chemical reaction engineering houston

Chemical Reaction Engineering-
Houston
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
Chemical
Reaction
Engineering-
Houston
Vern
W.
Weekman,
Jr.,
EDITOR
Mobil
Research
and
Development
Company
Dan
Luss,
EDITOR
University
of
Houston
The Fifth Internationa l Symposiu m
on
Chemica l Reactio n Engineerin g
co-sponsore d by the America n
Chemica l Society , the America n

Institute
of Chemica l Engineers ,
the Canadia n
Society
for Chemica l
Engineering , and the Europea n
Federatio n of Chemica l Engineering ,
held
at the Hyat t
Regency
Hotel ,
Houston ,
TX, Marc h
13-15,
1978 .
ACS SYMPOSIUM SERIES
65
AMERICAN CHEMICAL SOCIETY
WASHINGTON, D.C. 1978
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
Library
of
Congress
CIP
Data
International Symposium on Chemical Reaction
Engi-
neering,
5th, Houston, Tex., 1978. Chemical reaction

engineering—Houston.
(ACS
symposium
series;
65 ISSN
0097-6156)
Bibliography:
p.
Includes index.
1. Chemical engineering—Congresses. 2. Chemical
reactions—Congresses.
I. Weekman,
Vern
W. II. Luss, Dan, 1938-
III. American Chemical Society. IV
American
Chemical Society. ACS symposiu
TP5.I67 1978
660.2'9'9
77-25340
ISBN
0-8412-0401-2
ACSMC
8 65 1-619
(1978)
Copyrigh t
© 1978
America n
Chemica l Societ y
All

Right s Reserved .
The
appearanc e
of the
code
at the
botto m
of the
first
page
of
each
articl e
in
this volum e indicate s
the
copyrigh t owner' s
consent
that
reprographi c copie s
of
the articl e may
be
mad e
for
persona l
or
interna l
use or for the
persona l

or
interna l
use of
specifi c clients . Thi s
consent
is
give n
on the
condition , however ,
that
the
copie r
pay the
stated
per
copy
fee
throug h
the
Copyrigh t Clearanc e Center ,
Inc. for
copyin g beyon d
that
permitte d
by
Sections
107 or 108 of the
U.S. Copyrigh t Law . Thi s
consent
does

not
exten d
to copyin g
or
transmissio n
by any
means—graphi c
or
electronic—fo r
any
other purpose ,
such
as for
genera l distribution ,
for
advertisin g
or
promotiona l purposes ,
for
creatin g new
collectiv e works ,
for
resale ,
or for
informatio n
storage
and retrieva l systems .
The citatio n
of
trade

names
and/o r
names
of
manufacturer s
in
this publicatio n
is not to be
construe d
as an
endorsemen t
or as
approva l
by ACS of the
commercia l product s
or
service s
reference d herein ;
nor
shoul d
the
mere
reference
herei n
to any
drawing ,
specification ,
chemica l process ,
or
othe r data

be
regarde d
as a
licens e
or as a
conveyanc e
of
any righ t
or
permission ,
to
the holder , reader ,
or
any othe r perso n
or
corporation ,
to
manufacture , repro -
duce,
use, or
sell
any
patente d inventio n
or
copyrighte d wor k
that
may in any way be
relate d thereto .
PRINTED IN THE UNITED STATES OF AMERICA
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;

ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
ACS
Symposium
Series
Robert
F.
Gould,
Editor
Advisory
Board
Kenneth
B. Bischoff
Donald
G. Crosby
Jeremiah P. Freeman
E.
Desmond Goddard
Jack
Halpern
Robert A. Hofstader
James
P. Lodge
John L. Margrave
Nina
I.
McClelland
John B. Pfeiffer
Joseph V. Rodricks
F. Sherwood Rowland
Alan

C. Sartorelli
Raymond
B. Seymour
Roy
L. Whistler
Aaron
Wold
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
FOREWORD
The ACS
SYMPOSIU
a medium for
publishing
symposia
quickly
in book form. The
format of the
SERIE S
parallels
that
of the continuing
ADVANCE S
IN
CHEMISTR Y
SERIE S
except
that
in order to save time the
papers are not typeset but are reproduced as they are sub-

mitted
by the authors in camera-ready form. As a further
means of saving time, the papers are not edited or reviewed
except
by the symposium chairman, who becomes editor of
the book. Papers
published
in the ACS
SYMPOSIU M
SERIE S
are original contributions not
published
elsewhere in whole or
major part and include reports of research as
well
as reviews
since symposia may embrace both types of presentation.
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
PREFACE
A
has, as in
past
symposia, provided
an
excellent forum
for
reviewing
recent
accomplishments

in
theory
and
application.
This
international
symposium
series
grew
out of the
earlier European Symposia
on
Chemi-
cal Reaction Engineering which began
in 1957. In 1966, as
part
of the
American
Chemical Society Industrial and Engineering Chemistry
Divi -
sion's Summer Symposium series,
a
meeting
was
devoted
to
chemical
reaction engineering
and
kinetics.

This
meeting highlighted
the
great
interest and activity
in
this
field
in
the
United
States,
and led
the
orga-
nizers
to
join
with
the
America
European Federation
of
Chemical Engineers
in
organizing International
Symposia
on
Chemical Reaction Engineering.
The

first symposium
was
held
in Washington in
1970
and was followed by symposia in Amsterdam
(1972),
Chicago
(1974),
and Heidelberg
(1976).
These meetings consistently
attract
experts
in the
field
who
have
submitted many more papers than
can be
accommodated.
This
year
was
no
exception with more than
130
papers being submitted, only
48
of which could

be
accepted.
Again,
the
international flavor
was
main-
tained with more than one-half the papers coming from Western Europe,
in
addition
to
one each from Russia, Japan, Australia, and Canada.
While
industrial participation
was not as
extensive
as
anticipated
(30% ), it
did show clearly
the
increasing and productive application
of
Reaction Engineering tools
to
industrial problems.
The
meeting format maintained
three
plenary review

lectures
each
morning
and
three
parallel, original paper sessions
in the
afternoon.
The
nine plenary review papers
are
being published
in the
American
Chemical
Society Symposium Series
as a
separate
volume.
We acknowledge financial support from
the
National Science Foun-
dation,
American Chemical Society-Petroleum Research
Fund,
Shell
Oil
Co.,
Mobi l
Oil

Corp.,
and Exxon Co.
VER N
W.
WEEKMAN ,
JR.
DA N
Luss
Mobile
Research
Corp.
University of Houston
Princeton, NJ Houston, TX
October
1977
xi
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
Organizing
Committee
for the
Fifth
International Symposium on
Chemica
Vern
W. Weekman, Jr., Editor
Dan
Luss, Editor
Members:
Chandler

H .
Barkelew (Shell Development Co.)
K.
B. Bischoff (University of Delaware)
John B. Butt (Northwestern University)
James
M .
Douglas (University of
Massachusetts)
Hugh
M .
Hulburt
(Northwestern University)
Donald
N .
Miller
(Dupont Co.)
xii
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
1
Design
and Operation of a
Novel
Impinging Jet Infrared
Cell-Recycle
Reactor
R.
LEUTE
and

I.
G.
DALLA
LAN A
Department of Chemical Engineering, University of Alberta, Edmonton, Alberta, Canada
In
the
study
of
chemisorbed species on catalyst surfaces,
the
application
of
infrared spectroscopic methods has developed from
the early
in
situ
studies
of
Eischens and
Pliskin
[1] to
rather
detailed surface kinetics
measurements
[5].
The variety of
techniques
which
have been described [1,2,3,4,5,6,7,8] increase

in
their effectiveness
with
their
ability
to
discriminate between
the
spectra
of
adsorbed species
which
are
relevant
to the
reaction
mechanism and spectra
of
spurious adsorbed species. These
approaches may
be
classified using this criterion
as
follows:
(i)
Intrinsic Rates/Surface Spectra Transients Measured
Directly.
Under reaction conditions where adsorbed
reactants,
intermediates, and products display significant IR absorption

band intensities,
the
transient intensities may
be
quantita-
tively
monitored. Considerable detailed studies
are
required
to
correlate
these
intensities
with
surface concen-
trations.
(ii)
Global
Rates/Surface Spectra Static
or
Transient.
By
carrying
out studies
in an
IR
cell
-
circulation
flow

reactor,
a
cause-and-effect relation between
reactant
concentration
and specific band intensities may
be
discerned. Such
mechanistic insights may
be
useful
in
developing more
reliable
forms
of
rate
expressions.
(iii)
Indirect Studies
of
Adsorption and Surface Reactions.
The
observation
of
selected spectral band intensities attributed
to chemisorbed species
are
assumed
to be

related
to the
surface reactions
involved.
If the
spectra
are
recorded
at
room
temperature,
the
presence
of
spurious spectra may occur.
Generally,
additional experimental evidence
is
required
to
demonstrate
the
relevance
of
such observations
to the
kinetics
of
the
catalytic reaction.

©
0-8412-0401-2/78/47-065-003$05.00/0
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
4
CHEMICA L
REACTIO N
ENGINEERING—HOUSTO N
Thi s
pape r
describe s
the
developmen t
of an
improve d
versio n
o f
the IR
cell-recycle
reacto r (typ e
(ii) )
whic h
is to be
use d
to
stud y
the
mechanis m
and
kinetics

of
reaction s
of
2-propano l
on
variou s
alumin a
catalysts .
Whil e
thi s
reactio n
doe s
not
hav e
direct
commercia l
implication s (dehydratio n
or
dehydrogenation) ,
it
exhibit s
many of the
characteristics
whic h
make it
ver y
suitabl e
to
demonstrat e
the

usefulnes s
of the IR
technique .
Desig n
Factor s
Th e yin
AAXU
techniqu e involve s catalys t
pellet s
in the
for m
o f ver y
thi n
wafers ,
abou t
40
mg/cm2
alumin a
content .
The
hig h
surfac e area ,
abou t
4
m^/cm^-
of IR beam
cross-section ,
enable s
sufficien t
adsorbe d

specie s
to
interac t
wit h
the IR beam
eve n
at
relativel y
low
surfac e
coverage
tha t
spectr a wit h
goo d
I n studyin g solid-catalyze d
gas-phas e
reactions ,
the
back -
groun d
spectr a
resultin g
fro m
the
gas-phas e
are
usuall y eliminate d
b y
use of a
double-bea m

IR
spectrophotometer ,
i n
whic h
the
sampl e
cel l
is
matche d
wit h
an
"identical "
referenc e
cel l
withou t
catalys t
in
it .
Variation s
in
pressur e
and/o r
temperatur e
betwee n
sampl e
and
referenc e
cell s
increas e
the

difficult y
of
matchin g
the
tw o
cells .
When
the
catalys t
wafe r
is
place d transvers e
to the
flo w
of
gase s
throug h
the IR
cell-reactor ,
the
flo w pattern s
withi n
the
cel l
lea d
to
concentratio n gradient s alon g
the
axi s
of

th e
IR beam, and
betwee n
the
fron t
and
rea r surfac e concentration s
o n
the
wafer .
Unde r
reactio n conditions , thes e aspect s
limi t
the
sensitivit y
of the
techniqu e
becaus e
of low
surfac e
coverage s
at
reactio n
temperatures .
The new
cel l
attempt s
to
eliminat e
many of

thes e objectionabl e features .
Figur e
la
describe s
a
typica l
geometr y
fo r previou s
cel l
designs .
It
shoul d
be
eviden t tha t i t
is
difficul t
to
obtai n
value s
of the
intrinsi c
reactio n rat e
becaus e
of the
uneve n
contactin g
betwee n
the gas and
wafe r
at

variou s point s
on the
wafe r
surface .
Hig h
recirculatio n
rate s withi n
suc h
a
steady -
stat e
recycl e reacto r provid e
differentia l
value s
of the
reactio n
rate ,
but
thes e globa l value s
are
unlikel y
to
equa l
intrinsi c
rate s
(neglecting ,
for the moment,
intraparticl e
diffusion) .
Compatibilit y

of
flo w pattern s
betwee n
the IR
cel l
and an
idea l
continuou s stirred-tan k reacto r
are
require d
as a
minimu m
condition .
Sinc e
the
mode
of
heatin g
the
wafe r
likel y
involve s
IR-transparen t
window s
bein g
at
temperature s
lowe r
tha n
thos e

of
th e
wafer ,
compensatio n
fo r
temperatur e
gradient s
may
als o
be
required .
Figur e
lb
describe s
the
propose d
geometr y
of the
improve d
IR
cell-reactor .
Thi s recycl e reacto r
is to be
capabl e
of
bein g
operate d
in
eithe r
ope n

(flow )
or
close d (batch )
modes of
operation .
The
reacto r uni t i s maintaine d
at the
reactio n
temper -
atur e
(up to 400°C) and the
pump
and
samplin g
syste m
are
maintaine d
a t
a
constan t usuall y lowe r
temperatur e
(220°C) to
ensur e
maximum
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
NEW
IR REACTOR CELL
OLDER TYPE

IR REACTOR CELLS
(b) (a)
Figure
1.
Geometrical
arrangement
and
flow
patterns
in
typical
and
improved
ir
cell-reactors
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
6
CHEMICA L
REACTIO N
ENGINEERING—HOUSTO N
longevit y
o f equipment .
Figur e
2
describe s
th e
informatio n
flo w
betwee n

th e I R
spectrophotomete r
an d a n
IBM/180 0
compute r syste m whic h ar e
interfaced .
Th e
spectra l
dat a ar e monitore d a t
wave
numbe r
interval s
a s lo w a s 0. 2 cm
- 1
ove r th e complet e
spectra l
sca n rang e
o f
th e spectrophotomete r (abou t 70 0 t o
400 0
cm""
1
,
correspondin g t o
a
maximum
o f abou t
16,00 0
dat a
points) .

Th e " %
transmission "
versu s
"wave
number "
point s
ar e
transmitte d
i n
digitize d
for m t o
th e
compute r fro m absolut e encoders . A t present , th e complet e
spectra l
sca n ma y b e monitore d an d
store d
i n a
dis k
fil e
an d
retrieve d
a t a
late r
time . Th e couple d Mode l 62 1 spectrophoto -
mete r
wit h
IBM/1800-compatibl e
interfac e
wa s purchase d
some

tim e
ag o fro m Perkin-Elmer .
Th e improve d
cel l
utilize s
axisymmetri c
jet s
o f fee d ga s
impingin g
upo n
bot h
side
len t
fiel d
ove r
mos t
o f
reactio n
rate s
t o approximat e
intrinsi c
reactio n
rate s
a t hig h
flow-rate s
an d i n th e absenc e o f por e
diffusion .
Th e ne w
configuratio n
show n

i n
Figur e
1 i s house d i n a n oven -
typ e
enclosur e
controlle d
a t th e temperature , T3 , b y
interna l
ai r
circulation .
I n
additio n
t o th e ove n heater , a secon d heate r abou t
th e
inle t
section ,
packe d
wit h
glas s
beads ,
raise s
th e
circulatin g
ga s temperatur e fro m th e reduce d temperatur e i n th e
pump
compart -
ment
t o Tj . Becaus e o f hea t
losse s
fro m th e I R

windows ,
th e
tem -
peratur e
difference ,
T
3
-T
2
,
coul d
rang e a s hig h a s
50°C
Thi s no t
onl y
change s th e
densit y
o f th e
flowin g
ga s bu t
als o
result s
i n a
considerabl e
deviatio n
o f th e
tru q
temperatur e o f th e
catalys t
wafe r fro m th e

measure d
value s T
2
.
Additiona l
heater s place d
aroun d th e
end s
o f th e tw o
cylindrica l
section s
compensate d fo r th e
windo w
hea t
losses .
I n
thi s
way , th e temperatures , T
2
an d T3 ,
coul d
b e
matche d
withi n
0.5°C,
an d th e
wal l
temperatur e woul d b e
expecte d
t o

diffe r
fro m T
2
(o r T3 )
onl y
i f th e
catalyti c
reactio n
exhibite d
sever e therma l
effects .
Wit h
greatl y
improve d
mass
transfe r
rate s
norma l t o th e wafe r
surface ,
on e woul d
als o
expec t
fro m
similarit y
consideration s
enhance d hea t
transfe r
betwee n
th e
wafe r

surfac e
an d th e impingin g ga s
jet .
Suc h adjustment s
among
th e
thre e monitore d temperature s enable d th e referenc e
cel l
I R
beam
t o
compensat e
nearl y
exactl y
fo r th e sampl e
cel l
ga s phas e
absorptio n
spectra .
By changin g th e
configuratio n
o f th e tw o
cell s
i n th e sampl e
compartmen t
o f th e I R spectrophotomete r
thi s
enable s th e
deter -
minatio n

o f
eithe r
recirculatin g
ga s compositio n o r
plottin g
o f th e
baselin e
spectru m fo r th e
catalys t
wafer . Wit h th e tw o
cell s
i n
th e
double-bea m
mode,
th e
catalys t
baselin e
an d
surfac e
spectr a
ar e
recorded .
I f th e referenc e
cel l
wa s place d i n th e sampl e
beam
an d
an ai r ga p i n th e referenc e
beam,

quantitativ e
absorptio n
spectro -
scop y wa s
possible .
Th e I R
cell s
thu s provid e
informatio n
leadin g
t o
bot h
reactio n
rate s
an d mechanisti c
insight s
concernin g adsorbe d
specie s
a t
reactio n
conditions .
When
use d a s a
recirculatin g
batc h
reactor ,
th e spectrophoto -
meter-compute r
interfac e
ca n monito r bu t no t

recor d
th e " %
trans -
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
LEAUT E
AND
DALL A LAN A
Infrared
Cell-Recycle
Reactor
Linear
Encoder
Shaft
Encoder
IR
Spectra
Source
~T T
X|
lY
I I
i_t
X-Y
Recorder
X = Wave Length (abscissa)
Y
= Transmittance
(ordinate)
X

I
= analog signal, Wave Length, 5 digits
Y
1
= analog signal, Transmittance, 3 digits
Y1
I
Encoder
Readout/
I
nterf
ace
X Display
Y
Display
i
l
Figure
2.
Information
flow
between
ir
spectrophotometer
and
digital
computer
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
8

CHEMICA L
REACTIO N
ENGINEERING—HOUSTO N
mission " a t a
fixe d
"spectra l frequency " (usuall y tha t o f a speci -
fie d
absorptio n
band) .
A t present , th e
dru m
char t o n th e I R
recorde r plot s th e
tim e
- absorptio n
ban d
intensit y
relatio n
cor -
respondin g t o transien t reactio n conditions . Th e
tim e
constan t o f
th e
spectrophotomete r
thermocoupl e
senso r wa s
sufficientl y
smal l
tha t
th e transien t reactio n rate s coul d b e recorded .

Experimenta l
Performanc e
1.
Mas s
Transfe r
Performanc e
test s
wer e
designe d t o
tes t
fo r
micromixin g o r fo r
mass
transfe r
performanc e
an d thus , t o
facili -
tat e
definitio n
o f th e
cel l
desig n
specifications .
Limite d reac -
tio n
dat a ha d
bee n
recorde d fo r th e 2-proposa l reactio n
ove r
alumina .

Figur e 3
summarize
wer e
observe d
i n a protptyp
wafe r
material . Ai r flow s
betwee n
1 0 an d 5 0
t/m±n
wer e
passe d
throug h
th e
cel l
an d th e correspondin g sublimatio n rates ,
mg/min,
wer e
recorded . Sinc e th e
cel l
geometr y
wa s hel d constan t fo r a
serie s
o f flo w rate s an d th e
temperature s
wer e
alway s
a t
roo m
temperature ,

th e coordinate s o f Figur e 3
show
th e
measure d
subli -
matio n
rate s versu s flo w rat e rathe r
tha n
Reynold s
number .
Th e
exponen t
o f th e flo w
paramete r
(give n b y th e slop e o f th e
line )
i s
see n
t o
remai n
nearl y constan t
ove r
a
wid e
rang e
o f condition s
verifyin g
tha t th e turbulen t flo w
regim e
i s maintained . Th e

influenc e
o f
changin g
th e
orific e
siz e
use d
t o creat e th e
jets ,
an d
o f th e spacin g
betwee n
th e
orific e
an d th e
wafe r
upo n
mass
transfe r
rate s
ar e als o
shown .
I n additio n t o th e
abov e
test s wit h th e ne w design ,
mass
transfe r
rate s
wer e
als o

observe d
fo r
cell-reactor s
o f th e ol d
type , wit h
wafer s
positione d
bot h
paralle l
an d transvers e t o flows .
Thes e
test s
sugges t
tha t i n
suc h
geometrie s
much
o f th e
strea m
bypasse s
th e
wafe r
surfac e
makin g
i t
difficul t
t o obtai n
intrinsi c
rate s
o f reaction .

Furthermore ,
contactin g o f th e ga s flo w wit h
localize d
portion s o f th e peripher y o f th e
wafe r
resulte d i n
abnormall y hig h
loca l
mass
transfe r rates . Figur e 3
demonstrate s
tha t
ol d typ e
cel l
design s provid e
mass
transfe r
performanc e
inferio r
t o tha t
observe d
wit h th e impingin g
jets .
By
calculatin g
mass
transfe r
coefficient s
fro m
th e usua l

equatio n fo r th e rat e o f sublimation ,
gmol/(min)( g
catalyst) .
r
= k a ( C . -C )
s g surfac e o
an d usin g th e bul k ga s
phas e
concentration , C =0 , an d externa l
area , a=1 0 cm
2
,
some
experimenta l
coefficient s
coul d b e
compare d
t o
value s estimate d
fro m
publishe d correlations .
Tabl e
1
show s
thes e
results .
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
Tabl e 1
Compariso n

o f
Mass
Transfe r
Coefficient s
(cm/sec )
Mode l
Flo w
= 1 0 l/mi n
Flo w
= 5 0 l/mi n
Fla t
plat e
i n 1. 3
cm/se c
2. 9
cm/se c
perpendicula r
flo w
Spher e
o f equa l are a 2. 1 5. 4
i n
a
packe d
be d
Experimenta l wafer , 1. 6 3. 0
conventiona l
geometr y
Experimenta l wafer , 3. 7 9. 2
ne w
geometr y

Tabl e 1 an d Figur e 3 bot h
illustrat e
th e
marke d
superiorit y
o f th e
ne w I R
cell-reacto r
desig n i n promotin g
mass
transfe r
a t th e
wafe r surface .
However ,
i t
stil l
remain s
t o b e demonstrate d tha t
unde r
reactio n
conditions ,
intrinsi c
rate s o f
reactio n
ma y b e
obtaine d
a t th e flo w rate s
mentioned .
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

10
CHEMICA L
REACTIO N
ENGINEERING—HOUSTO N
2. Mixin g Withi n
Cel l
Th e analysi s o f
performanc e
withi n a
differentia l
bed-recycl e reacto r i s usuall y
compare d
t o tha t o f a
continuou s stirred-tan k reactor . B y operatin g th e reacto r wit h a n
iner t
wafe r
an d b y introducin g alcoho l t o th e fee d a s a ste p
chang e
i n concentration , th e mixin g
performanc e
o f
thi s
reacto r
ma y b e
compare d
t o tha t predicte d fo r a n
idea l
CSTR
o f
comparabl e

volume .
Figur e 4
illustrate s
suc h
a
compariso n
an d indicate s
substantia l
agreemen t
wit h th e
idea l
behaviour . I t ma y b e
expecte d
tha t
channelling , stagnatio n o f
some
flow , etc . ar e
absen t
fro m
th e recycl e reacto r withi n th e
rang e
o f
performanc e
o f th e
pump.
3.
Double-bea m
Compensatio n
fo r Ga s
Phas e

Absorptio n
When
recordin g I R spectr a a t reactio n
temperature ,
th e I R
beams
ar e
attenuate d b y th e
numbe r
o f
molecule s
i n th e
beam
path . Sinc e
th e ga s
phas e
populatio n i s
likel y
onl y on e o r tw o order s o f
magnitud e
greate r
tha n
th
th e
wafe r
surface , i t i
uatio n i n th e tw o
cell s
b e balance d a s wel l a s possible . Fo r
example ,

a pressur e
dro p
betwee n
th e tw o
cell s
necessitate s heatin g
th e
upstrea m
cel l
t o
reduc e
it s ga s densit y t o tha t i n th e
down-
strea m
cell .
Similarly ,
difference s i n
temperatur e
betwee n
th e
cell s
mus t
als o b e
compensated .
Suc h
inbalance s
betwee n
referenc e an d
sampl e
cel l

ga s
phase s
require d
calibrations^t o
determin e
th e value s o f T ^ required , fo r
a
fixe d
valu e o f T
2
( = T3 ) an d give n
circulatio n
rat e a t variou s
isopropano l concentration s i n th e
gas-phase ,
t o blan k ou t ga s
phas e
absorptio n spectra . Figur e 5
show s
ho w spectra l
band s
i n
th e
1200-150 0
cm
- 1
regio n
fro m
ga s
phas e

isopropano l ca n b e altere d
b e
changin g
Tj .
Curv e
B represent s near-extinctio n o f th e
back -
groun d
wherea s
curve s A an d B represen t
under -
an d
over-compensa -
tion ,
respectively .
4. Dehydratio n o f
2-Propano l
ove r
Alumin a
Th e preliminar y
measurement s
o f spectr a fo r
adsorbe d
specie s
wil l
b e
use d
t o
illustrat e
ho w th e

mechanis m
o f reactio n ma y b e
clarified .
Th e
mai n
featur e o f th e I R
cell-flo w
reacto r i s it s
capabilit y
o f
determinin g spectr a a t reactio n conditions .
Mos t
publishe d
wor k
o n th e dehydratio n o f isopropano l b y
alumin a
describe s Zn
A^Lta
studie s
wit h spectr a recorde d wit h th e
cel l
a t
roo m
temperature .
Figur e 6 reveal s absorptio n
band s
i n selecte d region s o f th e
spectru m
fo r severa l concentratio n
level s

o f isopropano l
vapour .
Eac h
curve , A , B , o r C , represent s a spectra l
sca n
a t steady-stat e
reactio n
condition s wit h al l reactio n
parameter s
excep t
fee d
compositio n o f isopropano l bein g
kep t
constant . I f
differen t
curve s (A , B , an d C )
result ,
th e
adsorbe d
specie s associate d wit h
th e spectr a ar e considere d t o b e
german e
t o th e reactio n
mechanism .
I n th e
even t
tha t th e spectra l
band s
d o no t
chang e

th e
adsorbe d
specie s ar e considere d t o b e spurious .
Subsequently ,
th e reacto r
ma y b e operate d i n a batc h
mode
an d th e questionabl e
ban d
moni -
tore d
continuously . Th e
failur e
o f
thi s
ban d
t o
chang e
wit h th e
exten t o f reactio n
woul d
provid e extr a suppor t t o th e
vie w
tha t
th e
ban d
i s associate d wit h a b y produc t specie s no t involve d i n
th e dehydratio n
mechanism .
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;

ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
LEAUT E
AN D DALL A LAN A
Infrared
Cell-Recycle
Reactor
Figure
4.
Comparison
between
ideal
CSTR
and
improved
cell-reactor
to
step
change
in
input
concentration
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
12
CHEMICA L REACTIO N ENGINEERING—HOUSTO N
110
100
c
|
80

(0
70
60
1
1 1
1 1 1
T2
=
287.8° C
-
Alcohol
5.5%
A:
T1 =
226.5°C
B:
T1=251.5°C
C:
T1 =
273.5°C
-
-4
1
u
1 1
\J
I I I
1500
1400 1300 1200 1100
Frequency, cm"

1
1000
Figure
5.
Compensation
of
gas-phase
adsorbance
between
ref-
erence
and
sample
cells
Catalyst
Weight
= 0.151 g
Temperature
=
246.1°C
Baseline
without
alcohol
Alcohol
Partial
Pressure
= 2.1
cmHg
Alcohol
Partial

Pressure
= 3.2
cmHg
Free OH Groups CH
3
Stretching Low Frequency Region
3800 3600 3000 2800 1600 1400
Frequency ,
cm"
1
Figure
6.
Steady-state
spectral
scans
for
dehydration
of
iso-
propanol
at
reaction
conditions
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
1.
LEAUT E
AN D
DALL A
LA N A

Infrared
Cell-Recycle Reactor
13
Th e steady-stat e spectra l
scan s
when
recorde d
on
th e
IBM/100 0
ma y
be
processed .
(i )
to
subtrac t th e baselin e
of
th e catalys t
wafe r
fro m
eac h
spectra l
sca n
a t varyin g
partia l
pressure s
of
th e isopro -
panol ;
(ii )

to
subtrac t
one
spectra l
sca n
a t
(P - ,
-) i
fro m
anothe r
o f th e
chang e
i n
ban d
intensitie s
at
give n
ban d
frequencies .
A preliminar y interpretatio n o f th e spectr a
show n
i n Figur e
6
woul d
sugges t
th e followin g observations .
The
fre e hydroxy l
group s
on

th e surfac e
of
alumin a
progressivel y disappear ,
A
t o
B
t o
C,
wit h increasin g reactan t concentration , isopropanol . Thi s
implie s
tha t th e alcoho l
hydroge n
bond s
to
thes e hydroxy l
site s
bu t i t i s no t clea r
whethe r
th e alcoho l
0 or H
ato m
i n it s
hydroxy l
grou p
i s involved
Th e stretchin g vibration
pano l als o displa y
direc t
correspondenc e

betwee n
thei r
surfac e
concentratio n an d tha t
of
th e isopropano l
vapou r
concentration .
Thi s informatio n
suggest s
tha t isopropano l adsorptio n
on
yalumin a
involve s
more
tha n
on e adsorptio n
band ,
i.e .
bot h
hydroxy l
and
emthy l
group s
ar e
bonde d
an d
likel y
t o
differen t

site s
on
th e
surfac e
of
alumina .
I n th e lo w frequenc y region , regio n
I
relate s
to
carbo n
chai n
skeleta l
vibration s an d regio n I I t o symmetrica l
C-H
deformatio n vibration s i n th e
methy l
group .
Bot h
of
thes e
observation s ar e i n accor d wit h
a
multi-sit e
adsorptio n
model .
Regio n
II I
show s
th e stretchin g

vibratio n
fo r
a
carboxylat e specie s
forme d
on
th e surface . Sinc e th e
ban d
intensitie s
i n regio n II I
do no t
chang e
wit h isopropano l
vapou r
concentration , th e spectr a
ar e considere d
incidenta l
t o th e reactio n
mechanism .
Wit h
JC Q
additiona l
experiments , i t shoul d
be
possibl e t o distinguis h
whic h
surfac e
site s
on
th e

alumin a
ar e
specificall y
involve d an d
thu s
to
propos e
a
reactio n
mechanis m
compatibl e wit h
suc h
chemica l
evidence . Durin g th e
abov e
spectra l
measurements ,
steady-stat e
reactio n
rate s i n th e
recirculatio n
reacto r
wer e
als o
determined .
Thes e
rate s
may
the n
be

use d
t o
tes t
th e
kineti c
mode l
resultin g
fro m
observation s
of
th e spectr a
of
adsorbe d
species .
1.
The
us e
of a
"single-wafer
1 1
catalyti c
recycl e reacto r
syste m
require s
stric t
attentio n t o operatin g
parameters ,
i f
one
aspire s

t o obtai n
intrinsi c
rate s
of
reaction .
By
modifyin g
th e flo w pas t th e
wafe r
t o
ensur e
highl y turbulen t condition s
o n
bot h
side s
of
th e
wafer ,
mass
transfe r rate s
may be
more
tha n
double d
ove r
thos e
observe d
i n th e ol d desig n
of
cell s

in
whic h
flo w i s transvers e
to
th e
wafe r
surface . Thi s indicate s
tha t
th e
utilizatio n
o f
bot h
side s
of
th e
wafe r
i s greatl y
improve d
an d tha t th e
averag e
mass
transfe r rate s ar e als o
enhanced .
spectra l
sca n
at
Comments
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
14

CHEMICA L
REACTIO N
ENGINEERING—HOUSTO N
2 . Idea l
mixin g
(CSTR )
i s obtaine d wit h th e recirculatin g rate s
availabl e
fro m
th e
bellow s
pump
use d
t o
thi s
system .
Th e
correspondin g
residenc e
tim e
distributio n functio n i s no t o f
valu e
i n th e analysi s o f th e kinetic s sinc e i t i s anticipate d
tha t non-linea r rat e
expression s
wil l
b e
encountered .
3 . Th e usefulnes s o f a
combine d

IR-kinetic s
stud y
i n establishin g
a
more
reliabl e
kineti c
mode l
i s
apparent .
Th e processin g o f
suc h
dat a
t o ascertai n
whic h
spectra l
band s
ar e significan t
i s
usuall y a
ver y
tediou s
chore .
B y interfacin g th e I R
spectrophotomete r
t o a
digita l
computer ,
a
numbe r

o f
dat a
processin g simplification s ar e evident .
Ful l
us e o f
thi s
situatio n
ha s no t ye t
bee n
attaine d i n
thi s
program .
Whethe r
o r no t
improve d
resolutio n o f
mino r
spectra l
band s
result s
fro m
a n onlin e
compute r
facilit y
stil l
remain s
t o b e
demon-
strate d
fo r

thi s
reactio
4 . Th e
proble m
o f
isolatin
reactio n rate s
measure d
i n a single-wafe r reacto r
appear s
t o
hav e
bee n
reduce d
bu t no t necessaril y solved . I f
relativ e
intensitie s o f absorptio n
band s
exhibite d b y
reactant s o r reactio n intermediate s ca n b e ascertaine d a s a
functio n o f
time ,
i t ma y b e possibl e t o
chec k
rat e
expression s
base d
upo n
a singl e ste p
bein g

rate-controlling .
5 .
Many
extension s
o f
thi s
techniqu e
(usin g th e ne w reactor ) ar e
eviden t i n th e
stud y
o f
catalyti c
kinetics .
Some
aspect s
wort h
pursuin g
include :
(i )
a
stud y
o f
por e
diffusio n
unde r
controlle d conditions ;
varyin g
wafe r
thicknes s a t
constan t

porosit y
shoul d
provid e a direc t
means
o f calculatin g th e effectivenes s
facto r
a s a functio n o f
wafe r
thickness .
(ii )
th e rol e o f trac e
amount s
o f catalys t
promoter s
o r
inhibitor s
ma y b e
examine d
usin g I R
technique s
an d
correlate d
directl y
wit h steady-stat e reactio n rates .
Acknowledgement s
Financia l
suppor t
o f
thi s
projec t b y th e Nationa l

Researc h
Counci l o f
Canad a
i s gratefull y
acknowledged .
Literature Cited
1 .
Eiechens ,
R.P. ,
Pliskin,
W.A. ,
Advan .
Cata. ,
(1957) ,
9 , 662 .
2 .
Heyne ,
H. ,
Tompkins ,
F.G. ,
Proc .
Roy . Soc. ,
(1966) ,
A292 ,
460 .
3 .
Baddour ,
R.F. , Modell , M. , an d
Goldsmith ,
R.L., J .

Phys .
Chem.
(1968) ,
72 , 3621 .
4 .
Dent ,
A.L. , an d
Kokes ,
R.J. , J .
Phys .
Chem,
(1970) ,
74 ,
3653 .
5 .
Tamaru ,
K. , Onishi , T. ,
Fukada ,
K. ,
Noto ,
Y. ,
Trans .
Farada y
Cos. ,
(1967) ,
63 ,
2300 .
6.
Thornton ,
R. , Ph.D . thesis , Universit y o f

Delawar e
1973 ,
7. Shih , Stuar t
Sha n
San , Ph.D . thesis ,
Purdu e
University , 1975 .
8 .
London ,
J.W. ,
Bell ,
A.T. , J . Cat. ,
(1973) ,
31 ,
36-109 .
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
2
Performances
of
Tubular
and
Loop
Reactors in
Kinetic
Measurements
GERHAR D
LUFT ,
RAINER
RÖMER,

and
FRITZ
HÄUSSER
Institut
für
Chemische
Technologie
der
Technischen
Hochschule
Darmstadt,
61
Darmstadt,
Petersengstrasse 15,
West
Germany
Industrial
reactor
terogenous
catalyti
sitive if the reaction conditions or the cooling
rates
are suddenly changed.
They
can be operated only in a
small
range in order to avoid damage to the apparatus
or
to the catalyst by super heating,
also

to avoid
loss
in
yield by
side
reactions, favoured at high tempera-
tures.
In
addition, poor accuracy in the rate data, as well
as in the
mass
and
heat
transfer parameters,do not allow
to calculate the
exact
concentration and temperature
profiles inside the reactor.
This
leads
to incorrect
prediction
of the reactor's dynamic behaviour.
There-
fore
these
data should be determined as accurately as
possible.
For
the measurement of reaction rates, differential

reactors having extremely short catalyst
beds
or inte-
gral
reactors with relative long catalyst
beds
are
often used. In the first type of experimental reactor,
the concentration and temperature gradients within the
catalyst
beds
are negligibly small. Due to this fact,
the reaction rate point data can be measured, provided
the small concentration differences can be accurately
analyzed.
In the integral reactor, the change in con-
centration is much higher. There is in general no dif-
ficulty
analyzing the concentrations of the reacting
species
but, the reaction
rates
have to be determined
from
the concentration curves by calculation and cannot
often be related to the
fast
changing temperature. Be-
cause
of

these
obvious disadvantages, the so-called
loop reactors are being used more and more in kinetic
studies. In loop reactors, the extremely small concen-
tration
and temperature gradients desired within the
short catalyst bed, along with sufficiently high con-
centration difference
between
the reactor inlet and the
©
0-8412-0401-2/78/47-065-015$05.00/0
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
CHEMICA L REACTIO N ENGINEERING—HOUSTO ^
Integral
Tubular Reactor
Differential
Reactor
Loop
Reactor Stirred-Tank Reactor
Figure
1.
Types
of
laboratory
reactors
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
2.

LUF T
ET
AL.
Tubular and Loop Reactors
17
outlet ,
can be
realize d
by
recyclin g
a
par t
of the
reac -
tio n
products .
In
orde r
to see how
thes e advantage s
coul d
be
reali -
zed
in
practice ,
the
performanc e
of a
loo p

reacto r
was
compared
wit h
tha t
of a
conventionally-buil t
integra l
reactor.I n
thi s
compariso n
the
capabilit y
to
handl e
actua l
industria l
catalysts ,
the
settlin g
tim e
of
chan -
gin g
experimenta l
conditions ,
the
difficult y
of the
mathematica l

evaluatio n
of the
measured
dat a
were
con -
sidered .
The
accurac y
of the
data s for
scal e
up
prob -
lems
was
checke d
in a
pilo t
plant .
For the
reaction ,
the
oxidatio n
of
o-xylen e wit h
a
vanadiumpentoxid e
ca-
talyst ,

an
industriall y
importan t process ,
was
chosen .
Apparatu s
The desig n
of th
changes
in the
reactio
re ,
concentratio n
and
throughpu t
in a
wide
range .
Its
cor e
is a
differentia l
reacto r
directl y
couple d
to the
blower .
It
suck s
the

reactant s
throug h
the
catalys t
bed
and
recycle s
par t
of
it .
Thi s
desig n allow s
onl y
a
smal l
dea d
volume
and a
smal l
pressur e dro p acros s
the
catalys t
bed
even
at
high flo w
rates .
Furthermore ,
the
whole

apparatu s
is
compact
and
therefor e
it can
easil y
be maintaine d
at
constan t temperature .
The
smal l
tem-
peratur e
and
concentratio n
gradient s
withi n
the
catalys t
bed, necessar y for
the
kineti c
measurements,
can^be
realize d
by
recyclin g
par t
of the gas

abou t
12 m
/h.
It
i s
ver y
larg e
compared
to the
fee d
and
correspond s
to
recycl e
ratio s
of loo to 5oo,
als o
sufficien t
for
the
appropriat e
stud y
of
highly-exothermi c
reactions .
The
recycl e
rati o
can be
changed

wit h
respec t
to the
reactio n
condition s
by
changin g
the
spee d
of
rotatio n
of
the
blower .
The blowe r
is
drive n
by an
asynchronousmoto r
whose
roto r
is
fixe d
to the
shaf t
of the
blower .
It is
sepa -
rate d

fro m
the
stato r
by
means
of a
pressur e tube
in
orde r
to
exclud e
any
leakage .
The
integra l
apparatu s (Fig .
3)
consist s
of a
tubu -
la r
reacto r
of 1
m-length .
In
orde r
to
measure
the
con-

centration ,
abou t
2o
samplin g tap s
are
installe d
alon g
the
tube . Throug h eac h samplin g tap ,
a
thermocoupl e
is
passe d
to
determin e
the
temperatur e
profile .
The
air
i s
fed
throug h
driers ,
flo w meter s
and
heater s befor e
enterin g
vaporizer ,
where

xylen e
is
evaporated .
Thi s
air-xylen e
mixture ,
containin g
abou t
o.9 i:ol%
xylene ,
is fed to the top of the
tubula r
reactor .
The
phthalicanhydride ,
leavin g
the
reacto r
is
washed
and
condense d
by
wate r
in a
spra y tower .
The
reactio n
hea t
i s

removed
by an
efficien t
coolin g
syste m
in
whic h di-
pheny l
(Dow
therm)
is
vaporized .
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
18
CHEMICAL
REACTION ENGINEERING—HOUSTON
Figure
2.
Loop
reactor
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.
2.
LUFT
ET AL.
Tubular
and
Loop
Reactors

19
1 Xylene storage
2 Filter
3 Metering pump
4 Reactor
5
Spray
absorber
6 Vaporizer
7 Air pre heater
j
8 Rotameter
9
10
Adsorber
- Dryer
11 Air regulater
12 Savety switch
13 Filter
Figure
3.
Tubular
reactor
for the
oxidation
of
xylene
In Chemical Reaction Engineering—Houston; Weekman, V., et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 1978.