23
Intensification of the Steam Cracking Process Mohamed Ellob, Jonathan Lee, Arthur Gough School of Chemical Engineering and Advanced Materials University of Newcastle Does Steam Cracking Need Steam

Intensification of the Steam Cracking · PDF file · 2007-06-01Intensification of the Steam Cracking Process Mohamed Ellob, Jonathan Lee, Arthur Gough School of Chemical Engineering

Embed Size (px)

Citation preview

Intensification of the Steam Cracking Process

Mohamed Ellob, Jonathan Lee, Arthur Gough

School of Chemical Engineering and Advanced Materials

University of Newcastle

Does Steam Cracking Need Steam

Presentation outline

IntroductionCatalytic plate reactorsCoke formationObjectivesBenefitsMethodologyExperimental WorkResultsConclusions

IntroductionOlefins demand in year 2005 :

Ethylene (107 million tons )propylene (67.1 million tons )

Olefins demand growth during years(2005 –2010):Ethylene about 4.3% per yearpropylene about 5.4% per year

Olefins production capacity growth:Ethylene about 5.4% per yearPropylene about 5.1% per year

Typical Steam Cracking Furnaces

Total Number of cracking tubes about 600

Total Reaction Volume about 45 m³

Total firebox volume about 9,000 m³

Residence Time 0.25 to 0.75 s

Firebox Efficiency about 65%

Steam function and process limitation

Enhance heat transferReduce coke formation and depositionImprove selectivity towards ethyleneOperation purposesCoke deposition is the main process limitations due to:- High tube skin temperature- High pressure drop

Exothermic channel gas

Endothermic channel gas

Heat

Exothermic reaction catalyst

Endothermic reaction catalyst

Catalytic Plate Reactor

Thin plate

Source: Velocys,(2005),olefins by high intensity oxidation, http: // www.velocys.com/Img/pdf.2250.pdf

Advantages Of Catalytic Plate Reactor

High Surface to volume RatioLaminar flow ConditionsHigh Heat transfer CoefficientThin Catalyst Layer Minimize Diffusion LimitationSurface Temperature only few degrees above the process temperatureImproved Safety and Environmental ImpactScale-up by Numbering –upLow Capital and operating Costs

Coke formationMetal-catalyzed cokeNon-catalytic coke from tarsSmall chemical species (coke precursors) react with free radicals on the coke surface

Chemisorbed

Surface reaction

Heterogeneous Catalytic

HC. Molecule

Heterogeneous non catalytic

RadicalMolecule

Radical on coke surface

Coke particle

Homogeneous non catalytic

Tar droplet

Objectives

1-Study and investigate the possibility of intensifying the thermal cracking of propane to produce ethylene through the use of the catalytic plate reactors.

2- Reducing the coke formation and deposition.

3- Reducing the use of steam.

4- Modelling and simulation for propane cracking using Catalytic Plate Reactor.

Benefits

Lower environmental and safety impacts.(NOx, contaminated water, CO2, H2S)

Improved energy efficiency.

Lower capital cost.

Improved overall plant economics

Experimental setup design criteria

Allows for accurate coke measurement

Constant and uniform temperature along the reactor

Very fast cooling of reaction products

Easy to change reactor size and material

V-1 V-2 Filter-1

Nitrogen

V-3 V-4 Filter-2

Propane

V-5 V-6 Filter-3

Air

V-7 V-8 Filter-4

Argon

V-9

FIC-1

Reactor

V16

V17

Knock-out Vent

To GCs

PI-1

FIC-2

V-10

FIC-3

V12 V14

To CO2

analyser

PI-2

Propane crackershowing flow paths

during decoke

V13

V15

FIC-4

Manometer

Catchpot

Catchpot

Experimental variables

Reactor materials and internal coatings

Reactor channel size

Process variables ( temperature, pressure, and flow rate)

Run time length

lowhigh low

high60

65

70

75

80

85

90

95

100

Con

vers

ion,

% Pressure Temperature

Low flow rate

lowhigh low

high60

65

70

75

80

85

90

95

100

Con

vers

ion,

%

Pressure Temperature

High flow rate

Conversion at low and high operating parameters

Coke yield vs Flow at 835 oC and 1.35 bar

0

0.2

0.4

0.6

0.8

1

1.2

1.5 2 2.5 3 3.5 4 4.5 5

Flow, l/h

Cok

e yi

eld

( mg/

g-pr

opan

e re

acte

d)

Coke yield vs Temperature. at 3.5 l/h and 1.35 bar

0.2

0.4

0.6

0.8

1

1.2

810 820 830 840 850 860

Temperature,oC

Cok

e yi

eld

(mg/

g-pr

opan

e re

acte

d)

Coke yield vs Pressure at 3.5 l/h and 835 oC

0.1

0.3

0.5

0.7

0.9

1.1

0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8

Pressure, bar

Cok

e yi

eld

(mg/

g-pr

opan

e re

acte

d)

Coke yield at low and high operating parameters

LowHigh Low

High0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

Cok

e yi

eld

(mg/

g-pr

opan

e re

acte

d)

pressure Temperature

Low flow rate

LowHigh Low

High0.2

0.3

0.4

0.5

0.6

Cok

e yi

eld

(mg/

g-pr

opan

e re

acte

d)

pressure Temperature

High flow rate

Inlet of the reactor

Middle of the reactor

Outlet end of the reactor + Quenching line

UY of Ethylene at low and high operating parameters

lowhigh low

high40

42

44

46

48

50U

Y o

f Eth

ylen

e, W

eigh

t %

Pressure Temperature

Low flow rate

lowhigh low

high40

42

44

46

48

50

UY

of E

thyl

ene,

Wei

ght

%

Pressure Temperature

High flow rate

FeedIn Propane MassProduced Ethylene MassYield Ultimate =

Assuming that the unreacted propane andthe ethane produced by one pass throughthe reactor, are recycled to the feed

Conclusions • Conversion of about 90 % can be achieved in 2 mm internal

diameter fused silica reactor without any significant pressure drop.

• Steam use can be reduced or possibly eliminated.

• High olefins yield can be obtained without steam.

• Low acetylene and C4+ yield.

• Run length of about 14 – 20 days was estimated to be possible before any decoking is required. This run length was achieved with no steam.

THANK YOU