ó 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Zhang Y et al. Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed. Appl Energy (2015), http://dx.doi.org/10.1016/j.apenergy.2015.01.009

2

Y. Zhang et al. / Applied Energy xxx (2015) xxx–xxx

Presssure sensor

Evacuation

Mass spectrometer

Gas cooler Gas filter-drier Water collector

Computer Fluidizing medium

Sample container

Thermocouple

Eletromagnetic valve

Furnace Mass flow meter

Reactor

Gas cylinder

Gas valve

Mass flow meter Three-way valve

Steam generator

Pump

Water tank

Fig. 1. A schematic diagram of the micro-fluidized bed reactor analyzer.

Table 1 Proximate and ultimate analyses of tested samples. Samples

Petroleum coke Fugu coal Dewatered BL

Proximate analysis (wt.%, ad)

Ultimate analysis (wt.%, daf)

HHV (MJ/kg)

M

V

A

FC

C

H

N

S

O

5.76 4.57 0.64

9.97 33.75 47.45

0.17 4.44 43.38

84.10 57.24 9.17

92.66 82.92 55.78

4.09 4.66 3.74

1.65 1.26 1.64

0.48 0.22 0.67

0.95 10.94 38.17

35.22 31.90 9.85

M: moisture; V: volatile matters; A: ash; FC: fixed carbon; ad: air dry; daf: dry ash-free; HHV: higher heating value; O: by difference.

Table 2 Composition analysis of BL ash by XRF. Components

Na2O

K2O

SiO2

SO3

Cl

Al2O3

P2O5

CaO

F 2 O3

TiO2

Content (wt.%, d)

49.73

4.48

4.39

3.62

2.64

0.11

0.09

0.06

0.04

0.01

gasification. Black liquor (BL) is generated in pulping process and contains usually alkaline salts and some organic compounds (lignin, cellulose and hemicellulose). Naqvi et al. [11–13] have investigated BL gasification by direct causticization for syngas and liquid fuel production integrated with a pulp mill that have shown to have good catalytic effect on carbonaceous materials [14–17]. Guo et al. [15] suggested that the sodium salts (e.g., NaOH and Na2CO3) in BL could be an effective catalyst on the alkali lignin pyrolysis and gasification. Zou et al. [16] demonstrated that the gasification rate of petroleum coke was more effectively enhanced by wet grinding with BL than by dry grinding. In a thermogravimetric analyzer (TGA), Zhan et al. [17] found that loading 5 wt.% BL into petroleum coke has made its reactivity with CO2 higher than that of Shenfu coal at 1273 K. These studies on co-gasification of petroleum coke and BL are conducted in TGA and using CO2 reagent, while there was almost no studies on the reaction kinetics. The understanding of gasification reactivity and kinetics for petroleum coke is of not only scientific value but also of practical importance on the gasifier design. The characteristics and kinetics for gasification of the carbonaceous fuels, like, coal, biomass and their chars, have been extensively

studied [18–21] also with various gas–solid reaction models. Zhang et al. [19] studied the reactivity and kinetics of gasification with steam and CO2 for six typical Chinese anthracite chars in TGA using homogeneous model and shrinking core model. Generally, petroleum coke has different pore properties from that of coal or biomass chars. It has usually a highly condensed layer of carbon but few pores on its surface when comparing with that of coal and biomass chars. The reaction models for coal or biomass char might not be applicable to the petroleum coke gasification. The kinetics from TGA is based on weight loss of a sample in a specified heating program. The sample is confined into a fixed crucible during the reaction, which leads to serious inhibition on mass and heat transfers. Thus, the experimental results from TGA cannot be directly extrapolated to the conditions of fluidized bed which are greatly different from these in TGA. The so-called micro fluidized bed reaction analyzer (MFBRA) [22] has been developed to enable the pulse feed of reactant particles in micro grams, rapid heating of reactant and effective suppression of external diffusion, while it allows also the real-time measurement of the gaseous product through an on-line mass spectrometer (MS). The MFBRA has already been used to measure different kinds of gas–solid reactions [23,24], such as biomass

Please cite this article in press as: Zhang Y et al. Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed. Appl Energy (2015), http://dx.doi.org/10.1016/j.apenergy.2015.01.009

3

Y. Zhang et al. / Applied Energy xxx (2015) xxx–xxx

2. Experimental section

feeding system and product analysis part. The micro fluidized bed reactor was made of quartz tube and had an inner diameter of 20 mm and a total length of about 160 mm. The reactor was divided into three sections by two gas distributors, which were, from the bottom, a preheating section filling with inert Al2O3 balls, a reaction area with fluidizing medium (inert silica sand with particle diameter of 100–150 lm) and a sedimentation section to reduce fine particle entrainment. Silica sand was acid-washed, filtered and calcined to remove the possible impurities before it was used as the fluidizing medium. High-purity argon (99.999%), oxygen (99.999%) and deionized water were used in the experiments. A stream of argon at 200 ml/min was the purging gas and also the calibration gas for determining the product gas volume of gasification. The steam gasification reagent was fed at 0.3 g/min and it, together with argon, also served as the fluidizing gas. When it is necessary, oxygen was introduced into the steam to form the blending gasification reagent. For petroleum coke gasification tests, 4 g silica sand was used as the bed material and about 50 mg sample, petroleum coke or its mixture with BL, was put into the pulse-feeding container. After connecting all gas lines and purging the reaction system, the reactor was heated to the preset gasification temperature under uniform fluidization of sand particles. In turn, the sample was injected into the reactor in milliseconds through the electromagnetic valve. The formed gasification products were quickly stripped out of the reactor by the fluidizing gas, which were then filtered, condensed and dried for the analysis by an on-line mass spectrometer (MS, PROLINE AMETEK). The composition of the effluent gas was monitored until no more gasification gas was detected via MS. The actual gas concentration in MS was determined by calibrating gas according to the internal standard method. Besides, it was verified by collecting the gas into a gas bag and further analyzing the gas in a GC (Micro 3000, Agilent). The total carbon of the gasification-formed gas was calculated according to the volume and composition of the gas, and it was then compared with the feedstock to evaluate the carbon balance. The carbon balance was found to be over 90 wt.%, mostly up to 95 wt.%. Each experiment was repeated three times to ensure the relative error of measurement to be less than 3%, and the average value of the three experiments were used to determine the kinetic parameters.

2.1. Apparatus and operation

2.2. Raw materials and analysis method

Petroleum coke gasification was conducted in the MFBRA using steam or a mixture of steam and oxygen as the reagent. Fig. 1 shows the schematic diagram of the MFBRA system, which mainly consists of the gas and steam supply unit, micro fluidized bed reactor, pulse

The tested petroleum coke was from Daqing Refinery Plant of China, and the black liquor (BL) was from a local pulp mill. As one of the commonly used gasification feedstocks in China, the

100

Conversion [%]

80

Without BL 5% BL 10% BL 15% BL 20% BL Fugu Coal Char

60

40

20

Temperature: 900 Particle size: 50-100 µ m 0

0

20

40

60

80

100

Time [min] Fig. 2. Variation of conversion time with black liquor loading amount for steam gasification of petroleum coke.

pyrolysis, CaCO3 decomposition, and graphite combustion. Recently, Zeng et al. [25] has characterized char gasification with CO2 in MFBRA and TGA, and they demonstrated the technical advantages of isothermal differential reaction ensured by MFBRA. We have previously [26] investigated the regeneration characteristics of coked FCC catalyst via steam gasification in MFBRA and analyzed the reaction kinetics using the shrinking core model and homogeneous model, which showed further the adaptability of MFBRA for kinetic studies. This work is devoted to investigating the reaction rate and kinetics of steam gasification for petroleum coke mixed with BL in MFBRA, and meanwhile to verify the feasibility of syngas production from gasifying petroleum coke facilitated with BL and ‘‘steam + O2’’ reagent in fluidized bed. It expects not only to offer an understanding of steam gasification for petroleum coke blended with BL from the aspect of isothermal differential reaction realized in MFBRA but also to justify the effectiveness of MFBRA for investigating steam-involving reactions.

120

CH4

5%

10%

CO

CO2

100

Gas composition [vol.%]

C(002)

Intensity

H2

Petroleum coke with 10% wt.BL

C(10)

80

60

40

20

Pure petroleum coke 10

20

30

40

0 50

60

70

80

90

2θ [deg] Fig. 3. XRD patterns of petroleum coke without and with 10 wt.% BL.

0%

15%

20%

Coal

BL loading amounts [wt.%] Fig. 4. Effect of BL loading amount on product gas composition for steam gasification of petroleum coke.

Please cite this article in press as: Zhang Y et al. Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed. Appl Energy (2015), http://dx.doi.org/10.1016/j.apenergy.2015.01.009

4

Y. Zhang et al. / Applied Energy xxx (2015) xxx–xxx

characterize the crystal structure of petroleum coke. The XRD analyzer was equipped with a 1.5408 Å (k) Cu Ka radiator working at 40 kV and 20 mA, and its scanning rate was 4°/min in the range of 5–90°. The morphology of petroleum coke was visualized by scanning electron microscopy (SEM, JSM-6700F JEOL). By gasification, the hydrocarbon components in petroleum coke and BL were converted into H2, CO, CO2 and CH4. The concentration of these formed gas species can be determined according to their response values in MS. The carbon conversion of petroleum coke gasification was estimated by calculating the carbon-containing gas species CO, CO2 and CH4 in syngas, as

100

60

Without BL With 10% BL 150-250 μm 100-150 μm 50-100 μm

40

20

X¼

Temperature: 900 0

0

20

40

60

80

100

Time [min] Fig. 5. Variation of conversion time with particle size for steam gasification of petroleum coke.

Conversion [%]

Fugu bituminite was chosen to compare its reactivity of gasification with that of petroleum coke catalyzed by BL. Table 1 shows the proximate and ultimate analyses of the tested petroleum coke, Fugu coal and dewatered BL, and Table 2 shows the result of BL ash analysis on dry basis (by XRF). The ash contains plenty of sodium and potassium species, and these AAEM and also the iron oxides inherently containing in BL could be the effective catalysts for gasification of carbonaceous materials. Petroleum coke was first pulverized to small particles and then dried at 105 °C for 4 h. Finally the particles were sieved to form samples with different size ranges for non-catalytic gasification tests including 150–250 lm, 100–150 lm and 50–100 lm. For the samples of catalytic gasification, the pulverized petroleum coke was first impregnated quantitatively with dewatered BL in the deionized water for 24 h and then dried at 105 °C for 12 h. Here, the dewatered BL was used to get the precise catalyst load. After this, the samples were sieved to the same size ranges shown above and stored for the catalytic gasification experiments. The BL loading amounts ranged from 5 wt.% to 20 wt.% of the total mass of petroleum coke and BL on dry basis. In addition to the above-mentioned gas analysis by MS and GC, the BL ash was analyzed through X-ray fluorescence spectrometer (XRF, AXIOS, PANalytical) for the composition of metal oxides. The X-ray diffraction (XRD, X’Pert MPD Pro, PANalytical) was used to

12VðC CO þ C CH4 þ C CO2 Þ 273:15  100%  22:4½mo ð1  aÞwPC þ mo awBL  273:15 þ T o

3. Results and discussion 3.1. Catalytic effect of black liquor Fig. 2 shows the time-series conversion of petroleum coke (50– 100 lm) in its steam gasification catalyzed by BL at loadings of 0– 20 wt.% at 900 °C. The loading of BL effectively catalyzed the petroleum coke gasification, and it reduced the reaction time from over 100 min for the case without BL to less than 40 min for the loading of 10 wt.% BL. When catalyzed by 10 wt.% BL, the time for complete conversion of petroleum coke was less than that of Fugu coal char. This indicates that the mixture of petroleum coke and BL has good gasification reactivity so that it can potentially be a feedstock for coal gasifiers. Further increasing the BL loading from 10 wt.% to 20 wt.% only slightly decreased the completion time of petroleum coke gasification. Raising the BL loading has to lower the petroleum coke feedstock, thus decreasing the treatment capacity of the gasifier for petroleum coke. The 10 wt.% BL loading of petroleum coke appeared optimal and was adopted for all tests herein. Fig. 3 shows the XRD patterns of petroleum coke without and with 10 wt.% BL impregnated. Pure petroleum coke exhibited sharp band of C(0 0 2) at the peak of 25°. This band is generally recognized as the stacking of graphitic basal plans of char crystallite.

100

100

80

80

60

40

20

Without BL 850 950

900 1000

With 10% BL 850 950

900 1000

0 0

20

40

60

80

100

120

140

160

Time [min] Fig. 6. Time-series conversion at different temperatures for steam gasification of petroleum coke without BL and with 10 wt.% BL.

ð1Þ

where X (%) is the conversion; V (L) represents the volume of gas product at the end of reaction; CCO, C CH4 and C CO2 (vol.%) denote respectively the volume fractions of CO, CH4 and CO2 in the formed gas; mo (g) is the sample mass and a (wt.%) is the mass fraction of BL in the sample; wPC and wBL (wt.%) are the carbon fractions in the petroleum coke and BL, and To (°C) is the environment temperature.

Conversion [%]

Conversion [%]

80

60

40

20

0

0

10

20

Without BL 850 950

900 1000

With 10% BL 850 950

900 1000

30

40

50

60

Time [min] Fig. 7. Time-series conversion at different temperatures for gasification of petroleum coke using steam and 5% oxygen without BL and with 10 wt.% BL.

Please cite this article in press as: Zhang Y et al. Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed. Appl Energy (2015), http://dx.doi.org/10.1016/j.apenergy.2015.01.009

5

Y. Zhang et al. / Applied Energy xxx (2015) xxx–xxx

3.0

100

(a)

CO2 2.5

80 2.0 1/3

3[1-(1-X) ]

Gas composition [vol.%]

CO 60

CH4 40

1.5

850 1.0

900

H2

950

0.5

20

1000 0.0 0

0

5

0

40

80

120

160

200

Time [min]

10

Oxygen content in gasification reagent [%] Fig. 8. Effect of oxygen content in steam agent on product gas composition of petroleum coke gasification at 950 °C.

850

(b)

8

900 950

C þ H2 O ! CO þ H2

1

DH0 ¼ þ131 kJmol 0

ƒƒƒ ƒ DH ¼ þ172 kJmol C þ CO2 ƒƒƒ ƒ!2CO ƒƒƒ ƒ 2 þ H2 CO þ H2 O ƒƒƒ ƒ!CO

0

ð2Þ

1

ð3Þ 1

DH ¼ 41 kJmol

ð4Þ

1000

6

-ln(1-x)

The interplanar spacing d002 at C(0 0 2) is about 0.346 nm, higher than that of graphite (0.335 nm), meaning that petroleum coke is of graphite-like structure. The C(0 0 2) and C(1 0) bands were weakened by impregnating 10 wt.% BL, and their crystalline parameters were calculated with the Scherrer equation. The result shows that the carbon structure of petroleum coke was less ordered when blending with 10 wt.% BL, explaining its higher gasification reactivity [27]. Zhan et al. [17] have also found that the mechanochemical interaction of petroleum coke and BL in grinding them together decreased the graphitic structure to show the higher gasification reactivity. Zou et al. [16] indicated that petroleum coke can absorb more mechanical energy in the existence of additive to facilitate the crystalline transformation. Both the destroyed graphite structure and alkaline metal oxides should be responsible for the improvement on gasification reactivity in petroleum coke gasification using BL as catalyst. Fig. 4 shows product gas composition of petroleum coke steam gasification varying with the loading amount of BL, including a comparison with that of coal-steam gasification. The plotted concentration presents the time-integrated average gas concentration, and it is equivalent to the concentration measured by collecting the gas into a gas bag. The hydrogen content was slightly higher when BL was used. For example, the H2 concentration was about 59 vol.% for pure petroleum coke gasification and it elevated to about 68 vol.% with 10 wt.% BL additive. The CO content correspondingly decreased from 19 vol.% to about 9 vol.%. The involved reactions include those shown by Eqs. (2)–(4), which shows obvious the contribution of water gas shift (WGS) reaction (4) to the formation of H2 through converting CO. Furthermore, the catalysis of BL, especially of its AAEM species [7–10] effectively accelerates Eq. (2) to produce more H2 and CO via petroleum coke gasification. The formed CO2 via WGS also facilitates the carbon–CO2 gasification. The sum of H2 and CO was up to 77 vol.% in Fig. 4 for the petroleum coke gasification with 10 wt.% BL, while the H2 concentration (68 vol.%) in the syngas was even higher than that of Fugu coal char gasification (64 vol.%). This shows that the syngas from BL-catalyzed petroleum coke gasification can be a potential hydrogen source for hydro-processing in refinery.

4

2

0

0

40

80

120

160

200

Time [min] Fig. 9. Correlation of experimental data according to reaction models of (a) SCM and (b) HM for steam gasification of petroleum coke without BL (particle size: 100– 150 lm).

3.2. Characteristics of petroleum coke gasification Fig. 5 shows the time-series conversion of steam gasification for petroleum coke with different sizes and without BL or with 10 wt.% BL at 900 °C. The reaction time for complete conversion was shorter for smaller coke particles, but the difference of completion time between the small particles was not as obvious as that of large particles. The sample with small particle sizes might have high specific surface area to provide more chances to interact with the gasification reagent (i.e., steam). Petroleum coke has highly condensed carbon structure and quite limited pores, and the outer surface area takes about half of its total surface area. Zou et al. [28] found that the BET surface area of petroleum coke was only 0.744 m2/g, in which the external surface accounts for about 50%. Wu et al. [29] reported that the BET surface area of petroleum coke was about 3–4 m2/g. via N2-adsorption in BET we did not get a value of surface area for our tested petroleum coke. It is true that the petroleum coke in Zou et al. was from Jingling Co. and that in Wu et al. was from Qilu Co., which are different from ours from Daqing Co. Overall, the effect from decreasing particle size was much less than from using the catalytic effect of BL. Blending 10 wt.% BL into petroleum coke caused the completion time, for a given size range, to be about 30% of the non-catalytic gasification. Fig. 6 shows the time-series conversion at temperatures of 850– 1000 °C for the non-catalytic and catalytic steam gasification of petroleum coke. The reaction completion time was greatly

Please cite this article in press as: Zhang Y et al. Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed. Appl Energy (2015), http://dx.doi.org/10.1016/j.apenergy.2015.01.009

6

Y. Zhang et al. / Applied Energy xxx (2015) xxx–xxx

3.0

(a) 2.5

1/3

3[1-(1-X) ]

2.0

1.5

1.0

850

0.5

950

900 1000 0.0

0

10

20

30

40

50

60

Time [min] 10

(b)

850

900

950

1000

8

-ln(1-x)

6

4

2

0

0

10

20

30

40

50

60

Time [min] Fig. 10. Correlation of experimental data according to reaction models of (a) SCM and (b) HM for steam gasification of petroleum coke with 10 wt.% BL (particle size: 100–150 lm).

Oxygen can be mixed into steam to facilitate gasification reaction because the use of steam only is hard to realize high reaction temperature. Fig. 7 presents the carbon conversion versus reaction time for the case with 5% oxygen in steam. Comparing with Fig. 6, it is obvious that the inclusion of such a small amount of O2 in the gasification reagent greatly facilitated the reaction. The largest acceleration effect was shown for the non-catalytic gasification at low temperatures. For example, at 850 °C the completion time was reduced from 160 min for pure steam (Fig. 6) to only 60 min (Fig. 7) via adding 5% oxygen into steam. In the case with BL catalyst, the time needed for complete conversion was further shortened by about 50%, say, from about 50 min in Fig. 6 for pure steam to below 30 min in Fig. 7 when using oxygen at the gasification temperature of 850 °C. As expected, the promotion effect from adding O2 became less pronounced at relatively high temperatures. Fig. 8 shows the effect of oxygen content in ‘‘steam + O2’’ reagent on gas composition for petroleum coke gasification at 950 °C but without BL catalyst. The obvious change in gas composition was the sharp decrease of H2 content from 53 vol.% for pure steam gasification to about 44 vol.% for the case with 5 vol.% O2 in steam and further to 33 vol.% with 10 vol.% O2. The CO2 content correspondingly increased from 11 vol.% to 25 vol.% and further to 40 vol.%, respectively. The presence of O2 in the gasification agent causes the reactions (5)–(8). Thus, CO and H2 are easy to be burned via reactions (7) and (8), while reactions (5) and (6) occur to facilitate C conversion. Consequently, the decreases in CO and H2 concentrations and increase in CO2 content show in fact that the added O2 promoted more the oxidation reactions (5), (7) and (8). The internal oxidation provides also part of the endothermic heat for steam gasification of petroleum coke, showing that a certain amount of oxygen in the steam-base gasification agent is necessary for not only the clarified effect on product gas composition but also the generation of reaction heat. The suitable oxygen amount should be subject to the required reaction heat for gasification.

C þ O2 ! CO2

DH0 ¼ 394 kJmol

1

ð5Þ 1

decreased at high temperature, and this effect is particularly evident for the non-catalytic gasification. Without BL the reaction completion time was over 160 min at 850 °C and it decreased to about 35 min at 1000 °C. The completion time of petroleum coke steam gasification in MFBRA is also shorter than in a fixed-bed reactor or TGA. The heating rate in MFBRA is up to 103–104 °C/s for fine particles in lm [22], while it also effectively removes the external diffusion limit toward reaction. With 10 wt.% BL as catalyst, the reaction completion time was only about 30% of the time required for the case without BL at each tested temperature. Thus, for achieving the same carbon conversion the gasification temperature was greatly decreased via adding 10 wt.% BL. The lower temperature is beneficial to the exothermic water gas shift reaction, thus facilitating the production of H2. It means that the adoption of BL catalyst not only catalyzed the reaction but also facilitated the production of H2 via steam gasification of petroleum coke.

C þ 0:5O2 ! CO DH0 ¼ 111 kJmol CO þ 0:5O2 ! CO2

0

ð6Þ 1

ð7Þ

1

ð8Þ

DH ¼ 283 kJmol

H2 þ 0:5O2 ! H2 O DH0 ¼ 242 kJmol

3.3. Kinetic characterization Gasification kinetics of petroleum coke gasification via steam was studied using the shrinking core model (SCM) and homogenous model (HM). The SCM suggests that gasification reaction occurs only on the surface of spherical reactant particles and the reaction rate is subject to the un-reacted surface area or remaining amount of the reactant [30,31]. When chemical reaction is the controlling step, the reaction order is 2/3 so that the reaction rate can be expressed as,

Table 3 Reaction rate of petroleum coke steam gasification with and without BL correlated according to the SCM and HM. Cases

Without BL

Temperature (°C)

SCM

850 900 950 1000

With 10 wt.% BL HM

SCM

HM

k (min1)

R2

k (min1)

R2

k (min1)

R2

k (min1)

R2

0.0136 0.0205 0.0419 0.0633

0.9878 0.9852 0.9929 0.9599

0.0252 0.0571 0.1013 0.1412

0.8670 0.9591 0.9469 0.9480

0.0414 0.0605 0.0799 0.1195

0.9962 0.9995 0.9702 0.9537

0.0799 0.1427 0.1816 0.2850

0.8748 0.9328 0.9672 0.9725

Please cite this article in press as: Zhang Y et al. Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed. Appl Energy (2015), http://dx.doi.org/10.1016/j.apenergy.2015.01.009

Y. Zhang et al. / Applied Energy xxx (2015) xxx–xxx

7

Fig. 11. SEM photographs of (a1 and a2) raw and (b1 and b2) spent petroleum coke after partial steam gasification at 950 °C (magnification: 1000 for a1 and b1 and 5000 for a2 and b2).

-2.0

 3 dX 1 ¼ kð1  X Þ2=3 or X ¼ 1  1  kt dt 3

(a)

The HM supposes that the active sites evenly distribute inside the solid particles and the particle size remains constant in the process of reaction that causes the particle density to uniformly change [32,33]. The reaction order of HM is 1, and when the chemical reaction is the controlling step we have

-2.4 -2.8

ln k

ð9Þ

-3.2

dX ¼ kð1  XÞ or X ¼ 1  ekt dt

-3.6

In Eqs. (9) and (10), X, t and k are the reaction conversion (%), reaction time (min) and reaction rate constant (min1), respectively. We further have the Arrhenius Eq. (11):

HM

-4.0

SCM -4.4 0.00078

0.00081

0.00084

0.00087

0.00090

-1

1/T [K ] -1.2

(b)

SCM HM

-1.6

ln k

-2.0

-2.4

-2.8

-3.2 0.00078

ð10Þ

0.00081

0.00084

0.00087

0.00090

-1

1/T [K ] Fig. 12. Arrhenius equation for activation energy E according to SCM and HM models for steam gasification of petroleum coke (a) without BL and (b) with 10 wt.% BL.

  E E or ln k ¼ lnðAÞ  k ¼ A exp  ; RT RT

ð11Þ

where A, E, R and T are the pre-exponential factor, activation energy (kJmol1), gas constant and temperature (K), respectively. Fig. 9 correlates the data of pure petroleum coke gasification with the SCM and HM, respectively. The reaction rate constant k was calculated from the slope of the curve 3[1  (1  X)1/3]  t in Fig. 9(a) for the SCM and from the curve of ln(1  X)  t in Fig. 9(b) for the HM. Fig. 10 shows the situation for catalytic gasification of petroleum coke based on the data from Fig. 6, with Fig. 10(a) and (b) respectively for the SCM in and HM. Table 3 summarizes the reaction rate constant k and its corresponding linear correlation factor R2 estimated from Figs. 9 and 10. It shows that both models allowed good linear fitting to make the linear correlation factor R2 be over 0.96 for the SCM and mostly about 0.90 for the HM. Obviously, the SCM allowed the better fitting than the HM did. The tested petroleum coke had very few pores and its surface area was very low, as shown by its BET and SEM analyses. This thus causes the gasification reaction of petroleum coke particles to be very close to the assumption of SCM so that the reactions between carbon and steam or ‘‘steam + oxygen’’ mainly occur on the particle

Please cite this article in press as: Zhang Y et al. Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed. Appl Energy (2015), http://dx.doi.org/10.1016/j.apenergy.2015.01.009

8

Y. Zhang et al. / Applied Energy xxx (2015) xxx–xxx

Table 4 Kinetic parameters of petroleum coke steam gasification under different conditions determined with the SCM and HM. Cases

Gasification reagent

Catalyst

Case Case Case Case

Pure steam Pure steam Steam with 5% oxygen Steam with 5% oxygen

Without BL With 10% BL Without BL With 10% BL

a b c d

surface to shrink its size gradually. Zhang [19] and Zeng et al. [25] have studied coal char gasification with steam and CO2 in TG and also found that the SCM enabled the better fitting than the HM did. Variation in the gasification agent (steam or ‘‘steam + oxygen’’) and catalyst (with or without BL) did not change this reaction nature so that the SCM showed the better fitting for all the tested conditions. The petroleum coke samples were characterized with SEM to show the morphology variation in gasification, as is shown in Fig. 11. The original samples (a1 and a2) had highly condensed layers and there were some tiny fragments on the surface. Being partially gasified by steam, the surface became greatly corrugated to show some pores. Thus, the surface area of the spent petroleum coke should be larger than the original sample. It means that the carbon gasification may initiate from the surface of nonporous grains and then extended to the pore surface within the solid [34]. Table 3 clarifies that the reaction rate constant k for the gasification of petroleum coke with BL as catalyst was higher than that of the non-catalytic gasification, especially at low conversion. For example, the k from the SCM at 850 °C was 0.0136 min1 without BL but was 0.0414 min1 with 10 wt.% BL. The gasification rate was promoted by 3–4 times by BL, agreeing with the shorter completion time shown in Fig. 6. There are also some differences for the k from the SCM and HM, showing that different kinetic models would lead to different kinetics data, but we should choose the model with the better linear fitting, that is, the SCM for the tested petroleum coke gasification. Both activation energy E and pre-exponential factor A were obtained from the Arrhenius equation. Fig. 12(a) corresponds to Fig. 9 for the non-catalytic gasification and Fig. 12(b) is for Fig. 10 of the catalytic gasification. The values of E and A obtained from Fig. 12 were listed in Table 4 for all the four cases tested. It is worthwhile to mention that the calculation procedure for the case with 5% oxygen in the gasification agent is similar to that for the cases a and b shown in Figs. 9 and 12. The non-catalytic steam gasification of petroleum coke (case a) has E of about 120 kJ/mol and 137 kJ/mol from the SCM and HM, respectively. When 10% BL was adopted (case b), the value of E decreased by about 40 kJ/mol for both the models. Karimi et al. [35] has reported E of about 210 kJ/mol for steam gasification of coke from Athabasca bitumen in TGA, and the adoption of K2CO3 and Na2CO3 as catalyst considerably decreased the activation energy to about 120 kJ/mol and 130 kJ/mol, respectively. Nemanova [36] found that E of 121.5 kJ/ mol for pure steam gasification of petroleum coke in TGA, which is similar to 120.04 kJ/mol in our tests (case a) from the SCM. The E of petroleum coke gasification from MFBRA was overall lower than that from TGA, and the reason may be related to the different effects of intra-particle diffusion on the determined kinetics in MFBRA and TGA. The gasification agent with 5% oxygen further lowered the activation energy of non-catalytic steam gasification of petroleum coke to about 100 kJ/mol for both the SCM and HM (case c). The E was only about 63 kJ/mol when BL was further used (case d). While these results agree with our knowledge that catalysis decreases activation energy, it further clarifies that the decrease in the activation energy of steam gasification by including certain

E (kJmol1)

A (min1)

SCM

HM

SCM

HM

120.04 76.55 100.43 62.97

137.46 96.59 105.92 62.54

5.26  103 153.62 2.21  103 76.65

6.84  104 453.25 7.32  103 142.59

O2 into the steam is due to the change of the reactions from having only reactions (2)–(5) to having more reactions from (2)–(8). Indeed, the gasification with O2 should be easier to occur than with steam, which makes the activation energy of gasification reactions using ‘‘steam + O2’’ be between those using O2 and steam alone. 4. Conclusions Petroleum coke gasification using black liquor (BL) as catalyst was studied in the so-called micro fluidized bed reaction analyzer (MFBRA) to characterize the steam-involved reactions and also to determine the kinetics of the reactions at 850–1000 °C. It was found that the steam gasification of petroleum coke was shortened by about 40 min through blending 10 wt.% BL into the coke as catalyst, which accounted for about 40% shorter comparing to the original reaction completion time of about 100 min for the gasification of petroleum coke without BL. The produced syngas (or fuel gas) was rich in H2 and adding 10 wt.% BL increased the H2 content of the syngas by about 9 vol.%. Parametric investigation demonstrated that the steam gasification of petroleum coke was enhanced to some extent through using small coke particles, elevating reaction temperature and including certain oxygen into the steam agent. Correlating the experimental data from MFBRA with reaction kinetic models clarified that the shrinking core model enabled the better fitting for experiments than the homogeneous model did. Estimating the reaction kinetic parameters showed that the activation energy for steam gasification of petroleum coke was reduced not only by using BL catalyst but also from including a small amount of oxygen into steam. Overall, the study also verified the technical feasibility of H2-rich syngas production from gasifying petroleum coke in fluidized bed with ‘‘steam + O2’’. Acknowledgements The study was supported by National Instrumentation Grant Program (2011YQ120039), National Nature Science Foundation of China (21406264), National Basic Research Program of China (2012CB224801) and Science Foundation of China University of Petroleum, Beijing (No. 2462013YJRC021). References [1] Murthy BN, Sawarkar AN, Deshmukh NA, Mathew T, Joshi JB. Petroleum coke gasification: a review. Can J Chem Eng 2014;92(3):441–68. [2] Speight JG. New approaches to hydroprocessing. Catal Today 2004;98(1– 2):55–60. [3] Hallale N, Liu F. Refinery hydrogen management for clean fuels production. Adv Environ Res 2001;6(1):81–98. [4] Wu Y, Wang J, Wu S. Potassium-catalyzed steam gasification of petroleum coke for H2 production: reactivity, selectivity and gas release. Fuel Process Technol 2011;92(3):523–30. [5] Zhang J, Wang Y, Dong L, Gao S, Xu G. Decoupling gasification: approach principle and technology justification. Energy Fuel 2010;24(12):6223–32. [6] Jing X, Wang Z, Yu Z, Zhang Q, Li C, Fang Y. Experimental and kinetic investigations of CO2 gasification of fine chars separated from a pilot-scale fluidized-bed gasifier. Energy Fuel 2013;27(5):2422–30. [7] Kitsuka T, Bayarsaikhan B, Sonoyama N, Hosokai S, Li C, Norinaga K, et al. Behavior of inherent metallic species as a crucial factor for kinetics of steam gasification of char from coal pyrolysis. Energy Fuel 2007;21(2):387–94.

Please cite this article in press as: Zhang Y et al. Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed. Appl Energy (2015), http://dx.doi.org/10.1016/j.apenergy.2015.01.009

Y. Zhang et al. / Applied Energy xxx (2015) xxx–xxx [8] Wang J, Jiang M, Yao Y, Zhang Y, Cao J. Steam gasification of coal char catalyzed by K2CO3 for enhanced production of hydrogen without formation of methane. Fuel 2009;88(9):1572–9. [9] Kajita M, Kimura T, Norinaga K, Li C, Hayashi J. Catalytic and noncatalytic mechanisms in steam gasification of char from the pyrolysis of biomass. Energy Fuel 2010;24(1):108–16. [10] Chen S, Yang R. Unified mechanism of alkali and alkaline earth catalyzed gasification reactions of carbon by CO2 and H2O. Energy Fuel 1997;11(2):421–7. [11] Naqvi M, Yan J, Dahlquist E. Black liquor gasification integrated in pulp and paper mills: a critical review. Bioresource Technol 2010;101(21):8001–15. [12] Naqvi M, Yan J, Dahlquist E. Bio-refinery system in a pulp mill for methanol production with comparison of pressurized black liquor gasification and dry gasification using direct causticization. Appl Energy 2012;90(1):24–31. [13] Naqvi M, Yan J, Dahlquist E. Synthetic gas production from dry black liquor gasification process using direct causticization with CO2 capture. Appl Energy 2012;97:49–55. [14] Kuang J, Zhou J, Zhou Z, Liu J, Cen K. Catalytic mechanism of sodium compounds in black liquor during gasification of coal black liquor slurry. Energy Convers Manage 2008;49(2):247–56. [15] Guo D, Wu S, Liu B, Yin X, Yang Q. Catalytic effects of NaOH and Na2CO3 additives on alkali lignin pyrolysis and gasification. Appl Energy 2012;95:22–30. [16] Zou J, Yang B, Gong K, Wu S, Zhou Z, Wang F, et al. Effect of mechanochemical treatment on petroleum coke-CO2 gasification. Fuel 2008;87:622–7. [17] Zhan X, Zhou Z, Wang W. Catalytic effect of black liquor on the gasification reactivity of petroleum coke. Appl Energy 2010;87(5):1710–5. [18] Gomez A, Mahinpey N. A new model to estimate CO2 coal gasification kinetics based only on parent coal characterization properties. Appl Energy 2015;137(1):126–33. [19] Zhang L, Huang J, Fang Y, Wang Y. Gasification reactivity and kinetics of typical Chinese anthracite chars with steam and CO2. Energy Fuel 2006;20(3):1201–10. [20] Ahmed II, Gupta AK. Kinetics of woodchips char gasification with steam and carbon dioxide. Appl Energy 2011;88(5):1613–9. [21] Ren L, Yang J, Gao F, Yan J. Laboratory study on gasification reactivity of coals and petcokes in CO2/steam at high temperatures. Energy Fuel 2013;27(9):5054–68.

9

[22] Yu J, Yue J, Liu Z, Dong L, Xu G, Zhu J, et al. Kinetics and mechanism of solid reactions in a micro fluidized bed. AIChE J 2010;11(56):2905–12. [23] Yu J, Zhu J, Xu G. Biomass pyrolysis in a micro-fluidized bed reactor: characterization and kinetics. Chem Eng J 2011;168:839–47. [24] Yu J, Zeng X, Zhang J, Zhong M, Zhang G, Wang Y, et al. Isothermal differential characteristics of gas–solid reaction in micro-fluidized bed reactor. Fuel 2013;103:29–36. [25] Zeng X, Wang F, Wang Y, Li A, Yu J, Xu G. Characterization of char gasification in a micro fluidized bed reaction analyzer. Energy Fuel 2014;28(3):1838–45. [26] Zhang Y, Yao M, Sun G, Gao S, Xu G. Characteristics and kinetics of coked catalyst regeneration via steam gasification in a micro fluidized bed. Ind Eng Chem Res 2014;53(15):6316–24. [27] Lu L, Sahajwalla V, Harris D. Characterization of chars prepared from various pulverized coals at different temperatures using drop-tube furnace. Energy Fuel 2000;14:869–76. [28] Zou J, Zhou Z, Wang F, Zhang W, Dai Z, Liu H, et al. Modeling reaction kinetics of petroleum coke gasification with CO2. Chem Eng Process 2007;46:630–6. [29] Wu Y, Wu S, Gu J, Gao J. Differences in physical properties and CO2 gasification reactivity between coal char and petroleum coke. Process Saf Environ 2009;87(5):323–30. [30] Levenspiel O. Chemical reaction engineering. Ind Eng Chem Res 1999;38(11):6140–3. [31] Dasappa S, Paul PJ, Mukunda HS, Shrinivasa U. The gasification of wood-char spheres in CO2–N2 mixtures analysis and experiments. Chem Eng Sci 1994;49(2):223–32. [32] Wen C. Noncatalytic heterogeneous solid–fluid reaction models. Ind Eng Chem 1968;60(9):34–54. [33] Ye D, Agnew JB, Zhang D. Gasification of South Australian low rank coal with carbon dioxide and steam, kinetics and reactivity studies. Fuel 1998;77(11):1209–19. [34] Bhatia SK, Perlmutter DD. A random pore model for fluid–solid reactions: I. Isothermal, kinetic control. AIChE J 1980;26(3):379–86. [35] Karimi A, Semagina N, Gray MR. Kinetics of catalytic steam gasification of bitumen coke. Fuel 2011;90(3):1285–91. [36] Nemanova V, Abedini A, Liliedahl T, Engvall K. Co-gasification of petroleum coke and biomass. Fuel 2014;117:870–5.

Please cite this article in press as: Zhang Y et al. Reactivity and kinetics for steam gasification of petroleum coke blended with black liquor in a micro fluidized bed. Appl Energy (2015), http://dx.doi.org/10.1016/j.apenergy.2015.01.009