close

Вход

Забыли?

вход по аккаунту

?

Nickel An Element with Wide Application in Industrial Homogeneous Catalysis.

код для вставкиСкачать
Nickel: An Element with Wide Application
in Industrial Homogeneous Catalysis
By Wilhelm Keim *
Dedicated to Professor Giinther Wilke on the occasion of his 65th birthday
The efficiency and future development of the chemical industry are closely linked to catalysis.
It has been estimated, for example, that 60 to 70% of all industrial chemicals have involved
the use of a catalyst at some point during their manufacture. In the past two decades the share
of the market credited to homogeneous transition metal catalysis increasead to 10- 15 %.
Besides cobalt, which is used mainly in hydroformylation reactions, nickel is the most frequently used metal. Many carbon-carbon bond formation reactions can be carried out with
high selectivity if catalyzed by organonickel complexes. Such reactions include, inter alia,
carbonylation reactions, cyclic and linear oligomerization and polymerization reactions of
monoenes and dienes, and hydrocyanation reactions. It was Reppe and Wilke who pioneered
and shaped the field of homogeneous nickel catalysis. Great impetus was also given to the
development of organonickel chemistry by Wilke and his students. Research in this area has
contributed immensely towards an understanding of the reactions involved in catalysis.-This
review is primarily concerned with nickel-catalyzed reactions which are of interest both preparatively and industrially; some mechanistic aspects are also dealt with.
1. Introduction
In the 16th century, miners working in the neighborhood
of Annaberg in Sachsen tried to recover copper from a reddish ore which in fact contained only nickel and arsenic.
They never suspected that this “enchanted copper”, coppernickel, contained a new element that would later play a significant role in industrial catalysis. Nickel, meaning “bad
partner (boy)” in the German language, was later isolated in
pure form from this “enchanted rock” by Cronstedt in 1754.
More than a further hundred years were to pass before
Mond, in 1888, discovered the first carbonylnickel complex,
Ni(CO), . This discovery initiated a vigorous development of
organonickel chemistry. A metal, which existed in the gas
phase and which could be isolated in pure form in this way
fascinated Mond’s contemporaries, one of whom even
wrote: “Mond gave wings to a metal”.
In 1906, Subatier reported the nickel-catalyzed hydrogenation of carbon monoxide to methane, a reaction which
is still used today on an industrial scale to remove carbon
monoxide from a stream of hydrogen gas, as, e.g., in the
production of ammonia. The reverse reaction, the cleavage
of methane (steam reforming), also proceeds via nickelbased catalysis. Sabatier’s biggest contribution to chemistry
was the recognition of these hydrogenation properties, thus
paving the way for catalysis. One of the oldest hydrogenation catalysts is Raney-nickel, which was first used industrially in 1925, and which subsequently became one of the most
important catalysts. In more recent times, however, dissatisfaction with the pyrophoric properties of Raney-nickel led to
the development of Ziegler systems,“] in which, for instance,
nickel salts are reduced with alkylaluminum compounds.
The nature of these catalysts has not been studied in great
[‘I
Prof. Dr. W. Keim
Institut fur Technische Chemie
und Petrolchemie der Technischen Hochschule
Worringerweg 1. 5100 Aachen (FRG)
A n g a . . Chem. I n [ . Ed. Engl. 29 (1990) 235-244
detail, and the ‘‘solutions’’ obtained after the reduction appear to be homogeneous to the eye. A process developed by
Rh6ne Poulenc for the hydrogenation of benzene to cyclohexane using this catalyst is used on a large scale by many
companies.
Many inherent drawbacks of highly dispersed nickel catalysts can be avoided by supporting nickel salts on various
carriers. In this connection, particular mention must be
made of the nickel catalysts used for hydrogenating fats and
those used, in the form of Ni . Mo/Al,O, and Ni/AI,O,, in
crude oil conversions and hydrogenations. Among the industrial applications, the processes for the hydrogenation of
carbon-carbon double bonds are the most widely used, followed by those for the hydrogenation of carbonyl groups,
with those for the hydrogenation of nitrogen-containing
compounds third.
The development of homogeneous nickel catalysis began
over fifty years ago with the pioneering work of Reppe.
2. Nickel in Reppe Chemistry
On searching for possible applications of carbon monoxide, produced as a byproduct in the manufacture of acetylene, Reppe discovered a series of homogeneous catalytic reactions called carbonylations [Eq. (a)-(~)].[”-”~Reaction (a)
HC-CH + CO
+
[Nil
HY
H2C=CH-C-Y
II
(a)
0
H2CzCH2 + CO
+
HY
[Nil
H$-CH2-C-Y
II
(b)
0
R-CH20H + CO
[Nil
R-CH2-C-OH
II
(C)
0
Y = OH. OR. NR2, RS. 02CR
0 VCH Verlugsgeselkhafi mhH. 0-6940
Weinheim. 1990
0570-0X33/90/0303-0235 $02.5010
235
represented the most important route used for acrylic acid
production until the end of the 1960’s. Depending upon the
amount of Ni(CO), used in the reaction, distinction was
made between a stoichiometric, a modified, and a catalytic
variant, the latter of which is still used today by BASE New
methods of acrylic acid synthesis, however, favor a route via
propene oxidation.
The use of ethylene [Eq. (b)] leads to propionic acid, a
process employed by BASF which produces about 50,000
tons per year.
The carbonylation of alcohols [Eq. (c)], especially that of
methanol in the synthesis of acetic acid, has developed into
one of the most important industrial processes. Originally,
nickel was used as catalyst; in the first industrial plant at
BASF, however, cobalt complexes were used as catalyst. Later the Monsanto Process, using rhodium catalysts, proved
superior, and it is currently the preferred process for acetic
acid synthesis.
The carbonylation of esters yields acid anhydrides, a reaction which is used by Tennessee Eastman for the synthesis of
acetic acid anhydride.”. ‘I Rhodium catalysts are also used in
this homogeneous process. Numerous efforts are being made
in industry, however, to replace rhodium by the more economical nickel, and these efforts appear very promising in
the case of the latter process. The variety of potential applications of Equations (a)-(c) is immense, the fundamental
“building blocks’’ are inexpensive, and the process technology is well worked out. It can be anticipated that many further
uses of carbonylation reactions may evolve in the future,
especially if the raw materials derived from crude oil should
be replaced by coal, natural gas, or biomass in the manufacture of synthesis gas. For instance, processes for the manufacture of ethylene from acetic acid, which in the final analysis originates from synthesis gas, are already known; two
steps are involved, namely hydrogenation, yielding ethanol,
followed by dehydration.
One of the most impressive reactions discovered by Reppe
is perhaps the tetramerization of acetylene to cyclooctatetraene.“’] Although this reaction has not, as yet, found commercial application, it has nevertheless played a decisive role
in the development of theoretical chemistry. The planar
C,HiO dianion with its 10 n-electrons and its numerous
complexes have stimulated a plethora of fundamental theoretical investigations.[’ ’1
The element nickel, especially Ni(CO), , heralded the beginning of homogeneous transition metal based catalysis.
Reppe himself wrote: “It became evident throughout these
investigations that it was a lucky choice using nickel carbony1 as a starting material from the beginning of this
work.” [841
3. Nickel in Wilke Chemistry“
-
Up to now, one of the most important discoveries in chemistry in our century has been that of Holzkamp, who, in 1952,
observed that nickel salts can alter the A1R3-catalyzed
“growth reaction” of ethylene to cc-olefins in such a way that
only dimers (butenes) are formed, a phenomenon referred to
in the literature as the “nickel effect”.“ This observation
consequentially led to the discovery of “Ziegler catalysis”, a
reaction which is of utmost importance, both in industry and
in academic research for the production of a variety of polymers.
In 1959, Wilke studied the possibility of using Ziegler catalysts prepared from nickel acetylacetonate and AlEt,OEt
for the polymerization of butadiene. A mixture of products
was obtained: (E,E)-I ,5-cyclooctadiene (COD) (25 YO),4vinylcyclohexene (VCH) (1 0 %), and (E,E,E)-I
,5,9-cyclododecatriene (CDT) (65%). Had Wilke tried only the Ziegler
system NiC12/Et2A1C1,which produces polybutadiene, an
exciting area of organonickel chemistry might have been
overlooked. Cyclic compounds containing large ring systems
were in great demand by preparative chemists, and the synthesis, especially of eight- and twelve-membered ring derivatives, was an exceptional challenge. Following Wilke’s discovery, various ring systems could be produced economically and quite elegantly. Not long after, Wilke observed that
the ratio of C0D:CDT:VCH could be altered by addition of
R,P-ligands (see Table
One of Wilke’s greatest accomplishments was to have recognized for the first time the route
to controlling the distribution of oligomers by ligand varia-
Table 1 . Reaction of butadiene with nickel acetylacetonate and AIEt,OEt;
control of oligomerization using R,P ligands Conditions: 3 0 T , 5 bar.
Product
(c-C,H, J3P
4-Vinylcyclohexene(VCH)
1,s-Cyclooctadiene (COD)
1,5,9-Cyclododecatriene (CDT)
Higher oligomers
(0-Ph-C,H,O),P
without
[”/.I
WI
[“/I
39.6
41.2
14.4
4.8
3.1
96.5
0.2
0.2
10
25
65
-
Wilhelm Keim was born in Oberhausen in 1934. He studied chemistry at the Universities of
Miinster and Saarbriicken before moving to the Max-Planck-Institut fur Kohlenforschung in
Mulheim an der Ruhr, where he obtained his doctorate under the supervision of Professor
G. Wilke. After one year as a postdoc with Professor I: Katz at the Columbia University, New
York, he joined the Shell Development Company in Emeryville, CA. In 1967 he was promoted to
leader ofthe petrochemistry research group, in 1969 to leader of the Department of Petrochemistry, and,finally, in 1972 to leader of Basic Research. In 1973 he accepted an appointment as
Professor and Director of the Institut fur Technische Chemie und Petrolchemie der R W T H
Aachen. Wilhelm Keim is, among other things, a member of the board of the Deutsche
Gesellschaft fur Mineralolwissenschafzen und Kohlechemie, Gesellschaft Deutscher Chemiker,
and the Deutsche Gesellschaft fur Chemisches Apparatewesen, Chemie Technik und Biotechnologie. He also is a member of the supervisory board of Degussa AG.
236
Angew Chem. Int. Ed. Enzl. 29 (1990) 235-244
tion. This so-called “ligand tailoring” is of considerable importance to all scientists working in this area of chemistry.
The ligands are often precisely “tailored” to the process envisaged. Two noteworthy examples of industrial uses are:
Shell’s hydroformylation reaction for converting olefins into
linear alcohols,lt8]and Du Pont’s hydrocyanation of butadiene to adipodinitrile (see Section 4.5).
Over the years, Wilke and his co-workers have studied the
syntheses of the above ring systems quite extensively.
Scheme 1 shows examples of the various possibilities. De-
ou
y
A
L d
\AG
H O O C ( C H 2 Il0COOH
I
0
c--
Piana
c =c
N -
Ves ta m id
8
Scheme 2. Nylon-I 2 derivatives from cyclododecatriene
crowns from hard plastics, and printing molds from nitrile
rubber.
-
metathesis
Scheme 1. Ring syntheses from butadiene.
pending on the reaction conditions, linear and/or cyclic compounds are obtainable, whereby cyclooctadiene and cyclododecatriene are of special industrial importance. Fundamental investigations leading to an understanding of the
impact of various ligands on controlling product selectivity
go back to the work of Heirnb~ch.“~’
Cyclododecatriene is used for the preparation of Nylon12. Scheme 2 outlines this synthesis as practiced by Hiils AG
(18,000 tons per year) and Du P~nt.[’’~
Valuable textiles are
produced from Du Pont’s Qiana@.Owing to its stability of
shape, corrosive resistance, and low-temperature properties,
Vestamid@,manufactured by Hiils AG, has numerous applications in compression cylinders for brakes in automobiles,
in fuel pressure lines, in the soles of sports shoes, in cable
coverings, in anti-corrosive coatings (coatings for wires, steel
pipe construction units) and in the linings of jackets and
coats. Hiils AG also manufactures the speciaI rubber Vestenamer“ from cyclooctadiene, [Eq. (d)] . Characteristic features of Vestenamer are its partial crystallinity and its relatively low molecular weight. The crystallinity is dependent
on the number of trans double bonds present. In blends with
other rubbers, Vestenamer improves the viscosity and is thus
of great help in complicated production procedures, such as
those of brake hoses from EPDM (ethylene-propene-diene
elastomer), corrugated bellows from polychloroprene, tooth
Angru Chem Int Ed Engl 29 (1990) 235-244
Shell produces a selection of industrially interesting olefins
based on cyclooctadiene and cyclododecatriene by the Feast
Process, [Eqs. (e)-(g)] .[201
1
O+lI
C
C
___)
c=c
c-
(el
If)
The successful reactions of organonickel complexes with
dienes automatically led to an extension of their range of
237
8
application by including reactions with monoenes. One of
the most impressive examples of ligand-nickel-control, the
dimerization of propene, was first reported by BogdunoviC
and Wilke et a1.[16*21,231
With trialkyl- or arylphosphanemodified nickel systems, hexenes, 2-methylpentenes, or 2,3dimethylbutenes can be produced depending upon the substituent on the phosphorus (see Table 2). The catalytic
activity is exceptionally high, and even at low temperatures
good turnover numbers can be obtained. It should be noted,
however, that the selectivity leading to 2Jdimethylbutene in
the case of tert-butyldiisopropylphosphanecan be increased
to 96.4% at 70 “C. A mixture rich in branched dimers has
proven useful for increasing the octane number in gasoline,
and is used in the Dimersol process (see Section 4.4.1). 2,3Dimethylbutene has also attracted some industrial attention
since it can be dehydrogenated to 2,3-dimethylbutadiene, a
potential monomer for polymerization. The oxidation of
2,3-dimethylbutenes could provide a route to pinacol.
materials for the production of nylon, is worth mentioning.
Acrylic acid methyl ester dimerizes smoothly with complex 2
to yield monounsaturated esters of adipic acid.r161In general, such a dimerization of functionalized olefins with Ziegler
catalysts is difficult because in most cases the catalysts will
react with the functionalizing group. Unfortunately, the
turnover numbers achieved so far using complexes of type 2
are too low for this synthesis to be applied industrially.
~
Table 2. Dimerization of propene with C,H,NiCI/AICIEt,. Influence of the
ligand R,P on the selectivity. Conditions: 25°C. 10 bar.
R,P
Hexenes [%]
2-Methylpentenes [%]
2,3-Dimethylbutenes [%I
21 6
99
01
73 9
80 3
19 0
81 5
45
98
80 9
11 5
~
Ph,P
Me,P
rPr,tBuP
rPrtBu,P
10
It is not surprising that the possibilities for controlling
carbon-carbon linkages via nickel-phosphane ligand systems
have also led to the investigation of enantioselective syntheses with organonickel complexes.1221
The catalysts consist of
1 :1 adducts of q3-allylnickel halides and optically active
phosphanes. For instance, as depicted in Equation (h), ee
values of 80% could be achieved for 1. Racemic 1 is of
interest as third component (“diene”) in EPDM. A noteworthy example of “ligand tailoring” is displayed by the enantioselective dimerization of styrene with ethylene; (R)-(
- )-3phenyl-I-butene is formed with ee values greater than
95 %.I161
In this context, also Pino’s work on the asymmetric
oligomerization of propylene is of
Other pioneering contributions by Wilke and his co-workers include their mechanistic investigations related to
organonickel chemistry, which have greatly improved our
understanding of the reactions involved in the catalysis. A
large variety of complexes could be isolated which are considered to be reactive intermediates. For the mechanism of
the cyclotrimerization of butadiene, it could be shown that
n-complexes and q ‘q3-allyl systems play a crucial role in the
reaction pathways. Scheme 3 illustrates this for cyclododecatriene synthesis. Using bis(q3-ally1)nickel as starting complex, three molecules of butadiene orient themselves coordinatively around a “naked” nickel atom with displacement of
1,5-hexadiene. The nickel atom with its six possible coordination sites can coordinate up to three molecules of butadiene, bound either as q4-or q2-coordinated ligands. Two
butadiene molecules combine via a carbon+arbon linkage
to yield complex 3, in which a C,-chain with two q3-allyl
I.
L = isopropyldirnenthylphosphane
Almost exclusively monoenes and dienes are used in the
aforementioned C-C linkage reactions. The inclusion of
functionalized olefins would considerably expand the scope
of application of homogeneous nickel catalysis. Wilke and
co-workers have also attempted to include functionalized
olefins in their catalytic investigations. In this connection,
the linear dimerization of acrylic acid, which leads to starting
238
H
05
Scheme 3. Mechanism of cyclododecatriene synthesis.
Angew. Chem. Inl. Ed. Engl. 29 (1990) 235-244
ligands is bonded to the nickel atom. Formation of an additional C-C linkage leads to complex 4, in which a C12-chain
is arranged around the nickel atom. In this stepwise mechanism the intermediate 5 is formed via a third C-C linkage.
Butadiene displaces the nickel atom from 5; cyclododecatriene is formed and “naked” nickel is liberated and reacts
again with butadiene, thus completing the catalytic cycle.
The complexes 4 and 5 could be isolated, and their structures
have been determined by X-ray crystallography.[221
If phosphane ligands are added to the system (see Table I),
coordination sites for butadiene are blocked and the intermediates 6-8 are formed, in which the nickel atom is so
modified that only two butadiene molecules fit into the coordination sphere of the nickel (see Scheme 4). Cyclooctadiene,
divinylcyclobutane, and vinylcyclohexene are easily derived
from complexes 6-8.
The expression “naked” nickel was introduced by Wilke
in order to describe one of the most important principles of
catalysis, namely that of the necessity of free coordination
sites. There is a very close relationship here to heterogeneous
catalysis which requires free sites for chemisorption-a phenomenon which had already been postulated by Taylor and
is illustrated in 9[271.Some 60 years ago Taylor put forward
the theory that single nickel atoms stick out randomly from
the metal surface, and thus form the “free” sites for the
chemisorption (coordination) of molecules.
Ni
I
Ni
Ni-Ni-Ni
I I
I
Ni
Ni-Ni
Ni
Ni-Ni
I
I 1
I
I I
9
-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-
1 1 1 1 1 1 1 1 1 1 1 1 1
-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-Ni-
1 1 1 1 1 1 1 1 1 1 1 1 1
6
I?
8
i
i
b
Scheme 4. Product distribution in the dimerization of butddiene in dependence
of phosphane addition L=R,P (see Table 1).
The ongoing studies of Wilke’s q’,q3-aIlyl systems led to
numerous “one component catalysts”, which are far less
complex than Ziegler catalysts. These studies opened up a
new area of research, mainly focused on the use of
organometallic compounds as precursor complexes in
homogeneous catalysis, and thus leading to new reactions
which provide a better understanding of catalysis.
4. Further Examples of Homogeneous
Nickel Catalysts Used in Industrial Processes
4.1. Dimerization and Oligomerization
of Monoolefins
Several important criteria, first recognized and postulated
by Wilke, emerge from the catalytic cycles described in
Schemes 3 and 4:
ql- and q3-allyl complexes are reactive intermediates
from which C-C linkages can be derived. A continual
change in the formal oxidation state occurs through oxidative addition and reductive elimination.
- “naked” metal atoms or ligand-metal systems facilitate a
“pre-orientation” of the reactants, thus influencing the
stereochemistry (selectivity) and consequently the type of
product formed. The necessary free coordination sites are
created as a result of ligand dissociation via q l - and q3allyl equilibria.
- isolated, well characterized organometallic complexes can
be used as precursor complexes leading to the “true” catalyst.
The complex chemistry of ql- and q3-allyl systems as we
know it today goes back to the fundamental work carried out
at the Max-Planck-Institut fur Kohlenforschung. Numerous
allyl complexes were synthesized, and their reaction behavior
monitored in detail in a series of systematic investigat i o n ~ . [ ~ ~ In
- ~ this
‘ ] connection, the numerous X-ray structure analyses performed by Kriiger deserve special mention,
for they impressively support formation of the various reaction intermediates proposed for the catalytic cycles.
~
A n p > i . .Chem. lnr. Ed. Engl. 29 (1990) 235 -244
With Wilke’s discoveries, a new era of chemistry developed based on organometallic chemistry applied to homogeneous transition metal catalysis. There are first of all the
students of Wilke who provided a rapid transfer of technological knowhow to industry. But there are also other research groups who-inspired by Wilke’s publications--have
developed their own ideas and concepts.
4.1. I . Dimersol Technology
One of the earliest works on propene dimerization, and
following the Wilke tradition, goes back to Ewers.[291Also
Chauvin from the Institut Franqais du Petrole (IFP) developed nickel catalysts (Ni2e/AlEtC12) that dimerize propene
and/or n-b~tene.[~’-~’]
These systems are highly active,
hence the recycling of catalyst can be dispensed with. The
propene dimers are used extensively, especially in the USA,
for improving the octane number in gasoline and to provide
heptenes and octenes for plasticizers. Since the first plant was
established in the USA in 1977, IFP have given about 45
licences, which include the following variants : in the Dimersol G process, propene, produced by catalytic cracking, is
dimerized to yield C,-olefins useful in octane-number improvements (about 20 plants). The Dimersol X variant has
found application in the dimerization of butene or codimer239
ization of propylene/butene for the preparation of plasticizers. In the Dimersol E process, high-octane gasoline is prepared form FCC-based olefins (FCC = fluid catalytic cracking process for the processing of crude oil).
In this context, the Hiils Octol process must be menti~ned.[~
I-Butene
~]
is first separated off from the corresponding C,-fraction, and the remaining 2-buteneln-butane
mixture is converted into (remarkably linear) octenes on a
supported nickel catalyst. Two commercial processes underline the importance of this new technology, which is in direct
competition with the homogeneous Dimersol route.
Other recent examples include the replacement of oxygen
by sulfur in acetylacetonate ligands to give dithio-p-diketonate species. Nickel complexes of the dithio ligand give rise
to extremely active olefin oligomerization catalysts.[34'
Figure 1 shows the X-ray structure of complex 10, which
fulfills all of the requirements in Scheme 5.[401 Indeed, when
4.i.2. SheIl Higher Olefin Process (SHOP)
While employed at the Shell Oil Company in 1965, the
author was given the task of converting ethylene into value
added chemicals such as a-olefins. As a former student of
Wilke familiar with the background of ligand tailoring the
author used the model shown in Scheme 5 for tackling the
U
Scheme 5 . Proposed model for the formation of free coordination sites 0.
problem. Chelates X'ir were chosen since they favor squareplanar structures with potentially free coordination sites for
formation of an octahedral system.1351
The atoms X and Y
should behave as an electron donor and acceptor, respectively. Finally, it should be possible to exchange the ligands
L and L' for ethylene. In the course of the investigations,
this concept led to the Shell Higher Olefin Process
(SHOP),[36-391 which, with approximately a one million ton
capacity in 1990, is one of the largest applications of homogeneous catalysis by a transition metal, especially nickel.
Fig. 1. Molecular structure of complex 10 in the crystal [see Eq. (k)]
10 is dissolved in toluene and allowed to react with ethylene,
oligomers are produced which are up to 99 YOlinear, of which
about 98% are formed as a-olefins. The olefins up to C,,,
and present in geometric distribution, were isolated by gas
chromatography. Activities of 6000 moles of ethylene per
mole of complex 10 have been observed. If, however, the
reactions are carried out with complex 10 suspended in nhexane, linear polyethylene is produced. The directly marketable a-olefins produced in the oligomerization part of
Shell's SHOP-process are separated by distillation. The remaining, unmarketable a-olefins are isomerized and converted in a third step by metathesis reactions into linear
olefins with internal double bonds. An overview of the products manufactured by Shell is given in Scheme 6. By com-
ethylene---e
1. oligornerization
2,isomerization
dobanol ethoxylates
t
t
- - I&
a-olefins
SHOP1
dobanol
-
Cethoxylation
f I
internal
olefins
3.metathesis
Scheme 6 . Products and secondary
products of the Shell Higher Olefin
Synthesis (SHOP). Dobane, dobanol,
dobanol ethoxylate and dobanol
ethoxysulfate are Shell tradenames for
surface active products (detergents).
240
Angew. Chem. In!. Ed. Engl. 29 (1990) 235-244
bining oligomerization, isomerization, and metathesis, a
high degree of flexibility in producing linear olefins of any
carbon chain length is possible.
To gain a better insight into the reaction mechanism of the
SHOP-process, our research group at Aachen synthesized a
number of FO nickel chelate complexes and tested them as
catalytic precursors.[4L1In general, it may be stated that: all
complexes active in the SHOP-process consist of a chelate
moiety and an organyl ligand part, as in Figure 2. The organo part only stabilizes the complex, while the chelate ligand controls the catalytic activity and selectivity.[42.431
I
c=c
a-Olefine
c=c
organo
part
cheiate
port
\
-e
Fig. 2. General form of precursor complexes leading to catalysts
Assuming that nickel hydride complexes are the active
catalysts, the reactions (i)-(k) offer a plausible explanation
for Ni-H formation, starting from various complexes as
catalyst precursors. In reaction (i) temperatures of 130 "C are
needed for the insertion of ethylene into the cyclopentadienyl
group and formation of a nickel hydride, whereas the nickelcrotyl complex in reaction 6)readily yields a nickel hydride
through butadiene elimination at 40 "C, and, finally, in reaction (k) ethylene inserts into the nickel-phenyl group followed by elimination of styrene. All the reactions (i)-(k)
outlined could be verified chemically. As predicted, the complexes 10-12 only become active for ethylene oligomerization at the temperatures needed for nickel hydride formation
as depicted in (i)-(k).
Ph, ,Ph
n-C=C
Scheme 7. Postulated mechanism of ethylene oligomerization
complex. In the case of complex 13, only in situ NMR studies
indicated that a nickel hydride is
while the
nickel hydride 14 (Fig. 3 ) could be trapped by using
Ph, ,Ph
11
Ph, ,Ph
g1j.i
c k H 3
Ph
Fig. 3 . Molecular structure of the Ni-H complex 14 in the crystal
12
P
h
,
Ph
PPh 3
,Ph
+ C-C 170°C
- Ph3P
*
[;>Ni
H
(k)
Ph
- @c=c
10
Scheme 7 presents a postulated mechanism for ethylene
oligomerization, using complex 13 as a catalyst precursor
Angeu. Chein. h i . Ed. Engl. 29 (1990) 235-244
Ph,PCH,C(CF,),OH as a ligand.14'] It was also possible to
isolate the corresponding ethyl complex of 14, formed by an
ethylene insertion into the nickel hydride bond. On heating,
the nickel-ethyl complex splits off ethylene to give complex
14 again. A variety of nickel hydrides are reported in the
literature that can be considered as additional support for
the postulated mechanism.["* 16,461
The success with TO chelate ligands suggested the use of
other ligands of this type to be included in a systematic study.
The chelate complex 15 catalyzes the linear oligomerization
241
of a - o l e f i n ~ . To
[ ~ our
~ ~ knowledge,
~~~
this is the only complex (catalyst) that produces highly linear dimers ( > 80%),
but unfortunately, until now, the turn over numbers are too
low for an industrial application.[481In this context, also the
work of Cuvell and Masters concerning olefin oligomerization deserves mentioning.'341
E
N
1
<2
+
c=c
,N-SiMe3
(SIMe312N-P,N-S,Me
3
17
polyethylene
P,h ,Ph
cF3
15
14
4.2. Monoolefin Polymerization
The formation of polyethylene using nickel catalysts was
already reported in a patent in 1953.[49JAlso, BogdunoviC
and Wilke reported the formation of polyethylene on using
the catalyst q3-C3H,NiX/AIX3/P(tC,H,),
In our investigations with various chelate ligands, e.g. derivatives of 1,3d i p h o ~ p h a p r o p e n e [ ~or
' ~ (C6H, ,),PCH,COOH,1511 we frequently observed polyethylene formation, especially when a
solvent was used in which the complex was insoluble.[401
Until now, no general predictions can be made as to what
ligand properties are needed to synthesize either oligomeric
olefins or polyethylene, respectively. Generally, it can be said
that starting from a nickel hydride, growth occurs via
ethylene insertion and S-elimination leading to the a-olefins
as shown in Scheme 7, thus giving rise to a geometric distribution (Schulz-Flory distribution) of products.[2s1In agreement with this, and under the assumption that no chain
transfer occurs, the product pattern can consist predominantly of dimers, higher oligomers or even polymers.
Following our work, and in an attempt to react functionalized monoolefins such as acrylic esters, Klubunde also employed complexes of the type shown in Figure 2 in polymerization and copolymerization reactions.[52.531
Of Klubunde's investigations, the formation of polyketones from ethylene and carbon monoxide using complex
10 should be emphasized.1541We were able to demonstrate
that the ligand 16 in combination with (cod),Ni also yielded
,CO-Ph
P h 3 A s = C,
CO-Ph
16
polyketones from ethylene and carbon monoxide. Such
polyketones have potential application as novel industrial
polymers. Also worth noting in this connection are the works
of Senf5'], N o ~ a k iand
~ ~ Drent[591
~]
of Shell, who have used
palladium instead of nickel in their catalytic studies aimed at
the production of polyketones from ethylene and carbon
monoxide.
By reacting amino-bis(imin0)phosphorane 17, (cod),Ni
and ethylene acording to Equation (I), a system results which
catalyzes a novel ethylene polymerization (2,w-polymerization).[601According to Fink, who has thoroughly investigat242
ed and interpreted this system,I6'I polymers with methylbranches arranged at regular intervals are formed [Eq. (m)].
The methyl groups of the polymer chain are formed from the
C(1)-atom of the a-olefins. This can be explained in terms of
a 1,2-hydride shift with concommitant migration of the catalytically active nickel atom along the carbon chain. It
should be emphasized that this type of polymerization-in
contrast to that with Ziegler catalysts-is carried out with
aluminum-free systems.
4.3. Dimerization and Oligomerization of Dienes
The success of the trimerization of butadiene to cyclododecatriene led a number of research groups to further investigations with other 1,3-dienes such as, for example, isoprene.f4, 3 , "1 Until now, however, poor selectivities have
precluded larger industrial applications. An example of an
industrial use is the trimerization of isoprene to 1,5,9-trimethyl-I ,5,9-cyclododecatriene reported by the Takasago
Perfumery
The nickel-catalyzed codimerization of butadiene with
ethylene leading to 1,4-hexadiene, which is useful for the
manufacture of EPDM, has been thoroughly investigated.[,*. 631. Here, rhodium catalysts proved to be superior for
industrial production.'641
4.4. Polymerization of Dienes
A large variety of nickel complexes used for the polymerization of butadienes has been reported.['7. 64*651 Depending
on the nature of the nickel starting salt, cis-1,4-, trans-1,4- or
1,2-polybutadienes and/or mixtures are formed. An explanation for the stereoselectivity may rest on the syn and anti
forms of the q3-allyl system [Eqs. (n, o)]. The choice of anion
@
R
R
'-..-g+
NI
-
os-product
(n)
trans- product
(01
also plays a crucial role; e.g. with q3-C3H,NiI, 1,4-transpolybutadiene is obtained. TeyssV et a1.[6s1contributed to a
better understanding of the observed selectivities, by showAngew. Chem. Int. Ed. Engl. 29 (1990) 235-244
ing that variation of X in q3-allylnickel-X leads to a dramatic change in activity in the series X = CH,COO’ <
CC1,COO’ < CF,C00°.[661
The Goodyear Tire and Rubber Company introduced a
nickel-BF, system for the synthesis of 1,4-cis-polybutadiene.[”l
Bunawerke Huls markets, “Polyol Huls”, a low molecular
weight 1,4-cis-butadiene polymer. This oligomer is prepared
by a nickel-catalyzed reaction and is used as an anticorrosion
agent and as an adhesive material.[681
are released through reductive elimination with re-initiation
of the catalytic cycle. This reaction is another example of
ligand control. By using Ph,P instead of (PhO),P, the reaction path changes [Eq. (r)], and no nickel hydride is formed;
(Ph,P),Ni(CN),
precipitates, and leads to “catalyst
death”.[751
(Ph,P),Ni
+ 2HCN
- 2 PhzP
(Ph,P),Ni(CN),
4.6. Telomerization[671 3 .
4.5. Hydrocyanation
The nickel-catalyzed addition of hydrocyanic acid to
mono- and di-olefins is an elegant method for the synthesis
of alkyl nitriles, which are important starting materials for
the production of amides, amines, and acids. Hydrocyanation. first reported by Taylor,[691
was used by Drinkard of Du
Pont for the reaction with butadiene according to Equations
(p) and (q).[70,711
This reaction is currently used at three
2 8 2 761
A reaction studied in great detail is that of 1,3-dienes with
nucleophiles HY, as shown for butadiene in Equation (s).
The telomer 19 or octatriene is formed. Although these reac-
Z H Y
+
HCN
(---v
[Nil
(PI
CN
y
CN
+
HCN
“I
1
NC{CH&CN
(s)
tions lead to a wide range of products, their use on an industrial scale has not yet been reported.
5. Outlook
Ni -CN
L”L
mechanism of
+ C=C-C=C-C-C-C=C
(ql
plants by Du Pont for producing the starting materials for
nylon. It may be assumed that the primary amines supplied
by Du Pont are produced likewise via a-olefin hydrocyanation. Complexes such as NIL, (L = aryl phosphite) are employed as catalysts for the hydrocyanation. An explanation
of the reaction mechanism goes back to the work of Tolman,
who thoroughly investigated this process.[72- 741 The proposed mechanism is shown in Scheme 8. HCN reacts with
Scheme 8. Proposed
+ H,
butadiene
hydrocyanation. L =
P(OCeH,)s.
NIL, with dissociation of two phosphite ligands to yield 18,
which can be isolated.r731Reaction of 18 with butadiene
yields an q’-nickel complex, from which butene carbonitriles
Angrw Chem Int. Ed. Engl 29 (1990) 235-244
Aside from cobalt, which is used mainly in hydroformylation reactions, nickel is probably the most frequently used
metal in homogeneous transition metal catalysis. This is undoubtedly due, on the one hand, to its economical
and, on the other, to the abundant and diverse reaction possibilities and selectivities. In the present article, primarily
only industrial processes have been considered. About 15 YO
of the established catalytic processes used in the chemical
industry can be ascribed to homogeneous transition metal
catalysis, and many products can be obtained only by these
processes. Nickel plays an important role in this area which
may even increase in the future. Generally, one can expect
that homogeneous transition metal catalysis will attract increasing interest as value added chemicals grow in importance in the chemical industry.[78.791 Considering the potential in the area of specialty chemicals, many uses of nickel
catalysts may be foreseen. Organonickel complex chemistry
is progressing at a considerable rate, and this could provide
the nutrient for many future applications. Of course, the
question may be raised: “Is the current importance of the
element nickel in catalysis due to its nature or due to the
amount of work which has been undertaken in organonickel
complex chemistry?”
Homogeneous catalysis by nickel began with the work of
Reppe, and it is to Wilke’s great merit to have introduced
organonickel chemistry and its application into catalysis. By
isolating model compounds, he was able to gain insights into
the mechanistic pathways of homogeneous catalysis and catalytic cycles. The areas of homogeneous transition metal
catalysis and organonickel chemistry are greatly indebted to
both Reppe and Wilke. In conclusion it would not be amiss
to quote Max Planck:[83J
243
“During the stormy development of all branches of science, it has become apparent that the progress of science
rests first of all upon the success of the individual scientist”.
Received: September 14, 1989 [A 751 IE]
German version: Angew. Chem. 102 (1990) 251
[l] S. J. Lapporte, W. R. Schuett, J. Org. Chem. 28 (1963) 1947.
[2] A. Mullen in J. Falbe (Ed.): New Syntheses with Carbon Monoxide,
Springer, Berlin 1980, pp. 243-308.
[3] G. N. Schrauzer, Adv. Organomet. Chem. 2 (1964) 1-45.
[4] a) G. Wilkinson, F. G. A Stone, E. W. Abel (Eds.): Comprehensive
Organometallic Chemistry, Pergamon, Oxford 1982; b) P. W Jolly in [4a],
Vol. 8 , pp. 773-796.
[5] T. Ohara, T. Sato, N. Shimizu in: Ullmann’s Encyclopedia of Industrial
Chemistry, Vol. A 1, VCH Verldgsgesellschaft. Weinheim 1985, pp. 161 174.
[6] a) A. Mortreux, F. Petit (Eds.): Industrial Applications of Homogeneous
Catalysis, Reidel, Dordrecht 1988; b) M. Roper in [6a], pp. 1- 15.
171 S. W. Polichnowski, J. Chem. Educ. 63 (1986) 206.
181 W. Keim in D. R. Fahey (Ed.): Industrial Chemicals via C,-Processes. ACS
Symp. Srr. 328 (1987) 1-16.
[9] W Keim (Ed.): Catalysis in C , Chemistry, Reidel, Dordrecht 1983.
[lo] J. P. Collman, L. S. Hegedus, J. R. Norton, R. G. Finke: Principles and
Applications of Organo~ransitionMetal Chemistry, University Science
Books, Mill Valley 1987, pp. 613-615.
[ l l ] M. A. Bennett, Adv. Organomet. Chem. 4 (1966) 375-384.
[12] P. W. Jolly, G. Wilke: The Organic Chemistry of Nickel, Vol. 1, Academic
Press, New York 1971.
P. W. Jolly, G. Wilke: The Organic Chemistry of Nickel, Vol. 2, Academic
Press, New York 1975.
P. W. Jolly in [4a], Vol. 8, pp. 615-709.
P. W. Jolly in [4a], Vol. 6 , pp. 3-223.
a) G . Wilke, Angew. Chem. lOO(1988) 190; Angew. Chem. lnt. Ed. Engl. 27
(1988) 185; b) G. Wilke in M. Tsutsui (Ed.): Fundamental Research in
Homogeneous Catalysis, Vol. 3, Plenum, New York 1979, pp. 1-24.
a) P. Heimhach, R. Traunmiiller: Metall-Olefin-Komplexe, Verlag Chemie,
Weinheim 1970, b) P. Heimbach, T. Bartik: An Ordering Concept on rhe
Basis of Alternative Principles in Chemistry, Springer, Berlin 1989.
L. H. Slaugh, R. D. Mullineaux, US-Pat. 3239 569 (1966). Shell Oil Co.
K. Weissermel, H. J Arpe: Industrielle Organische Chemie, VCH Verlagsgesellschaft, Weinheim 1988, pp. 94, 259-260, 279-282.
R. Streck, J. Mol. Carol. 46 (1988) 305.
B. BogdanoviC, Adv. Organomet. Chem. 17 (1979) 105-137.
B. Bogdanovic, B. Henc, B. Meister, H. Pauling, G. Wilke, Angew. Chem.
84 (1972) 1070; Angew. Chem. Inr. Ed. Engl. 11 (1972) 1023.
P. Pino, P. Cioni, M. Galimberti, J. Wei, N. Piccolrovazzi in W. Kaminsky.
H. Sinn (Eds.): Olefrn Pofymerizarion.Springer, Berlin 1988, pp. 269-279.
G. Wilke, B. Bogdanovic, P. Borner, H. Breil, P. Hardt, P. Heimbach, G.
Hermann, H.-J. Kaminsky, W. Keim, M. Kroner, H. Miiller, W. Oberkirch,
E. W. Miiller, J. Schneider, J. Stedefeder, K. Tanaka, K. Weyer, Angew.
Chem. 75 (1963) 10; Angew Chem. I n [ . Ed. Engl. 2 (1963) 105.
G. Wilke, B. Bogdanovic, P. Hardt, P. Heimbach, W. Keim, M. Korner,
W. Oberkirch, K. Tanaka, E. Steinriicke, D. Walter, H. Zimmermann,
Angew. Chem. 78 (1966) 157; Angew. Chem. Int. Ed. Engl. 5 (1966) 151.
G. Wilke, J Organomer. Chem. 200 (1980) 349.
H. S. Taylor, J. Phys. Chem. 30 (1926) 150.
W. Keim, A. Behr, M. Roper in [4a], Vol. 8, pp. 372-429.
J. Ewers, Angew. Chrm. 78 (1966) 593; Angew. Chem. Inr. Ed. Engl. 5
(1966) 584.
Y. Chauvin, J. F. Gaillard, D V. Quang, J. W. Andrews, Chem. Ind. (London) 1974. 375.
a) Y. Chauvin, J. E GaiIlard, J. Leonard, P. Bonnifay, J. W. Andrews, Hydrocarbon Process. 1982, 110; h) J. Leonard, J. G. Gaillard, Hydrocurbon
Process. 1981, 99.
B. Bogdanovik, B Henc, H.-G. Karmann, H . 4 . Ntissel, D. Walter, G.
Wilke, Ind. Eng. Chem. 62 (1970) 34.
R. H. Friedlander. D. J. Ward, F. Obenaus, F. Nierlich, J. Neumeister,
Hydrocarbon Process. 1986, 31.
a) K. J. Cavell, A. F. Masters, J. Chem. Res. Synop. 3 (1983) 72, b) A. F.
Masters, K. J. Cavell, US Pat. 4533651 CSIRO (1985); c) K. J. Cavell in
244
D. M. Bibley, C. D. Chang, R. F. Howe, S. Yurahak (Eds.): Studies in
Surface Science and Catalysis. Vol. 36, Elsevier, Ansterdam 1988, pp. 523.
[351 W Keim, B. Hoffmann, R. Lodewick, M. Peuckert, G. Schmitt, J. Flelschhauer, U. Meier, J. Mol. Curd. 6 (1979) 79.
[36] M. Peuckert, W. Keim, Organomeiallics 2 (1983) 594.
[371 W. Keim, Chem.-lng.-Tech. 56 (1984) 850.
[38] E. R. Freitas, C. R. Gum, Chem Eng. Prog. 75 (1979) 73.
[39] A. Behr, W. Keim, Arabian J. Sci. Eng. 10 (1984) 377.
(401 W. Keim, F. H. Kowaldt, R. Goddard, C. Kriiger, Angew. Chem. 90 (1978)
493; Angel*’.Chem. Int. Ed. Engl. 17 (1978) 466.
[41] W. Keim, A. Behr, B. Gruber, B. Hoffmann, F. H. Kowaldt, U. Kiirschner,
B. Limbacker, F. P. Sisting, Organometallics 5 (1986) 2356.
[42] W. Keim. J. Mol. Catal. 52 (1989) 19.
[43] W. Keim, New J. Chem. 11 (1987) 531.
[44] W. Keim in D. W. Slocum, W. R. Moser (Eds.): Catalytic Transition Meral
Hydrides, Ann. N.Y Acad. Sci. 415 (1983) 191.
[45] U. Muller, W. Keim, C. Kriiger, P. Betz, Angew. Chem. 101 (1989) 1066;
Angen’. Chem. In!. Ed. Engl. 28 (1989) 1011.
[46] D. M. Roundhill, Adv. Organomet. Chem. 13 (1975) 274-354.
I471 W. Keim, A. Behr, G. Kraus, J. Organomet. Chem. 25f (1983) 377.
[481 U. Freudenberg, W. Keim, Chem.-&-Tech. 60 (1988) 1058.
[49] E. F. Peters, A. Zletz, B. L. Evering, Ind. Eng. Chem. 49 (1957) 1879.
[SO] W Keim, R. Appel, S. Gruppe, F. Knoch, Angew. Chem. 99 (1987) 1042;
Angen,. Chem. Int. Ed. Engl. 26 (1987) 1012.
[51] W. Keim, S . Murata, unpublished results.
[52] U. Klabunde. S. D. Ittel, J. Mol. C a r d 41 (1987) 123.
[53] U. Klabunde, R. Mulhaupt, T. Herskovitz, A. H. Janowicz, J. Calabrese,
S. D. Ittel, J Polym. Sci. Polym. Phys. Ed. 25 (1987) 1989.
[54] U. Klabunde, T. H. Tulip, D. C. Roe, S. D. Ittel, J. Organomet. Chem. 334
(1987) 141.
[55] W Keim, A. Behr, B. Limbacker, C. Kriiger, Angew. Chem. 95 (1983) 505;
Angen,. Chem. Int. Ed. Engl. 22 (1983) 503.
[56] W Keim, B. Driessen, unpublished results.
[57] A. Sen, CHEMTECH 1986, 48.
[58] K. Nozaki, US-Pat. 3689460 (1972), Shell Oil Co.
[59] E. Drent US-Pat. 4740625 (19881, Shell Oil Co.
[60] W Keim, R. Appel, A. Storeck, C. Kriiger, R. Goddard, Angew. Chem. 93
(1981) 91; Angew. Chem. Int. Ed. Engl. 20 (1981) 116.
[61] V. M. Mohring, G. Fink, Angew. Chem. 97 (1985) 982; Angew. Chem Int.
Ed. Engl. 24 (1985) 1001.
[62] Jpn. Pat. 7531 138 (1975) Takasago Perfumery Co.; Chem. Abstr. 84
(1976) 164281; Shell Catalog “Shell Fine Chemicals” 1986, 39.
[63] A. C. L. Su, Adv. Organomel. Chem. 17 (1979) 269.
[64] G. W. Parshall: Homogeneous Catalysis. Wiley, New York 1980, p. 63.
[65] M. Julemont, P. Teyssie, Aspects Homogeneous Catal. 4 (1981) 99-141.
[66] J. P. Durand, F. Dawans, P. Teyssie, J. Polym. Sci. Part A 1, 8 (1970)
979.
[67] M. C . Throckmorton, F. S. Farson, Rubber Chem. Techno/. 45, (1972)
268.
[68] R. Streck im GDCh-Kurs Organische Synthesen durch MetallkompiexKatalyse, RWTH Aachen 1988; B. Schleimer, H. Weber, Angew. Makromol. Chem. 16/17 (1971) 253.
[69] B. W. Taylor, H. E. Swift, J. Catal. 26 (1972) 74.
[70] C. A. Tolman, J. Chrm. Educ. 63 (1986) 199.
[71] B. R. James in [4a], Vol.8 , pp. 285-369.
[72] J. C. Draliner, A. D. English, J. P. Jesson, P. Meakin, C. A. Tolman, J. Am.
Chem. SOC.98 (1976) 2156.
[73] C. A Tolman. Chem. Rev 77 (1977) 313.
[74] W. Keim, A. Behr, H.-0. Luhr, J. Weisser, J. Catul. 78 (1982) 209.
[75] W. Keim, A. Behr, J. P. Bioul. J. Weisser, Erdol Kohle, Erdgas, Perrochem.
35 (1982) 436.
[76] A. Behr, Aspects Homogeneous Catal. 5 (1984) 5-61.
[77] G. W Parshall, W A. Nugent. CHEMTECH 1988, 314.
[78] W Keim in [6a], pp. 335-349.
1791 W. Keim in M. Grauani, M. Giongo (Eds.): Fundamental Research in
Homogeneous Catalysis, Vol. 4, Plenum, New York 1984, pp. 131-141.
1801 P. W. Jolly in [4a], Vol. 8, pp. 713-772.
[81] G. P. Chiusoli, G. Salerno, Adv. Organomel. Chem 17 (1979) 195-243.
[82] R. Baker, Chem. Rev. 73 (1973) 487.
[83] Max Planck, opening speech at the 100th anniversary of the Gesellschaft
Deutscher Chemiker und Arzte, Leipzig 1922.
[84] W. Reppe: Neue Entwicklungen auf dem Gebiet der Chemie des Acetylens
und Kohlenoxyds, Springer, Berlin 1949.
Angew. Chem. Int. Ed. Engl. 29 (1990) 235-244
Документ
Категория
Без категории
Просмотров
2
Размер файла
959 Кб
Теги
nickell, elements, application, catalysing, homogeneous, industries, wide
1/--страниц
Пожаловаться на содержимое документа