catalyst separation, recovery and recycling

CA
T
A
LY
S
T
SE
P
AR
P
P
A
R
R
TION, RECOVE
AA
R
Y
R
R
A
ND RE
C
Y
C
LIN
G
C
ata
ly
s

i
s
by
Meta
l
Comp
l
exes
Vo
l
ume 3
0
E
dito
r
s:
Br
i
an James
,
U
niversity of British Columbia, Vancouver, Canada
Pi
et W. N. M. van Leeuwen
,

U
niversity of Amsterdam, The Netherland
s
A

d
visory Boar
d:
A
lbert S.C. Chan,
T
he Hon
g
Kon
g
Pol
y
technic Universit
y
, Hon
g
Kon
g
R
obert Crabtee,
Y
ale University, U.S.A.
D
avid Cole-Hamilton
,

U
niversity of St An
d
rews, Scotlan

d
István Horvát
h,

E
otvos Loran
d
University, Hungar
y
Kyo
k
o Noza
ki
, University o
f
To
k
yo, Japan
R
o
b
ert Waymout
h
,
S
tanford University, U.S.A
.
T
he titles publishe
d

in this series are liste
d
at the en
d
of this volume.
A
ND RE
C
Y
C
LIN
G
R
OBERT P. TOOZE
Fi
f
e, Scotlan
d
Edited b
y
Fi
f
e, Scotlan
d
an
d
EaStCHEM, School o
f
Chemistr
y

, Universit
y
o
f
St. Andrews,S t. Andrews,
D
AVID J. COLE-HAMILTO
N
CA
T
A
LY
S
T
SE
P
AR
P
P
A
R
R
TION, RECOVE
AA
R
Y
R
R
Sasol Technolo
gy

(UK) Ltd., St. Andrews
,
Chemistr
y
and Process Desi
g
n
A
C.I.P. Cata
l
ogue recor
d
for t
hi
s
b
oo
k

i
s ava
il
a
bl
e from t
h
e L
ib
rary of Congress.
I

SBN-10 1-4020-4086-5
(
HB
)

I
SBN-13 978-1-4020-4086-3
(
HB
)

I
SBN-10 1-4020-4087-3
(
e-
b
oo
k)

I
SBN-13 978-1-4020-4087-0 (e-book)
Pu
bli
s
h
e
d

b
y Spr

i
nger
,
P
.O. Box 17, 3300 AA Dor
d
rec
h
t, T
h
e Net
h
er
l
an
d
s
.
P
rinted on acid-
f
ree paper
A
ll Ri
g
hts Reserved
No part of t
hi
s wor
k

may
b
e repro
d
uce
d
, store
d

i
n a retr
i
eva
l
system, or transm
i
tte
d
i
n any form or
b
y any means, e
l
ectron
i
c, mec
h
an
i
ca

l
, p
h
otocopy
i
ng, m
i
crof
il
m
i
ng
,
r
ecordin
g
or otherwise, without written permission from the Publisher, with the exceptio
n
of an
y
material supplied specificall
y
for the purpose of bein
g
entere
d
a
n
d
execute

d
on a computer system, for exc
l
us
i
ve use
b
y t
h
e purc
h
aser of t
h
e wor
k
.
P
rint
ed
in th
e
N
e
th
e
rlan
ds.
www
.springer.com
ww

©
2006 Sprin
g
er
C
HAPTER 1 H
O
M
OG
E
N
E
OUS

C
ATALY
S
I
S
– AD
V
A
N
TA
G
E
S
A
N
D

PR
O
BLEM
S
1
1
.1
C
atal
y
s
i
s 1
1
.2
C
atalyst
S
tab
i
l
i
ty 4
1.2.1 THERMALLY IND
UC
ED DE
CO
MP
OS
ITI

O
N 4
1.2.2
C
HEMI
C
ALLY IND
UC
ED DE
CO
MP
OS
ITI
O
N
5

1.2.3 PHY
S
I
CA
L L
OSS
FR
O
M THE PR
OC
E
SS

6

1
.3 Layout o
f
the Book
6

1
.4 Re
f
erences
8

C
HAPTER 2
C
LA
SS
I
C
AL H
O
M
OG
E
N
E
OUS

C
ATALY
S
T
S
EPARATI
ON

TE
C
H
NO
L
OG
Y
9

2
.1.1 Covera
g
e of Chapter 9
2.2
G
eneral Process
C
onsiderations 9
2
.3 Ever
y
th

i
n
g

i
s a Reactor 10
2
.4 Overv
i
ew o
f

S
eparat
i
on Technolo
gi
es 10
2
.4.1 TR
A
DITI
O
N
A
L
CO
B
A
LT WITH

CA
T
A
LY
S
T DE
CO
MP
OS
ITI
O
N 10
2
.4.2 UNION CARBIDE-DAVY GAS RECYCLE PROCESS 11
2
.4.3 LI
Q
UID RE
C
Y
C
LE 12
2
.4.4 BIPHASIC SYSTEMS
;
WATER-ORGANIC 14
2
.4.5 IND
UC
ED PHA

S
E
S
EPARATI
O
N 14
2
.4.6 N
O
N-
AQ
UE
O
U
S
PH
AS
E
S
EP
A
R
A
TI
O
N 15
2.4.6.1 NAPS Usin
g
a Non-Polar Catal
y

st
.
16
2.4.6.2 NAPS Using a Po
l
ar Cata
l
yst
.
1
7
2.4.6.3 Ligan
d
Structure an
d

S
olubility Properties
.
17
2.5 Hypothetical processes - How Might the Product be Separated from the
C
atal
y
st? 18
2
.5.1 PR
O
PENE HYDR
O

F
O
RMYL
A
TI
O
N 19
2
.5.2 1-
OC
TENE HYDR
O
F
O
RMYL
A
TI
O
N 20
2
.5.3 ALLYL ALCOHOL 20
2
.5.4 METH
O
XY
V
INYLN
A
PHTH
A

LENE 21
2
.5.5
S
EP
A
R
A
TI
O
N TE
C
HN
O
L
OG
Y F
O
R LE
SS

S
T
A
BLE
CA
T
A
LY
S

T
S
22
2.5.5.1 Mitsu
b
is
h
i TPPO
/
TPP Separation
.
22
2.5.5.2 Organic Po
l
ymer
f
or Cata
l
yst Sta
b
i
l
ization
.
22
2.6 Real-World Com
p
lications 22
2
.6.1

O
R
G
AN
O
PH
OS
PH
O
R
US
LI
G
AND DE
G
RADATI
O
N
S
23
TABLE OF
C
ONTENT
S
2.6.1.1 Oxidation
.
23
2.6.1.2 Alk
y
ldiar

y
lphosphine Formation 2
3
2.6.1.3 Ligand Scrambling
24
2.6.1.4 Phosphine Reactions with Con
j
u
g
ate
d
S
y
stems
.
24
2
.6.1.5 Phos
p
hite Oxi
d
ation
.
24
2
.6.1.6 Simple Phosphite H
y
drol
y
sis

2
5
2
.6.1.7 Poisoning P
h
osp
h
ite Formation
25
2
.6.1.8 Al
d
ehy
d
e Aci
d
Formation
.
2
5
2
.6.1.9 Aci
d
ity Control
.
26
2.6.2
S
EPARATIN
G

BYPR
O
D
UC
T
S
FR
O
M REA
C
TANT
S

O
R PR
O
D
UC
T
S
27
2
.6.2.1 Alkene Hydrogenation
27
2
.6.
2
.
2A
lkene

I
some
r
i
z
ation
27
2.6.2.3 Al
d
ehy
d
e Dimerization an
d
Trimerization
.
2
7
2.6.2.4 Formation o
f
Con
j
u
g
ate
d

C
arbon
y
ls 2

8
2.6.3 INTRIN
S
I
C

CA
T
A
LY
S
T DE
AC
TI
VA
TI
O
N 28
2.7 Further
S
eparat
i
on
C
hallenges 29
2.7.1 RECOVERY OF METAL VALUES FROM A SPENT CATALYST 29
2
.7.1.1 Catal
y
st Containment an

d
Capture Technolo
g
ies
.
30
2.8 Concludin
g
Remarks 35
2.
9 Re
f
erences 36
C
HAPTER 3
S
UPPORTED
C
ATALY
S
T
S
39
Immobilisation of Tailor-ma
d
e Homogeneous Catalysts
.
39
3
.1 Introduction

39
3
.2
S
hort Historical
O
verview 4
0
3
.3 Pol
y
st
y
rene
S
upported
C
atal
y
sts 41
3
.4
Si
l
i
ca
S
upported
C
atal

y
st 44
3
.5 Catalysis in Interphases 53
3
.6 Ordered Mesoporous
S
upport 58
3
.7 Non-covalentl
y
Supported Catal
y
sts 60
3
.8
S
upported Aqueous Phase
C
atal
y
s
i
s 63
3
.9 Process Des
ig
n [71] 65
3
.10 Concluding Remarks 68

3
.11 References
69
C
HAPTER 4
S
EPARATI
ON
BY
S
IZE-EX
C
L
US
I
ON
FILTRATI
ON
73
H
omo
g
eneous Catal
y
sts Applied in Membrane Reactors
.
73
4
.1 Introduction 73
4

.
2
Reactors 7
4

TA
BLE
O
F
CO
NTENT
S
v
i
4.2.1 DE
A
D-END FILTR
A
TI
O
N RE
AC
T
O
R
S
75
4.2.2
C
R

OSS
-FL
OW
FILTR
A
TI
O
N RE
AC
T
O
R
S
7
6
4
.
3
Membranes 78
4.3.1
C
LA
SS
IFI
C
ATI
O
N
O
F FILTRATI

O
N TYPE
S
78
4.4 Dendrimer Supported Catal
y
sts 80
4.4.1 KH
A
R
ASC
H
A
DDITI
O
N RE
AC
TI
O
N 81
4.4.2 ALLYLIC SUBSTITUTION REACTI
O
NS 82
4.4.3 HYDROVINYLATION REACTION 86
4
.4.
4
HYDROGENATION REACTION 88
4.4.5 MI
C

H
A
EL
A
DDITI
O
N
R
E
AC
TI
O
N 89
4.
5
Dendritic Effects
90
4.6
U
nmodified or
N
on-dendritic
C
atal
y
sts 94
4.6.1 HYDR
OG
ENATI
O

N 95
4.6.2 PH
AS
E TR
A
N
S
FER
CA
T
A
LY
S
I
S
97
4.7
S
oluble Polymer
S
upported
C
atalysts 98
4.8
C
onclud
i
ng Remarks 102
4.9 References 102
C

HAPTER 5 BIPHA
S
I
C

S
Y
S
TEM
S
:
W
ATER –
O
R
G
A
N
I
C
105
5
.2.1 GENERAL 106
5
.2.2 BIPH
AS
I
C

S

Y
S
TEM
S
107
5
.2.3 A
Q
UEOUS BIPHASIC CATALYSIS 108
5.2.3.2 Aqueous-p
h
ase Cata
ly
sis
a
s a Unit
O
peration
1
1
0
5
.2.4 EX
A
MPLE
S

O
F
AQ

UE
O
U
S
BIPH
AS
I
C

CA
T
A
LY
S
I
S
114
5
.2.4.1 Hy
d
roformylation (Ruhrchemie/Rhône-Poulenc[RCH/RP] process)

114
5
.2.4.2 Other In
d
ustrially Use
d
A
q

ueous-bi
p
hasic Processes
1
16
5
.2.4.3
Sh
ort
O
vervie
w

of
Ot
h
er Reaction 11
8
5
.2.5
O
THER PR
O
P
OSA
L
S
F
O
R W

A
TER - BIPH
AS
I
C

S
Y
S
TEM
S
119
5
.2.6 INTERL
U
DE - BIPHA
S
I
C

S
Y
S
TEM
S
:
O
R
G
ANI

C
-
O
R
G
ANI
C
123
5.3 Recycle and Recovery of Aqueous Catalysts 124
5
.3.1 RE
C
Y
C
LIN
G
126
5
.3.2 RE
CO
VERY 128
5
.3.3 E
CO
N
O
MI
CS

O

F
T
HE PR
OC
E
SS
132
5
.3.4 ENVIRONMENTAL ASPECTS 132
5.4 Concludin
g
Remarks 134
5
.
5
References 1
35
C
HAPTER 6 FLUOROU
S
BIPHA
S
I
C

C
ATALY
S
I
S

145
6
.1 Introduction 145
4
.3.2 MEMBRANE MATERIALS

7
9
5
.1 Introduct
i
o
n
.
1
05
5.2.3.1
W
ater as a
S
olvent
.
1
08
5
.2 Immob
i
l
i
zat

i
on w
i
th the Help o
f
L
i
qu
i
d
S
upports 10
6
TA
BLE
O
F
CO
NTENT
S
v
i
i
6.2 Alkene H
y
dro
g
enat
i
on 148

6.3 Alkene H
y
dros
i
lat
i
on 151
6.4 Alkene Hydroborat
i
on 151
6.5 Alkene Hydroformylation 152
6.6 Alkene Epoxidation 158
6.7
O
ther
O
xidation Reactions 161
6.8 All
y
l
i
c Alk
y
lat
i
on 163
6.9 Heck,
S
t
i

lle,
S
uzuk
i
,
S
onagash
i
ra and Related
C
oupl
i
ng React
i
ons 164
6.10 Asymmetr
i
c Alkylat
i
on o
f

A
ldehydes
166

6.11 Miscellaneous Catalytic Reactions 169
6.12 Fluorous Catal
y
sis Without Fluorous Solvents 170

6.13 Continuous Processin
g
171
6.14 Process
Sy
nthes
i
s
f
or the Fluorous B
i
phas
i
c H
y
dro
f
orm
y
lat
i
on o
f

1
-Octene
1
75
6.15
C

onclus
i
ons 178
6.16 Acknowledgement 179
6
.1
7
Re
f
erences 1
79

C
HAPTER 7
C
ATALY
S
T RE
C
Y
C
LIN
G
U
S
IN
G
IONI
C
LI

Q
UID
S
183
7
.1 Introduction 1
83

7.1.1 INTRODUCTION TO IONIC LI
Q
UIDS 183
7.1.2 INTR
O
D
UC
TI
O
N T
O
TR
A
N
S
ITI
O
N MET
A
L
CA
T

A
LY
S
I
S
IN I
O
NI
C

L
I
Q
UID
S
187
7.1.3 MULTIPHASIC CATALYSIS WITH IONIC LI
Q
UIDS – ENGINEERING
AS
PE
C
T
S
189
7.2 Liquid-liquid Biphasic, Rh-catalysed Hydroformylation Using Ionic
Liq
u
i
ds 192

7.3 Rhod
i
um
C
atalysed Hydro
f
ormylat
i
on Us
i
ng
S
upported Ion
i
c L
i
qu
i
d
Phase
S
ILP
)

C
atalys
i
s 201
7.3.1 SUPPORTED IONIC LI
Q

UIDS BY CHEMICAL BONDS 203
7.3.2
S
UPP
O
RTED I
O
NI
C
LI
Q
UID
S
BY IMPRE
G
N
A
TI
O
N 204
7.4
C
osts And Econom
i
cs 206
7.5
C
onclus
i
ons 209

7.6 References 210
C
HAPTER 8
SU
PER
C
RITI
C
AL FL
U
ID
S
215
C
ompressed Gases as Mobile Phase and Catal
y
st Support
.
2
1
5
8.1 Introduct
i
on to supercr
i
t
i
cal
f
lu

i
ds 215
8.2 Appl
i
cat
i
ons o
f
sc
C
O
2
in Catalyst Immobilisation 217
8
.2.1
CO
2
AS
THE
O
NLY M
ASS

S
EP
A
R
A
TIN
G

AG
ENT 217
8
.2.2 BIPHASIC SYSTEMS CONSISTING OF CO
2
A
ND LI
Q
UID PH
AS
E
S
223
8
.2.2.1 Water as the Liquid Phase
.
223
8
.2.2.2 Poly(ethyleneglycol) (PEG) as the Liqui
d
Phase
.
225
8
.2.2.3 Ionic Li
q
ui
d
s as the Li

q
ui
d
Phase
.
2
2
5
8
.2.3 BIPHA
S
I
C

S
Y
S
TEM
S

CO
N
S
I
S
TIN
G

O
F

CO
2
A
ND
SO
LID PH
AS
E
S
23
0
8.2.3.2
8
.2.3.1 Inorganic
S
upports
.
230
Organic Polymer Supports
231
T
A
BLE
O
F
CO
NTENT
S
v
ii

i
8.4
S
ummar
y
234
8
.
5
References 2
3
4
C
HAPTER 9 AREAS FOR FURTHER RESEARCH 237
9
.1 Introduct
i
on 237
9
.2 Conventional Se
p
aration Methods (See Cha
p
ter 2) 239
9
.3
C
atal
y
sts on Insoluble

S
upports
(C
hapter 3
)
240
9
.4
C
atalysts on
S
oluble
S
upports
(C
hapter 4
)
241
9
.5 Aqueous B
i
phas
i
c
C
atalys
i
s
(C
hapter 5

)
242
9
.6 Fluorous Biphasic Catalysis (Chapter 6) 243
9
.7 React
i
ons Involv
i
n
g
Ion
i
c L
i
qu
i
ds
(C
haoter 7
)
244
9
.8 Reactions Usin
g
Supercritical Fluids (Chapter 8) 245
9
.9
C
onclus

i
ons 247
9
.10 References 247
8
.3 Econom
i
c Evaluat
i
on and
S
ummar
y
232
8.3.1 P
O
TENTI
A
L F
O
R
SCA
LE-
U
P 232
T
A
BLE
O
F

CO
NTENT
S
ix
C
HAPTER 1 H
O
M
OG
E
N
E
OUS

C
ATALY
S
I
S
– AD
V
A
N
TA
G
E
S
A
N
D

PROBLEM
S
D. J. COLE-HAMILTON
a
AND R. P. T
OO
Z
E
b
a
EaStCHEM, School of Chemistry, Univesity of St. Andrews, St. Andrews,
a
F
ife, KY19 9ST, Scotland
b
Sasol Technology UK, Ltd., Purdie Building, North Haugh, St. Andrews,
F
i
f
e, KY19 9ST, Scotlan
d

1
.1
C
atal
y
s
i
s

Cata
l
ysts spee
d
up c
h
em
i
ca
l
react
i
ons
b
ut can
b
e recovere
d
unc
h
ange
d
at t
h
e en
d
of t
h
e
reaction. The

y
can also direct the reaction towards a specific product and allow
C
h
em
i
str
y
to
b
e carr
i
e
d
out at
l
ower temperatures an
d
pressures w
i
t
h

high
er se
l
ect
i
v
i

t
y

towards the desired product. As a result they are used very extensively in the Chemical
In
d
ustry. C
h
r
i
s A
d
ams, wr
i
t
i
ng for T
h
e Nort
h
Amer
i
can Cata
l
ys
i
s Soc
i
ety est
i

mates
that “35% of
g
lobal GDP depends on catal
y
sis, althou
g
h this excludes the emer
g
en
t

g
enet
i
c
b
us
i
ness. Conf
i
n
i
n
g
t
h
e ana
ly
s

i
s to t
h
e c
h
em
i
ca
l
s
i
n
d
ustr
y
, w
i
t
h

gl
o
b
a
l
sa
l
es of
p
erhaps $1.5 x 1

0
1
2
t
h
e proport
i
on of processes us
i
ng cata
l
ysts
i
s 80% an
d

i
ncreas
i
ng.
T
he catal
y
st market itself is US
$
10
10
,
so that catal
y

sis costs are much less than 1% of
t
h
e sa
l
es revenue from t
h
e pro
d
ucts w
hi
c
h
t
h
e
y

h
e
l
p create. Sma
ll
won
d
er t
h
at t
h
e

cata
ly
st mar
k
et
i
s
i
ncreas
i
n
g
at 5% per annum” [1]
TABLE 1.1 Comparison of homo
g
eneous and hetero
g
eneous catal
y
sts

H
etero
g
eneous
H
omo
g
eneous
C

atalyst form Solid, often metal or metal oxide Metal complex
M
o
d
e of use F
i
xe
d

b
e
d
or s
l
urry D
i
sso
l
ve
d

i
n react
i
on me
di
um
So
l
vent Usua

lly
not requ
i
re
d
Usua
lly
requ
i
re
d
– can
b
e pro
d
uct o
r

by
pro
d
uc
t
Se
l
ect
i
v
i
ty Usua

ll
y poor Can
b
e tune
d

Sta
bili
ty Sta
bl
e to
hi
g
h
temperature Often
d
ecompose < 100
o
C
R
ec
y
clabilit
y
Eas
y
Can be ver
y
difficult
S

pecial reactions Haber process, exhaust clean up etc. Hydroformylation of alkenes, methanol
c
ar
b
on
yl
at
i
on, as
y
mmetr
i
c s
y
nt
h
es
i
s etc
T
h
ere are two
ki
n
d
s of cata
l
ysts. Heterogeneous cata
l
ysts are

i
nso
l
u
bl
e
i
n t
h
e me
di
um
in which the reaction is takin
g
place so that reactions of
g
aseous or liquid rea
g
ents
o
ccur at the surface, whilst homo
g
eneous catal
y
sts are dissolved in the reaction
m
e
di
um an
d

h
ence a
ll
cata
l
yt
i
c s
i
tes are ava
il
a
bl
e for react
i
on. Some of t
h
e propert
i
es
o
f catal
y
sts are collected in Table 1.1, where hetero
g
eneous and homo
g
eneous catal
y

sts
a
re com
p
ared.
©
2006 Springer. Printe
d
in the Netherlan
d
s.
1
1–8
.

D
.
J
J
J
C
o
C
C
l
e
-
H
a
H

H
m
i
l
t
o
n
a
n
d
R
.
P
.
T
o
T
T
o
z
e (eds.), Catalyst Separation, Recovery and Recycling,

D
.
J
.
CO
LE-HAMILT
O
N AND R. P. T

OO
ZE
2
Heterogeneous cata
l
ysts are genera
ll
y meta
l
s or meta
l
ox
id
es an
d
t
h
ey ten
d
to g
i
ve
rather unselective reactions. The
y
are ver
y
stable towards heat and pressure, so can be
used at high temperature. Only the surface atoms are available for reaction.
Homo
g

eneous catal
y
sts, on the other hand are usuall
y
complexes, which consist of a
meta
l
centre surroun
d
e
d

b
y a set of organ
i
c
li
gan
d
s. T
h
e
li
gan
d
s
i
mpart so
l
u

bili
ty an
d

s
ta
bili
ty to t
h
e meta
l
comp
l
ex an
d
can
b
e use
d
to tune t
h
e se
l
ect
i
v
i
ty of a part
i
cu

l
a
r

catalyst towards the synthesis of a particular desirable product. By varying the size,
s
hape and electronic properties of the li
g
ands, the site at which the substrate binds can
b
e constra
i
ne
d

i
n suc
h
a wa
y
t
h
at on
ly
one of a
l
ar
g
e num
b

er of poss
ibl
e pro
d
ucts can
b
e pro
d
uce
d
. As an examp
l
e, F
ig
ure 1.1 s
h
ows a ran
g
e of pro
d
ucts t
h
at m
igh
t
b
e
p
ro
d

uce
d
from a m
i
xture conta
i
n
i
ng an a
lk
ene, car
b
on monox
id
e,
h
y
d
rogen an
d
an
alcohol. All of the products have their uses, but it is a triumph of homo
g
eneous
catalysis that any one of the products can now be made with > 90 % selectivity by
careful selection of the metal centre, li
g
ands, reaction conditions and in some cases
s
u

b
strate [2]
R
CO

H
2
Me
O
H
R
C
H
O
R
C
H
O
*
R
CH
2
OH
R
C
H
2
OH
R
CO

2
M
e
R
CO
2
M
e
*
*
R
R
O
*
R
O
*
R
*
R
O
R
M
e
O
O
H
R
n
R

R
n
Me
CO
2
Me
Me
CO
2
H
E
t
O
H
C
H
2
O
HOC
H
2
C
H
2
OH
Hyd
ro
f
o
rm

yl
at
i
on
Hyd
rocar
b
on
yl
at
i
on
M
et
h
ox
y
car
b
on
yl
at
i
on
CO
Hydrogenatio
n
R'
R
H

y
dro
g
enation
Iso
m
e
r
isation
Co-oli
g
omerisatio
n
Co-pol
y
merisation
P
o
ly
mer
i
sat
i
on
Methanol homolo
g
atio
n
Methanol carbon
y

latio
n
*
*
Fi
g
ure 1.1
.
Some of the products that can form from an alkene, carbon momoxide, h
y
dro
g
en and methanol.
T
he asterisks represent asymmetric centres in chiral molecules
Various different kinds of selectivity are represented in Figure 1.1. These include:
x
C
h
emose
l
ect
i
v
i
t
y
, t
h
e pro

d
uct
i
on of one pro
d
uct t
y
pe suc
h
as a
l
co
h
o
l
s rat
h
er
than aldeh
y
des
x Reg
i
ose
l
ect
i
v
i
ty, t

h
e pro
d
uct
i
on of a
li
near ester rat
h
er t
h
en one w
i
t
h
a
branchin
g
meth
y
l
g
roup
x
Stereose
l
ect
i
v
i

ty, t
h
e pro
d
uct
i
on of one enant
i
omer of a c
hi
ra
l
compoun
d

(
c
hi
ra
l
pro
d
ucts are mar
k
e
d
w
i
t
h

an aster
i
s
k

i
n F
ig
ure 1.1
)
In general, heterogeneous catalysts do not show the selectivity shown by chiral
catal
y
sts, althou
g
h current research on surface modifiers has shown that even
enantioselective reactions, albeit for a restricted ran
g
e of substrates is becomin
g
p
oss
ibl
e [3, 4]
Despite this selectivity advantage of homogeneous catalysts, almost all of the industrial
cata
ly
t
i
c processes use

h
etero
g
eneous cata
ly
sts,
b
ecause of t
h
e
i
r one ma
j
or a
d
vanta
g
e,
their ease of separation form the reaction product. Bein
g
insoluble in the reaction
H
O
M
OG
ENE
OUS

C
ATALY

S
I
S
– AD
V
ANTA
G
E
S
AND PR
O
BLEM
S
3
me
di
um,
h
eterogeneous cata
l
ysts can often
b
e use
d
as f
i
xe
d

b

e
d
s over w
hi
c
h
t
h
e
s
ubstrates flow continuousl
y
in the liquid or
g
aseous form. This means that the catal
y
st
can be contained within the reactor at all times. Not only does this mean that the
s
eparation of the products from the catal
y
st is built into the process, but also, the
cata
l
yst
i
s a
l
ways
k

ept un
d
er t
h
e con
di
t
i
ons of temperature, pressure, contact w
i
t
h
t
h
e
s
u
b
strate an
d
pro
d
ucts, for w
hi
c
h

i
t
h

as
b
een opt
i
m
i
se
d
.
For homogeneous catalysts, which are dissolved in the reaction medium containing
the substrates, products and dissolved
g
ases, the separation can be extremel
y
ener
gy
i
ntens
i
ve an
d
t
i
me consum
i
n
g
. On
ly
rare

ly
, w
h
en t
h
e pro
d
uct can
b
e evaporate
d
un
d
er
t
h
e react
i
on con
di
t
i
ons, can
h
omo
g
eneous cata
ly
t
i

c react
i
ons
b
e carr
i
e
d
out un
d
e
r

cont
i
nuous f
l
ow con
di
t
i
ons, w
h
ere t
h
e su
b
strates are
i
ntro

d
uce
d
cont
i
nuous
l
y
i
nto t
h
e
reactor whilst the products are continuousl
y
removed. More often, commercial
s
olution containin
g
the product(s), unreacted substrates and
c
atalyst is removed
o
perat
i
n
g
at
l
ower pressure t
h

an t
h
e reactor. T
h
e pro
d
ucts an
d
unreacte
d

subs
trat
es
ar
e

t
h
en separate
d
from t
h
e cata
l
yst an
d

l
ower

b
o
ili
ng
b
ypro
d
ucts
b
y fract
i
ona
l
di
st
ill
at
i
on
b
efore the fraction containing the catalyst is returned to the reactor. Since the
s
eparat
i
on
i
s carr
i
e
d

out un
d
er con
di
t
i
ons t
h
at are far remove
d
from t
h
ose for w
hi
c
h
t
h
e
catal
y
st has been optimised, there is a dan
g
er that the catal
y
st ma
y
precipitate, thus
c
l

o
ggi
n
g
p
i
pewor
k
or, worse st
ill
,
d
ecompose
i
n t
h
e rec
yli
n
g

l
oop
.
In general homogeneous catalysis has only been commercialised when there is no
h
eterogenous cata
l
yst t
h

at
i
s capa
bl
e of promot
i
ng t
h
e
d
es
i
re
d
react
i
on or w
h
en
s
electivit
y
to a hi
g
her added value product is possible usin
g
a homo
g
eneous catal
y

st.
Creat
i
ve c
h
em
i
sts an
d
process en
gi
neers
h
ave t
h
en
j
o
i
ne
d
forces to prov
id
e a cos
t

effective solution to the se
paration problem
.
Increas

i
ng env
i
ronmenta
l
concerns an
d

d
w
i
n
dli
ng supp
li
es of raw mater
i
a
l
s an
d

energy sources mean that there is now a significant pressure to introduce cleane
r

p
rocessin
g
in the chemical and pharmaceutical industries.
Ideall

y
reactions should have as man
y
as possible of the followin
g
properties:
x
Use renewa
bl
e fee
d
stoc
k
s
x Make a single produc
t

x

H
ave 100 % atom efficienc
y
(all the atoms in the startin
g
materials end up in
t
he products, althou
g
h expulsion of water mi
g

ht be acceptable)
x
O
perate un
d
er m
ild
con
di
t
i
ons
(
prefera
bl
y am
bi
ent temperature an
d
pressure
)
x
B
e tolerant of the
p
resence of ai
r

x
Pro

d
uce no waste or ot
h
er
by
pro
d
ucts
(
t
h
ese are often quant
i
f
i
e
d
us
i
n
g
t
h
e
E-
Factor, which is the mass (k
g
) of waste produced per k
g
of product formed.

For a fu
ll
er
di
scuss
i
on see C
h
apter 5, Sect
i
on 5.3.4
)

x

H
ave a ver
y
lon
g
-lived catal
y
st if, as is ver
y
likel
y
, one is required
x

H

ave a simple separation method for the catal
y
st from the products
x
O
perate un
d
er cont
i
nuous f
l
ow con
di
t
i
on
s
Th
e cata
l
ysts t
h
at come c
l
osest to meet
i
ng t
h
ese requ
i

rements are enzymes,
b
ut
i
n
g
eneral, the products of biolo
g
ical process are not separated from the medium in which
they are formed, rather being used in situ. The demand for high selectivity and low
environmental impact su
gg
ests that there will be a desire to commercialise more
P
art of t
h
e
li
qu
id
cata
l
yt
i
c
c
o
n
t
i

n
u
o
u
s
l
y
f
r
f
f
o
m
t
h
e
r
e
a
c
t
o
r
t
o
a
s
e
p
a

r
a
t
o
r
,
w
h
i
c
h
i
s
u
s
u
a
l
l
y
a
d
i
s
t
i
-
llation s
y
stem

p
rocesses are carried out under what we shal
l
r
e
f
e
r t
o
a
s

b
at
c
h
co
ntin
uous

co
n
d
iti
o
n
s.

D
.

J
.
CO
LE-HAMILT
O
N AND R. P. T
OO
ZE
4
p
rocesses us
i
ng
h
omogenous cata
l
ys
i
s an
d

h
ence t
h
e pro
bl
em of separat
i
ng, recover
i

ng
and/or rec
y
clin
g
the catal
y
sts must be addressed, perhaps usin
g
innovative solutions [5,
6
]
T
here are, then, three critical requirements of an
y
catal
y
st if it is to be exploited on
a commerc
i
a
l
sca
l
e; t
h
ese are act
i
v
i

ty, se
l
ect
i
v
i
ty an
d
sta
bili
ty. It
h
as
b
een w
id
e
l
y
d
emonstrate
d
an
d

g
enera
lly
accepte
d

t
h
at
h
omo
g
eneous cata
ly
sts are super
i
or to t
h
e
ir

h
eterogeneous counterparts
i
n terms of
b
ot
h
act
i
v
i
ty
(
certa
i

n
l
y un
d
er m
ild
react
i
on
conditions) and selectivit
y
(the classical example is chiral catal
y
sis).
Th
e fata
l
wea
k
ness t
h
at
h
as prevente
d
t
h
e pre
di
cte

d
pro
li
ferat
i
on of
h
omo
g
eneous
cata
ly
sts
i
s
l
ac
k
of sta
bili
t
y
. T
hi
s furt
h
er
ill
ustrates t
h

at a
ll
of t
h
e a
b
ove cr
i
ter
i
a nee
d
to
be
f
ul
f
illed.
1
.2 Catal
y
st Stabilit
y

Cata
ly
st sta
bili
t
y

can
b
e
d
ef
i
ne
d

i
n terms of turnover num
b
er
(
TON
)
. A text
b
oo
k

de
f
i
n
i
t
io
n
o

f t
his

is:

TON = mo
l
of fee
d
consume
d

b
efore act
i
v
i
t
y
cease
s
mol of catalyst utilised
In rea
li
ty t
h
e
li
m
i

t
i
ng case
i
t not comp
l
ete
l
oss of act
i
v
i
ty
b
ut rat
h
er re
d
uct
i
on o
f

act
i
v
i
t
y

b
e
l
ow a cr
i
t
i
ca
l
t
h
res
h
o
ld

d
eterm
i
ne
d

by
t
h
e econom
i
cs of an
y

gi
ven process
an
d
reactor
d
es
i
gn.
T
he TON can be reduced in a number of wa
y
s
x
T
hermall
y
induced decomposition.
x
C
h
em
i
ca
ll
y
i
n
d
uce

d

d
ecompos
i
t
i
on, of w
hi
c
h
two furt
h
er categor
i
es can
be considered namely substrate induced decomposition and poisoning by
i
mpur
i
t
i
es or pro
d
ucts.
x
P
hy
s
i

ca
l

l
oss from t
h
e process.
Th
ese w
ill
now
b
e cons
id
ere
d

b
r
i
ef
l
y
i
n turn
As ment
i
one
d
a

b
ove one of t
h
e fun
d
amenta
l
attr
ib
utes ascr
ib
e
d
to
h
omogeneous
catal
y
sts is superior activit
y
at low temperature. However, even within classes of such
catalysts, improvements in catalyst activity can be made allowing operation at lowe
r

temperatures, thus reducin
g
or avoidin
g
completel
y

this mode of catal
y
st deca
y
. One
s
uc
h
examp
l
e can foun
d

i
n recent a
d
vances
i
n pa
ll
a
di
um cata
ly
se
d
et
h
ene car
b

on
yl
at
i
on
(
E
q
uation 1.1).
C
H
2
=
CH
2
+
CO
+ R
O
HH-
{CH
2
CH
2
C(O)}
n
-
OR
E
quation 1.1.

E
t
h
ene car
b
on
yl
at
i
on
l
ea
di
n
g
to a
lkyl

(
R
)
prop
i
onates
(
n = 1
)
or to et
h
ene car

b
on monox
id
e
c
opo
ly
mers
(
n
i
s
l
ar
g
e
)
1.2.1 THERMALLY IND
UC
ED DE
CO
MP
OS
ITI
O
N
H
O
M
OG

ENE
OUS

C
ATALY
S
I
S
– AD
V
ANTA
G
E
S
AND PR
O
BLEM
S
5
Thi
s react
i
on to g
i
ve met
h
y
l
propanoate can
b

e cata
l
yse
d

b
y a com
bi
nat
i
on o
f

p
alladium acetate, triphen
y
lphosphine and methanesulphonic acid [7]. However in
order to obtain acce
p
table rates the reaction must be carried out at over 100
o

C
. At this
temperature cata
ly
st
li
fe
i

s s
h
ort
d
ue to a var
i
et
y
of s
id
e react
i
ons [8, 9] an
d
ver
y
expensive palladium is lost. Advances in catalyst design have identified alternative
p
hosphine ligands such as 1,2-
bis
(
ditertiarybutylphosphinomethylbenzene) that can
g
ive much increased activit
y
even at 3
0
o
lower tem
p

eratures and thus avoid thermal
decomposition [10]. Whilst this and other examples show that increased understandin
g

can
l
ea
d
to
i
mprovements
i
n cata
l
yst
d
es
i
gn suc
h
t
h
at reactor operat
i
ng con
di
t
i
ons can
b

e c
h
an
g
e
d
so as to avo
id

d
ecompos
i
t
i
on, no suc
h
act
i
v
i
t
y
en
h
ancement w
ill

i
mpact on
t

h
e separat
i
on process w
hi
c
h

i
s governe
d

b
y t
h
e p
h
ys
i
ca
l
propert
i
es of a
ll
pro
d
ucts an
d

reactants. An example here would be the h
y
droform
y
lation of alkenes. Scientific
advances have resulted in phosphine modified rhodium catalysts that exhibit much
g
reater activit
y
than for example unmodified cobalt catal
y
sts, but when this technolo
gy

i
s app
li
e
d
to
hi
g
h
er mo
l
ecu
l
ar we
i
g

h
t a
lk
enes
(
C10+
)
t
h
e separat
i
on of pro
d
ucts from
cata
ly
st
i
n suc
h
a wa
y
as to avo
id
extens
i
ve
d
ecompos
i

t
i
on of expens
i
ve cata
ly
st
i
s a
f
orm
id
a
bl
e tec
h
no
l
og
i
ca
l
c
h
a
ll
enge. D
i
fferent poss
ibl

e separat
i
on met
h
o
d
o
l
og
i
es for
l
ong chain aldehydes formed by hydroformylation reactions are explored in more detail
i
n the subse
q
uent cha
p
ters of this book
.
N
o cata
l
yst
h
as an
i
nf
i
n

i
te
li
fet
i
me. T
h
e accepte
d
v
i
ew of a cata
l
yt
i
c cyc
l
e
i
s t
h
at
i
t
p
roceeds via a series of reactive species, be they transient transition state type
s
tructures or relativel
y
more stable intermediates. Reaction of such intermediates with

either excess li
g
and or substrate can
g
ive rise to ver
y
stable complexes that are
ki
net
i
ca
lly

i
ncompetent of susta
i
n
i
n
g
cata
ly
s
i
s. T
h
e text
b
oo
k

examp
l
e of t
hi
s
i
s
tr
i
p
h
eny
l
p
h
osp
hi
ne mo
di
f
i
e
d
r
h
o
di
um
h
y

d
roformy
l
at
i
on, w
h
ere a p
l
ot of act
i
v
i
ty versus
l
i
g
and:metal ratio shows the classical “volcano plot” whereb
y
activit
y
reaches a peak a
t

a certain ratio but then falls off rapidl
y
in the presence of excess phosphine, see Fi
g
ure
1.2 [11]. On occas

i
on t
h
ese over
lig
ate
d
comp
l
exes are mater
i
a
l
s t
h
at can
b
e
id
ent
i
f
i
e
d

i
n so
l
ut

i
on or per
h
aps more te
lli
ng
l
y
i
so
l
ate
d
from cata
l
yt
i
c react
i
ons. Suc
h
react
i
ons
can often
b
e reverse
d

b

y remova
l
of t
h
e excess reagent. Suc
h
processes are not
considered in the context of this book as resulting in loss of overall turnover number.
F
igure 1.2
.
Typical plot of the effect of rate on the P:Rh ratio for a Rh/PPh
3
hydroformylation catalyst. The
e
xact
p
osition of the maximum de
p
ends on [Rh],
p
CO
an
d
T
1.2.2
C
HEMI
CA
LLY INDU

C
ED DE
CO
MP
OS
ITI
O
N
D
.
J
.
CO
LE-HAMILT
O
N AND R. P. T
OO
ZE
6
A more serious but potentiall
y
soluble problem is that of poisonin
g
b
y
impurities in the
f
ee
d
to a cata

l
yt
i
c react
i
on. Homogeneous cata
l
ysts are aga
i
n
b
e
li
eve
d
to
b
e more
s
usceptible to this mode of decomposition than their heterogeneous counterparts. This
pro
bl
em may
b
e so
l
ve
d

b

y t
h
e
d
eve
l
opment of more ro
b
ust cata
l
ysts,
b
ut a more usua
l
s
olution is feedstock
p
urification. An exam
p
le where this
p
urification is carried ou
t
in
sit
u
i
nvo
l
ves t

h
e a
ddi
t
i
on of
l
ar
g
e quant
i
t
i
es of a
l
um
i
n
i
um rea
g
ents
(
suc
h
as
met
h
y
l

a
l
um
i
noxane, MAO
)
to po
l
ymer
i
sat
i
on an
d
o
li
gomer
i
sat
i
on react
i
ons. T
h
e rat
i
o
of a
l
um

i
n
i
um to trans
i
t
i
on meta
l
can
b
e very
hi
g
h

(
>500
)
an
d
part of t
h
e purpose of t
hi
s
l
ar
g
e excess is believed to be removal of ox

yg
enates from alkene feeds [12]
.
T
he loss of expensive catal
y
st from the reactor s
y
stem can be fatal for an
y
process.
Physical loss involves the removal of active catalyst from the closed loop of the
process. T
hi
s can
i
nc
l
u
d
e t
h
e p
l
at
i
n
g
out of meta
l

or ox
id
es on t
h
e
i
nterna
l
surfaces o
f

t
h
e manufactur
i
ng p
l
ant, fa
il
ure to recover potent
i
a
ll
y act
i
ve cata
l
yst from purge
s
treams and the decomposition of active catal

y
st b
y
the process of product recover
y
.
T
he first two can be alleviated to some extent b
y
improvements in catal
y
st or process
d
es
ig
n, t
h
e
l
ast
i
s an
i
ntr
i
ns
i
c pro
bl
em for a

ll
manufactur
i
n
g
operat
i
ons an
d

i
s t
h
e
s
u
bj
ect of t
hi
s
b
oo
k
.
Catalysts are traditionally designed and optimised based on their performance in
the reactor and not for their ability to withstand traditional separation processes.
However, on takin
g
an
y

s
y
stem from the laborator
y
to the pilot plant and be
y
ond, this
need to isolate product whilst efficientl
y
recoverin
g
the catal
y
st often becomes the mos
t

i
mportant s
i
n
gl
e
i
ssue. T
h
e
b
est opt
i
on

i
s se
l
ect
i
on of a pro
d
uct
i
so
l
at
i
on met
h
o
d
t
h
a
t

maintains the integrity of the catalyst and requires no further treatment of the catalyst
prior to reintroduction into the reactor, or leaves the catal
y
st in the reactor at all times
.
A compromise solution can be that, althou
g
h a catal

y
st ma
y
not be in its active
f
orm
i
n t
h
e separat
i
on un
i
t,
i
t can
b
e recovere
d
an
d
re
g
enerate
d
eas
ily
at t
h
e pro

d
uct
i
on
f
ac
ili
ty. A f
i
na
l
opt
i
on
i
s t
h
at spent cata
l
yst can
b
e recovere
d
, concentrate
d
an
d
returne
d

to t
h
e or
i
g
i
na
l
supp
li
er for reprocess
i
ng. W
hil
st t
hi
s
i
s an expens
i
ve an
d

i
ne
l
egant
option, it remains the most pra
g
matic solution until technolo

g
ies described in this boo
k

reach maturit
y.
1
.
3
In this book, we report on the state of the art of methods for catalyst separation
recover
y
and rec
y
clin
g
, not
j
ust describin
g
the chemistr
y
, but also discussin
g
the
process
d
es
i
gn t

h
at wou
ld

b
e requ
i
re
d
to put t
h
e processes
i
nto pract
i
ce.
Convent
i
ona
l
processes
i
nvo
l
v
i
ng
di
st
ill

at
i
on of t
h
e pro
d
uct
di
rect
l
y from t
h
e reactor o
r

b
atc
h
cont
i
nuous o
p
erat
i
on w
h
ere t
h
e
di

st
ill
at
i
on
i
s carr
i
e
d
out
i
n a se
p
arate c
h
am
b
e
r

(
Chapter 2) provide the backdrop for the man
y
alternative processes that are bein
g

d
i
scussed.

1.2.3 PHY
S
I
C
AL L
OSS
FR
O
M THE PR
OC
E
SS

L
a
y
out o
f
the Boo
k

H
O
M
OG
ENE
OUS

C
ATALY

S
I
S
– AD
V
ANTA
G
E
S
AND PR
O
BLEM
S
7
Th
ese a
l
ternat
i
ve processes can
b
e
di
v
id
e
d

i
nto two ma

i
n categor
i
es, t
h
ose t
h
a
t

i
nvolve insoluble (Cha
p
ter 3) or soluble (Cha
p
ter 4) su
pp
orts cou
p
led with continuous
f
low operation or filtration on the macro – nano scale, and those in which the catalyst is
i
mmobilised in a separate phase from the product. These chapters are introduced b
y
a
di
scuss
i
on of aqueous

bi
p
h
as
i
c systems
(
C
h
apter 5
)
, w
hi
c
h

h
ave a
l
rea
d
y
b
een
commerc
i
a
li
se
d

. Ot
h
er c
h
apters t
h
en
di
scuss newer approac
h
es
i
nvo
l
v
i
ng f
l
uorous
s
olvents (Cha
p
ter 6), ionic li
q
uids (Cha
p
ter 7) and su
p
ercritical fluids (Cha
p

ter 8).
N
o attempt is made to provide comprehensive covera
g
e of all the work carried out in
t
h
ese
di
fferent me
di
a,
b
ut rat
h
er to
gi
ve a f
l
avour of t
h
e
ki
n
d
of s
y
stems for w
hi
c

h
t
h
e
di
fferent approac
h
es ma
y

b
e appropr
i
ate. In a
ll
t
h
e c
h
apters, a more
d
eta
il
e
d

di
scuss
i
on

of t
h
e r
h
o
di
um cata
l
yse
d

h
y
d
roformy
l
at
i
on of 1-octene to nonana
l
, as a representat
i
ve
example of the s
y
nthesis of a lon
g
chain aldeh
y
de with relativel

y
low volatilit
y
, is
p
rovided [13, 14]. This reaction has been chosen because:
x
It
i
s a react
i
on
d
emonstrat
i
n
g
100 % atom econom
y
x It
i
s a react
i
on w
hi
c
h
uses
b
ot

h
gas an
d

li
qu
id
su
b
strates
x
T
h
e
rat
e

o
f th
e
r
e
a
c
ti
o
n i
s

c

r
uc
ial f
o
r
success
f
u
l
co
mm
e
r
c
iali
s
ati
o
n
x
T
here are important issues relatin
g
to chemoselectivit
y
(aldeh
y
des or alcohols
ma
y

b
e t
h
e pro
d
ucts an
d
a
lk
ene
i
somer
i
sat
i
on
i
s a compet
i
n
g
s
id
e react
i
on,
w
hi
c

h
must
b
e re
d
uce
d
to a m
i
n
i
mum
)
an
d
reg
i
ose
l
ect
i
v
i
ty
(li
near a
ld
e
h
y

d
e
i
s
much
p
referred over branched)
x
It is a commerciall
y
important reaction as a step in the s
y
nthesis of nonanol,
an
i
mportant p
l
ast
i
c
i
zer a
l
co
h
o
l
. Ot
h
er

l
on
g
c
h
a
i
n a
l
co
h
o
l
s,
d
er
i
ve
d
from
p
ro
d
uct a
ld
e
h
y
d
es

b
y
h
y
d
rogenat
i
on are use
d
as t
h
e
b
as
i
s of soaps an
d

d
etergents
,
x Currentl
y
the reaction is carried out usin
g
cobalt based catal
y
sts with severe
p
ena

l
t
i
es
i
n terms of
h
ars
h
operat
i
n
g
con
di
t
i
ons
(
80
b
ar CO
/H
2
,

200
o
C
)

. In
addition, substantial loss of substrate
(
ca.
10%) to hydrogenation makes the
overa
ll
se
l
ect
i
v
i
ty to t
h
e
li
near a
l
co
h
o
l
ca.
8
0% [15]. R
h
o
di
um

b
ase
d
systems
are capable of
g
ivin
g
hi
g
her selectivities (>90%) to the desired linear alde
y
de
p
ro
d
uct un
d
er m
ild
er con
di
t
i
ons
(
20
b
ar, 10
0

o
C
)
[13]
x
Th
e react
i
on
h
as
b
een stu
di
e
d
us
i
ng a
ll
of t
h
e
di
fferent poss
ibl
e separat
i
on
methods and represents a s

y
stem where advanta
g
es and disadvanta
g
es of the
var
i
ous processes can
b
e compare
d
[5
]
x Desp
i
te t
h
e very attract
i
ve propert
i
es of t
h
e r
h
o
di
um-
b

ase
d
system, no
commerc
i
a
l
p
l
ants use
d

i
t
b
ecause t
h
e
l
ow sta
bili
ty of t
h
e cata
l
yst meant t
h
a
t

the catal
y
st separation problem prevented commercialisation. Ver
y
recentl
y
,
t
hi
s s
i
tuat
i
on
h
as c
h
an
g
e
d
w
i
t
h
t
h
e
i
ntro

d
uct
i
on of r
h
o
di
um-
b
ase
d
p
l
ant
by

Sasol in South Africa which uses technolo
gy
developed b
y
Kvaerner Process
T
ec
h
no
l
ogy
(
now Davy Process Tec
h

no
l
ogy
)
. T
hi
s
b
atc
h
cont
i
nuous p
l
an
t

o
ut
b
y
l
ow
pressure
di
st
ill
at
i
on [16-18]

In the final Cha
p
ter of the book (Cha
p
ter 9), all the different
p
rocesses are com
p
are
d

wi
t
h
a
di
scuss
i
on of t
h
e var
i
ous areas w
h
ere furt
h
er researc
h
w
ill

b
e requ
i
re
d
to
i
mprove t
h
e new processes to a po
i
nt w
h
ere t
h
ey may
b
e commerc
i
a
ll
y attract
i
ve.
p
r
o
d
u

d
c
e
s
m
e
d
i
u
m
- long chain aldehydes
m
and the separation is carried
o
o
D
.
J
.
CO
LE-HAMILT
O
N AND R. P. T
OO
ZE
8
1
.4 Re
f
erences

[
1
]
C. J. A
d
ams,
Th
e Nort
h
American Cata
l
yst Society
,
[
2] 'App
li
e
d
Homo
g
enous Cata
ly
s
i
s w
i
t
h
Or
g

anometa
lli
c Compoun
d
s', e
d
. B. Corn
il
s an
d
W. A.
Herrmann, VCH, Weinheim, 1996.
[
3] Q. H. X
i
a, H. Q. Ge, C. P. Ye, Z. M. L
i
u, an
d
K. X. Su,
Ch
em. Rev
.
,
2005
,

1
05
,

1603.
[
4] T. Burgi and A. Baiker,
A
ccounts Chem. Res
.
, 2004, 3
7
, 909.
[
5] D. J. Cole-Hamilton,
S
cience,
2003
,
299
,
1
7
02
.
[
6
]
C. C. Tzsc
h
uc
k
e, C. Mar
k

ert, W. Bannwart
h
, S. Ro
ll
er, A. He
b
e
l
, an
d
R. Haag,
A
ngew. C
h
em Int.
E
dit
.
,
2002
,
41
,
3964
.
[
7] E. Drent,
E
ur. Pat.
,

1984
,
0106.
[
8] R. P. Tooze, K. Whiston, A. P. Mal
y
an, M. J. Ta
y
lor, and N. W. Wilson,
J
.
C
hem.
S
oc Dalton
T
r
ans.
,
2000
,
3441.
[
9] W. G. Reman, G. B. J. De
b
oer, S. A. J. Van Langen, an
d
A. Na
h
u

ij
sen,
Eu
r
.

Pat.
,
1991
,
0411.
[
10] W. Cle
gg
, G. R. Eastham, M. R. J. Else
g
ood, R. P. Tooze, X. L. Wan
g
, and K. Whiston,
C
hem
.
C
ommun
.
,

1999
,
18

77.
[
11
]
K. L. O
li
ver an
d
F. B. Boot
h
,
Hud
r
oca
r
bon
Pr
ocess.
,
1970, 49, 112.
[
12] E. Y. X. C
h
en an
d
T. J. Mar
k
s,
Ch
em. Rev

.
,
2000
,
10
0
,
1391.
[
13] 'R
h
o
di
um cata
l
yse
d

h
y
d
roformy
l
at
i
on', e
d
. P. N. W. M. Van Leeuwen an
d
C. C

l
aver, K
l
uwer,
D
ordrecht, 2000.
[
14] C. D. Fro
hli
ng an
d
C. W. Ko
hl
pa
i
ntner,
i
n 'App
li
e
d
Homogeneous Cata
l
ys
i
s w
i
t
h
Organometa

lli
c
C
om
p
ounds', ed. W. A. Herrmann, VCH, Weinheim, 1996.
[
15] B. Corn
il
s,
i
n 'New synt
h
es
i
s w
i
t
h
Car
b
on Monox
id
e', e
d
. J. Fa
lb
e, Spr
i
nger Ver

l
ag, Ber
li
n, 1980.
[
16] J. A. Banister and G. E. Harrison,
US
Patent,
2004
,
0186323
.
[
17]
Ch
em. En
g
. News
.
,
1999
,
77
,
19.
[
18]
C
hem. En
g

. News
.
,
2004
,
82
,

29.
h
ttp://www.nacatsoc.or
g
/e
d
u_
info.asp?edu
_
infoID=1
.
C
HAPTER 2
C
LA
SS
I
C
AL H
O
M
OG

E
N
E
OUS

C
ATALY
S
T
S
EPARATI
ON
TE
C
H
NO
L
OG
Y
D
A
VID R. BRY
A
NT
I
ntellectual Property and Technology, Inc.
1201 Sha
d
y Way
Sout

h
C
h
ar
l
eston
,
WV 25309-2419
U
SA
d

Coverage of Chapter
W
hen considering a separation technique for a homogeneous catalytic process, one
must realize that catal
y
st/product/b
y
product separation is an inte
g
ral part of the entire
process. T
h
e se
l
ect
i
on an
d

d
es
i
gn of t
h
e separat
i
on tec
h
no
l
ogy goes
h
an
d
-
i
n-
h
an
d
w
i
t
h
catalyst design, often in an iterative fashion. That is, a catalyst is selected and tested in
a continuous unit, with rec
y
cle of streams, to discover if there are problems that will

necess
i
tate re
d
es
ig
n of t
h
e cata
ly
st. Re
d
es
ig
n
i
s more often t
h
e fact t
h
an t
h
e except
i
on.
T
h
e o
bj
ect

i
ve of t
hi
s c
h
apter
i
s to
d
eta
il
cons
id
erat
i
ons t
h
at must
b
e a
dd
resse
d

i
n
order to successfully marry a catalyst technology with catalyst/product separation tech
-
nolo
gy

. The focus of this chapter is h
y
droform
y
lation, but the
g
eneral principles should
app
ly
to man
y

h
omo
g
eneous prec
i
ous-meta
l
cata
ly
ze
d
processes.
2.2 General Process Considerations
T
here are four principal factors that are paramount in selectin
g
the best separation tech
-

n
i
que. T
h
e
y
are t
h
e ener
gy
requ
i
re
d
for t
h
e separat
i
on, t
h
e cap
i
ta
l
requ
i
re
d
for t
h

e
equ
i
pment use
d

i
n t
h
e separat
i
on, t
h
e eff
i
c
i
ency
/
effect
i
veness of t
h
e separat
i
on, an
d
t
h
e

vitality of the catalyst after the separation. General process considerations include:
x
T
ransitions of any type including temperature, pressure or phase changes
sh
ou
ld

b
e m
i
n
i
m
i
ze
d
.
x Cooling below 40 degrees Celsius becomes more expensive (river water can
-
n
ot
b
e use
d)
.
x Vacuum below 20 mm Hg is challenging.
x Bypro
d
uct format

i
on s
h
ou
ld

b
e m
i
n
i
m
i
ze
d
. S
i
ng
l
e pro
d
uct processes are
b
et
-
ter. A distillation column, or other ste
p
, will be re
q
uired for each material in

t
h
e m
i
xture.
2.1
©
2006 Springer. Printe
d
in the Netherlan
d
s.
9
9–37.
9
D
.
J
J
J
C
o
C
C
l
e
-
H
a
H

H
m
i
l
t
o
n
a
n
d
R
.
P
.
T
o
T
T
o
z
e
(
e
(
(
d
s
d
d
.

)
,
C
a
C
C
t
a
l
y
l
l
s
y
y
t
S
e
S
S
p
e
a
r
a
t
i
o
n
,

R
e
c
o
v
e
r
y
r
a
n
d
R
e
c
y
c
c
l
i
n
g


,
g
g
D
. R. BRY
A

NT
10
x
E
veryt
hi
ng feas
ibl
e s
h
ou
ld

b
e recyc
l
e
d
so as to m
i
n
i
m
i
ze waste.
x
Pressures s
h
ou
ld

b
e
k
ept
b
e
l
ow 35
b
ar, at
l
east
b
e
l
ow 100
b
ar, to m
i
n
i
m
i
ze
costs and because most process design experience is here.
x
Th
e use of rotat
i

n
g
equ
i
pment suc
h
as compressors or centr
i
fu
g
es s
h
ou
ld

b
e
minimiz
ed
t
o
minimiz
e
maint
e
nan
ce

cos
t

s.
x
Corros
i
ve mater
i
a
l
s, part
i
cu
l
ar
ly
c
hl
or
id
e, s
h
ou
ld

b
e avo
id
e
d
.
x Batch o

p
erations should be avoided.
x
Th
e
h
an
dli
ng of so
lid
s s
h
ou
ld

b
e avo
id
e
d
.
2
.3 Ever
y
th
i
n
g

i

s a Reactor
Thi
s may
b
e a goo
d
t
i
me to
i
ntro
d
uce a very s
i
mp
l
e pr
i
nc
i
pa
l
of process c
h
em
i
stry,
b
ut
one that is not widely recognized. It is taught in chemical engineering that the only

thin
g
s in chemistr
y
that matter are temperature and concentration. Ever
y
other variable
can
b
e re
d
uce
d
to t
h
ese two. For examp
l
e, t
i
me
i
s s
i
mp
l
y a ref
l
ect
i
on of c

h
ang
i
ng con-
ce
ntrati
o
n
.
N
ow a corollar
y
: since ever
y
piece of process equipment has associated with i
t

temperature an
d
concentrat
i
on, a
ll
p
i
eces of process equ
i
pment are reactors. State
d

di
f-
f
erent
l
y, everyt
hi
ng
i
s a reactor
.
There is a tendenc
y
to think that once the catal
y
st is removed from the reactor, all
c
h
em
i
stry ceases. C
h
em
i
stry
i
s occurr
i
ng t
h

roug
h
out t
h
e process, an
d
t
h
at
i
s w
h
y sepa
-
rat
i
on of pro
d
ucts cannot
b
e v
i
ewe
d

i
n
i
so
l

at
i
on from t
h
e process t
h
at ma
d
e t
h
em.
2
.4 Overv
i
ew o
f

S
eparat
i
on Technolo
gi
es
2
.4.1 TR
A
DITI
O
N
A

L
CO
B
A
LT
W
ITH
CA
T
A
LY
S
T DE
CO
MP
OS
ITI
O
N
T
raditional cobalt h
y
droform
y
lation separations will not be covered in detail since the
y

h
ave
b

een
d
escr
ib
e
d

i
n many exce
ll
ent references.[1] A
k
ey factor
i
n un
d
erstan
di
ng
co
b
a
l
t
h
y
d
roformy
l
at

i
on con
di
t
i
ons an
d
co
b
a
l
t
/
pro
d
uct separat
i
ons
i
s to recogn
i
ze t
h
a
t

cobalt is a relativel
y
unreactive catal
y

st that requires hi
g
h temperatures to achieve
commerc
i
a
ll
y v
i
a
bl
e rates. Co
b
a
l
t car
b
ony
l
s
h
ave
li
m
i
te
d
t
h
erma

l
sta
bili
ty. By us
i
ng
hi
g
h (200 bar) partial pressures of s
y
n
g
as (CO/
H
2
), thermal stabilit
y
is achieved durin
g
hyd
roform
yl
at
i
on. However, to separate t
h
e co
b
a
l

t cata
ly
st from t
h
e
hyd
roform
yl
at
i
on
p
ro
d
ucts t
h
e pressure must
b
e re
d
uce
d
. Separat
i
on
i
s ac
hi
eve
d

b
y
d
ecompos
i
ng t
h
e
catal
y
st in a step referred to as decobaltin
g
. There have been a variet
y
of techniques
di
sc
l
ose
d
for ac
hi
ev
i
n
g
t
hi
s

g
oa
l
.[2]
A ma
j
or a
d
vance
i
n
h
omogeneous cata
l
ys
i
s was t
h
e
i
ntro
d
uct
i
on of a tr
i
a
lk
y
l-

p
hosphine to supplement the role of carbon monoxide in catalyst stabilization.[3] A
lig
an
d
mo
di
f
i
er suc
h
as tr
i
a
lkyl
p
h
osp
hi
ne serves t
h
ree pr
i
nc
i
pa
l
ro
l
es

i
n a
h
omo
g
eneous
cata
l
yt
i
c process. It sta
bili
zes t
h
e meta
l
,
i
t
i
nf
l
uences t
h
e react
i
on rate, an
d

i

t
i
nf
l
uences
p
rocess selectivity.
C
L
ASS
I
CA
L
S
EP
A
R
A
TI
O
N TE
C
HN
O
L
OG
Y
11
In co
b

a
l
t
h
y
d
roformy
l
at
i
on, t
h
e tr
i
a
lk
y
l
p
h
osp
hi
ne prov
id
es a more t
h
erma
ll
y sta
bl

e
catalyst so that decobalting is not required. Its influence on reaction rate is not a desir
-
able one in that the TOF (turnover frequenc
y
) of the cobalt is reduced with the
consequence t
h
at
high
er operat
i
n
g
temperatures are nee
d
e
d
to ac
hi
eve commerc
i
a
l
rates. F
i
na
ll
y, t
h

e tr
i
a
lk
y
l
p
h
osp
hi
ne s
i
gn
i
f
i
cant
l
y a
l
ters process se
l
ect
i
v
i
ty. Rat
h
er t
h

an
making mainly aldehydes, as is the case with unmodified cobalt, the principal product
i
n phosphine-modified cobalt catal
y
sis is the correspondin
g
alcohol. For man
y
alkenes
t
hi
s
i
s not un
d
es
i
ra
bl
e s
i
nce
hi
g
h
er mo
l
ecu
l

ar we
i
g
h
t a
ld
e
h
y
d
es wou
ld
pro
b
a
bl
y
b
e re-
d
uced to alcohol in subsequent processing steps. For butanal, however, the
c
ir
cu
m
s
tan
ce
i
s

d
iff
e
r
e
nt
.
2
.4.2 UNI
O
N
CA
RBIDE-D
A
VY
GAS
RE
C
Y
C
LE PR
OC
E
SS

Buty
l
a
l

co
h
o
l

i
s not t
h
e pr
i
nc
i
pa
l
use of
b
utana
l
o
b
ta
i
ne
d

b
y propene
h
y
d

roformy
l
at
i
on.
Rather its major market is 2-ethylhexanol that is prepared via aldol condensation fol
-
l
owed b
y
h
y
dro
g
enation.
[
4
]
Thus formation of alcohols when aldeh
y
des are desired is
n
ot on
ly
a
di
rect eff
i
c
i

enc
y

l
oss,
b
ut a
l
so t
h
e a
l
co
h
o
l

i
mpur
i
t
y
w
ill
form
h
em
i
aceta
l

s
an
d
aceta
l
s t
h
at comp
li
cate ref
i
n
i
ng an
d

l
ea
d
to
i
ncrease
d
operat
i
ng costs.
A breakthrough in hydroformylation was achieved with the introduction of a tri-
ar
y
lphosphine-modified, in particular triphen

y
lphosphine-modified, rhodium
cata
l
yst.
[
5
] T
hi
s
i
nnovat
i
on prov
id
e
d
s
i
mu
l
taneous
i
mprovements
i
n cata
l
yst sta
bili
ty,

react
i
on rate an
d
process se
l
ect
i
v
i
t
y
. A
ddi
t
i
ona
lly
, pro
d
ucts cou
ld

b
e separate
d
from
cata
l
yst un

d
er
h
y
d
roformy
l
at
i
on con
di
t
i
ons. One var
i
ant
i
s
d
escr
ib
e
d
as Gas Recyc
l
e
(
Fi
g
ure 2.1) since the products are isolated from the catal

y
st b
y
vaporization with a
l
ar
g
e rec
y
c
l
e of t
h
e reactant
g
ases.[6] T
h
e rec
y
c
l
e
g
as
i
s c
hill
e
d
to con

d
ense
b
utana
l
s.
p
ropylene
CO
/
H
2
To Str
i
pper an
d
P
roduct Recover
y
Li
qu
id
-Vapor
S
eparato
r
Reacto
r
C
on

d
ense
r
C
y
cl
e
Com
p
ressor
Figure 2.1.
G
as Recycle Hydroformylation Process
In t
h
e pract
i
ce of
g
as rec
y
c
l
e
hyd
roform
yl
at
i
on [7], r

h
o
di
um comp
l
ex an
d
tr
i
p
h
en
yl-
ph
osp
hi
ne are
di
sso
l
ve
d

i
n a su
i
ta
bl
e so
l

vent. T
h
e reactor
i
s pressur
i
ze
d
w
i
t
h
t
h
e
D
. R. BRY
A
NT
12
reactants, carbon monoxide, hydrogen and propene. The entering gas is passed through
the catal
y
st solution and becomes saturated with aldeh
y
de. Gas exitin
g
the reactor is
c
hill

e
d
to con
d
ense
b
utana
l
, an
d
t
h
e reactant gases are compresse
d
an
d
returne
d
to t
h
e
r
e
a
c
t
o
r
.
One a

d
vanta
g
e of Gas Rec
y
c
l
e operat
i
on
i
s t
h
at t
h
e cata
ly
st rema
i
ns
i
n t
h
e reacto
r

an
d

i

s t
h
us a
l
ways wor
ki
ng. T
hi
s re
d
uces t
h
e
i
nventory of t
h
e expens
i
ve prec
i
ous meta
l
.
Another advanta
g
e is that part of the heat of reaction is used to vaporize aldeh
y
de prod-
ucts. A
d

owns
id
e
i
s t
h
e energy consumpt
i
on of t
h
e recyc
l
e compressor. Anot
h
er
d
ownside is that the large gas flows through the catalyst solution expand its volume
s
uch that a
g
reater reactor volume is required resultin
g
in increased capital cost.
Gas Rec
y
c
l
e
i
s a re

l
at
i
ve
ly
s
i
mp
l
e operat
i
on. Rat
h
er t
h
an
b
e
i
n
g
c
i
rcu
l
ate
d
t
h
rou

gh

a var
i
ety of p
i
pes, pumps an
d
co
l
umns, t
h
e cata
l
yst rema
i
ns
i
n one p
l
ace. A
k
ey contro
l

variable is maintaining a constant liquid level in the reactor. This is not as simple as it
mi
g
ht first seem because in addition to butanal isomers formin
g

, butanal condensation
p
ro
d
ucts
i
nc
l
u
di
ng
di
mers an
d
tr
i
mers a
l
so form to g
i
ve w
h
at are co
ll
ect
i
ve
l
y terme
d

“
h
e
a
v
i
es
”
.
Heav
i
es format
i
on
i
s acce
l
erate
d

b
y a var
i
ety of mater
i
a
l
s.[8] Successfu
l

Gas Re-
cycle operation depends on keeping the catalyst solution as pristine as possible to limit
heavies formation since in Gas Recycle there is no independent way to remove heavies.
T
here are a sin
g
le set of conditions for product formation, product removal and b
y
-
p
ro
d
uct
(h
eav
i
es
)
remova
l
. A
k
ey to successfu
l
operat
i
on
i
s
id

ent
i
fy
i
ng con
di
t
i
ons
under which the heavies can be removed essentially at their rate of formation. A down-
s
ide of Gas Rec
y
cle is that it ma
y
be difficult to recover from upsets in operation,
whi
c
h
resu
l
t
i
n t
h
e cata
ly
st so
l
ut

i
on conta
i
n
i
n
g
a
di
sproport
i
onate amount of
h
eav
i
es.
Gas Rec
y
cle technolo
gy
has been licensed worldwide b
y
Union Carbide-Dav
y
fo
r

t
h
e

hyd
roform
yl
at
i
on of propene.[9] It
h
as a
l
so
b
een operate
d

by
Un
i
on Car
bid
e fo
r

et
h
ene
h
y
d
roformy
l

at
i
on. Its use w
i
t
h

b
utene
i
s feas
ibl
e,
b
ut at t
h
e marg
i
n of opera
bil-
i
t
y
. Liquid Rec
y
cle, described below, is a better option for butene.
In sp
i
te of
i

ts
li
m
i
tat
i
ons, Gas Rec
y
c
l
e tec
h
no
l
o
gy
rema
i
ns a v
i
a
bl
e opt
i
on
i
n cer-
ta
i
n c

i
rcumstances w
h
ere
i
ts se
l
ect
i
on may
b
e favore
d

b
y p
l
ant sca
l
e or capac
i
ty. Ot
h
e
r

keys to the decision are as mentioned earlier: energy consumption, capital investment,
s
eparation efficienc
y

and catal
y
st vitalit
y
.
2
.4.3 LIQUID RECYCLE
In Liquid Recycle, the conditions for the reaction are decoupled from those for the
s
eparat
i
on s
y
stem.[10] D
i
st
ill
at
i
on
i
s a w
id
e
ly
pract
i
ce
d
an

d
we
ll
-un
d
erstoo
d
tec
h
no
l-
ogy, so
i
t
i
s genera
ll
y t
h
e f
i
rst cons
id
erat
i
on for any
h
omogeneous cata
l
yt

i
c process. A
t
y
pical Liquid Rec
y
cle s
y
stem is shown in Fi
g
ure 2.2
.
C
L
ASS
I
CA
L
S
EP
A
R
A
TI
O
N TE
C
HN
O
L

OG
Y
13
propylene
C
O
/
H
2
Catalyst Recycle
Heavy By-products
Al d
e
hyd
e
s
y
P
r
P
P
opane P
u
r
ge
1
3
4
5
2

Figure
2
.
2
. B
l
oc
k
F
l
ow D
i
agram for a L
i
qu
id
Recyc
l
e Process
Propene and s
y
n
g
as are fed to a reactor (1 in Fi
g
ure 2.2) where the
g
ases are intimatel
y

contacte
d
w
i
t
h
an or
g
anop
h
osp
h
orus-mo
di
f
i
e
d
r
h
o
di
um cata
ly
st. T
h
e exot
h
erm
i

c
h
ea
t

of react
i
on
i
s contro
ll
e
d
w
i
t
h

h
eat exc
h
anger
(
2
)
. Eff
l
uent from t
h
e reactor passes to a

column (3) where the solution is de
g
assed. Propane in the c
y
cle from h
y
dro
g
enation of
p
ropene
i
s vente
d
a
l
on
g
w
i
t
h
some propene an
d
s
y
n
g
as. From t
h

e
d
e
g
ass
i
n
g
co
l
umn,
the catalyst solution passes to column (4) where aldehyde products and condensation
b
y
products are separated from catal
y
st solution. The catal
y
st solution is rec
y
cled to the
reactor, an
d
t
h
e pro
d
uct m
i
xture

i
s transferre
d
to co
l
umn
(
5
)
w
h
ere
i
so
l
at
i
on of t
h
e
b
u-
tanal
occu
r
s.
Fee
d
to ta
il

s rat
i
o ma
y

b
e
d
ef
i
ne
d
as t
h
e rat
i
o
b
etween t
h
e
li
qu
id
fe
d
to co
l
umn
(

4
)
an
d
t
h
e
li
qu
id

i
n t
h
e cata
l
yst recyc
l
e. H
i
g
h
er fee
d/
ta
il
s rat
i
os contr
ib

ute to
hi
g
h
er con
-
version since with onl
y
catal
y
st and heav
y
solvent bein
g
rec
y
cled more of the reacto
r

vo
l
ume
i
s ava
il
a
bl
e for pro
d
uct.

Whereas in Gas Recycle the product must be removed at the same temperature and
p
ressure at w
hi
c
h

i
t
i
s forme
d
,
i
n L
i
qu
id
Rec
y
c
l
e t
h
e separat
i
on of pro
d
uct
(

an
d

by-
p
ro
d
ucts
)
from cata
l
yst
i
s
i
n
d
epen
d
ent of t
h
e con
di
t
i
ons un
d
er w
hi
c

h
t
h
e pro
d
ucts were
f
ormed. This added de
g
ree of control brin
g
s a variet
y
of benefits. Since lar
g
e
g
as flows
are no
l
on
g
er requ
i
re
d

i
n t
h

e reactor, t
h
e
li
qu
id
expans
i
on
d
ue to
g
ass
i
n
g

i
s re
d
uce
d
an
d

more catalyst can be contained in a specific reaction vessel. Reactor temperature and
reactant concentrations can be tuned for optimum catal
y
st performance. The conditions
i

n t
h
e separat
i
on system can
lik
ew
i
se
b
e tune
d
for opt
i
mum performance. In part
i
cu
l
ar,
more severe conditions will
p
ermit better control over the concentration of heavies in
t
h
e cata
ly
st so
l
ut
i

on.
T
h
e more concentrate
d
t
h
e cata
l
yst ex
i
t
i
ng t
h
e separat
i
on system
(
vapor
i
zer
)
an
d

bein
g
returned to the reactor, the hi
g

her the concentration of product in the effluen
t

f
rom t
h
e reactor. H
i
g
h
er pro
d
uct concentrat
i
on means fewer passes of t
h
e cata
l
ys
t

D
. R. BRY
A
NT
14
t
h
roug
h

t
h
e vapor
i
zer for a g
i
ven pro
d
uct
i
on, an
d
fewer passes means
hi
g
h
er eff
i
c
i
en-
cies in the conversion of raw materials to products since each time catal
y
st is remove
d

from t
h
e reactor some unconverte
d

reactants w
ill

b
e
l
ost.
Another advantage of Liquid Recycle is that multiple reactors may be arranged in
s
er
i
es w
i
t
h
t
h
e eff
l
uent from one pass
i
n
g
on to t
h
e next. T
h
e a
lk
ene concentrat

i
on
i
s
l
ess
i
n t
h
e
d
ownstream reactors,
b
ut react
i
on con
di
t
i
ons can
b
e a
dj
uste
d
to opt
i
m
i
ze

each reactor’s
p
erformance. In back mixed reactors in continuous o
p
eration, the efflu
-
ent from t
h
e reactor
i
s t
h
e same as t
h
e cata
l
yst so
l
ut
i
on t
h
roug
h
out t
h
e reactor. By
p
lacing reactors in series, the first reactor can be optimized for high rates and later reac
-

tors for hi
g
h conversion.
T
h
ere are some
d
owns
id
es to L
i
qu
id
Recyc
l
e operat
i
on. T
h
e f
i
rst
h
as
b
een referre
d

to as thermal de
g

radation [11] or thermal shock, althou
g
h this term su
gg
ests that onl
y
temperature
i
s respons
ibl
e,
b
ut remem
b
er t
h
at
i
n c
h
em
i
str
y
t
h
e two
k
e
y

var
i
a
bl
es are
temperature and concentration.[12] What one observes is that the catalyst may become
l
ess active or even less soluble when passed through a vaporizer or when exposed to
car
b
on monox
id
e an
d

hyd
ro
g
en
i
n t
h
e a
b
sence of a
lk
ene. T
h
e successfu
l

d
eve
l
opment
of a
h
omogeneous cata
l
yt
i
c process requ
i
res t
h
e c
l
ose cooperat
i
on of
b
ot
h
c
h
em
i
sts an
d

engineers to manage the tradeoffs as product is separated from catalyst.
2
.4.4 BIPHASIC SYSTEMS; WATER-ORGANIC
Cons
id
era
bl
e wor
k

h
as
b
een con
d
ucte
d
on a water-so
l
u
bl
e cata
ly
st us
i
n
g
su
l
fonate

d

ph
osp
hi
ne-mo
di
f
i
e
d
r
h
o
di
um. Deta
il
s of t
hi
s c
h
em
i
stry w
ill

b
e
d
escr

ib
e
d

i
n C
h
apter 5.
T
he general concept (Figure 2.3) is to make the catalyst water soluble, then after prod
-
uct formation, decant the product. In order for the water-soluble catal
y
st to be effective,
t
h
e a
lk
ene must
di
sso
l
ve
i
n t
h
e aqueous
l
ayer. T
hi

s
h
as
b
een
d
emonstrate
d
on a com
-
merc
i
a
l

b
as
i
s us
i
ng propene. T
h
e
l
ow so
l
u
bili
ty of
hi

g
h
er a
lk
enes
i
n t
h
e aqueous
catal
y
st la
y
er has proven problematic. The desirable characteristic of the li
g
and, water
s
o
l
u
bili
t
y
,
i
s nee
d
e
d

i
n t
h
e separat
i
on step
b
ut
i
s a
di
sa
d
vanta
g
e
i
n t
h
e react
i
on step
.
D
ecante
r
R
e
a
c

t
o
r
Or
g
an
i
cP
h
ase
A
q
ueous
Cata
l
ys
t
Ph
a
se
Figure 2.3. Water-Organic Biphasic Catalyst System
2
.4.5 INDU
C
ED PH
AS
E
S
EP
A

R
A
TI
O
N
An approac
h
t
h
at overcomes t
h
e
di
sa
d
vantage of
h
av
i
ng a
lk
ene an
d
cata
l
yst
i
n separate
p
hases in the reactor(s) is to use a phosphine li

g
and that is less hi
g
hl
y
sulfonated. One
C
L
ASS
I
CA
L
S
EP
A
R
A
TI
O
N TE
C
HN
O
L
OG
Y
15
can prepare cata
l
ysts w

i
t
h
monosu
l
fonate
d
p
h
osp
hi
nes w
hi
c
h
are organ
i
c so
l
u
bl
e. Dur-
i
ng hydroformylation higher alkenes will be in the same phase as the catalyst an
d

s
i
g
nificantl

y
hi
g
her rates will be obtained. To achieve separation, a small amount of
w
ater
i
s a
dd
e
d
so t
h
at p
h
ase separat
i
on occurs.[13] After pro
d
uct separat
i
on, t
h
e cata
-
l
yst
i
s
d

r
i
e
d
an
d
t
h
en returne
d
to t
h
e reactor
(
F
i
gure 2.4
)
.
olefi
n
CO
/
H
2


wa
t
e

r
p
roduct
cata
l
yst
l
aldehyde
y
rec ycl ed wa ter/NMP
y
cata
ly
st rec
y
c
le
yy
wate
r
P
ri mar
y
Water
/
Catalyst
Se
p
aratio
n

C
ata
l
yst
D
ry
i
ng
W
ate
r
E
xtra
c
t
or
dist illed
wate
r
R
eacto
r
Induced Phas
e
Se
p
arato
r
D
ec

ant
e
r
Fi
g
ure 2.4. Induced Phase Separation Flow Dia
g
ram
In t
hi
s process, cata
l
yst so
l
ut
i
on
l
eav
i
ng t
h
e reactor goes to a separator w
h
ere t
h
e sma
ll
amount of water is added to induce
p

hase se
p
aration. The mixture
p
asses to a decante
r

w
h
ere t
h
e cata
l
yst
i
s separate
d
from t
h
e pro
d
uct. T
h
e cata
l
yst stream passes t
h
roug
h

two dr
y
in
g
sta
g
es; the first sta
g
e produces distilled water that is fed to the water extrac
-
tor, t
h
e secon
d
stage comp
l
etes t
h
e
d
ry
i
ng of t
h
e cata
l
yst w
hi
c
h

t
h
en
i
s returne
d
to t
h
e
reactor. The
p
roduct
p
hase from the decanter is sent to the water extractor to remove
t
h
e NMP use
d
to fac
ili
tate so
l
u
bili
z
i
n
g
t
h

e cata
ly
st.
A
d
vantages of In
d
uce
d
P
h
ase Separat
i
on are t
h
at very
hi
g
h
mo
l
ecu
l
ar we
i
g
h
t a
l-
kenes can be h

y
droform
y
lated and the aldeh
y
de product and b
y
products can be
s
eparate
d
w
i
t
h
out t
h
e cata
l
yst suffer
i
ng “t
h
erma
l
s
h
oc
k
”. D

i
sa
d
vantages
i
nc
l
u
d
e a more
l
imited ligand selection and the removal of water that has a high heat of vaporization.
In a
ddi
t
i
on, t
hi
s tec
h
no
l
o
gy

i
s, as
i
s t
h

e water-so
l
u
bl
e su
l
fonate
d
cata
ly
st,
li
m
i
te
d
to t
h
e
f
ormation of non
p
olar
p
roducts.
2
.4.6 NON-AQUEOUS PHASE SEPARATION
A ma
j
or

b
rea
k
t
h
rou
gh

i
n separat
i
on of pro
d
ucts from cata
ly
st,
i
n part
i
cu
l
ar
h
eat sens
i
-
tive products, came with the discovery of the NAPS or Non-Aqueous Phase Separation
technolo
gy
. NAPS provides the opportunit

y
to separate less volatile and/or thermall
y
l
a
bil
e pro
d
ucts. It
i
s amena
bl
e to t
h
e separat
i
on of
b
ot
h
po
l
ar [14] an
d
non-po
l
ar [15]
p
roducts, and it offers the opportunity to use a very much wider array of ligands an
d

s
eparat
i
on so
l
vents t
h
an pr
i
or-art p
h
ase separat
i
on processes. T
h
e p
h
ase
di
str
ib
ut
i
on
c
h
aracter
i
st

i
cs of t
h
e
li
gan
d
can
b
e tune
d
for t
h
e process. Two
i
mm
i
sc
ibl
e so
l
vents are
D
. R. BRY
A
NT
16
n
orma
ll

y requ
i
re
d
to effect cata
l
yst
/
pro
d
uct separat
i
on,
b
ut
i
n some cases t
h
e pro
d
uct
i
tself ma
y
have the appropriate polarit
y
to behave as either the polar or non-polar sol-
vent. For examp
l
e, an a

li
p
h
at
i
c
h
y
d
rocar
b
on suc
h
as
h
exane
i
s a typ
i
ca
l
non-po
l
ar
s
olvent while acetonitrile and methanol are typical polar solvents. The catalyst system
i
s mo
di
f

i
e
d
to
h
ave po
l
ar
i
t
y
oppos
i
te to t
h
e pro
d
uct. T
h
e
lig
an
d
prov
id
es t
h
e
b
as

i
s for
t
h
e
d
es
i
re
d
cata
l
yst separat
i
on se
l
ect
i
v
i
ty.
2.4.6.1 NAPS Usin
g
a Non-Polar Catal
y
s
t

An alkene which will
g

ive a polar aldeh
y
de product and s
y
n
g
as are introduced into the
reactor conta
i
n
i
ng a non-po
l
ar
li
gan
d
mo
di
f
i
e
d
r
h
o
di
um cata
l
yst. Cata

l
yst so
l
ut
i
on ex
i
t-
i
ng the reactor enters a Flash stage where CO/
H
2
are purge
d
. T
h
e cata
l
yst so
l
ut
i
on t
h
en
enters an extractor where it is contacted with a polar solvent. The product aldeh
y
de is
capture
d

i
n t
h
e po
l
ar so
l
vent
i
n t
h
e extractor, t
h
en concentrate
d

i
n t
h
e So
l
vent Remova
l
Co
l
umn. Po
l
ar So
l

vent
i
s rec
y
c
l
e
d
to t
h
e Extractor. T
h
e Non-Po
l
ar cata
ly
st so
l
ut
i
on
i
s
recycled to the reactor (see Figure 2.5).
ole
f
i
n
CO/H
2

E
xtractor
N
on-Pola
r
C
ata lyst Recycle
F
la
s
h
CO/
H
2
P
ur
g
e
P
o
l
ar
S
o
l
ven
t
R
eactor
P

ro
d
uct
i
n
P
olar Solven
t
P
roduct
S
olven
t
R
emova
l
Column
Fi
g
ure 2.5. NAPS Flow Dia
g
ram for Polar Products