Methods for predicting the hydrocarbon yield in the ... · Methods for predicting the hydrocarbon yield in the catalytic cracking of lignite 8th International Freiberg Conference

M.Seitz | Catalytic Cracking of Lignites

23/06/16 | page 1

Methods for predicting the

hydrocarbon yield in the catalytic

cracking of lignite

8th International Freiberg Conference on IGCC & XtL Technologies

Innovative Coal Value Chains

12-16 June 2016, Cologne, Germany

Mathias Seitz*, Armin Engelhardt*, Jens Zimmermann*, Timo Stam-Creutz*, Thomas Hahn**

* University of Applied Science, Merseburg, Germany, **MLU Halle-Wittenberg, Halle, Germany


23/06/16 | page 2

BGR 2008

reserves

In Gt

assumed

accumulated

production

since 1950

resources reserves production

Introduction (1) Lignite: reserves, resources and production


23/06/16 | page 3

Introduction (2): different types of lignites


23/06/16 | page 4

oil

lignite

H rich

yellow lignite:

- high content of liptinite

high content of aliphatic molecules

Introduction (3): different types of lignites

1872 Riebeck‘schen Montanwerke am Standort Oberröblingen;

Quelle: LMBV

Historical oil and paraffin production

by low-temperature carbonisation


23/06/16 | page 5

Batch Concept of Catalytic Cracking

cracking in situ

at 400 – 600°C no use of hydrogen

no heavy tar formation

catalyst regeneration

catalyst coated beads

inert gas

solids

(char/catalyst separation)

vapor

(gas/liquid separation)


23/06/16 | page 6

meth

an

e

eth

an

e

eth

en

e

pro

pan

e

pro

pen

e

bu

tan

e

bu

ten

e

pen

tan

e

pen

ten

e

hexan

e

hexen

e

C7

C8

ben

zen

e

tolu

en

e

eth

ylb

en

zen

e

xyle

ne

meb

0,0

0,5

1,0

1,5

2,0

yie

ld [

%m

af]

without catalyst

catalyst powder

beads

excess powder

Typical products

Conditions: Batch:, T = 400 °C, τ = 100 s , coal(maf) /catalyst ratio = 20 : 1; MFI


23/06/16 | page 7

mining

NIR-

analysis

prediction

H/C ratio

prediction

HC yield

batch

prediction

HC yield

continuous

Three steps of modeling - from mining to hydrocarbons -

H/C ratio as a light

measurable factor

to predict the

hydrocarbon yield.

Easy statements

about the

achievable

products.

Prompt information

about the potential of

the lignite.


23/06/16 | page 8

Characterization of molar H/C ratio

NIR spectrum of lignite to determine the molar H/C-ratio Cepus, V., et. al. IR Spectroscopic Characterization of Lignite as a Tool to Predict the Product Range of Catalytic Decomposition, International Journal of Clean Coal and Energy (2016) 5, pp. 13-22


23/06/16 | page 9

Characterization of molar H/C ratio

Cepus, V., et. al. Journal of Clean Coal and Energy (2016) 5, pp. 13-22

Assumption:

- molar O/C ratio

has no influence


23/06/16 | page 10

Influence of the molar H/C ratio



23/06/16 | page 11

0.00 0.02 0.04 0.06 0.08 0.100

5

10

15

20

different CTC; H/C 1.3

variation modulus H/C; 1.3

different CTC; H/C 1.02

variation modulus H/C; 1.02yie

ld p

er

lig

nit

e (

ma

f) i

n M

a-%

amount of acid sites (maf) in mmol/g

Influence of the amount of acid sites


CTC: catalyst to

coal ratio (mass)

modulus: molar

Al2O3/SiO2 ratio

of the catalyst

modulus > 60 because of less

contact of catalyst and lignite

and coking


23/06/16 | page 12

Influence of the amount of …


- different catalysts

- different catalyst properties

- H/C ratio; O/C ratio

- temperature

- catalyst to coal ratio

- gas residence time

- solid residence time

- inert gas/solid ratio

- reactor operation mode; flash behaviour

- coal/catalyst contact

- water content of the lignite

Nearly all measurements

of the batch process

(except some catalyst

types) were used to find a

an as easy as possible

model !


23/06/16 | page 13

Batch model

4

6

8

10

12

14

16

18

20

22

24

ad

jus

ted

hy

dro

ca

rbo

n y

ield

C1

-C9

[%

ma

f]

0,8 0,9 1 1,1 1,2 1,3 1,4molar H/C-ratio [-]

0,2 0,4 0,6 0,8modificated acidic centre ratio [-]

450 500 550temperature [°C]

Adjusted Response Graph - Hydrocarbon yield C1-C9 (86 experiments)

a -1,13E+02 9,91%

b 4,14E-01 11,52%

c 3,13E+01 4,95%

d -5,37E-02 8,52%

e -3,16E-04 15,71%

R2 = 0,9827

𝒀𝑪𝟏−𝟗 = 𝒂 + 𝒃 ∙ 𝑻 °𝑪 + 𝒄 ∙𝒏𝑯𝒍𝒊𝒈𝒏𝑪𝒍𝒊𝒈

∙𝒏𝑨𝒄𝒎𝒐𝒅𝒎𝒍𝒊𝒈

+ 𝒅 ∙ 𝑻 ∙𝒏𝑨𝒄𝒎𝒐𝒅𝒎𝒍𝒊𝒈

+ 𝒆 ∙ 𝑻𝟐

HC yield

parity plot


23/06/16 | page 14

Continuous concept of catalytic cracking


23/06/16 | page 15

batch reactor conti reactor

From batch to a continuous reactor

carrier

gas

carrier

gas

Reaction parameters:

gas residence time: 1 sec

lignite residence time 40 sec

Reaction parameters:

gas residence time: 100 sec

lignite residence time 40 min


23/06/16 | page 16

0

2

4

6

8

oil

ph

as

e

me

b

xy

len

e

C7

eth

ylb

en

ze

ne

he

xe

ne

tolu

en

e

be

nze

ne

C8

he

xa

ne

yie

ld [

% m

af]

batch H/C 1.1

batch H/C 1.3

continuous H/C 1.3

pe

nte

ne

pe

nta

ne

bu

ten

e

bu

tan

e

pro

pe

ne

pro

pa

ne

eth

en

e

eth

an

e

me

tha

ne

Batch and continuous reactor


23/06/16 | page 17

Continuous model:


23/06/16 | page 18

Continuous model: parameter value error

3.207 ± 18 %

0.929 ± 18 %

2.284 ± 7,4 %

HC yield

parity plot


23/06/16 | page 19

Continuous model vs batch


23/06/16 | page 20

Continuous model vs batch

more

catalyst

activity has to be

adjusted to the

amount of volatile

matter and the gas

residence time


23/06/16 | page 21

H/C 1.3 (with transfer of the linear batch relation to the continuous model)

CTC 1:10; T 600°C

batch model: YHC = 27.7%

continuous model: YHC = 27.0%

continuous measured: YHC = 28.7%

Continuous model vs batch –

extrapolation/validation


23/06/16 | page 22

Summary

lignite is an interesting feedstock

- cheap, - available, - can obtain paraffins

catalytic conversion in situ without hydrogen to

olefins (about 11 % maf) and aromatics is possible

model for

- rapid prediction of molar H/C ratio is possible

- prediction of HC yield of batch process is possible

- prediction of HC yield of continuous process is possible

but work is to be done.


23/06/16 | page 23

- physico-chemical based models

- models to describe the single products

- extension to different lignites with higher O/C

- improvement of the continuous model in

consideration of the residence time

- catalyst and its properties

Outlook


23/06/16 | page 24

Acknowledgement

for financial support

for cooperation

03WKBZ05

Methods for predicting the hydrocarbon yield in the ... · Methods for predicting the hydrocarbon yield in the catalytic cracking of lignite 8th International Freiberg Conference

Documents