- Email: [email protected]
ZrO2-impregnated red mud as a novel catalyst for steam catalytic cracking of vacuum residue
Fuel 165 (2016) 462–467
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
ZrO2-impregnated red mud as a novel catalyst for steam catalytic cracking of vacuum residue Hak Sung Lee, Chinh Nguyen-Huy, Thanh-Truc Pham, Eun Woo Shin ⇑ School of Chemical Engineering, University of Ulsan, Daehakro 93, Nam-gu, Ulsan 680-749, South Korea
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
ZrO2-impregnated red mud were
used for steam catalytic cracking of vacuum residue. 3 wt% ZrO2-impregnated red mud exhibited the best performance. 3 wt% ZrO2-impregnated red mud shows a high surface area and good catalyst stability. 3 wt% ZrO2-impregnated red mud improves hydrogenation activity. ZrO2-impregnated red mud has good durability for steam catalytic cracking.
a r t i c l e
i n f o
Article history: Received 17 March 2015 Received in revised form 19 October 2015 Accepted 21 October 2015
Keywords: Red mud ZrO2 Vacuum residue Catalytic cracking Steam
a b s t r a c t ZrO2-impregnated red mud catalysts were employed as a novel catalyst for catalytic cracking of vacuum residue with steam. Under the batch reaction condition at 470 °C for 2 h with superheated steam, 3 wt% ZrO2-impregnated red mud exhibited the best performance for catalytic cracking of vacuum residue. Furthermore, under the same reaction conditions, the conversion and liquid yield of 3 wt% ZrO2impregnated red mud were higher than those of 3 wt% ZrO2-supporting Al–FeOx, a well-known catalyst for catalytic cracking of heavy oil with steam. 3 wt% ZrO2-impregnated red mud also showed better catalytic performance than 3 wt% ZrO2-supporting Al–FeOx under the fixed-bed reaction conditions at 500 °C for 2 h with steam atmosphere, resulting in higher conversion as well as liquid yield. The better catalytic performance of 3 wt% ZrO2-impregnated red mud was due to large surface area and high catalyst stability. The large surface area of 3 wt% ZrO2-impregnated red mud could generate more active sites for hydrogenation, which induced higher H/C ratio in liquid product. X-ray diffraction data of the spent catalysts showed that iron oxide phase in 3 wt% ZrO2-impregnated red mud maintained a hematite structure while it in 3 wt% ZrO2-supporting Al–FeOx was transformed to magnetite, inactive phase for catalytic cracking. Ó 2015 Published by Elsevier Ltd.
1. Introduction Refineries face the problem of increasing production of heavier cuts such as vacuum gas oil (VGO) and vacuum residue (VR) [1,2]. New technology that converts the heavy hydrocarbon feedstock
⇑ Corresponding author. Tel.: +82 52 259 2253; fax: +82 52 259 1689. E-mail address: [email protected] (E.W. Shin). http://dx.doi.org/10.1016/j.fuel.2015.10.083 0016-2361/Ó 2015 Published by Elsevier Ltd.
of VR into more valuable products with lower boiling points is highly sought after. Among several processes including coking, visbreaking, and catalytic cracking, catalytic cracking with steam has attracted attention since it uses steam as an alternative, inexpensive hydrogen source [3–10]. Several studies have been reported that provide water as an alternative hydrogen source for heavy oil upgrading including aquaconversion [7,8] and supercritical water treatment [9,10]. The catalytic cracking of
463
H.S. Lee et al. / Fuel 165 (2016) 462–467
heavy oil in a steam atmosphere has been developed with iron oxide supporting ZrO2 catalysts (ZrO2–FeOx), and this process succeeded in production of lighter fuels from atmosphericdistilled residual oil and vacuum-residual oil [3–6]. Red mud (RM), a solid waste product of the Bayer process, is composed of a mixture of iron, aluminum and titanium oxides, with significant silicon, calcium and sodium oxide contents. RM has been applied as a catalyst for liquefaction of coal [11,12], biomass [13] and hydrogenating anthracene oil [14]. In our previous studies, we developed and used RM as a catalyst for slurry-phase hydrocracking of [15–17]. Its main components are iron oxides and it is a potential catalyst for catalytic cracking of VR with steam by impregnation of ZrO2. In this work, ZrO2-impregnated RM was prepared and applied as a novel catalyst for catalytic cracking of VR with steam. To our knowledge, there are no reports in the literature on the catalytic cracking of VR with steam in the presence of ZrO2-impregnated RM. Our results demonstrated that ZrO2-impregnated RM is a good catalyst for steam catalytic cracking of VR.
2.3. Catalytic cracking of VR with steam Experiments were performed in a 100 ml batch type autoclave fitted with a stirrer. The reactor was loaded with 20 g of VR, 1 g of catalyst and 20 g of deionized water. The pressure of the system was approximately 3 MPa when the system was heated to 470 °C. Water exists as steam under these conditions. Before heating, the reactor was purged with N2 to ensure an oxygen-free atmosphere. After 2 h reaction time, the reactor was cooled to room temperature using a water cooler and a fan. The gas products were released to the air. Reactor contents were centrifuged to separate the liquid products from the solid–liquid mixture. Conversion of VR and the yield of each product were calculated by balancing the content of the desired boiling point products in the following equations:
Gas yield ðwt%Þ ¼
½input VR weight ½weight of all of the products Feed VR weight 100
Liquid yield ðwt%Þ ¼
100
2. Experimental 2.1. Catalysts and materials RM was supplied by the KC Corporation (Seoul, South Korea) and activated with the Pratt and Christoverson method [18]. The activated RM was denoted as ARM. ZrO2 was impregnated into the ARM by a conventional impregnation technique using an aqueous excess solution of ZrO2 precursor. Certain amounts of ZrO (NO3)26H2O (Sigma–Aldrich, 99%) salt was dissolved in a beaker and then 5 g of ARM was added to the solution. The solution was stirred at 30 °C for 1 h and slowly heated to 50 °C while stirring constantly to evaporate water. The paste samples were dried at 100 °C in an oven for 12 h and then calcined at 550 °C for 2 h with a ramp of 1.5 °C/min. The catalysts obtained are referred to as ZrO2(x)/ARM where x is the nominal weight percentage of ZrO2 (x = 1, 2, 3, 7, 11). A ZrO2-supporting complex metal oxide of Fe and Al catalyst was also synthesized for comparison, following to the procedure in literature [4]. The support was prepared by a coprecipitation method using aqueous iron (III) chloride and aluminum sulfate (concentration of FeCl36H2O = 4.8 wt%; Al2(SO4)3.14–18H2O = 4.8 wt%), and then treated with steam at 550 °C for 1 h (denoted as Al–FeOx). ZrO2 was impregnated into the Al–FeOx by a conventional impregnation technique using an aqueous excess solution of ZrO(NO3)26H2O. The final powder product was designated as ZrO2/Al–FeOx. The ZrO2 contents in each catalyst were measured by EDX analysis and provided in Table 1. A commercial VR was used as a feedstock for the reaction tests and its properties was described in Table S1 (see the supplementary material).
2.2. Catalyst characterization The structures of the catalysts were analyzed using an X-ray diffractometer (Rigaku RAD-3C, Japan) with Cu Ka radiation (l = 1.5418 Å) at a scan rate of 2° (2h)/min, operated at 35 kV and 20 mA. The composition of catalysts was determined by an energy-dispersive X-ray spectrometer (FE-SEM–EDX, JEOL, JSM600F, Japan). The surface area of the catalysts was evaluated from nitrogen adsorption isotherms measured at 195.6 °C (Micromeritics ASAP 2020 apparatus, USA).
½weight of naphtha þ diesel þ VGO products Feed VR weight
Coke yield ðwt%Þ ¼
½weight of coke produced during the reaction Feed VR weight 100
Each desired product yield ðwt%Þ ¼
½products weight of desired boiling point 100 Feed VR weight
Conversion ðwt%Þ ¼ ½Gas yield þ ½liquid yield þ ½Coke yield The distribution of desired liquid products – naphtha (<150 °C), diesel (<350 °C), VGO (<560 °C) and 560 °C+ fraction (>560 °C) – was measured by GC-SIMDIS (AC Analytical Controls, Agilent 7890). The analysis conditions were based on ASTM D-2887 (simulated distillation) and ASTM D-5307 methods [19,20]. The hydrogen and carbon contents in liquid products were analyzed by Thermo ScientificTM FLASH 2000 CHNS/O Analyzers. To evaluate the reproducibility of the experiments, each experiment was repeated several times. The remaining product in solid form was extracted with toluene to remove soluble deposits, oil, and remaining feed [21]. The toluene-insoluble fraction consisted of coke, catalyst, and inorganic matter. After washing, the solids were dried in an oven at 120 °C for 4 h, and then ground using an agate mortar. The coke yield was determined by TGA analysis (Fig. S1 in the supplementary material). For further understanding the reaction mechanism and evaluating the catalytic activity of ZrO2-impregnated red mud, ZrO2(3)/ ARM and ZrO2(3)/Al–FeOx were applied for steam catalytic cracking of VR in a fixed-bed reactor (Fig. S2 in the supplementary material). The fixed-bed reactor was loaded with 0.5 g of catalyst; reaction temperature was 500 °C and reaction pressure was 0.1 MPa. We diluted the VR with toluene to reduce its viscosity and used the resulting solution at 10 wt% of VR as feedstock. All of catalysts were confirmed to be inactive to toluene in a preliminary experiment. The time factor W/FR was 0.6 h, where FR is the flow rate of VR without toluene and W is the amount of catalyst. A mixture of steam and nitrogen was introduced into the reactor as a carrier gas. The flow rate of nitrogen was 75 cm3/min and the flow rate of steam was defined by vaporization of 0.3 ml/min
464
H.S. Lee et al. / Fuel 165 (2016) 462–467
Table 1 Composition of fresh and spent catalysts (batch reactor) measured by EDX (wt%). Element
O
Fe
Al
Si
Na
Ca
Ti
Zr
Others (V, Cu, S, Zn)
Zr/Fe ratio
Fresh catalysts
ARM ZrO2(1)/ARM ZrO2(2)/ARM ZrO2(3)/ARM ZrO2(7)/ARM ZrO2(11)/ARM ZrO2(3)/Al–FeOx
50.4 39.6 50.7 49.9 43.4 37.3 36.0
22.8 33.7 23.9 23.0 28.1 28.8 55.2
12.2 12.6 13.2 14.1 11.4 12.2 5.5
4.1 3.9 5.5 4.5 3.9 4.9 0.3
0.3 – 0.8 – – 1.1 –
1.5 1.5 0.8 1.2 1.5 1.6 –
3.8 5.4 2.8 4.2 5.3 4.4 –
– 1.2 2.2 2.7 5.0 9.8 3.0
– – – – – – –
– 0.04 0.09 0.12 0.18 0.34 0.05
Spent catalysts
ZrO2(1)/ARM ZrO2(2)/ARM ZrO2(3)/ARM ZrO2(7)/ARM ZrO2(11)/ARM ZrO2(3)/Al–FeOx
38.7 45.9 48.3 47.1 36.3 42.1
33.4 23.1 22.5 24.8 32.8 48.9
12.4 15.5 13.5 11.5 11.2 4.7
5.0 5.0 5.8 5.4 3.6 –
– – – – – –
1.6 1.2 1.3 1.5 1.7 –
5.2 4.4 3.6 4.0 5.3 –
1.2 1.8 2. 6 3.0 5.5 2.1
2.4 3.2 2.5 2.7 3.7 2.3
0.04 0.08 0.11 0.12 0.17 0.04
water at 100 °C by tape heater. After the reaction of VR for 2 h, the liquid and gas products were separated in an ice trap and analyzed by GC-SIMDIS (AC Analytical Controls, Agilent 7890) and a thermal conductivity detector gas chromatograph with HPPLOT Q column (TCD; Acme 6100 GC, Young Lin instrument Co. Ltd., Korea), respectively. 3. Results and discussion 3.1. Catalyst characterization The nitrogen adsorption–desorption isotherms and the corresponding pore size distributions of the catalysts are shown in
Fig. 1 and their textual properties are listed in Table 2. Fig. 1a shows that the each hysteresis loop of ZrO2(x)/ARM (x = 1, 2, 3, 7 and 11) and ARM corresponds to H3 type and a combination of H3 and H4 types (IUPAC classification), which are usually found in solid powders consisting of slit-shaped pores with good pore connectivity. Additionally, the step jump observed at a high relative pressure (>0.9) indicates the macropores existence [22]. Fig. 1b shows the pore size distributions of these catalysts. On ARM, one single narrow peak is centered at 3.37 nm while ZrO2 (x)/ARM catalysts show bimodal pore size distributions centered in 3.37 and 5 nm. This observation indicates that the larger mesopores also formed during impregnation of ZrO2 onto RM. However, the surface areas decreased with increasing the ZrO2 contents, probably due to the pore blockage by impregnated ZrO2. The XRD patterns of these catalysts are compared in Fig. 2. All of the catalysts show the same XRD patterns, corresponding to Fe2O3 hematite structure (JCPDS 33-0664). Other peaks were identified as: rutile (RU) (JCPDS 21-1276), TiO2; iron titanium oxide (JCPDS 43-1011), Fe3Ti3O10. However, there is no peak corresponding to ZrO2, implying high dispersion of ZrO2 in the Fe2O3 matrix. This behavior is consistent with results reported previously for ZrO2supporting iron oxide catalysts [23]. The composition of fresh catalysts and spent catalysts (after regeneration) are given in Table 1. Actual Zr contents are similar to those intended. After reaction, Zr element in each spent catalysts was still remained. The small decrease in Zr content stems from the deposition of the other elements such as V, S, Zn and Cu. However, the Zr/Fe ratio of spent ZrO2(7)/ARM and spent ZrO2(11)/ARM drastically decreased, indicating that the ZrO2 peeled off from these two catalysts. 3.2. Catalytic cracking of VR
Fig. 1. (a) N2 adsorption–desorption isotherms of fresh catalysts and (b) their pore size distributions.
To avoid the limitations of mass transfer, the catalytic cracking of VR was conducted under superheated steam conditions. Table 3 provides the product yields and conversion of the reaction of VR along with ZrO2 contents of RM. From 1 wt% to 3 wt% ZrO2 content, the liquid yields increased from 29 wt% to 38 wt%, while VR conversions increased from 64 wt% to 85 wt%. However, when the ZrO2 amount reached 7 wt% and 11 wt%, the liquid yield and conversion became lower than those of ZrO2(3)/ARM. The reaction mechanism for this model could be suggested with four main reactions: thermal cracking, hydrogenation which was generated by iron oxides [15–17], catalytic cracking which was related to acidic sites on catalysts and oxidative cracking which was generated by ZrO2 and iron oxides [3–6]. ZrO2 promotes the generation of active oxygen and hydrogen species from H2O [3,4]. These oxygen species spill over to the surface of the iron oxide resulting in oxidative cracking of heavy oil with active oxygen species on the iron oxide [4,6]. Therefore, the catalyst performance depends on the ZrO2 con-
465
H.S. Lee et al. / Fuel 165 (2016) 462–467 Table 2 Textural characteristics of catalysts obtained by nitrogen adsorption.
Fresh catalyst Surface area (m2/g) Pore diameter (nm) Pore volume (cm3/g)
ARM
ZrO2(1)/ARM
ZrO2(2)/ARM
ZrO2(3)/ARM
ZrO2(7)/ARM
ZrO2(11)/ARM
ZrO2(3)/Al–FeOx
107 8.1 0.21
113 10.1 0.28
100 10.1 0.25
91 9.2 0.21
82 9.0 0.18
78 8.7 0.16
28 24.8 0.17
6 5.3 0.007
6 6.7 0.008
4 6.2 0.005
15 5.5 0.014
13 6.7 0.011
3 5.7 0.004
Spent catalyst (batch reactor) Surface area (m2/g) – Pore diameter (nm) – Pore volume (cm3/g) –
Fig. 2. X-ray diffractograms for fresh catalysts. Fig. 3. Effect of ZrO2 contents and catalyst surface areas on VR conversions.
tent as well as the active sites on the iron oxide. Fig. 3 shows the relationship of the ZrO2 content and the catalyst surface area with VR conversion. With increasing the ZrO2 content, the surface areas of the catalysts decreased. Consequently, the active sites on the iron oxides were covered with excessively impregnated ZrO2. Therefore, the decrease in surface area and pore plugging (shown in Fig. 1b) with increase in ZrO2 content may possibly lead to decrease in active sites on the iron oxides. These results suggest that catalytic activity to crack VR compounds was a combination of the ZrO2 content on the iron oxide and the active sites on the iron oxide surface. However, increasing ZrO2 content also led to increasing coke yield. Coke was generated by two main reactions, thermal cracking and catalytic cracking. However, only catalytic cracking, which had a definite effect on coke generation, is related to catalyst. Therefore, the reason behind the increase of coke yield can be related to the acidity of catalyst. The addition of ZrO2 to metal oxide Fe2O3 can increase the acidity of the catalysts [24]. Therefore, catalytic cracking also can be occurred and increased with the increase of ZrO2 contents, resulting in increase of coke yield and high conversion. However, acidity cannot be considered as the main reason behind the difference in coke yield when
ZrO2 amount reached 7 wt% and 11 wt%, resulting in a lower coke yield as well as conversion. The reason can be related to the decrease of surface area with increase of ZrO2 content. The decrease of surface area leads to a decrease of active sites and hence a coke yield and conversion were decreased. Furthermore, the peeling off of ZrO2 in ZrO2(7)/ARM and ZrO2(11)/ARM catalysts, which was confirmed by a decrease of Zr/Fe ratio by EDX (Table 1), possibly leads to the decrease in VR conversion. To practically evaluate the catalytic activity of ZrO2(3)/ARM, a well-known catalyst for the reaction, ZrO2(3)/Al–FeOx [4] was prepared and tested in a batch reactor and a fixed-bed reactor for comparison. The reaction results of ZrO2(3)/ARM and ZrO2(3)/Al– FeOx in a batch reactor and a fixed-bed reactor were shown in Tables 3 and 4, respectively. VR conversion and liquid yield of ZrO2(3)/ARM are higher than those of ZrO2(3)/Al–FeOx in both the case, implying that the ZrO2-impregnated RM are a comparable catalyst for steam catalytic hydrocracking of VR. The main products of the reaction were diesel and VGO, and coke formation was much suppressed in fixed-bed reactor, due to better diffusion ability of VR after dilution with toluene. The better catalytic perfor-
Table 3 Product distributions and conversions obtained from steam catalytic cracking of VR in batch reactor (reaction temp. = 470 °C, reaction time = 2 h).
Yield (wt%) Coke Gas Nap. (<150 °C) Diesel (150–350 °C) VGO (350–560 °C) 560 °C+ Liquid yield (wt%) Conversion (wt%)
ARM
ZrO2(1)/ARM
ZrO2(2)/ARM
ZrO2(3)/ARM
ZrO2(7)/ARM
ZrO2(11)/ARM
ZrO2(3)/Al–FeOx
23.9 11.3 4.3 16.2 9.1 35.2 29.6 64.8
29.3 12.4 3.7 18.9 8.4 27.3 31.0 72.7
29.2 11.2 5.1 20.6 8.9 25.0 34.6 75.0
34.4 12.7 4.3 22.3 11.5 14.8 38.1 85.2
25.7 12.0 4.3 18.3 10.0 29.7 32.6 70.3
21.6 12.5 4.4 18.2 9.1 34.2 31.7 65.8
26.9 16.2 3.0 19.3 12.9 21.7 35.2 78.3
466
H.S. Lee et al. / Fuel 165 (2016) 462–467
Table 4 Product distributions, conversions and H/C ratio obtained from steam catalytic cracking of VR in fixed-bed reactor (reaction temp. = 500 °C, reaction time = 2 h, W/FR = 0.6 h). ZrO2(3)/Al–FeOx Yield (wt%) Coke Gas Nap. (<150 °C) Diesel (150–350 °C) VGO (350–560 °C) 560 °C+ Conversion (wt%) H/C ratio of liquid products
1.1 6.0 2.8 9.2 36.4 44.5 55.5 1.54
ZrO2(3)/ARM 1.9 5.3 0.8 18.4 41.6 32.0 68.0 1.76
Table 5 Gas composition after steam catalytic cracking of VR over ZrO2(3)/ARM and ZrO2(3)/ Al–FeOx in fixed-bed reactor (reaction temp. = 500 °C, reaction time = 2 h, W/FR = 0.6 h). Products
Relative yield (mol%)
Methane Ethane Propylene Propane C4 products
ZrO2(3)/ARM
ZrO2(3)/Al–FeOx
30.8 13.0 6.5 6.3 43.4
24.4 8.3 4.7 6.4 56.2
Table 6 Product distributions and conversions obtained from steam catalytic cracking of VR in fixed-bed reactor with 2 times reaction and regeneration cycles (reaction temp. = 500 °C, reaction time = 2 h, W/FR = 0.6 h).
Yield (wt%) Coke Gas Nap. (<150 °C) Diesel (150– 350 °C) VGO (350– 560 °C) 560 °C+ Conversion (wt%)
Fig. 4. X-ray diffractograms for (a) fresh and spent catalysts of ZrO2(3)/ARM and ZrO2(3)/Al–FeOx after the steam catalytic cracking of VR in batch reactor (reaction temp. = 470 °C, reaction time = 2 h) and (b) spent catalysts of ZrO2(3)/ARM and ZrO2(3)/Al–FeOx after the steam catalytic cracking of VR in fixed-bed reactor (reaction conditions: reaction temp. = 500 °C, reaction time = 2 h, W/FR = 0.6 h).
mance of ZrO2(3)/ARM is due to the high surface area and good stability of iron oxide. XRD patterns of the spent ZrO2(3)/ARM and ZrO2(3)/Al–FeOx were examined to confirm the stability of the iron oxide phase, shown in Fig. 4. Fig. 4a and b shows the XRD patterns of spent catalysts after 2 h reaction in a batch reactor and in a fixed-bed reactor, respectively. ZrO2(3)/ARM still maintained a hematite structure after the reaction while iron oxide of ZrO2(3)/ Al–FeOx was transformed into a magnetite structure. The transformation of iron oxide phase in both of catalysts after reaction indicates that the oxidative cracking occurred. Masuda et al. [3] suggested that the lattice oxygen in the iron oxide structure was consumed by decomposing residual oils over the surface of the catalyst. Furthermore, an iron oxide matrix that was losing lattice
ZrO2(3)/ ARM
ZrO2(3)/ARM, 1st reg.
ZrO2(3)/ARM 2nd reg.
1.9 5.3 0.8 18.4
0.8 6.3 3.3 14.8
3.4 3.8 2.4 10.4
41.6
38.1
43.0
32.0 68.0
36.7 63.3
37.0 63.0
oxygen must be supplied with active oxygen species, which were generated from H2O over ZrO2, to maintain the catalyst activity [6,25]. A short supply of active oxygen species led to the transformation of the iron oxide matrix from a hematite structure to a magnetite structure, resulting in catalyst deactivation [3,6]. ZrO2(3)/ARM with a high surface area can provide better ability to supply the active oxygen species for the losing lattice oxygen in the iron oxide structure. Therefore, the catalytic stability of ZrO2(3)/ARM is better than that of ZrO2(3)/Al–FeOx because of less phase transformation from hematite to magnetite. However, the CO2 was not detectable in gas phase analysis of reaction over both catalysts in fixed-bed reactor (shown in Table 5), indicating that oxidative cracking is not dominant in this steam catalytic cracking reaction. On the other hand, iron oxide is well-known as an active phase for hydrogenation [15–17]. Table 4 shows the H/C ratio in liquid products after steam catalytic cracking of VR over different catalysts in a fixed-bed reactor. As a result, the use of the catalysts improved the quality of liquid products in view of higher H/C ratios (H/C ratio of VR is 1.34). The H/C ratio of liquid phase with ZrO2(3)/ ARM catalyst (1.76) is higher than that of liquid phase with ZrO2(3)/Al–FeOx catalyst (1.54). The increase in H/C ratio with ZrO2(3)/ARM catalyst indicates that more hydrogenation reactions occurred with this catalyst. ZrO2(3)/ARM with high surface area and good stability of iron oxide can generate more active sites for hydrogenation reaction. However, the ZrO2(3)/ARM produced more coke than ZrO2(3)/ Al–FeOx, probably due to the higher acidity of the ZrO2(3)/ARM. The ZrO2(3)/ARM contains a higher amount of Al and Si (see Table 1) and much higher surface area than those of ZrO2(3)/Al– FeOx, which can provide more acidic sites. Thus, ZrO2(3)/ARM providing higher acidity can enhance the catalytic cracking, resulting in higher coke amount and higher conversion than ZrO2(3)/Al– FeOx. The surface area of spent catalysts after reaction was provided in Table 2. All of catalysts showed a very low surface area after reaction. The loss of surface area in these catalysts could be attributed to coke deposits on pore structure of catalyst.
H.S. Lee et al. / Fuel 165 (2016) 462–467
The catalyst regeneration ability was tested with ZrO2(3)/ARM in fixed-bed reactor by repetition of reaction and regeneration. The reaction with VR was allowed to proceed for 2 h, and the used catalyst was the regenerated at 600 °C in an air stream for 2 h followed by treatment in steam for 1 h at the same temperature as that for air treatment. The product distribution and conversion obtained from 2 times reaction and regeneration cycles were shown in Table 6. The result shows that the conversion was slightly decreased after the first regeneration. However, it is almost unchanged after the second regeneration. It is therefore concluded that the ZrO2(3)/ARM catalyst shows good durability. 4. Conclusions ZrO2-impregnated red mud catalysts were applied as a novel catalyst for steam catalytic cracking of VR. Among them, ZrO2(3)/ ARM exhibited the highest catalytic activity and liquid yield for VR decomposition under a batch reaction conditions. Furthermore, ZrO2(3)/ARM showed better catalytic performance than ZrO2(3)/ Al–FeOx under the both of batch reaction conditions and fixedbed reaction conditions, which was due to the high surface area and good stability of the iron oxide structure in RM. High surface area and good stability of ZrO2(3)/ARM led to improving hydrogenation activity, resulting in higher H/C ratio in liquid product. The repetition of reaction and regeneration in fixed-bed reactor showed that ZrO2(3)/ARM catalyst also has good durability for steam catalytic cracking. Acknowledgement This work was supported by the Research Fund of University of Ulsan. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.fuel.2015.10.083. References [1] Chen S-L, Jia S-S, Luo Y-H, Zhao S-Q. Mild cracking solvent deasphalting: a new method for upgrading petroleum residue. Fuel 1994;73:439–42. [2] Dutriez T, Courtiade M, Thiébaut D, Dulot H, Bertoncini F, Vial J, et al. Hightemperature two-dimensional gas chromatography of hydrocarbons up to nC60 for analysis of vacuum gas oils. J Chromatogr A 2009;1216:2905–12.
467
[3] Fumoto E, Tago T, Tsuji T, Masuda T. Recovery of useful hydrocarbons from petroleum residual oil by catalytic cracking with steam over zirconiasupporting iron oxide catalyst. Energy Fuels 2004;18:1770–4. [4] Fumoto E, Tago T, Tsuji T, Masuda T. Production of lighter fuels by cracking petroleum residual oils with steam over zirconia-supporting iron oxide catalysts. Energy Fuels 2006;20:1–6. [5] Fumoto E, Tago T, Masuda T. Production of light oil by oxidative cracking of oil sand bitumen using iron oxide catalysts in a steam atmosphere. Energy Fuels 2011;25:524–7. [6] Funai S, Fumoto E, Tago T, Masuda T. Recovery of useful lighter fuels from petroleum residual oil by oxidative cracking with steam using iron oxide catalyst. Chem Eng Sci 2010;65:60–5. [7] Pereira P, Flores C, Zbinden H, Guitian J, Solari RB, Feintuch H, et al. Aquaconversion technology offers added value to E. Venezuela synthetic crude oil production. Oil Gas J 2001;99:79–85. [8] Thomas M, Fixari B, Perchec PL, Princic Y, Lena L. Visbreaking of Safaniya vacuum residue in the presence of additives. Fuel 1989;68:318–22. [9] Dejhosseini M, Aida T, Watanabe M, Takami S, Hojo D, Aoki N, et al. Catalytic cracking reaction of heavy oil in the presence of cerium oxide nanoparticles in supercritical water. Energy Fuels 2013;27(8):4624–31. [10] Cheng Z-M, Ding Y, Zhao L-Q, Yuan P-Q, Yuan W-K. Effects of supercritical water in vacuum residue upgrading. Energy Fuels 2009;23:3178–83. [11] Mastral AM, Mayoral C, Izquierdo MT, Pardos C. Iron catalyzed hydrogenation of high sulphur content coals. Fuel Process Technol 1993;36:177–84. [12] Yokoyama S, Yamamoto M, Maekawa Y, Kotanigawa T. Catalytic activity of sulphate for hydroliquefaction of coal by using diphenylether and diphenylmethane. Fuel 1989;68:531–3. [13] Klopries B, Hodek W, Bandermann F. Catalytic hydroliquefaction of biomass with red mud and COO–MOO3 catalysts. Fuel 1990;69:448–55. [14] Llano JJ, Rosal R, Sastre H, Diez FV. Catalytic hydrogenation of anthracene oil with red mud. Fuel 1994;73:688–94. [15] Nguyen-Huy C, Kweon H, Kim H, Kim DK, Kim D-W, Oh SH, et al. Slurry-phase hydrocracking of vacuum residue with a disposable red mud catalyst. Appl Catal A 2012;447–448:186–92. [16] Nguyen-Huy C, Kim H, Kweon H, Kim DK, Kim D-W, Oh SH, et al. Modification of disposable red mud catalysts for slurry-phase hydrocracking of vacuum residue. Chem Eng Technol 2013;36:1365–70. [17] Nguyen-Huy C, Pham VH, Kim DK, Kim D-W, Oh SH, Chung JS, et al. Role of macroporosity in macro-mesoporous red mud catalysts for slurry-phase hydrocracking of vacuum residue. Appl Catal A 2013;468:305–12. [18] Pratt KC, Christoverson V. Hydrogenation of a model hydrogen-donor system using activated red mud catalyst. Fuel 1982;61:460–2. [19] ASTM D 2887. Annual book of ASTM standards, vol. 05.02. Philadelphia; 1989. [20] ASTM D 5307. Annual book of ASTM standards, vol. 05.02. Philadelphia; 1992. [21] Matsumura A, Sato S, Kondo T, Saito I, de Souza WF. Hydrocracking Marlim vacuum residue with natural limonite. Part 2: experimental cracking in a slurry-type continuous reactor. Fuel 2005;84:417–21. [22] Sing KSW. Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity. Pure Appl Chem 1985;57:603–19. [23] Fumoto E, Mizutani Y, Tago T, Masuda T. Production of ketones from sewage sludge over zirconia-supporting iron oxide catalysts in a steam atmosphere. Appl Catal B 2006;68:154–9. [24] Jung-Chung W, Der-Sun L, An-Nan K. Dehydrogenation of ethylbenzene over TiO2–Fe2O3 and ZrO2–Fe2O3 mixed oxide catalysts. Catal Lett 1993;20:191–201. [25] Sonoyama N, Nobuta K, Kimura T, Hosokai S, Hayashi J, Tago T, et al. Production of chemicals by cracking pyrolytic tar from Loy Yang coal over iron oxide catalysts in a steam atmosphere. Fuel Process Technol 2011;92:771–5.
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
ZrO2-impregnated red mud as a novel catalyst for steam catalytic cracking of vacuum residue Hak Sung Lee, Chinh Nguyen-Huy, Thanh-Truc Pham, Eun Woo Shin ⇑ School of Chemical Engineering, University of Ulsan, Daehakro 93, Nam-gu, Ulsan 680-749, South Korea
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
ZrO2-impregnated red mud were
used for steam catalytic cracking of vacuum residue. 3 wt% ZrO2-impregnated red mud exhibited the best performance. 3 wt% ZrO2-impregnated red mud shows a high surface area and good catalyst stability. 3 wt% ZrO2-impregnated red mud improves hydrogenation activity. ZrO2-impregnated red mud has good durability for steam catalytic cracking.
a r t i c l e
i n f o
Article history: Received 17 March 2015 Received in revised form 19 October 2015 Accepted 21 October 2015
Keywords: Red mud ZrO2 Vacuum residue Catalytic cracking Steam
a b s t r a c t ZrO2-impregnated red mud catalysts were employed as a novel catalyst for catalytic cracking of vacuum residue with steam. Under the batch reaction condition at 470 °C for 2 h with superheated steam, 3 wt% ZrO2-impregnated red mud exhibited the best performance for catalytic cracking of vacuum residue. Furthermore, under the same reaction conditions, the conversion and liquid yield of 3 wt% ZrO2impregnated red mud were higher than those of 3 wt% ZrO2-supporting Al–FeOx, a well-known catalyst for catalytic cracking of heavy oil with steam. 3 wt% ZrO2-impregnated red mud also showed better catalytic performance than 3 wt% ZrO2-supporting Al–FeOx under the fixed-bed reaction conditions at 500 °C for 2 h with steam atmosphere, resulting in higher conversion as well as liquid yield. The better catalytic performance of 3 wt% ZrO2-impregnated red mud was due to large surface area and high catalyst stability. The large surface area of 3 wt% ZrO2-impregnated red mud could generate more active sites for hydrogenation, which induced higher H/C ratio in liquid product. X-ray diffraction data of the spent catalysts showed that iron oxide phase in 3 wt% ZrO2-impregnated red mud maintained a hematite structure while it in 3 wt% ZrO2-supporting Al–FeOx was transformed to magnetite, inactive phase for catalytic cracking. Ó 2015 Published by Elsevier Ltd.
1. Introduction Refineries face the problem of increasing production of heavier cuts such as vacuum gas oil (VGO) and vacuum residue (VR) [1,2]. New technology that converts the heavy hydrocarbon feedstock
⇑ Corresponding author. Tel.: +82 52 259 2253; fax: +82 52 259 1689. E-mail address: [email protected] (E.W. Shin). http://dx.doi.org/10.1016/j.fuel.2015.10.083 0016-2361/Ó 2015 Published by Elsevier Ltd.
of VR into more valuable products with lower boiling points is highly sought after. Among several processes including coking, visbreaking, and catalytic cracking, catalytic cracking with steam has attracted attention since it uses steam as an alternative, inexpensive hydrogen source [3–10]. Several studies have been reported that provide water as an alternative hydrogen source for heavy oil upgrading including aquaconversion [7,8] and supercritical water treatment [9,10]. The catalytic cracking of
463
H.S. Lee et al. / Fuel 165 (2016) 462–467
heavy oil in a steam atmosphere has been developed with iron oxide supporting ZrO2 catalysts (ZrO2–FeOx), and this process succeeded in production of lighter fuels from atmosphericdistilled residual oil and vacuum-residual oil [3–6]. Red mud (RM), a solid waste product of the Bayer process, is composed of a mixture of iron, aluminum and titanium oxides, with significant silicon, calcium and sodium oxide contents. RM has been applied as a catalyst for liquefaction of coal [11,12], biomass [13] and hydrogenating anthracene oil [14]. In our previous studies, we developed and used RM as a catalyst for slurry-phase hydrocracking of [15–17]. Its main components are iron oxides and it is a potential catalyst for catalytic cracking of VR with steam by impregnation of ZrO2. In this work, ZrO2-impregnated RM was prepared and applied as a novel catalyst for catalytic cracking of VR with steam. To our knowledge, there are no reports in the literature on the catalytic cracking of VR with steam in the presence of ZrO2-impregnated RM. Our results demonstrated that ZrO2-impregnated RM is a good catalyst for steam catalytic cracking of VR.
2.3. Catalytic cracking of VR with steam Experiments were performed in a 100 ml batch type autoclave fitted with a stirrer. The reactor was loaded with 20 g of VR, 1 g of catalyst and 20 g of deionized water. The pressure of the system was approximately 3 MPa when the system was heated to 470 °C. Water exists as steam under these conditions. Before heating, the reactor was purged with N2 to ensure an oxygen-free atmosphere. After 2 h reaction time, the reactor was cooled to room temperature using a water cooler and a fan. The gas products were released to the air. Reactor contents were centrifuged to separate the liquid products from the solid–liquid mixture. Conversion of VR and the yield of each product were calculated by balancing the content of the desired boiling point products in the following equations:
Gas yield ðwt%Þ ¼
½input VR weight ½weight of all of the products Feed VR weight 100
Liquid yield ðwt%Þ ¼
100
2. Experimental 2.1. Catalysts and materials RM was supplied by the KC Corporation (Seoul, South Korea) and activated with the Pratt and Christoverson method [18]. The activated RM was denoted as ARM. ZrO2 was impregnated into the ARM by a conventional impregnation technique using an aqueous excess solution of ZrO2 precursor. Certain amounts of ZrO (NO3)26H2O (Sigma–Aldrich, 99%) salt was dissolved in a beaker and then 5 g of ARM was added to the solution. The solution was stirred at 30 °C for 1 h and slowly heated to 50 °C while stirring constantly to evaporate water. The paste samples were dried at 100 °C in an oven for 12 h and then calcined at 550 °C for 2 h with a ramp of 1.5 °C/min. The catalysts obtained are referred to as ZrO2(x)/ARM where x is the nominal weight percentage of ZrO2 (x = 1, 2, 3, 7, 11). A ZrO2-supporting complex metal oxide of Fe and Al catalyst was also synthesized for comparison, following to the procedure in literature [4]. The support was prepared by a coprecipitation method using aqueous iron (III) chloride and aluminum sulfate (concentration of FeCl36H2O = 4.8 wt%; Al2(SO4)3.14–18H2O = 4.8 wt%), and then treated with steam at 550 °C for 1 h (denoted as Al–FeOx). ZrO2 was impregnated into the Al–FeOx by a conventional impregnation technique using an aqueous excess solution of ZrO(NO3)26H2O. The final powder product was designated as ZrO2/Al–FeOx. The ZrO2 contents in each catalyst were measured by EDX analysis and provided in Table 1. A commercial VR was used as a feedstock for the reaction tests and its properties was described in Table S1 (see the supplementary material).
2.2. Catalyst characterization The structures of the catalysts were analyzed using an X-ray diffractometer (Rigaku RAD-3C, Japan) with Cu Ka radiation (l = 1.5418 Å) at a scan rate of 2° (2h)/min, operated at 35 kV and 20 mA. The composition of catalysts was determined by an energy-dispersive X-ray spectrometer (FE-SEM–EDX, JEOL, JSM600F, Japan). The surface area of the catalysts was evaluated from nitrogen adsorption isotherms measured at 195.6 °C (Micromeritics ASAP 2020 apparatus, USA).
½weight of naphtha þ diesel þ VGO products Feed VR weight
Coke yield ðwt%Þ ¼
½weight of coke produced during the reaction Feed VR weight 100
Each desired product yield ðwt%Þ ¼
½products weight of desired boiling point 100 Feed VR weight
Conversion ðwt%Þ ¼ ½Gas yield þ ½liquid yield þ ½Coke yield The distribution of desired liquid products – naphtha (<150 °C), diesel (<350 °C), VGO (<560 °C) and 560 °C+ fraction (>560 °C) – was measured by GC-SIMDIS (AC Analytical Controls, Agilent 7890). The analysis conditions were based on ASTM D-2887 (simulated distillation) and ASTM D-5307 methods [19,20]. The hydrogen and carbon contents in liquid products were analyzed by Thermo ScientificTM FLASH 2000 CHNS/O Analyzers. To evaluate the reproducibility of the experiments, each experiment was repeated several times. The remaining product in solid form was extracted with toluene to remove soluble deposits, oil, and remaining feed [21]. The toluene-insoluble fraction consisted of coke, catalyst, and inorganic matter. After washing, the solids were dried in an oven at 120 °C for 4 h, and then ground using an agate mortar. The coke yield was determined by TGA analysis (Fig. S1 in the supplementary material). For further understanding the reaction mechanism and evaluating the catalytic activity of ZrO2-impregnated red mud, ZrO2(3)/ ARM and ZrO2(3)/Al–FeOx were applied for steam catalytic cracking of VR in a fixed-bed reactor (Fig. S2 in the supplementary material). The fixed-bed reactor was loaded with 0.5 g of catalyst; reaction temperature was 500 °C and reaction pressure was 0.1 MPa. We diluted the VR with toluene to reduce its viscosity and used the resulting solution at 10 wt% of VR as feedstock. All of catalysts were confirmed to be inactive to toluene in a preliminary experiment. The time factor W/FR was 0.6 h, where FR is the flow rate of VR without toluene and W is the amount of catalyst. A mixture of steam and nitrogen was introduced into the reactor as a carrier gas. The flow rate of nitrogen was 75 cm3/min and the flow rate of steam was defined by vaporization of 0.3 ml/min
464
H.S. Lee et al. / Fuel 165 (2016) 462–467
Table 1 Composition of fresh and spent catalysts (batch reactor) measured by EDX (wt%). Element
O
Fe
Al
Si
Na
Ca
Ti
Zr
Others (V, Cu, S, Zn)
Zr/Fe ratio
Fresh catalysts
ARM ZrO2(1)/ARM ZrO2(2)/ARM ZrO2(3)/ARM ZrO2(7)/ARM ZrO2(11)/ARM ZrO2(3)/Al–FeOx
50.4 39.6 50.7 49.9 43.4 37.3 36.0
22.8 33.7 23.9 23.0 28.1 28.8 55.2
12.2 12.6 13.2 14.1 11.4 12.2 5.5
4.1 3.9 5.5 4.5 3.9 4.9 0.3
0.3 – 0.8 – – 1.1 –
1.5 1.5 0.8 1.2 1.5 1.6 –
3.8 5.4 2.8 4.2 5.3 4.4 –
– 1.2 2.2 2.7 5.0 9.8 3.0
– – – – – – –
– 0.04 0.09 0.12 0.18 0.34 0.05
Spent catalysts
ZrO2(1)/ARM ZrO2(2)/ARM ZrO2(3)/ARM ZrO2(7)/ARM ZrO2(11)/ARM ZrO2(3)/Al–FeOx
38.7 45.9 48.3 47.1 36.3 42.1
33.4 23.1 22.5 24.8 32.8 48.9
12.4 15.5 13.5 11.5 11.2 4.7
5.0 5.0 5.8 5.4 3.6 –
– – – – – –
1.6 1.2 1.3 1.5 1.7 –
5.2 4.4 3.6 4.0 5.3 –
1.2 1.8 2. 6 3.0 5.5 2.1
2.4 3.2 2.5 2.7 3.7 2.3
0.04 0.08 0.11 0.12 0.17 0.04
water at 100 °C by tape heater. After the reaction of VR for 2 h, the liquid and gas products were separated in an ice trap and analyzed by GC-SIMDIS (AC Analytical Controls, Agilent 7890) and a thermal conductivity detector gas chromatograph with HPPLOT Q column (TCD; Acme 6100 GC, Young Lin instrument Co. Ltd., Korea), respectively. 3. Results and discussion 3.1. Catalyst characterization The nitrogen adsorption–desorption isotherms and the corresponding pore size distributions of the catalysts are shown in
Fig. 1 and their textual properties are listed in Table 2. Fig. 1a shows that the each hysteresis loop of ZrO2(x)/ARM (x = 1, 2, 3, 7 and 11) and ARM corresponds to H3 type and a combination of H3 and H4 types (IUPAC classification), which are usually found in solid powders consisting of slit-shaped pores with good pore connectivity. Additionally, the step jump observed at a high relative pressure (>0.9) indicates the macropores existence [22]. Fig. 1b shows the pore size distributions of these catalysts. On ARM, one single narrow peak is centered at 3.37 nm while ZrO2 (x)/ARM catalysts show bimodal pore size distributions centered in 3.37 and 5 nm. This observation indicates that the larger mesopores also formed during impregnation of ZrO2 onto RM. However, the surface areas decreased with increasing the ZrO2 contents, probably due to the pore blockage by impregnated ZrO2. The XRD patterns of these catalysts are compared in Fig. 2. All of the catalysts show the same XRD patterns, corresponding to Fe2O3 hematite structure (JCPDS 33-0664). Other peaks were identified as: rutile (RU) (JCPDS 21-1276), TiO2; iron titanium oxide (JCPDS 43-1011), Fe3Ti3O10. However, there is no peak corresponding to ZrO2, implying high dispersion of ZrO2 in the Fe2O3 matrix. This behavior is consistent with results reported previously for ZrO2supporting iron oxide catalysts [23]. The composition of fresh catalysts and spent catalysts (after regeneration) are given in Table 1. Actual Zr contents are similar to those intended. After reaction, Zr element in each spent catalysts was still remained. The small decrease in Zr content stems from the deposition of the other elements such as V, S, Zn and Cu. However, the Zr/Fe ratio of spent ZrO2(7)/ARM and spent ZrO2(11)/ARM drastically decreased, indicating that the ZrO2 peeled off from these two catalysts. 3.2. Catalytic cracking of VR
Fig. 1. (a) N2 adsorption–desorption isotherms of fresh catalysts and (b) their pore size distributions.
To avoid the limitations of mass transfer, the catalytic cracking of VR was conducted under superheated steam conditions. Table 3 provides the product yields and conversion of the reaction of VR along with ZrO2 contents of RM. From 1 wt% to 3 wt% ZrO2 content, the liquid yields increased from 29 wt% to 38 wt%, while VR conversions increased from 64 wt% to 85 wt%. However, when the ZrO2 amount reached 7 wt% and 11 wt%, the liquid yield and conversion became lower than those of ZrO2(3)/ARM. The reaction mechanism for this model could be suggested with four main reactions: thermal cracking, hydrogenation which was generated by iron oxides [15–17], catalytic cracking which was related to acidic sites on catalysts and oxidative cracking which was generated by ZrO2 and iron oxides [3–6]. ZrO2 promotes the generation of active oxygen and hydrogen species from H2O [3,4]. These oxygen species spill over to the surface of the iron oxide resulting in oxidative cracking of heavy oil with active oxygen species on the iron oxide [4,6]. Therefore, the catalyst performance depends on the ZrO2 con-
465
H.S. Lee et al. / Fuel 165 (2016) 462–467 Table 2 Textural characteristics of catalysts obtained by nitrogen adsorption.
Fresh catalyst Surface area (m2/g) Pore diameter (nm) Pore volume (cm3/g)
ARM
ZrO2(1)/ARM
ZrO2(2)/ARM
ZrO2(3)/ARM
ZrO2(7)/ARM
ZrO2(11)/ARM
ZrO2(3)/Al–FeOx
107 8.1 0.21
113 10.1 0.28
100 10.1 0.25
91 9.2 0.21
82 9.0 0.18
78 8.7 0.16
28 24.8 0.17
6 5.3 0.007
6 6.7 0.008
4 6.2 0.005
15 5.5 0.014
13 6.7 0.011
3 5.7 0.004
Spent catalyst (batch reactor) Surface area (m2/g) – Pore diameter (nm) – Pore volume (cm3/g) –
Fig. 2. X-ray diffractograms for fresh catalysts. Fig. 3. Effect of ZrO2 contents and catalyst surface areas on VR conversions.
tent as well as the active sites on the iron oxide. Fig. 3 shows the relationship of the ZrO2 content and the catalyst surface area with VR conversion. With increasing the ZrO2 content, the surface areas of the catalysts decreased. Consequently, the active sites on the iron oxides were covered with excessively impregnated ZrO2. Therefore, the decrease in surface area and pore plugging (shown in Fig. 1b) with increase in ZrO2 content may possibly lead to decrease in active sites on the iron oxides. These results suggest that catalytic activity to crack VR compounds was a combination of the ZrO2 content on the iron oxide and the active sites on the iron oxide surface. However, increasing ZrO2 content also led to increasing coke yield. Coke was generated by two main reactions, thermal cracking and catalytic cracking. However, only catalytic cracking, which had a definite effect on coke generation, is related to catalyst. Therefore, the reason behind the increase of coke yield can be related to the acidity of catalyst. The addition of ZrO2 to metal oxide Fe2O3 can increase the acidity of the catalysts [24]. Therefore, catalytic cracking also can be occurred and increased with the increase of ZrO2 contents, resulting in increase of coke yield and high conversion. However, acidity cannot be considered as the main reason behind the difference in coke yield when
ZrO2 amount reached 7 wt% and 11 wt%, resulting in a lower coke yield as well as conversion. The reason can be related to the decrease of surface area with increase of ZrO2 content. The decrease of surface area leads to a decrease of active sites and hence a coke yield and conversion were decreased. Furthermore, the peeling off of ZrO2 in ZrO2(7)/ARM and ZrO2(11)/ARM catalysts, which was confirmed by a decrease of Zr/Fe ratio by EDX (Table 1), possibly leads to the decrease in VR conversion. To practically evaluate the catalytic activity of ZrO2(3)/ARM, a well-known catalyst for the reaction, ZrO2(3)/Al–FeOx [4] was prepared and tested in a batch reactor and a fixed-bed reactor for comparison. The reaction results of ZrO2(3)/ARM and ZrO2(3)/Al– FeOx in a batch reactor and a fixed-bed reactor were shown in Tables 3 and 4, respectively. VR conversion and liquid yield of ZrO2(3)/ARM are higher than those of ZrO2(3)/Al–FeOx in both the case, implying that the ZrO2-impregnated RM are a comparable catalyst for steam catalytic hydrocracking of VR. The main products of the reaction were diesel and VGO, and coke formation was much suppressed in fixed-bed reactor, due to better diffusion ability of VR after dilution with toluene. The better catalytic perfor-
Table 3 Product distributions and conversions obtained from steam catalytic cracking of VR in batch reactor (reaction temp. = 470 °C, reaction time = 2 h).
Yield (wt%) Coke Gas Nap. (<150 °C) Diesel (150–350 °C) VGO (350–560 °C) 560 °C+ Liquid yield (wt%) Conversion (wt%)
ARM
ZrO2(1)/ARM
ZrO2(2)/ARM
ZrO2(3)/ARM
ZrO2(7)/ARM
ZrO2(11)/ARM
ZrO2(3)/Al–FeOx
23.9 11.3 4.3 16.2 9.1 35.2 29.6 64.8
29.3 12.4 3.7 18.9 8.4 27.3 31.0 72.7
29.2 11.2 5.1 20.6 8.9 25.0 34.6 75.0
34.4 12.7 4.3 22.3 11.5 14.8 38.1 85.2
25.7 12.0 4.3 18.3 10.0 29.7 32.6 70.3
21.6 12.5 4.4 18.2 9.1 34.2 31.7 65.8
26.9 16.2 3.0 19.3 12.9 21.7 35.2 78.3
466
H.S. Lee et al. / Fuel 165 (2016) 462–467
Table 4 Product distributions, conversions and H/C ratio obtained from steam catalytic cracking of VR in fixed-bed reactor (reaction temp. = 500 °C, reaction time = 2 h, W/FR = 0.6 h). ZrO2(3)/Al–FeOx Yield (wt%) Coke Gas Nap. (<150 °C) Diesel (150–350 °C) VGO (350–560 °C) 560 °C+ Conversion (wt%) H/C ratio of liquid products
1.1 6.0 2.8 9.2 36.4 44.5 55.5 1.54
ZrO2(3)/ARM 1.9 5.3 0.8 18.4 41.6 32.0 68.0 1.76
Table 5 Gas composition after steam catalytic cracking of VR over ZrO2(3)/ARM and ZrO2(3)/ Al–FeOx in fixed-bed reactor (reaction temp. = 500 °C, reaction time = 2 h, W/FR = 0.6 h). Products
Relative yield (mol%)
Methane Ethane Propylene Propane C4 products
ZrO2(3)/ARM
ZrO2(3)/Al–FeOx
30.8 13.0 6.5 6.3 43.4
24.4 8.3 4.7 6.4 56.2
Table 6 Product distributions and conversions obtained from steam catalytic cracking of VR in fixed-bed reactor with 2 times reaction and regeneration cycles (reaction temp. = 500 °C, reaction time = 2 h, W/FR = 0.6 h).
Yield (wt%) Coke Gas Nap. (<150 °C) Diesel (150– 350 °C) VGO (350– 560 °C) 560 °C+ Conversion (wt%)
Fig. 4. X-ray diffractograms for (a) fresh and spent catalysts of ZrO2(3)/ARM and ZrO2(3)/Al–FeOx after the steam catalytic cracking of VR in batch reactor (reaction temp. = 470 °C, reaction time = 2 h) and (b) spent catalysts of ZrO2(3)/ARM and ZrO2(3)/Al–FeOx after the steam catalytic cracking of VR in fixed-bed reactor (reaction conditions: reaction temp. = 500 °C, reaction time = 2 h, W/FR = 0.6 h).
mance of ZrO2(3)/ARM is due to the high surface area and good stability of iron oxide. XRD patterns of the spent ZrO2(3)/ARM and ZrO2(3)/Al–FeOx were examined to confirm the stability of the iron oxide phase, shown in Fig. 4. Fig. 4a and b shows the XRD patterns of spent catalysts after 2 h reaction in a batch reactor and in a fixed-bed reactor, respectively. ZrO2(3)/ARM still maintained a hematite structure after the reaction while iron oxide of ZrO2(3)/ Al–FeOx was transformed into a magnetite structure. The transformation of iron oxide phase in both of catalysts after reaction indicates that the oxidative cracking occurred. Masuda et al. [3] suggested that the lattice oxygen in the iron oxide structure was consumed by decomposing residual oils over the surface of the catalyst. Furthermore, an iron oxide matrix that was losing lattice
ZrO2(3)/ ARM
ZrO2(3)/ARM, 1st reg.
ZrO2(3)/ARM 2nd reg.
1.9 5.3 0.8 18.4
0.8 6.3 3.3 14.8
3.4 3.8 2.4 10.4
41.6
38.1
43.0
32.0 68.0
36.7 63.3
37.0 63.0
oxygen must be supplied with active oxygen species, which were generated from H2O over ZrO2, to maintain the catalyst activity [6,25]. A short supply of active oxygen species led to the transformation of the iron oxide matrix from a hematite structure to a magnetite structure, resulting in catalyst deactivation [3,6]. ZrO2(3)/ARM with a high surface area can provide better ability to supply the active oxygen species for the losing lattice oxygen in the iron oxide structure. Therefore, the catalytic stability of ZrO2(3)/ARM is better than that of ZrO2(3)/Al–FeOx because of less phase transformation from hematite to magnetite. However, the CO2 was not detectable in gas phase analysis of reaction over both catalysts in fixed-bed reactor (shown in Table 5), indicating that oxidative cracking is not dominant in this steam catalytic cracking reaction. On the other hand, iron oxide is well-known as an active phase for hydrogenation [15–17]. Table 4 shows the H/C ratio in liquid products after steam catalytic cracking of VR over different catalysts in a fixed-bed reactor. As a result, the use of the catalysts improved the quality of liquid products in view of higher H/C ratios (H/C ratio of VR is 1.34). The H/C ratio of liquid phase with ZrO2(3)/ ARM catalyst (1.76) is higher than that of liquid phase with ZrO2(3)/Al–FeOx catalyst (1.54). The increase in H/C ratio with ZrO2(3)/ARM catalyst indicates that more hydrogenation reactions occurred with this catalyst. ZrO2(3)/ARM with high surface area and good stability of iron oxide can generate more active sites for hydrogenation reaction. However, the ZrO2(3)/ARM produced more coke than ZrO2(3)/ Al–FeOx, probably due to the higher acidity of the ZrO2(3)/ARM. The ZrO2(3)/ARM contains a higher amount of Al and Si (see Table 1) and much higher surface area than those of ZrO2(3)/Al– FeOx, which can provide more acidic sites. Thus, ZrO2(3)/ARM providing higher acidity can enhance the catalytic cracking, resulting in higher coke amount and higher conversion than ZrO2(3)/Al– FeOx. The surface area of spent catalysts after reaction was provided in Table 2. All of catalysts showed a very low surface area after reaction. The loss of surface area in these catalysts could be attributed to coke deposits on pore structure of catalyst.
H.S. Lee et al. / Fuel 165 (2016) 462–467
The catalyst regeneration ability was tested with ZrO2(3)/ARM in fixed-bed reactor by repetition of reaction and regeneration. The reaction with VR was allowed to proceed for 2 h, and the used catalyst was the regenerated at 600 °C in an air stream for 2 h followed by treatment in steam for 1 h at the same temperature as that for air treatment. The product distribution and conversion obtained from 2 times reaction and regeneration cycles were shown in Table 6. The result shows that the conversion was slightly decreased after the first regeneration. However, it is almost unchanged after the second regeneration. It is therefore concluded that the ZrO2(3)/ARM catalyst shows good durability. 4. Conclusions ZrO2-impregnated red mud catalysts were applied as a novel catalyst for steam catalytic cracking of VR. Among them, ZrO2(3)/ ARM exhibited the highest catalytic activity and liquid yield for VR decomposition under a batch reaction conditions. Furthermore, ZrO2(3)/ARM showed better catalytic performance than ZrO2(3)/ Al–FeOx under the both of batch reaction conditions and fixedbed reaction conditions, which was due to the high surface area and good stability of the iron oxide structure in RM. High surface area and good stability of ZrO2(3)/ARM led to improving hydrogenation activity, resulting in higher H/C ratio in liquid product. The repetition of reaction and regeneration in fixed-bed reactor showed that ZrO2(3)/ARM catalyst also has good durability for steam catalytic cracking. Acknowledgement This work was supported by the Research Fund of University of Ulsan. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.fuel.2015.10.083. References [1] Chen S-L, Jia S-S, Luo Y-H, Zhao S-Q. Mild cracking solvent deasphalting: a new method for upgrading petroleum residue. Fuel 1994;73:439–42. [2] Dutriez T, Courtiade M, Thiébaut D, Dulot H, Bertoncini F, Vial J, et al. Hightemperature two-dimensional gas chromatography of hydrocarbons up to nC60 for analysis of vacuum gas oils. J Chromatogr A 2009;1216:2905–12.
467
[3] Fumoto E, Tago T, Tsuji T, Masuda T. Recovery of useful hydrocarbons from petroleum residual oil by catalytic cracking with steam over zirconiasupporting iron oxide catalyst. Energy Fuels 2004;18:1770–4. [4] Fumoto E, Tago T, Tsuji T, Masuda T. Production of lighter fuels by cracking petroleum residual oils with steam over zirconia-supporting iron oxide catalysts. Energy Fuels 2006;20:1–6. [5] Fumoto E, Tago T, Masuda T. Production of light oil by oxidative cracking of oil sand bitumen using iron oxide catalysts in a steam atmosphere. Energy Fuels 2011;25:524–7. [6] Funai S, Fumoto E, Tago T, Masuda T. Recovery of useful lighter fuels from petroleum residual oil by oxidative cracking with steam using iron oxide catalyst. Chem Eng Sci 2010;65:60–5. [7] Pereira P, Flores C, Zbinden H, Guitian J, Solari RB, Feintuch H, et al. Aquaconversion technology offers added value to E. Venezuela synthetic crude oil production. Oil Gas J 2001;99:79–85. [8] Thomas M, Fixari B, Perchec PL, Princic Y, Lena L. Visbreaking of Safaniya vacuum residue in the presence of additives. Fuel 1989;68:318–22. [9] Dejhosseini M, Aida T, Watanabe M, Takami S, Hojo D, Aoki N, et al. Catalytic cracking reaction of heavy oil in the presence of cerium oxide nanoparticles in supercritical water. Energy Fuels 2013;27(8):4624–31. [10] Cheng Z-M, Ding Y, Zhao L-Q, Yuan P-Q, Yuan W-K. Effects of supercritical water in vacuum residue upgrading. Energy Fuels 2009;23:3178–83. [11] Mastral AM, Mayoral C, Izquierdo MT, Pardos C. Iron catalyzed hydrogenation of high sulphur content coals. Fuel Process Technol 1993;36:177–84. [12] Yokoyama S, Yamamoto M, Maekawa Y, Kotanigawa T. Catalytic activity of sulphate for hydroliquefaction of coal by using diphenylether and diphenylmethane. Fuel 1989;68:531–3. [13] Klopries B, Hodek W, Bandermann F. Catalytic hydroliquefaction of biomass with red mud and COO–MOO3 catalysts. Fuel 1990;69:448–55. [14] Llano JJ, Rosal R, Sastre H, Diez FV. Catalytic hydrogenation of anthracene oil with red mud. Fuel 1994;73:688–94. [15] Nguyen-Huy C, Kweon H, Kim H, Kim DK, Kim D-W, Oh SH, et al. Slurry-phase hydrocracking of vacuum residue with a disposable red mud catalyst. Appl Catal A 2012;447–448:186–92. [16] Nguyen-Huy C, Kim H, Kweon H, Kim DK, Kim D-W, Oh SH, et al. Modification of disposable red mud catalysts for slurry-phase hydrocracking of vacuum residue. Chem Eng Technol 2013;36:1365–70. [17] Nguyen-Huy C, Pham VH, Kim DK, Kim D-W, Oh SH, Chung JS, et al. Role of macroporosity in macro-mesoporous red mud catalysts for slurry-phase hydrocracking of vacuum residue. Appl Catal A 2013;468:305–12. [18] Pratt KC, Christoverson V. Hydrogenation of a model hydrogen-donor system using activated red mud catalyst. Fuel 1982;61:460–2. [19] ASTM D 2887. Annual book of ASTM standards, vol. 05.02. Philadelphia; 1989. [20] ASTM D 5307. Annual book of ASTM standards, vol. 05.02. Philadelphia; 1992. [21] Matsumura A, Sato S, Kondo T, Saito I, de Souza WF. Hydrocracking Marlim vacuum residue with natural limonite. Part 2: experimental cracking in a slurry-type continuous reactor. Fuel 2005;84:417–21. [22] Sing KSW. Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity. Pure Appl Chem 1985;57:603–19. [23] Fumoto E, Mizutani Y, Tago T, Masuda T. Production of ketones from sewage sludge over zirconia-supporting iron oxide catalysts in a steam atmosphere. Appl Catal B 2006;68:154–9. [24] Jung-Chung W, Der-Sun L, An-Nan K. Dehydrogenation of ethylbenzene over TiO2–Fe2O3 and ZrO2–Fe2O3 mixed oxide catalysts. Catal Lett 1993;20:191–201. [25] Sonoyama N, Nobuta K, Kimura T, Hosokai S, Hayashi J, Tago T, et al. Production of chemicals by cracking pyrolytic tar from Loy Yang coal over iron oxide catalysts in a steam atmosphere. Fuel Process Technol 2011;92:771–5.