Kinetics and Reaction Engineering Aspects of Syngas Production by the Heterogeneously Catalysed Reverse Water Gas Shift Reaction
Full text
1
Kine ics and Reac ion Enginee ing Aspec s o Syngas
P oduc ion by he He e ogeneously Ca alysed
Re e se Wa e Gas Shi Reac ion
1
Von de Fakul ä ü Angewand e Na u wissenscha en
de Uni e si ä Bay eu h
zu E langung de Wü de eines
Dok o -Ingenieu s (D .-Ing.)
genehmig e Disse a ion
1
o geleg e on
M.Tech. Unde Rajabhau Baji ao
aus
Undewadi (Indien)
E s gu ach e : P o . D .-Ing. And eas Jess
Zwei gu ach e : P o . D . e . na . Pe e Wasse scheid
Tag de mündlichen P ü ung 11. June 2012
Leh s uhl ü Chemische Ve ah ens echnik
Uni e si ä Bay eu h
2012
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 1
Acknowledgemen s
Fi s o all I would like o exp ess my since ely hank o P o . D . And eas Jess o gi ing
me he oppo uni y o wo k in his g oup, o making my wish become ealis ic, o his
excellen guidance and suppo , o he en husiasm in supe ision and i edless co ec ion o
my disse a ion. His ad ice and insigh in o hings h oughou my doc o al p ojec ha e been
in aluable.
My hanks go o P o . D . Pe e Wasse scheid o ag eeing co- e e ee o my hesis, and also
o P o . D . Ru h F ei ag and P o . D . Ral Moos o accep ing o be in my examina ion
commi ee.
I hank o D . Ch is oph Ke n o his suppo in he modelling, o p o iding aluable
sugges ions, and o he co ec ion o my disse a ion and also o D . Wol gang Ko h o
co ec ion o my disse a ion and help ul commen s as well as many use ul discussions.
I also hanks o D . Leonid Da se ich o his ui ul discussions, M . Jö g Ge chau o much
help o expe imen al se -up and he ope a ion o ins umen s, and o he help o sol ing
compu e p oblems, M s. Bi gi B unne o he help and suppo du ing my esea ch wo k.
I hank Sec e a y M s. Ri a Pannek o he adminis a i e assis ances.
I hank my colleagues om Chai o Chemical Enginee ing Johannes Thiessen and Lisa
Schilde o a good ime o sha ing he o ice wi h hem and o hei help, Flo ian Heym,
Amadeus Rose, Anne Piegsa, S e an F i z, S ephan Aschaue , Philipp Kaise , Pe e
F eme ey and Susanne F i schi o hei help, suppo and p o iding pleasan and iendly
esea ch a mosphe e o e he yea s.
I hank o all my iends and my coun y-ma es in Bay eu h o hei lo e, suppo s and
encou agemen .
I am o e e indeb ed o my pa en s, pa en s in law, amily membe s and iends o hei
lo e, p aye s, suppo s and encou agemen .
Finally, I hank o my lo ely wi e, Sudha, o he suppo , unde s anding, encou agemen
and endless pa ience.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 1
Pa s o his wo k we e al eady published:
1
1. A. Jess, P. Kaise , C. Ke n, R. B. Unde and C. on Olshausen, Conside a ions
conce ning he Ene gy Demand and Ene gy Mix o Global Wel a e and S able
Ecosys ems. Chemie Ingenieu Technik 83, 1777–1791 (2011).
2. R. B. Unde, C. Ke n and A. Jess, High empe a u e CO
2
hyd ogena ion o e Ni Ca alys .
8
h
Eu opean Cong ess o Chemical Enginee ing, P ocessNe , Be lin, Sep embe 25–29
(2011).
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
i
Table o Con en s
Lis o symbols .......................................................................................................................
1
1.
1
In oduc ion ................................................................................................................. 1
1
2.
1
Basic heo y and backg ound o he wo k ................................................................ 3
1
2.1 He e ogeneous ca alysis ............................................................................................... 3
1
Di usion p ocesses in he e ogeneous ca alysis ...................................................... 3
1
2.1.1
In luence o in e nal and ex e nal mass anspo on he e ogeneously ca alysed
2.1.2
eac ions ................................................................................................................. 5
1
Ca alys deac i a ion phenomena .......................................................................... 13
1
2.1.3
2.2 U iliza ion o CO
2
o p oduc ion o chemicals and uels .......................................... 16
1
2.3 Concep o p oduc ion o liquid uels om CO
2
ia e e se wa e gas shi (RWGS)
and Fische T opsch syn hesis (FTS) ........................................................................ 20
1
2.4 The modynamics o CO
2
con e sion ( e e se wa e gas shi and me hana ion) ...... 27
1
2.5 Kine ics o e e se wa e gas shi eac ion and me hana ion .................................... 32
1
3.
1
Objec i e and scope o he wo k .............................................................................. 39
1
4.
1
Expe imen al me hod and da a analysis ................................................................. 41
1
4.1 Expe imen al se up ..................................................................................................... 41
1
4.2 Expe imen al p ocedu e ............................................................................................. 43
1
4.3 Ca alys cha ac e isa ion ............................................................................................ 45
1
4.4 E alua ion o he expe imen al da a ........................................................................... 47
1
CO
2
hyd ogena ion eac ion .................................................................................. 47
1
4.4.1
4.4.1.1 Con e sion o CO
2
and yield o CO and CH
4
.............................................. 47
1
4.4.1.2 De e mina ion o he in insic kine ic pa ame e s ........................................ 48
1
4.4.1.3 In e nal mass anspo calcula ions ............................................................. 49
1
4.4.1.4 Calcula ion o ex e nal mass anspo limi a ions ...................................... 51
1
CO hyd ogena ion (me hana ion) eac ion ............................................................ 52
1
4.4.2
4.4.2.1 Con e sion o CO
and yield o CO
2
and CH
4
.............................................. 52
1
4.4.2.2 Kine ic analysis and mass anspo calcula ions ......................................... 52
1
Wa e gas shi (WGS) eac ion ............................................................................ 53
1
4.4.3
4.4.3.1 Con e sion o CO
and yield o CO
2
and CH
4
.............................................. 53
1
4.4.3.2 Kine ic analysis and mass anspo calcula ions ......................................... 53
1
5.
1
Resul s and discussion ............................................................................................... 55
1
iii
Dimensionless numbe s
Nu Nussel numbe
P P and l numbe
9A
Reynolds numbe
3B
Schmid numbe
3
C
She wood numbe
Abb e ia ions
CCC ca bon cap u e and con e sion
FTS Fische T opsch syn hesis
HTE high empe a u e elec olysis o wa e
MFC mass low con olle
NTP no mal empe a u e and p essu e (20 °C and 1 a m)
RWGS e e se wa e gas shi
oe onnes o oil equi alen , 1 oe = 42 GJ
TPR empe a u e p og ammed educ ion
TPS echnical pho osyn hesis
WGS wa e gas shi
XRD x- ay di ac ion
In oduc ion
1
1. In oduc ion
Ca bon dioxide is he mos impo an g eenhouse gas being emi ed in o he a mosphe e
om ossil uel combus ion and o he an h opogenic ac i i ies. Acco ding o he Na ional
Oceanic and A mosphe ic Adminis a ion (NOAA), he CO
2
concen a ion in he
a mosphe e was 392 ppm in Feb ua y 2011 [1] which is e y high compa ed o 379 ppm in
2005 and a om he na u al ange o he las 650,000 yea s [2]. The annual emissions o
CO
2
ha e g own be ween 1970 and 2004 by abou 80% om 21 o 39 giga onnes [2]. The
g owing concen a ion o ca bon dioxide in he a mosphe e inc eased he impac on he
en i onmen such as global wa ming and o cing a clima e change. The main g eenhouse
gases a e wa e apou (H
2
O), ca bon dioxide (CO
2
), me hane (CH
4
), ni ous oxide (N
2
O),
hyd o luo oca bons (HFCs), pe luo oca bons (PFCs), and sulphu hexa luo ide (SF
6
), bu
CO
2
con ibu es abou 77% o he wo ld’s g eenhouse gas emissions (excluding wa e
apou ) in o he a mosphe e in 2004 [2].
Al e na i e ene gy sou ces a e no only impo an wi h ega d o global wa ming bu also o
main ain he ising cos o c ude oil. Bo h aspec s ha e mo i a ed esea che s o look o
solu ions o educe he g eenhouse gas emission and/o i s u iliza ion. The mi iga ion o
g eenhouse gas emissions is also an in e es ing challenge in explo ing new concep s and
new oppo uni ies o ca alysis and indus ial chemis y. Recen ly a ious ca bon cap u e
and s o age echnologies (CCS) a e de eloped o a signi ican educ ion o CO
2
emissions
in o he a mosphe e [3]. These echnologies a e mainly sui able o cap u e CO
2
om la ge
indus ial sou ces such as ossil uel i ed powe plan s. Bu only wi h hese e o s i is no
possible o educe and con ol CO
2
emissions [4], and he a ailabili y o su icien s o age
capaci y o cap u e ca bon dioxide is also s ill an open ques ion. Howe e , e y li le
a en ion has been paid by indus y and academia in u iliza ion o CO
2
because o i s high
he modynamic s abili y. Fo he educ ion o g eenhouse gas emission a ious s a egies
ha e been sugges ed ha includes p omo ing ene gy end-use e iciency, suppo ing he use
o enewable ene gy esou ces as well as sus ainable anspo a ion and was e managemen [5].
One e ec i e app oach o a oid CO
2
accumula ion in o he a mosphe e is he eco e y o
ca bon dioxide om lue gases and i s ecycle by con e ing o use ul chemicals [6, 7]. The
In oduc ion
2
con e sion o CO
2
o CO by ca aly ic hyd ogena ion has been ecognized as a e y
p omising p ocess. In indus y, syn hesis gas con aining H
2
and CO can be used o p oduce
me hanol as well as long chain hyd oca bons ia he Fische -T opsch syn hesis. The e o e,
he e e se wa e gas shi (RWGS) eac ion (CO
2
+ H
2
5 CO + H
2
O) is an impo an
op ion o CO p oduc ion.
The ans o ma ion o CO
2
and H
2
in o CO and H
2
O depends upon se e al ac o s such as
ca alys selec ion, a io o CO
2
/H
2
, and eac ion empe a u e and p essu e. The e o e, he
main ocus o his s udy was o selec a sui able ca alys and eac ion condi ions o CO
2
hyd ogena ion in a ixed bed eac o . The expe imen s we e designed such ha he RWGS
eac ion could be examined in he o wa d and e e se di ec ion, i.e. he “no mal” wa e gas
shi eac ion (CO + H
2
O 5 CO
2
+ H
2
) was also s udied. The consecu i e CO
hyd ogena ion, i.e. me hana ion eac ion (CO + 3H
2
5 CH
4
+ H
2
O) also had o be
conside ed.
In he e ogeneous ca alysis, anspo p ocesses (ex e nal bounda y laye di usion, po e
di usion) may ha e an in luence on he e ec i e eac ion a e, and we e also s udied in his
wo k in de ail.
Basic heo y and backg ound o he wo k
3
2. Basic heo y and backg ound o he wo k
In he ollowing, some basic aspec s o he e ogeneous ca alysis and phenomena o mass
ans e in he e ogeneous ca alysis (gas-solid di usion in a ixed bed eac o ), deac i a ion
phenomena o ca alys s a e b ie ly ou lined. Some CO
2
u iliza ion p ocesses including
e e se wa e gas shi , and he he modynamics o CO
2
con e sion a e also discussed.
2.1 He e ogeneous ca alysis
The s udy o he e ogeneous ca alysis da es back o he ea ly 1800s whe e Fa aday was one
o he i s scien is s who pe o med he abili y o pla inum o acili a e oxida ion eac ions
[8]. A e ha , he ield o he e ogeneous ca alysis has been g own con inuously and
a ac ed se e al Nobel p izes ( o example, in 2007, Ge ha d E l, a Ge man physical
chemis was awa ded he Nobel P ize in chemis y o his con ibu ion in he a ea o su ace
science). The p oduc ion o low cos and high quali y aw ma e ials, p oduc ion o
anspo a ion uels, and pollu ion con ol a e he majo a eas in which he e ogeneous
ca alysis has shown a ema kable impac .
I is well known ha he ca alys e iciency is di ec ly p opo ional o i s su ace a ea. Many
o he he e ogeneous ca alys s used in oday’s indus y consis o one o mo e ca aly ically
ac i e componen s deposi ed on he su ace o a suppo ma e ial ha ing a high su ace a ea,
high po osi y, and a sui able he mal and mechanical s eng h. He e ogeneous eac ions
occu in a sys em in which wo o mo e phases a e p esen (e.g. he solid ca alys and liquids
o gases as eac an s) and he eac ions occu a he in e ace be ween hese phases.
Di usion p ocesses in he e ogeneous ca alysis
2.1.1
The chemical eac ions o he e ogeneous ca alysis occu be ween he adso bed eac an s and
he su ace o he solid ca alys pa icle. The di usion o eac an s o he ca alys su ace
(mos ly o he inne su ace o po es) and he di usion o p oduc s om his su ace a e
pu ely physical phenomena. In case o ca aly ic gas-solid eac ions, he a e o eac ion
wi hin a po ous ca alys s ongly depends upon he accessibili y o eac an s o he ac i e
si es dispe sed h oughou he po ous s uc u e o he ca alys (Fig. 2-1).
Basic heo y and backg ound o he wo k
4
Fig. 2-1: Sequen ial s eps in ol ed in he e ogeneous ca aly ic gas phase eac ion.
Conside a simple gaseous eac ion (A
→
B) occu ing inside a eac o con aining po ous
ca alys pa icles. In o de o con e he eac an A in o p oduc B, he ollowing physical
and chemical p ocesses a e impo an :
1. Di usion o eac an A om he bulk gas phase h ough he bounda y laye o s agnan
gas ilm su ounding he ca alys pa icle o he ex e nal su ace o he ca alys ( ilm
di usion).
2. Di usion o eac an A in o he po ous s uc u e o he ca alys pa icle o he poin
whe e adso p ion/ eac ion akes place (po e o in apa icle di usion).
3. Adso p ion o eac an s on he inne su ace o he ca alys .
4. Su ace eac ion o adso bed eac an A o adso bed p oduc B a he ca alys su ace.
5. Deso p ion o p oduc B om he inne su ace o he ca alys in o he po es.
6. Di usion o o med p oduc s h ough he po ous ne wo k s uc u e o he ex e nal
su ace o he ca alys (in apa icle di usion).
Basic heo y and backg ound o he wo k
5
7. Di usion o p oduc B om he ex e nal su ace o he ca alys h ough he bounda y
laye in o he bulk gas phase.
As a esul o hese s eps, a concen a ion p o ile may exis in he ca alys pelle and ou side
in he ilm laye . The s eps 3, 4, and 5 a e chemical p ocesses, which s ongly depend on
empe a u e ypically wi h e ec i e ac i a ion ene gies o 20 o 200 kJ/mol. The s eps 1, 2,
6, and 7 a e di usional p ocesses o mass ans e wi h ela i e low empe a u e dependence
compa ed o he chemical p ocesses. I hese s eps a e e y as , hen he e is no esis ance
o mass ans e om he bulk phase o he ex e nal su ace a ea o he pa icles and o he
ac i e si es inside he po e. So he concen a ion a he in e nal ca alys su ace is he same
as ha o he bulk phase, and he mass ans e does no a ec he eac ion a e. Bu in case
ha mass ans e is slow ela i e o he chemical eac ions, he e may be an in luence o
con ol o he o e all eac ion a e by he mass ans e . The ne kine ics ob ained om all
eac ion s eps (1 o 7) is hus called e ec i e kine ics o mac o-kine ics.
In luence o in e nal and ex e nal mass anspo on he e ogeneously ca alysed
2.1.2
eac ions
In he e ogeneous ca aly ic eac ions, he conside a ion o in e nal and ex e nal mass ans e
limi a ions is e y impo an when expe imen s a e pe o med o de e mine in insic kine ic
pa ame e s. Acco ding o A henius law, he in insic chemical a e is nea ly an exponen ial
unc ion o empe a u e and he mass ans e a e is less s ongly in luenced by he
empe a u e, e.g. o gas di usion he a e is p opo ional o abou T
1.5
. Fig. 2-2 shows an
A henius plo o he empe a u e dependence o he e ec i e a e cons an o a ca aly ic
eac ion o he h ee eac ion a e con olling egimes.
Regime o in insic kine ic:
A low empe a u es, he a e o he chemical eac ion is slow compa ed o mass anspo
p ocesses. The e o e only a negligible concen a ion g adien will be es ablished in he
ex e io and in e io o he ca alys pa icle. The eac ion empe a u e also emains cons an
o e he en i e c oss-sec ion o he ca alys pelle and also in he bounda y laye . This is
called as an in insic egion whe e he ue in insic kine ics o eac ions is measu ed. Fo
his egion, he slope o he cu e ln k
m
s 1/T is p opo ional o he chemical ac i a ion
Basic heo y and backg ound o he wo k
6
ene gy (E
A
). The eac ion a e o gas-solid eac ions is o en desc ibed wi h accu acy by a
powe law equa ion:
D
E
F
4
5
F
E
1
(2.1)
The eac ion a e usually depends upon he a e cons an (k
m
) ela ed o he mass o ca alys
(m
ca
), concen a ion o eac an (
), and he in insic o de o eac ion (n). The empe a u e
dependence o he eac ion a e cons an is gi en by he A henius equa ion:
E
F
E
A
!"
1
(2.2)
In he A henius equa ion, k
m,0
is he equency ac o and E
A
is he in insic ac i a ion
ene gy o he chemical eac ion.
Regime o po e di usion:
A in e media e empe a u es, he chemical eac ion a e is as e han he inne mass
anspo (po e di usion). Be ween his egime and he egime o in insic kine ic, a
ansi ion egime is loca ed whe e he slope o he cu e changes wi h empe a u e. In he
po e di usion egion, he concen a ion o eac an s a he po e mou h is much highe han
ha inside he po e and d ops dis inc ly. He e he en i e ca aly ic su ace is no accessible o
he same concen a ion. The e o e he e ec i e eac ion a e will be less as compa ed o he
a e wi hou mass ans e limi a ions.
Fo his egion, he e ec i e ac i a ion ene gy is oughly one hal o he ue ac i a ion
ene gy (E
A
/2). In gene al, he e is no only he change in e ec i e ac i a ion ene gy bu he
e ec i e o de o he eac ion also changes when he ansi ion om kine ic o di usion
con ol occu s. Fo an n
h
o de eac ion, he e ec i e eac ion o de app oaches he alue o
(n + 1)/2 in case o s ong limi a ion by po e di usion.
Basic heo y and backg ound o he wo k
7
Fig. 2-2: Typical A henius plo o empe a u e dependence o he e ec i e a e cons an o
ca aly ic eac ion; h ee egimes o eac ion a e con ol o low, in e media e and
high eac ion empe a u e.
Fo a i s o de eac ion he e ec i e a e o chemical eac ion is gi en as
D
E
#$$
F
%
E
&
1
(2.3)
whe e
&
' he concen a ion a he ca alys su ace and 6 is he e ec i eness ac o which
is a unc ion o he Thiele modulus. The e ec i eness ac o 2 is de ined as he a io o he
obse ed a e o eac ion o he a e in he absence o any di usional esis ance, i.e.
%
F
(B)*(+
DA(B),-4
D()A
,4)D,4.,B
D()A
-/
B
C
A,B(+
DA(B),-4
F
D
E
#$$
D
E
1
(2.4)
The e a e se e al ac o s which in luences he e ec i eness ac o , such as po e shape and
po e s uc u e (mic o-mac o), pa icle size dis ibu ion, and change in olume upon eac ion
[9]. The magni ude o he e ec i eness ac o anges be ween 0 and 1 which indica e he
ela i e impo ance o di usion and eac ion limi a ions [10]. The in e nal e ec i eness
ac o a ies o di e en ca alys geome ies (see Fig. 2-3) and o di e en eac ion o de s
(see Fig. 2-4).
Basic heo y and backg ound o he wo k
8
Fig. 2-3: Rela ionship o he e ec i eness ac o e sus gene alized Thiele modulus
(Eq. (2.7)) o di e en shapes o ca alys .
Fig. 2-4: Rela ionship o he e ec i eness ac o e sus gene alized Thiele modulus
(Eq. (2.8)) o simple-o de eac ions [11].
The e ec i eness ac o as a unc ion o he Thiele modulus o a la pla e and an
iso he mal, i e e sible i s -o de eac ion is gi en as:
2324
234
4
2324 234 4 42 422
E ec i eness ac o [1]
Thiele modulus [2]
567819678A
B9CADA
EF6AD
Basic heo y and backg ound o he wo k
9
%
F
)(4
C
8
8
1
(2.5)
Fo he sphe ical pa icle, an e ec i eness ac o is gi en as:
%
F
0
8
1
0
)(4
C
2
3
8
4
5
0
2
3
8
4
6
1
(2.6)
The ela ion be ween Thiele modulus and e ec i eness ac o o a la pla e in Eq. (2.5)
gi es he app oxima ion ha 2 equals 1/8 o la ge alues o he Thiele modulus (8 8 2).
The equa ion o an e ec i eness ac o o a la pla e can be used in a good app oxima ion
o any pa icle geome y wi h cha ac e is ic leng h (L
p
), which is gi en by Eq. (2.9).
The dimensionless Thiele modulus (8) plays a key ole in de e mining po e di usion
limi a ions. Fo gene al shape, i s o de and i e e sible eac ion, he Thiele modulus is
gi en as [9]:
8
F
7
8
9
E
:
8
;
#$$
1
(2.7)
Fo a i e e sible eac ion and o de n 7 1, he Thiele modulus is gi en as:
8
F
7
8
9
1
4
<
0
=
6
E
:
8
>
?
@
;
#$$
1
(2.8)
whe e 6
p
is he densi y o pa icle, D
i,e
is he e ec i e di usi i y o species i, and L
p
is he
cha ac e is ic leng h o a ious shapes which is gi en as:
7
8
F
A-+*A
-/
.
C
(BA
AC)AD4(+
(DA(
-/
.
C
(BA
1
(2.9)
The cha ac e is ic leng h o cylinde s is L
p
= d
p
/4, o sphe es L
p
= d
p
/6 and o la pla e L
p
(whe e d
p
is he pa icle diame e and 2L
p
is he pla e hickness).
The poin a which a pelle o op imal dimension shows he ansi ion om chemical o
di usion con ol egion is p ac ically impo an . Small pa icles ha e a low cha ac e is ic
leng h, which dec eases he alue o he Thiele modulus (see Eq. (2.7) and Eq. (2.8)) and
hus inc ease he e ec i eness ac o , and lowe s he po e di usion esis ance. On he o he
hand, a small pa icle size c ea es a high p essu e d op in a ixed bed eac o . The e o e om
Basic heo y and backg ound o he wo k
16
Alloying:
In alloying, a combina ion o wo o mo e me als akes place a high empe a u e
(pa icula ly unde educ ion condi ions). Alloying may change he ac i i y and s abili y o
ca alys . The Cu-Zn alloy o ma ion du ing he educ ion o he ca alys in he empe a u e
ange o 230 o 500 °C wi h a mix u e o H
2
and N
2
deac i a es he ca alys and educes i s
ac i i y o he wa e gas shi [27]. O he example a e o ma ion o RhAl
2
O
4
in P -Rh/Al
2
O
3
ca alys s in ca aly ic con e e s o educ ion o ca engine emissions, o ma ion o Ni
2
Al
2
O
4
du ing s eam e o ming o e Ni/Al
2
O
3
, and o ma ion o KAlO
2
in FTS.
2.2 U iliza ion o CO
2
o p oduc ion o chemicals and uels
The ene gy ela ed CO
2
emissions ha e inc eased wi h an a e age annual g ow h a e o
1.9% om 1990 o 2007 and he emissions a e expec ed o ise wi h an annual a e o 1.3%
un il 2035 [8]. The global CO
2
emissions om ossil uels combus ion inc ease om nea by
ze o in 1870 o 29.7 billion me ic ons in 2007 and a e p ojec ed o inc ease up o 42 billion
me ic ons in 2035 [8]. The analysis o wo ld ene gy- ela ed ca bon dioxide emissions by
he consump ion o ossil uels (Fig. 2-6) shows ha coal is he la ges sou ce o ca bon
dioxide (12.5 billion me ic ons), ollowed by liquid uels (11.3 billion me ic ons), and
na u al gas (5.9 billion me ic ons).
Fig. 2-6: The wo ld ene gy- ela ed ca bon dioxide emissions by uel ype [8].
Basic heo y and backg ound o he wo k
17
To a oid CO
2
emissions, a ious measu es such as eco e y, emo al, and s o age disposal
ha e been p oposed. Ca bon cap u e and s o age needs la ge amoun s o ene gy o i s
cap u e, anspo a ion, and seques a ion. The e o e he u iliza ion o CO
2
in chemical
con e sion p ocesses may become an impo an op ion o sus ainable de elopmen ,
mi iga ion o ca bon emissions, and o a oid global wa ming. The a ious di ec and indi ec
uses o CO
2
a e shown in Fig. 2-7. The po en ial o CO
2
in he di ec use is e y low
compa ed o indi ec use. The indi ec u iliza ion o CO
2
has ad an ages such as p oduc ion
o alue added chemicals, en i onmen iendly p ocessing, and non-haza ds p ocess
u iliza ion o CO
2
.
Fig. 2-7: Va ious di ec and indi ec pa hways o CO
2
u iliza ion (modi ied a e [5]).
Un il oday, CO
2
is no used in i s ulles po en ial e en in he indi ec way o u iliza ion
because o i s high he modynamic and kine ic s abili y. The use o e icien ca alys s and
selec i e eac ion pa hways a e needed o p omo e he eac ion a e. Undoub edly, he
chemical indus y can only make li le di ec con ibu ion owa ds he educ ion o o e all
CO
2
emission. Acco ding o cu en es ima es, he chemical indus y could con ibu e o
con e a ound 1% o global CO
2
emissions in o chemical p oduc s [3]. Table 2-2 shows he
u iliza ion o CO
2
in a ious chemical con e sion p ocesses in 2006.
Basic heo y and backg ound o he wo k
18
The ixa ion o CO
2
in o hese o ganic compounds e e s o eac ions ha use he en i e
molecule. The de ail desc ip ion o he syn hesis o hese o ganic compounds using CO
2
is
gi en in appendix A. All hese p ocesses use CO
2
o p oduce alue added p oduc s in an
en i onmen ally iendly way whe e u iliza ion po en ial assis ed in e ms o less ene gy use
and lowe was e p oduc ion.
Table 2-2: Indus ial p ocess u iliza ion o CO
2
as a aw ma e ial o syn hesis o o ganic
compounds [28].
Indus ial p ocesses ha
u ilize CO
2
as aw ma e ial
Wo ld capaci y pe yea
[million onnes]
Amoun o ixed CO
2
[million onnes]
Chemical syn hesis:
Salicylic acid 0.07 0.025
U ea 143 105
Cyclic ca bona es 0.080 0.04
Poly (p opylene ca bona e) 0.070 0.03
Fuel syn hesis:
Me hanol 20 2
Syn he ic na u al gas - -
O he uels - -
Table 2-2 clea ly indica es ha he amoun o CO
2
u ilized oday o he p oduc ion o
o ganic chemicals and uels (me hanol) is e y small (a ound 100 million onnes) compa ed
o oday’s global CO
2
emissions o a ound 30 billion onnes.
I should also be no ed ha only a ound 10% o he global c ude oil consump ion is used
oday in he chemical indus y. The majo ly is used as liquid uels such as gasoline, diesel
and hea y oil. Hence, an e ec i e use o CO
2
wi h ega d o a no iceable educ ion o he
global ne emissions o CO
2
can only be eached, i he CO
2
(e.g. sepa a ed om lue gases
o in he e y a u u e om ai ) is u ilized o uels, e.g. by e e se wa e gas shi and
subsequen Fische T opsch syn hesis.
Basic heo y and backg ound o he wo k
19
The e e se wa e gas shi eac ion echnology is a simple and e ec i e way o u ilize
ca bon dioxide in many indus ies [29, 30]. This eac ion occu s a high empe a u es whe e
CO
2
is con e ed wi h H
2
in o CO and wa e .
i
P
<
j
P
k
i
<
j
P
i
1
!
j
l
Pmn
o
F
p0
[
=
q
O
-+
1
(2.21)
Se e al mos ly unwan ed pa allel and side eac ions may also ake place:
The Saba ie eac ion:
i
P
<
p
j
P
k
j
<
=
j
P
i
1
!
j
l
Pmn
o
F
5
0Y`
q
O
-+
1
(2.22)
The Bosch eac ion:
i
P
<
=
j
P
k
<
=
j
P
i
1
!
j
l
Pmn
o
F
5
Z
[
=
q
O
-+
1
(2.23)
Boudoua d eac ion:
=s
k
<
i
P
1
!
j
l
Pmn
o
F
5
0 =
[
Y
q
O
-+
1
(2.24)
Me hana ion eac ion:
s
<
3
j
P
k
j
<
j
P
i
1
!
j
l
Pmn
o
F
5
=ZY
q
O
-+
1
(2.25)
The e e se wa e gas shi eac ion has been known o chemis y since he mid 1800's bu
no expe imen al wo k was done o e eal i s iabili y [31]. Since he las wo decades,
s udies ha e been ocused on ca aly ic con e sion o CO
2
o indus ially impo an
chemicals such as ligh ole ins and liquid hyd oca bons. Depending on he eac ion ou e
and ca alys used, i is di ided in o wo g oups. One is he hyd ogena ion o CO
2
o
hyd oca bons ia me hanol syn hesis [32, 33] which combines wo eac ion s eps: me hanol
syn hesis om CO
2
and subsequen con e sion o me hanol o gasoline by he MTG
p ocess:
i
P
<
3
j
P
k
j
u
ij
<
j
P
i
1
!
j
l
Pmn
o
F
5
p
[
`
q
O
-+
1
(2.26)
=
j
u
ij
k
j
u
i
j
u
<
j
P
i
(D-(),B.
w
[
(2.27)
Al e na i ely, me hanol can also be used di ec ly a liquid uel, bu he in as uc u e is no
ye es ablished. The second ou e o con e CO
2
in o liquid hyd oca bons is Fische
T opsch syn hesis [34-36], which also combines wo s eps: hyd ogena ion o ca bon dioxide
Basic heo y and backg ound o he wo k
20
by e e se wa e gas shi eac ion (Eq. (2.21)) and, hen, u he hyd ogena ion o CO o
hyd oca bons:
i
<
=
j
P
2
5
j
P
5
4
<
j
P
i
1
!
j
l
Pmn
o
F
5
0`=
q
O
-+
(2.28)
The e m (-CH
2
-) ep esen s a me hylene g oup o a pa a in. This ou e is discussed in some
de ails in he subsequen chap e .
2.3 Concep o p oduc ion o liquid uels om CO
2
ia e e se
wa e gas shi (RWGS) and Fische T opsch syn hesis (FTS)
The concep o p oduc ion o liquid uels om CO
2
ia RWGS and FTS was ecen ly
desc ibed and discussed [37], whe e some mo e de ails can be ound. He e only some main
aspec s should be ou lined.
Elec ici y p oduced by sola ene gy (o o he enewables) can be used o p oduce liquid
uels like diesel oil wi h no o li le ne CO
2
p oduc ion by he ollowing s eps [38-42]:
a) Sepa a ion o CO
2
om lue gases (o in he long un e en om he a mosphe e).
b) H
2
p oduc ion by high empe a u e wa e (s eam) elec olysis and non- ossil elec ici y:
3
j
P
s
3
j
P
<
0
[
`
i
P
1
!
j
l
Pmn
o
F
3=
q
O
-+
1
(2.29)
c) CO p oduc ion by e e se wa e gas shi (Eq. (2.21)).
d) Fische T opsch syn hesis o hyd oca bons (p e e ably o diesel oil) (Eq. (2.28)).
In summa y we ge :
i
P
<
j
P
s
2
5
j
P
5
4
<
0
[
`
i
P
1
!
j
l
Pmn
o
F
p
q
O
-+
1
(2.30)
I we compa e Eq. (2.30) wi h na u al pho osyn hesis
i
P
<
j
P
s
0
Y
x
j
@P
i
x
<
i
P
1
(2.31)
we may ega d his p ocess as a echnical pho osyn hesis (TPS, Fig. 2-8). This ou e is also
called ca bon cap u e and con e sion (CCC) [42]. Compa ed o he p oduc s o na u al
pho osyn hesis (e.g. wood), hose o he echnical pho osyn hesis ha e a much highe ene gy
densi y ela ed o mass ( ac o 3) and e en mo e ela ed o olume ( ac o 6).
Basic heo y and backg ound o he wo k
21
Fig. 2-8: The simpli ied lowshee o a plan o p oduc ion o liquid uels om CO
2
by
Fische T opsch syn hesis and sola ene gy.
The echnical pho osyn hesis may also be help ul o ene gy s o age and anspo . Liquid
uels ha e excellen s o age, loading and anspo capabili ies, and app op ia e la ge-scale
s o age echnologies will be needed in u u e o he e icien use o enewable ene gies:
(1) Many enewable ene gy sou ces (mos no ably sola and wind) a e subjec o na u al
luc ua ions. These ha e o be compensa ed o by s o ing excess supply peaks, which
inc eases he e iciency and economic alue o enewable ene gy, and keeps
ins an aneous elec ical gene a ion and consump ion in a be e balance.
(2)
The economically mos p ac ical me hod o de eloping al e na i e ene gy sou ces is o
make use o he ea h's sunbel and high-wind zones, bu hese egions a e mos ly a
away om consume s. Hence, he e is a need o a sui able ca ie o ene gy in o de o
anspo he ene gy om he sou ce o sui able ma ke s.
One could also hink o using he hyd ogen gene a ed by elec olysis di ec ly o s o age and
anspo o sola -elec ical ene gy. Bu he ene gy densi y o liquid hyd ogen is only 0.2
oe/m
3
(and e en only 0.1 oe/m
3
o comp essed H
2
a 700 ba and 25 °C) compa ed o
diesel oil wi h 0.8 oe/m
3
( oe: onnes o oil equi alen , 1 oe = 42 GJ). The lique ac ion and
Basic heo y and backg ound o he wo k
22
anspo o H
2
is no easy: High sa e y equi emen s and a new in as uc u e a e needed,
e.g. 20% o 30% o he ene gy con en is consumed o lique ac ion [43] and 8% o
anspo as comp essed gas ia pipelines (pe 1000 km) [44].
Fo a p elimina y basic layou o a TPS plan o he p oduc ion o liquid uels om CO
2
by
sola ene gy and Fische T opsch syn hesis (FTS) he ollowing assump ions we e made:
Hyd ogen p oduc ion
H
2
is gene a ed by high empe a u e solid oxide elec olysis (HTE) a 830 °C. Acco ding o
S oo s e al. 124 MJ elec ical ene gy a e consumed pe kg H
2
(= 13.8 MJ pe kg con e ed
H
2
O) [45]. I we use he ene gy eleased by he exo he mic Fische T opsch syn hesis o he
p oduc ion o sa u a ed s eam (22 ba , 215 °C) (see below), he ene gy needed o o e hea
he s eam (215 5 830 °C) is 1.4 MJ/kg. Hence in o al, 137 MJ elec ical ene gy is needed
pe kg H
2
(= 15.2 MJ pe kg con e ed H
2
O).
Desalina ion o seawa e
I eshwa e is no a ailable, seawa e has o be desalina ed o p oduce he eed o he
elec olysis. 7 kJ elec ical ene gy is su icien pe kg wa e , i e e se osmosis is used [46].
This equi emen is only 0.04 % o he ene gy o he subsequen elec olysis. Fo
compa ison: 170 kJ/kg is needed, i mul i-s age lash dis illa ion is used [44], which is s ill
only 0.9% compa ed o wha he elec olysis consumes.
CO
2
p oduc ion
CO
2
has o be sepa a ed om lue gases (o on he long un e en om ai ). The inescapable
ene gy equi emen (in J/mol CO
2
) o concen a ing CO
2
om a gas mix u e is gi en by he
laws o he modynamics, as we ha e a leas o o e come he di e ence o he en opy o he
mix u e and o he pu e compounds (see Fig. 2-9):
y
E
F
L
z
3
F
5
{
|
}
~
9
L
+4
U
{
|
}
~
W
5
U
0
5
{
|
}
~
W
9
L
+4
U
0
5
{
|
}
~
W
{
|
}
~
1
(2.32)
Basic heo y and backg ound o he wo k
23
Fig. 2-9: Inescapable ene gy equi emen o concen a ing CO
2
om lue gases
(11 ol.-% CO
2
) and ai (300 ppm CO
2
) a 25 °C (Eq. (2.32)).
Fo sepa a ion o CO
2
om lue gases wi h ypically 11 ol-% CO
2
, he minimum ene gy
equi emen is 7.8 kJ/mol (177 kJ/kg) and o ai 22.4 kJ/mol CO
2
(509 kJ/kg). In he ideal
case, each mol o CO
2
is inally con e ed by FTS in o one mol o CH
2
-g oups (hyd oca -
bons) wi h a hea ing alue o 595 kJ/mol = 42500 kJ/kg. Hence, he ene gy equi emen o
CO
2
sepa a ion is 3.8% (ai ) and 1.3% ( lue gases) o he ene gy con en o he liquid uels
p oduced by FTS. O cou se in eali y, much mo e ene gy is consumed a he cu en s a us
o echnology. He e we use a alue o 1.2 MJ/kg CO
2
as gi en by Gö liche and P uschek
[47] o CO
2
sepa a ion om lue gases by chemical abso p ion. This alue is 7 imes highe
han he inescapable ene gy equi emen . Fo CO
2
sepa a ion om ai , a alue o 3.6 MJ/kg
CO
2
(= 7 E
min
) may be aken as a i s es ima ion. In he ideal case o a echnical pho osyn-
hesis, 3 mol wa e a e con e ed by elec olysis pe mol o CO
2
(Eq. (2.21) and (2.29)), i.e.
1.23 kg H
2
O/kg CO
2
. Hence, o he con e sion o 1 kg o CO
2
24.5 MJ a e needed o
elec olysis o wa e compa ed o 1.2 MJ o CO
2
sepa a ion om lue gases and 3.6 MJ
om ai .
Re e se wa e gas shi
H
2
and CO
2
a e con e ed o H
2
O and CO by RWGS a a empe a u e o abou 800 °C. A
he eac o ou le , he he modynamic equilib ium is eached as p o en by espec i e
expe imen s wi h a Ni-ca alys (see sec ion 5.1). The eac o is hea ed (e.g. elec ically). I
Basic heo y and backg ound o he wo k
24
hea losses a e neglec ed, he equi ed ene gy is 41 kJ/mol CO
2
(Eq. (2.21)) o 932 kJ/kg
CO
2
.
Fische T opsch syn hesis
P oduc ion o syn he ic uels ia FTS has he po en ial o p oduce uels like gasoline and
diesel oil as well as pe ochemicals om ossil and enewable sou ces. In ecen yea s, he
a ailabili y o cheap na u al gas, coal, and biomass has gi en momen um o FT echnology.
The wo ldwide FT plan capaci ies will inc ease in u u e, oday wi h na u al gas a ou ed as
eeds ock. In 2015, he wo ldwide annual p oduc ion o liquid uels ia FT will be a ound 30
million onnes, mainly p oduced in coun ies like Sou h A ica, Malaysia, and Qa a .
Beside o he main eac ion o he FT syn hesis (Eq. (2.28)), me hana ion (Eq. (2.25)) is
o en conside ed as a sepa a e eac ion. The hi d eac ion ha plays an impo an ole (a
leas i i on based ca alys s a e used) p oduces unwan ed CO
2
by he WGS eac ion:
i
<
j
P
i
k
i
P
<
j
P
1
!
j
l
Pmn
o
F
5
p0
[
=
q
O
-+
1
(2.33)
Two eac o ypes a e cu en ly a ou ed o FTS, he mul i- ubula ixed bed and he slu y
bubble column. FT eac o s a e usually cooled by boiling wa e . Sa u a ed s eam is p oduced
and may be used a e u he hea ing as eed o he HTE. A ypical FT cooling empe a u e
is 215 °C [48]. The equi ed he mal ene gy is hen 2.7 MJ pe kg sa u a ed s eam = 48.6
kJ/mol. Hence, 3.1 mol o s eam can be gene a ed pe mol CO con e ed by FTS (1jl
Pmno
= - 152 kJ/mol CO), which is he amoun needed o elec olysis.
Typical selec i i ies (in C-%) o he FTS wi h Fe-ca alys s (30% CO con e sion pe pass,
i.e. ecycle o uncon e ed syngas has o be ins alled) a e: 18% CO
2
, 5% me hane, 11% C
2
o C
4
, 10% gasoline, 17% diesel and 39% waxes [48-50]. The waxes a e u he con e ed
(mainly) o diesel oil by mild hyd oc acking a 70 ba and 350 °C [51].
All ligh hyd oca bons (me hane o C
4
) and also he CO
2
o med by he wa e gas shi
eac ion in he FTS (Eq. (2.33)) may be ecycled o he RWGS uni o be inally con e ed
by s eam e o ming ( e e se o Eq. (2.25)) o by he RWGS (Eq. (2.21)) o CO and H
2
,
espec i ely. In summa y we ge he ollowing mass balance: Fo each mol o esh CO
2
, we
ha e o ecycle app oxima ely 0.18 mol CO
2
and 0.18 mol ca bon (as C
1
o C
4
) in o he
RWGS eac o . Hence, abou 2 imes mo e ene gy is needed o un he RWGS eac o
Basic heo y and backg ound o he wo k
25
(1.18 x 41 kJ/mol CO
2
and 0.16 x 206 kJ/mol C as C
1
o C
4
compa ed o 41 kJ/mol wi hou
ecycle).
Based on he abo e lis ed assump ions and ac s, an es ima ion o he ene gy equi emen s
o he p oduc ion o liquid uels om CO
2
by FTS and sola ene gy is possible (Table 2-3).
The ene gy equi ed o he con e sion o CO
2
o CO is calcula ed by including he
con e sion o ecycled CO
2
and C
1
o C
4
hyd oca bons (coun ed as CH
4
), bo h unwan ed by-
p oduc s o FTS, and RWGS and s eam e o ming ( e e se o Eq. (2.25)), espec i ely.
Table 2-3: Es ima ion o ene gy equi emen s o he p oduc ion o liquid uels om CO
2
by FTS and sola ene gy (con e sion o 1 kg CO
2
o 0.32 kg liquid uels).
P ocess/ene gy ou pu Ene gy [kJ] Commen /assump ion
Sepa a ion o 1 kg CO
2
om lue gas (ai ) 1200 (3600) Value om [47],which is 7 imes he
he modynamic minimum (Fig. 2-9)
Con e sion o CO
2
o CO
(RWGS) 2144 Only coun ing en halpy o eac ion
Desalina ion o 1.23 kg
seawa e 9 Re e se osmosis plan
HT-elec olysis o 1.23 kg
wa e 18700
13.8 MJ/kg H
2
O elec ical ene gy o
elec olysis [45] and 1.4 MJ/kg o
o e hea sa u a ed s eam gene a ed in
he FTS (215 5 830 °C)
FT syn hesis - Hea o eac ion is used o gene a e
sa u a ed s eam (215 °C, 22 ba )
To al equi ed ene gy 21773 (24173)
P oduc ion o 0.32 kg liquid
uels 13600 Hea ing alue o uel = 42500 kJ/kg
P ocess e iciency
62%
(= 13600/21773)
CO
2
om lue gases
56%
(= 13600/24173)
CO
2
om ai
Basic heo y and backg ound o he wo k
32
he p oduc ion o uels ia Fische T opsch syn hesis. Exempla ily, he equilib ium
composi ion o syngas is shown in Fig. 2-16. Fo empe a u e abo e 750 °C, a mosphe ic
p essu e, and H
2
/CO
2
inle a io o 2, he CO
2
hyd ogena ion can gi e he equi ed a io o
syngas o he p oduc ion o uel. Fo example, a 800 °C and a mosphe ic p essu e, he
equilib ium composi ion a he eac o ou le is 57% CO, 28% H
2
, and 15 ol.-% CO
2
.
Hence, he CO/H
2
a io is abou 2.
Fig. 2-16: Equilib ium composi ion o syngas a e emo al o wa e o e empe a u e in
CO
2
hyd ogena ion eac ion wi h me hana ion (p = 1 a m, H
2
/CO
2
= 2).
2.5 Kine ics o e e se wa e gas shi eac ion and me hana ion
Two gene al mechanisms ha e been p oposed o me hana ion o CO
2
o e Ni con aining
ca alys s
[33, 57-59]. In he i s mechanism o me hane o ma ion, ini ially CO
2
unde goes
apid dissocia e adso p ion wi h he o ma ion o CO and a omic oxygen on he ca alys
su ace and, hen, u he s eps coincide wi h CO hyd ogena ion s eps, whe e u he
dissocia ion o CO occu s in o in e media e ca bon and oxygen. The in e media e ca bon
species is hen hyd ogena ed in a chain mechanism o o m me hane. In his case, CO
2
hyd ogena ion p inciple would ha e o become simila as ha o CO hyd ogena ion. The
mechanism p oposed o bo h CO
2
and CO me hana ion [58] a e shown in Table 2-4.
Basic heo y and backg ound o he wo k
33
Table 2-4: Mechanism o me hane o ma ion om CO
2
and CO [58].
E
1A8C7781A C7!"1
s
P
k
s
<
s
1#4$ 1
P
k
=
1#$ 1
s
k
s
1#$ 1
s
k
<
s
1#$ 1
<
k
1#$ 1
<
k
P
1#$ 1
P
<
k
u
1#$ 1
u
<
k
1#$ 1
k
1#$ 1
i
<
j
k
ij
1#42$
ij
<
j
k
j
P
i
1#44$
j
P
i
k
j
P
i
1#4$
The second mechanism in ol es he pa hway ha does no equi e ini ial ans o ma ion o
CO
2
o CO. Ins ead, he hyd ogena ion o CO
2
o me hane occu s di ec ly h ough he
o ma ion o in e media es (Table 2-5). This explana ion was gi en on he basis o he
di e ence in speci ic ac i i y and selec i i y wi h espec o me hane o ma ion in CO and
CO
2
hyd ogena ion eac ion [59]. Acco ding o [59], he s eps 13 and 14 a e as and s ep 15
is slow, and he subsequen s eps 16 o 23 a e as and i e e sible.
Basic heo y and backg ound o he wo k
34
Table 2-5: Mechanism o me hane o ma ion wi hou CO o ma ion [59].
E
1A8C7781A C7!"1
s
P
k
s
P
1
2034
P
k
=
1
20p4
s
P
<
k
ss
1
20`4
ss
<
k
s
<
s
1
20Y4
s
<
k
<
s
1
20 4
<
k
P
1
20K4
P
<
k
u
1
20 4
u
<
k
1
2=Z4
k
1
2=04
ij
<
j
k
j
P
i
1
2==4
j
P
i
k
j
P
i
1
2=34
The di ec hyd ogena ion o CO
2
o hyd oca bons (Eq. (2.39)) using lase gene a ed i on
ca bide ca alys s has been ecen ly in es iga ed by Fia o e al. [60]. This eac ion is
p oposed o p oceed ia dissocia i e adso p ion o ca bon dioxide in o CO and oxygen
ollowed by u he hyd ogena ion o he adso bed species on he ca alys su ace ia chain
eac ion mechanism.
i
P
<
3
j
P
2
5
j
P
5
4
<
=
j
P
i
1
j
l
Pmn
o
F
5
00Z
[
K
q
O
-+
(2.39)
The ans o ma ion o CO
2
and H
2
depends upon many ac o s. Se e al a emp s ha e been
made by se e al in es iga o s o o e come he limi a ions o RWGS eac ion. Se e al noble
me als (e.g. P , Ru, Pd) and ansi ion me als (e.g. Cu, Ni, Fe, Co) suppo ed on di e en
me al oxides (e.g. 8-Al
2
O
3
, CeO
2
, SiO
2
) ha e been explo ed o ca y ou RWGS eac ion
[30]. The a ious ca alys s in es iga ed as:
Basic heo y and backg ound o he wo k
35
Coppe ca alys :
A ca alys wi h 12 w .-% Cu loaded on alumina has shown a good ac i i y and excellen
selec i i y o CO in CO
2
hyd ogena ion [61], whe e 28% con e sion o CO
2
wi h 100% CO
selec i i y a 350 °C and a mosphe ic p essu e was es ablished. The eac ion was ca ied ou
wi h H
2
/CO
2
eed a io o 4 and a a space eloci y o 100 ml/g
ca
/min. The exclusi e
selec i i y o CO was ob ained on Cu compa e o o he ca alys such as Ni, Ru, Rh and P on
alumina. These ca alys s p oduce only CH
4
while Pd ca alys s gi e bo h CH
4
and CO unde
he same eac ion condi ion.
A ca alys wi h 5 w .-% Cu suppo ed on alumina (H
2
/CO
2
eed a io o 3, space eloci y o
100 ml/g
ca
/min, 220 °C, 3 MPa) p oduces CO and CH
3
OH wi h he selec i i y o abou 87%
and 7.5%, espec i ely [62]. Some imes, he ca alys suppo may al e he selec i i y o
p oduc . Wi h he same amoun o Cu on SiO
2
he selec i i ies a e abou 87% and 13% o
CO and CH
3
OH, espec i ely while on TiO
2
he alues a e 92.5% and 4.3% o CO and
CH
3
OH, espec i ely and 2.9% o CH
4
. Simila ly, when 5 w .-% Cu/SiO
2
ca alys ope a ed
a 350 °C (p essu e o 0.2 ba , gas ecycle loop ha ing a H
2
/CO
2
a io o 4), he selec i i y o
CO was abou 97% wi h small amoun o CH
4
(<1.5%), C
2
H
4
(<0.5%) and C
2
H
6
(<0.4% )
[63]. The addi ion o a small amoun o Ni o a Cu ca alys enhances he ca aly ic ac i i y
bu also s a s CH
4
o ma ion. When he 5 w .-% Cu/SiO
2
ope a ed a 280 °C (6 MPa,
H
2
/CO
2
a io o 3, space eloci y o 50 ml/g
ca
/min), he selec i i y o CO dec eases and he
me hanol selec i i y inc eases o 76% and 24%, espec i ely [64].
I on ca alys :
The CO
2
hyd ogena ion on Fe ca alys s has been s udied by Lee and cowo ke s [35, 65-70]
and Cubei o e al. [71]. A 400 °C (2 MPa, space eloci y o 31.67 ml/g
ca
/min), he
Fe/Al
2
O
3
ca alys gi es
abou
49% con e sion o CO
2
and 12.4% and 87.6% o selec i i y
owa ds CO and hyd oca bons, espec i ely [35]. The addi ion o po assium o Fe
ema kably inc eases he ca alys ac i i y and selec i i y owa ds C
2+
–hyd oca bons. Unde
he same eac ion condi ion wi h K/Fe mola a io o 0.5, CO
2
hyd ogena ion ac i i y
inc eased o abou 70% while selec i i y changes o abou 4% and 96% o CO and
hyd oca bons, espec i ely. Fu he inc ease in mola a io o K/Fe o 1 showed a small
dec ease in ac i i y and hyd oca bon selec i i y while he selec i i y o CO sligh ly
inc eased. The addi ion o alumina i sel ac s as a p omo e when he Fe-Cu-K/Al
2
O
3
Basic heo y and backg ound o he wo k
36
ca alys was p epa ed by he co-p ecipi a ion me hod [65]. The long e m s abili y es on
Fe-K/Al
2
O
3
ca alys showed a signi ican dec ease in CO
2
con e sion and selec i i y o
hyd oca bons [66], whe e he deac i a ion was a ibu ed mainly due o he ca bonaceous
deposi ion on he ca alys su ace. A simila beha iou was ob ained wi h a Fe-Cu-K/Al
2
O
3
ca alys , whe e he ca alys deac i a ion was caused by he g ow h o c ys alli es and he
esul ing dec ease in dispe sion o ca alys componen s (p omo e s) [67].
To imp o e he s eng h o ca alys s, binde s a e some imes added. A high empe a u es,
binde s a e no chemically ine . The addi ion o an alumina binde o a Fe-K/Al
2
O
3
ca alys
showed excellen ac i i y and selec i i y owa ds highe hyd oca bons (C
5+
)
while ac i i y
and selec i i y d ama ically dec eased wi h silica binde . The eason o his in luence on
he ac i i y and selec i i y was he change in acidi y o ca alys , s uc u e and me al-suppo
in e ac ion [68]. The s ong me al-suppo in e ac ion be ween Fe-K/9-Al
2
O
3
makes he
ca alys mo e ac i e and selec i e owa ds long chain hyd oca bons and ligh ole ins [69]. A
s udy on Fe/Al
2
O
3
ca alys p epa a ion me hods such as imp egna ion, p ecipi a ion, and
physical mixing and ca alys cha ac e iza ion was pe o med by Cubei o e al. [71]. The
e ec o K p omo e in bo h CO
2
and CO hyd ogena ion was also in es iga ed.
Gold ca alys :
Gold ca alys s suppo ed on a ious me al oxides such as ZnO, Fe
2
O
3
, CeO
2
, TiO
2
, Z O
2
La(OH)
3
NiO, and Co
3
O
4
we e used by Saku ai e al. [72] o CO
2
hyd ogena ion. All he
ca alys s es ed o CO
2
hyd ogena ion we e ea ed a 150 o 400 °C (8 MPa, H
2
/CO
2
a io
o 3, space eloci y o 50 ml/g
ca
/min). Among all he ca alys s, Au suppo ed on TiO
2
,
Fe
2
O
3
,
and ZnO we e he mos ac i e ca alys s o he e e se wa e gas shi eac ion, whe e
he equilib ium con e sion o CO
2
was ob ained a 400 °C. Bu e en a lowe empe a u es
be ween 150 and 200 °C, Au/TiO
2
gi es equilib ium yield o CO. Howe e , in he
empe a u e ange o 150 o 200 °C, Au/Fe
2
O
3
and Au/ZnO ca alys s p oduce much
me hanol. The ac i i y o he ca alys and he selec i i y g ea ly di e depending upon he
eac ion p essu e and na u e o oxide suppo [73], i.e. when he p essu e was dec eased
om 5 o 0.1 MPa, he CO selec i i y inc eased om 86% o 99% on Au/TiO
2
ca alys
(250 °C).
Basic heo y and backg ound o he wo k
37
Cobal ca alys :
The CO
2
hyd ogena ion was also s udied o e Co Fische T opsch ca alys [74, 75]. The
Co/SiO
2
ca alys a 210 °C (2.4 MPa, H
2
/CO
2
a io o 4, space eloci y o 83.3 ml/g
ca
/min)
showed deac i a ion o e ime and also p oduces mo e han 70% o me hane [74]. The
compa a i e CO
2
hyd ogena ion on Fe, Co, and Ni ca alys using Al
2
O
3
as s uc u al suppo
showed highes ac i i y o e Co/Al
2
O
3
ca alys . The selec i i y o de o CO was Fe/Al
2
O
3
> Co/Al
2
O
3
> Ni/Al
2
O
3
[76].
Nickel ca alys :
Cu en ly, Ni is one o he mos s udied ca alys s o CO
2
hyd ogena ion eac ion because o
i s high ac i i y and compa a i ely low cos . A low empe a u es, he CO
2
hyd ogena ion
o e Ni ca alys p incipally p oduces CH
4
as a main p oduc [58, 77-80]. A kine ic s udy o
CO
2
hyd ogena ion o e Ni/SiO
2
ca alys showed ha he ac i a ion ene gy shi s om 89 o
39 kJ/mol as empe a u e is inc eased om 227 o 327 °C [58]. The p epa a ion me hod is
also impo an . The speci ic ac i i y o Ni/Al
2
O
3
cop ecipi a ed ca alys dec eases as he
me al loading inc eases whe eas he ac i i y inc eases wi h me al loading o an imp egna ed
ca alys [77]. The speci ic ac i i y ( ela ed o Ni only) inc eases wi h me al loading up o
30% and hen i dec eases o Ni/Z O
2
ca alys p epa ed by an ul asound assis ed me hod
[78]. Du ing he eac ion, ans o ma ion o zi conium o zi conium dioxide and he
c ys alliza ion o me allic Ni pa icles occu s [80]. A a high Ni con en , nickel exis s as a
bulk NiO and co espondingly p oduces mo e me hane [81]. The Ni ca alys p epa ed by
cop ecipi a ion me hod wi h a loading om 0 o 20% on CeO
2
showed ha he 2 w .-% Ni-
CeO
2
exhibi s excellen ca aly ic ac i i y, selec i i y, and s abili y o he e e se wa e gas
shi eac ion [81]. A 600 °C and a mosphe ic p essu e wi h H
2
/CO
2
a io o 1 and a qui e
high space eloci y o 2 l/g
ca
/min, CO
2
con e sion ob ained o e 2 w .-% Ni/CeO
2
ca alys
was abou 35% wi h 100% CO selec i i y. The o he RWGS eac ion ca alys (10 w .-%
NiO/ZnO) showed an e en highe ac i i y (CO
2
con e sion o 38%, which is e y nea o
equilib ium con e sion o 40%) and a selec i i y owa ds CO o abou 98% [82].
38
Objec i e and scope o he wo k
39
3. Objec i e and scope o he wo k
As men ioned in he p e ious chap e , he consump ion o ossil uels and he elease o CO
2
in o he a mosphe e is con inuously g owing. E e y yea billion ons o an h opogenic
ca bon dioxide a e eleased in o he a mosphe e. The abili y o sepa a e CO
2
om lue gases
and o s o e a se e al billion ons o CO
2
emi ed pe yea is ques ionable as a e he
en i onmen al consequences. The e o e he con e sion o CO
2
o use ul chemicals and uels
is an a ac i e op ion o CO
2
mi iga ion.
A possible a enue o sus ainable de elopmen is he ca aly ic con e sion o CO
2
o liquid
uels by Fische T opsch syn hesis and sola o wind ene gy. The e e se wa e gas shi
eac ion (RWGS) is he i s s ep, and he p oduced CO u he con e ed o liquid uels by
Fische T opsch syn hesis. The ans o ma ion o CO
2
and H
2
depends upon se e al ac o s
such as ca alys selec ion, a io o CO
2
/H
2
, and eac ion empe a u e and p essu e. The e o e,
he main ocus o his wo k was o use a sui able ca alys and o de e mine he op imal
eac ion condi ions o CO
2
hyd ogena ion in a ixed bed eac o .
The expe imen s we e designed such ha he RWGS eac ion could be examined in bo h he
o wa d and e e se di ec ion. In addi ion o hese expe imen s, he consecu i e eac ion o
CO o me hane was also conside ed. The s udies o all h ee eac ions we e pe o med using
a comme cial Ni ca alys in a lab scale ixed bed eac o .
Since anspo p ocesses (ex e nal bounda y laye di usion and po e di usion) may ha e
s ong in luence, his aspec was also conside ed.
The mos a ailable s udies o ca aly ic hyd ogena ion o CO
2
we e pe o med a low
empe a u es whe e me hane is he main p oduc . The e o e, in his wo k CO
2
hyd ogena ion
is s udied a high empe a u es whe e he e e se wa e gas shi eac ion is a ou ed
he modynamically and p oduces CO and H
2
O as main p oduc s. Fo he basic in es iga ion
o ca aly ic ac i i y, pu e alumina and some alumina suppo ed C , Zn-Cu, and Ni ca alys s
we e used. A e compa ing he ac i i y o all hese ca alys s, he comme cial Ni ca alys
supplied by Süd-Chemie (sphe ical pelle , d
p
= 5 o 7 mm) was chosen o a de ailed kine ic
s udy.
Objec i e and scope o he wo k
40
The pa ame e s ha in luence he eac ion a e such as empe a u e, eac an concen a ion,
esidence ime, and pa icle size o he ca alys we e s udied in a lab scale ixed bed qua z
eac o . To de e mine he kine ic, he expe imen s we e conduc ed a condi ions su icien ly
a away om equilib ium. No only he RWGS eac ion, bu also CO hyd ogena ion
(me hana ion) (CO + 3H
2
5 CH
4
+ H
2
O) and WGS eac ion (CO + H
2
O 5 CO
2
+ H
2
) we e
s udied in de ail.
Finally, a one dimensional adiaba ic and iso he mal ixed bed eac o model was de eloped
a echnical condi ions (high empe a u e) o e Ni/Al
12
O
19
as well as Al
2
O
3
ca alys sys em
o simula e he pe o mance o hyd ogena ion o CO
2
o CO o he p oduc ion o liquid
uels ia Fische T opsch syn hesis. The axial p o iles o eac o pe o mance ( empe a u e
and con e sion) we e simula ed along he ca aly ic ixed bed.
Expe imen al me hod and da a analysis
41
4. Expe imen al me hod and da a analysis
In he ollowing, he expe imen al se up and he p ocedu e used in CO
2
hyd ogena ion, CO
hyd ogena ion o me hane as well as wa e gas shi eac ion, and i s co esponding
calcula ion me hods a e in oduced.
4.1 Expe imen al se up
The expe imen al se -up wi h eac o appa a us cons uc ed o his s udy is shown in Fig. 4-1.
I mainly consis s o a qua z eac o , mass low con olle s, a hea ing u nace and a
empe a u e con ol sys em, a wa e sa u a o , a cooling sys em, and an online gas analyse .
Fig. 4-1: Expe imen al se -up used o CO
2
hyd ogena ion, CO hyd ogena ion and wa e gas
shi eac ion.
A pho og aph and a schema ic diag am o he bench-scale ixed bed eac o (2 cm inne
diame e and 45 cm leng h) used o he expe imen al s udies is shown in Fig. 4-2. In he
Expe imen al me hod and da a analysis
48
me hane a e he only de ec able p oduc s o CO
2
hyd ogena ion eac ion, CO
2
con e sion is
calcula ed as:
i
P
B-4AAD.,-4
2
|}
~
4
F
2
i
4
FJ
<
2
j
4
FJ
2
i
4
FJ
<
2
j
4
FJ
<
2
i
P
4
FJ
0ZZ
1
(4.3)
Simila ly, he yield o CO and CH
4
can be calcula ed (no CO and CH
4
in he eed) as:
i
{,A+
2
|}
4
F
2
i
4
FJ
2
i
4
FJ
<
2
j
4
FJ
<
2
i
P
4
FJ
0ZZ
1
(4.4)
j
{,A+
2
|
4
F
2
j
4
FJ
2
i
4
FJ
<
2
j
4
FJ
<
2
i
P
4
FJ
0ZZ
1
(4.5)
4.4.1.2 De e mina ion o he in insic kine ic pa ame e s
The in insic kine ic expe imen s we e ca ied ou in a qua z glass ixed bed eac o wi h an
inne diame e o 2 cm, illed wi h 1.2 g ca alys o 5 o 7 mm diame e . The iso he mal
expe imen s we e ca ied ou a a mosphe ic p essu e. The inle eed passed h ough he
eac o (H
2
/CO
2
= 6) was dilu ed wi h 23 ol.-% o ni ogen. The expe imen s we e
pe o med un il cons an con e sion o CO
2
was ob ained.
Fo he bimolecula CO
2
hyd ogena ion eac ion (CO
2
+ H
2
5 CO + H
2
O), he in insic
eac ion a e o e solid ca alys is calcula ed as:
D
E
|
}
~
F
E
|
}
~
|}
~
~
E
1
(4.6)
Fo he eac ion o de n 7 1, he eac ion a e cons an
E|}~
is gi en as:
E
|
}
~
F
2
0
5
~
4
@
?
5
0
2
5
0
4
E
|}
~
@
~
E
1
(4.7)
This calcula ed
E|}~
in Eq. (4.7) may no be he in insic a e cons an , which can be
con i med by calcula ing he e ec i eness ac o .
The modi ied esidence ime
E
as a unc ion o mass o ca alys and olume low a e a
eac ion empe a u e and o al p essu e o gas is gi en as:
E
F
6
5
2
L
FQ
4
1
(4.8)
Expe imen al me hod and da a analysis
49
The ac i a ion ene gy E
A
and equency ac o k
m,0
can be calcula ed om he A henius
equa ion:
E
|
}
~
F
E
F
A
!"
1
(4.9)
4.4.1.3 In e nal mass anspo calcula ions
As men ioned in chap e 2, he po e di usion limi a ions may occu , and he measu ed
kine ics may no be he in insic bu he e ec i e one. The e ec i e a e o chemical eac ion
is calcula ed wi h he help o he e ec i eness ac o 2 is gi en as:
D
E
#$$
F
%
D
E
1
(4.10)
The e ec i eness ac o 2 o n
h
o de i e e sible eac ion is gi en by:
%
F
D
E
#$$
D
E
F
)(4
C
8
8
1
(4.11)
The Thiele modulus o he ca alys wi h homogeneous dis ibu ion and o n
h
o de ,
i e e sible, bi-molecula CO
2
hyd ogena ion eac ion is gi en as
8
F
7
8
9
1
4
<
0
=
6
E
:
8
|}~
@
~
E
;
|}
~
#$$
1
(4.12)
whe e he cha ac e is ic leng h L
p
is he a io o he olume o he pa icle o he ex e nal
su ace o he pa icle ( o sphe ical pa icle L
p
= D
8
O3). Fo a shell ca alys he modi ied
Thiele modulus (8
4 is gi en as [83]
8
F
7
8
9
1
4
<
0
=
6
E
:
8
|}~
@
~
E
;
|}
~
#$$
1
(4.13)
whe eas he cha ac e is ic leng h (L
p
) o a shell ca alys is gi en as
7
8
F
0
5
e
0
5
)
&
D
8
h
u
D
8
3
1
(4.14)
whe e D
8
is he adius o pa icle and )
&
is he shell hickness o ac i e ma e ial. The e ec i e
di usi i y o CO
2
in a po ous ca alys is gi en as:
Expe imen al me hod and da a analysis
50
D
|}
~
#$$
F
E
8
7
8
D
|}
~
8FG#
F
E
8
7
8
e
0
D
|}
~
EFQ
<
0
D
|}
~
IJ
h
@
1
(4.15)
In he p esen s udy, mass anspo in he po es depends bo h on molecula di usion and
Knudsen di usion. The molecula di usion coe icien s o CO
2
in he gas mix u e we e
calcula ed by a compu e da abase p og am while Knudsen di usion coe icien s we e
calcula ed by using Eq. (4.16) as
D
|}~IJ
F
8FG#
3
9
K
9
L
M
N
|}
~
1
(4.16)
whe e
8FG#
is he po e diame e , R is he gas cons an , T is he empe a u e, and N
|}~
he
molecula weigh o CO
2
. The o uosi y ac o depends upon na u e o in e connec ing pa hs
which is in luenced by he connec i i y, shapes, and speci ic limi ing cons ic ions. In his
wo k, a ypical alue o 7
8
o 1.6 was assumed o Ni ca alys . The calcula ion o he
modi ied Thiele modulus o he shell ca alys needs he modi ied eac ion a e
cons an U
E
W, which is gi en o a sphe ical ca alys pa icle o adius D
8
and shell
hickness )
&
as:
E
F
8
&
E
6
8
6
&
E
1
(4.17)
The ac o V
p
/V
s
is he a io o olume o he pa icle o he olume o he shell:
6
8
6
&
F
0
5
e
0
5
)
&
D
8
h
u
@
1
(4.18)
Fo he Ni ca alys used in his wo k V
p
/V
s
is abou 2.4.
By Eq. (4.17) and Eq. (4.13),
E
can be calcula ed as
E
F
0
5
e
0
5
)
&
D
8
h
u
@
E
1
(4.19)
Inse ion o he cha ac e is ic leng h (7
8
) om Eq. (4.14) and o he modi ied a e cons an
(
E
4 om Eq. (4.19) in o Eq. (4.13), yields he modi ied Thiele modulus o a shell ca alys
ha ing ac i e ma e ial only in he shell:
Expe imen al me hod and da a analysis
51
8
F
0
5
e
0
5
)
&
D
8
h
u
[
¡
D
8
3
9
1
4
<
0
=
6
E
:
8
|}~
@
~
E
;
|}
~
#$$
1
(4.20)
Fo he case o )
&
FD
8
, Eq. (4.19) leads o he same equa ion ha is used o he calcula ion
o he Thiele modulus o a ca alys wi h a homogeneous dis ibu ion o he ac i e
componen (Eq. (4.12)).
4.4.1.4 Calcula ion o ex e nal mass anspo limi a ions
The e ec i e eac ion a e due o ex e nal mass ans e is calcula ed as
D
E
|
}
~
#$$
F
R
S
E
#T
U
|}
~
5
|}
~
&
W
1
(4.21)
The ex e nal mass ans e coe icien 3 (m/s) depends upon he pa icle size and geome y,
molecula di usion coe icien o gas (D
|}~EFQ
4, and he hyd odynamic condi ions such as
eloci y and iscosi y o he luid. 3 as a unc ion o he dimensionless She wood numbe
(
¢
) o he mass ans e can be calcula ed as
R
F
3
C
D
|}
~
EFQ
8
1
(4.22)
The ex e nal su ace a ea A
m,ex
(m
2
/kg) o sphe ical pa icle is
S
E
#T
F
Y
8
:
8
1
(4.23)
whe e d
p
and 6
p
a e he pa icle diame e and densi y o pa icle, espec i ely. When gas
lows h ough a ca alys ixed bed, he She wood numbe is gi en as
whe e 4
b
is he bed po osi y. The pa icle She wood numbe (
¢
£
) o gas low a ound he
pa icle and o lamina low, co ela ed o he Schmid numbe (
¢¤
) and he Reynolds
numbe (Re) is gi en as [20]:
3
C
8
F
=
<
Z
[
Y
]
9A
]
3B
^
1
(4.25)
The Reynolds numbe (
¥
) depends upon he in e s i ial eloci y u (m/s), he kinema ic
iscosi y o gas
¦
(m
2
/s), and he pa icle diame e d
p
(m):
3
C
F
_
0
<
0
[
`
2
0
5
H
a
4
b
3
C
8
1
(4.24)
Expe imen al me hod and da a analysis
52
9A
F
*
8
c
1
(4.26)
Simila ly, Schmid numbe (
¢¤
) depending upon he kinema ic iscosi y 7 (m
2
/s) and he
molecula di usion coe icien o gas D
|}~EFQ
(m
2
/s) as:
3B
F
c
D
|}
~
EFQ
1
(4.27)
Now he e ec i e eac ion a e due o he combined in luence o in e nal and ex e nal mass
ans e esis ances is gi en in app oxima ion ( o an almos i e e sible eac ion) by
d
E
#$$
F
e
0
%
E
|}
~
~
E
<
0
R
g
E
#T
~
h
@
1
(4.28)
CO hyd ogena ion (me hana ion) eac ion
4.4.2
4.4.2.1 Con e sion o CO
and yield o CO
2
and CH
4
Simila o CO
2
hyd ogena ion, he con e sion o CO and he yield o CO
2
and CH
4
a gi en
eac ion condi ion is calcula ed depending upon he ou le ca bon pe cen age alues ob ained
o e he analyse . As he me hane and ca bon dioxide a e he main p oduc s o CO
hyd ogena ion eac ions, he CO con e sion is calcula ed as:
i
B-4AAD.,-4
2
|}
4
F
2
i
P
4
FJ
<
2
j
4
FJ
2
i
4
FJ
<
2
j
4
FJ
<
2
i
P
4
FJ
0ZZ
1
(4.29)
and he yield o CO
2
and CH
4
in CO hyd ogena ion is calcula ed as:
i
P
{,A+
2
|}
~
4
F
2
i
P
4
FJ
2
i
4
FJ
<
2
j
4
FJ
<
2
i
P
4
FJ
0ZZ
1
(4.30)
j
{,A+
2
|
4
F
2
j
4
FJ
2
i
4
FJ
<
2
j
4
FJ
<
2
i
P
4
FJ
0ZZ
1
(4.31)
4.4.2.2 Kine ic analysis and mass anspo calcula ions
The calcula ions wi h ega d o he in insic kine ics and in e nal and ex e nal mass ans e
limi a ions o CO hyd ogena ion we e pe o med by using he equa ions applied o CO
2
hyd ogena ion (see sec ion 4.4.1).
Expe imen al me hod and da a analysis
53
Wa e gas shi (WGS) eac ion
4.4.3
4.4.3.1 Con e sion o CO
and yield o CO
2
and CH
4
The con e sion o CO and he yield o CO
2
and CH
4
in wa e gas shi eac ion we e
calcula ed by using he same equa ions as o he CO hyd ogena ion (Eq. (4.29) o (4.31)).
4.4.3.2 Kine ic analysis and mass anspo calcula ions
The kine ic analysis o he WGS eac ion was also done by using he equa ions o CO
2
hyd ogena ion eac ion (see sec ion 4.4.1).
54
Resul s and discussion
55
5. Resul s and discussion
As al eady men ioned in Chap e 4, his wo k is di ided in o ou pa s. In he i s pa
(Chap e 5.1), expe imen al esul s a e discussed o CO
2
hyd ogena ion s udied by using a
ixed bed eac o . This pa mainly includes he eac ion pa ame e s udy, he in insic
kine ics o CO
2
hyd ogena ion, and he in luence o po e di usion and ilm di usion on he
e ec i e eac ion a e o CO
2
hyd ogena ion. The second pa (Chap e 5.2) consis s o a
simila pa ame ic s udy o CO hyd ogena ion (me hana ion eac ion). Simila o he i s
and second pa , he hi d pa (Chap e 5.3) includes he pa ame e s udy o wa e gas shi
eac ion (as he e e se eac ion o CO
2
hyd ogena ion), again he in insic kine ics as well
as he in luence o po e di usion and ilm di usion. The ou h and las pa (Chap e 5.4)
includes he ixed bed eac o modelling o CO
2
hyd ogena ion a echnical condi ions.
5.1 CO
2
hyd ogena ion (RWGS)
The s udies on CO
2
hyd ogena ion (CO
2
+ H
2
9 CO + H
2
O) we e ca ied ou in a ixed bed
qua z eac o o e a wide empe a u e ange (300 – 900 °C) and a a mosphe ic p essu e.
The comme cial Ni ca alys was used. Fo compa ison, 8-Al
2
O
3
was also es ed o
in es iga e he ca aly ic ac i i y o he suppo only.
E ec o educ ion empe a u e on CO
2
hyd ogena ion (RWGS)
5.1.1
The comme cial NiO/Al
12
O
19
(G-90.B) ca alys was educed a wo di e en empe a u es
(500 and 800 °C) o examine he e ec o he educ ion empe a u e on he ca alys
pe o mance and he yield o p oduc s. Fig. 5-1 shows he con e sion o CO
2
and he yield
o CO and CH
4
ob ained o e he ca alys educed a 500 and 800 °C. A s ong dependence
o ac i i y and p oduc yield on he educ ion empe a u e was obse ed. On he Ni ca alys
educed a a low empe a u e o 500 °C o 15 h, me hane was o med almos exclusi ely
along wi h CO in he low eac ion empe a u e ange (below 600 °C), while on he ca alys
samples educed a high empe a u e o abou 800 °C o 3 h, he me hane o ma ion
s ongly dec eased in he same ange o eac ion empe a u e. The o e all con e sion o CO
2
ob ained a any eac ion empe a u e is high o he ca alys educed a 500 °C compa ed o
800 °C. This could be due o he deac i a ion o he ac i e species (pa icula ly me hane
o ming species) a high educ ion empe a u es.
Resul s and discussion
56
Fig. 5-1: CO
2
con e sion and CO and CH
4
yield e sus empe a u e; open symbol: ca alys
educed a 800 °C o 3 h; closed symbol: ca alys educed a 500 °C o 15 h
(p = 1 a m, m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
,
d
p
= 6 mm, y
H2
= 66 ol.-%,
y
CO2
= 11 ol.-%, es N
2
, gas low a e = 138 l/h (NTP)).
O e he educed ca alys bo h a 500 °C and 800 °C, me hana ion eac ion occu s (mos
a ou ed he modynamically a low empe a u e) along wi h he e e se wa e gas shi
(RWGS) eac ion. The yield o CO inc eases con inuously wi h inc ease in empe a u e bu
no he con e sion o CO
2
. This is a clea indica ion o di ec me hane o ma ion ia CO
2
hyd ogena ion (Saba ie eac ion) wi hou in e media e CO o ma ion. The o e all CH
4
o ma ion obse ed could be due o bo h CO
2
and CO hyd ogena ion bu i is qui e di icul
o quan i y because bo h eac ions occu simul aneously.
Fig. 5-2 shows he XRD pa e n o he Ni/Al
12
O
19
ca alys (11 w .-% Ni). The di ac og am
(A) o he esh Ni ca alys shows s ong NiO lines a 2B alues o 37.3, 43.3, and 62.9
con i ming he p esence o ee nickel oxide [84]. Fo bo h he educed and used ca alys no
ee nickel oxide lines we e obse ed. On he di ac og am (B) o he educed ca alys , he
lines we e obse ed o nickel c ys alli es a 44.5 and 51.8 while o he used ca alys (C) a
line was obse ed a 44.5. F om he di ac og am (B) i is clea ha he ca alys s ge
comple ely educed e en a low empe a u e o 500 °C. On he di ac og am (C) o he
used ca alys , he ee NiO lines we e no obse ed which means ha no oxida ion o Ni
happens by he wa e o med du ing he eac ion a 500 °C.
0
20
40
60
80
100
300 400 500 600 700 800 900
XCO2, YCO, YCH4 [C-%]
Tempe a u e [3C]
Equilib ium con e sion o CO2
(bo h o CO and CH4)
YCO
YCH4
XCO2
Resul s and discussion
57
Fig. 5-2: XRD pa e ns o esh NiO/Al
12
O
19
(A), educed ca alys a 500 °C (B), and used
ca alys a 500 °C (C).
To de e mine he exac empe a u e needed o he educ ion o he Ni ca alys and he
numbe o educible species p esen in he ca alys , a TPR expe imen was pe o med by
using C
HEMBET
-3000. TPR analysis begins by passing he analysis gas (10% hyd ogen in
Resul s and discussion
64
Ni ca alys 40% CO
2
con e sion is eached a abou 9 kg s m
-3
(Fig. 5-5). Hence, he ac i i y
o he Ni ca alys is abou 10 imes highe .
Fig. 5-9: E ec o modi ied esidence ime on CO
2
con e sion and selec i i y o CO and
CH
4
(T = 700 °C, p = 1 a m, m
ca
= 10 g, ca alys = Al
2
O
3
, d
p
= 1 – 3 mm,
y
H2
= 66 ol.-%, y
CO2
= 11 ol.-%, es N
2
).
Acco ding o he da a ob ained in his s udy, me hane o ma ion du ing CO
2
hyd ogena ion
may be due o he di ec me hana ion (CO
2
+ 4H
2
5 CH
4
+ 2H
2
O) o o he consecu i e
eac ion o CO hyd ogena ion (CO + 3H
2
5 CH
4
+ H
2
O). The consecu i e eac ion
mechanism in CO
2
hyd ogena ion implies he di icul y ha he op imum eac ion condi ions
o each o he eac ions a e di e en (high empe a u e o RWGS and low empe a u e o
CO hyd ogena ion).
E ec o ca alys pa icle size (RWGS and consecu i e me hana ion)
5.1.5
To de e mine he e ec o pa icle size on he ca alys pe o mance, pa icles o Ni ca alys
wi h less han 0.5 mm diame e (c ushed ca alys ) and 6 mm diame e we e used. Fig. 5-10
shows he con e sion o CO
2
and he yield o CO and CH
4
a di e en empe a u es.
The ca alys pa icles wi h diame e o less han 0.5 mm show a highe con e sion han he
6 mm pa icles wi h he ac i e ma e ial only a he shell side which may due o wo easons.
One is he esh ca alys which is no ea ed a high empe a u e o long ime has highe
2
2
2
2
2
422
2 2 422 42 22 2
XCO2, SCO, SCH4 [C-%]
Modi ied esidence ime [kg s m-3]
Equilib ium con e sion o CO2
SCO
SCH4
XCO2
Resul s and discussion
65
ac i i y han he used one (see Fig. 5-4). On he hand he in luence o di usion esis ances
may be smalle o small pa icles. This will be analysed in mo e de ail in sec ion 5.1.7.
Fig. 5-10: CO
2
con e sion and CO and CH
4
yield a di e en pa icle size in CO
2
hyd ogena ion (p = 1 a m, m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
, gas low a e =
138 l/h (NTP), y
H2
= 66 ol.-%, y
CO2
= 11 ol.-%, es N
2
,).
S abili y o he Ni ca alys o he RWGS eac ion a high empe a u e
5.1.6
The ime on s eam s abili y o Ni/Al
12
O
19
ca alys was es ed a 900 °C and a mosphe ic
p essu e wi h he o al gas low a e o 138 l/h in he ixed bed eac o . The es was ca ied
ou o ou days wi h in e media e cooling cycles. A he beginning o he expe imen , he
Resul s and discussion
66
ca alys bed empe a u e was inc eased om ambien empe a u e o 900 °C a he a e o 10
°C/min wi h a small low a e o N
2
and hen held cons an o 30 min o ge a s able
empe a u e. Then he equi ed gas composi ion (CO
2
/H
2
/N
2
) in he eac ion mix u e was
adjus ed and ed o he eac o sys em.
Fig. 5-11 shows he con e sion o CO
2
ob ained o e ime on s eam o he eac ion mix u e.
A such a high empe a u e o 900 °C, he CO
2
is p ac ically only con e ed by he e e se
wa e gas shi eac ion and CO is he main p oduc . On he i s day o ime on s eam, he
ca alys showed some deac i a ion whe e CO
2
con e sion dec eased om 68% o 66% in 8
h. A e ha he ca alys was cooled o 300 °C and kep cons an o e nigh . No deac i a ion
was obse ed on u he ime on s eam, bu a small loss in ca alys ac i i y du ing hea ing
and cooling cycle was obse ed which may be due o su ace modi ica ions.
Fig. 5-11: Con e sion o CO
2
e sus ime on s eam a 900 °C in CO
2
hyd ogena ion
(p = 1 a m, m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
, d
p
= 6 mm, y
CO2
= 11 ol.-%,
y
H2
= 66 ol.-%, es N
2
, gas low a e = 138 l/h (NTP)).
Kine ic analysis o he RWGS on he Ni ca alys
5.1.7
In insic kine ics
Fo he kine ic analysis, he concen a ions o CO
2
and H
2
we e a ied (Fig. 5-12 and Fig. 5-13).
The empe a u e was kep low (340 °C) o a oid an in luence o mass anspo on he
e ec i e eac ion a e, which hen equals he in insic a e.
Resul s and discussion
67
Fig. 5-12: E ec o CO
2
concen a ion on he CO
2
con e sion and he CO and CH
4
yield in
CO
2
hyd ogena ion (T = 340 °C, p = 1 a m, m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
,
d
p
= 6 mm, y
H2
= 66 ol.-%, y
CO2
= 11 – 22 ol.-%, N
2
= emaining p opo ion,
gas low a e = 48 l/h (NTP)).
Fig. 5-13: E ec o H
2
concen a ion on he CO
2
con e sion and he CO and CH
4
yield in
CO
2
hyd ogena ion (T = 340 °C, p = 1 a m, m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
,
d
p
= 6 mm, y
H2
= 44 – 66 ol.-%, y
CO2
= 11 ol.-%, N
2
= emaining p opo ion,
gas low a e = 48 l/h (NTP)).
2
4
4
34 3 3
XCO2, YCO, YCH4 [C-%]
CCO2 [mol/m3]
0E
1E
1E2
nCO2= 0
nCO2= 1
2
4
4
3 423 43
XCO2, YCO, YCH4 [C-%]
CH2 [mol/m3]
0E
1E
1E2
mH2= 0
mH2= 1
Resul s and discussion
68
The eac ion o de wi h espec o each eac an can be de e mined by using he ollowing
equa ion:
~
F
0
5
§
E
|}
~
¨
~
E
2
5
0
4
|}
~
@
<
0
©
@
@
(5.1)
Fig. 5-14 shows he plo o X
CO2
e sus C
CO2
. The o de was calcula ed by he in eg al
me hod i.e. he alue o he eac ion o de was de e mined by he bes i o he measu ed
da a. Fo he expe imen al condi ions used, he eac ion o de o CO
2
is 0.1.
Fig. 5-14: Dependence o CO
2
con e sion upon CO
2
concen a ion (T = 340 °C, p = 1 a m,
m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
, d
p
= 6 mm, y
CO2
= 11 – 22 ol.-%,
y
H2
= 66 ol.-%, N
2
= emaining p opo ion, gas low a e = 48 l/h (NTP)).
Simila ly, he eac ion o de wi h espec o hyd ogen was de e mined (Fig. 5-15), which
leads o a alue o 0.4. The eac ion a e and he eac ion a e cons an o CO
2
hyd ogena ion
was calcula ed by using Eq. (4.6) and Eq. (4.7), espec i ely.
The A henius plo is shown in Fig. 5-16. Only he alues o k
m,CO2
o T < 410 °C a e
included, because hen he eac ion a e is solely de e mined by he in insic kine ics. Abo e
his empe a u e, he eac ion a e is hen also con olled by mass anspo limi a ions.
0.02
0.04
0.06
0.08
0.10
0.12
0.14
3 3 3
XCO2
CCO2 [mol/m3]
n = -0.1
n = 0.1
n = 0.3
nCO2= 0.1
Fixed poin
Resul s and discussion
69
Fig. 5-15: Dependence o CO
2
con e sion upon H
2
concen a ion (T = 340 °C, p = 1 a m,
m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
, d
p
= 6 mm, y
H2
= 44 – 66 ol.-%,
y
CO2
= 11 ol.-%, N
2
= emaining p opo ion, gas low a e = 48 l/h (NTP)).
Fig. 5-16: A henius plo (1/T e sus ln k
m,CO2
) o CO
2
con e sion (p = 1 a m, m
ca
= 1.2 g,
ca alys = Ni/Al
12
O
19
, d
p
= 6 mm, y
H2
= 66 ol.-%, y
CO2
= 11 ol.-%, es N
2
,
gas low a e = 138 l/h (NTP)).
The equency ac o and ac i a ion ene gy (Fig. 5-16) a e 5.55 C 10
2
m
1.5
mol
0.5
kg
-1
s
-1
and
65 kJ/mol, espec i ely. The alue o E
A
(65 kJ/mol)
ob ained in his s udy is wi hin he
ange ound by o he au ho s o CO
2
hyd ogena ion eac ion o e Ni ca alys s (Table 5-1).
0.07
0.08
0.09
0.10
0.11
0.12
42 44 4 4 4
XCO2
C
H2
[mol/m3]
0.4
0.7
mH2= 0.4
Fixed poin
Resul s and discussion
70
Table 5-1: Ac i a ion ene gies epo ed o CO
2
hyd ogena ion on nickel ca alys s.
Ca alys Tempe a u e (°C) P essu e (a m) E
A
(kJ/mol) Re e ences
75.3-84.6% Rany Ni 160 – 270 1 88 – 91 [33]
20% Ni/Al
2
O
3
250 – 400 1 87 [76]
3% Ni/SiO
2
227 – 277 1 81 [87]
59% Ni-88 283 – 397 2 – 30 58 – 55 [88]
58% Ni-104 277 – 318 6 – 18 61 [89]
33.6% Ni (G-65) 200 – 230 1 106 [90]
42% NiO/Al
2
O
3
210 – 315 1 80 – 92 [91]
11% Ni
(G.90-B) 305 – 410 1 65 This wo k
In luence o in e nal mass anspo (po e di usion)
The po e di usion limi a ions we e de e mined o bo h he shell and non-shell ca alys
(homogeneous ma e ial). The model pa ame e s equi ed in he calcula ion o he e ec i e
eac ion a e a e gi en in Table 5-2.
The e ec i e a e o chemical eac ion is gi en as:
D
E
|
}
~
#$$
F
%
D
E
|
}
~
1
(5.2)
The Thiele modulus o an i e e sible, iso he mal, n h o de eac ion and o he sphe ical
pa icle o 0.5 mm and 6 mm diame e (non-shell ca alys ) was calcula ed by using Eq. (4.12)
while o he shell ca alys Eq. (4.17) was used. The e ec i eness ac o as a unc ion o he
Thiele modulus was de e mined by Eq. (4.11).
The Knudsen di usion coe icien D
|}~IJ
was calcula ed by using Eq. (4.16) and he
combined po e di usion coe icien D
|}~8FG#
was calcula ed by using Eq. (4.15). The
di usion coe icien o CO
2
in he ca alys po e a 500 °C and 1 ba is shown in Fig. 5-17.
Fo he a e age po e diame e o 230 nm, he di usion o CO
2
in he po es o he Ni ca alys
is de e mined by bo h Knudsen and molecula di usion ( ansi ion a ea). The Knudsen
Resul s and discussion
71
di usion domina es he anspo phenomenon a po e diame e o less han 100 nm while
molecula di usion domina es abo e he po e diame e o 10000 nm.
Table 5-2: Model pa ame e s used o he calcula ion o e ec i e eac ion a e o CO
2
hyd ogena ion o e Ni ca alys (p = 1 a m).
Pa ame e s 2345671
F equency ac o (k
m,0
) 5.55 · 10
2
m
1.5
mol
0.5
kg
-1
s
-1
Ac i a ion ene gy (E
A
) 65 kJ/mol
Pa icle densi y (6
p
) 1910 kg/m
3
Po osi y o pa icle (4
p
) 0.33
To uosi y o pa icle A
p
(assump ion) 1.6
Molecula di usion coe icien (D
CO2,mol
) a 773 K 1.83 · 10
-4
m
2
/s
Knudsen di usion coe icien (D
CO2,knu
) a 773 K 4.68 · 10
-5
m
2
/s
Po e di usion coe icien (D
CO2,po e
) a 773 K 3.72 · 10
-5
m
2
/s
Kinema ic iscosi y o CO
2
(7
CO2
) a 773 K 1.63 · 10
-4
m
2
/s
Fig. 5-17: Di usion coe icien o CO
2
in he po es o he Ni/Al
12
O
19
and Al
2
O
3
ca alys s
de e mined a 500 °C and 1 ba .
Resul s and discussion
72
The plo o he e ec i eness ac o e sus empe a u e is shown in Fig. 5-18. In e nal mass
ans e esis ance in he ca alys po es is no negligible o T > 400 °C e en o he shell
ca alys wi h 0.5 mm shell hickness o ac i e ma e ial (Ni).
Fig. 5-18: E ec i eness ac o o ca alys pa icle e sus empe a u e (p = 1 a m,
ca alys = Ni/Al
12
O
19
, y
CO2
= 11 ol.-%, y
H2
= 66 ol.-%, es N
2
, gas low a e
= 138 l/h (NTP)).
In luence o ex e nal mass anspo (bounda y laye di usion)
To de e mine he in luence o ilm di usion, he e ec i e eac ion a eD
E|}~#$$
as a
unc ion o ex e nal mass ans e coe icien 3 and ex e nal su ace a ea pe mass o he
ca alys A
m,ex
has been calcula ed by Eq. (4.21). All he calcula ions o he ex e nal mass
ans e coe icien (3), ex e nal su ace a ea o he ca alys (A
m,ex
),
She wood numbe (Sh),
Reynolds numbe (Re), and Schmid numbe (
¢¤
) a e gi en in sec ion 4.4.1.4. The e ec i e
eac ion a e due o he combined in luence o in e nal and ex e nal mass anspo is
calcula ed as:
d
E
#$$
F
e
0
%
E
|
}
~
|}
~
[
@
~
[
<
0
R
g
E
#T
U
~
5
~
ª«
W
h
@
1
(5.3)
Fig. 5-19 shows a good ag eemen o he calcula ions and he measu emen s o he used
shell ca alys . I also e eals ha o he hypo he ic case o a non-shell ca alys (pa icle
2324
2342
4322
22 22 22 22 22 22 22
E ec i eness ac o (1)
Tempe a u e [8C]
o pa icle (dp= 0.5 mm)
in luence o po e di usion
o ca alys o 0.5 mm
shell hickness
o pa icle (dp= 6 mm)
homogeneous dis ibu ion
Resul s and discussion
73
diame e o 6 mm) he e ec i e eac ion a e is lowe han o he shell ca alys . Fo a e y
small pa icle size (d
p
= 0.5 mm) he e ec i e eac ion a e is only in luenced by po e
di usion a empe a u e abo e 700 °C while no in luence o ex e nal mass anspo was
obse ed in he s udied ange o empe a u e (no shown he e). Conside ing 6 = 1 in Eq. (5.3)
enables o sepa a e he kine ic limi a ion om in e nal mass ans e limi a ion. In Fig. 5-19,
he e ec i e eac ion a e wi hou in e nal mass ans e limi a ion di e s signi ican ly om
he e ec i e eac ion a e wi h in e nal mass anspo limi a ion o he shell ca alys as well
as o he non-shell ca alys o 6 mm pa icle diame e . Fig. 5-19 shows ha he shell
ca alys ope a es in a mixed egime wi h bo h in e nal and ex e nal mass ans e . The
ca alys pe o mance is mo e and mo e con olled by ex e nal mass ans e abo e a
empe a u e o abou 900 °C.
No e ha Eq. (5.3) is only an app oxima ion because o wo easons:
• The e e se eac ion (CO + H
2
O 5 CO
2
+ H
2
) is no conside ed in he po e di usion
e m.
• Fo he ex e nal mass ans e , he equilib ium concen a ion is assumed a he ex e nal
su ace. In eali y, his alue is highe and only eached, i he chemical eac ion is e y
as .
Eq. (5.3) is he e o e only exac i only po e di usion o only ex e nal mass ans e plays a
ole. Fo he egime in be ween hese wo ex emes, Eq. (5.3) is only a hope ully good
app oxima ion.
Fu he mo e, in Fig. 5-19, in insic a e o he chemical eac ion (
m
) Eq. (4.6), he eac ion
a e o di usion bounda y laye (
m,ex
) Eq. (4.21), and he e ec i e eac ion a e due o
in e nal and ex e nal mass anspo (
m,e
) Eq. (5.3) a e shown. The e ec i e eac ion a es
(
m,e (calcula ed)
) we e calcula ed based on he inle concen a ions while he measu ed alues
(
m,e (measu ed)
) we e calcula ed based on he con e sion, inle concen a ion, and mass o
ca alys .
No e ha o empe a u es be ween abou 500 °C and 900 °C, he equilib ium con e sion is
no 100% (see Fig. 2-13 o R = 6) and eaches a minimum alue o a ound 75% a 600 °C.
This was no conside ed o he calcula ions o
m,e
and was only included o calcula e
m,ex
(FRg
E#T
2
~
5
~ª«
4).
Resul s and discussion
80
wi h empe a u e un il abou 450 °C a e eached. No ca alys deac i a ion was obse ed. In
he low empe a u e ange, he eac ion is con olled kine ically and no coke o ma ion ook
place. A low empe a u es (below 450 °C) a small amoun o CO
2
was p oduced by he
wa e gas shi eac ion [93]. A a e y high empe a u e o abou 800 °C, also no ca alys
deac i a ion was obse ed, bu he coke o med in he empe a u e ange o 480 o 710 °C
may no comple ely gasi ied a 800 °C, and he e o e he con e sion o CO was low and a
om equilib ium. The dec ease in con e sion due o he ca alys deac i a ion o e ime on
s eam is shown below (see sec ion 5.2.2 (Fig. 5-26)).
The coke o ma ion may occu ia Boudoua d eac ion (2CO k C + CO
2
) which is ela ed o
CO dissocia ion ac i i y in CO hyd ogena ion [94] and/o di ec ca bon monoxide
hyd ogena ion (CO + H
2
k C + H
2
O) o e he Ni ca alys [95]. Fig. 5-24 shows he
equilib ium con e sion o CO o e he empe a u e by bo h Boudoua d eac ion and di ec
CO hyd ogena ion. A high empe a u es bo h eac ions a e limi ed he modynamically.
Fig. 5-24: The equilib ium con e sion o CO in he Boudoua d eac ion (CO = 1 mole) and
di ec hyd ogena ion eac ion (H
2
/CO = 1) as unc ion o empe a u e and a
a mosphe ic p essu e.
S abili y o he Ni ca alys in CO me hana ion
5.2.2
Fig. 5-25 shows ime on s eam beha iou o he Ni ca alys o he CO con e sion and he
yield o CO
2
and CH
4
(450 °C, a mosphe ic p essu e). The maximum deac i a ion o he
2
2
2
2
2
422
22 22 22 22 22 22 22 4222
XCO, Equilib ium [C-%]
Tempe a u e [3C]
2CO 1 C + CO2
CO + H21 C + H2O
Resul s and discussion
81
ca alys s occu s gene ally in he ini ial s age o he eac ion [93, 96]. In his expe imen , no
deac i a ion was obse ed in 2 h ime on s eam a 450 °C, and he s eady s a e o eac ion
was eached wi hin some minu es. When he empe a u e inc eased u he (keeping all o he
condi ions cons an ), he coke o ma ion s a s abo e 480 °C, whe e CO con e sion
dec eases o e ime on s eam in he empe a u e ange o 480 o 710 °C (Fig. 5-26). Fo
example, a 620 °C, he con e sion o CO dec eases om 7 o 6.3% in 100 min o eac ion
ime. A he empe a u e o 710 and 810 °C, a small inc ease in con e sion o e ime on
s eam was obse ed which may be due o he gasi ica ion o coke o med in he empe a u e
ange o 480 o 710°C.
Fig. 5-25: CO con e sion and he yield o CO
2
and CH
4
e sus ime on s eam in CO
hyd ogena ion (T = 450 °C, p = 1 a m, m
ca
= 1.2 g, d
p
= 6 mm, y
H2
= 66 ol.-%,
y
CO
= 11 ol.-%, es N
2
, ca alys = Ni/Al
12
O
19
, gas low a e = 138 l/h (NTP)).
A e he comple ion o he expe imen a 810 °C, he eac o empe a u e was dec eased o
300 °C in he p esence o small N
2
s eam. Then he eac o empe a u e was inc eased again
om 300 o 700 °C a he a e o 10 °C/min by adding 10% O
2
in N
2
. The p esence o CO
and CO
2
species in he ou le gas s eam du ing he oxida ion o ca alys shows ha he coke
o med was no comple ely gasi ied a 810 °C.
2
4
4
2
2 2 2 2 2 422 42 42
XCO, YCO2, YCH4 [C-%]
Time on s eam [min]
XCO
YCO2
YCH4
Resul s and discussion
82
Fig. 5-26: Time on s eam s abili y o ca alys in CO hyd ogena ion a di e en empe a u e
(p = 1 a m, m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
, d
p
= 6 mm, y
H2
= 66 ol.-%,
y
CO
= 11 ol.-%, es N
2
, gas low a e = 138 l/h (NTP)).
E ec o esidence ime on me hana ion
5.2.3
The e ec o he esidence ime on he con e sion o CO and yield o CO
2
and CH
4
o e
empe a u e (300 – 800 °C) was s udied a wo di e en gas low a es o 48 l/h and 138 l/h.
In bo h cases, a simila end o CO con e sion as well as yield o CO
2
and CH
4
was
ob ained. In he low empe a u e ange o 300 o 450 °C, me hane p oduc ion inc eases
apidly wi h empe a u e and he eac ion is con olled kine ically. Abo e 450 °C, ca alys
deac i a ion due o coke o ma ion may also dec ease CO con e sion and he yield o CH
4
.
A a empe a u e o abou 800 °C and a low gas low a e o 48 l/h, he con e sion o CO
almos eached equilib ium. The e ec o he esidence ime on he ca alys pe o mance a
cons an empe a u e is gi en below (sec ion 5.2.4) in de ail.
Resul s and discussion
83
Fig. 5-27: CO con e sion and CO
2
and CH
4
yield a di e en gas low a e e sus
empe a u e in CO hyd ogena ion (p = 1 a m, ca alys = Ni/Al
12
O
19
, d
p
= 6 mm,
m
ca
= 1.2 g, y
H2
= 66 ol.-%, y
CO
= 11 ol.-%, es N
2
).
The e ec o dec ease in CO con e sion wi h inc easing empe a u e is no he esul o he
equilib ium (a leas o T < 700 °C) and also no o a a he slow deac i a ion by coke
o ma ion (sees Fig. 5-26 and Fig. 5-27). Hence, he inc ease o he empe a u e ob iously
leads o a e e sible deac i a ion o he ca alys o he me hana ion eac ion. This s ong
shi in selec i i y wi h empe a u e (less me hane) was also ound in case o CO
2
hyd ogena ion (Fig. 5-4).
I mus be no ed ha he wo expe imen s shown in Fig. 5-27 we e done one a e he o he ,
i.e. he expe imen wi h he lowe low a e (48 l/h) was done a e he expe imen wi h
Resul s and discussion
84
138 l/h. In be ween hese expe imen s, he ca alys was only oxidised (coke bu n-o ) and
hen educed wi h H
2
(600 °C). In bo h expe imen s, he ac i i y p o ile is simila . Hence, a
high empe a u e does no lead an i e e sible deac i a ion and only o a s ong dec ease o
he CH
4
selec i i y ( o T > 400 °C). Whe he and o wha ex en his selec i i y shi is
induced by he empe a u e and/o by he o ma ion o coke is s ill an open ques ion.
E ec o esidence ime a cons an empe a u e
5.2.4
The in luence o modi ied esidence ime on ca alys pe o mance as well as on he yield o
CO
2
and CH
4
was measu ed by a ying he o al gas low a e a 405 °C and a mosphe ic
p essu e. As shown in Fig. 5-28, he con e sion o CO and he yield o CH
4
and CO
2
inc eases wi h an inc easing modi ied esidence ime.
Fig. 5-28: In luence o modi ied esidence ime on CO con e sion and yield o CO
2
and
CH
4
in CO hyd ogena ion (T = 405 °C, p = 1 a m, m
ca
= 1.2 g, d
p
= 6 mm,
ca alys = Ni/Al
12
O
19
, y
H2
= 66 ol.-%, y
CO
= 11 ol.-%, es N
2
).
In e es ingly he e was no in luence o he modi ied esidence ime (and hus o he CO
con e sion) on he selec i i y o CO
2
and CH
4
(Fig. 5-29). The ex apola ion owa ds ze o
CO con e sion a ze o esidence ime esul ed in o no change in selec i i y o CH
4
and CO
2
.
Hence mos p obably, di ec con e sion o CO o CO
2
(2CO + 2H
2
5 CO
2
+ H
2
O) may ake
place (pa allel o he CO me hana ion i.e. CO + 3H
2
5 CH
4
+ H
2
O). By compa ing he
2
42
4
2
2
2 42 2 2 2 2 2
XCO, YCO2, YCH4 [C-%]
Modi ied esidence ime [kg s m-3]
XCO
YCH4
YCO2
Resul s and discussion
85
con e sion da a ob ained o CO hyd ogena ion (Fig. 5-28) and o CO
2
hyd ogena ion (Fig. 5-7)
a 405 °C, i is clea ha he ca alys has a simila ac i i y o CO and CO
2
hyd ogena ion,
bu less CH
4
and mo e CO a e ob ained in CO
2
hyd ogena ion compa ed o he CH
4
and CO
2
in he CO hyd ogena ion.
Fig. 5-29: CO con e sion e sus selec i i y o CO
2
and CH
4
o e Ni ca alys in CO
hyd ogena ion (T = 405 °C, p = 1 a m, 4
m
= 10 – 49.7 kg s m
-3
, m
ca
= 1.2 g,
d
p
= 6 mm, ca alys = Ni/Al
12
O
19
, y
H2
= 66 ol.-%, y
CO
= 11 ol.-%, es N
2
).
E ec o ca alys pa icle size on me hana ion
5.2.5
The pe o mance o he Ni ca alys o a pa icle diame e o less han 0.5 mm and 6 mm
o e he empe a u e is shown in Fig. 5-30. Fo bo h sizes o ca alys pa icles a simila end
o CO con e sion and yield o CO
2
and CH
4
was ob ained. The con e sion o CO inc eases
up o 450 °C and hen con inuously dec eases due o he dec ease o he me hane selec i i y
by coke o ma ion and/o inc easing empe a u e.
2
2
2
2
2
422
2 42 4 2 2
SCO2, SCH4 [C-%]
XCO [C-%]
SCH4
SCO2
Resul s and discussion
86
Fig. 5-30: CO con e sion and CO
2
and CH
4
yield a di e en pa icle size o ca alys e sus
empe a u e in CO hyd ogena ion (p = 1 a m, m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
,
y
H2
= 66 ol.-%, y
CO
= 11 ol.-%, es N
2
, gas low a e = 138 l/h (NTP)).
The p e iously used ca alys pa icles o 6 mm diame e in CO
2
hyd ogena ion s udy we e
applied in his expe imen . As men ioned in sec ion 5.1.2, he long- ime ea men o his
ca alys a high empe a u e educed i s me hana ion ac i i y. The 0.5 mm size ca alys
pa icles applied in his expe imen we e used be o e only in he pa icle size e ec s udy in
CO
2
hyd ogena ion. The e o e he ca alys pa icles o 0.5 mm ha e a highe ac i i y han
he 6 mm pa icles. Due o his eason, in he low empe a u e kine ic egime a highe
con e sion o CO and a highe yield o CH
4
was ob ained o e 0.5 mm size ca alys . Hence,
Resul s and discussion
87
he di ec compa ison o eac ion a e o e pa icle size wi h di e en ea men is no
possible.
Kine ic analysis and in luence o in e nal and ex e nal mass ans e on
5.2.6
me hana ion
In insic Kine ics
The in luence o CO and H
2
concen a ions on he me hana ion is shown in Fig. 5-31 and
Fig. 5-32. To de e mine he eac ion o de wi h espec o each eac an , a simila equa ion
ha was al eady applied in CO
2
hyd ogena ion (Eq. (5.1)) was used:
F
0
5
§
E
|}
¨
~
2
£
5
0
4
|}
8
@
<
0
©
@
@
8
(5.4)
The con e sion o CO was measu ed by a ying he espec i e concen a ion o he eac an s
CO and H
2
a 340 °C and a mosphe ic p essu e. Fi ing his da a by Eq. (5.4) shows a
nega i e eac ion o de o CO (p = -0.3) and a posi i e o de o H
2
(q = 0.7) ( o de ail see
Appendix B.1.1). These alues a e in good ag eemen wi h ypical alues epo ed in he
li e a u e [97, 98].
Fig. 5-31: E ec o CO concen a ion on he CO con e sion and he CO
2
and CH
4
yield
(T = 340 °C, p = 1 a m, m
ca
= 1.2 g, d
p
= 6 mm, ca alys = Ni/Al
12
O
19
,
y
H2
= 66 ol.-%, y
CO
= 13.2 – 22 ol.-%, N
2
= emaining p opo ion, gas low a e
= 48 l/h (NTP)).
2
4
3 3 3
XCO, YCO2, YCH4 [C-%]
CCO [mol/m3]
0E
1E
1E2
pCO = 1
pCO = 0
Resul s and discussion
88
Fig. 5-32: E ec o H
2
concen a ion on he CO con e sion and he CO
2
and CH
4
yield
(T = 340 °C, p = 1 a m, m
ca
= 1.2 g, d
p
= 6 mm, ca alys = Ni/Al
12
O
19
,
y
H2
= 66 ol.-%, y
CO
= 11 ol.-%, y
H2
= 44 – 66 ol.-%, N
2
= emaining
p opo ion, gas low a e = 48 l/h (NTP)).
The in insic a e o CO hyd ogena ion eac ion is he e o e gi en as
D
E
|}
F
E
|}
|}
[
¬
~
[
u
1
(5.5)
whe e k
m,CO
is he eac ion a e cons an o CO hyd ogena ion eac ion which can be
calcula ed as:
E
|}
F
2
0
5
4
@
5
0
2
£
5
0
4
E
|}
8
@
~
Æ
C
ADA
¯
B
F
5
Z
[
3
°
F
Z
[
±
1
(5.6)
The A henius plo is shown in Fig. 5-33. Only he alues o k
m,CO
o T < 420 °C a e
included whe e he eac ion a e is exclusi ely de e mined by he in insic kine ic. The
ac i a ion ene gy E
A
and equency ac o k
m,0
calcula ed om he A henius equa ion
(Eq. (4.9)) a e 102 kJ/mol and 2.35 · 10
5
m
1.2
mol
0.6
kg
-1
s
-1
, espec i ely. The alue o E
A
ob ained in his s udy (102 kJ/mol) o CO hyd ogena ion (me hana ion) is wi hin he ange
ound by o he au ho s using Ni ca alys s (see Table 5-5).
2
4
3 423 43
XCO, YCO2, YCH4 [C-%]
CH2 [mol/m3]
0E
1E
1E2
qH2 = 1
qH2 = 0
Resul s and discussion
89
Fig. 5-33: Reac ion a e cons an (1/T e sus ln k
m,CO
) o CO hyd ogena ion (p = 1 a m,
m
ca
= 1.2 g, d
p
= 6 mm, ca alys = Ni/Al
12
O
19
,
y
H2
= 66 ol.-%, y
CO
= 11 ol.-%,
es N
2
, gas low a e = 138 l/h (NTP)).
Table 5-5: Ac i a ion ene gies epo ed in li e a u e o CO hyd ogena ion on Ni ca alys s.
Ca alys Tempe a u e (°C) P essu e (a m) E
A
(kJ/mol) Re e ences
Monoli hic Ni 200 – 350 6.8 104 – 89 [98]
3% Ni/SiO
2
227 – 277 1.38 96 [99]
Ni/MgAl
2
O
4
205 – 290 1.4 96.7 [100]
10% Ni/SiO
2
270 – 320 1 112 [101]
1.04Ni/K-Al
2
O
4
339 – 364 1 94
²
3
[102]
11% Ni (G.90-B) 305 – 345 1 102 This wo k
In luence o mass ans e limi a ion
The in luence o in e nal mass anspo on he e ec i e eac ion a e o CO hyd ogena ion
was de e mined o he shell ca alys as well as non-shell ca alys (6 mm pa icle diame e
wi h homogeneous dis ibu ion). Fo ex e nal mass anspo , ex e nal su ace pe mass o
2242
(42
(
(
(
(
(
23224 23224 23224 232244 23224 23224
T [8C]
ln km,CO [m1.2 mol-0.6 kg-1 s-1]
1/T [1/K]
T endline
in insic eac ion a e cons an
EA= 102 kJ/mol
Km,o = 2.35 2 105m1.2 mol0.6 kg-1 s-1
Resul s and discussion
96
Fig. 5-38: In luence o modi ied esidence ime on CO con e sion and he yield o CO
2
and
CH
4
in wa e gas shi eac ion (T = 405 °C, p = 1 a m, m
ca
= 1.2 g, d
p
= 6 mm,
ca alys = Ni/Al
12
O
19
, y
H2O
= 44 ol.-%, y
CO
= 11 ol.-%, es N
2
).
S abili y o he Ni ca alys in he WGS eac ion
5.3.3
5.3.3.1 Low empe a u e s abili y
In he wa e gas shi eac ion, he Ni ca alys beha es di e en ly a di e en empe a u e.
No deac i a ion was obse ed below 450 °C, bu he ca alys shows deac i a ion due o coke
o ma ion in he empe a u e ange o 450 o 700 °C. So, depending upon his beha iou he
ime on s eam s abili y o ca alys was s udied a bo h low and high empe a u es.
The low empe a u e ca aly ic s abili y o he Ni/Al
12
O
19
ca alys in he WGS eac ion was
es ed o abou 100 min a 340 °C and 450 °C, and a mosphe ic p essu e. Fig. 5-39 shows
he con e sion o CO o e he ime on s eam. The Ni/Al
12
O
19
ca alys showed a good
s abili y a hese low empe a u es (340 °C and 450 °C).
0
10
20
30
40
0 10 20 30 40
XCO, YCO2, YCH4 [C-%]
Modi ied esidence ime [kg4s/m3]
T = 405 0C
XCO 3 YCO2
YCH4
Resul s and discussion
97
Fig. 5-39: CO con e sion e sus ime on s eam in WGS eac ion a low empe a u e
(p = 1 a m, m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
, d
p
= 6 mm, y
CO
= 11 ol.-%,
y
H2O
= 44 ol.-%, es N
2
,
gas low a e = 48 l/h (NTP)).
5.3.3.2 High empe a u e s abili y
The ime on s eam beha iou o he Ni ca alys in he high empe a u e ange o 500 °C o
900 °C and a a mosphe ic p essu e o wa e gas shi eac ion is shown in Fig. 5-40. I was
di icul o de e mine he amoun o coke o med ia he mog a ime ic analysis because a
e y small amoun o coke is o med and he mass loss by bu ning o he coke and he mass
gain by ca alys oxida ion akes place simul aneously.
The expe imen al p ocedu e was as ollows:
A i s , he wa e gas shi eac ion was pe o med a a ce ain empe a u e, e.g. a 500 °C
(Fig. 5-40). Then he empe a u e was inc eased o 900 °C, whe e no only he WGS
eac ion akes place bu also he gasi ica ion o he coke o med a he low empe a u e
expe imen (Fig. 5-41).
Fig. 5-40 indica es ha a 900 °C, no coke is o med, i.e. he coke ha may be o med is
gasi ied wi h s eam a an app op ia e a e. Hence, he ca alys s used o WGS eac ion a
lowe empe a u es o abou 100 min egain hei ac i i y by gasi ica ion, i he empe a u e
o 900 °C is adjus ed a e wa ds. This is shown in Fig. 5-41.
2
4
2
2 2 2 2 2 422 42
XCO [C-%]
Time on s eam [min]
450 4C
340 4C
Resul s and discussion
98
Fig. 5-40: CO con e sion e sus ime on s eam in WGS eac ion a high empe a u es
(p = 1 a m, m
ca
= 1.2 g, ca alys = Ni/Al
12
O
19
, d
p
= 6 mm, y
CO
= 11 ol.-%,
y
H2O
= 44 ol.-%, es N
2
, gas low a e = 48 l/h (NTP)).
Fig. 5-41: CO con e sion e sus ime on s eam in WGS eac ion a 900 °C a e he lowe
empe a u e expe imen s (p = 1 a m, m
ca
= 1.2 g, d
p
= 6 mm, y
CO
= 11 ol.-%,
y
H2O
= 44 ol.-%, es N
2
, ca alys = Ni/Al
12
O
19
, gas low a e = 48 l/h (NTP)).
In ano he expe imen , he eshly educed Ni ca alys (800 °C) was applied o 6 h in he
eac ion a 650 °C whe e he ca alys shows deac i a ion wi h ime on s eam (Fig. 5-42).
Then his used ca alys was oxidised in he same eac o using 5.5 ol.-% O
2
in N
2
by
inc easing he empe a u e om 300 o 700 °C a a a e o 10 °C/min. The p esence o CO
2
4
2
2
2 2 2 2 2 422 42 42
XCO [C-%]
Time on s eam [min]
500 4C
700 4C
800 4C
600 4C900 4C
2
42
2
2
2
2
2
2 2 2 2 2 422 42
XCO [C-%]
Time on s eam [min]
2215E
2215E
2215E
2215E
TWGS = 900 4C
Resul s and discussion
99
and CO
2
species in he ou le gas s eam du ing oxida ion o ca alys (coke bu ning) we e
analysed using a mic o-gas analyse and he amoun o coke o med was quan i ied om he
CO
2
p oduced du ing he ca alys oxida ion. The ac i i y o he ca alys was hen eco e ed
by oxida ion/ educ ion. The egene a ed ca alys was again used o 3 h and he same
p ocedu e was epea ed o con i m he coke o ma ion.
Fig. 5-42: Time on s eam s abili y and coke o ma ion o e eshly educed ca alys in
WGS eac ion (T = 650 °C, p = 1 a m, m
ca
= 1.2 g, d
p
= 6 mm, y
CO
= 11 ol.-%,
y
H2O
= 44 ol.-%, es N
2
, ca alys = Ni/Al
12
O
19
, gas low a e = 48 l/h (NTP)).
Fig. 5-42 clea ly shows ha he egene a ion by coke bu n-o is possible. The amoun o
coke is a he small compa ed o he ca bon (as CO) ha has passed h ough he eac o .
Du ing he 6 h expe imen , abou 17 g o ca bon (as CO) ha e en e ed he eac o bu only
0.45 mg o coke was o med. Ne e heless his small amoun o coke is su icien o a
s ong deac i a ion o he ca alys , al hough he amoun o coke (0.04 mmol C) is s ill small
compa ed o he amoun o Ni (2 mmol), bu a leas in he same o de o magni ude (and no
all o he Ni ma e ial is accessible by eac an s).
Resul s and discussion
100
Kine ic analysis and in luence o in e nal and ex e nal mass ans e on he WGS
5.3.4
eac ion
In insic kine ics
The eac ion o de s o he wa e gas shi eac ion wi h espec o eac an and p oduc
componen s o e he Ni/Al
12
O
19
ca alys we e de e mined by i ing he expe imen al da a o
Eq. (5.8).
F
0
5
§
E
|}
¨
~
}
&
~
J
|}
~
³
|}
G
@
2
d
5
0
4
<
0
©
@
@
G
(5.8)
The espec i e expe imen al esul s a e shown in he Fig. 5-43 o Fig. 5-46. The eac ion
o de s wi h espec o eac an CO and H
2
O a e 0.8 and 0.4 while he p oduc s H
2
and CO
2
shows a -0.15 and -0.1 o de dependencies on he eac ion a e o WGS eac ion. (see
Appendix B.1.2). Bo h o hese nega i e o de dependencies o p oduc s we e no
conside ed in he in insic a e calcula ion.
Fig. 5-43: E ec o CO concen a ion on he CO con e sion and he CO
2
and CH
4
yield in
WGS eac ion (T = 340 °C, p = 1 a m, m
ca
= 1.2 g, d
p
= 6 mm, y
H2O
= 44 ol.-%,
y
CO
= 5.5 – 14.3 ol.-%, N
2
= es p opo ion, ca alys = Ni/Al
12
O
19
, gas low a e
= 48 l/h (NTP)).
2
42
4
2
2
432 434 34 34
XCO, YCO2, YCH4 [C-%]
C
CO
[mol/m3]
0E
1E
1E2
CO = 1
CO = 0
Resul s and discussion
101
Fig. 5-44: E ec o H
2
O concen a ion on he ca alys ac i i y and he CO
2
and CH
4
yield in
WGS eac ion (T = 340 °C, p = 1 a m, m
ca
= 1.2 g, d
p
= 6 mm, y
CO
= 11 ol.-%,
y
H2O
= 34 – 54 ol.-%, N
2
= es p opo ion, ca alys = Ni/Al
12
O
19
, gas low a e
= 48 l/h (NTP)).
Fig. 5-45: E ec o H
2
addi ion in o he eed gas o WGS eac ion (T = 340 °C, p = 1 a m,
m
ca
= 1.2 g, d
p
= 6 mm, ca alys = Ni/Al
12
O
19
, y
CO
= 11 ol.-%, y
H2O
= 44 ol.-%,
y
H2
= 0 – 18.75 ol.-%, N
2
= es p opo ion, gas low a e = 48 l/h (NTP)).
2
42
4
2
2
3 3 4232
XCO, YCO2, YCH4 [C-%]
CH2O [mol/m3]
0E
1E
1E2 sH
2
O = 0
sH2O = 1
2
4
4
4
4
2 43 3 3
XCO, YCO2, YCH4 [C-%]
C
H2
[mol/m3]
0E
1E
1E2
uH2= 0
Resul s and discussion
102
Fig. 5-46: E ec o CO
2
addi ion in o he eed gas o WGS eac ion (T = 340 °C, p = 1 a m,
m
ca
= 1.2 g, d
p
= 6 mm, ca alys = Ni/Al
12
O
19
, y
CO
= 11 ol.-%, y
H2O
= 44 ol.-%,
y
CO2
= 0 – 16.67 ol.-%, N
2
= es p opo ion, gas low a e = 48 l/h (NTP)).
The in insic eac ion a e o WGS eac ion has been calcula ed as
D
E
|}
F
E
|}
|}
[
n
~
}
[
F
E
A
!"
|}
[
n
~
}
[
1
(5.9)
whe e k
m,CO
is he eac ion a e cons an o WGS eac ion which can be calcula ed as:
E
|}
F
2
0
5
4
@
[
n
5
0
2
Z
[
K
5
0
4
E
|}
[
n
@
~
}
[
1
(5.10)
The empe a u e dependency o k
m,CO
is shown in Fig. 5-47.
To de e mine he in insic kine ic, only he alues o k
m,CO
o T < 345 °C a e included
whe e no mass anspo o he modynamic limi a ions occu . The equency ac o k
m,0
and
appa en ac i a ion ene gy o eac ion E
A
a e 3.46 · 10
5
m
3.6
mol
-0.2
kg
-1
s
-1
and 96 kJ/mol,
espec i ely. The ac i a ion ene gy da a epo ed in li e a u e o he wa e gas shi eac ion
o e Ni con aining ca alys is gi en in Table 5-7. The alue o E
A
(96 kJ/mol) ob ained in
his s udy is wi hin he ange ound by o he au ho s.
2
4
4
4
4
2 43 3 3
XCO, YCO2, YCH4 [C-%]
C
CO2
[mol/m3]
0E
1E
1E2
CO2= 0
Resul s and discussion
103
Fig. 5-47: Reac ion a e cons an (1/T e sus ln k
m,CO
) o he WGS eac ion (p = 1 a m,
m
ca
= 1.2 g, d
p
= 6 mm, ca alys = Ni/Al
12
O
19
, y
H2O
= 44 ol.-%,
y
CO
= 11 ol.-%, es N
2
, gas low a e = 48 l/h (NTP)).
Table 5-7: Ac i a ion ene gies epo ed o wa e gas shi eac ion on nickel ca alys s.
Ca alys Temp (°C) P essu e (a m) E
A
(kJ/mol) Re .
5% Ni/Al
2
O
3
monoli h 300 – 1000 1 85 [104]
5% Ni/5% Ce/Al
2
O
3
monoli h 300 – 1000 1 85 [104]
5 a .% Ni/Ce(10% La)O
x
275 – 300 1 38 [105]
Black NiO 200 – 300 1 96
²
p
[106]
G een NiO 200 – 300 1 89
²
p
[106]
11% Ni
(G.90-B) 305 – 345 1 96 This wo k
In luence o mass anspo
To de e mine he in luence o mass anspo on he e ec i e eac ion a e o wa e gas shi
eac ion, all he calcula ions we e made by using he equa ions om sec ion 4.4.1. Simila o
22422222
(
(3
(
(3
(
(3
(
(3
(
232242 23224 23224 23224 23224 23224
T [8C]
ln km,CO [m3.6 mol-0.2 kg-1 s-1]
1/T [1/K]
EA= 96 kJ/mol
km,0 = 3.46 61105m3.6 mol-0.2 kg-1 s-1
T endline
in insic eac ion a e cons an
Resul s and discussion
104
CO
2
and CO hyd ogena ion, he in luence o mass ans e in WGS eac ion was calcula ed
o bo h shell ca alys ( hickness o 0.5 mm) and non-shell ca alys (d
p
= 6 mm). The
e ec i e eac ion a e due o he in luence o bo h in e nal and ex e nal mass anspo was
calcula ed as:
d
E
#$$
F
e
0
%
E
|}
|}
[
n
~
}
[
<
0
R
g
E
#T
U
5
ª«
W
h
@
1
(5.11)
The model pa ame e s equi ed in he calcula ion o e ec i e eac ion a es o wa e gas
shi eac ion a e gi en in Table 5-8. The de ailed po e di usion calcula ion o Ni/Al
12
O
19
ca alys is gi en in appendix B.2 (Fig. B-8).
Table 5-8: Model pa ame e s used o he calcula ion o e ec i e eac ion a e o he wa e
gas shi eac ion o e Ni ca alys (p = 1 a m).
Pa ame e s Values
F equency ac o (k
m,0
) 3.46 · 10
5
m
3.6
mol
-0.2
kg
-1
s
-1
Ac i a ion ene gy (E
A
) 96 kJ/mol
Pa icle densi y (6
p
) 1910 kg/m
3
Po osi y o pa icle (4
p
) 0.33
To uosi y o pa icle A
p
(assump ion) 1.65
Molecula di usion coe icien o CO (D
CO,mol
) a 773 K 1.07 · 10
-4
m
2
/s
Knudsen di usion coe icien (D
CO,knu
) a 773 K 5.86 · 10
-5
m
2
/s
Po e di usion coe icien (D
CO,po e
) a 773 K 3.78 · 10
-5
m
2
/s
Kinema ic iscosi y o CO (7
CO
) a 773 K 8.17 · 10
-5
m
2
/s
In he wa e gas shi eac ion, he in luence o po e di usion on he e ec i e eac ion a e
was e alua ed om e ec i eness ac o 6. Fig. 5-48 shows he e ec i eness ac o e sus
empe a u e o he shell as well as o a hypo he ic non-shell ca alys . Fo bo h ca alys s he
e ec i eness ac o dec eases s ongly wi h inc easing empe a u e close o 345 °C. Fig. 5-49
compa es he e ec i e eac ion a es (conside ing 6 = 1 and wi h ac ual 6). The e is a clea
in luence o po e di usion o bo h he shell and non-shell ca alys .
Resul s and discussion
105
Fig. 5-48: E ec i eness ac o o ca alys e sus empe a u e (p = 1 a m, m
ca
= 1.2 g,
ca alys = Ni/Al
12
O
19
, y
CO
= 11 ol.-%, y
H2O
= 44 ol.-%, es N
2
, gas low a e
= 48 l/h (NTP)).
Fig. 5-49: In luence o in e nal and ex e nal mass anspo on he e ec i e eac ion a e o
WGS eac ion (p = 1 a m, m
ca
= 1.2 g, y
CO
= 11 ol.-%, y
H2O
= 44 ol.-%,
es N
2
, ca alys = Ni/Al
12
O
19
, gas low a e = 48 l/h (NTP)).
23224
2324
234
4
22 22 22 22 22 22
E ec i eness ac o (1)
Tempe a u e [8C]
po e di usion egion
*D197D8 6A1#911""$
C"'AA3!1!8D438
*D197D8 6A1*1231""
!CA6618C &A!!11
Resul s and discussion
112
Fig. 5-51: Axial p o iles o CO
2
con e sion o he Ni/Al
12
O
19
ca alys and h ee di e en
cases (Eq. (5.19);
E|}~
» R », and egula model (H
2
/CO
2
= 3).
The simula ed adiaba ic empe a u e p o iles a di e en gas supe icial eloci ies along he
axial eac o coo dina e a e shown in Fig. 5-52 bo h o Ni/Al
12
O
19
and Al
2
O
3
as ca alys .
The eac ion empe a u e dec eases along he leng h due o he p e ailing endo he mic
Resul s and discussion
113
RWGS eac ion, i.e. he ene gy consump ion by he eac ion leads o a cooling o he gas.
Fo an inle gas empe a u e o 1200 °C and he highes supe icial gas eloci y o 15 m/s (a
T
0
), he adiaba ic inal empe a u e o 998 °C is eached a a eac o leng h o abou 0.4 m
o he Ni ca alys and a abou 25 m o he Al
2
O
3
ca alys . The espec i e esidence imes
( ela ed o he emp y ube) a e only abou 30 ms o he Ni ca alys and abou 2 s o he
Al
2
O
3
. Fo lowe gas eloci ies, he equi ed leng hs (and esidence imes) a e smalle .
Fig. 5-52: Axial empe a u e p o iles o an adiaba ic eac o o Ni/Al
12
O
19
and Al
2
O
3
as
ca alys s (H
2
/CO
2
= 3).
Simila ly, he axial p o iles o he CO
2
con e sion a e shown in Fig. 5-53 o he simula ed
adiaba ic and iso he mal case bo h o he Ni and Al
2
O
3
ca alys . The con e sion o CO
2
and
Resul s and discussion
114
hus he yield o CO inc eases along he eac o leng h and eaches he maximum
(equilib ium) alue a he eac o ou le . As he gas supe icial eloci y inc eases, he eac o
leng h equi ed o each he equilib ium also inc eases. As al eady s a ed, an inc easing gas
eloci y also inc eases he equi ed leng h o he eac o o each (almos ) he inal
equilib ium alue. Fo he Ni ca alys and an inle empe a u e o 1200 °C, an inc ease om
0.5 m/s o 15 m/s leads o an inc ease o he eac o leng h om abou 0.013 m o 0.4 m o
he adiaba ic eac o and om 0.01 m o 0.3 m o he iso he mal eac o .
Fig. 5-53: Con e sion o CO
2
o e Ni/Al
12
O
19
and Al
2
O
3
ca alys s a di e en gas
supe icial eloci ies e sus axial eac o co-o dina e (H
2
/CO
2
= 3).
The con e sion o CO
2
o CO ob ained a he eac o ou le is abou 82% o he adiaba ic
eac o and 87% o he iso he mal eac o .
Resul s and discussion
115
Fig. 5-54 shows he con e sion o CO
2
along he eac o leng h a di e en H
2
/CO
2
a ios o
he adiaba ic and iso he mal simula ion (again bo h o he Ni and Al
2
O
3
ca alys ). As he
a io H
2
/CO
2
inc eases, he con e sion o CO
2
aises along he eac o leng h. Fo example,
i he inle a io o H
2
/CO
2
is inc eased om 1 o 6, he con e sion o CO
2
a he eac o
ou le inc eases om abou 55% o 91% o he adiaba ic eac o and om 62% o 93% o
he iso he mal eac o . Bu high inle H
2
/CO
2
a ios also inc ease he H
2
/CO a io o he
p oduced syngas, which hen may no be op imal o a subsequen Fische T opsch
syn hesis. The H
2
/CO a io o he syngas a he eac o ou le (R
syngas
) o di e en H
2
/CO
2
a ios o he eed o he RWGS eac o is shown in Fig. 5-55.
Fig. 5-54: Con e sion o CO
2
o e Ni/Al
12
O
19
and Al
2
O
3
ca alys s a di e en inle gas
composi ion e sus axial eac o co-o dina e (u
s
= 15 m/s, R = H
2
/CO
2
).
Resul s and discussion
116
Fig. 5-55: Mola H
2
/CO a io o he syngas (R
syngas
) a he eac o ou le o di e en mola
H
2
/CO
2
a ios o he eed (R
eed
) o he RWGS eac o (p = 1 ba ).
Fig. 5-55 indica es ha an almos op imal H
2
/CO a io o he syngas o sligh ly abo e wo
( o accoun o he me hana ion, which always akes place du ing Fische T opsch) is
eached o a H
2
/CO
2
a io o he eed o abou h ee. This a io was he e o e used o
u he simula ions.
As men ioned be o e, he eac o leng h equi ed o ge a pa icula con e sion depends upon
he gas inle empe a u e, he gas supe icial eloci y, and he inle gas composi ion. Fo he
RWGS eac ion o e he Ni/Al
12
O
19
ca alys (inle empe a u e o 1200 °C, inle H
2
/CO
2
a io o 3), he eac o leng h equi ed o each 99% o he equilib ium con e sion in he
adiaba ic and iso he mal case is shown in Fig. 5-56.
Fig. 5-57 shows he simila pa e n o Al
2
O
3
ca alys . As he e ec i e a e o he RWGS
eac ion o e Al
2
O
3
is by abou a ac o o 40 lowe compa ed o he Ni ca alys , he eac o
leng h equi ed is much highe compa ed o Ni/Al
12
O
19
ca alys o ge he same con e sion.
2
4
4
Rsyngas = H2/CO ( eac o ou le )
R eed = H2/CO2
Iso he mal
Adiaba ic
T0= 1200 4C
T0= 1000 4C
Resul s and discussion
117
Fig. 5-56: Reac o leng h equi ed o 99% o equilib ium con e sion in adiaba ic eac o
(X
CO2
= 81%) and iso he mal eac o (X
CO2
= 86%) e sus supe icial gas
eloci y o e Ni/Al
12
O
19
ca alys (H
2
/CO
2
= 3, p = 1 ba ).
Fig. 5-57: Reac o leng h equi ed o 99% o equilib ium con e sion in adiaba ic eac o
(X
CO2
= 81%) and iso he mal eac o (X
CO2
= 86%) e sus supe icial gas
eloci y o e Al
2
O
3
ca alys (H
2
/CO
2
= 3, p = 1 ba ).
Resul s and discussion
118
P essu e d op in ixed bed echnical RWGS eac o s
5.4.3
Acco ding o E gun [107], he p essu e loss in a ixed (packed) bed is gi en by
z
B
a
F
e
/
a
7
8
:
$
*
&
P
=
h
1
(5.27)
whe e :
$
is he densi y o he luid (kg/m
3
), /
a
he ic ion ac o o a packed bed, which is
gi en by he ollowing equa ion based on he pa icle Reynolds numbe , 9AF*
&
8
Oc:
/
a
F
2
0
5
¼
a
4
¼
a
u
1
3
[
`
<
2
0
5
¼
a
4
3ZZ
9A
6
1
(5.28)
Fo non-sphe ical pa icles an equi alen pa icle diame e is used
8
F
e
Y
6
8
S
E
#T
h
1
(5.29)
which is he diame e o a sphe e wi h he same ex e nal su ace a ea pe uni olume as he
ac ual pa icle.
Fo a packed bed o sphe es (equal diame e , po osi y ¼
a
= 0.4), Eq. (5.28) simpli ies o
/
a
F
33
<
0 ZZ
9A
1
(5.30)
The pa icle diame e and he luid eloci y ha e a s ong in luence onzB
a
. As we can see
by inse ion o Eq. (5.30) in o Eq. (5.27), 1B
a
is p opo ional o *
&
O
8
P
o low alues o
9Aand o *
&P
O
8
o high alues.
The p oduc ion o syngas by CO
2
hyd ogena ion would mos p obably been done a he
ypical p essu e o he subsequen p oduc ion o liquid uels ia Fische T opsch syn hesis.
Hence, a o al p essu e o 30 ba , a maximum p essu e d op o 1 ba , and an inle
empe a u e o 1200 °C a e assumed o he inal simula ion o a echnical RWGS eac o . In
addi ion, only adiaba ic ope a ion is subsequen ly conside ed, as his ype o ope a ion is
much easie o ealize in a echnical eac o compa ed o iso he mal ope a ion. The luid
densi y and iscosi y a e calcula ed a 1100 °C (a e age empe a u e in adiaba ic eac o ).
Values o he equi ed pa ame e s o calcula e zB
a
a e gi en in Table 5-11. Fig. 5-58 shows
ha he p essu e d op pe me e leng h s ongly inc eases wi h inc easing supe icial gas
eloci y. In case o he Al
2
O
3
ca alys and an adiaba ic ixed bed eac o , he supe icial gas
Resul s and discussion
119
eloci y should be less han 1.7 m/s o limi he p essu e d op o he assumed alue o 1 ba .
The eac o leng h is hen 2.8 m ( o ge 99% o he equilib ium con e sion) (Fig. 5-57). A
his eloci y, zB
a
O7 is abou 0.38 ba /m and so 1B
a
D 1 ba o e he o al eac o leng h. In
case o he Ni/Al
12
O
19
ca alys and a same supe icial gas eloci y o 1.7 m/s, a eac o
leng h o 0.05 m (Fig. 5-56) would be su icien and zB
a
would only be abou 0.02 ba .
The espec i e gas eloci y o he Ni/Al
12
O
19
ca alys , whe e he assumed limi ing alue o
zB
a
o 1 ba is eached, is 6.7 m/s ( eac o leng h o 0.18 m).
Table 5-11: Model pa ame e s used o calcula e he p essu e d op o a RWGS eac o .
Pa ame e s (30 ba , 1100 °C) Values
Kinema ic iscosi y (7), m
2
/s 6 · 10
-6
Fluid densi y (3
), kg/m
3
4.65
Pa icle diame e (d
p
), m 0.006
Pa icle Reynolds numbe (Re) a u
s
= 1 m/s 1000
F ic ion ac o (
b
) a u
s
= 1 m/s 34.7
Fig. 5-58: P essu e d op in ixed bed eac o pe me e leng h o di e en supe icial gas
eloci ies.
23224
2324
234
4
42
422
234 4 42
5pb/L [ba /m]
Supe icial gas eloci y, us[m/s]
T = 1100 4C
p = 30 ba
5pb 31 ba (Al2O3)
5pb 30.02 ba (Ni/Al12O19)
5pb 31 ba (Ni/Al12O19)
Resul s and discussion
120
Es ima ion o he size o he RWGS eac o wi h ega d o a subsequen Fische
5.4.4
T opsch ixed bed eac o
I is in e es ing o es ima e he size o an adiaba ic RWGS eac o (bo h o he in es iga ed
Ni- and he Al
2
O
3
ca alys ) needed o a subsequen Fische -T opsch (FT) eac o . Acco -
ding o he da a gi en by Jess and Ke n [48], he pa ame e s o a echnical FT eac o a e as
ollows:
o The FT eac o is a mul i ubula eac o wi h 8000 ubes each wi h a diame e o 5 cm.
o The o al olume a e o esh (d y) syngas is abou 160,000 m
3
/h (NTP).
o The o al p essu e is a ound 30 ba .
The eac o leng h equi ed o ge 99% o he equilib ium con e sion o CO
2
is calcula ed as
7
mm
½|}
P
#
F
7
*
&
111111111111(i: Ni- o Al
2
O
3
-ca alys )1
(5.31)
And he leng h o each a ce ain zB
a
(in his calcula ion zB
a
= 1 ba ) as
7
z
8
¾
F
=
z
B
a
8
/
a
:
$
*
&
P
1
(5.32)
Fo he es ima ion o he design o he adiaba ic RWGS eac o , he ollowing assump ions
we e made:
o Fo he assumed olume a e o d y syngas o 160,000 m
3
/h (NTP), he olume a e
h ough he RWGS eac o is 201,000 m
3
/h (NTP), because o a CO
2
con e sion o 81%
( eached in adiaba ic ope a ion) 20% o he syngas is wa e , which is sepa a ed. A a
mean empe a u e o 1100 °C and 30 ba , his co esponds o 34,000 m
3
/h (= 9.4 m
3
/s).
o Fo he Ni/Al
12
O
19
ca alys , he esidence ime equi ed o ge 99% o he equilib ium
con e sion o CO
2
is 0.027 s (see Fig. 5-56); he espec i e alue o he Al
2
O
3
ca alys
is 1.65 s (see Fig. 5-57). (Rema k: Po e di usion
is only de e mined by Knudsen
di usion. Hence, he e is p ac ically no in luence o he o al p essu e on he e ec i e
eac ion a e (cons an ) and hus also no on he equi ed esidence ime.)
o Fo he p essu e d op in he RWGS eac o a limi ing alue o 1 ba is assumed.
The dimensions o he ixed bed o he RWGS eac o a e hen es ima ed as ollows.
(a) Ni/Al
12
O
19
ca alys : A a supe icial gas eloci y o 6.7 m/s, he eac o leng h equi ed
o ge he 99% o he equilib ium con e sion o CO
2
is 0.18 m and he assumed p essu e
Resul s and discussion
121
d op o 1 ba is also jus eached (Fig. 5-59). As his leng h is a he small, a leng h o
0.5 m can be assumed o be app op ia e ( o be on he sa e side wi h ega d o bypass
e ec e c.), e en i he p essu e d op hen inc eases o 2.5 ba . The c oss-sec ional a ea
o he RWGS eac o is 1.4 m
2
( a io o olume a e o 9.4 m
3
/s and gas eloci y o
6.7 m/s), he eac o diame e 1.3 m, and he olume o he ca alys is 0.7 m
3
.
Fig. 5-59: Reac o leng h e sus supe icial gas eloci y: dashed-do ed line: leng h, whe e a
p essu e d op o 1 ba is jus eached; lines: leng h equi ed o each 99% o
equilib ium con e sion (X
CO2
= 81%).
(b) Al
2
O
3
ca alys : Now a lowe supe icial gas eloci y o 1.7 m/s has o be adjus ed o
each 99% o he equilib ium con e sion o CO
2
and also a p essu e d op o 1 ba
(Fig. 5-59). The co esponding leng h o he ixed bed is abou 2.8 m, and he c oss-
sec ional a ea 5.5 m
2
(2.6 m diame e ). Inc easing he eac o leng h by 1 m ( o al leng h
o 3.8 m) o be on he sa e side hen co esponds o a olume o he Al
2
O
3
ca alys o
abou 21 m
3
, which is by a ac o o 30 highe compa ed o he Ni-ca alys .
These es ima ions clea ly show ha o bo h ca alys s, he equi ed olume is a he small o
deli e he syngas o a huge echnical FT eac o . The ques ion, whe he he Ni- o he Al-
ca alys should be used is hen a ques ion o he cos s o he ca alys and e en mo e
impo an o he long e m s abili y o he ca alys , which should be de e mined by u he
in es iga ions.
234
4
42
2 42
Reac o leng h [m]
Supe icial gas eloci y, us[m/s]
5pb= 1 ba
Al2O3
5pb= 1 ba
Adiaba ic eac o
5pb= 1 ba
Ni/Al12O19
Zusammen assung und Ausblick
128
In de o liegenden A bei wu de de d i e Sch i , die RWGS-Reak ion, un e such . Dabei
wu de o allem ein aluminiumoxidge äge e Nickelka alysa o (Ni/Al
12
O
19
) e wende .
Da die RWGS-Reak ion endo he m e läu , wu de die Reak ion bei hohen Tempe a u en
du chge üh , um CO (und H
2
O) als Haup p oduk e zu e hal en.
Übe sich zu den expe imen ellen Un e suchungen: Die expe imen ellen Un e suchungen
alle Teilsch i e (Me hanisie ung, Wasse -Gas-Shi bzw. RWGS) wu den in einem Fes -
be eak o un e Ve wendung des oben genann en Nickelka alysa o s bei No mald uck
du chge üh . Neben de Tempe a u wu den auch die Konzen a ionen de Reak an en, die
Ve weilzei und die Ka alysa o pa ikelg öße a iie . Fü jede Reak ion e olg e eine
Bes immung sowohl de in insischen als auch de e ek i en Reak ionskine ik. Neben dem
Nickelka alysa o wu den auch ande e Ka alysa o en wie z.B. Al
2
O
3
ge es e .
Die expe imen ellen E gebnisse lassen sich olgende maßen zusammen assen.
CO
2
-Hyd ie ung: Zunächs wu den e schiedene komme zielle Ka alysa o en un e such
(siehe Tab. 5-4), um ein g undsä zliches Ve s ändnis ü die Ak i i ä de Ka alysa o en ü
die CO
2
-Hyd ie ung zu e hal en. Hie bei zeig e de Ni/Al
12
O
19
-Ka alysa o die höchs e
Ak i i ä . Dahe wu den die wei e gehenden kine ischen Un e suchungen zu CO
2
-
Hyd ie ung mi diesem Ka alysa o du chge üh .
E wa ungsgemäß e höhen sich de CO
2
-Umsa z und die CO-Ausbeu e mi s eigende
Tempe a u (300 – 950 °C). Bis zu eine Tempe a u on 500 °C s eig zunächs auch die
Me hanausbeu e an, um dann au g und de he modynamischen Limi ie ung wiede abzu-
allen. Bei hohen Tempe a u en wu den Umsä ze e ziel , die seh nahe am Gleichgewich
liegen. So konn e beispielsweise bei 950 °C ein Umsa z on 94 % des Gleichgewich s-
umsa zes e hal en we den (lee oh bezogene Ve weilzei : 18 ms). Die Pa ikelg öße wu de
mi wenige als 0,5 mm so gewähl , dass de äuße e S o anspo keinen Ein luss ha . De
Nickelka alysa o zeig e bei 900 °C zudem eine seh gu e Langzei s abili ä . Die Be echnung
de in insischen kine ischen Pa ame e e gab eine Ak i ie ungsene gie on 65 kJ/mol und
einen Häu igkei s ak o on 5,55 · 10
2
m
1,5
mol
0,5
kg
-1
s
-1
. Die Reak ions a e de CO
2
-
Hyd ie ung kann du ch olgenden Ansa z besch ieben we den:
D
E
|
}
~
F
E
|
}
~
|}
~
@
~
1
2
¿
04
Zusammen assung und Ausblick
129
Die Be echnung des S o anspo ein lusses au die e ek i e Reak ions a e de CO
2
-Hy-
d ie ung zeig , dass die expe imen ellen und be echne en E gebnisse gu übe eins immen.
Bei dem e wende en Schalenka alysa o (Du chmesse 3 mm, Schich dicke 0,5 mm) beein-
luss sowohl die Po endi usion als auch de ex e ne S o anspo die e ek i e Reak ions-
a e. Bei den expe imen ell einges ell en ge ingen Gasgeschwindigkei en wi d die Ka aly-
sa o leis ung obe halb on 900 °C zunehmend du ch den ex e nen Massen ans e bes imm .
CO-Hyd ie ung (Me hanisie ung): Die Me hanisie ung is exo he m und somi bei nied igen
Tempe a u en he modynamisch begüns ig . Dahe s eig de CO-Umsa z mi wachsende
Tempe a u an, um ab ca. 450 °C schließlich zu allen. Zwischen 480 und 710 °C deak i ie
de Ka alysa o au g und de hie einse zenden Koksbildung.
Die Reak ionso dnungen wu den bei 340 °C und A mosphä end uck bes imm . Fü CO
e gib sich ein We on -0,3 und ü Wasse s o on 0,7. Die Ak i ie ungsene gie be äg
102 kJ/mol und de Häu igkei s ak o ha einen We on 2,35 · 10
5
m
1,2
mol
0,6
kg
-1
s
-1
.
Die Be echnungen zum S o anspo lassen ü den Schalenka alysa o einen Ein luss de
Po endi usion obe halb on 420 °C e wa en. Die be echne en We e de e ek i en
Reak ions a e weichen alle dings bei hohen Tempe a u en on den expe imen ellen We en
ab. Dies is au die dann au e ende Deak i ie ung des Ka alysa o s zu ückzu üh en.
Wasse -Gas-Shi Reak ion (WGS): Die WGS-Reak ion wu de eben alls un e such , da es
sich hie um die Rück eak ion de CO
2
-Hyd ie ung (RWGS) handel . Die Haup p oduk e
de WGS-Reak ion sind dahe Wasse s o und Kohlendioxid. Diese Reak ion is exo he m
und wi d olglich bei nied igen Tempe a u en he modynamisch begüns ig .
Bis zu eine Tempe a u on 450 °C e höhen sich sowohl de CO-Umsa z als auch die CO
2
-
Ausbeu e. Gleichzei ig konn e in dem un e such en Tempe a u ens e wede eine
Me hanbildung noch eine Deak i ie ung des Ka alysa o s beobach e we den. Im
Tempe a u be eich on 500 bis 700 °C zeig de Nickelka alysa o eine Deak i ie ung, die
du ch eine Ve kokung des Ka alysa o s he o ge u en wi d.
Die Zugabe beide Reak an en (CO, H
2
O) üb einen posi i en Ein luss au die WGS-
Reak ion aus, wäh end (wie zu e wa en) die Beimischung de P oduk e (CO
2
, H
2
) einen
nega i en Ein luss au die Reak ions a e au weis . Die Bes immung de Reak ionso dnung
ü die einzelnen Komponen en e gab: 0,8 (CO), 0,4 (H
2
O), -0,1 (CO
2
) und -0,15 (H
2
). Die
Bildung on Me han and im un e such en Pa ame e be eich nu un e geo dne s a .
Zusammen assung und Ausblick
130
Die Ak i ie ungsene gie de WGS-Reak ion (Ni-Ka alysa o ) be äg 96 kJ/mol. De We
des Häu igkei s ak o s lieg bei 3,46 · 10
5
m
3,6
mol
-0,2
kg
-1
s
-1
. S o anspo be echnungen
ü den Schalenka alysa o zeigen obe halb on 350 °C einen Ein luss de Po endi usion.
Beding du ch eine Ka alysa o deak i ie ung nimm die e ek i e Reak ions a e im
Tempe a u be eich zwischen 500 und 700 °C ab. Obe halb on 900 °C wi d die e ek i e
Reak ionsgeschwindigkei meh und meh du ch den ex e nen Massen ans e bes imm .
Zu Modellie ung eines echnischen Reak o s zu He s ellung on Syn hesegas du ch die
CO
2
-Hyd ie ung wu de ü beide Ka alysa o en (Ni/Al
12
O
19
; Al
2
O
3
) das Modell eines ein-
dimensionalen Fes be eak o s e wende , d.h. die axialen Tempe a u - und Umsa z-P o ile
wu den be echne . Dabei wu de de adiaba e und auch de iso he me Fall be ücksich ig .
Die e ek i e Reak ions a e de RWGS-Reak ion am Ni/Al
12
O
19
-Ka alysa o is im Ve -
gleich zum Al
2
O
3
-Ka alysa o um den Fak o 40 höhe . Fü einen gegebenen Umsa z kann
die Reak o länge ü den Ni-Ka alysa o en sp echend wesen lich kü ze gewähl we den.
Ausblick
Ausgehend on den in diese A bei e hal enen E gebnissen soll en in wei e üh enden
Un e suchungen sowohl ü den Ni-Ka alysa o als auch ü den Al
2
O
3
-Ka alysa o ih e
Langzei s abili ä en un e such we den. Hie zu sind Ve suchszei en on einigen Wochen
anzus eben, um aussagek ä ige Da en zu gene ie en. Wei e hin soll en Tempe a u en on
meh als 1000 °C einges ell we den, da dann de e eichba e CO
2
-Umsa z bei de RWGS
allein schon aus he modynamischen G ünden noch höhe is (> 90%). Hie zu muss die
bes ehende Ve suchsanlage mi einem Reak o , de diesen Tempe a u en s andhäl , sowie
einem en sp echenden Heizsys em ausges a e we den.
Appendix A
131
Appendix A
A.1 U iliza ion o CO
2 in ca aly ic con e sion p ocesses
The de ailed desc ip ion o he syn hesis o alue added p oduc s using CO
2
is gi en in his
sec ion.
A.1.1 Salicylic acid o ma ion
The di ec syn hesis o salicylic acid ia he coupling o CO
2
wi h phenol is a “g een
p ocess”, use ul o he chemical ixa ion o CO
2
. Salicylic acid is widely used in he
indus y as a aw ma e ial and in e media e o he p oduc ion o pha maceu ical and ine
chemicals.
x
j
¡
ij
<
i
P
À
x
j
2
ij
4
iij
1
2g
¿
04
Salicylic acid can be ob ained by ca boxyla ion o phenol wi h a ca bon dioxide in p esence
o Lewis acid ca alys s. AlB
3
as a ca alys showed he bes ac i i y and selec i i y owa ds
salicylic acid among he a ious Lewis acids examined [108]. A 80 °C and 8 MPa, he yield
o salicylic acid eached o 60% wi h he selec i i y o 100% in 1 h. Se e al ypes o base
me al oxides such as alumina, zi conia, ce ia, and alkali me al and alkaline ea h me al sal s
ha e been used o he syn hesis o salicylic acid [109, 110]. Po assium ca bona e as a
ca alys unde op imum condi ion yields 68% o salicylic acid wi h 99% selec i i y. The
sys ema ic in es iga ion o he di ec eac ion o supe c i ical CO
2
and phenol o e he
ca alys s like Z O
2
, TiO
2
, and KF/NaY+Mn
2+
, Z O
2
was in es iga ed [1].
A.1.2 Syn hesis o u ea and u ea de i a i es
The p ocess o u ea syn hesis om ca bon dioxide and ammonia was de eloped in 1922,
and is called he Bosch-Meise u ea p ocess. This non-ca aly ic p ocess s ill exis s and
ope a es a ela i ely high p essu e and high empe a u e.
=
Á
j
u
<
i
P
À
j
P
ÁiiÁ
j
1
2g
¿
=4
j
P
ÁiiÁ
j
À
Á
j
P
i
Áj
P
<
j
P
i
1
2g
¿
34
Appendix A
132
The p oduc ion o u ea in ol es o ma ion o ammonium ca bama e (NH
2
COONH
4
) using
ammonia and CO
2
. Ammonium ca bama e subsequen ly ge s dehyd a ed in o u ea and wa e
by he applica ion o hea . The i s eac ion is as , exo he mic, and basically goes o
comple ion, while he second one is ela i ely slow, endo he mic, and does no go o
comple ion [111]. Bo h o hese eac ions a e e e sible and he e o e ammonia and ca bon
dioxide exi he eac o along wi h ammonium ca bama e and u ea. Thus, he con e sion
based on CO
2
is usually in he o de 50 o 80% [111]. The componen s o his mix u e in he
p oduc s eam a e hen sepa a ed by s ipping o gaseous ammonia ollowed by ca bon
dioxide.
The wo ldwide u ea p oduc ion in 2008 was 146 million me ic ons which co esponds o
107 million me ic ons o CO
2
consump ion pe annum [112]. Hence in he chemical
indus y, only his p ocess has a ce ain po en ial o CO
2
educ ion. U ea is mainly used o
e ilize s bu u ea de i a i es a e used in he a ious o he a eas such as ag ochemicals,
pha maceu icals, an ioxidan s in gasoline, and co osion inhibi o s [113]. Comme cially u ea
de i a i es a e syn hesised by using oxic phosgene as well as ca bon monoxide. In a new
app oach, u ea de i a i es a e p oduced by using amines and CO
2
as a ca bonyl agen ia
ca aly ic and non-ca aly ic p ocesses. The ca bonyla ion o amines using CO
2
is simple, sa e,
and a clean p ocess which a oids he use o poisonous compounds like phosgene and CO
[114, 115]. The eac ion o CO
2
wi h p ima y alipha ic amines also leads o u ea de i a i e
[114]. The p ocess using he ionic liquid [Bmim]Cl wi h CsOH as ca alys yields abou 98%
o an u ea de i a i e [115].
A.1.3 S y ene syn hesis
Comme cially s y ene is p oduced by e hylbenzene dehyd ogena ion using po assium
p omo ed i on oxide as a ca alys wi h a la ge excess o supe hea ed s eam a 600 o 650 °C.
The s eam plays an impo an ole in e hylbenzene dehyd ogena ion as i shi s he
equilib ium owa ds highe con e sions, dec eases he amoun o coke o ma ion and supply
he hea o he eac ion. A new ou e uses ca bon dioxide o he dehyd ogena ion:
Appendix A
133
Con en ional ou e:
1
2g¿p4
New CO
2
-based ou e:
1
2g¿`4
The ca aly ic con e sion o e hylbenzene using CO
2
ha e been discussed ecen ly by Pa k e
al. [116]. The di ec u iliza ion o CO
2
as an oxidan in he dehyd ogena ion o e hylbenzene
o s y ene o e s se e al ad an ages. The main ad an ages shown in di e en s udies [117-119]
a e he accele a ion o he eac ion a e, he d op in empe a u e, he enhancemen o he
con e sion o e hylbenzene and o he selec i i y. Mo eo e , he ene gy consump ion o he
e hylbenzene dehyd ogena ion using ca bon dioxide is much lowe han he cu en ly
ope a ing p ocess using s eam [116].
The dehyd ogena ion o e hylbenzene o s y ene in he p esence o CO
2
was also s udied
o e MnO
2
-Z O
2
[117], SnO
2
–Z O
2
mixed oxide nano-composi e [118], c omia based
ca alys [119], Co, Mo and CoMo ca alys s suppo ed on na u al and aluminium-pilla ed
clays [120], and a CeO
2
–Z O
2
mixed oxide suppo ed SBA-15 ca alys [121]. In hese
ca aly ic sys ems, acidic and basic si es coope a i ely ac i a e CO
2
as well as e hylbenzene,
which lead o a highe con e sion o e hylbenzene. An a emp o use CO
2
as a diluen and
oxidan o e an ac i a ed ca bon-suppo ed i on ca alys ha e made by Sugino e al. [122].
The addi ion o 20 o 30 mol% li hium ni a es o he i on ca alys esul ed in a signi ican
inc ease in he ca aly ic ac i i y.
A.1.4 Dime hyl ca bona e (DMC) syn hesis
The con en ional DMC syn hesis use oxic phosgene. The use o ca bon dioxide in he
syn hesis o DMC p esen s an en i onmen ally iendly and a ac i e app oach since i
eplaces phosgene and chlo ine [123, 124].
Appendix A
134
Con en ional ou e:
i
+
P
<
=
j
u
ij
j
u
iii
j
u
<
=
j+
2g
¿
Y4
New CO
2
-based ou e:
i
P
<
=
j
u
ij
j
u
iii
j
u
<
j
P
i
2g
¿
4
DMC has been syn hesized using CO
2
and me hanol o e solid oxide ca alys s like Z O
2
[125, 126]. The dissocia i e adso p ion o CO
2
occu s as e han he adso p ion o me hanol
on Z O
2
and species o med om me hanol a e bound mo e s ongly. DMC also has been
syn hesized using base ca alys s like K
2
CO
3,
Na
2
CO
3,
Cs
2
CO
3
and CH
3
I as a p omo e [127].
The use o an ionic liquid along wi h CH
3
I as a p omo e o e KOH ca alys shows an
inc ease in he yield o DMC [128]. The u iliza ion o CO
2
o syn hesis o DMC is e y
bene icial because DMC i sel is a unique molecule and can be used in new en i onmen ally
iendly eac ions o eplace some o he en i onmen ally ha m ul p ocesses like ans-
es e i ica ion o DMC wi h phenol ha p oduces me hyl phenyl ca bona e [129].
A.1.5 Me hanol syn hesis
Me hanol is used as an impo an chemical eeds ock o he syn hesis o many compounds
such as ace ic acid, o maldehyde, me hyl e -bu yl e he (MTBE), me hyl me hac yla e
(MMA) and chlo ome hane. The i s indus ial plan o me hanol p oduc ion using
syn hesis gas was cons uc ed by he BASF in 1923. This p ocess was ope a ing unde high
p essu e o abou 20 MPa and a 300 °C using a zinc oxide/ch omium oxide ca alys [5].
Due o he en i onmen al impac , me hanol syn hesis h ough he ca bon dioxide
hyd ogena ion has a ac ed wo ldwide esea ch in e es . The po en ial use o CO
2
o eplace
CO in he me hanol syn hesis is also an e ec i e way o CO
2
u iliza ion [130]. The
me hanol syn hesis om CO
2
has wo possible mechanisms. One is he di ec con e sion o
ca bon dioxide o me hanol (Eq. (A-8)) and he second one is he con e sion o CO
2
o CO
and H
2
O ia he RWGS eac ion wi h subsequen con e sion o CO and H
2
o me hanol
(Eq. (A-9)). Comme cially me hanol has been p oduced om coal and na u al gas
con aining small amoun o CO
2
along wi h CO and H
2
as a eeds ock [131].
i
P
<
3
j
P
j
u
ij
<
j
P
i
1
j
l
Pmn
o
F
5
p
[
`K
q
O
-+
2g
¿
K4
Appendix A
135
i
<
=
j
P
j
u
ij
1
j
l
Pmn
o
F
5
Z
[
q
O
-+
2g
¿
4
The me hanol o ma ion is an exo he mic eac ion, so he modynamically a o ed a low
empe a u es and high p essu es. Du ing he eac ion hyd ogen can be consumed by CO
2
in
e e se wa e gas shi eac ion and educe he me hanol o ma ion. A ecen s udy show ha
a mix u e o a p ope p opo ion o CO
2
and CO o me hanol syn hesis is no only inc eases
he me hanol yield bu also dec ease he appa en ac i a ion ene gy o he eac ion [132].
The p esence o CO
2
could main ain he ac i e coppe si es in he oxida ion s a e o p e en
an o e - educ ion o he ZnO componen o Cu/Zn ca alys s du ing me hanol syn hesis
[132]. Table A-1 shows he da a epo ed o con e sion o CO
2
and he selec i i y o
me hanol using CO
2
o e Cu/Zn/Z O
2
con aining ca alys .
Table A-1: Me hanol syn hesis by CO
2
hyd ogena ion using Cu/Zn/Z O
2
con aining
ca alys .
Ca alys P epa a ion me hod T, °C X
CO2
, % S
MeOH
, % Re .
Cu/Zn/Z O
2
cop ecipi a ion 250 19.4 29.3 [133]
Cu/Zn/Z O
2
cop ecipi a ion 220 21 68 [134]
Cu/Zn/Z O
2
cop ecipi a ion 200 5.9 47.5 [135]
Cu/Zn/Z O
2
u ea ni a e combus ion 240 17 56.2 [136]
Cu/Zn/Z O
2
glycine ni a e combus ion 220 12 71.1 [137]
Cu/Zn/Al/Z O
2
cop ecipi a ion 240 18.7 47.2 [138]
Cu/Zn/Ga/Z O
2
cop ecipi a ion 250 - 75 [139]
The p esence o Z O
2
enhances Cu dispe sion which leads o an imp o ed ca aly ic ac i i y
and s abili y owa ds me hanol o ma ion [133, 136]. The addi ion o an op imum amoun o
me al oxides like Ga
2
O
3
, Al
2
O
3
, Z O
2
, and C
2
O
3
as a modi ie s inc eases he me hanol
syn hesis ac i i y and he s abili y o Cu/ZnO-based e na y ca alys s [135, 138-140].
A no el app oach was employed using low empe a u e o abou 170 °C and 5 MPa p essu e
in a semi-ba h au ocla e eac o wi h 2-bu anol as a sol en [141]. CO
2
con e sion and
Appendix A
136
selec i i y owa ds me hanol ob ained o e a Cu ca alys was abou 26% and 73%,
espec i ely.
Al hough much a en ion has been gi en o he me hanol syn hesis om CO/H
2
, CO
2
/H
2
and
CO/CO
2
/H
2
using ixed bed echnology, some impo an subjec s o in es iga ion a e s ill
unde discussion: (1) whe he he ac i e si e is Cu
0
o Cu
n+
, (2) whe he he ca bon sou ce
o me hanol syn hesis is CO
2
o CO, (3) whe he he e is special in e ac ion be ween he Cu
me al and oxide suppo , and (4) whe he he wa e o med du ing me hanol syn hesis has
any in luence on he ca aly ic ac i i y.
A.1.6 Fo mic acid syn hesis
The syn hesis o o mic acid is ano he al e na i e o con e ca bon dioxide in o aluable
p oduc s. Fo mic acid is widely used as a pickling agen , as a educing agen , as an
an ibac e ial agen , a mo dan in he dyeing indus y, disin ec an and p ese a i e agen in
sani a y s a ions, and as a neu alize in he anning indus y [142]. I has been also used as a
aw ma e ial in he chemical indus y o he p oduc ion o o ma e es e , which is impo an
eeds ock o he syn hesis o a ious o ganic de i a i es like aldehydes, ke ones, amides,
and ca boxylic acids [142]. Fo mic acid has been p oduced di ec ly by he hyd a ion o
ca bon monoxide, and by he hyd olysis o me hyl o ma e. The syn hesis o o mic acid by
di ec CO
2
hyd ogena ion was i s epo ed by Fa low and Adkins in 1935 using Raney
nickel as a ca alys unde 20 o 40 MPa and 80 o 150 °C in p esence o amines [143]. A a
empe a u e o 250 °C, he shee b ass used in ab ica ing line s o s eel eac ion essel ac s
as an ac i e ca alys o he hyd ogena ion eac ion:
i
P
<
j
P
À
jiij
1
2g
¿
0Z4
Mos o he ca aly ic s udies o o mic acid syn hesis a e based he me al complexes o he
second and hi d ow ansi ion me als, whe e he me al like u henium, palladium, and
hodium has been used wi h he combina ion o halides o hyd ides as anionic ligands and
phosphines as neu al ligands [142].
The ca aly ic hyd ogena ion o CO
2
o o mic acid o e ansi ion me al ca alyzed
complexes RuH
2
(PPh
3
)
4
and Pd(Ph
2
PCH
2
CH
2
PPh
2
)
2
, has been epo ed [144]. Simila ly,
o mic acid has been p oduced by ca bonyla ion o CO and H
2
O using a me al complex, i.e.
Appendix A
137
[Ru
II
(EDTA-H)CO]
-
in he WGS eac ion [145]. The highly ac i e me al complex
RuCl(O
2
CMe)(PMe
3
)
4
as a ca alys wi h acidic alcohols was used o enhance he a e o he
hyd ogena ion eac ion. The o ganic bases wi h in e media e basici y we e added o he
eac ion mix u e o ex ac he o mic acid [146]. Al hough much a en ion has been paid o
syn hesize o mic acid using a homogeneous ca alys , he p oblem o sepa a ion o o mic
acid om he ca alys and he base s ill emains.
The hyd ogena ion o CO
2
o o mic acid o e alumina suppo ed u henium hyd oxide as a
he e ogeneous ca alys was ca ied ou using ie hylamine and e hanol as a sol en a 80 °C
and 13.5 MPa [142]. Also he Ru henium(II) complex ca alyzed hyd ogena ion o CO
2
o
o mic acid was heo e ically in es iga ed using cis-RuH
2
(PH
3
)
4
as a model ca alys
[147, 148]. The hyd ogena ion o e his ca alys ook place h ough he inse ion o CO
2
in o
he Ru-H bond ollowed by he H-OCOH educ i e elimina ion, whe e CO
2
inse ion was
he a e de e mining s ep.
A.1.7 CO
2
e o ming o me hane
The e o ming o CO
2
wi h me hane has also a ac ed a con inuous esea ch in e es . This
p ocess was ca ied ou no only o he u iliza ion o wo undesi able g eenhouse gases
(CO
2
, CH
4
) bu also o he p oduc ion o uels by Fische T opsch and me hanol syn hesis.
The d y e o ming o me hane wi h CO
2
is no ye easible, and gi es incomple e con e sion
o CO
2
due o he modynamic cons ain s [149].
D y e o ming:
i
P
<
j
=
i
<
=
j
P
1
j
l
Pmn
o
F
=p
q
O
-+
2g
¿
004
S eam e o ming:
j
<
j
P
i
i
<
3
j
P
1
j
l
Pmn
o
F
=ZY
q
O
-+
2g
¿
0=4
The d y e o ming o CO
2
wi h me hane was i s s udied by Fische and T opsch in 1928.
A e ha many esea che s in es iga ed his eac ion using a ious ca alys s o he
p oduc ion o syngas. Syngas can be p oduced wi h CO/H
2
a io o abou a uni y depending
upon he eac ion condi ions. This is an endo he mic eac ion, gene ally ca ied ou in he
empe a u e ange o 300 o 830 °C and a a mosphe ic p essu e [150]. The g oup VIII
me als (e.g. Ni, Co, P , Pd, I , Ru, Rh e c.) a e mo e o less ca aly ically ac i e o he ca bon