Kinetic Modeling of the Hydrotreating and Hydrocracking Stages for

Subscriber access provided by CMU Libraries - http://library.cmich.edu

Article

Kinetic modeling of the hydrotreating and hydrocracking stages for upgrading scrap tires pyrolysis oil (STPO) towards high quality fuels Idoia Hita, Andres Tomas Aguayo, Martin Olazar, Miren J. Azkoiti, Javier Bilbao, José María Arandes, and Pedro Castaño Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.5b01502 • Publication Date (Web): 20 Oct 2015 Downloaded from http://pubs.acs.org on October 21, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Kinetic modeling of the hydrotreating and hydrocracking stages for upgrading scrap tires pyrolysis oil (STPO) towards high quality fuels Idoia Hita, Andrés T. Aguayo, Martin Olazar, Miren J. Azkoiti, Javier Bilbao, José M. Arandes, Pedro

5

Castaño* Department of Chemical Engineering, University of the Basque Country (UPV/EHU), P.O. Box 64448080 Bilbao, Spain. * E-mail: [email protected]

ABSTRACT The upgrading of scrap tires pyrolysis oil (STPO) has been studied in order to produce 10

high-quality alternative fuels, conceived as the second and third stage of an industriallyorientated valorization pathway for tires: pyrolysis, hydrotreating and hydrocracking. The experiments have been carried out in a fixed bed reactor under the following experimental conditions: (i) for hydrotreating: NiMo/Al2O3 catalyst; time on stream of 0-8 h; 300-375 ºC, 65 bar, H2:Oil ratio of 1,000 v/v, and space time of 0-0.5 gcat h gfeed-1;

15

and (ii) for hydrocracking: PtPd/SiO2-Al2O3 catalyst; time on stream of 0-6 h, 440500 ºC, 65 bar, space time of 0-0.28 gcat h gfeed-1, H2:Oil ratio of 1,000 v/v. From the results, lump-based kinetic models have been established for both stages, considering the reactions of (i) hydrodesulfurization (HDS), (ii) hydrocracking (HC) and (iii) hydrodearomatization (HDA) in each one of them. Catalyst deactivation is insignificant

20

in the hydrotreating stage but important for the hydrocracking stage. Therefore, deactivation has been considered in the corresponding kinetic equations. The computed deactivation constants have allowed for quantifying the contribution of each lump/composition fraction to coke formation.

25

Keywords: Kinetic modeling, hydrotreating, hydrocracking, bifunctional catalyst, pyrolysis oil, tires

1

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 36

1. INTRODUCTION

The valorization of tires has attracted increasing interest in the last years and is 30

becoming a major issue within the current energetic scenario. Approximately every year, 7·106 t of waste tires are disposed worldwide

1

and, only in the EU, an increase

6

from 2.1 to 3.3·10 t of disposed waste tires has been suffered within the 2003-2010 period, with a total disposal cost of 600 million euros.2 This fact, together with (i) the necessity to avoid the environmental damage caused by their non-biodegradability and 35

uncontrolled disposal, (ii) an increasing demand for recycling hydrocarbons, and (iii) the high calorific power that waste tires have (that could be recovered for their use as a fuel),3,4 turn tires into the perfect recycling target. In the recent years, chemical recovery routes have attracted increasing interest as an alternative to the most common material and energy recovery routes for tire recycling,5,6 and especially considering the necessity

40

to avoid the traditional land filling and incineration strategies, which are no longer environmentally acceptable.7 In this context, the greatest challenge that refineries are facing nowadays is to adapt and scale-up to the new recovery routes for petroleumderived wastes, in order to achieve a more energetically sustainable scenario. Incineration allows for recovering part of the calorific power, but it is no longer an

45

acceptable option due to its high level of emissions.8 As an alternative among the chemical recovery routes, pyrolysis presents more environmentally respectful characteristics,9,10 and also promising features not only for valuable chemical and calorific power recovery but also for its industrial scale implementation. Pyrolysis allows for treating diverse types of wastes like tires, plastics or biomass (either alone or

50

co-fed) with a high energetic efficiency.8 Three product fractions are obtained from scrap tire pyrolysis, each one with a wide range of applications, which contribute to the global economic viability of the process.11 Gases (10-20 wt%) can be used for achieving the energetic self-sufficiency of the pyrolysis process,12 while the solid product (char, 35-50 wt%) can be used for obtaining activated carbon through a variety of activation

55

routes

13,14

and also recovering the original carbon black used in tire manufacturing.15

From both an energetic and economical perspective the scrap tire pyrolysis oil (STPO), which is generally the most abundant pyrolysis product fraction (40-55 wt%), is also the most interesting of all, considering its great potential for being used as a blend in the refinery.16 However, STPO shows a series of compositional barriers that hinder its 2


Page 3 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

60

direct application as a fuel, and evidence the necessity for its upgrading. These barriers can be summarized as: (i) high sulfur content, (ii) concentration of aromatics and (iii) high amount of heavy molecules within the gasoil boiling point range (BP > 350 °C). Among all the processes taking place in a refinery, hydroprocessing is the only one that allows for solving the main STPO drawbacks simultaneously while increasing the value

65

of the feedstock. In fact, catalytic hydrotreaters are being used intensively for this purpose.17,18 The high sulfur content in STPO requires of a previous hydrotreating using low cost catalysts (NiMo).19 The hydrotreated scrap tires pyrolysis oil (HT-STPO) can be then further hydroprocessed using noble metal catalysts (more active towards hydrocracking reactions) for obtaining fuels with the adequate composition for a

70

refinery blend.20,21 The three reactions that are interesting regarding the main hydroprocessing

goals

for

STPO

are:

hydrodesulfurization

(HDS),

hydrodearomatization (HDA) and hydrocracking (HC). Trickle bed reactors are the most commonly used in hydroprocessing, consisting in fixed beds of catalyst with cocurrent downflow of gas and liquid phase reactants.22 A common industrial strategy is 75

to place different catalyst beds in series, with increasing order of activity in order to protect the most active catalyst from poisoning and deactivation.23,24 HDS kinetics can generally be modeled as a Langmuir-Hinshelwood (LH) mechanism, commonly used for heterogeneous catalyzed processes.25,26 HC and HDA kinetics can be modeled through different approaches with varying level of complexity, as reviewed

80

by Ancheyta.27 The experience of our group focused on the HC/HDA of model compounds28-30 and pyrolysis gasoline.31 Due to the very heterogeneous composition of refinery streams, considering each compound independently implies designing a kinetic model of great complexity that may present several analytical drawbacks, as a great amount of parameters need to be estimated. To minimize this problem, compounds can

85

be grouped in lumps, according to a boiling point or similar reactivity criteria.32,33 Indeed, lumping is the most common approach for modeling hydrocracking processes considering catalyst deactivation due to coke formation.34 Elizalde and Ancheyta35 proposed a 3-stage approach for modeling start-of-run, middle-of-run and end-of-run deactivation of the heavy oil hydrotreating.

90

This work delves with the kinetic modeling of STPO upgrading through a 2-stage hydrotreating-hydrocracking strategy. On the first stage aiming the sulfur removal, a 3


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

NiMo/Al2O3 hydrotreating catalyst has been used. On the second stage targeting the conversion of the heavier molecules, a PtPd/SiO2-Al2O3 hydrocracking catalyst has been used, togeather with the feed: a low-sulfur hydrotreated STPO (HT-STPO) 95

obtained in the first stage. 2. EXPERIMENTAL 2.1. Catalyst preparation and characterization. On the first hydrotreating stage, a commercial hydroprocessing NiMo catalyst supported on Al2O3 (2-3 wt% Ni, 7-10 wt% Mo) has been used for carrying out the hydroprocessing of STPO. The catalyst was

100

tableted and sieved to the desired particle size of 0.15-0.30 mm. For the second hydrocracking stage, a PtPd catalyst (1 wt% Pt, 0.5 wt% Pd) supported on SiO2-Al2O3 (Sigma Aldrich) has been prepared through wet impregnation at 80 ºC using Pt(NH3)4(NO3)2 (Alfa Aesar) and Pd(NH3)4(NO3)2 (Stem Chemicals) as precursors. The required amount of 1M precursor solutions for obtaining the desired metallic charge has

105

been added drop-by-drop on the support suspended in deionized water. pH = 7 was regulated by adding NH3 or NH4NO3 solution drops when necessary. The suspension containing the support and precursors is stirred for 24 h and once the adsorption equilibrium has been reached, excess water is removed in a Rotavapor at 80 ºC. Subsequently, the catalyst is then dried at 110 ºC for 12 h in an oven and calcined for

110

3 h at 500 ºC in air, following a ramp of 2.5 ºC min-1. Finally, the catalyst was tableted and sieved to the desired particle size of 0.15-0.30 mm. The main physico-chemical properties of the two catalysts are set out in Table 1. The techniques used for determining these catalytic properties have been described in previous works.19,20

115

Table 1. Physico-chemical properties of the catalysts 2.2. Kinetic runs. Hydrotreating and hydrocracking runs have been carried out in a laboratory scale fixed bed reactor (PID Tech Microactivity Pro) described elsewhere,19 consisting of 4 zones: (i) gas feeding, (ii) liquid feeding, (iii) reaction and (iv) sample collection zones. The runs have been carried out under the following conditions for

120

hydrotreating of STPO: time on stream of 0-8 h; 300-375 ºC, 65 bar, H2:Oil ratio of 1,000 v/v, and space time of 0-0.5 gcat h gfeed-1; and for hydrocracking of HT-STPO: time on stream of 0-6 h; 440-500 ºC, 65 bar, H2:Oil ratio of 1,000 v/v, and space time of 4


Page 4 of 36

Page 5 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

0-0.28 gcat h gfeed-1. Prior to the experiments, the NiMo catalyst is sulfided in-situ with 50 mL min-1 of a H2S:H2 mixture (10 vol%) raising the temperature at 5 ºC min-1 to 400 125

ºC and maintaining that temperature for 4 h. The PtPd catalyst, on the other hand, is reduced under atmospheric pressure in a stream of H2:N2 (30 mL min-1 H2, 50 mL min1

N2) while temperature is raised at a rate of 5 ºC min-1 from room temperature up to

400 ºC, and then maintained for 4 h. After activating the catalyst, the different experimental stages have been configured sequentially and registered using Processa 130

software. The feeds and reaction products have been analyzed offline by means of bi-dimensional gas chromatography (GCxGC) coupled in line with mass spectrometry (MS) using an Agilent 7890A gas chromatograph coupled with an Agilent 5975C series MS, and consisting of two columns of different polarities connected through a flow modulator.

135

The first column is a non-polar DB-5 ms J&W 122-5532 (length, 30 m; internal diameter, 0.25 mm), and the second one is a polar TRB-50 HT (length, 6 m; internal diameter, 0.25 mm). The outlet flow goes through both a flame ionization detector (FID) and a mass selective detector (MSD). Sulfur content has been measured by means of an Agilent 7890A GC equipped with

140

both a FID and a pulsed flame photometric detector (PFPD) specific for sulfur. The column was an Agilent Technologies HP PONA (length, 50 m; internal diameter, 0.2 mm). Total sulfur content was calculated by means of a calibration line (R2 > 0.995) that correlated the sulfur concentration with a ratio between the areas under the PFPD and FID chromatograms. This calibration line was obtained using a benzothiazole

145

standard (Aldrich, 96 %), in the same split and signal attenuation conditions of the product analysis. Based on the results obtained through GCxGC/MS, and equally to what was specified in a previous work,19 the composition of STPO is: 25.7 wt% naphtha (35-216 ºC), 44.5 wt% diesel (216-350 ºC) and 29.8 wt% gasoil (> 350 ºC). According to a chemical

150

group classification, STPO can also be divided into: 2.4 wt% paraffins and isoparaffins (P+iP), 7.1 wt% olefins (O), 34.7 wt% naphthenes (N), 38.1 wt% 1-ring aromatics (A1) and 17.7 wt% 2-ring aromatics (A2). Elemental sulfur accounts for 11,600 ppm on STPO. Benzothiazole (BTZ) is the most abundant sulfur compound (accounting for 3,794 ppm of elemental sulfur), and comes from sulfur-containing compounds added to 5


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

155

Page 6 of 36

the original tire as vulcanization accelerating agents. Dibenzothiophene-type structures (DBT) are divided into four groups according to the number of C atoms on their substituents, with M3DBT and M4DBT groups the most abundant groups with 3,460 and 3,412 ppm, respectively, followed by M2DBT (752 ppm) and M1DBT species (182 ppm).

160

Different conversions have been defined for the evaluation of the catalyst activity. The expressions quantifying the evolution of each reaction are: X HDS =

X HC =

(1)

S feed

x Gasoil

X HDA =

165

S feed − S prod

feed

− x Gasoil

x Gasoil x Arom

feed

(2)

prod

(3)

feed

− x Arom

x Arom

prod

feed

Where S is the total amount of sulfur in ppm, xG the mass fraction of the gasoil lump, and xArom the mass fraction of total aromatics. 3. RESULTS AND DISCUSSION 3.1. Kinetic models. Based on the sulfur species, product lumps and composition fractions; 3 kinetic schemes have been proposed for the reaction routes, as shown in

170

Figure 1 for the hydrotreating (Figure 1a-c) and hydrocracking (Figure 1c-e) stages. Figure 1a depicts the HDS of the most abundant sulfur compound in the STPO (BTZ) as representative of the rest of compounds. Figures 1b and 1c shows the HDA and HC kinetic schemes, respectively.

These are sequential hydrogenation and cracking

pathways from aromatics to aliphatics and from heavier to lighter products. Figure 1d 175

shows an example of the HDS of a M2DBT, as expected for HT-STPO. These schemes are based on other previous one established by Gutierrez et al.36 for the hydroprocessing of pyrolysis gasoline and others reviewed by Ancheyta.27 For HC reactions, gasoil ↔ diesel and diesel ↔ naphtha transformations have been considered (Figure 1b) (with an additional naphtha ↔ LPG transformation on the second stage, Figure 1e), as well as a

180

partial formation of naphtha (Figure 1b) or LPGs (Figure 1e) from heavy compounds in the gasoil lump. For HDA, hydrogenations of A2 and A1 groups together with ring 6


Page 7 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

opening of A1 have been proposed, considering the corresponding equilibriums, as well as chain scission reaction of A1 compounds to form paraffins (Figure 1c). Figure 1. Kinetic schemes 185

3.1.1. Kinetic equations for the STPO hydrotreating stage. Assuming that adsorbed sulfur (S*) is in equilibrium with the concentration of H2S and H2 in the liquid phase, we can express this concentration (xS*) as a function of the equilibrium constant (Φ):

x S* =

Φ x x H H 2S

(4)

2

190

Where xH2 the molar concentration of H2 and xH2S the molar concentration of H2S in the liquid phase. Then, Eq. 4

can be substituted in the Langmuir-Hinshelwood (LH)

expression, which is the most common approach for HDS kinetics.37,38:

k 'i x i x H dx i 2 = dτ 1+ K S*x S*

(

)

2

=

k 'i x i x H

2

  1+ K Φ x  S*  x H H2S   2

2

(5)

Where xi is the concentration of each sulfur species in ppm, ki the kinetic constant for 195

each species. The exponent 2 of the denominator in the equation is representative of the number of active sites in the adsorption for H2S. Considering that the equilibrium constant (Φ) and the concentration of H2 remain virtually constant in the reaction conditions employed, Eq. 6 can be expressed as a function of the concentration of H2S in the reaction media and the adsorption constant

200

for H2S (KH2S): dx i = dτ

k i' x i x H2 (1 + K H2S x H2S ) 2

=

ki xi (1 + K H2S x H2S ) 2

(6)

The kinetic equations proposed for describing the HC kinetic scheme (Figure 1b) have been elemental first order equations, as follows: dxG = −k1x G x H - k 3x G x H + k -1x D x H dτ

(7)

7


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

dx D = k1x G x H - k -1x D x H + k -2 x N x H - k 2 x D x H dτ

205

Page 8 of 36

(8)

Where xN, xD and xG are the mass fractions of naphtha, diesel and gasoil, and k is the corresponding kinetic constant for each stage. And the ones proposed for describing the HDA kinetic scheme (Figure 1c) are: dx A 2 = -k1x A 2 x H + k -1x A1x H dτ

210

(9)

dx A1 = k1x A 2 x H - k -1x A1x H - k 2 x A1x H + k - 2 x N x H - k 4 x A1x H dτ

(10)

dx N = k 2 x A1x H − k −2 x N x H − k 3 x N x H + k −3 x P x H dτ

(11)

Where xP, xN, xA1 and xA2 are the mass fractions of paraffins and isoparaffins, naphthenes, 1-ring aromatics and 2-ring aromatics, respectively. The equilibrium constants have been established according to the Van’t Hoff equation 215

(Eq. 12), and the kinetic constant has been defined by the reparametrized Arrhenius equation (Eq. 13), considering Tref as 390 ºC. Ki =

 ∆H  1 1  ki  = K i, Tref exp -   k −i  R  Tref T 

− E  1 1  − k i = k i,Tref ·exp  R  T Tref

  

(12)

(13)

3.1.2. Kinetic equations for the HT-STPO hydrocracking stage. The kinetic equation 220

proposed for the HDS reaction has the same expression as the one proposed on the hydrotreating stage (Eq. 6). For the HDA reaction, the kinetic scheme is also the same (Figure 1c) and consequently the same kinetic equations have also been proposed for both stages (Eqs. 9-11). For HC reaction, the formation of a LPG lump has been considered (Figure 1e) and consequently the system of equations at zero time on stream,

225

t0, is: dxG = −k1x G x H - k 4 x G x H + k -1x Dx H dτ dx D = k1x G x H - k -1x D x H + k -2 x N x H - k 2 x D x H dτ 8


(14)

(15)

Page 9 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

dx N = k 2 x D x H - k -2 x N x H + k -3 x LPG x H - k 3 x N x H dτ

(16)

It is to mention that, while in the hydrotreating stage over a NiMo catalyst no 230

deactivation has been observed, catalyst deactivation has become significant on the HC and HDA reactions on the HT-STPO hydrocracking stage using a PtPd catalyst. Catalyst deactivation is habitual in hydrocracking over bifunctional catalysts, where heavy aromatic molecules are responsible for coke deposition during the first hours of reaction,39,40 while for higher time on stream values metal-bearing species are the main

235

cause of permanent deactivation of the catalyst.41,42 Numerous kinetic studies have been conducted for hydrocracking reactions,39,43,44 but most of them in steady state conditions, with only a few considering catalyst deactivation during the first hours of reaction. To consider the effect of deactivation on the HT-STPO hydrocracking stage, an activity

240

parameter, a, has been included in the kinetic equations (Eq. 6, Eqs. 9-11 and Eqs. 1416), defined as the ratio between the reaction rate at a specific time, t, and the reaction rate at zero time on stream, t0.

a=

(r ) (r )

j t

j t 0

 dX   j  dτ   t =  dX   j  dτ  

(17)

t0

where j is referred to sulfur species in HDS, product lumps in HC (LPG, naphtha, diesel 245

or gasoil) and compositional groups in HDA (paraffins and isoparaffins, naphthenes, and 1- or 2-ring aromatics). According to Eq. 17, the rate of the reactions affected by deactivation at t time on stream is computed multiplying the reaction rates at zero time on stream, t0, (Eq. 6, Eqs. 9-11 and Eqs. 14-16) by the activity. To complete the kinetic model, a deactivation

250

kinetic equation is required to relate the activity with time on stream. Knowing from a previous work that in this hydrocracking stage deactivation is due mainly to coke formation,45 coke precursors need to be included in the deactivation model, since their concentration affects catalyst activity. For determining the suitable expression, different equations have been studied considering (i) a single coke precursor with only one

255

deactivation constant, (ii) various coke precursors with only one deactivation constant, 9


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 36

and (iii) various coke precursors with independent deactivation constants. After carrying out the corresponding statistical (SSE) and significance analysis (F-test), it has been concluded that the best deactivation equations for the HC and HDA models are the ones that consider all the product lumps and composition fractions as coke precursors with 260

independent deactivation constants For HC reactions: −

da = (k d1 x G + k d 2 x D + k d 3 x N + k d 4 x LPG ) a dt

(18)

For HDA reactions: −

265

da = (k d1 x A 2 + k d 2 x A1 + k d 3 x N + k d 4 x P ) a dt

(19)

Where xG, xD, xN and xLPG are the mass fractions of each product lump in HC; xA2, xA1, xN and xP the mass fractions of each compositional group in HDA; and kdi the independent deactivation constants associated to each one, which are also a function of temperature: − Ei k d i = k di ,Tref ·exp  R

270

1 1  − T T ref 

   

(20)

The deactivation equations have then been included in the corresponding kinetic models, and the values of the independent deactivation constants have allowed for quantifying the contribution of each product lump and compositional group to deactivation in order to determine which one is more likely to lead to coke formation. For the HDS reactions, the best fitting has been obtained by using a deactivation

275

equation independent of the concentration of the components in the reaction media: −

da = kda dt

(21)

3.2. Methodology for kinetic parameter computation. The computation of the kinetic parameters (activation energies and kinetic constants at the reference temperature of 390 ºC) has been carried out by fitting the experimental data of the evolution with time 280

on stream of the sulfur concentration (for HDS reactions) and mass fractions (for HC and HDA reactions) to the corresponding values calculated by integration of the 10


Page 11 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

differential equations of the models. The computed kinetic parameters minimize the following function: nc

f=

p

∑∑ (x

* i, j

− x i, j )2

(22)

i =1 j =1

285

Where x*i,j is the mass fraction of the component i at the experimental condition j, xi,j is the corresponding predicted value, nc is the number of components in the kinetic scheme and p is the total amount of experimental conditions. 3.3. Kinetic model fitting and validation 3.3.1. Hydrotreating of STPO. Table 2 lists the values computed for the kinetic

290

parameters corresponding to the models proposed for the hydrotreating of STPO. Focusing on HDS reactions in the first place, the model predicts that HDS of M4DBT occurs the fastest, followed by M1DBT and M2DBT (both species reacting at a very similar rate) and being M3DBT species those that react the slowest, and the ones that have the highest activation energy. Even though, according to their substituent number,

295

a higher reaction rate could be predicted for M3DBT species, in practice HDS of M4DBT species has occurred twice as rapid as that of M3DBT due to the position of their substituents. Table 2. Kinetic parameters for the first HT stage The quality and accuracy of the fitting can be evaluated from the parity diagram

300

(Figure 2a) representing experimental and calculated data, as well as from the comparison between the results predicted by the model equations (lines) and the experimental data (dots) in Figure 3, which shows the evolution of the different sulfur species at different temperatures and space times. As observed, an increase on space time (and therefore in the amount of catalyst loaded in the reactor), significantly affects

305

sulfur removal, obtaining the highest removal rates at higher space times. Yan et al.46 and Kallinikos et al.47 reported very similar effect of space time on the HDS of gasolinediesel type fractions, and gasoil, respectively. At lower space time, and therefore less catalyst loaded on the reactor, contact time decreases and less reaction time is provided. Considering the original sulfur concentrations in STPO, and the most favorable HDS

310

conditions, at 0.5 gcat h g-1feed the M4DBT-type species are the ones that have been removed in a greater extent, with individual removal conversions of 95-99 %, followed 11


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

by M1DBT and M2DBT (92-98 wt%) and with M3DBT being the most difficult species to eliminate (85-97 %). Sulfur molecules with a higher number of C atoms in their substituents are more 315

resistant towards HDS than less heavily substituted DBTs.19,48 However, comparing Figure 3c and Figure 3d (with almost identical initial concentrations), for a specific temperature and space time conditions, removal of M4DBT species occurs faster than that of M3DBT ones. From this, once again the determining role of the substituent position on HDS is evidenced. Substituent groups located close to the C-S bond can

320

strongly hinder HDS.49,50 Considering this, it can be deduced that the methyl- and ethylgroups in the remaining M3DBT species are mainly located in 2,4- positions (the closest to the S atom), while M4DBT species have a much lower proportion of molecules with substituents in these positions and can be removed easily. Kim et al.51 reported the important differences between “reactive” S compounds (BTs and substituted BTs) and

325

“refractive” S compounds (4- and/or 6- position substituted DBTs) on the deep HDS of a Saudi Arabian gasoil, confirming that HDS of refractive molecules appears as the great challenge. Figure 2. Parity diagrams for stage 1 Figure 3. Experimental data and model fitting for HDS

330

HDS kinetics has been widely studied with model compounds.52-55 Egorova and Prins56 proved that, for Mo catalysts, the direct desulfurization route of 4,6-DMDBT could be significantly enhanced by promotion with Ni and Co. According to the results obtained by Li et al.55 on the HDS of the same compound, the limiting reaction rate turned out to be hydrogenation and hydrogenolysis of 4,6-DMDBT itself, with higher reaction rates

335

on the rest of the mechanism steps. Regarding real heavy feedstock, Ferreira et al.57 recently developed a model that predicted very similar effect of the space time on the evolution of S content in a heavy vacuum residue, also considering the content in metals of the feed. Positive effect of higher temperatures, pressures and space times was also reported by Jarullah et al.38 on

340

the simulation on a trickle bed reactor of the HDS reaction of a crude oil. The LH-based model proposed by Laredo and Cortés58 studied the kinetics of differently substituted DBTs on the HDS of a gasoil narrow-cut fraction and once again corroborated that a substituent in position 4- greatly hinders HDS. Froment et al.59 proved that considering 12


Page 12 of 36

Page 13 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

structural contribution parameters of substituted DBTs allowed for obtaining a more 345

complete and accurate modeling approach on the HDS reaction of real feeds. Table 2 also summarizes the kinetic parameters obtained for the HC reaction (Figure 1b). As observed, at 390 ºC the gasoil to diesel conversion rate is higher than the rest, being twice as rapid as the reverse reaction (k1 > k-1). Conversion of diesel to naphtha, however, is less favored and considerably slower than its reverse reaction

350

(k2 < k-2), as very little amount of light compounds within the naphtha boiling point range are being formed, especially below 340 °C. According to the value of k3 (higher than k2), it can be deduced that the slight increase in the amount of naphtha is mainly due to the hydrogenation and cracking of compounds originally present in gasoil. The good fitting of the model can be proved from the parity diagram shown in

355

Figure 2b, as well as from the experimental data (dots) and model prediction (lines) in Figure 3. Diesel is the most abundant lump at all temperatures. For values of space time below 0.24 gcat h g-1feed, significant variations can be seen, especially regarding diesel and gasoil yields. The former increases in approximately 6-9 wt%, while the latter decreases in 7.5-12 wt% upon increasing space time and temperature. The highest diesel

360

yield and lowest gasoil yields are obtained at 375 °C and 0.5 gcat h g-1feed, with 54.2 wt% and 20.8 wt%, respectively. Variations with space time in naphtha occur in a much lesser extent, and become more important upon increasing temperature. While at 300 °C the amount of naphtha remains constant compared to the original feed amount, at 340 °C it increases up to 24.3 wt% and at 375 °C up to 25.0 wt% for 0.5 gcat h g-1feed

365

conditions. According to these results, at the entrance of the reactor products consist mainly of gasoil and diesel but, in further positions of the reactor, naphtha becomes more important, especially at temperatures above 340 °C. In any case, and as observed by Rayo et al.60 in the hydroprocessing of a Maya crude oil, preferentially middle distillates like diesel (more than 50 wt%) are obtained with this kind of supports, with

370

low Brönsted site concentration. Figure 4. Experimental data and model fitting for HC Less information for HC modeling is available in the literature compared to that reported for HDS. Ramírez et al.61 proposed a 5-lump model for thermal HC of heavy oils, equal to that previously developed by Sánchez et al.62 for moderate HC. They

375

determined that vacuum residue (VR) conversion had a higher selectivity towards heavy 13


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

fractions like VGO, compared to diesel or naphtha. Similar conclusion was reported by Martínez and Ancheyta34 applying the same 5-lump model on the HC of a heavy residue in a CSTR considering catalyst deactivation. Purón et al.44 proposed a 4-lump model that considered coke formation on the HC of a vacuum residue (VR) from Maya crude 380

oil over a NiMo/Al2O3 catalyst in a batch reactor. They observed that higher reaction temperatures led to the formation of products with a lower boiling point (10 wt% more of < 450 °C fraction when increasing temperature from 400 to 450 °C), and that the stages of gas formation from coke and equilibrium between coke and lower boiling point products should be considered.

385

Regarding HDA, the computed kinetic parameters are also listed in Table 2. According to the values of k1, k-1, and k2, it can be deduced that hydrogenation of 2-ring aromatics represents for the most relevant reaction, and occurs more rapidly than its reverse reaction (k1 > k-1). However, in the same conditions, the reversibility of A1 aromatics is more favored towards dehydrogenation (k2 < k-2). Due to the low cracking ability of the

390

catalyst, ring opening and aliphatic chain scission reactions occur in a much lesser extent, as deduced from their corresponding low kinetic constants. The accuracy of the model prediction can be seen in Figure 2c and Figure 5, corresponding to the parity diagram and fitting of the model prediction, respectively. Again, significant differences are observed with space time. These changes occur more

395

rapidly at 340 and 375 °C, when the steady state is reached earlier. The main variations occur in the case of paraffinic compounds, observing an increase from 2.4 wt% in STPO to 31.7 wt%, 32.7 wt% and 32.9 wt% at 300, 340 and 375 °C, respectively, as temperature has a positive effect on cracking reactions. The rapid olefin hydrogenation has a great contribution to the increase in paraffins. Furthermore, naphthene and

400

aromatics hydrogenation and further ring-opening also play an important role. Taking the data at 340 °C as an example, within the 0-0.3 gcat h g-1feed range, 1-ring aromatics decrease 11 wt%, 2- aromatics 7.4 wt%, while a smaller decrease of 4.8 wt% is observed in the case of naphthenes. According to this, together with olefins, a significant amount of aromatics are being hydrogenated and converted into saturated

405

cycles, to follow subsequent ring opening reactions and finally crack to form paraffins. Figure 5. Experimental data and model fitting for HDA

14


Page 14 of 36

Page 15 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

In Figure 5, a clear effect of the temperature is observed, as steady state is reached more rapidly when working at higher temperatures. Additionally, temperature has an effect on final compositions. At higher temperatures, lower amount of total paraffinic compounds 410

is obtained, together with higher total aromatics amount. The composition of naphthenes is not significantly affected. For a given space time conditions, similar trend for aromatics was reported by Tang et al.63 on the low temperature mild hydrotreating of a coal distillate. Several authors64-66 have reported the effect of space time on aromatic hydrogenation kinetics, also predicting higher hydrogenation rates at higher space time

415

values. 3.3.2. Hydrocracking of HT-STPO. The HT-STPO used as feed in this second hydroprocessing stage for the STPO upgrading consists of a mixture of the reaction products collected on the first stage. The aforementioned HT-STPO has been analyzed using the same chromatographic techniques previously specified, and consists of:

420

27.9 wt% naphtha, 50.4 wt% diesel and 21.8 wt% gasoil. According to its chemical groups, its composition is: 35.4 wt% paraffins and isoparaffins (P+iP), 22.0 wt% naphthenes (N), 31.9 wt% 1-ring aromatics (A1) and 10.7 wt% 2-ring aromatics (A2). All the remaining sulfur in HT-STPO is in the form of alkylated DBT species, distributed as: M1DBT, 82 ppm; M2DBT, 144 ppm; M3DBT, 1,019 ppm; and M4DBT,

425

784 ppm. The computed values for the kinetic constants and activation energies corresponding to the HDS reaction are listed in Table 3. As observed, the model predicts a more rapid HDS of the M1DBT species, followed by M4DBT, M2DBT and M3DBT, respectively (k1 > k4 > k2 > k3). Again, as already discussed on the first hydrotreating stage, M3DBT

430

is the least reactive sulfur species. Table 3. Kinetic parameters for the second HC stage The parity diagram of Figure 6a shows the quality of the fitting, and the evolution of the different sulfur species with TOS for the different space time conditions is shown in Figure 7 in terms of both experimental data (dots) and kinetic model prediction (lines).

435

Deactivation of the metallic phase takes place and increasing amounts of sulfur are detected in the products upon increasing TOS. However, space time has a more remarkable effect than temperature, as total sulfur increases up to ca. 800 ppm at the lowest space time of 0.05 gcat h g-1feed and TOS = 6 h. On the other hand, when working 15


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

at 0.28 gcat h g-1feed, less than 140 ppm total (mainly M3DBT compounds) were 440

measured at all TOS values, with a much lower activity decay. This can be explained considering that, at lower space times, fewer amounts of metallic sites are available for HDS as less catalyst is loaded in the reactor. For all space times and due to their lower reactivity, the main remaining sulfur species were M3DBT and M4DBT. Figure 6. Parity diagrams for stage 2 Figure 7. Experimental data and model fitting for HDS reaction

445

Accurate predictions can be obtained from the model, especially for lower space time values. However, for higher values of space time (0.16-0.28 gcat h g-1feed), the model predicts higher removal rates than that obtained experimentally. Therefore, the lower the total amount of sulfur in the products, the higher error it is to assume in the model 450

prediction. Regarding the results corresponding to HC reactions (Figure 1e). The kinetic parameters computed are also listed in Table 3. As observed, the hydrogenation of 2-ring and 1-ring aromatics is favored at 390 °C compared to their reverse reactions (k1 >> k-1 and k2 >> k-2), which can be considered negligible. Additionally, according to the values of

455

k3 and k-3, the transformation between naphtha and LPG is greatly favored towards naphtha yield (occurring even two orders of magnitude faster). These values confirm the higher activity of PtPd/SiO2-Al2O3 catalyst towards hydrogenation and cracking reactions compared to those computed using a NiMo/Al2O3 catalyst in a previous STPO hydroprocessing stage.

460

The good fitting of the model is proved by the parity diagram in Figure 6b, as well as from the comparison between the experimental data (dots) and the model prediction (lines) in Figure 8 for the evolution of product lumps with TOS at 440 °C. Qualitatively, increasing TOS, and therefore the progression of deactivation, has the same effect at all space time values, with decreasing amounts of naphtha and LPG (the lightest lumps)

465

and increasing diesel and gasoil yields due to loss of cracking ability. As observed in Figure 8, an increase in space time implies obtaining higher naphtha yields and lower diesel yields in fresh catalyst conditions (TOS = 0 h), up to 83 wt% at 0.28 gcat h g-1feed, with naphtha being the main lump. As the reaction occurs and the cracking function of the catalyst deactivates, this trend inverts and diesel becomes the main product lump 16


Page 16 of 36

Page 17 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

470

and significant amounts of gasoil are detected, except for 0.28 gcat h g-1feed conditions at which naphtha prevails during the whole reaction time and gasoil yield does not surpass 6 wt%. Very similar effect of time on stream on deactivation was observed by Elizalde et al.39 on the hydrocracking of atmospheric residue using the lumping approach. As has been observed on the results corresponding to the evolution with time on stream of the

475

different lump yields (Figure 8), the yield of naphtha rapidly decreases and diesel and gasoil yields increase until approximately a value of TOS = 6 h. Then, deactivation reaches an apparent equilibrium, and later on progresses very slowly. The obtained kinetic model quantifies deactivation during that initial 6 h period. Figure 8. Experimental data and model fitting for HC

480

An identical approach has been followed regarding the HDA reaction. According to the computed kinetic parameters summarized in Table 3, at 390 °C, the two aromatic hydrogenation stages are clearly favored over their reverse reactions, and the last ones can be considered negligible due to their low kinetic constant values. The same thing occurs with ring opening and its reverse cyclization reactions, with the former taking

485

place at a much faster rate than the latter. Furthermore, and as a consequence of the higher activity of the PtPd/SiO2-Al2O3 catalyst, the differences between the forward reactions (ki, hydrogenation and cracking) and their reverses (k-1, dehydrogenation and olygomerization) are more significant in this case than when using the NiMo/Al2O3 catalyst.

490

The accuracy of the model is proved by the parity diagram of Figure 6c and the evolution with TOS of the model fitting (lines) and experimental data (dots) in Figure 9. The effect of TOS has already been analyzed, observing a decrease in paraffins and isoparaffins and increasing amounts of naphthenics and aromatics. Upon increasing TOS, cyclic compounds appear in a greater extent, except for 0.28 gcat h g-1feed

495

conditions, when paraffins account for 52 wt% of the total. More saturated compounds, like paraffins and naphthenes, are obtained upon increasing space time, while unsaturated compounds (like aromatics) decrease. Equally to the HC model, the proposed kinetic model for HDA can be applied for predicting the evolution with time on stream of the composition fractions during the first reaction hours.

500

Figure 9. Experimental data and model fitting for HDA 17


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3.4. Comparison of the reaction rates of the two stages. Considering the computed kinetic constant values for the three reactions (HDS, HC and HDA) on both upgrading stages, relative reaction rates can be calculated to compare the kinetic behavior of both hydroprocessing catalysts (Table 4). For the comparison, removal of M1DBT, the gasoil 505

to diesel transformation and hydrogenation of the 2-ring aromatic fraction have been established as references (value 1) for HDS, HC and HDA, respectively. Table 4. Comparison of relative reaction rates For HDS, M3DBT compounds have been the least reactive in both hydroprocessing stages, particularly using the PtPd catalyst, due to (i) the longer substituent chains and

510

(ii) the position of the substituents with 4,6 positions being the least reactive. M4DBT species have been more reactive than M3DBT in both hydroprocessing stages, which might indicate that the substituents in M3DBT are mainly located in positions 4 and 6. Regarding HC results, clear differences are observed as, for the PtPd catalyst on hydrocraking, and due to the much higher cracking activity of the catalyst, the

515

transformation of gasoil and diesel to lighter lumps is clearly favored over the reverse reactions, even by 2 orders of magnitude. In the first hydroprocessing stage, however, gasoil transforms to diesel only twice as rapid as its reverse reaction, and further transformation of diesel into naphtha is hindered by reaction conditions. Very similar behavior of the PtPd catalyst has been observed for HDA, with a much higher activity

520

in hydrogenation reactions, with negligible dehydrogenations for both 1-ring aromatics and naphthenics. The significantly higher ring opening and chain scission rates of the PtPd catalyst compared to hydrogenation reactions should be highlighted, as they have greatly favored the removal of aromatics compared to the NiMo catalyst. 4. CONCLUSIONS

525

The experimental study that has been carried out has allowed for exploring the possibilities for STPO upgrading through a two-stage catalytic strategy (hydrotreating and hydrocracking). Furthermore, through kinetic modeling, and by means of quantifying the kinetics of HDS, HC and HDA reactions, it has been possible to establish the relation between the different lumps in the reaction media and the relative

530

importance of individual reactions.

18


Page 18 of 36

Page 19 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Accurate predictions for the evolution of the different sulfur species with time on stream and space time at different temperature conditions can be achieved by assuming a LH kinetic mechanism for HDS on the two stages, including H2 concentration in the media and the inhibition effect of H2S in the kinetic equation. Computed kinetic constants 535

predict the lowest reactivity of highly substituted DBT species, particularly M3DBT species, due to alkyl- substituents located in positions 4- and 6-. On the other hand, a sequential gasoil ↔ diesel ↔ naphtha scheme with an additional gasoil → naphtha

transformation

(for

STPO

hydrotreating)

and

a

gasoil ↔ diesel ↔ naphtha ↔ LPG scheme considering a gasoil → LPG transformation 540

(for HT-STPO hydrocracking), have been suitable for predicting the evolution with space time of product lump yields of the HC reactions. In the hydrotreating stage, at 390 °C, the model predicts that gasoil → diesel transformation is greatly favored over its reverse, contrary to what happens with the diesel → naphtha transformation, which is twice as slower as its reverse. In the hydrocracking stage, on the other hand, it has been

545

proven that the contribution to deactivation of the different product lumps follows the same trend as their average boiling point, with gasoil lump being the main responsible for deactivation in HC reactions (highest deactivation constant). A sequential A2 ↔ A1 ↔ N ↔ P + iP scheme for HDA reaction considering an additional A1 → P + iP stage is suitable for accurately predicting the evolution of the

550

composition fractions in both stages. In the hydrotreating stage, at 390 ºC the hydrogenation of heavier aromatics (A2) is more favored than that of lighter ones (A1) and ring opening reactions of naphthenes, while chain scission reaction can be considered negligible. At higher temperatures the hydrogenation of aromatic compounds is less favored, evidencing a strong thermodynamic control in these

555

reactions. In the hydrocracking stage, however, hydrogenation reactions are more favoured and ring opening and chain scission reactions acquire a more significant extent due to the higher catalyst activity. Furthermore, it has also been proven that, 2-ring aromatics have the greatest contribution to deactivation in HDA transformations, followed by 1-ring aromatics, naphthenics and paraffins.

560

The results obtained in this work show good perspectives for the valorization of scrap tires in a large scale. On the other hand, it should be considered that the economical viability of the hydroprocessing of STPO requires its co-feeding with other refinery 19


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

streams to the refinery units, which is encouraging for continuing this work with further research in this respect. 565

ACKNOWLEDGEMENTS This work was carried out with the financial support of the Ministry of Economy and Competitivity of the Spanish Government (CTQ2010-19623, CTQ2012-35192 and CTQ2013-46172-P), the FEDER funds, the Basque Government (SAIOTEK SA2011/00098 and SA-2013/00173_IT748-13), and the University of the Basque Country

570

(UFI 11/39). Idoia Hita is also grateful for her Basque Government research training grant (BFI2010-223). REFERENCES

575

580

585

590

595

600

(1) Portofino, S.; Donatelli, A.; Iovane, P.; Innella, C.; Civita, R.; Martino, M.; Matera, D. A.; Russo, A.; Cornacchia, G.; Galvagno, S., Steam gasification of waste tyre: Influence of process temperature on yield and product composition. Waste Manage. 2013, 33 (3), 672-678. (2) Sienkiewicz, M.; Kucinska-Lipka, J.; Janik, H.; Balas, A., Progress in used tyres management in the European Union: A review. Waste Manage. 2012, 32 (10), 17421751. (3) Fernández, A. M.; Barriocanal, C.; Alvarez, R., Pyrolysis of a waste from the grinding of scrap tyres. J. Hazard. Mater. 2012, 203-204, 236-243. (4) Acevedo, B.; Barriocanal, C., Fuel-oils from co-pyrolysis of scrap tyres with coal and a bituminous waste. Influence of oven configuration. Fuel 2014, 125 (0), 155-163. (5) Gieré, R.; Smith, K.; Blackford, M., Fuel 2006, 85 (16), 2278-2285. (6) Lee, H. J.; Roh, H. S., The use of recycled tire chips to minimize dynamic earth pressure during compaction of backfill. Construction and Building Materials 2007, 21 (5), 1016-1026. (7) Jang, J.-W.; Yoo, T.-S.; Oh, J.-H.; Iwasaki, I., Discarded tire recycling practices in the United States, Japan and Korea. Resources, Conservation and Recycling 1998, 22 (1–2), 1-14. (8) Sharma, V. K.; Fortuna, F.; Mincarini, M.; Berillo, M.; Cornacchia, G., Disposal of waste tyres for energy recovery and safe environment. Appl. Energy 2000, 65 (1-4), 381-394. (9) Zabaniotou, A.; Antoniou, N.; Bruton, G., Analysis of good practices, barriers and drivers for ELTs pyrolysis industrial application. Waste Management 2014, 34 (11), 2335-2346. (10) Lombardi, L.; Carnevale, E.; Corti, A., A review of technologies and performances of thermal treatment systems for energy recovery from waste. Waste Management (0). (11) Antoniou, N.; Zabaniotou, A., Features of an efficient and environmentally attractive used tyres pyrolysis with energy and material recovery. Renewable and Sustainable Energy Reviews 2013, 20 (0), 539-558. (12) Kaminsky, W.; Sinn, H., PYROLYSIS OF PLASTICS WASTES AND USED TYRES IN A FLUIDISED BED REACTOR. Kunstst Ger Plast 1978, 68 (5), 14-17. 20


Page 20 of 36

Page 21 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

605

610

615

620

625

630

635

640

645

650

(13) Acevedo, B.; Barriocanal, C., Texture and surface chemistry of activated carbons obtained from tyre wastes. Fuel Processing Technology 2015, 134 (0), 275-283. (14) Acevedo, B.; Barriocanal, C.; Lupul, I.; Gryglewicz, G., Properties and performance of mesoporous activated carbons from scrap tyres, bituminous wastes and coal. Fuel 2015, 151 (0), 83-90. (15) Mikulova, Z.; Sedenkova, I.; Matejova, L.; Večeř, M.; Dombek, V., Study of carbon black obtained by pyrolysis of waste scrap tyres. Journal of Thermal Analysis and Calorimetry 2013, 111 (2), 1475-1481. (16) de Marco Rodriguez, I.; Laresgoiti, M. F.; Cabrero, M. A.; Torres, A.; Chomón, M. J.; Caballero, B., Pyrolysis of scrap tyres. Fuel Processing Technology 2001, 72 (1), 9-22. (17) Castañeda, L. C.; Muñoz, J. A. D.; Ancheyta, J., Comparison of approaches to determine hydrogen consumption during catalytic hydrotreating of oil fractions. Fuel 2011, 90 (12), 3593-3601. (18) Tian, K. P.; Mohamed, A. R.; Bhatia, S., Catalytic upgrading of petroleum residual oil by hydrotreating catalysts: a comparison between dispersed and supported catalysts. Fuel 1998, 77 (11), 1221-1227. (19) Hita, I.; Gutiérrez, A.; Olazar, M.; Bilbao, J.; Arandes, J. M.; Castaño, P., Upgrading model compounds and Scrap Tires Pyrolysis Oil (STPO) on hydrotreating NiMo catalysts with tailored supports. Fuel 2015, 145 (0), 158-169. (20) Hita, I.; Rodríguez, E.; Olazar, M.; Bilbao, J.; Arandes, J. M.; Castaño, P., Prospects for Obtaining High Quality Fuels from the Hydrocracking of a Hydrotreated Scrap Tires Pyrolysis Oil. Energy and Fuels 2015, 29 (8), 5458-5466. (21) Gutiérrez, A.; Arandes, J. M.; Castaño, P.; Olazar, M.; Bilbao, J., Preliminary studies on fuel production through LCO hydrocracking on noble-metal supported catalysts. Fuel 2012, 94 (0), 504-515. (22) Ranade, V. V.; Chaudhari, R. V.; Gunjal, P. R., Chapter 3 - Reaction Engineering of Trickle Bed Reactors. In Trickle Bed Reactors, Ranade, V. V.; Gunjal, R. V. C. R., Eds. Elsevier: Amsterdam, 2011; pp 77-116. (23) Alvarez, A.; Ancheyta, J.; Muñoz, J. A. D., Comparison of quench systems in commercial fixed-bed hydroprocessing reactors. Energy and Fuels 2007, 21 (2), 11331144. (24) Castaño, P.; Gutiérrez, A.; Pawelec, B.; Fierro, J. L. G.; Aguayo, A. T.; Arandes, J. M., Effect of the support on the kinetic and deactivation performance of Pt/support catalysts during coupled hydrogenation and ring-opening of pyrolysis gasoline. Appl. Catal. A: Gen. 2007, 333 (2), 161-171. (25) Topsoe, H.; Clausen, B. S.; Massoth, F. E., Hydrotreating catalysis. Berlin, 1996; Vol. 11. (26) Castaño, P.; van Herk, D.; Kreutzer, M. T.; Moulijn, J. A.; Makkee, M., Kinetic and deactivation modelling of biphenyl liquid-phase hydrogenation over bimetallic PtPd catalyst. Appl. Catal. B: Environ. 2009, 88 (1-2), 213-223. (27) Ancheyta, J., Modeling and simulation of catalytic reactors for petroleum refining. Hoboken, New Jersey, 2011. (28) Castaño, P.; Arandes, J. M.; Pawelec, B.; Fierro, J. L. G.; Gutiérrez, A.; Bilbao, J., Kinetic model discrimination for toluene hydrogenation over noble-metal-supported catalysts. Ind. Eng. Chem. Res. 2007, 46, 7417-7425. (29) Castaño, P.; Gayubo, A. G.; Pawelec, B.; Fierro, J. L. G.; Arandes, J. M., Kinetic modelling of methylcyclohexane ring-opening over a HZSM-5 zeolite catalyst. Chem. Eng. J. 2008, 140 (1-3), 287-295. 21


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

655

660

665

670

675

680

685

690

695

700

(30) Castaño, P.; Arandes, J. M.; Pawelec, B.; Azar, M.; Bilbao, J., Kinetic modeling for assessing the product distribution in toluene hydrocracking on a Pt/HZSM-5 catalyst. Ind. Eng. Chem. Res. 2008, 47 (4), 1043-1050. (31) Gutiérrez, A.; Castaño, P.; Azkoiti, M. J.; Bilbao, J.; Arandes, J. M., Modelling product distribution of pyrolysis gasoline hydroprocessing on a Pt-Pd/HZSM-5 catalyst. Chem. Eng. J. 2011, 176-77, 302-311. (32) Botchwey, C.; Dalai, A. K.; Adjaye, J., Kinetics of bitumen-derived gas oil upgrading using a commercial NiMo/Al2O3 catalyst. Canadian Journal of Chemical Engineering 2004, 82 (3), 478-487. (33) Krishna, R.; Saxena, A. K., Use of an axial-dispersion model for kinetic description of hydrocracking. Chemical Engineering Science 1989, 44 (3), 703-712. (34) Martínez, J.; Ancheyta, J., Kinetic model for hydrocracking of heavy oil in a CSTR involving short term catalyst deactivation. Fuel 2012, 100 (0), 193-199. (35) Elizalde, I.; Ancheyta, J., Application of a three-stage approach for modeling the complete period of catalyst deactivation during hydrotreating of heavy oil. Fuel 2014, 138 (0), 45-51. (36) Gutiérrez, A.; Castaño, P.; Azkoiti, M. J.; Bilbao, J.; Arandes, J. M., Modelling product distribution of pyrolysis gasoline hydroprocessing on a Pt–Pd/HZSM-5 catalyst. Chem. Eng. J. 2011, 176–177, 302-311. (37) Contreras-Valdez, Z.; Mogica-Betancourt, J. C.; Alvarez-Hernández, A.; Guevara-Lara, A., Solvent effects on dibenzothiophene hydrodesulfurization: Differences between reactions in liquid or gas phase. Fuel 2013, 106 (0), 519-527. (38) Jarullah, A. T.; Mujtaba, I. M.; Wood, A. S., Kinetic parameter estimation and simulation of trickle-bed reactor for hydrodesulfurization of crude oil. Chemical Engineering Science 2011, 66 (5), 859-871. (39) Elizalde, I.; Ancheyta, J., Modeling catalyst deactivation during hydrocracking of atmospheric residue by using the continuous kinetic lumping model. Fuel Processing Technology 2014, 123 (0), 114-121. (40) Castaño, P.; Gutiérrez, A.; Hita, I.; Arandes, J. M.; Aguayo, A. T.; Bilbao, J., Deactivating species deposited on Pt-Pd catalysts in the hydrocracking of light-cycle oil. Energy Fuels 2012, 26 (3), 1509-1519. (41) Hauser, A.; Stanislaus, A.; Marafi, A.; Al-Adwani, A., Initial coke deposition on hydrotreating catalysts. Part II. Structure elucidation of initial coke on hydrodematallation catalysts. Fuel 2005, 84 (2-3), 259-269. (42) Idei, K.; Takahashi, T.; Kai, T., Estimation of coke and metal deposition distribution within hydrodesulfurization catalyst pore at the last stage of operation. Sekiyu Gakkaishi (Journal of the Japan Petroleum Institute) 2003, 46 (1), 45-52. (43) Elizalde, I.; Rodríguez, M. A.; Ancheyta, J., Modeling the effect of pressure and temperature on the hydrocracking of heavy crude oil by the continuous kinetic lumping approach. Applied Catalysis A: General 2010, 382 (2), 205-212. (44) Puron, H.; Arcelus-Arrillaga, P.; Chin, K. K.; Pinilla, J. L.; Fidalgo, B.; Millan, M., Kinetic analysis of vacuum residue hydrocracking in early reaction stages. Fuel 2014, 117, Part A (0), 408-414. (45) Hita, I.; Rodriguez, E.; Olazar, M.; Bilbao, J.; Arandes, J. M.; Castaño, P., Prospects for obtaining high quality fuels from the hydrocracking of a hydrotreated scrap tires pyrolysis oil. In. (46) Kan, T.; Wang, H.; He, H.; Li, C.; Zhang, S., Experimental study on two-stage catalytic hydroprocessing of middle-temperature coal tar to clean liquid fuels. Fuel 2011, 90 (11), 3404-3409. 22


Page 22 of 36

Page 23 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

705

710

715

720

725

730

735

740

745

750

(47) Kallinikos, L. E.; Bellos, G. D.; Papayannakos, N. G., Study of the catalyst deactivation in an industrial gasoil HDS reactor using a mini-scale laboratory reactor. Fuel 2008, 87 (12), 2444-2449. (48) Navarro, R. M.; Castaño, P.; Alvarez-Galvan, M. C.; Pawelec, B., Hydrodesulfurization of dibenzothiophene and a SRGO on sulfide Ni(Co)Mo/Al2O3 catalysts. Effect of Ru and Pd promotion. Catal. Today 2009, 143 (1-2), 108-114. (49) Song, C.; Ma, X., New design approaches to ultra-clean diesel fuels by deep desulfurization and deep dearomatization. Applied Catalysis B: Environmental 2003, 41 (1–2), 207-238. (50) Shafi, R.; Hutchings, G. J., Hydrodesulfurization of hindered dibenzothiophenes: an overview. Catalysis Today 2000, 59 (3–4), 423-442. (51) Kim, T.; Ali, S. A.; Alhooshani, K.; Park, J.-I.; Al-Yami, M.; Yoon, S.-H.; Mochida, I., Analysis and deep hydrodesulfurization reactivity of Saudi Arabian gas oils. Journal of Industrial and Engineering Chemistry 2013, 19 (5), 1577-1582. (52) Lewandowski, M., Hydrotreating activity of bulk NiB alloy in model reaction of hydrodesulfurization 4,6-dimethyldibenzothiophene. Applied Catalysis B: Environmental 2014, 160–161 (0), 10-21. (53) Castillo-Araiza, C. O.; Chávez, G.; Dutta, A.; de los Reyes, J. A.; Nuñez, S.; García-Martínez, J. C., Role of Pt–Pd/γ-Al2O3 on the HDS of 4,6-DMBT: Kinetic modeling & contribution analysis. Fuel Processing Technology (0). (54) García-Martínez, J. C.; Castillo-Araiza, C. O.; De los Reyes Heredia, J. A.; Trejo, E.; Montesinos, A., Kinetics of HDS and of the inhibitory effect of quinoline on HDS of 4,6-DMDBT over a Ni–Mo–P/Al2O3 catalyst: Part I. Chemical Engineering Journal 2012, 210 (0), 53-62. (55) Li, X.; Wang, A.; Egorova, M.; Prins, R., Kinetics of the HDS of 4,6dimethyldibenzothiophene and its hydrogenated intermediates over sulfided Mo and NiMo on γ-Al2O3. Journal of Catalysis 2007, 250 (2), 283-293. (56) Egorova, M.; Prins, R., Hydrodesulfurization of dibenzothiophene and 4,6dimethyldibenzothiophene over sulfided NiMo/γ-Al2O3, CoMo/γ-Al 2O3, and Mo/γAl2O3 catalysts. Journal of Catalysis 2004, 225 (2), 417-427. (57) Ferreira, C.; Tayakout-Fayolle, M.; Guibard, I.; Lemos, F., Hydrodesulfurization and hydrodemetallization of different origin vacuum residues: New modeling approach. Fuel 2014, 129 (0), 267-277. (58) Laredo, G. C.; Córtes, C. M., Kinetics of hydrodesulfurization of dimethyldibenzothiophenes in a gas oil narrow-cut fraction and solvent effects. Applied Catalysis A: General 2003, 252 (2), 295-304. (59) Froment, G. F.; Castaneda-Lopez, L. C.; Marin-Rosas, C., Kinetic modeling of the hydrotreatment of light cycle oil and heavy gas oil using the structural contributions approach. Catalysis Today 2008, 130 (2–4), 446-454. (60) Rayo, P.; Ramírez, J.; Torres-Mancera, P.; Marroquín, G.; Maity, S. K.; Ancheyta, J., Hydrodesulfurization and hydrocracking of Maya crude with P-modified NiMo/Al2O3 catalysts. Fuel 2012, 100, 34-42. (61) Ramírez, S.; Martínez, J.; Ancheyta, J., Kinetics of thermal hydrocracking of heavy oils under moderate hydroprocessing reaction conditions. Fuel 2013, 110 (0), 8388. (62) Sánchez, S.; Rodríguez, M. A.; Ancheyta, J., Kinetic model for moderate hydrocracking of heavy oils. Industrial and Engineering Chemistry Research 2005, 44 (25), 9409-9413. 23


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

755

760

(63) Tang, W.; Fang, M.; Wang, H.; Yu, P.; Wang, Q.; Luo, Z., Mild hydrotreatment of low temperature coal tar distillate: Product composition. Chemical Engineering Journal 2014, 236 (0), 529-537. (64) Gutiérrez, A.; Arandes, J. M.; Castaño, P.; Olazar, M.; Barona, A.; Bilbao, J., Effect of space velocity on the hydrocracking of Light Cycle Oil over a Pt-Pd/HY zeolite catalyst. Fuel Process. Technol. 2012, 95, 8-15. (65) Melis, S.; Erby, L.; Sassu, L.; Baratti, R., A model for the hydrogenation of aromatic compounds during gasoil hydroprocessing. Chemical Engineering Science 2004, 59 (22–23), 5671-5677. (66) Kishore Kumar, S. A.; John, M.; Pai, S. M.; Niwate, Y.; Newalkar, B. L., Low temperature hydrogenation of aromatics over Pt–Pd/SiO2–Al2O3 catalyst. Fuel Processing Technology 2014, 128 (0), 303-309.

24


Page 24 of 36

Page 25 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

765

Hydrotreatment of STPO

Hydrocracking of HT-STPO

a) HDS N

d) HDS

ki

ki

S

b) HC Gasoil >350 ºC

S

e) HC

k3 k1 k-1

k2

Diesel 215-350 ºC

k-2

Gasoil

Naphtha

k2 Diesel

k-1

35-215 ºC

c) HDA A2

k4 k1

k3 Naphtha

k-2

LPG k-3

k4 k1 k-1

A1

k2 k-2

N

k3

P+iP

k-3

Figure 1. Proposed kinetic schemes for the HDS, HC and HDA routes respectively in the hydrotreating (a-c), and hydrocracking (d-f) stages for STPO upgrading.

25


Energy & Fuels

3500

a)

Sulfurcalc (ppm)

3000 2500 2000 1500

M1DBT

1000

M2DBT M3DBT

500

M4DBT

0 0

1000

2000

3000

Sulfurexp (ppm)

0.55

b)

0.50

-1

xi, Calc (gi gtotal)

0.45 0.40 0.35

Naphtha Diesel Gasoil

0.30 0.25 0.20 0.2

0.3

0.4

0.5

-1

xi, Exp (gi gtotal) 0.35

c) 0.30

-1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 36

0.25 0.20

P+iP N A1

0.15

A2

0.10 0.10

0.15

0.20

0.25

0.30

0.35

-1

xi, Exp (gi gtotal) 770

Figure 2. Parity diagram for the kinetic models proposed for the HDS (a), HC (b) and HDA (c) reactions on the hydrotreating of STPO.

26


Page 27 of 36

200

800

a) M1DBT

175

300 ºC 340 ºC 375 ºC

b) M2DBT

700 600

Sulfur (ppm)

Sulfur (ppm)

150 125 100 75 50 25

500 400 300 200 100

0

0 0.0

0.1

0.2

0.3

0.4

0.5

0.0

-1

0.2

0.3

0.4

0.5

Space time (gcat h gfeed) 3500

3500

c) M3DBT

3000

d) M4DBT

3000 2500

Sulfur (ppm)

2500 2000 1500 1000

2000 1500 1000 500

500

0

0 0.0

0.1

0.2

0.3

0.4

0.5

-1

0.0

0.1

0.2

0.3

0.4

0.5

-1

Space time (gcat h gfeed)

775

0.1

-1


Sulfur (ppm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels


Figure 3. Experimental data (dots) and model fitting (lines) for the evolution with space time of the sulfur species at different temperatures on the hydrotreating of STPO (65 bar, TOS, 8 h).

27


Energy & Fuels

0.60 0.55

0.60

a) 300 ºC

0.50

Naphtha Diesel Gasoil

0.40

-1

0.45

xi (gi gtotal)

-1

b) 340 ºC

0.55

0.50

xi (gi gtotal)

0.35 0.30 0.25

0.45 0.40 0.35 0.30 0.25

0.20

0.20 0.0

0.1

0.2

0.3

0.4

0.5

0.0

-1

0.1

0.2

0.3

0.4

0.5

-1



0.60 0.55

c) 375 ºC

0.50

-1

xi (gi gtotal)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 36

0.45 0.40 0.35 0.30 0.25 0.20 0.0

0.1

0.2

0.3

0.4

0.5

-1


780

Figure 4. Experimental data (dots) and model fitting (lines) for the evolution with space time of the product lumps at different temperatures on the hydrotreating of STPO (65 bar, TOS, 8 h).

28


Page 29 of 36

P+iP N A1

0.5

a) 300 ºC 0.4

0.5

b) 340 ºC 0.4

A2

xi (gi gtotal)

0.3

0.3

-1

-1

xi (gi gtotal)

0.2 0.1

0.2 0.1

0.0 0.0

0.1

0.2

0.3

0.4

0.0

0.5

0.0

-1

0.1

0.2

0.3

0.4

0.5

-1



0.5

c) 375 ºC 0.4 0.3

-1

xi (gi gtotal)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

0.2 0.1 0.0 0.0

0.1

0.2

0.3

0.4

0.5

-1


785

Figure 5. Experimental data (dots) and model fitting (lines) for the evolution with space time of product composition at different temperatures on the hydrotreating of STPO (65 bar, TOS, 8 h).

29


Energy & Fuels

500

a) SulfurCalc (ppm)

400 300

M1DBT

200

M2DBT 100

M3DBT M4DBT

0 0

100

200

300

400

500

SulfurExp (ppm) 1.0

b)

-1


0.8 0.6 0.4

Gasoil Diesel Naphtha LPG

0.2 0.0 0.0

0.2

0.4

0.6

0.8

1.0

-1

xi, Exp (gi gtotal) 0.8 0.7

c)

0.6 -1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 36

0.5 0.4

P+iP N A1

0.3 0.2 0.1

A2

0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 -1

xi, Exp (gi gtotal)

790

Figure 6. Parity diagram for the kinetic models proposed for the HDS (a), HC (b) and HDA (c) reactions on the hydrocracking of HT-STPO.

30


Page 31 of 36

500

500 -1

-1

a) 0.05 gcat h gfeed

b) 0.09 gcat h gfeed

400

Sulfur (ppm)

Sulfur (ppm)

400 300 200 100 0

M1DBT

300

M2DBT M3DBT

200

M4DBT

100 0

0

1

2

3

4

5

6

0

1

2

TOS (h)

3

4

5

6

4

5

6

TOS (h)

500

500 -1

-1

c) 0.16 gcat h gfeed

d) 0.28 gcat h gfeed

400

Sulfur (ppm)

400

Sulfur (ppm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

300 200 100 0

300 200 100 0

0

1

2

3

4

5

6

0

1

2

TOS (h)

3

TOS (h)

Figure 7. Experimental data (dots) and model fitting (lines) for the evolution with TOS of the sulfur species at different space times on the hydrocracking of HT-STPO (65 bar, 440 ºC). 795

31


Energy & Fuels

-1


0.8

xi (gi gtotal)

0.6 -1

-1

xi (gi gtotal)

-1


0.8

0.6 0.4 0.2

0.4 0.2

0.0

0.0 0

2

4

6

0

TOS (h)

LPG Naphtha Diesel Gasoil

-1


-1

0.2

0.4 0.2

0.0

0.0 0

2

4

6

0

TOS (h)

800

6

0.6

xi (gi gtotal)

-1

0.4

4

0.8


0.6

2

TOS (h) -1

0.8

xi (gi gtotal)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 36

2

4

6

TOS (h)

Figure 8. Experimental data (dots) and model fitting (lines) for the evolution with TOS of the product lumps at different space times on the hydrocracking of HT-STPO (65 bar, 440 ºC).

32


Page 33 of 36

0.8

0.8 -1

P+iP N A1

0.6

-1


Composition (wt%)

Composition (wt%)


A2 0.4

0.2

0.0

0.6

0.4

0.2

0.0 0

2

4

6

0

TOS (h)

2

4

0.8 -1

-1


Composition (wt%)


0.6

0.4

0.2

0.0

0.6

0.4

0.2

0.0 0

2

4

6

0

TOS (h)

805

6

TOS (h)

0.8

Composition (wt%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

2

4

6

TOS (h)

Figure 9. Experimental data (dots) and model fitting (lines) for the evolution with TOS of the compositional fractions at different space times on the hydrocracking of HTSTPO (65 bar, 440 ºC).

33


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 36

Table 1. Physico-chemical properties of the fresh NiMo and PtPd catalyst

810

Property

NiMo/Al2O3

PtPd/SiO2-Al2O3

Ni (wt%) Mo (wt%) Pt (wt%) Pd (wt%)

2.23 7.97

SBET (m2 gcat-1) Pore volume (cm3 g-1) VP/P0=0.2Ads (cm3 g-1) VP/P0=0.5Ads (cm3 g-1) VP/P0=0.5Des (cm3 g-1) Average pore diameter (Å)

278 0.550 80 117 123 82

468 0.683 136 217 212 56

Total acidity (mmolt-BA g-1)

0.525

0.288

0.76 0.48

Table 2. Kinetic parameters for the 3 models proposed for the hydrotreating of STPO k (gfeedgcat-1h-1)

E (kJ mol-1) HDS

M1DBT M2DBT M3DBT M4DBT

-5

(2.17±0.70)·10 (1.89±0.59)·10-5 (1.44±0.19)·10-5 (2.79±0.39)·10-5

34.30±6.22 33.49±1.41 36.10±3.07 32.38±4.16 HC

Gasoil→Diesel, k1 Diesel→Gasoil, k-1 Diesel→Naphtha, k2 Naphtha→Diesel, k-2 Gasoil→Naphtha, k3

10.23±0.52 4.08±0.43 4.95±0.69 9.83±0.85 7.79±0.49

11.83±0.76 1.87±1.67 2.90±1.73 3.57±0.17 12.6±3.36 HDA

Hydrogenation, k1 Dehydrogenation, k-1 Hydrogenation, k2 Dehydrogenation, k-2 Ring opening, k3 Cyclization, k-3 Chain Scission, k4

(3.71±0.15)·102 (1.49±0.15)·102 (1.57±0.36)·102 (1.66±0.36)·102 (1.71±0.11)·102 16.3±0.11 1.71±0.11

34


0.61±0.29 92.01±8.43 22.78±2.15 69.73±2.56

Page 35 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Table 3. Kinetic parameters for the 3 models proposed for the hydrocracking of HTSTPO k (h-1)

E (kJ mol-1) HDS


(4.03±0.28)·10-4 (2.31±0.07)·10-4 (1.46±0.10)·10-4 (2.81±0.09)·10-4

11.78±1.32 1.09±0.23 7.33±1.25 11.14±2.03 HC

Gasoil→Diesel, k1 Diesel→Gasoil, k-1 Diesel→Naphtha, k2 Naphtha→Diesel, k-2 Naphtha→LPG, k3 LPG→Naphtha, k-3 Gasoil→LPG, k4

43.41±7.25 1.88±0.15 21.58±1.31 3.03±0.54 7.57±1.85 (3.25±0.66)·102 5.59±0.89

43.07±11.2 5.75±2.30 44.14±7.52 39.94±4.65 30.61±6.32 2.80±2.35 65.94±11.25 HDA


8.73±1.22 (8.49±2.16)·10-1 8.47±2.11 (3.18±0.37)·10-2 19.40±4.10 1.34±0.41 1.40±0.36

815

35


33.01±7.25 17.71±4.36 31.76±9.50 33.78±4.52

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 36

Table 4. Comparison between the relative HDS, HC and HDA reaction rates for the hydrotreating and hydrocracking stages NiMo/Al2O3

PtPd/SiO2-Al2O3 HDS


1 0.87 0.66 1.29

1 0.57 0.36 0.70 HC

Gasoil→Diesel Diesel→Gasoil Diesel→Naphtha Naphtha→Diesel Gasoil→Naphtha/LPG

1 0.40 0.49 0.96 0.76

1 4.3·10-2 0.50 6.9·10-2 0.17 HDA


1 0.40 0.42 0.45 0.46 4.4·10-2 4.6·10-3

820

36


1 9.7·10-2 0.97 3.6·10-3 2.22 0.15 0.16

Kinetic Modeling of the Hydrotreating and Hydrocracking Stages for

Recommend Documents