Kinetic evaluation of hydrodesulfurization and hydrodenitrogenation

Subscriber access provided by HACETTEPE UNIVERSITESI KUTUPHANESI

Article

Kinetic evaluation of hydrodesulfurization and hydrodenitrogenation reactions via lumped model Yiqian Yang, Fei Dai, Chunshan Li, Shuguang Xiang, Muhammad Yaseen, and Suo-Jiang Zhang Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b00496 • Publication Date (Web): 11 Apr 2017 Downloaded from http://pubs.acs.org on April 15, 2017

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 26

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

1

Kinetic evaluation of hydrodesulfurization and

2

hydrodenitrogenation reactions via lumped model Yiqian Yang,a,b Fei Dai,b Chunshan Li,b,c*,Shuguang Xiang, a Muhammad

3

Yaseenb and Suojiang Zhang,b*

4 5

a

Hi-Tech Institute for Petroleum, Chemical Industry, Qingdao University of Science, Technology,

6 7

Qingdao Shandong 266042, PR China b

Beijing Key Laboratory of Ionic Liquids Clean Process, State Key Laboratory of Multiphase

8

Complex System, The National Key Laboratory of Clean and Efficient Coking Technology, Institute of

9

Process Engineering, Chinese Academy of Sciences, Beijing 100190, PR China

10

c

School of Chemistry and Chemical Engineering, University of Chinese Academy of Sciences,

11 12

Beijing 100049, PR China

ABSTRACT

13

Multi-lump kinetic models, for model compounds representative of various

14

sulfur (S) and nitrogen (N) compounds in coker gas oil (CGO), were developed to

15

describe coker gas oil hydrodesulfurization(HDS) and hydrodenitrogenation(HDN)

16

process, respectively. Models parameters were obtained from fitting of operating data.

17

Validation results revealed that models could predict contents of S and N in products

18

accurately, and hydrogen consumption of HDS and HDN can be estimated. The effect

19

of temperature, hydrogen/oil ratio and pressure on the reaction performance were also

20

investigated.

21

1. INTRODUCTION

22

With the increasingly stringent environmental requirements, it is imperative to

23

produce higher quality fuel. Hydrorefining is the most effective method of removing

24

sulfur (S), nitrogen (N) and other impurities from petroleum oil.[1] Kinetic research

25

provides a significant guidance for parameter optimization of hydrogenation refining.

26

Many reports are available on kinetic models for hydrodesulfurization (HDS) and

27

hydrodenitrogenation (HDN)[2-5], which are mainly divided into two types, power

28

index model and Langmuir-Hinshelwood model. Nguyen[6] proposed a new

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

1

Langmuir-Hinshelwood model including L-V equilibrium and confirmed the reaction

2

scheme of HDN. Ferdous[7] used a single simplified L-H equation for HDS and HDN

3

reactions. In their work, it was proposed that the effect of ammonia adsorption on the

4

HDN and HDS reactions can be neglected while N and H2S adsorption had enormous

5

impact on HDN and HDS reactions. In general, the L-H model is not applicable to the

6

complicated real system attributed to presence of many coefficients which are difficult

7

to determine[8]. On the contrary, power index model for real system is widely adopted.

8

Power index model parameters can be obtained via the least squares fitting of the

9

experimental data. A pseudo-first-order kinetic model for total nitrogen, basic nitrogen

10

and non-basic nitrogen was presented by Wei[9]. They investigated the effects of

11

different operating conditions on HDN conversion which increased with increase in

12

temperature, hydrogen/oil ratio and pressure. Xiang[10] established a macro reaction

13

kinetic model for the HDS of coker gas oil (CGO) in a slurry reactor.

14

Due to the presence of large amount of N and S compounds in the CGO[9], it is

15

difficult to build specific kinetic models to describe the reaction process. Generally,

16

lump method is regarded as the most useful way to characterize the complex mixtures.

17

Torrisi[11] believed that S compounds with different boiling point showed different

18

reaction rate, normally reaction rate increased as its boiling point decreased. Tang[12]

19

proposed power index kinetic models to describe the process of HDS and HDN. The

20

S and N compounds in feed were lumped according to the reactivity into three or four

21

group cuts to obtain three- and four-lump kinetic models. And results showed that

22

three-lump and four-lump models predicted the performance of HDS and HDN

23

effectively with changing operational conditions and feedstock. Therefore, multi-lump

24

kinetic models are suitable for HDS and HDN of CGO. Generally, the more amount

25

of lumps in a kinetic model, the larger number of parameters need to be estimated by

26

using more detailed experimental data.[13] Thus, selecting a proper division method of

27

S and N compounds is the key to establish kinetic model accurately.

28

As for sulfur compounds in coker gas oil, based on the analysis of CGO a


Page 2 of 26

Page 3 of 26

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

1

conclusion can be drawed is that most of sulfur compounds exist in state of

2

benzothiophene (BT), dibenzothiophene (DBT) and benzonaphthothiophene (BNT),

3

Bannatham[14] established multi-lumped kinetic model to evaluate the process of HDS

4

in a trickle-bed reactor. The sulfur compounds in petroleum oil were divided into

5

several groups according to the boiling-point to establish multi-lumped models for

6

HDS. The predicted results were reliable and described the process of HDS accurately.

7

When it comes to nitrogen compounds in CGO, most of which exist in state of

8

pyridine, acridine, quinoline, carbazole and their derivatives. The multi-lumped

9

were divided into several groups according to its basic and reaction rate.

10

The study of catalyst, reactor, and reaction of CGO hydrorefining have been

11

widely reported [9,10,15,16]. As for heavy oil hydrorefining and kinetic model, our group

12

has done a lot of relevant researches[17-25]. It was proved that lumping method was a

13

well-established approach for characterizing complicated system. Results predicted by

14

models were in good agreement with the experimental data. In addition, experimental

15

data obtained from plant were objective reflection of performance of CGO

16

hydrorefining and provide sufficient data for estimating model parameters.

17

In this study model compounds were used to describe the constituents of S and N

18

in CGO and develop power index kinetic models suitable for CGO hydrorefining,

19

respectively. The effect of temperature, pressure, liquid hourly space velocity( LHSV)

20

and hydrogen/oil ratio on the reaction performance were also investigated and

21

considered in this model. In addition, hydrogen consumption can be calculated by

22

using empirical equations.

23

2. Methodology

24

For modeling the HDS and HDN, S and N compounds in CGO were divided into

25

two times three groups according to conversion of hydrogenation reaction and

26

reactivity, and two times three-lump kinetic models were established, respectively.

27 28

2.1 Description of HDS Model HDS of CGO is complicated process due to different type of S constituents, i.e.


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 4 of 26

1

Benzothiophene (BT), dibenzothiophene (DBT) and benzonaphthothiophene (BNT)

2

and their derivatives having different reaction rate and boiling range. Therefore, the

3

feed in the model can be divided into BT, DBT and BNT and their derivatives, which

4

react to H2S, as shown in Figure 1 assuming no mutual reaction between S

5

compounds.

6

2.2 Description of HDN Model

7

Similar to HDS, N compounds in HDN reaction are lumped into three groups

8

according to their reactivity. Pyridine, acridine, quinoline and their derivatives, mostly

9

consist in basic nitrogen with higher conversion, are grouped into lump 1. Carbazole

10

and its derivatives, having most of the non-basic nitrogen with low conversion, can be

11

regarded as lump 2. The other nitrogen compounds with highest conversion are

12

considered as lump 3. No reaction among individual N compounds was assumed for

13

the model.

14 15 16 17

18 19 20

2.3Kinetic Models On the basis of the above lump division, the n-order kinetic model of each lump can be described as follows: E (− a ) dCi a = -k0i ×（H2 / Oil） × Pb × e R×T × Cin 1 d( ) LHSV

(1)

where Ci represents the concentration of sulfur for lump 1, lump 2 and lump 3, µg/g, k0i is pre-exponential factors, h*MPa-b, H2/Oil refers to the H2/oil ratio in V/V, P is operating pressure in MPa, Ea is apparent activation energy in J/mol, LSHV is liquid

21

hourly space velocity, (h-1), R is gas constant, J/(mol*K),T is temperature in K, n

22

represents reaction order and a and b are fit parameters.

23 24 25

In the equation 1, the effect of operation conditions on the S(N) conversion are considered in the kinetic

Ci = ( C0ni−1 + ( 1 − n) × P b × （ H / Oil）a × k0 × e

26

−(

Ea ) R×T

×

1 1 )1− n (2) LHSV

where C0i refers to concentration of S or N in feed for lump 1, lump 2 and lump 3,


Page 5 of 26

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

1 µg/g. The three-lump kinetic model for hydrogenation of CGO is shown in equation 3: n1 −1 C total = ( C 01 + (1 − n1 ) × P b1 ×（ H 2 / Oil）a1 × k 01 × e

2

+ (C

n 2 −1 02

+ (C

n3 − 1 03

−(

Ea 1 ) R ×T

a2

b3

a3

+ (1 − n3 ) × P ×（ H 2 / Oil） × k 03 × e

1 ) 1− n1 LHSV

−(

Ea 2 ) R ×T

−(

Ea 3 ) R ×T

+ (1 − n 2 ) × P ×（ H 2 / Oil） × k 02 × e b2

1

×

5

1

1 × ) 1 − n3 LHSV (3)

3 4

1

1 × ) 1− n 2 LHSV

Based on the nonlinear regression and least-square method, the kinetic parameters were estimated which could describe HDS and HDN reactions accurately.

2.4 Estimation of Chemical Hydrogen Consumption

6 7

The chemical hydrogen consumption of CGO hydrorefining is mainly derived

8

from process of HDS and HDN. Based on the result of three-lump kinetic model, the

9

content of sulfur and nitrogen in product can be predicted under different operational

10

conditions. Empirical equations were regarded as an optimal method to estimate

11

chemical hydrogen consumption for CGO hydrorefining without detailed analytical

12

data. Equations are shown as following:

13

H HDS = α × (S f − SP )

(4)

14

H HDN = β × ( N f − N P )

(5)

15

where HHDS and HHDN is hydrogen consumption in HDS and HDN reaction, Nm3/m3,

16

respectively. Sf and Nf represent the content of S and N in feed respectively. Sp and Np

17

represent the content of sulfur and nitrogen in product, α and β are relevant

18

parameters, according to relevant literature[26], α=18~23, and β=62.

19

3. RESULTS AND DISCUSSION

20

3.1Effect of reaction temperature on HDN and HDS

21

As temperature is the most critical factor influencing HDS and HDN, the effect

22

of reaction temperature ranging from 500K- 680K at a pressure of 10 MPa, LHSV of

23

1 h-1, and hydrogen/oil ratio of 500 was investigated. Results are shown in Figure 3-4.

24

Conversion of S and N compounds increased with increasing reaction temperature

25

which became stable above 660K. Moreover, different types of S compounds showed

26

significant variations for HDS. Lump 1 showed the highest conversion and improve

27

rapidly with increasing temperature. Conversion of lump 2 was low below 540 K. At

28

temperature over 540K, catalytic activity improved rapidly, resulting in higher


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

1

conversion. The trend of lump 3 was similar to those of lump 2 and the only

2

difference was that conversion increase rapidly when temperature reached 570K. A

3

similar effect of temperature on HDN was observed. Conversions of lump 2 and lump

4

3 were more than those of lump 1 while Lump 1 gave higher conversion at high

5

temperature. Thus, the conversion of total nitrogen mainly depended on temperature.

6

However, the increase of coking and metal deposition were promoted with rising

7

temperature, resulting in catalyst deactivation rapidly which led to decreased

8

conversion. Thus, conversion and catalyst-life should be considered when selecting

9

the optimal reaction temperature. Based on Figure 3 and 4, HDS and HDN showed

10 11

satisfied conversion in a temperature range of 630K – 660K.

3.2 Kinetics of HDN and HDS

12

The contents of S and N compounds in product of CGO hydrorefining are

13

strongly affected by temperature, pressure, hydrogen/oil ratio and LHSV. Different

14

plant data sets of various level of operational conditions(Table 3-4) were fitted with

15

three-lump kinetic models. Kinetics parameters of three-lump models for CGO HDN

16

and HDS were determined and shown in Table 1 and 2.

17

The comparison of experimental data and calculated results are shown in Figure

18

5. The predicted content of S and N compounds in product from the three-lump model

19

were in good agreement with those obtained from experimental data. It appeared that

20

three-lump model established in present work can predict the performance of HDS

21

and HDN accurately at different operating conditions. The average absolute relative

22

deviations(AARD) of HDS and HDN was 0.34% and 1.71%, respectively.

23

In order to validate the model, seven sets of experimental data at different

24

operating conditions with same feedstock were obtained and shown in Figure 6. The

25

three-lump kinetic model also predicted the conversion of sulfur and nitrogen

26

compounds accurately. The AARD of HDS and HDN was 4.06% and 1.31%

27

respectively. These results proved that three-lump models were suitable for CGO

28

HDS and HDN and can accurately predict concentration of S and N compounds in

29

product stream at fixed operating conditions.

30

The same model can also be used to estimate S and N compounds in CGO

31

hydrorefining process at different operating conditions. Data for these investigations

32

are shown in Figure 7-9, which indicate that S compounds were preferentially

33

removed, and temperature and pressure were the most influential factors on the


Page 6 of 26

Page 7 of 26

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

1

performance of HDS and HDN. Figure 7 shows that concentration of S compounds

2

decreased with increasing time. This decrease was rapid initially and 0.1h onward

3

became linear. This trend was very much linear and gradual for HDN regarding N

4

compounds decreased. Compare with HDS and HDN, the effect of temperature on

5

HDN was more remarkable than HDS and nitrogen compounds are more difficult to

6

be removed.

7

Figure 8 shows the effect of hydrogen/oil ratio on HDN and HDS. The content of

8

S and N in product reduce with increasing the amount of hydrogen under the same

9

operational condition. But little variation can be seen in performance of HDS and

10

HDN with the change in hydrogen/oil ratio. It can be proved that hydrogen/oil ratio

11

was

12

hydrodenitrogenation. The increase of hydrogen/oil ratio is favourable to more

13

hydrogen participate in reaction, promoting the degree of hydrorefining. However,

14

residence time of reactant on catalyst bed decreases as hydrogen/oil ratio increases,

15

leading to reducing reaction time and decreasing the degree of hydrorefining.

16

Therefore, a proper hydrogen/oil ratio facilitate the process of hydrorefining.

not

a

critical

factor

affecting

the

CGO

hydrodesulfurization

and

17

Figure 9 shows the effect of pressure on performance of HDS and HDN. It was

18

observed that pressure predominantly affected HDN as compared to HDS while in

19

both the cases, increase in pressure resulted in higher performance of both the

20

processes for CGO hydrorefining. This is because that improving pressure can

21

increase hydrogen partial pressure, which is benefit with HDS and HDN. Besides, the

22

increase of hydrogen partial pressure promotes the hydrogenation saturation of

23

aromatics, which hydrogenation saturation of aromatics is first step to remove

24

nitrogen compounds, and finally accelerate the reaction rate of HDN .

3.3Estimation of Hydrogen consumption

25 26

The lump model was also used to predict the amount of hydrogen consumption

27

in hydrorefining process. The total amount of hydrogen consumption predicted was

28

0.66% which was in good agreement with the experimental data from plant. To

29

illustrate the applicability of the empirical equations, the empirical equations were

30

used for other experimental data, and the results were shown in Figure 10. Reference

31

literature data were compared with the empirical equations for hydrogen consumption,

32

and the AARD was 7.99%. Calculation showed good agreement with experimental

33

data, proving that the empirical equations possess wide adaptation.


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

1

4. CONCLUSIONS

2

On the basic of the experimental data of CGO hydrodesulfurization and

3

hydrodenitrogenation, which were recorded at the steady state, a three-lump kinetic

4

models were proposed, in which the sulfur and nitrogen compounds in the feed were

5

respectively grouped into three lumps according to composition of CGO and reaction

6

mechanism. Results showed that three-lump model can effectively predict

7

conversions of sulfur and nitrogen. The effect of operational conditions on HDS and

8

HDN were discussed. It can be concluded that temperature and pressure were as

9

major influencing factors on the performance of HDS and HDN. Combined with

10

empirical equations for hydrogen consumption and predicting the content of sulfur

11

and nitrogen in product by using three-lump kinetic model, the chemical hydrogen

12

consumption can be obtained. The present model can help in formulation of reaction

13

modeling, mechanism and products distribution in CGO hydrorefining process on

14

industrial level.

15

ACKNOWLEDGEMENTS

16

Project supported by the National Science Fund for Excellent Young Scholars

17

(21422607), The National Natural Science Funds (No. 21576261). National key R &

18

D program (2016YFB0601303), Innovation Fund of Petro-China (2015D-5006-0406)

19

NOMENCLATURE

20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

a= fit parameter b=fit parameter Ccal= calculated content, µg/g Cexp= experimental content , µg/g Ci=the concentration of sulfur for lump i, i=1,2, 3, µg/g C0i= concentration of sulfur(nitrogen) in feed for lump 1, lump 2 and lump 3, µg/g Ea=apparent activation energy, kJ·mol-1 H= the H2/oil ratio, V/V HHDS= hydrogen consumption in hydrodesulfurization reaction, Nm3/m3 HHDN= hydrogen consumption in hydrodesulfurization reaction, Nm3/m3 k0i= pre-exponential factors, h-1MPa-bi LSHV=liquid hourly space velocity,h-1 Nf= the mass fraction of nitrogen in feed Np= the mass fraction of nitrogen in product n = reaction order P=operation pressure, PMa


Page 8 of 26

Page 9 of 26

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

1 2 3 4

R=gas constant, 8.314 j·mol-1·K-1 Sf= the mass fraction of sulfur in feed Sp= the mass fraction of sulfur in product

5

T=reaction temperature, K α and β=relevant parameters

6

Reference

7

(1)Martínez J, Ancheyta J. Modeling the kinetics of parallel thermal and catalytic

8

hydrotreating of heavy oil. Fuel 2014; 138:27-36.

9

(2)Ferdous D, Dalai A K, Adjaye J. Hydrodenitrogenation and Hydrodesulfurization

10

of Heavy Gas Oil Using NiMo/Al2O3 Catalyst Containing Boron: Experimental and

11

Kinetic Studies. Canadian Journal of Chemical Engineering 2005; 83(83):855-864.

12

(3)Vonortas A, Papayannakos N. Kinetic Study of the Hydrodesulfurization of a

13

Heavy Gasoil in the Presence of Free Fatty Acids Using a CoMo/γ-Al2O3 Catalyst.

14

Ind. Eng. Chem. Res 2014; 53(23):9646-9652.

15

(4)Hensen

16

hydrodesulfurization over carbon-supported transition metal sulfides. Journal of

17

Catalysis 1996; 163(2):429-435.

18

(5)Ferreira C, Tayakout-Fayolle M, Guibard I, et al. Hydrodesulfurization and

19

hydrodemetallization of different origin vacuum residues: New modeling approach.

20

Fuel 2014; 129(7):267-277.

21

(6)Nguyen M T, Tayakoutfayolle M, Gerhard Pirngruber, Chainet F, Geantet C.

22

Kinetic Modeling of Quinoline Hydrodenitrogenation over a NiMo(P)/Al2O3 Catalyst

23

in a Batch Reactor. Ind. Eng. Chem. Res. 2015; 58(38):9278-9288.

24

(7)Ferdous D, Dalai A K, Adjaye J. Hydrodenitrogenation and Hydrodesulfurization

25

of Heavy Gas Oil Using NiMo/Al2O3 Catalyst Containing Boron: Experimental and

26

Kinetic Studies.Ind. Eng. Chem. Res.2006; 45:544-552.

27

(8)Whitehurst D D, Takaaki I, Mochida, I. Present State of the Art and Future

28

Challenges in the Hydrodesulfurization of Polyaromatic Sulfur Compounds.

29

Advances in Catalysis 1998; 42: 344-368

30

(9)Wei Q, Wen SH, Tao XJ, Zhang T, Zhou YS, Chung K, Xu CM.

31

Hydrodenitrogenation of basic and non-basic nitrogen-containing compounds in coker

32

gas oil. Fuel Processing Technology 2015; 129:76-84.

33

(10)Xiang HD, Wang TF. Kinetic study of hydrodesulfurization of coker gas oil in a

34

slurry reactor. Frontiers of Chemical Science & Engineering 2013;7(2):139-144.

35

(11)Torrisi S, Remans T, Swain J. The challenging chemistry of ultra-low-sulfur

36

diesel. Process. Technol. Catal. 2009;1−4.

37

(12)Tang X, Li SY, Yue CT, He JL, Hou JL. Lumping kinetics of hydrodesulfurization

E,

Vissenberg

M

J.

Kinetics

and

mechanism


of

thiophene

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

1

and hydrodenitrogenation of the middle distillate from Chinese shale oil. Oil Shale

2

2013; 30(4):517-535.

3

(13)Dai F, Gao MJ, Li CS, Xiang SG, Zhang SJ. Detailed Description of Coal Tar

4

Hydrogenation Process Using the Kinetic Lumping Approach. Energy & Fuels 2011;

5

25(11):4878-4885.

6

(14)Bannatham P, Teeraboonchaikul S, Patirupanon T, et al. Kinetic Evaluation for

7

Hydrodesulfurization via Lumped Model in a Trickle-Bed Reactor[J]. Industrial &

8

Engineering Chemistry Research, 2016, 55(17).

9

(15)Yang YQ, Wang HY, Dai F, Xiang SG*, Li CS*. Simplified catalyst lifetime

10

prediction model for coal tar in hydrogenation process. Energy & Fuels 2016; 30(7):

11

6034-6038.

12

(16)Wang G, Liu Y, Wang X, Xu C, Gao J. Studies on the Catalytic Cracking

13

Performance of Coker Gas Oil. Energy & Fuels 2009; 23(4):1942-1949.

14

(17)Dai F, Gong MM, Li CS, Li ZX, Zhang SJ. Molecular Kinetic Model of Coal Tar

15

Hydrogenation Process Based on the Carbon Number-based Component Approach.

16

Appl. Energy 2015; 137(1): 265-272.

17

(18)Wang HY, Dai F, Li ZX, Li CS. Upgrading shale oil distillation to clean fuel by

18

coupled hydrogenation and ring opening reaction of aromatics on W-Ni/γ-Al2O3

19

catalysts. Energy & Fuels 2015;29(8):4902-4910.

20

(19)Dai F, Wang HY, Gong MM, Li CS, Muhammad Y, Li ZX. Modeling of

21

kinetic-based catalyst grading for upgrading shale oil hydrogenation. Fuel 2016;

22

166(15): 19-23.

23

(20)Dai F, Wang HY, Gong MM, Li CS, Li ZX, Zhang SJ. Carbon-Number-Based

24

Kinetics, Reactor Modeling, and Process Simulation for Coal Tar Hydrogenation.

25

Energy & Fuels 2015, 29: 7532−7541.

26

(21)Dai F, Yaseen M, Gong X, Li CS, Li ZX, Zhang SJ. Low-temperature and

27

low-pressure fuel hydrodesulfurization by solid catalyst coupling with ionic liquids.

28

Fuel 2014; (134): 74-80.

29

(22) Wang HY, Jiao TT, Li ZX, Li CS, Zhang SJ, Zhang JL. Study on palmoil

30

hydrogenation for clean fuel over Ni-Mo-W/γ-Al2O3-ZSM-5 catalyst. Fuel Processing

31

Technology 2015; 139: 91-99.

32

(23) Gaurav A, Ng F T T, Rempel G L. A New Green Process for Biodiesel

33

Production from Waste Oils via Catalytic Distillation using a Solid Acid

34

Catalyst-Modelling. Economic and Environmental Analysis. Green Energy &

35

Environment 2016;1(1):62-74

36

(24)Wang HY, Li CS, Peng ZJ, Zhang SJ. Study on the Combustion Kinetic

37

Characteristics of Coal Tar under Catalysts. Energy Sources, Part A: Recovery,


Page 10 of 26

Page 11 of 26

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

1

Utilization, and Environmental Effects, DOI:10.1080/15567036.2011.613892.

2

(25)Dai F, Gong MM, Li CS, Li ZX, Zhang SJ. New kinetic model of coal tar

3

hydrogenation process via carbon number component approach. Applied Energy 2015;

4

137(C):265-272.

5

(26)Li DD. Hydrotreating Processes and Engineering. China Petrochemical Press

6

2004.

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


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

1

Figure captions:

2

Figure 1. Conversion of sulfur compounds via HDS

3

Figure 2. Conversion of nitrogen compounds via HDN

4 5

Figure 3. Effect of temperature on the conversion of total, BT, DBT, and BNT Figure 4. Effect of temperature on the conversion of total, BN, NBN, and other

6

compounds

7

Figure 5. Comparison of predicted and experimental data

8

Figure 6. Comparison of predicted value from the calculation and 7 sets of

9

experimental data

10

Figure 7. The effect of temperature on CGO HDS and HDN

11

Figure 8. The effect of H2/Oil on CGO HDS and HDN (P: 10MPa,LHSV: 1 h-1, T: 643 K)

12

Figure 9. The effect of pressure on CGO HDS and HDN (hydrogen/oil: 500, LHSV: 1 h-1,

13 14

T: 643)

Figure 10. Comparison of predicted value from the calculation and literature data

15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35


Page 12 of 26

Page 13 of 26

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

1 2 3 4 5 6 7 8 9 10 11 12 13

Lump1

14

(BT and derivatives)

Lump3

15 H2S+CnHm

16

Lump2

17

(DBT and derivatives)

18

(BNT and derivatives)

Figure 1. Conversion of sulfur compounds via HDS

19 20 21 22 23 24 25 26 27 28 29 30 31


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 14 of 26

1 2 3 4 5 6 7 8 9 10 11 12

Lump 1 (Pyridine, acridine,

Lump 3

quinoline and derivatives)

13 14

NH3+CnHm

(The other nitrogen compounds)

Lump 2 (Carbazole and its derivatives)

15 16

Figure 2. Conversion of nitrogen compounds via HDN

17 18 19 20 21 22 23 24 25 26 27 28


Page 15 of 26

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

1 2 3 4 5 6 7 8 9 10 11 12

13 14 15

Figure 3. Effect of temperature on the conversion of total, BT, DBT, and BNT compounds( X is conversion rate: X=

Cexp -Ccal Cexp

16 17 18 19 20 21 22 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

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

18 19 20

Figure 4. Effect of temperature on the conversion of total, BN, NBN, and other compounds

21 22 23 24 25 26 27 28


Page 16 of 26

Page 17 of 26

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

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

Figure 5. Comparison of predicted and experimental data

26 27 28 29 30 31 32 33 34 35 36 37


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

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

Figure 6. Comparison of predicted value from the calculation and 7 sets of experimental data

28 29 30 31 32 33 34 35 36 37 38


Page 18 of 26

Page 19 of 26

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

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Figure 7. The effect of temperature on CGO HDS and HDN

25

(P: 10MPa, LHSV: 1 h-1, hydrogen/oil: 500 )

26 27 28 29 30 31 32 33 34 35 36 37 38


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

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

Figure 8. The effect of H2/Oil on CGO HDS and HDN

23

(P: 10MPa,LHSV: 1 h-1, T: 643 K)

24 25 26 27 28 29 30 31 32 33 34 35 36 37 38


Page 20 of 26

Page 21 of 26

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

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

Figure 9. The effect of pressure on CGO HDS and HDN

27

(hydrogen/oil: 500, LHSV: 1 h-1, T: 643)

28 29 30 31 32 33 34 35 36 37


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

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

16 17

Figure 10. Comparison of predicted value from the calculation and literature data

18 19 20 21 22 23 24 25 26 27 28


Page 22 of 26

Page 23 of 26

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

1

Table captions:

2

Table 1. Kinetic parameters of HDN reaction

3

Table 2. Kinetic parameters of HDS reaction

4

Table 3. Operating Data for HDN of CGO

5

Table 4. Operating Data for HDS of CGO

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


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 24 of 26

1 2 3 4 5 6 7

Table 1. Kinetic parameters of HDN reaction

Parameters

Lump 1

Lump 2

Lump 3

a

0.30

0.11

0.14

b

1.70

1.14

1.25

n

1.09

1.54

1.13

Ea, J/mol

1.56×105

1.25×105

1.23×105

k0, mol1-nLn-1h-1

5.70×109

3.35×107

9.57×107

8 9 10 11 12 13 14 15

Table 2. Kinetic parameters of HDS reaction

Parameters

Lump 1

Lump 2

Lump 3

a

0.31

0.30

0.31

b

0.59

0.59

0.63

n

1.69

1.54

1.16

Ea, J/mol

1.33×105

1.35×105

1.52×105

k0, mol1-nLn-1h-1

7.36×108

3.17×108

3.84×1010

16 17 18 19 20 21


Page 25 of 26

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

1 2 3

Table 3. Operating Data for HDN of CGO

P

Product, µg/g

Feed, µg/g

H/Oil

LHSV

K

V/V

h-1

10

638

500

1

569

260

775

927

1159

2069

10

643

500

1

484

221

659

927

1159

2069

10

648

500

1

397. 5

182

541

927

1159

2069

10

653

500

1

305

139

414

927

1159

2069

10

658

500

1

240

110

327

927

1159

2069

10

643

500

0.5

174

79

237

927

1159

2069

10

643

500

0.7

313

143

426

927

1159

2069

10

643

500

1

484

221

659

927

1159

2069

10

643

500

1.2

610

278

830

927

1159

2069

10

643

500

1.5

670

306

912

927

1159

2069

6

643

500

1

834

381

1135

966

1209

2158

8

643

500

1

660

301

898

966

1209

2158

10

643

500

1

484

221

659

927

1159

2069

12

643

500

1

387

176

526

927

1159

2069

14

643

500

1

272

124

370

966

1209

2158

10

643

700

1

435

198

591

927

1159

2069

10

643

1000

1

443

202

603

927

1159

2069

MPa

T

Lump1 Lump2 Lump3 Lump1 Lump2 Lump3

4 5 6 7 8 9 10 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

Page 26 of 26

1 2 3 4

Table 4. Operating Data for HDS of CGO

Product, µg/g

Feed, µg/g

P

T

H/Oil

LHSV

MPa

K

V/V

h-1

10

633

500

1

16

295

1109

5696

2779

3055

10

638

500

1

12

227

851

5696

2779

3055

10

643

500

1

9

166

622

5696

2779

3055

10

648

500

1

6

117

440

5696

2779

3055

10

653

500

1

4

83

310

5696

2779

3055

10

658

500

1

3

63

238

5696

2779

3055

10

643

500

0.5

3

56

211

5696

2779

3055

10

643

500

0.7

6

110

412

5696

2779

3055

10

643

500

1

9

166

622

5696

2779

3055

10

643

500

1.2

14

255

959

5696

2779

3055

10

643

500

1.5

15

281

1054

5696

2779

3055

6

643

500

1

14

259

972

5612

2738

3010

8

643

500

1

11

215

808

5612

2738

3010

10

643

500

1

9

166

622

5696

2779

3055

14

643

500

1

7

139

522

5612

2738

3010

10

643

500

1

9

166

622

5696

2779

3055

10

643

700

1

8

154

577

5696

2779

3055

Lump1 Lump2 Lump3 Lump1 Lump2 Lump3

5 6


Kinetic evaluation of hydrodesulfurization and hydrodenitrogenation

Recommend Documents