Reactions Catalyzed by Pentacyanocobaltate II

This article was downloaded by:
On: 20 January 2011

Access details: Access Details: Free Access
Publisher Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-
41 Mortimer Street, London W1T 3JH, UK
Catalysis Reviews
Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713597232
Reactions Catalyzed by Pentacyanocobaltate(II)

Jack Kwiateka
a
U.S. Industrial Chemicals Co. Division of National Distillers and Chemical Corp, Cincinnati, Ohio
To cite this Article Kwiatek, Jack(1968) 'Reactions Catalyzed by Pentacyanocobaltate(II)', Catalysis Reviews, 1: 1, 37 — 72
To link to this Article: DOI: 10.1080/01614946808064700
URL: http://dx.doi.org/10.1080/01614946808064700
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
Reactions Catalyzed by
Pentacyanocobaltate[ll]
JACK KWIATEK
U.S. Industrial Chemicals Co.
Diuision of National Distillers and Chemical C o w .
Cincinnati. Ohio
Downloaded At: 12:43 20 January 2011
I . INTRODUCTION ANDSCOPE ..................... 38

.
I1 PENTACYANOCOBALTATE(I1) . . . . . . . . . . . . . . . . . . . . 38
.
A Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
.
B Reaction with Writer . . . . . . . . . . . . . . . . . . . . . . . . . 39
.
C Reaction with Hydrogen ...................... 40
ID. REACTIONS WlTH INORGANIC COMPOUNDS . . . . . . . . . . 41
A . Hydrogen Peroxide ......................... 41
B . Oxygen ................................. 42
C. Halogen ................................. 44
D. Metal Ions ............................... 44
E . Ligand Substitution ......................... 45
IV . REACTIONS WITH ORGANIC COMPOUNDS . . . . . . . . . . . 46
A . OrganicHalides ........................... 46
B . Activatedolefins .......................... 51
C . Vinyl Halides ............................. 64
D. Other Compounds .......................... 65
E . Borohydride Reductions ...................... 67
F . Polymerization ............................ 67
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
37
38 JACK KWIATEK
I. INTRODUCTION AND SCOPE
A s early a s a century ago Descamps [ 11 noted that solutions of

cobaltous salts evolved hydrogen in the presence of cyanide ion. Yet
it was not until 1942 [2,3] that Iguchi observed that such solutions
were able to absorb molecular hydrogen and transfer it to substrates,
This homogeneous activation of hydrogen was surprising, although
a few y e a r s previously Calvin [4,5] had demonstrated that cuprous
acetate in quinoline catalyzed the reduction of cupric ion a s well a s
benzoquinone by hydrogen. In the past decade an ever increasing
number of such systems, most of which a r e composed of transition
metal complexes, have been recognized and studied. The subject has
been reviewed periodically by Halpern L6-91 as well as others [10,11].
Homogeneous catalytic reactions a r e of interest not only in their
own right, but also because they may be studied stoichiometrically
on a molecular scale. The sequence of elementary steps comprising
the overall reaction for a homogeneous process i s thus more readily
determined than for a comparable heterogeneous process. A variety

of reaction sites may be present on a solid catalyst surface, and the
individual steps involved a r e extremely difficult to resolve.
The r e s u l t s of such studies a r e not wholly applicable to hetero-
geneous systems but nevertheless provide considerable insight into
various aspects of the catalytic process. These include factors in-
fluencing the r a t e of reaction, due to structure and electron config-
uration of the catalyst o r the substrate, the mechanism of chemisorp-
tion, the nature of the chemisorbed species, the mode of product r e -
lease, selectivity and stereoselectivity , and poisoning and reactiva-
tion. It has become accepted in recent y e a r s that the chemistry in-
volved in similar processes, whether catalyzed by solid metal s u r -
faces, coordination complexes, or enzyme systems, may be inti-
mately related [ 12- 141.
Excellent reviews of transition metal cyanides [ 15,161 and hy-
drides [ 17,181 covering the general chemistry of pentacyanocobaltate
(11) are now available. The present review, which surveys the litera-
ture until December 1966, w i l l focus on the catalytic reactions of
this complex. Although noncatalytic reactions will also be discussed
where they relate, overlap with previous reviews will be avoided
where possible.
11. PENTACYANOCOBALTATE(I1)
A. Preparation
Solutions of [Co(CN),IS- a r e simply prepared by addition of excess
cyanide to solutions of cobaltous salts in an inert atmosphere, The
PENT ACYANOCOBALTATE(I1) 39
light brown cobaltous cyanide which initially precipitates redissolves,

forming an olive-green solution of the pentacoordinated complex an-
ion, [Eq. (l)][ 191. Evidence for the formation of tri- and tetracyano
intermediates has been recently obtained 1201. Various alkali metal
or alkaline earth cyanides, a s well a s various cobaltous salts, may
be employed. Practically all work with this complex has been car-
ried out in aqueous media due to limited solubility in other solvents.
However, methanol [21-231 and a methanol-glycerol mixture [24]
have also been used.
Freshly prepared solutions of this complex have a magnetic mo-

ment of 1.72 BM/cobalt, corresponding to one unpaired electron
[ 19,251. The paramagnetism decreases with time [25] due to the ex-
t r e m e reactivity of the complex which is readily oxidized even by
water (Section 1I.B). The violet crystalline solid which may be pre-
cipitated from these solutions i s diamagnetic and therefore formu-

lated as the dimeric complex [Co,(CN),,IG- [26]. Several dimer s a l t s
have been prepared: Na,LCo,(CN),,]4H20, K,H,~CO,(CN)~~]~H,O, and
K6[Co,(CN),,~6H,0 [27]. They redissolve in water to give solutions
of [Co(CN),I3--.
B. Reaction with Water
The reduction in paramagnetic susceptibility of [Co(CN),P- solu-

tions is accompanied by a loss in capacity for hydrogen absorption
[25] (Section 1I.C). This "aging" process occurs very slowly in dilute
solutions, but quite rapidly a t 0.1 M concentrations o r higher; it may
be accompanied by hydrogen evolution, which i s promoted by cesium
ion [22,28].
These effects have been shown to result from the homolytic cleav-
age ofwaterby [CO(CN),]~-,[Eq. (2)] 122,291. The hydrido complex so
formed is the source of the hydrogen evolved (Section 1I.C). The r e -
action is reversible (Section III.A.2), and kinetic evidence indicates
the possible involvement of the dimeric species ~CO,(CN),,]~- [30].
Some speculation as to the nature of alkali metal ion promotion h a s
been presented [31].
~[CO(CN),]~-
i- 2H,O LCo(CN),HI3- + [CO(CN),OH,]~-+ OH- (2)
The complex [CO(CN),OH,]~-is occasionally formulated a s its con-

jugate base LCo(CN),OHI3-, and there is evidence that these species
a r e pH dependent, [Eq. (3)] [32]. For a more complete discussion,
s e e Chadwick and Sharpe ([ 151, p. 134).
40 JACK KWIATEK
[CO(CN),OH,]~- + OH- =+ [Co(CN),OHI3- + H,O (3)
At elevated temperatures and in the presence of excess cyanide,

[Co(CN),J3- may be formed from [Co(CN) 5]3- in good yields [Eq. (4)].
Since cyanide rapidly replaces water in [CO(CN),OH,]~- at elevated
temperatures, and the hydrido complex decomposes with the evolu-
tion of hydrogen, the reaction is considered to involve the cleavage
of water as an intermediate step.
~ [ C O ( C N ) , ] ~+- 2CN- + 2H,O - 2[Co(CN),I3- + H, + 20H- (4)
The formation of a complex analyzing for [ C O ( C N ) $ C ~ H ~ ]from

~-
the reaction of thiophenol with [Co(CN),I3- appears to be related to
the water reaction [24].
C . Reaction with Hvdrogen

Solutions of [Co(CN),I3- prepared in a hydrogen atmosphere ab-

s o r b up to 1/2 mole of hydrogen per mole of complex. This high re-
activity toward molecular hydrogen, first noted by Iguchi [2], is ba-
sic to the function of this complex as a hydrogenation catalyst. While
some doubt remains concerning the detailed mechanism involved,
the overall process is that of a reversible homolytic cleavage of hy-
drogen to produce the same hydrido complex formed in the cleavage
of water [Eq. (5)] 122,291.
2[Co(CN),I3- + H, =+ Z[CO(CN),H]~- (5)
The product obtained by reduction of [Co(CN),I3- with borohydride

is also considered to be the hydrido complex 1221. No mechanism
has been proposed for t h i s reaction. It is disturbing that although
complete conversion to a metal hydride with borohydride was ob-
served by proton resonance only a small percentage was detected
with hydrogen [33]. Perhaps saturation with hydrogen was not main-
tained during the proton resonance studies. Attempts to isolate the
hydrido complex by alcohol precipitation result in hydrogen evolution
and the separation of [Co,(CN),,]6- [22,331. Other methods of isola-
tion a r e under investigation 1341.
Studies of the hydrogen-deuterium exchange reaction a s well as
para-ortho hydrogen conversion [25,35-371 have been made. The
hydrido complex exchanges rather slowly in heavy water [38]. Kinet-
ic evidence indicates that it is the dimer C O ~ ( C N ) ~which
~ ] ~ - reacts
with hydrogen 1301; species such as [Co(CN),H,13- 122,301, and
[CO,(CN),,H,]~- [ 3 1,391 have been considered as intermediates.
PENTACYANOCOBALTATE (11) 41
The hydrido complex has been formally considered to contain co-

balt in the +3 oxidation state 1221. This oxidation state is thought to
be preferentially stabilized by the cyanide ligands [40]. The H-Co
bond-dissociation energy [30]and the magnetic and spectroscopic
properties of [Co(CN),HIS- have been cited in favor of this oxidation
state (Halpern in 1391, p. 93). On the other hand, polarographic prop-
e r t i e s indicate that the oxidation state of cobalt remains a t the +2
level and the hydrogen i s present in a “stabilized” atomic form [39,
41,421. The hydrido complex has been likened to hydrogen absorbed
on transition metal surfaces [30,39].
Although not originally recognized as a homogeneous activation
of hydrogen, catalytic hydrogenation by [ Co,(CO),] was observed
many y e a r s ago [43]. Hydrogen has recently been reported to take
p a r t in the alkylation of cobaloximes(II), the chemistry of which r e -
sembles that of vitamin B,, and [Co(CN),J3- [44]. Mixed ligand co-
balt(I1) complexes of cyanide and amines, one of which has been iso-
lated 1451, a r e reported to activate hydrogen [46-491. The reduction
of various ligands in cobalt(I1) and cobalt(II1) amine complexes by

hydrogen has also been observed [46,47]. This activation of hydro-
gen by a cobalt(II1) complex appears to be unique.
111. REACTIONS WITH INORGANIC COMPOUNDS
A. Hydrogen Peroxide
1. Reaction with ICo(CN),13-
Hydrogen peroxide r e a c t s rapidly with [CO(CN),]~-

to form
[CO(CN),OH,]~-[Eq. (6)] [ 501 identical to the complexformedwithwater.
It i s not known if this reaction takes place with the intermediate for-
mation of hydroxyl radicals. The formation of alkoxyl radicals in
the reduction of alkyl hydroperoxides (Section III.A.2) indicates such
a possibility. Oxidation of [CO(CN),]~-with persulfate also yields the
aquo complex [ 501.
~ [ C O ( C N ) , ]+~ -H,O, + 2H,O -.~[CO(CN),OH,]~-+ 2 0 H - (6)
2. Catalvtic Reduction
Hydrogen peroxide is catalytically and quantitatively reduced by

hydrogen in the presence of [CO(CN),]~-[ 511. The amount of peroxide
added at any given time must be insufficient to convert all the
[CO(CN),]~-and [CO(CN),H]~- present to [CO(CN),OH,]~-,which is inert
42 JACK KWIATEK
to hydrogen. The reaction i s considered to involve the intermediate

formation of [Co(CN),0H,j2- (or [Co(CN),OHI3-), which r e a c t s further
with [Co(CN),HI3- to form water [Eqs. (7)-(9)]. The autocatalytic re-
duction of [CO(CN),OH,]~-has also been effected [24], thus demon-
strating that the cleavage of water by [ C O ( C N ) ~ ]is
~ -reversible
[Eq. @)I.
+ H,O,
[CO(CN),H]~- - LCO(CN),OH,]~-+ OH- (7)
[Co(CN),HI3- + [CO(CN),OH,]~-+ OH- * Z[CO(CN),]~-+ 2H,O (8)
2[Co(CN),J3- + H, + 2[Co(CN),HI3- (9)
Although [Co(CN),OH,Y- is reducedby [Co(CN),HJ3-, [CO(CN),]~-is

not 1241. This difference in reactivities i s also observed in reductions
of cobalt(II1) complexes at the dropping-mercury electrode [52]. The
mechanism of the interaction of hydrido and aquo complexes has not
been elucidated. The description of the hydrido complex as contain-

ing “stabilized” atomic hydrogen (Section 1I.C) is of particular inter-
est in view of the demonstrated reductions of cobalt(II1) complexes
by atomic hydrogen [53J. These a r e considered to occur by attack of
the hydrogen atom on a ligand. Unfortunately, a comparison between
reducibilities of [CO(CN),OH,]~-and [Co(CN),IS- by atomic hydrogen
has not been made.
It may be assumed that any compound (e.g., persulfate) reacting
with [Co(CN),I3- to form [CO(CN),OH,]~-may be catalytically reduced
by hydrogen in the presence of [Co(CN),13-. The catalytic reductions
of nitrite ion, hydroxylamine, and organic hydroperoxides probably
involve formation of this intermediate [241. The formation of meth-
ane and acetophenone from a-cumyl hydroperoxide and of caproic
acid from cyclohexanone peroxide indicates that alkoxyl radicals are
also intermediates in these reductions.
B. Oxygen
1. Reaction with [Co(CN),13-

-
Solutions of [CO(CN)~]~- r e a c t with excess oxygen to form a brown

binuclear peroxy complex [Eq. (lo)] which may be isolated [ 501 and
purified by recrystallization [32]. The species [Co(CN),0,13- has
been proposed as an intermediate [54J.
2[Co(CN),J3- + 0, - [(NC),COOOCO(CN),]~- (10)

However, gradual introduction of oxygen into solutions of

~ C O ( C N ) , ] r~e-s u l t s in the formation of the aquo complex [Eq. (ll)].
This change i n product is due to the fact that the binuclear peroxide
initially formed readily oxidizes [Co(CN),I3- [Eq. (12)] [50].
t 0, + 6H,O
~LCO(CN),]~- - ~[CO(CN),OH,]~-+ 40H- (11)
+ 2[Co(CN),J3- + 6H,O
[(NC),COOOCO(CN),]~- -
+ 40H-
~[CO(CN),OH,]~- (12)
Acid-catalyzed hydrolysis, resulting in the formation of hydrogen

peroxide, h a s been employed in the quantitative determination of the
peroxy complex [50J. T h e s t e p s involved LEqs. (13)-(15)] include for-
mation of a hydroperoxy complex [CO(CN),OOH]~-,which was a l s o
p r e p a r e d f r o m [Co(CN),HIS- 1551 (Section III.B.2). T h e r m a l decom-
position of the peroxy complex, which is accompanied by oxygen evo-
lution [24], presumably takes place in a s i m i l a r manner.
[(NC),COOOCO(CN),]~-+ H
' = ~(NC),COOOHCO(CN),]~- (13)
[(NC),COOOHCO(CN),]~-+ H,O - [Co(CN),OOHI3-

+ [Co(CN),0H2J2- (14)
[CO(CN),OOH]~-+ H30+ - [CO(CN),OH,]~-+ H,O, (15)
F u r t h e r oxidation of the peroxy complex with molecular oxygen

1501 o r other oxidizing agents [32,50] r e s u l t s in the formation of a
superoxide [(NC),CoOOCo (CN),15-.
Bridged binuclear complexes have a l s o been reported t o f o r m by
reaction of LCO(CN),]~-with NO; [56], CO, 1571, SO,, SnC1, [58],
HC=CH 1591, and CF,=CF, 1601.
2 . Catalytic Reduction
Oxygen is reduced by hydrogen in the presence of [Co(CN)$-

[51,55]. The s a m e hydroperoxy complex obtained by hydrolysis of
the binuclear peroxide (Section LII.B.l) is considered to be the initial
oxygen-containing complex formed 551. F u r t h e r reaction yields the
aquo complex, which is reduced as described in the reduction of hy-
drogen peroxide (Section III.A.2) [Eqs. (16)-(1'7)].
[CO(CN),H]~-+ 0, -.[CO(CN),OOH]~- (16)

44 JACK KWIATEK
[Co(CN),HIS- + [Co(CN),00H13- f 2H' - ~[CO(CN),OH,]~- (17)
In contrast to the reduction of hydrogen peroxide, oxygen i s only

partially reduced, resulting in the absorption of -70% of the hydrogen
required for complete conversion to water [24]. Presumably, some
oxidation product other than [Co(CN),OH,]*- i s also formed in the r e -
action.
Elemental sulfur is reduced quantitatively by hydrogen in the pres-
ence of [Co(CN),IS- 124,511. The reduction presumably involves the in-
termediate formation of sulfur-cobalt bonded complexes. Color
changes with sulfur have been previously observed 1611.
C. Halogen
-
1. Reaction with [Co(CN),13-
Solutions of ~CO(CN),]~-a r e oxidized by iodine, bromine, and chlo-

rine to the corresponding halopentacyanocobaltate (111) complexes [ 621

LEq. (1811.
2[Co(CN),IS- + X, - 2[Co(CN),XIS- (18)
Side reactions occur, with formation of [ Co(CN),IS- and LCO(CN)~OH,]~-,

especially in the reaction with chlorine which gives a poor yield of
[Co(CN),C1IS-. The substitution of halide by water in these complexes
is fairly rapid at room temperature 132,631.
2. Catalvtic Reduction
Iodine and bromine are catalytically reduced by hydrogen in the

presence of alkaline solutions of [Co(CN),IS- [24,51]. The correspond-
ing halopentacyanocobaltate(II1) complexes, assumed to be intermedi-
a t e s in the reaction, a r e also reduced [24], possibly via the aquo com-
plex.
D. Metal Ions
Permanganate , dichromate , and f erricyanide a r e catalytically re-

duced by hydrogen in the presence of [Co(CN),IS- to manganous hydrox-
ide [24], chromic hydroxide [24], and ferrocyanide [ 511, respectively.
The initial reaction with ferricyanide is considered to be formation
of the bridged binuclear complex [(NC),CoNCFe(CN),]'-. This is fol-
lowed by hydrolysis to [Fe(CN),I4- and [CO(CN),OH,]~-,which is fur-
ther reduced by [Co(CN),HI3- as in the reduction of hydrogen peroxide
PENTACYANOCOBALTATE (111 45
(Section III.A.2). The bridged complex may be isolated [ 501 and cata-
lytically reduced by hydrogen in the presence of [Co(CN),13- under
conditions comparable to those employed in the reduction of ferricy-
anide [24]. However, the mechanism of reduction may be more com-
plex than that presented, as hydrolysis of this binuclear ion is re-
ported to require prolonged heating at 80°C. 1501.
Reductions of cobalt(II1) complexes by hydrogen 1471 and stannate
(11) ion [64], a s well a s reductions of cobalt(I1) complexes by hydro-
gen [47] and carbon monoxide [65], are catalyzed by cyanide ion. Al-
though [Co(CN),13- may be involved in certain of these reactions,
there is evidence that this may not be the c a s e in the hydrogen re-
ductions [46]. Further work i s obviously required to establish the
routes by which such catalyzed reactions take place.
E . Ligand Substitution
The substitution of various pentammine cobalt(II1) complexes

[CO(NH,),X]~' by cyanide ion i s catalyzed by [Co(CN),13- [62]. Two

types of substitution behavior, depending on the nature of the ligand
X, have been distinguished. When X = C1-, Br-, I-, N,, N C S , o r OH-,
the reaction results in the formation of [CO(CN),X]~- and involves an
inner-sphere electron transfer via the bridged intermediate
[(NC),CoXCo(NH,),]- [EqS. (19)- (20)] [66].
-
LCO(CN),]~-+ [CO(NH,),X]~+ [CO(CN),X]~-
+ [Co(NH,),]" (19)
[Co(NH,),]'+ + 5CN- - LCO(CN),]~-+ 5NH, (20)
When X = PO:-, COi-, SO:-, NH,, or OAc-, the reaction r e s u l t s in

the formation of [Co(CN),13- and proceeds through an outer-sphere
electron transfer involving [ CO(CN),]~-,assumed to be in equilibrium
with [Co(CN),I3- and cyanide ion [Eqs. (21)-(23)] [66].
[Co(CN),13- + CN- * ~CO(CN),]~- (21)
[Co(CN),I4- + [Co(NH,),X]*' -.[Co(CN),13- + [Co(NH,),X]+ (22)
[Co(NH,),X]' + 5CN- - [Co(CN),I3- + 5NH, + X- (23)
For X = F-, NO;, and CN-, both reaction pathways are observed
[66,67]. The pentammine cobalt(I1) complexes formed a r e substitution-
labile, a s opposed to cobalt(II1) complexes which undergo substitutions
much more slowly.
Evidence that ligand transfers [Eq. (19)] occur via bridged inter-
46 JACK KWIATEK
mediates has been obtained by detection of the transient intermedi-

ates [Co(CN),ONO]’- and LCo(CN),NC]’- 1671. These species were
formed from the nitrogen- and carbon-bonded pentammine cobalt(II1)
complexes, respectively, indicating attack of LCo(CN)J3- on the ex-
posed electron pair of the ligand X. A number of such cyanide sub-
stitutions have been studied [ 68-72]. Evidence for the participation
of [ CO(CN),]~-in charcoal-catalyzed substitutions of LCo(NH,),]’+ by
cyanide ion [73] has been presented [74], but not wholly accepted [75].
In the preceding discussion it is seen that ligand transfer from
one metal to another is often involved in ligand substitution reactions.
Transfers of organic ligands, from chromium to pentacyanocobaltate
(11) [76], and from the latter to mercury [21], have also been observed.
The substitution of the dimethylglyoxime ligands in organocobaloximes
(111) by cyanide ion i s catalyzed by [Co(CN),]’-, but it has not been de-
termined whether transfer of the ligand is involved [24].
IV. REACTIONS WITH ORGANIC COMPOUNDS
A. Organic Halides
1. Reaction with LCO(CN)~]~-
Organic halides react with [ C O ( C N ) ~ to

] ~ -form mixtures of organo-
and halopentacyanocobaltates (111) (Table 1). The reaction takes place
TABLE 1
2[Co(CN),I3- + RX - [Co(CN)5RI3- + [Co(CN)5XI3-
Identificationa
Halide, R X of [Co(CN),RI3- Ref.
CN,I A B F G H I
C4HSI A B
cyclo-C3H5Br A C
n-C3H,I A C F H I
iso-C3H,I A C H
n-C,H91 A B
iso-C4HSI AC
neo-C5H111 A B C
C&,C (CH3)zCH,I A C
Cg H5CHzCIIzI A
ClCHZCHzI AH
HOCHzCH21 A B D H
C,H~OCHZCH~I A D
(C2HSO)ZCHCHzI A D
c6 H5CHzI(Br) A B C H I [21,78,80]
T A B L E 1 (Cont’d)
Identificationa
Halide, RX of 1Co(CN),RI3- Ref.
p-CH,C6H,CH21(Br) A C
p-CH,OC,H,CHzI AC
P-CH~OCHZC,H4CH2I AC
p - o o c C 6 H,CHzBr A
CH2=CHCH21(Br, C1) A B D E L78,80,81]

CH&H =CHC H z B r A B D E J 178,80,811
HOCH,CH=CHCHZCl AD 1241
CH2=CBrCH2C1 Ah [241
HC= CCH,Br (C1) Ah HC [241
CH3COCHzBr A D
CH3COCH (CH3)Br A D
c6 H5COCH2Br A B D I
C6H5COCH(CH3)Br A B D
-OOCCHzI(Br,C1) A
-OOCCH(CH3)Br A
-OOCCH(CHzCOO-)Br A
-OOCCH2CH21(Br) A
CH300CCHzI( B r ,C1) A
C H 3 0 0 C C H (CH3)Br A
CH3OOCCHzCHzI A B
H,NCOCH,I(Br,Cl) A
-03SCHZI A
NC C Hz C1 A B I
NCCH(CH,)Br A B
NCC (CH.J21(Cl) A B
NCCII(C,II,)Br A B
NCCHZCHZI A B H
(Y -C,H,NI A F H I 1771
aThe l e t t e r s d e s i g n a t e v a r i o u s m e t h o d s u s e d for
identification of t h e c o m p l e x e s f o r m e d : A , i n f r a r e d ;
B , NMR; C , a c i d - b a s e r e a r r a n g e m e n t t o RCN; I), a c i d
c l e a v a g e ; E , [Co(CN),HJ3- c l e a v a g e ; F , I, c l e a v a g e ; G ,
HgC1, c l e a v a g e ; H, p y r o l y s i s ; I , a n a l y s i s ; J , d i s p r o p o r -
tionation. See Section IV.B.2 for a g e n e r a l d i s c u s s i o n
of p r o p e r t i e s .
bAbsorption band at 1915 cm-’ indicative of a n a l l e n e
structure.
‘CH3C=CH f o r m e d .
48 JACK KWIATEK
in two steps [Eqs. (24)-(25)], the first involving halogen abstraction

by [Co(CN),]S- with formation of an organic radical [77-791.
[Co(CN),I3- + RX- [Co(CN),X]’- + R . (24)
[Co(CN),]’- + R - - [Co(CN),RI3- (25)
Alkylpentacyanocobaltates have been obtained almost exclusively

from primary alkyl iodides, although low yields of the isopropyl and
cyclopropyl derivatives were isolated from the corresponding iodide
and bromide, respectively. Neopentyl and neophyl iodides yield un-
rearranged organocobalt complexes. Secondary and t e r t i a r y alkyl
iodides form mixtures of alkanes and alkenes, rather than the organo-
cobalt complexes which are assumed to be unstable [78,79]. This ap-
pears to be the result of [Co(CN),]’--catalyzed radical disproportion-
ations in which [Co(CN),]’- s e r v e s as the hydrogen c a r r i e r [Eqs. ( 2 6 ) -
(27)] 1791. The exclusive formation of alkenes when such reactions
are c a r r i e d out in the presence of efficient hydrido-complex scaven-

g e r s [79] and the formation of only 10% monodeuterated alkane
when the reaction is conducted in D,O 1241 indicate the formation and
role of [Co(CN),H]’- as an intermediate. Products of radical-trapping
have also been observed [79].
[Co(CN),]’- + RCH(CH,) . -.[Co(CN),H]’- + RCH= CH, (26)
[Co(CN),H]’- + RCH(CH,) . - [Co(CN),IS- + RCH,CH3 (27)
Vicinal dibromides are dehalogenated by [CO(CN)~]’-[ 77,781. A

stepwise mechanism involving the intermediate formation of the 2-
haloalkyl radical has been proposed [77], and in the case of l-chloro-
2-iodoethane, formation of the 2-chloroethyl-cobalt complex has been
observed [24]. Structures of the complexes formed from polyhalogen
compounds, such as methylene bromide and trichloroacetic acid, have
not been determined.
P r i m a r y benzyl bromides and iodides form benzylpentacyanoco-
baltate(II1). No complex could be isolated from the reaction of a-
bromo-ethylbenzene, which is converted to a mixture of styrene and
ethylbenzene [24].
Ally1 chloride, bromide, and iodide form o-allylpentacyanoco-
baltate(II1). A gradual transition of the o-bonded ally1 group to the
n-ally1 form has been observed in proton magnetic resonance spectra
of solutions of this complex in heavy water [78,81]. Displacement of
a cyanide ligand is assumed. Addition of cyanide ion, but not chloride
o r hydroxide, reconverts the complex to the u-form, indicating an
equilibrium [Eq. (28)].
-
H
/cH = CH2
CH2
Nc\l/cN -CN-
CN-
NC
CN
Methylallyl chlorides form mixtures of isomeric butenes and buta-

diene with [Co(CN),13-. A poor yield of the 2-butenylcobalt complex
i s obtained from crotyl bromide. This complex decomposes in solu-
tion (inert atmosphere) to an approximately equimolar mixture of
butenes and butadiene [ S l ] . The decomposition proceeds via forma-
tion of [CO(CN),H]~-[Eqs. (29)-(30)]. A similar role for the hydrido
complex in radical disproportionations has been previously discussed

[Eqs. (26)-(27)].
[CO(CN),C,H,]~-+ [Co(CN),H]’- + 2[Co(CN),I3- + C,H, (30)
Halides of resonance-stabilized organic radicals which cannot

undergo disproportionation (trityl, benzhydryl, y-phenylallyl, cyclo-
heptatrienyl) form d i m e r s on dehalogenation by [CO(CN),)~-[Eq. (31)]
~781.
2[Co(CN),13- + 2RX - 21Co(CN),XjS-+ RR (31)
Chlorodiphenylacetonitrile f o r m s a dimer (tetraphenylsuccinoni-

tri1e)aswell a s a N-amidocobalt complex. The reaction proposed in-
volves formation of a keteniminocobalt complex from the resonance-
stabilized radical [ 791 [Eq. (32)]. a-Chloro-a-methyl-phenylacetoni-
r i l e r e a c t s analogously to form an amidocobalt complex [24]. Such
complexes are also formed from N-bromoamides [79].
L(C,H,),CC=N - (C,H,),C=C=N - 1 + [Co(CN),13--.

L(C,H,),C=C=NCO(CN),]~-
(C,HJ,CHCONHCo (CN)J3- (32)
Other a-cyanoalkyl halides and a-haloketones and acids form the

corresponding carbon-cobalt bonded complexes. It is significant that
50 JACK KWIATEK
TABLE 2
e dy d r o g e n o l y s i s of O r g a n i c H a l i d e s
[C~(CN)~]~--CatalyzH
Products Ref.
C H31
CZH51
C3HTI
is0-C 3H,I
tert -C4H91
neo-C,HliI
CHzBrz
CICHzCH,I
ICHZCHZCHzI
C6H5CHzBr
C6H5CH(CH3)Br
C6H&H(CH,OH)C1
p-BrCHzC6H4CH2Br
(C6H5)3CC1
CH2=CHCH21(Br, C1) C3H6

CH3CH =C HCHzBr(Cl) C4H8
CH,=CHCH(CH,)C1 C4H8
CH,=C(CH3)CHzBr iso-C4H8
(CH3),C =CHCH2Br is0 -C5Hi,
C6H5CH=CHCHzBr(C1) C,H,CH:=CHCH,
CH2= C (C6H5)CHzBr C6H5CH(CH3)2
ClCHZCH=CHCH2C1 C4H8
CICHzC ECCHZCI C4H8
HCZCCHzCl CH3C zCH, CH,CH=CHz
C6H5COCH2Br C6HSCOCH3
C,H,COCH(CH,)Br C6HsCOC H2C H3
,coo-
NCCH2CI
NCCH(CH,)Br -
NCC(CH,),I (CH3)ZCHCN
NCCH(C6H5)Br -
the only stable t e r t i a r y complex formed [CO(CN),C(CH,),CN]~-con-

tains the highly electronegative nitrile group. Concurrent formation
of methacrylonitrile, isobutyronitrile , and the stable o-cyanoisopro-
pylcobalt complex from a-iodoisobutyronitrile [24] indicates that the
organocobalt complex is not an intermediate in the radical dispropor-
tionation.
a-Iodopyridine is the only aromatic halide reported to react with
[ C O ( C N ) ~ ][77].
~ - Phenyldiazonium s a l t s were employed in the forma-
tion of the phenylcobalt complex [78].
2 . Catalytic Reduction
The same organic halides which react with [Co(CN),IS- are also
reduced by alkaline solutions of [Co(CN),HI3- in a hydrogen atmo-
sphere (Table 2). The reductions are considered to involve organic
radicals, formed from [Co(CN),IS- present in equilibrium with
[CO(CN),H]~-[Eqs. (33)-(34)] 1791.
[Co(CN),]'- + RX + [Co(CN),XJS-+ R . (33)
[Co(CN),HI3- + R . - LCO(CN),]~-+ RH (34)
Organocobalt complexes a r e ruled out as intermediates in most of

these reductions a s they a r e stable to cleavage by [CO(CN),H]~- [78,
791. However) allylic cobalt complexes are readily cleaved. There
is evidence that such complexes may be intermediates in allylic
halide reductions under certain conditions [ 791.
An alternate hydride-halide exchange mechanism not involving
radical formation has been considered. Loss of optical activity upon
reduction of an@-bromoacid is supporting evidence for the radical
intermediate [79].
Yields of reduced substrate are excellent in many cases, but con-
current formation of organocobalt complexes occurs with certain
substrates. This characteristic has been utilized in the preparation
of purified organocobalt complexes [ 781.
The hydrogenolysis of a number of allylic derivatives (alcohols,
e t h e r s , acetates, amines) has also been effected [ 811. No mechanism
has been presented for these reactions.
B. Activated Olefins
1. Catalytic Hydrogenation and Organocobalt Complex Formation
The catalytic hydrogenation of a variety of activated olefins, in-

cluding conjugated dicnes, styrenes, a, /3-unsaturated aldehydes,
TABLE 3
[C~(CN),]~--CatalyzedHydrogenation of Activated Olefins
Olefin Product Ref.
CHZ =CHCH=CH, C4H8

CHZ=C(CH,)CH=CH, C5Hi0
CHz=CHCH=CHCH, C5H10
CHz=C(CH,)C(CH,)=CH, C6Hi2
CH,CH=CHCH=CHCH, C6HIZ
CHz=CHCH=CHCH,CHzCH=CHz C8H14
Cyclopentadiene -1,3 Cyclopentene
Cyclohexadiene-l,3 Cyclohexene
3-Methylenecyclohexene 1- an5 3-Methylcyclohexenes
1,2-Dimethylenecyclohexane Dimethyl cycbheseaeY, im ethyl -
methylenecyclohexane
Cyclooctatetraene A bicyclooctadiene
Norbornadiene Nortricyclene, norbornene
CdH,CH=CH, C,H,C,H,
C~H~CH-CHCHZOH C,H,CH,CH,CH,OH
C,H,C(CH,)=CH, CdHjCH(C H3)z
C,H,C(OCH,)=CH, C6H5 CH(OCH,)CH,
(c6 H5)ZC=CHZ (C6H5)ZCHCH3
p-NC C ~ H QHC=CHZ p-NCCgH,CzH,
a -Vinylpyridine a-Ethylpyridine
CHz=C( CH3)CHO (CH3)zCHCHO

CH,CH=CHCHO CH,CHzCH,CHO
CH3CH=C(CH,)CHO CH,CHzCH(C H3) CHO
C,H,CH=C(CH3)CHO CZHSC HZCH(CH,)CHO
C,H,CH=CHCHO C,HSCHZCHzCH=C( CHzC6H5)CHO
TABLE 3 (Cont’d)
Olefin Product Ref.
-~
CH, =C (CH,)COCH, (CH,),CHCOCH,

CHZ=C(CH,)COC~H, (CH&CHCOC,H,
CHz=CHCOC,Hs CzH5COC,H,
CGH,CH=CHCOCH, C,H,CH,CH,COCH3
CaH&OCH=CHCOC,H5 CsH,COCH,CH,COCgH,
p-Benzoquinone p-Hgdroquinone
Anthsaquinone Hydranthr aquinone
Indigo Indigo white
CHZ=CHCOO- CH3CHZCOO-
CHz= C( CH3)COO- (CH,),CHCOO-
CHZ=C(CH,)COOCH, (CH,),CHCOOCH,, (CH3)zCHCOO-
CHz=C(CH,COO-)CO@ CHSCH(CH,COO-)COD-
CHz=C(C,H,)C03- CH,CH(C,H,)COO-
CH3CH=CHCH=CHCOO- Hexenoates
C@H,CH=CHCOO- C~H~CHZCH~COO-
cis-OOCGH=CHCOO- -0OCCHzCHzCOO-
I
OCC(C&,)=C(C,H,)~O -OOCCH(C~H~)CH(C~HS)COO-
-0-
-0occ-ccoo- -0OCCHZ CHZ COO-,

trans- OOC CH= CHCOO-
CHz=C(C,H,)CN CH,CH(C,H,)CN
C H,= C(OC2H5)CN CH3CH(OCzH,)CN
CH,=C= CHCN C Hs CH =CHCN
TL4BLE 4
[ C ~ ( C N ) ~ H l ~ - - O l e fAdducts
in
[Co( CN) ,R1 3-

Olefin R= Identificationa Ref.
CHFZCFZ I X-ray
CH3CH=CHCH, A B D E J
p-NCCsHdCH(CH3) Ab
a - C ,H*NC H(CH3) Ab,'
CijH,C=CH, A B Dd
CeHrjC =CHCOO- A B Dd
CH,COCH=CH, CH3COCH( CH3) ABD

C,H,COCH=CH, CGHsCOCH(CH3) ABD
2 -Norbornenone 6-(2-oxonorbornyl) A C D
CH, =CHCOO- CH,CH(COO-) AB

CHz= CHCOOCH, CH3CH (COOCH,) AB
CH, =C( CH,)COOCH, (CH3)2C(C00CH3) Ab7' J
cis-ROOCCH=CHCOOR ROOCCHZCH(CO0R) Ab
trans-ROOCCH=CHCOOR ROOCCH,CH(C083R) Ah
CH,=CHCN CH3CH(CN) ABI

CHZ =C(CH,)CN (CH3)2C(CN) AB
CH,CH=CHCN CH,CHzCH(CN) Ab
C,H,CH=CHCN CsH,CHzCH( CN) Ab
TABLE 4 (Cont’d)
[ C o(CN) 5R]3-
Olefin R= Identificationa Ref.
1 -C:yanocyclohexene 1-Cyanocyclohexyl AB P I
CH,=C(CGH,)CN CH,CH( CsH5)CONH A Be I791
CH,=C(OCzH,)CN CH3CH(OC2Hs)CONH A Be W91
p-HON=C,H,=C( CsHj) CN P-H~NC,H,CH(C,H~)CONH A [791
%ee Table 1 f o r explanation of letters.
bForinlila assumed.
‘Deuteriogenation supports formula.
dAcetophenone formed.
eHydrolyzed t o amide.
56 JACK KWIATEK
ketones, acids, and acid derivatives by [Co(CN),I3- has been demon-

strated (Table 3). Nonconjugated olefins are neither reduced nor iso-
merized.
Initial work with this catalyst system indicated that catalytic re-
duction was controlled by small differences in olefin structure. Po-
tassium methacrylate was quantitatively reduced, but under compa-
rable conditions no reduction of potassium acrylate was observed [51,
821. Since then, it has been recognized that many substrates, e.g.,
acrylate, form organocobalt complexes with [CO(CN),H]~-(Table 4)
which a r e stable to further reaction with [CO(CN),H]~-.The catalyst
may be considered to be “poisoned.” This section discusses the in-
fluence of structure on catalytic hydrogenation versus organocobalt
complex formation.
u . Conjugated Dienes. Conjugated dienes, including those with

fixed s - t r a n s a s well as s-cis conformations, a r e reduced selectively
to monoolefins. Exceptions a r e 2,5-dimethyl-2,4-hexadiene and cyclo-
octadiene- 1,3 [81].

Butadiene formed an unstable adduct with LCO(CN),H]~-,identical
to that prepared by the reaction of [ C O ( C N ) ~ ]with
~ - crotyl bromide
(Section IV.A.l). The complex was isolated and characterized by its
P M R spectrum as ~-2-butenylpentacyanocobaltate(III)[ 78,811. At-
tempts to isolate [ Co(CN),HIS- adducts of other reducible conjugated
dienes failed [24].
The reduction of butadiene proceeds via formation of this complex
[Eqs. (35)-(37)] [51] [Cf. EqS. (29)-(3O)I.
~ ~*
~ [ C O ( C N )+~H, - ~[CO(CN),H]~- (35)
[Co(CN),HIS- + C,H, =[Co(CN),C,H,I3- (36)
[Co(CN),HIS- + LCU(CN),C,H,]~- - 2[Co(CN),J3- + C,H, (37)
Results of deuteriogenation experiments [ 51,811 and disproportion-

ation of the butenylcobalt complex (Section 1V.A. 1) indicate that ad-
duct formation is reversible.
An interesting control of stereoselectivity by cyanide ion is ob-
served in these reductions [51,83,84]. Trans-2-butene is the major
product obtained in hydrogenations of butadiene c a r r i e d out in the
presence of little f r e e cyanide ion (CN-/Co < 5.5). 1-Butene is the
major product formed when considerable f r e e cyanide is present
(CN-/Co > 5.5). Similarly, a greater proportion of terminal olefin is
obtained in the presence of excess cyanide in reductions of isoprene
[79,86], 1,3-pentadiene “791, and 2,3-dimethylbutadiene [24]. It has
been proposed [ 51,79,81] that this effect is due to a cyanide-dependent

equilibrium between u- and n-allylic intermediates as observed in
the case of the unsubstituted ally1 complex (Section 1V.A. 1). A ration-
ale for the butene product isomers formed from butadiene has been
presented [79,81]. Unfortunately, a u-?T equilibrium has not been ob-
served with the 2-butenylcobalt complex, solutions of which a r e un-
stable.
L e s s stereoselectivity is observed in the product of 2,3-dimethyl-
butadiene reduction a t low values of CN-/Co [24]. This may be due to
a sterically unfavorable antimethyl group in the n-allylic form of its
[Co (CN),HI3- adduct.
The observation that the same mixture of norbornene and nortri-
cylene (7:93) is obtained in the reduction of norbornadiene at either
low or high CN-/Co suggests that a n-homoallylic complex cannot be
an intermediate. The high selectivity for nortricylene formation,
compared to results obtained in reductions of the norbornenyl-nor-
tricyclyl radical system [96,97], indicates that the [Co(CN),13--cata-
lyzed reduction takes place via an organocobalt complex, rather than

a radical.
A hydroxyl ion effect has also been noted in reductions of dienes
[24,84,85]. In the presence of OH- somewhat smaller quantities of
cyanide are required to effect predominant formation of 1-butene
from butadiene. This may be attributed to the suppression of cyanide
hydrolysis by OH-. Isoprene product distributions resulting from re-
ductions c a r r i e d out in a hydrogen atmosphere a r e considerably dif-
ferent from those c a r r i e d out in a nitrogen atmosphere at the same
cyanide to cobalt ratio [24]. The factors involved in this change have
not been clarified.
b . Styrenes. A number of styrenes have been reduced to the cor-

responding ethylbenzenes. Unstable [Co(CN),HI3- adducts were iso-
lated from p-cyanostyrene and a-vinylpyridine, each of which contain
electronegative aromatic groups. Some interesting structural speci-
ficities have been noted. Although a number of p-substituted styrenes
( propenylbenzene, indene, stilbene) could not be reduced, cinnamyl
alcohol was reduced to 3-phenylpropanol and cinnamyl acetate under-
went hydrogenolysis yielding propenylbenzene [24,82]. The reduction
of 1,l-diphenylethylene, formerly reported as not occurring [ 511, was
very slow.
c. Acetylenes. Acetylene does not add LCO(CN),H]~-to form the

vinylcobalt complex [24] (Section IV.C), although it does form a
bridged binuclear complex [ (NC),COCH=CHCO(CN),]~-with [Co(CN),13-
[59]. This complex rearranges in solution to form a nitrile-contain-
58 JACK KWIATEK
ing organocobalt complex whose structure has not been determined

1241. A s i m i l a r bridged complex [ (NC),COCF,CF,CO(CN),]~- has been
obtained with perfluoroethylene [ 60,951. Phenylacetylene and phenyl-
propiolic acid salt w e r e not reduced but did f o r m adducts with
[CO(CN),H]~-[24] (Table 4).
d . Aldehydes. Only a few a,@-unsaturated aldehydes have been

reduced, and these in poor yields. The reactions a r e complicated by
condensations, and presumably [CO(CN),H]~-adduct formation which
was not investigated [51]. Propionaldehyde formed 2-methylpen-
tenal via reduction of the aldol condensation product, 2-methyl-2-pen-
tenal. Reductive aldol condensations were a l s o observed with butyral-
dehyde at elevated temperatures, and with mixed aldehydes [ 51,821.
e . Kelones. Interesting structural effects were observed in r e -

actions with ketones. Isopropenyl ketones were readily reduced, and
no lCo(CN),HJS-adduct could be isolated. The attempted formation of
organocobalt complexes [Co(CN) C(CH,),COR]3- by reaction of 1-bro-

moisopropyl ketones with [Co(CN),J3- also failed 1241. In contrast,
vinyl ketones reduced poorly and yielded stable adducts [Co(CN),-
CH(CH3)C0R]3-which could also be obtained from the reaction of 1-
bromoethyl ketones with [CO(CN),]~-[ 781 (compare Tables 1 and 4).
Ketones such as dibenzoylethylene [87] and benzoquinone [ 511 were
catalytically reduced only in the presence of OH-. Presumably, inter-
mediates subject to hydrolysis to [CO(CN),OH]~-a r e formed 1511.
Norbornenone formed a [Co(CN),HI3- adduct, the protonation r e -
actions of which were especially interesting. Acidification of the corn-
plex released 1-hydroxynortricyclene . Subsequent treatment with
base yielded a ketonitrile (Section IV.B.2.a and b). The adduct p r e -
sumably consists of both endo and exo forms. The former would be
expected to eliminate 1-hydroxynortricyclene [Eq. (38)]by analogy
with the stereospecific acid-induced homoenolization of norbornanone
[98]. The latter, in which the cobalt atom is not favorably situated
for elimination, would be expected to undergo acid-base rearrange-
ment to nitrile [Eq. (39)]. The position of the nitrile group in the
ketonitrile, however, has not been determined. The formation of 1-
hydroxynortricyclene supports the mechanism of ketone elimination
by proton attack on oxygen [80]. This reaction sequence may be of

value in the preparation of cyclopropanols f r o m other homoconjugated
unsaturated ketones such as 3-methylenecyclobutanone.
f. Acids a d Acid Derivalives, Structural effects on the reducibil-

ity of @#-unsaturated acids and e s t e r s have been observed [ 51,79,93,
991. Although a-substituted acrylic acids a r e readily reduced, acrylic
acid itself is not, except at elevated temperatures, due to formation
of a stable [Co(CN),HI3- adduct. Cinnamic acid and sorbic acid are
readily reduced 0-substituted acrylic acids which a r e activated by
other unsaturated centers in the molecule. Methyl methacrylate is

readily reduced, and an unstable organocobalt adduct with LCO(CN),H]~-
may be isolated. It disproportionates to methyl methacrylate and
methyl isobutyrate.
The reduction of cinnamate ion has been formulated a s involving
a radical intermediate [Eqs. (40)-(43); S = cinnamate] [39,91,92].
The mechanism differs from that offered for the reduction of buta-
diene [51] in which an organocobalt complex is considered to be an in-
termediate. Although the semireduced cinnamate radical anion may
form an organocobalt complex in a fast equilibrium with [ CO(CN),]~-
[Eq. (43)], the kinetic results indicate that such a complex plays no
essential role in the hydrogenation process.
~ [ C O ( C N ) , ]+~ H,
- * Z[CO(CN),H]~- (40)
[Co(CN),HJ3--1 S 4 [Co(CN),J3-+ HS . (41)
[CO(CN),H]~-+ HS --LCo(CN),J3-+ H,S (42 1
[CO(CN)J+ + HS . * [Co(CN),SH13- (43)
A recent deuteriogenation study [24] has shown that formation of the

radical [Eq. (4111 must be reversible.
6. Nitvilcs. a$- Unsaturated nitriles substituted in the a-position

by alkoxy or phenyl groups are reduced when gradually added to
[CO(CN),H]~- “791. When these nitriles a r e present in excess during
60 JACK KWIATEK
formation of [Co(CN),HI3-, N-amidocobalt complexes, analogous to

those obtained from tertiary chlorophenylacetonitriles (Section 1V.A.
l ) ,a r e obtained instead. The reductions, as well as N-amido-com-
plex formation, are considered to take place via radical intermediates
[Eqs. (44)-(46); R = C,H, o r OC,H,]. In contrast, acrylonitrile and
methacrylonitrile are not reduced, nor do they form N-amidocobalt
complexes; instead they yield stable carbon-cobalt bonded complexes.
As radical intermediates would be expected to undergo at least par-
tial reduction, they a r e considered not to be formed with these com-
pounds. Attempts to reduce nitriles at elevated temperatures result
i n the formation of products of hydrolysis of the nitrile group [99].
Apparently [ Co(CN),I3- catalyzes the hydrolysis.
CH,=CRCN + [CO(CN),H~~- CH,CRCN + [ C O ( C N ) ~ ] ~ - (44)

-).
CH.$RCN + [Co(CN),HJS- - CH,CHRCN + [CO(CN)~],- (45)

HO
CH,CRCN + [Co(CN),I3- -+ [CH,CR=C=NCo(CN),]'- 2
[CH,CHRCONHCO(CN),]~- (46)
2. Properties of Organocobalt Complexes
Only a few organopentacyanocobaltate complexes have been ob-

tained in pure form [21,77]. The presence of such derivatives in cy-
anocobaltate(II1) mixtures is readily detected by observation of a
characteristic cyanide stretching band in the vicinity of 2100 cm-',
approximately 25 cm-l l e s s than the band observed for inorganic com-
plexes. Structures of the organic moieties bonded to cobalt a r e gen-
erally determined by proton magnetic resonance spectra J f solutions
in heavy water [21,78,79].
The order of stabilities of such complexes is XCH, > 'XHCH,
> XC(CH3)2,where X is either an electropositive or electronegative
group. The dependence of stability on the nature of the X group i s
in the o r d e r CN > COOR > COR > CH=CH, = R > C,H, [79]. Un-
stable complexes, which nevertheless could be isolated in certain
c a s e s , decompose with disproportionation of the organic group. Stable
complexes yield products apparently arising from homolytic cleavage
when decomposed by heat o r reaction with iodine [21,77,78].
The complexes may be placed in three categories, depending on
their mode of reaction with acids: (a) those which rearrange and r e -
lease nitriles on further treatment with base, (b) those which release
the organic moiety or derivative thereof directly, and (c) those which
a r e stable to acid.
a . Alkyl, benzyl, vinyl, and phenyl pentacyanocobaltates form
the corresponding nitriles on treatment with acid, followed by base

124,801. Yields of nitrile a r e far from quantitative, and amide a s well
a s acid a r e often formed concurrently. The reaction has been em-
ployed in detecting the presence of s m a l l quantities of organocobalt
complexes (e.g., the isopropyl derivative) [79] and in determining the
stereoconfiguration of vinyl derivatives (Section 1V.C). Complexes
containing groups highly susceptible to rearrangement (e.g., neo-
phyl) yielded the unrearranged nitrile indicating that the migrating
group i s tightly bound throughout the reaction. A reaction scheme in-
volving a 1,2-shift of the organic group to an electron-deficient c a r -
bon has been proposed [80] [Eq. (47)].
R R
I H' I
[(NC),COCN]~- - 7 [(NC),CO-&NH]~--
OH-
[ (NC),&O-CR=NHI~-- RCN (47)
Reaction by such a scheme should result in the concomitant forma-

tion of a reduced cyanocobalt complex, and indeed it i s found that the
residual complexes catalyze hydrogenations. The rearrangement is
suggestive of the carbonyl insertion reaction observedwith alkyl t r a n s -
ition metal carbonyls [ 100,101]. It is interesting that vinylacetyl
chloride r e a c t s with LCO(CN),H]~-
with elimination of CO 1813.
b. Ally1 and 2-0xo-, 2-hydroxy-, and 2-alkoxy-alkylpentacyanoco-
baltates release the organic moiety on treatment with acid [24,80].
A reaction scheme involving proton addition to the organic group to
form a carbonium ion which undergoes 1,2-elimination has been sug-
gested [Eqs. (48)-(50)] 1801.
[ CO(CN),CH,C H=CHRJ3--
H'
[CO( CN),CH26HCH,Rj2- -
CH;=CHCH,R (48)
[Co(CN),CH2CORJ3--
H'
LCO(CN),CH,~OHR]~- -
CH3COR (49)
This elimination reaction has been employed in determining

amounts of ailylic complexes present in mixtures. The exclusive
formation of 1-butene from the 2-butenylcobalt complex indicates
62 JACK KWIATEK
proton attack at the y-position [80,81], just a s formation of 1-hydroxy

nortricyclene from the norbornenone hydrido-complex adduct indi-
cates attack of proton on the oxygen atom (Section 1V.B.l.e). Although
butadiene is evolved from the 4-hydroxy-2-butenylcobaltcomplex,
p-xylylene is not formed from the p-methoxymethyl benzylcobalt
complex L241. The a-styryl complex formed from phenyl acetylene
and the hydrido complex yields acetophenone on acidification [24].
The reaction scheme proposed for these protonations indicates
that the oxidation state of cobalt remains +3 after elimination of the
organic moiety. Nevertheless, when the resultant solutions a r e made
basic, they catalyze hydrogenations. Possibly a portion of the com-
plex is converted to the cobaltous state a s in the case of rearrange-
ment to nitriles.
c . 1-Cyanoalkyl- and a-pyridylcobalt complexes a r e stable to acid
treatment [77,79].
Organocyanocobaltates , other than those undergoing spontaneous
decomposition, a r e stable to [ Co(CN),HI3- with minor exceptions. The
ally1 complex readily releases propylene. Attack of hydrogen on the

y-carbon of allylic groups has been invoked in rationalizing the stereo-
specificities observed in diene reductions [ 79,811. 1-Benzoylethylcobalt
complex releases propiophenone in low yield on treatment with the
hydr ido complex.
3 . Mechanism of Hydrogenation
The structures of stable organocobalt complexes show that hydrido

complex adds t o a,p-unsaturated acids, ketones, and nitriles a s though
the cobalt-hydrogen bond were polarized Cot-H-. Whether such addi-
tions a r e cis o r trans has not been determined. Activated olefins
which are catalytically hydrogenated generally f o r m unstable organo-
cobalt complexes o r none a t all. Evidence for direction of addition
must be obtained indirectly in these cases.
Deuteriogenations of butadiene [51] and sorbate [89] have been re-
ported. Additions of [CO(CN),D]~- to these compounds as well as to
styrene, a-methylstyrene, a-vinylpyridine, methyl methacrylate, and
cinnamate a r e reversible, a s indicated by the quantity of deuterium
incorporated in the products and recovered olefins [24]. Product in-
corporation of > 1 deuterium /3 and < 1 deuterium a to the activating
group revealsthat initialadditions of deuterium must be to the p -
carbon. This direction of addition is also found with olefins forming
stable organocobalt complexes.
Some of these results have been selected for comparison with
TABLE 5
Deuterium Distributions in Saturated Products f r o m
Deuteriogenations of CH,=C(a)X
Catalyst X a aa B” PI@ Ref.
[Co(CN) 5]3-b C6H5 H 0.85 2.42 2.85 ~ 4 1

Pd-C C6H5 €I 0.58 1.41 2.43 [lo21
[Co(CN) 5] 3- C6H5 CH3 0.79 4.72 6.00 ~ 4 1
Pd-C C6H5 CH3 0.63 1.47 2.34 [1021
[ C O ( C N ) ~ ] ~ - ~ CH,O,C CH, 0.08 1.22 14.7 ~ 4 1
Pd-C CH30,C CH3 0.15 1.82 12.1 [lo21
[ C O ( C N ) , ] ~ - ~ CPC~H~N H 0.79 1.93 2.45 [241
_. ~-
“Indicates a v e r a g e number of deuterium a t o m s found in a and [3 positions
relative to X in t h e product.
bD, and D,O employed.
deuteriogenations over a heterogeneous catalyst [ 1021 in Table 5. It

may be seen that the highest ratios of p to LY deuterium incorporation
in product a r e obtained with methyl methacrylate for either type of
catalyst. Especially striking i s the s m a l l amount of deuterium enter-
ing the &-position of methyl methacrylate compared to the styrenes.
This difference has been observed in a number of reductions over
palladium and is one of the features of the so-called “phenyl effect”
[ 1021.
It has been suggested that styrenes hydrogenate over palladium
via a different route than do a,p-unsaturated e s t e r s due to ‘IT-com-
plexing of the phenyl group to the metal surface [ 1021. Although sim-
ilarities in mechanisms involving ‘IT-allylic complexes for homoge-
neous and heterogeneous hydrogenations of dienes have been indi-
cated [13,79], there is no evidence for an arene cyanocobaltate.
It is tempting to attribute the differences observed in the deuterio-
genation studies to differing stabilities of the semihydrogenated or-
ganometal species. Those species containing highly electronegative
groups would be expected to form stronger carbon-metal bonds [ 791.
The observation that deuterium distributions obtained from palladium-
catalyzed deuteriogenations of C,F,CH=CH, and C,F,C(CH,)=CH,
resembled those of CH,O,CCH=CH, and CH,O,CC(CH,)=CH,, but not
C,H,CH=CH, and C,H,C(CH,)=CH, [ 1021, supports this view. Unfor-
tunately, these substrates have not been deuteriogenated in the pres-
ence of [ CO(CN),]~-.Similar deuterium distribution patterns might be
expected for a-vinylpyridine and methyl methacrylate, a s well, since
64 JACK KWIATEK
unstable [CO(CN),H]~- adducts were isolated from each of these olefins.

However, [ C~(CN),]~--catalyzed deuteriogenations revealed a distri-
bution pattern for a-vinylpyridine which resembled that obtained with
styrenes (Table 5). Further investigations along these lines should
help to point out common features of homogenous and heterogeneous
hydrogenations.
A general scheme which incorporates the observations made in
[ C~(CN),]~--catalyzed hydrogenations of butadiene , cinnamate , and
@+-unsaturated nitriles may be outlined. The hydrogen of the hydrido
complex approaches the positive p-carbon of an activated double bond.
The four-center transition state thus formed may collapse to either
a radical o r an organocobalt complex, depending on the nature of the
olefin. Final reduction to the product occurs by transfer of hydrogen
from [Co(CN),HI3- to the radical o r , in the case of allylic complexes,
to the organocobalt complex. This scheme accommodates the evidence
that formation of stable organocobalt complexes need not involve the
formation of radical intermediates (Section 1V.B.1.g) and that radical
disproportionation need not require the formation of an organocobalt

intermediate (Section 1V.A. 1 ) .
co
C. Vinyl Halides
Vinyl chlorides and bromides react with alkaline [Co(CN),HI3-

solutions to f o r m vinylcobalt complexes [24,78]. A two-step addition-
elimination sequence has been assumed [Eqs. (51)-(52)].
[CO(CN),HI3- + CH,=CHX -+ [Co(CN),CHXCH313- (51)
[CO(CN),CHXCH,]~- + [Co(CN),CH=CH,13- + HX (52)
If the above sequence is operative, reactions of vinyl halides with

deuterium in heavy water would be expected to result in the formation
of a largely monodeuterated vinylcobalt complex. The reaction of p-
bromostyrene in this manner yielded ethylbenzene and an organoco-
balt complex which on treatment with acid and base gave cinnamoni-
trile containing 0.17 deuteriums/mole 1241. As the deuterium incor-

porated may be due to some reversible addition in the reduction r e -
action, the above mechanism is not supported.
Another consequence of the above scheme is theexpectation that
cis- and trans-vinyl halides would f o r m identical vinylcobalt com-
plexes since their [ Co(CN),HI3- adducts would be identical. The cis/
t r a n s isomer ratios of the crotononitriles obtained by rearrangement
of the complexes formed from 1-bromopropene mixtures indicate that
the reaction is stereospecific [24]. Obviously, further work is re-
quired to elucidate the mechanism of vinylic complex formation.
A reaction with trans- 1,2-dichloroethylene, designed to form the
bridged complex [(NC),COCH=CHCO(CN),]~-[ 591, yielded a complex
exhibiting absorption bands at 3270 and 1980 cm-’ [24]. A possible
structure is [ CO(CN),C=CH]~-. A stable adduct [Co(CN),CF,CF,HI3-
of the hydrido complex with CF,=CF, has been isolated [95] and its
structure determined by X-ray analysis [94].
D. Other Compounds
A number of other compounds have been reduced by hydrogen in

the presence of [Co(CN),]”. These a r e listed in Table 6. The major-
ity of the reactions involve reductions of multiply bonded structures.
1. Carbonyls
As in reductions of the olefinic bond, the carbonyl group must be

activated for reduction to occur. The examples shown indicate that
conjugation with an aromatic or other carbonyl group is required.
Benzophenone and acetophenone a r e not reduced.
2. Oximes
Reductive amination of ketones is effected a t 40-70” and 700 psi

hydrogen [90,103]. These reactions proceed via the oximes which
have been shown to be reduced to amines [90,99,104]. Hydrazones
are reduced similarly.
3. Nitro Compounds
Aromatic nitro compounds have been reduced to azoxy, azo, and

hydrazo compounds, as well as hydroxylamines and amines, depend-
ing on the conditions of reaction and the structures of the nitro com-
pounds. Azoxybenzene [82] and azobenzene [51] have also been r e -
duced. Aliphatic nitro compounds and N-nitroso compounds a r e re-
duced to amines.
66 JACK KWIATEK
TABLE 6
[ C ~ ( C N ) ~ ] ~ - - C a t a l y z Reductions
ed by Hydrogen
Compound Products Ref.
CgH5CHO CBH~CH~OH
CH3COCOCH3 CH3CH(OH)COCH,
2,5 -dimethylhydroquinone
CeH,CH(OH)COC6H,
Dioxindole
Methylcyclohexanones + NH3 1LIethylcyclohexylamines

CH3CO'ZOO- + NH3 CH,CH(NH,) COO-
C,H,CH,COCOO- + RNHZ C,H5CH2CH(NHR)COO-
C,H,CH-CHCOR + NH3 C,H,CH2CH,CH(NHz)R
C,H&OCOC,H, + NH3 C,H,CH(OH)CH(NH,)C,H5,
C,H,CH(NH~)CH(NH~)COH~
(CH,),C=NOH (CH3)2CHNH2
C,H,C(CH,)=NOH CaH,CH(CH,)NHz
C,H,CHzC( COO-)=NOH C,H,CH,CH( COO-)NH,
C,H,C(CN)=NOH C,H,CH(COO-)NH2
C,H,CHzC(CN)=NNHC,H, C,H,CHzCH( COO-)NHNHCsH,
Dimethylfurazone T e t r amethylpiperazine
C6H5N02
o -HO C, HdNOz
P-HOC~H~NO,
CZHSNOZ
(CH3)3CN0Z
(CH3)zNNO
(C6H5)2NN0
Cyclohexene oxide Cyclohexanol

Styrene oxide
4. Oxiranes
Epoxide rings a r e cleaved to alcohols. It is interesting that sty-

rene oxide yielded p-phenylethanol; the a-isomer was not detected.
E. Borohydride Reductions
A number of compounds have been reduced by aqueous borohydride

in the presence of [CO(CN)&~-. Included a r e a,p-unsaturated acids
[ 1051, nitrobenzene, azoxybenzene, azobenzene [ 1061, and nitrosodi-
methylamine L1071. The reduction of dyes by borohydride [lo81 and
dithionates [ 109,1101 is similarly catalyzed.
A s borohydrides a r e known to r e a c t with [Co(CN),13- to form
[Co( CN),H]S- [22,33], the hydrido complex has been considered to be
the reducing agent in the reaction with a#-unsaturated acids [ 1051.
However, some acids (e.g., crotonic, fumaric, phenylpropiolic) r e -
duced by borohydride in the presence of [ C O ( C N ) ~ ]a~r -e not reduced
by [Co(CN),HI3- alone. These reductions, as well as the interaction

of borohydride with [Co(CN)$- itself, warrant further study. Reduc-
tions by borohydride in the presence of other transition metals have
also been reported L107,111].
F . Polymerization
Butadiene, isoprene, styrene, methyl methacrylate, and acryloni-

trile a r e catalytically polymerized by aqueous [Co(CN),13- in the p r e s -
ence of a surfactant [112,113]. Other transition metal cyanides have
been employed as polymerization accelerators [ 1141.
Head-to-tail dimerizations of acrylic and methacrylic acids a t
125" have also been observed 1511.
Acknowledgement
It i s a pleasure to acknowledge the support of the A i r Force Office

of Scientific Research [Contract No. A F 49(638)-12141, the encourage-
ment of D r . I. L. Mador, and the assistance of J. K. Seyler.
REFERENCES
113 A. Descamps, Comfit. Rend., 67, 330 (1868).

[2] M. Iguchi, J . C h e m . SOC. Ja&zn, 63, 634 (1942); C A , 41, 2975d
(1947).
68 JACK KWIATEK
131 M. Iguchi, J . Chem. Soc. Japan, 63, 1752 (1942); C A , 41, 3353c
(1947).
[4] M. Calvin, Trans. Faraday SOC., 34, 1181 (1938).
[5] M. Calvin, J. A m . Chem. SOC., 61, 2230 (1939).
[6] J. H a l p e r n , Quart. Rev. (London), 10, 463 (1956).
[7] J. H a l p e r n , Advan. Catalysis, 11, 301 (1959).
[8] J. H a l p e r n , Ann. Reu. Phys. Chem., 16, 103 (1965).
[9] J. H a l p e r n , in Proceedings of the 3rd International Congress
on Catalysis, Amsterdam, 1964, Vol. 1 (W. M. H. S a c h t l e r ,
G. C . A. Schuit, and P. Zwietering, eds.), North-Holland, Am-
sterdam, 1965, p . 146.
[ l o ] S. W e l l e r a n d G. A. Mills, Advan. Catalysis, 8, 163 (1956).
[ l l ] M. E. Winfield, Rev. Pure Appl. Chem., 5, 217 (1955).
[12] G . C . Bond a n d P. B. W e l l s , Advan. Catalysis, 15, 210 (1964).
[13] B. J. J o i c e , J. J . m o n e y , P. B. W e l l s , a n d G . R. Wilson,
Discussions Faraday S O C . , 41, 223 (1966).
[14] M. Orchin, Aduan. Catalysis, 16, 1 (1966).
[15] B. M. Chadwick and A. G. S h a r p e , Advan. Inorg. Chem. Radio-

chem., 8, 83 (1966).
[16] M. H. Ford-Smith, The Chemistry of Complex Cyanides, H. M.
Stationery Office, London, 1964.
[17] A. P. G i n s b e r g , in Transition Metal Chemistry, Vol. 1 (R. L.
C a r l i n , e d . ) , D e k k e r , New York, 1965, p. 111,
1181 M. L. H. G r e e n and D. J. J o n e s , A d v a n . Inorg. Chewz. Radio-
chem., 7, 115 (1965).
[19] A. W. A d a m s o n , J . Am. Chem. SOC., 7 3 , 5710 (1951).
[20] T. Suzuki and T, Kwan, Nippon Kaggakzc Zasshi, 87, 395 (1966);
C A , 6 5 , 13190e (1966).
[ 2 l ] J . H a l p e r n a n d J. P. M a h e r , J . A m . Chem. Soc., 86, 2311 (1964).
[22] N. K. King a n d M. E . Winfield, J . A m . Chem. S O C . , 83, 3366
(1961).
1231 A. F. Mabrouk, H. J. Dutton, a n d J. C . Cowan, J . A m . Oil
Chemists’ SOC.,41, 153 (1964).
[24] J. Kwiatek and J. K. S e y l e r , unpublished r e s u l t s , 1963-1966.
[25] G. A. M i l l s , S. W e l l e r , a n d A. W h e e l e r , J . Phys. Chem., 63,
403 (1959).
[26] W. P . Griffith and G. Wilkinson, J . Inorg. NLLCZ.Chem., 7, 295
(1958).
[27] R. N a s t , H. Ruppert-Mesche, a n d M. Helbig-Neubauer, 2 .
Anovg. Allgem. Chew., 312, 314 (1961).
[28] N. K. King and M. E . Winfield, J . An?. Cbzem. Soc., 80, 2060
(1958).
1291 B. d e V r i e s , Koninkl. Ned. Akad. Wetenschafi. PYOC.Ser. B63,
443 (1960).
PENTACY ANOCOBALTATE ffI) 69
[30] B. d e V r i e s , J . Catalysis, 1, 489 (1962).

[31] 0. P i r i n g e r and A. Farcas, 2. Physik. Chem. (Frankfurt), 46,
190 (1965).
[32] J. H. Bayston, R. N. Beale, N. K. King, a n d M. E . Winfield,
Australian J . Chem., 16, 954 (1963).
[33] W. P. Griffith a n d G. W i l k i n s o n , J . Chem. Soc., 1959, 2757.
1341 K. N. Raymond and F. Basolo, Znorg. Chem., 5 , 949 (1966).
[35] T. Mizuta and T. Kwan, Nippon Kagaku Zasshi, 86, 1010 (1965);
C A , 64, 11914e (1966).
[36] U. Schindewolf, Ber. Bunsenges. Physik. Chem., 67, 219 (1963).
[37] U. Schindewolf,J. Chim. Phys., 60, 124 (1963).
[38] W. P. Griffith, L. Pratt, and G. Wilkinson, Nature, 182, 466
(1958).
[39] L. Simandi and F. Nagy, inproceedings of the Symposium on
Coordination Chernist,vy, Tihany , Hungary, 1964, Akademiai
Kiado, Budapest, 1965, p . 83.
[40] J. H a l p e r n , Chem. Eng. News, 44(45), 68 (1966).
[41] L. Simandi, Magy. Kem. Folyoirat, 71, 141 (1965); CA, 63,
6613f (1965).
[42] L. Simandi a n d F. Nagy, Acta Chim. Acad. Sci. Hung., 46, 1 0 1
(1965).
[43] M. O r c h i n , Aduan. Catalysis, 5, 385 (1953).
[44] G. N. S c h r a u z e r and R. J. Windgassen, J . A m . Chem. SOC., 88,
3738 (1966).
[45] R . Ripan, A. Farcas, and 0. Piringer, 2 . Anorg. AElgem.
Chem., 346, 211 (1966).
[46] J. Kang, Nippon Kagaku Zasshi, 84, 55 (1963); CA, 59, 15208d
(1963).
[47] M. M u r a k a m i , J. Kang, H. Itatani, S. Senoh, and N. Matsusato,
Nippon Kagaku Zasshi, 84, 48 (1963); CA, 59, 152078 (1963).
[48] 0. P i r i n g e r a n d A. Farcas, Nature, 206, 1040 (1965).
[49] 0. P i r i n g e r and A. Farcas, 2. Physik. Chem. (Frankfurt), 49,
3 2 1 (1966).
[50] A. H a i m and W. K. W i l m a r t h , J . Am. Clzem. SOC., 83, 509 (1961)
[51] J. Kwiatek, I. L. M a d o r , a n d J. K. S e y l e r , Aduan. Chem. S e r . ,
37, 201 (1963).
[521 D. N. Hume and I. M. Kolthoff, J . A m . Chrm. SOC., 71, 867
(1949).
[53] J. Halpern and J. R a b a n i , J . Am. Chem. SOC.,88, 699 (1966).
[54] J. H. Bayston, F. D. Looney, and M. E . W i n f i e l d , A u s l r a l i a n J .
Chem., 16, 557 (1963).
[55] J. H. Bayston a n d M. E. Winfield,J. Calalysis, 3, 123 (1964).
[56] E . P a g l i a , A t t i Accad. Nazl. Lincei Rend. Classe Sci. Fis. Mat.
Nat., 24, 725 (1958); C A , 53, 12908c (1959).
70 JACK KWIATEK
[57] T . Suzuki and T. Kwan, Nippon Kagaku Zasshi, 87, 342 (1966);
C A , 65, 14487f (1966).
[58] A. A. Vlcek and F. Basolo, Inorg-. Chem., 5, 156 (1966).
[59] W. P. Griffith and G. Wilkinson, J . Chcm. Soc., 1959, 1629.
[60] M. J. Mays and G. Wilkinson, J . Chem. SOC., 1965, 6629.
1611 F. Burriel-Marti and R. Gallego-Andreu, Mikrochim. Actu.,
1962, 376.
[62] A. W. Adamson,J. An?. Chem. Soc., 7 8 , 4260 (1956).
[63] A. Haim, R. J. Grassi, and W. K. Wilmarth, Advaiz. Chem.
Ser., 49, 31 (1965).
[64] R. C. Pinkerton and F. R. Duke, J . A m . Chem. Soc., 74, 1535
(1952).
1651 A. A. Blanchard and P. Gilmont,J. A m . Chem. Soc., 62, 1192
(1940).
[66] J. P. Candlin, J. Halpern, and S. Nakamura, J . Am. Chem. SOC.,
85, 2517 (1963).
1671 J. Halpern and S. Nakamura, J . A m . Chem. Soc., 87, 3002 (1965).
[68] J. L. Burmeister, Inorg. Chem., 3, 919 (1964).

[69] J. L. Burmeister and M. Y. Al-Janabi, Inorg. Chenz., 4, 962
(1965).
1701 J. L. Burmeister and D. Sutherland, Chem. Commun.., 1965,
175.
[711 S . C. Chan and M. L. Tobe, J . Chem. Soc., 1963, 966.
1721 G. W . Chantry and R . A . Plane, J . C k m . PhYs., 33, 736 (1960).
[73] H. S. Nagarajaiah, A. G. Sharpe, and D. B. Wakefield, Proc.
Clzevn.. SOC., 1959, 385.
[74] F. P. Dwyer and A. M. Sargeson, Natuve, 187, 1022 (1960).
[75] H. E . Swift and B. E . Douglas J . Inorg. Nucl. Clzem., 26, 601
(1964).
1761 J. K. Kochi and D. Buchanan, J . An?. Chem. Soc., 87, 853 (1965).
[77] J. Halpern and J. P, Maher, J . Atn. Clzem. Soc., 87, 5361 (1965).
[78] J. Kwiatek and J. K. Seyler, J . O?Fganoinetal.Chem. ( A m s t e r -
dam), 3, 421 (1965).
[79] J. Kwiatek and J. K. Seyler,Aduan. Chcm. S e r . , 70, (1967),
in p r e s s .
[80] J. Kwiatek and J. K. Seyler, J . Organomelal. Clzenz. (Amster-
d a m ) , 3 , 433 (1965).
[81] J. Kwiatek and J. K. Seyler, in Proceedings of the 8th Inter-
national Conference on Coordination Chemistry, Vienna, 1964
(V. Gutmann, ed.), Springer, Vienna, 1964, p. 308.
[82] J. Kwiatek, I. L. Mador, and J. K. Seyler, J . Am. Clzem. Soc.,
84, 304 (1962).
[831 M. S. Spencer and D. A . Dowden (to Imperial Chemical Indus-
t r i e s Ltd), U.S. P a t . 3,009,969 (1961).
[84J T . Suzuki and T. Kwan, NipDon Kagaku Zasshi, 86, 713 (1965);
CA, 64, 6473e (1966).
1851 T. Suzuki a n d T. Kwan, Nippon Kaguku Znsshi,86, 1198 (1965);
CA, 64, 11070c (1966).
[86] T. Suzuki and T. Kwan, Nippon Kagaku Z a s s h i , 86, 1341 (1965);
C A , 65, 12097d (1966).
[87] J. Kwiatek and I. L . Mador (to National D i s t i l l e r s and Chemical
Corp.), US. Pat. 3,205,266 (1965).
[88] J. Kwiatek and I. L. Mador (to National D i s t i l l e r s and C h e m i c a l
Corp.), U.S. Pat. 3,205,217 (1965).
[89] A. F. Mabrouk, E . Selke, W. K. Rohwedder, and H. J. Dutton,
J . A m . Oal Chemists’ Soc., 42, 432 (1965).
[go] M. Murakami a n d J. Kang, Bull. Chem. Soc. Japan, 36, 763
(1963).
[91] L. Simandi and F. Nagy, Acta Chim. Acad. Sci. Hung., 46, 137
(1965).
[92] L . Simandi and F. Nagy, Mugy. Kem. Folyoirat, 71, 6 (1965);
CA, 62, 13006b (1965).

[93] M. Murakami, K. Suzuki, and J. Kang, Nippan Kugaku Zasshi,
83, 1226 (1962); CA, 59, 13868a (1963).
[94] R. Mason and D. R. R u s s e l l , Chem. Commun., 1965, 182.
[95] M. J. Mays a n d G. Wilkinson, Nutuve, 203, 1167 (1964).
[96] D. I. Davies, J. N. Done, and D. H. Hey, Chem. Commun., 1966,
725.
[97] C. R. W a r n e r , R. J. Strunk, and H. G. Kuivila,J. Ovg. Chem.,
31, 3381 (1966).
[98] A. Nickon, J. L. Lambert, R. 0. W i l l i a m s , a n d N. H. Werstiuk,
J . A m . Chem. SOC.,88, 3354 (1966).
[99] M. Murakami, R. Kawai, and K. Suzuki, Nippon Kagaku Zasshi ,
84, 669 (1963); CA, 60, 4053b (1964).
[loo] D. S. Breslow and R. F. Heck, Chem. Ind. (London), 1960, 467.
[loll T. H. Coffield, J. Kozikowski, a n d R. D. C l o s s e n , d. 0%.
Chem., 22, 598 (1957).
[lo21 G. V. Smith and J. A. Roth, J . A m . Chem. Soc., 88, 3879 (1966).
[103] M. M u r a k a m i , K. Suzuki, M. Fujishige, and J. Kang,Nippon
Kuguku Zasshi, 85, 235 (1964); CA, 61, 13408b (1964).
[104] M. Murakami, K. Suzuki, H. Itatani, M. Kyo, and S. Senoh (to
Asahi Chemical Industry Co., Ltd.), Japan. Pats. 21,058-21,059
(1962).
[ 1051 A. K a s a h a r a and T. Hongu, Yumugata Daiguku Kiyo, Shizen
Kuguku, 6, 263 (1965); CA, 65, 2 1 2 0 ~(1966).
[ 1061 A. K a s a h a r a and T. Hongu, Nifipon Kugakzl Zasshi, 86, 1343
(1965); CA, 65, 13584c (1966).
[ 1071 D. G o e r r i g , Ger. Pat. 1,099,506 (1961).
72 JACK KWIATEK
[lo81 (To Ciba Ltd.), Meth. Appl. 6,400,332 (1964).

[ 1091 A. Blum and G. Meyer (to Badische Anilin- und Soda- Fabrik
A.-G.), Belg. Pat. 613,287 (1962).
[ l l O ] D. Goerrig, Brit. Pat. 954,597 (1964).
11111 A. A. Vlcek and A. Rusina, Proc. Chew. Soc., 1961, 161.
[ 1121 F. Martin, M. Osgan, and P. Teyssie (to Institut Francais du
P e t r o l e , des Carburants e t Lubrifiants), Fr. Pat. 1,421,283
(1966).
[113] M. Takahashi, Bull. Chew. SOC. Japan, 36, 622 (1963).
[114] M. A. Youker (to E. I. du Pont de Nemours and Co.), Brit. Pat.
590,191 (1947).

Reactions Catalyzed by Pentacyanocobaltate II

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Reactions Catalyzed by Pentacyanocobaltate II

Transféré par

Droits d'auteur :

Formats disponibles

This article was downloaded by:

On: 20 January 2011

Reactions Catalyzed by Pentacyanocobaltate(II)

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

I . INTRODUCTION ANDSCOPE ..................... 38

I. INTRODUCTION AND SCOPE

A s early a s a century ago Descamps [ 11 noted that solutions of

determined than for a comparable heterogeneous process. A variety

light brown cobaltous cyanide which initially precipitates redissolves,

Freshly prepared solutions of this complex have a magnetic mo-

cipitated from these solutions i s diamagnetic and therefore formu-

B. Reaction with Water

The reduction in paramagnetic susceptibility of [Co(CN),P- solu-

The complex [CO(CN),OH,]~-is occasionally formulated a s its con-

[CO(CN),OH,]~- + OH- =+ [Co(CN),OHI3- + H,O (3)

At elevated temperatures and in the presence of excess cyanide,

~ [ C O ( C N ) , ] ~+- 2CN- + 2H,O - 2[Co(CN),I3- + H, + 20H- (4)

The formation of a complex analyzing for [ C O ( C N ) $ C ~ H ~ ]from

C . Reaction with Hvdrogen

Solutions of [Co(CN),I3- prepared in a hydrogen atmosphere ab-

2[Co(CN),I3- + H, =+ Z[CO(CN),H]~- (5)

The product obtained by reduction of [Co(CN),I3- with borohydride

The hydrido complex has been formally considered to contain co-

of various ligands in cobalt(I1) and cobalt(II1) amine complexes by

111. REACTIONS WITH INORGANIC COMPOUNDS

1. Reaction with ICo(CN),13-

Hydrogen peroxide r e a c t s rapidly with [CO(CN),]~-

~ [ C O ( C N ) , ]+~ -H,O, + 2H,O -.~[CO(CN),OH,]~-+ 2 0 H - (6)

Hydrogen peroxide is catalytically and quantitatively reduced by

to hydrogen. The reaction i s considered to involve the intermediate

[Co(CN),HI3- + [CO(CN),OH,]~-+ OH- * Z[CO(CN),]~-+ 2H,O (8)

2[Co(CN),J3- + H, + 2[Co(CN),HI3- (9)

Although [Co(CN),OH,Y- is reducedby [Co(CN),HJ3-, [CO(CN),]~-is

been elucidated. The description of the hydrido complex as contain-

1. Reaction with [Co(CN),13-

Solutions of [CO(CN)~]~- r e a c t with excess oxygen to form a brown

2[Co(CN),J3- + 0, - [(NC),COOOCO(CN),]~- (10)

However, gradual introduction of oxygen into solutions of

Acid-catalyzed hydrolysis, resulting in the formation of hydrogen

lution [24], presumably takes place in a s i m i l a r manner.

[(NC),COOOHCO(CN),]~-+ H,O - [Co(CN),OOHI3-

[CO(CN),OOH]~-+ H30+ - [CO(CN),OH,]~-+ H,O, (15)

F u r t h e r oxidation of the peroxy complex with molecular oxygen

Oxygen is reduced by hydrogen in the presence of [Co(CN)$-

[CO(CN),H]~-+ 0, -.[CO(CN),OOH]~- (16)

[Co(CN),HIS- + [Co(CN),00H13- f 2H' - ~[CO(CN),OH,]~- (17)

In contrast to the reduction of hydrogen peroxide, oxygen i s only

1. Reaction with [Co(CN),13-

Solutions of ~CO(CN),]~-a r e oxidized by iodine, bromine, and chlo-

rine to the corresponding halopentacyanocobaltate (111) complexes [ 621

2[Co(CN),IS- + X, - 2[Co(CN),XIS- (18)

Side reactions occur, with formation of [ Co(CN),IS- and LCO(CN)~OH,]~-,

Iodine and bromine are catalytically reduced by hydrogen in the

Permanganate , dichromate , and f erricyanide a r e catalytically re-

The substitution of various pentammine cobalt(II1) complexes

[CO(NH,),X]~' by cyanide ion i s catalyzed by [Co(CN),13- [62]. Two

[Co(NH,),]'+ + 5CN- - LCO(CN),]~-+ 5NH, (20)

When X = PO:-, COi-, SO:-, NH,, or OAc-, the reaction r e s u l t s in

[Co(CN),13- + CN- * ~CO(CN),]~- (21)

[Co(CN),I4- + [Co(NH,),X]*' -.[Co(CN),13- + [Co(NH,),X]+ (22)

[Co(NH,),X]' + 5CN- - [Co(CN),I3- + 5NH, + X- (23)

mediates has been obtained by detection of the transient intermedi-

IV. REACTIONS WITH ORGANIC COMPOUNDS