An Overview of Industrial Processes For The Production of Olefins - C4 Hydrocarbons
An Overview of Industrial Processes For The Production of Olefins - C4 Hydrocarbons
An Overview of Industrial Processes For The Production of Olefins - C4 Hydrocarbons
of Olefins C4 Hydrocarbons
Michael Bender[1]
Abstract
The survey describes industrial processes for the
production of light olefins emphasizing on C4 hydrocarbons. Value chains in the petrochemical and
the oil refining industry are strongly interrelated
with regards to olefin production. An account is
www.ChemBioEngRev.de
[1]
136
a)
Dehydrogenation
2%
Steam cracking,
Ethane & LPG
11%
Steam cracking,
liquid feed
28%
Fluid-catalytic
cracking
59%
b)
Fluid / Deep
Catalytic Cracking
5%
Steam cracking,
Ethane & LPG
38%
Steam cracking,
Naphtha / GO
57%
Ethylene
156 Mmto / +4.5% p.a.
c)
Propane
Dehydrogenation
5%
Metathesis
5%
Fluid / Deep
Catalytic Cracking
39%
d)
Steam cracking,
Ethane & LPG
4%
Delayed
Coking
1%
Fluid-catalytic
cracking*
69%
Steam cracking,
liquid feed
22%
Butane
Dehydrogenation
4%
Crude C4 yield, total
132 Mmto / +4 % p.a.
www.ChemBioEngRev.de
processes for olefin production yield C4 hydrocarbons in different ratios. Steam cracking almost exclusively yields C4 olefins,
while FCC units are operated such that the total C4 olefin content and the isobutane content of the raffinate 1 stream are in a
stoichiometric one-to-one balance. This ratio is required to further process these components into alkylate gasoline. n-Butane
does not react under the alkylation process conditions. Other
ratios of C4 hydrocarbons are obtained, e.g., in the upcoming
dehydrogenation processes (Tab. 1).
Beyond butadiene production, raffinates are further processed to a number of different products. Raffinate 1 is used to
manufacture methyl tert-butyl ether (MTBE) and ethyl tert-butyl ether (ETBE) by acid-catalyzed, selective etherification of
isobutene with methanol and ethanol, respectively. Both MTBE
and ETBE are large-scale products used as important additives
to the gasoline pool.
Alternatively, C4 olefins can be catalytically converted with
isobutane in raffinates yielding alkylate gasoline, another important additive to the gasoline pool. Additional isobutene
volumes can be generated from n-butenes by acid-catalyzed
skeletal isomerization. As an example for this process type
CDTechs ISOMPlus process can be named. ISOMPlus operates
a ferrierite catalyst at 340 to 440 C to generate isobutene from
n-butenes by rearranging the molecular carbon skeleton [13].
Isobutene can be recovered in substance by acid-catalyzed
cleavage of MTBE or iso- and tert-butanol, respectively. Isobutene is consumed in substance, e.g., for the production of polyisobutene (PIB).
1-Butene can be isomerized by acid catalysis or by hydroisomerization to 2-butene. N-olefins can be produced in substance
by catalytic distillation. They are used for some specific, mostly
chemical production processes.
137
primary generation
Bio mass
Syn
Gas
MeOH
BD
Extract
Butanediol
Methanol-toOlefins
Catalytic
Distillation
N-butene Isom.
Skeletal
Isomer.
Butene-1
Alkylate
gasoline
Full C4 Hydro
Ethanol
Isobutene
Butanes
n-BDH
Iso-BDH
EthanoltoButadiene
(Hydro)Isom.
Butene-2
MTBE / TBA
cleavage
Coal
Acetylene
Delayed Coking
Vacuum
Residue
FCC Cracking
Raff II [butene-1-rich]
Crude Oil
Gas Oil
PIB
Higher Olefin
Cracking
Naphtha
Alkylation
Methane
BD
SHU
XTBE /
TAME
Condensate
products
Oxidative DeHydrogenation
Crude C4
[Butadienerich]
Raff I [iso-butene-rich]
Butanes
conversion processes
Ethylene Dimerisation
Ethane /
Propane
Steam cracking
Natural Gas
crudes
Butadiene
Figure 2. Network of petrochemical and refinery processes for the generation and conversion of C4 hydrocarbons; Starting from raw materials a number of crudes are produced first that are used as feedstock for the primary generation of olefins. C4 olefins are further processed from their raffinates and in some cases are recovered in substance before being used as a chemical feedstock.
Table 1. Composition of raw C4 hydrocarbon streams generated in different petrochemical and refinery processes; the red boxes indicate the difference in butane to butene ratios between steam cracking and fluid catalytic cracking.
C4
Boiling
point [C]
Yield [wt %]
SC, low
severity
SC, high
severity
FCC
cracker
n-Butane
BDH
Ethanol
ETB
n-Butene
ODH
MTO
[ZSM-5]
1,3-Butadiene
4.4
28
49
0.2
913
90
58
Isobutene
6.9
32
22
24
~ 30*
cis-Butene-2
+3.7
11
3741
10
~ 60
trans-Butene-2
+0.9
Butene-1
6.3
20
14
15
Isobutane
11.7
37
n-Butane
0.5
[7]
[7]
Source
Most of the 1-butene is consumed in the production of alkylate gasoline, while only one quarter is used in the production
of chemicals and polymers. After isomerization to 2-butene
1-butene volumes are also consumed to produce propylene by
metathesis with ethylene. Similar to isobutylene volumes of
1-butene will grow mainly along with alkylate gasoline.The
supply and demand situation for 2-butene is very similar to
1-butene. 2-Butene is also produced mainly in refineries and is
converted by approx. three quarters into alkylate gasoline. The
FCC process yields a 2-butene to 1-butene ratio much closer to
www.ChemBioEngRev.de
15
1.3
12
~ 50
[8]
[9]
[10]
< 10**
[11, 12]
138
a)
a)
Isobutane
Dehydrogenation
14%
Propylene
via metathesis
24%
b)
Fluid-catalytic
cracking, total
62%
Steam cracking,
liquid feed
24%
Methyl Ethyl
Ketone,
6%
Alkylate
gasoline
70%
Isobutylene by source
30 Mmto / +4 % p.a.
2-Butene by use
14 Mmto / +4 % p.a.
b)
b)
Synthetic
Rubber
9%
Chemicals,
other
3%
Fluid-catalytic
cracking, total
86%
Steam cracking,
14%
MTBE / ETBE
41%
Alkylate
gasoline
47%
a)
Alkylate
gasoline
7%
Ethylene
dimerisation
4%
Butane
7%
XTBE
2%
Fluid-catalytic
cracking, total
73%
Steam
cracking
23%
2-Butene by source
14 Mmto / +4 % p.a.
Isobutylene by use
30 Mmto / +4 % p.a.
Bioethanol
10%
Gasoline
fractions
75%
Gasoline, total
1,000 Mmto / 1 % p.a.
Butene-1 by source
18 Mmto / +4 % p.a.
Figure 6. The global gasoline pool, volumes by component. Individual components and their physical properties relevant to
gasoline blending are summarized in Tab. 2 [22, 23].
b)
Chemicals
6%
Polybutene-1,
3%
Polyethylene
(hdPE / lldPE)
9%
Alkylate
gasoline
73%
Propylene
via metathesis
10%
Butene-1 by use
18 Mmto / +4 % p.a.
www.ChemBioEngRev.de
139
Table 2. Individual components and their physical properties relevant to gasoline blending [22, 23]
Component
Net energy
[%]
Crude gasoline
88
611
100
Ethanol
112
1822
65
Isobutanol
102
45
82
Alkylate gasoline
95
45
95
C4 raffinate
86104
5070
~ 105
MTBE
110
80
ETBE
111
83
ponent for the gasoline pool, a trend still in its infancy. Biobutanol boosts the octane number of gasoline, but unlike bioethanol, without increasing its
Reid vapor pressure. This would relieve the demand for alkylate gasoline as a low-vapor pressure
blending component and, thus, could make C4 olefins from refineries available for the petrochemical
industry.
Since all three butenes are expected to grow by
about 4 % a1, it will be interesting to see how the
corresponding volumes will be produced. With
much lower growth rates in gasoline volumes FCC
units will not be able to satisfy this growing demand
alone. Also, with ethylene plants being converted
from liquid to gas feedstock butene availability from
steam cracking will also decrease, potentially leaving
behind a supply gap in C4 olefins.
a)
b)
Alkylate gasoline, total
65 Mmto / 4 % p.a.
When turning our view to alternative processes for manufacturing C4 olefins dehydrogenation of the corresponding butane
is the first process that must be mentioned. Iso- and n-butane
can be dehydrogenated to raffinate-2 and raffinate-1-compatible product streams, respectively (Fig. 8). Butadiene-rich crude
C4 streams can be generated from n-butenes by oxidative dehydrogenation and can be further processed conventionally by
butadiene extraction. Butanes are available as feedstock in large
amounts from natural gas, from crude oil distillation, or as a
residual stream from raffinate processing in oil refineries and
petrochemical plants.
Industrial processes for dehydrogenation of light paraffins
are available from various licensors (Tab. 3). Due to thermodynamics of the dehydrogenation reaction operating temperatures are high at around 600 C and the processes are operated
at relatively low pressure. Most of the processes use a combination of platinum and tin as the active ingredient of the dehydrogenation catalyst while support materials are different for
different licensors. As an exemption the Catofin process uses
an alumina-supported chromium oxide catalyst.
Dehydrogenation processes typically operate at partial conversion of the paraffin feedstock between 50 and 60 %. Under
these conditions the processes reach C4 olefin selectivities of
about 90 %. The remainder of the converted feedstock (~10 %)
forms coke deposits on the catalyst surface. These deposits are
used in some process types like in the Catofin and in the Oleflex process to generate internal process heat by combustion
during catalyst regeneration. In contrast, the STAR process uses
steam injection into the reactant stream to minimize coke deposits by internal steam reforming. Depending on the coke
management concept cycle times between two catalyst regeneration steps vary between a few minutes and several hours per
cycle for the various process types.
www.ChemBioEngRev.de
140
UOP oleflex
Clariant Catofin
Linde-BASF process
Reactor
Fixed-bed, adiabatic
Fixed-bed, isothermal
T [C]
580650
570590
560650
550650
p [bar]
0.30.5
>1
Cycle time
510 d
7h
620 min
~9h
50
~35
6065
C3 :30
S(C4 ) [%]
n-C4 : 81;
i-C4: 91
> 9193
90
C3 :90
Catalyst
Pt-Sn on Al2O3
Cr2O3 : Al2O3
Pt-Sn support
Endothermal dehydrogenation allows only partial conversion of ethane, propane, and of butane feedstock in the technically relevant range of operating temperatures up to 800 C
(Fig. 9). Heat of reaction can be introduced into the reaction
volume in three different ways (Fig. 10): the reaction volume
can be heated externally, e.g., by gas burners. Alternatively, the
reactant gas flow can be preheated to temperatures high
enough above the kinetic onset temperature to carry heat into
the reaction volume by its heat capacity. The heat capacity of
the reactant gas stream may be further increased by steam dilution to increase the amount of heat introduced. Thirdly, heat of
reaction can be generated internally by selective combustion of
the hydrogen that is released by the dehydrogenation reaction.
Oxygen that is mixed into the reactant gas stream reacts with
the hydrogen on the dehydrogenation catalyst.
100
Butane
Propane
Conversion [%]
75
Ethane
50
25
0
400
500
600
Temperature [C]
700
Figure 9. Thermodynamically limited conversion of light paraffins in endothermal dehydrogenation. The gray area indicates
the catalytically relevant range of operating temperatures;
from: [27].
Variable economics of olefin-producing processes also depend on by-product yields. As can be seen in Tab. 4 (yellow
boxes), when steam cracking propane and butanes, about one
quarter of the feedstock is converted into light gas, mainly synthesis gas and methane. However, when dehydrogenating these
paraffins, the resulting light-gas yield is much lower. Hence, C3
www.ChemBioEngRev.de
Adiabatic
Isothermal
high Tin
Low Tin
Autothermal
Low Tin
z, t
RH = CP,gas T
RH = RHcomb.
RH = RHcomb.
preheated
reactant gas
external gas
firing
H2 combustion
in-situ
Figure 10. Heat management concepts for endothermal dehydrogenation processes; from [27].
The multitude of possibilities to produce methanol from carbon-containing raw materials via synthesis gas grants a flexible
access to light olefins from all four raw material sources natural gas, crude oil, coal, and biomass.
While ExxonMobils original methanol-to-gasoline (MTG)
process was dedicated primarily to gasoline production from
methanol, later developments like UOPs methanol-to-olefin
141
Table 4. Industrial processes for the generation of light olefins and liquid hydrocarbons: product yields in percent of feedstock and
global volumes in Mmto [16, 28].
SC, C2
SC, C3
SC, C5
SC, naphtha
SC, GOs
Fuel-FCC
80
28
19
~ 250
35
570
30
10
10
28
7.5
4.1
53
5.7
1.3
4.0
C2 [Mmto]
42
11
~ 80
~2
~6
C2 yield [%]
53
40
32
~ 32
25
~ 0.4
20
C3 [Mmto]
3.7
40
36
C3 yield [%]
1.2
13.3
20
~ 16
~ 14
~6
1520
45
C4 yield [%]
2.2
13
10
10
15
19
45
~ 22
~ 36
~ 55
~ 35
Co-crack
35
12
45
50
50
Light gas
~8
25
23
~ 16
~ 11
Liquids
Olefin-FCC n-Propane
PDH
n-Butane
Economics of Olefin-producing
Processes
Table 5. Industrial processes for the catalytic cracking of methanol to hydrocarbons (MT processes).
ExxonMobil MTG
UOP/HYDRO MTO
Lurgi MTP
Sinopec S-MTO
T [C]
400420
340540
425
350550
p [bar]
~4
13
1.5
~1
Catalyst
S(C2) [%]
1.1
37
5035
S(C3) [%]
4.5
37
64
3045
S(C4) [%]
C4 : 1.1C4 : 11.9
~ 12 (lab:2025)
~8
~ 11
S(C5 ) [%]
~7
~ 25
< 10
Source
[29]
[30]
[31]
[32]
www.ChemBioEngRev.de
142
bution margin 1 (CM1). For the present paper CM1 was calculated based on published product yields for the various
processes and on U.S. market prices for feedstock and products, respectively. All other cash expenses were neglected for
now. CM1 represents a rough estimate for the operating cashflow and, thus, compares profitability between different processes.
At current US market prices steam cracking of ethane yields
by far the highest CM1 margin, followed by propane dehydrogenation and by the two olefin-generating MT processes. CM1
margins of all other processes are significantly lower. For most
of the processes CM1 margins are driven by the product mix,
while for ethane cracking it is driven mainly by low feedstock
costs.
Regional differences, in general, play a major role in the selection of suitable processes to produce light olefins and other
basic petrochemicals (Fig. 14). Differences in feedstock prices
result from geological differences in raw material abundance.
In contrast, regional differences in market prices and product
demand result from differences in population and economic
power. Regions rich in raw material like the Middle East, the
former GUS countries, Australia, or Latin America possess a
comparatively small population and generate a low absolute
GDP. Petrochemicals that are produced in these regions are
predominantly exported, while countries poor in raw materials,
but rich in population and with a strong GDP like Europe or
Asia are net importers of these goods.
As an exception, the U.S. currently enjoy significant raw material resources and at the same time a strong demand for these
products in a large population with a strong economy. In addition, U.S. petrochemical producers increasingly export their
products globally.
1,000
800
600
400
C2 Dimer.
MTP
ETE
MTO
MTG
C4 Dehydro
Deep FCC
C3 Dehydro
FCC
GO SC
Naphtha SC
Butane SC
Ethane SC
Propane SC
200
Europe
North America
Asia Pacic
Middle East
Africa
Populaon
GDP
South America
Coal reserves
Australia
Oil reserves
Natural Gas reserves
Figure 14. Regional differences in population, GDP and raw material abundances; the global figures have been composed from individual Wikipedia entries by country for population, GDP and for the raw material reserves, respectively. Full scale of each graph represents a
60 % share of the region in the respective global figure.
www.ChemBioEngRev.de
143
Due to these regional differences it is necessary to take transport costs for raw materials and for petrochemical products into account when deciding on investments into petrochemical
process equipment and on the location of the individual investment (Tab. 6).
High specific costs are related to the transport of hydrocarbon equivalents in particular for liquefied natural gas (LNG)
and for liquefied petroleum gas (LPG). However, methanol and
biomass, such as corn, also come with relatively high transport
costs when calculated by the ton of hydrocarbon equivalent.
For LNG and LPG, costs are high mainly due to an extensive
infrastructure transport of methanol, and biomass is expensive
because they contain large amounts of water that must be
transported alongside. Due to cheap infrastructure, crude oil
and gasoline but also coal can be transported relatively cheap.
The differences in specific transport costs require petrochemical processes that are based on natural gas, ethane, and
LPG to be operated in the region of raw material abundance,
e.g., in the Middle East, but also in the United States. Ethane
cracking is currently big in these two regions. In contrast, oilor coal-based processes are often operated in the region where
petrochemical products are to be marketed, e.g., in Europe or
in Asia. Olefin production in these two regions relies predominantly on naphtha cracking and in Asia industrial MT plants
that use coal-based methanol feedstock have been installed
worldwide for the first time.
a)
nitrile rubber
4%
SBD block
copolymers
6%
ABS polymers
12%
Other
4%
SB.
elastomers
28%
polybutadiene
26%
Butadiene
10 Mmto / 4.5% p.a.
SB latex
12%
b)
Other (ODH)
4%
Steam cracking,
Ethane & LPG
17%
FCC, total
1%
Steam cracking,
Naphtha & GO
78%
Butadiene
10 Mmto / 4.5% p.a.
Table 6. Overseas shipment costs for different raw materials and petrochemical products; cost figures have been collected from [3438].
Commodity
Shipment cost
[US $ t1]
Energy content
[MMBtu t1]
Shipment costs
[US $ MMBtu]
Shipment costs
[US $ t1H2C eq.]
140
53.4
2.64
160
100 (Panama)
150200 (Cape Horn)
47.3
2.114.22
100200
1117
39.7
0.280.43
1020
U.S. gasoline
1012
41.2
0.410.49
1012
U.S. coal
1020
~ 27
0.370.74
U.S. methanol
3060
19
1.583.16
70140
U.S. ethanol
3060
25.6
1.172.34
5080
U.S. corn
3060
~ 15
2.004.00
www.ChemBioEngRev.de
144
5000
4000
Olefin prices
2010-2013
3000
2000
1000
Figure 16. Prices for butadiene (red curve) versus those for ethylene (blue), propylene (velvet), and naphtha (black) between
2010 and 2013; amplitudes in butadiene prices are much larger
than for the latter three materials [40].
ing with coal-based Reppe chemistry, butadiene could be obtained by catalytic dehydration of butanediol, originally a process that was operated in the 1940s to produce synthetic rubber. A somewhat more conventional way to butadiene would
be the dimerization of ethanol [41] after Lebedev or after Ostromislensky. Bioethanol can be produced routinely from biomass by fermentation. However, ethanol dimerization has not
been optimized for industrial yields. These processes were often
operated in war times or in isolated national economies. Hence,
their profitability is often not competitive in todays world market.
Beyond steam cracking of liquids, additional butadiene can
be generated by oxidative dehydrogenation (ODH) of n-butenes (Fig. 17).
,
O2, H2O
In the past, technical developments resulted in a number of fermentative processes for the production of industrial chemicals.
With a 10% share in the global gasoline pool, bioethanol has already arrived in the group of very large industrial chemical
products. In recent years, several start-up companies have begun to develop fermentation processes for the production of
higher industrial alcohols such as isobutanol and butanediol.
Typical yields of these processes range from 0.3 to 0.5 talcohol
per ton of sugar or corn starch [46] (Tab. 8).
Table 8. Fermentative yields of different industrial alcohols
(column 1 and 2, from: [46]) and CM1 margins of different olefins obtained from these alcohols by catalytic dehydration (column 3 and 4). It is assumed for simplification that the dehydration process yields 100% olefins. *Prices [$ t1]: 2000 (i-butylene)
and 2500 (butadiene).
+ H2O
Bio-alcohol
Yield
Olefin from
bio-alcohol
CM1 [$ t1]
Ethanol
0.45 t t1corn
Ethylene
~ 110
Isobutanol
0.35 t t1sugar
Isobutylene*
500 +500
1,4-Butanediol
0.5 t t1sugar
Butadiene*
~ 300
Catalytic dehydration of the corresponding industrial monoalcohols easily yields ethylene [47] and isobutylene [48], while
the same reaction on 1,4-butanediol does not easily produce
butadiene [49] (Fig. 18).
Table 7. Survey of four different ODH processes; additional ODH processes were developed by BP Chemicals, Mitsubishi Chemicals and
Nippon Zeon.
Phillips O/X/D
Petro/Tex Oxo/D
SK Energy ODH
BASF ODH
X (n-C4 ) [%]
7580
75
83
~ 95
S (BD) [%]
90
77.3
96
~ 95
512:1:1
24:2:1
15:1>0.75
015:1:2
24780
5501090
900
~ 450
Catalyst
Li-Sn-POx
ZnFeFeO4
Bi-Mo-Fe-oxides
Source
[42]
[43]
[44]
[45]
www.ChemBioEngRev.de
145
Conclusion
Today, a number of processes for the generation, the conversion and the separation of C4 olefins in petrochemicals and
in oil refineries are already available on
industrial scale. At present the majority
of global C4 olefin volumes are generated
and further processed in oil refineries
making blend components for the gasoline pool. Their availability for petrochemical production purposes hinges on
Figure 18. Catalytic dehydration of various alcohols to the corresponding light olefins.
their valuation relative to their alternative
production by petrochemical processes.
In future, butane dehydrogenation will yield increasing vola)
umes of C4 olefins. Smaller volumes of butadiene will be generated by oxidative dehydrogenation of butenes obtained by dehydrogenation before. MT processes may also be tuned for C4
olefin production depending on further process development.
Bio-based routes to C4 olefins are still in their infancy. Their
success will largely depend on the development of market prices for biomass raw materials and on a potential solution of the
conflict of food versus fuel.
The author has declared no conflict of interests.
b)
Figure 19. Long-term market prices for sugar (a) [50] and for
maize (b) [51].
In any case, it can be concluded that the comparatively simple dehydration of bio-based industrial alcohols can generate
positive CM1 margins and, depending on market prices, may
become as profitable as conventional petrochemical processes.
However, prices of bio-based raw materials for the fermentative production of alcohols may undergo drastic changes that
could easily undermine process profitability on the long run.
After a long period of relatively stable prices soft commodities
have recently experienced price hikes that doubled or even
tripled market prices for certain carbohydrates (Fig. 19).
Despite of these uncertainties additional efforts will be spent
in future developments of production processes for petrochemicals on the basis of biomass can be expected. Hopefully, some
of these processes will also grant access to bio-based C4 olefins.
www.ChemBioEngRev.de
References
[1]
[2]
[3]
[4]
[5]
146
www.ChemBioEngRev.de
147