close

Вход

Забыли?

вход по аккаунту

?

Dinitrogen-Transition Metal Complexes Synthesis Properties and Significance.

код для вставкиСкачать
[51] Th. Kauffmonn, Angew. Chem. 76, 206 (1964); Angew. Chem. internat.
Edit. 3. 342 (1964).
[52] Cf. M . J . S. Dewar, Angew. Chem. 83,859 (1971); Angew. Chem. internat.
Edit. 10, 761 (1971).
[53] R. Eidenschink and Th. Kauffmann, Angew. Chem. 84,292 (1972): Angew.
Chem. internat. Edit. 1 1 , 292 (1972).
[54] W Bannwarth, Diplomarbeit, Universitat Miinster 1974; W Bannwarth,
R . Eidenschink, and Th. Kauffmann, Angew. Chem. 86, 476 (1974); Angew.
Chem. internat. Edit. 13, 468 (1974).
[55] A cycloaddition of this type might take place in the formation of
a cyclic dimer (39%) during the actlon of sodium on a-methylstyrene ( M .
Kolobielski and H. Pines, J. Amer. Chem. SOC.79, 5820 (1957)).
[56] G Boche and D. Martens, Angew. Chem. 84, 768 (1972); Angew. Chem.
internat. Edit. I I , 724 (1972)
[57] G Boche, personal communication 1973
[58] J . E. Mulraney and D . Savage, J. Org- Chem. 36, 2592 (1971).
[59] T h e reactions of (63) and similar compounds with alkenes and other
unsaturated compounds are now in the course of thorough investigation:
cf. ref. [54].
[60] Rearrangement of 1,3-diphenylpropyne to 1.3-diphenylallene by bases:
E. V Dehmlow. Chem. Ber. 100. 3260 (1967): H . A. Staab and H.-A. Kurmeier.
ibid. 101, 2697 (1968).
[61] D. Serbach, Angew. Chem. 81. 690 (1969): Angew. Chem. internat.
Edit. 8. 639 (1969). and literature cited therein: mechanism: K . D. Berlin,
8.S . Rathore and M Peterson, J. Org. Chem. 3U, 226 (1965).
[62] R . B. Bates, L. M . Kroposki, and D. E . Potter, J. Org. Chem. 37,
560 (1972).
[63] A. Alemagno, T Baccetti, and C. Rizzi, Gazz. Chim. Ital. 102, 311
(1972).
[64] L. Claisen, Ber. Deut. Chem. Ges. 36, 3673 (1903); J . Goerdelrr, J .
Ohm, and 0 . Tegrmeyer, Chem. Ber. 89, 1534 (1956): R. L. Letsinger and
D. F . Polhrt, J. Amer. Chem. SOC. 78, 6079 (1956); H . Kloosrerzirl, J . A.
A. can Drunen, and P. Galama, Chem. Commun. 1969, 885; P. Eberhard
and R. Huisgen, J. Amer. Chem. SOC. 94, 1345 (1972).
[65] 1,3-Diphenyl-2-azaallylsodiumand -potassium were prepared in solution by the thermal ring opening of cis-1,2-diphenyl-l-aziridinylsodiumand
-potassium in THF at 20°C 1191, a s well a s by the action of N a N H l or
K N H 2 on N-benzylidenebenzylamine in ether at 0 ° C [17, 661 A solution
of 1,3-diphenyl-2-azaallylmagnesiumbromide was obtained by the action
of mesitylmagnesium bromide on N-benzylidenebenzylamine in THF at 67°C
C661.
[66] D. Beryrr and Th. Kauflmann, unpublished work.
[67] Cf. P. Hrrmbach and R. Traunmiiller Chemie der Metall-Olefin-Komplexe. Verlag Chemie, Weinheim 1970; W P. Giering and M . Rosenblum,
J . Amer. Chem. SOC.93,5299 (1971): A. Rosau, M Rosenblum, and J . Tancredu,
ibid. 95, 3062 ( 1 973).
Dinitrogen-Transition Metal Complexes: Synthesis, Properties, a d
Significance
By Dieter
Sellmann[*]
Compared to the large number of CO complexes, N 2 complexes are still rare. In certain
cases they may be formed from N 2 gas and metal compounds under physiological conditions,
and are therefore frequently considered as possible intermediates in N z assimilation. However,
despite numerous attempts it has not yet been possible to reduce the N Z ligand of a fully
characterized N z complex to N H 3. This negative evidence recently prompted the discoverers
of the first dinitrogen complex, the [Ru(NH&N2lZ@ ion isolated in 1965, to express doubt
whether such complexes really d o play a part in the enzymatic reduction of N2. The latest
findings nevertheless incline toward a somewhat more optimistic view. Thus, the mild partial
reduction of the N 2 ligand in metal complexes with inert gas configuration, hitherto considered
nonreducible,justifies the hope that a suitable system permitting the catalytic reduction of molecular nitrogen under normal conditions will one day be found. Even apart from any technical
potential, the N complexes constitute an interesting chapter of modern inorganic chemistry.
1. Introduction
Under normal conditions molecular nitrogen is usually found
to be chemically inert. Thus, reactions involving the Nzmolecule generally require drastic conditions, e. g. elevated temperatures and pressures, as in the Haber-Bosch process.
However, nature is capable of converting atmospheric nitrogen
enzymatically during N 2 assimilation into ammonia and
amino acids under mild physiological conditions ( 1 5 "C,
0.8atm. N2, aqueous medium). In chemistry, only one defined
reaction of the N 2 molecule is currently known which also
proceeds under physiological conditions: the reaction with
metal compounds to yield dinitrogen complexes. This reaction
[*] Dr. D. Sellmann
Anorganisch-chemisches Laboratorium der Technischen Universitat
8 Miinchen 2. Arcisstrasse 21 (Germany)
Angew. Chem. internat. Edit. / Vol. 13 ( 1 9 7 4 ) I N O . 10
can clearly be considered as a model for the absorption of
atmospheric nitrogen by the Nz-assirnilating enzymes in
nature. Hence the fundamental significance of the reaction
and its products.
What features characterize the reaction partner, which allows
reduction of the normally inert molecular nitrogen during
the course of assimilation?
Biochemists are trying to answer this question by the isolation
and study of the nitrogen-assimilating enzymes, the nitrogenases; purely chemical means are employed by chemists
attempting to simulate nature in uitro by fixing molecular
nitrogen to synthetic compounds. Since transition metalsespecially molybdenum and iron-are present in the nitrogenases, systems containing these metals are of particular
interest. What the chemist wants is a defined system in which
he can follow the reduction of the Nzmolecule step by step.
639
Isolable and unambiguously characterizable model substances
thus appear especially important.
Nz-fixing chemical systems can be roughly divided into protic
and aprotic. The impetus which led to the investigation of
the aprotic group was the observation that Ziegler-Natta catalysts and similar compounds absorb nitrogen and reduce it
to nitride-type products. The fundamental difference between
these systems and the nitrogenases lies in their aprotic character: the hydrolytic susceptibility of the organometallic components ( e . g. Grignard reagents, alkylaluminum derivatives,
alkali metal naphthalides) makes rigorous exclusion of water
essential. The reactions which proceed in such systems are
therefore reminiscent of the well-known reaction between
nitrogen and lithium to give Li3N. Coupled with the fact
that, with few exceptions, unambiguously characterized compounds could not be isolated, this proved to be a disadvantage
compared with the second group, the protic systems. These
are capable of existing in aqueous or other proton-containing
solvents, and the reaction with molecular nitrogen yields welldefined products, the N 2 complexes, which will be discussed
below I.
2. Definition, Nomenclature, Possible Modes of Bonding
Dinitrogen complexes are coordination compounds which
contain the Nz molecule as ligand. The first example of this
class of compounds was discovered by Allen and Senoff in
1965[’]. The attempted synthesis of ammine complexes from
RuC13 and hydrazine hydrate did not yield the expected
[Ru(NH3)6I2@ion but the stable complex [(NH&RuN2]CI2
instead.
Long before thisdiscovery, the realization that Nz-assimilating
bacteria require molybdenum in order to fix atmospheric
nitrogen had led to the hypothesis that dinitrogen-transition
metal complexes play a part in this process[3! However,
numerous unsuccessful attempts to prepare such complexes
as well as subsequent theoretical considerations led to the
assumption that these compounds were not stable.
The isolation of the first stable N z complex, and the hope
ofobtaining information on N 2 assimilation through the study
of this new class of compounds, stimulated an intensive search
for further examples. The result is a respectable number of
new dinitrogen complexes; compared with the hundreds of
complexes formed by the isoelectronic carbon monoxide, however, they must still be considered a rarity.
According to the IUPAC rules14a1,the N2 molecule is named
“dinitrogen” when it is a formally uncharged ligand in coordination compounds. Other proposals such as “nitrogenyl”
(by analogy with “carbonyl”), “nitrogeno”, or “nitrogen”
should therefore be avoided.
In principle, several possibilities are conceivable for the linkage
between dinitrogen and metals (Scheme 1). These are in particular : terminal (“end-on”) bonding [ ( I ) ] , lateral (“side-on”,
or “edge-on”) bonding [ ( 2 ) ] , bridge bonding via one of the
N atoms [(3)], terminal bridge bonding via both N atoms
with a metal atom on each nitrogen [ ( 4 ) , ( 5 ) ] o r two metal
atoms on each nitrogen [ { 6 ) ] . ( 1 ) and (3) correspond to
the bonding in terminal and bridged carbonyl complexes respectively, (2) is comparable to the bonding in acetylene com-
640
plexes. In (I), (2),and ( 4 ) the triple bond in the N2 molecule
is largely retained, but not so in ( 3 ) , ( 5 1 , and (6). ( 5 ) is
therefore to be regarded as a complex of the doubly depro-
Scheme 1. Linkage possibllities of the N 2 ligand
tonated diimine HN=NH, and (6) as a fully deprotonated
hydrazido complex, rather than as dinitrogen complexes.
So far, only bonds of types ( I ) , ( 4 ) , and (6) have been
unequivocally demonstrated in Nz complexes; (2) has been
postulated. The structural unit (6) is present in the complex
[{(PhLi)3Ni)2N2.2Et2OI2[*1.
Here the two Ni atoms and
the two atoms of the N2 molecule occupy the corners of
a distorted tetrahedron: two such [Ni2N2] units are bridged
cia two Li atoms. A side-on 71 bond between the N 2 molecule
and the metal centers is considered as a possibility; accordingly, ethylene readily displaces the N z ligand from the
complex. The length of the N-N bond was found by X-ray
structure analysis to be 1.35A, the highest value yet encountered in a N2 complex. This value approaches that of an
N-N single bond (1.47 8)and points to a considerable activation of the N 2 molecule[4b~.
A Pt complex, formed from ~is-[PtC12(PPh3)~]
and hydrazine,
contains the structural unit (3) in protonated form‘51(Fig.
1).
Flgure 1. Schematic structure of [jPt(PPh,)>N2Hi2]”.
Molecules related to terminal N2 complexes include N 2 0 ,
HN,, CH2N2,C,H5NYXe, and 1,10-Bi,H,(N2)2,which may
be considered as an internal diazonium salt[h1.These nonmetallic compounds containing the N grouping are not discussed
here, nor are substances such as azobenzene, Ph-N=N-Ph,
which are only formally related to bridged N complexes.
3. Preparation of Nitrogen Complexes
Table I summarizes the data on unequivocally characterized
N complexes. Apart from chloride and ammonia, phosphanes
and hydride ions appear remarkably often as coligands of
Nz. It was not until very recently that N 2 complexes were
synthesized with arene and CO ligands. In view of the high
sulfur content of the nitrogenases, it is surprising that so
far only one very unstable and poorly characterized N Z
[*] For abbreviations see Table I , footnote [a].
Angew. Chem. internat. Edit. 1 Vol. 13 (1974)
1 No.
10
Table I . Some dinitrogen complexes.
~~~~
.
~~
.-
- ... .
-__
. .
Complex [a]
. .
~
Color
...
_-
.
2145 [h]
2132 [b]
21 12 [h]
.
.
. ...
yellow-orange
d"
2040
orange
dark red
yellow
d 'l
d'
d"
orange
d ',
2000 [L]
1925 [L]
1911 [d]
199x
1957 [d]
[L]
I975
~
_..
~~~
reddish-brown
yellow
yellow
yellow
yellow
red
.
. ... -.
..
~
~
.
-
2169 [b]
2141 [b]
1925 [el
19x0 [c]
2060 [d]
7060
d"
d"
d"
d"
d"
d'
-
.
ycllow
orange
d6
d"
d
white
white
yellowish
ycllouish
yellowish
d"
d ''
d"
d"
d6
2147 rgl
2 I 70
21 14 [g]
H hite
d
2157 [g]
d"
20x5
20x2
2057
202x
2 I20
217s
[d]
[g]
2093
[fl
19x9 [gl
2057
2090
?'
rg]
-
r110 [g]
2220 [g]
2190
M
d ''
d"
hitr
nhite
yellowish
~
d ''
d6
...
~~
.
2090 [d]
1x75 [f]
2 IS2 [g]
2105 [c]
.
- -
orange
~
-.
. ..-
. ...
-
2076 [d]
~~
hlue
red
red
brown
brown
.-
. -.
[d]
~~
red
rcd-orange
hlack
yellowish
).ellow
-
[g]
.-
-
12x0
1910 [h]
1910
-~
dark red
yellowish
red-blac k
red
gold yellow
gold brown
-
187.5
2054 [h]
1660
1761
2100 [h]
20x0
dark hroiln
dark red
dark red
[a] Ahhre\iations: acac = 3.5-pcntanedionate, Cy = cyclohexyl ; das =ethylenehisldiphenylarsanc)Ph,AsCH2CH2AsPh2, dcpe = cthylcnebis(dieth~1phoaphaiic)
Lt, PCH,CH PEt ; diphos = ethylenchis(dipheny1phosphane) Ph, PCH > C H , PPh, : dmpe = ethylenehis(dimethy1phosphane) Me, PCH ,CH PMe, ; dpsc = 1.2his(pheny1thio)ethaiie PhSCH,CH,SPh; en=ethylenediaminc; Et -ethyl; h m b = hexamethylhenrenc, M =metal: Mc=methyl. mes=mesitylenc: Ph-phenyl;
THF=tetrahydrofuran; tol=toluene.
[b] In herane. [c] in toluene; [d] in benrenc; [el in chloroform. [f] in tetrahydrofuran; [g] in n u j o l ; [h] in the Raman spectrum.
[I] Recently the first N, complex with a porphyrin ligand was prepared: ~~?r~i~~-o~taethylporphinato(dinitrogen)tetrahydrofur~inosniium(i~).
see J . W Birchlrr. a n d
P. D. Snirrh. Angew Chem.. in press.
, ,
,
complex with a sulfur-containing ligand is known: trans[M O ( N ~ ) A P M ~(PhSCH
~ P ~ ) 2CH
~ SPh)].
Table 1 does not include N 2 complexes obtained by matrix
isolation at low temperatures (ca. - 196"C), e . g . CrN2, NiN2,
PtN 2I 7 I , Ni(N dn,Pd(N2),181, and Ni(CO)3N2l9].
Angew. Chrm. internat. Edit.
Vol. 13 ( 1 9 7 4 ) I No. 10
Thus far three possibilities exist for the preparation of N,
complexes:
1 ) Direct reaction of N z with a metal complex, frequently
under strongly reducing conditions.
641
2) Conversion of a metal complex with ligands containing
N-N bonds, e. g. N2H4 or N?.
3) Formation of an N2 triple .bond in a metal complex by
reaction of ligands such as NH3 with a nitrogen-containing
reagent, e. g. H N 0 2 .
3.3. Formation of an N 2 Ligand from an N 1 Ligand
Ammine complexes can be diazotized with nitrous acid, but
this method is of very limited application.
[Os(NH,)5N21Ze
+ HNOz
+
c . i s - [ O ~ ( N H , ) ~ ( N 2 ) 2+] ~2~H 2 0
(15)
3.1. Direct Methods
4. Structure of Dinitrogen Complexes
A metal complex in the presence of excess ligand is reduced
under N, by metals, metal amalgams, or hydrides.
N,
[WCI,(PMe,Ph),]+Na/Hp
--j
A
toluene
[MoCI,(PMe,Ph),]+Na
+ NaBH,
[FeHCl(depe),]
[W(N,),(PMe,Ph),]
(1)
[Mo(N,),(PMe,Ph),]
(2)
[FeH(N,)(depe),]BH,
a,,~~ne
+ NaCl
(3)
Alkylmetal compounds may also be used as reducing agents.
Presumably they introduce an alkyl group into the metal
complex; the ensuing olefin elimination produces a hydride
complex which then reacts with the NZ. Suitable reactants
include AIEt,(OEt), AlEt,, and Al(iBu),. AIMe, yields mainly
methyl complexes.
4.1. Molecular Structure
That the metal-N-N or metal-N-N-metal unit in terminal
N z complexes is approximately linear is apparent from the
strong IR absorption of the N z stretching vibration; a side-on
linkage of the N 2 ligand, as in ethylene or acetylene complexes,
would produce only weak IR absorption. Clear evidence for
a linear arrangement is provided by X-ray structure analysis
(Fig. 2), although accurate determination of bond lengths
failed in many cases owing to disorder phenomena.
N
I
N
C)
Figure2. Schematic structures of(a)[CoH(N&PPh,),], (b) [ O S ( N H ~ ) ~ N ~ ] " ,
and (c) [ { R u ( N H ~ ) ~ } z N ~ ) ~ " .
Labile ligands in metal complexes can be displaced by Nz.
These are the clearest reactions of molecular nitrogen; they
proceed under mild conditions and are frequently reversible.
[CoH,(PPh,),]
+ N,
EtOH
+ N,
[Ru(NH,),H,O]'"
[Ru(NH,),N,]~"
2[C5H,Fe(dmpe)acetone]"
+ N,
-
+ H,O
(8)
acetone
[/C,H5Fe(dmpe)},N,]2@
[(hrnb)Cr(CO),]
17)
d [CoH(N,)(PPh,),l +.H,
+ 2acetone
+ T H F 3 [(hrnb)Cr(CO),THF] + CO
2[(hrnb)Cr(CO),THF]
THF
+ N,
[ j(hmb)Cr(CO),/,N,]
(9)
(10a)
+ 2THF ( l o b )
In some cases the intermediates cannot be isolated but immediately react further with Nz, as in the reaction according
to eqs. (1Oa) and (lob).
3.2. Conversion of a Ligand with N-N
Bonds into N2
+ PhCON,
0-C
CHCI,
truns-[lr(N,)CI(PPh,),]
/O\
[(PPh,),Cl,Re--N=N-CPh]
+ ZPPh,
truns-[ReC1(N,)(PPh3),]
[C,H,Mn(CO),N,H,]
(RuCl(das),N,]PF,
+ 2H,02
+ NOPF,
-
+ HCI + PhC0,Me
(11)
(12)
CU'".THF
-40
[C,H,Mn(CO),N,]
[RuCI(N,)(das),]PF,
642
+ PhCONCO
+4H,O
+ by-products
Since the nitrogenases also reduce acetylene (to ethylene),
it should be pointed out that the structures of (end-on) N Z
complexes bear not the slightest resemblance to the acetylene
complexes in which the two C atoms of the C z H 2 are linked
to the metal in the same manner. Instead, they are similar
to acetylide complexes, e. g. Kz[Cr(C0)3(C-CH)3][561.
4.2. Bonding
Most systematic syntheses of N, complexes were successful
by this method.
rruns-[lr(CO)C1(PPh,),]
Table 2 summarizes the most important data.
The N-N distances of coordinatively bound dinitrogen are
only marginally greater than in the free N2 molecule. The
complex [{(PhLi)3Ni) 2Nz-2Et20]2[4bJ,which contains the
structural unit (6) (Scheme 1) and in which ~ N - N = 1.35A,
is an exception, as is [ReC1(PMezPh)4-N z-MoC14(0Me)]
with ~ N - N = 1.21 A[67b! For comparison, the length of the
NN double bond (in diazomethane) is 1.24A and that of
the N N single bond (in hydrazine) is 1.47
(13)
(14)
The structures of N 2 complexes largely resemble those of
the analogous isoelectronic C O complexes, thus revealing similarities in electronic structure. However, while carbon monoxide forms numerous mono- and poly-CO complexes, only
a few monocomplexes are known for dinitrogen; among the
higher complexes, only bis-N, species have been reported.
The reason for this, and for the generally greater stability
of CO complexes, lies in the electronic differences between
N,and CO.
The following discussion of the bonding properties of NZ
and CO should throw some light on this aspect. Figure 3
shows the molecular orbital scheme of dinitrogen, as derived
Angew. Chem internat. Edit.
1 Vol. 13 (1974) 1No. I0
Table 2. Molecular data for some dinitrogen compounds
1.0976
1.12 [b]
1.12
1.106
1.06 [b]
2.10
1.84
1.89
1.97
1.81
1.11
1.10
2.01
1.124
1.12
1.35
1.21
1.928
1.79
1.79; 1.89
1.34
1.24
1.18
1.13
1.134
1.128
180
180
180
1481
[481
1481
[a] M =metal, C, N. or 0.
[b] No accurate determination possible.
from UVr571an photoelectron spectra[581, as well as from
theoretical calculations[59!
0
0
1 xg antibonding
degenerate pair in xy- and xz-planes, - 7 eV,
unoccupied
@
30, weakly bonding
- 15 58 eV, occupied (2e.)
me
0 0
a=N
1 K, bonding
degenerate pair in xy- and
xz-planes, - 17.1 eV,
occupied (4 eC)
C-0 bond from 1.128 to 1.115A in the ionization CO-+CO*.
The ionization N,+NY, on the other hand, results in a lengthClearly
ening of the N-N bond from 1.098 to 1.117
the electron distribution in C O and N2 will also differ because
carbon and oxygen have different radii, electronegativities,
ett., whereas dinitrogen is a homonuclear molecule. C O possesses a small dipole moment (p=0.2D). Calculations of the
electron-density distribution also show that the C atom exhibits a greater lone electron pair charge density than the 0
atom~"*].
The metal-dinitrogen bond in N 2 complexes can be described
according to the MO method by overlapping of the 30,
and In, orbitals of the dinitrogen with suitable d orbitals
of the metal (Fig. 4).
20" weakly antibonding
-18.7eV
occupied ( 2 e - l
2 0 ~bonding
- 35.5 eV (estimated)
occupied ( 2 e . ~ )
Figure 3. Molecular Orbital scheme of dinitrogen
The exceptional stability of dinitrogen compared to other
diatomic molecules is a consequence of its low-lying three
highest occupied bonding molecular orbitals. The actual
H O M O is a bonding 3 0 , orbital of very low energy. Hence
the ionization energy of N is considerable (1 5.58 eV), approximating to that of argon (15.75 eV). The LUMO, 1 n,, lies
8.6 eV above the 3 G~orbital, i. e. so high that it can be occupied
at best by electrons from alkali metals and other strongly
electropositive metals. The absence of orbitals in the energy
range between the two, i.e. precisely that range in which
the occupied and unoccupied orbitals participating in chemical
reactions are located, largely explains the low reactivity of
the N, molecule.
While the M O scheme of carbon monoxide is very similar
to that of dinitrogen, essential differences are noted in the
much lower energetic stability of the highest occupied a and
of the lowest unoccupied x* orbitals ( - 14.0 and 6 eV respectively) and in the smaller energy difference between these
two orbitals, which is lower by 0.6 eV. In addition, the highest
occupied 0 orbital in CO possesses weak antibonding character; this can be deduced also from the shortening of the
~
Angew. Chem. internat. Edit.
1 Vol. 13
(1974)
1 No. I0
Figure 4. Scheme of the M-N2
bond.
As in the case of the M-CO bond, a donor bond is postulated
involving the 3 0, electrons of dinitrogen and an unoccupied
metal d orbital. The resulting increased electron density at
the metal is reduced by back-bonding from nonbonding occupied metal d orbitals to the antibonding l n , orbital of the
dinitrogen. The result is a synergetic 0-71double bond, i.r.
a strengthening of the 0 bonding leads to the simultaneous
strengthening of the II bonding.
This type of bonding explains why N 2 complexes are less
stable than C O complexes. Most metal complexes possess
d orbitals (e. g. e, and t2, orbitals) whose energies correspond
better to those of the 0 and n* orbitals of CO than to those
of the N Z molecule. This enhanced interaction increases the
stability of the C O complexes.
The bond scheme illustrated in Figure 4 also explains why
electronic variations result in largely analogous effects in N,
and C O complexes.
643
What physical changes are to be expected in the N2 molecule
as a result of coordination if the above model of the M-N2
bond is correct'?
The N-N bond should become weaker on complex formation,
since the electrons of the M-N o bond come from a bonding
orbital ofthe N 2and those of the M-N n bond are transferred
to an antibonding N2 orbital. However, this conclusion has
recently been disputed. According to theoretical calculations,
the removal of electrons from the 3o, orbital should reduce
the electron repulsion between the two N atoms and thereby
strengthen the N 2 CT bondlh1I. The weakening of the N2 triple
bond would then be mainly due to the x back-bonding.
The overall weakening of the N, triple bond and consequent
activation of the N, molecule on complexing are clearly
demonstrated by the marked drop in the N, stretching
frequency, which falls by several hundred wave numbers from
the value of 2331 cm- I recorded for the free molecule (Raman
spectrum). Furthermore, the N2 molecule in terminal N 2 complexes is subject to strong electronic polarization, which
accounts for the high intensity of the vN2 absorption in the
IR spectrum. This polarization is also confirmed by the ESCA
spectra of N 2 complexes. Metal-bound and terminal N atoms
exhibit separate N-Is emissions, e. g. at 399.9 and 397.9 eV
for [ReC1(N2)(diphos)z], whereas the emission of the free N2
molecule is observed at 41 1 eV. These values afford a charge
difference of approximately 0.4 e between the two N atoms,
the terminal one bearing the greater negative chargeIb2.631.
Attempts to determine the properties of the N2-bonding coordination site in N2 complexes have been based primarily
on comparative studies, and particularly on systematic variation of the coligands ofdinitrogen. However, since the stability
limits of N z complexes are very narrow owing to the unfavorable energy of the N 2orbitals, only a few examples are available
for studying the effect of the variation of coligands on the
metal-N2 bond. Thus the reaction according to eq. (1 1 ) can
only be effected with triphenylphosphane; tris(p-toly1)phosphane or methyldiphenylphosphane no longer yield isolable
has not been finally resolved. Mossbauer spectra of trans[FeH(X)(de~e)~lBPh4(X=
N2, C0)suggest that CO is a better
o-donor and n-acceptor ligand than NLrzz1.The same can
be concluded from IR intensity measurements on analogous
CO and N2 complexedh6'. Such behavior can also be deduced
from the electronic properties, i. e. the energy of the molecular
orbitals, of the two molecules. However, for the compounds
[CSHSM(CO)ZNZ] (M = Mn, Re) and [ArCr(C0)2N2]
(Ar = benzene, mesitylene, and hexamethylbenzene) it appears
from the 'H-NMR and IR spectra that both the cyclic ligands
and the CO groups must accommodate a greater electron
density in the N2 complexes than in the corresponding tricarbonylmetal complexes. In these compounds therefore the ratio
of o-donor to n-acceptor strength must be greater for dinitrogen than for carbon monoxide[' I h . "1.
'.
Surprisingly, almost the same N2 stretching vibration frequency is often observed in dinuclear N 2 complexes as in
the corresponding mononuclear complexes; examples include
[ ( R U ( N H ~ ) ~ } ~ N(2100cm-')
~]~'
and [ R U ( N H ~ ) ~ N ~ ] ' ~
(2130cm-I). Since the N-N triple bond of the N2 molecule
is weakened by coordination to one transition metal, bonding
to two transition metals was expected to result in further
weakening. This expectation is fuIfilled only in some cases.
An explanation for this phenomenon is provided by the bonding scheme of the M-N-N-M
bond. If the CT bond component is regarded as largely constant, and the changes in
M-N and N-N bonds are attributed mainly to variations
in the x system, the scheme shown in Figure 5 applies to
the TC bonds ofcomplexes with approximate fourfold symmetry,
such as the complex [(RU(NH,),},N.,]~@ or the complex
[( PMe, Ph),ClRe-N,-CrCI,(THF),]
lS5.
"'I.
Experience has shown that the central metal of an N2-bonding
complex must exist in a low oxidation state and possess a
relatively high electron density. A fitting example is
[Ru(NH~)~CO]", the CO analog of [ R U ( N H ~ ) ~ N ~ ] ~ ' [ ~ ~ I .
Despite the formal double positive charge on the ruthenium,
the CO compound exhibits one of the lowest vco frequencies
(1919cm-') ever observed in terminal CO complexes.
The effect of electron density is also apparent from the series
[ ( b e n ~ e n e ) C r ( C O ) ~ N(VN]
~ ] = 2145 cm- I), [(mesitylene)Cr(CO),N,] (vN2= 2132 cm-'), and [(hexamethylbenzene)Cr(CO)2N2] (vN,=21 12cm- I). Introduction of the "ekctrondonating" CH3 substituents into the benzene ring increases
the electron density at the chromium. As a result the frequency
of the N 2 stretching vibration decreases; the bond scheme
in Figure 4 indicates that this phenomenon should be associated with a strengthening of the M-N bond. Accordingly,
the stability of the N2 complexes increases in the above
sequence" '1.
N 2 and CO ligands influence the central metal in very similar
fashion; thus, both exert approximately the same trans effect
on the Ir-X bond in the complexes rran~-[IrX(X')(PPh~)~]
(X = CI, Br ; X'= N,, CO)1641.The much-discussed question
of the o-donor and n-acceptor properties of Nz and CO
644
6x3
___
3e
2b
Figure 5. Qualitative molecular orbital scheme for dinuclear N L complexes
with approximate fourfold symmetry.
The four-center molecular orbitals indicated in Figure 5 are
obtained by linear combination of Md,,, Np,, and M'd,,
orbitals. The z-axis is determined by the CT bond. A set of
energy levels, whose energies increase with increasing nodal
number of the molecular orbitals, corresponds to the molecular
orbitals. Since, in addition to the orbitals shown, there exists
an equivalent set of x molecular orbitals, formed from the
Md,,, Np,, and M'd,, orbitals, the energy levels are degenerate.
For the sake of simplicity they are designated as le, 2e,
erc. The 6 bonds formed from the Md,, orbitals contribute
Angew. Chem. internat. Edit. / Vol. 13 ( 1 9 7 4 ) / No. 10
practically nothing to the bonding; they correspond to the
1 b and 2 b levels. For energetic reasons the I e and 4e levels
correspond largely to the bonding and antibonding x molecular orbitals I x,, and 1 xIIrrespectively, of the Nz molecule
in Figure 3. The 2e and 3e levels possess mainly metal character, as d o the t b and 2 b levels.
and of the N-N bonds
The strengths of the M-N-N-M
thus depend chiefly on the occupancy of the e levels. Four
electrons of the two N atoms fill the 1 e level. This possesses
bonding character for the N atoms; owing to its partial metal
character, however, a weaker N-N bond results than in
the free Nr molecule. The progressive occupation of higher
levek with the 12d electronsofthe two Ru" centers ( M = M ' )
leads to a marked weakening of the N-N bond in the 2e
level, as the orbitals are antibonding with respect to the N-N
bond. The filling of the I b and 2 b levels leaves the N-N
bond unaffected, while the occupancy of the 3e level
strengthens the N-N bond since the orbitals are bonding
with respect to the N atoms. Altogether there are thus five
bonding filled orbitals-the bonding character of the filled
M-N-N-M
CT orbital must of course also be taken into
account here-as against two antibonding filled orbitals with
respect to the N-N
bond: there is therefore hardly any
difference between the N z bonding in [ I R u ( N H ~2Nr]40
)~~
and that in [Ru(NH,)jN,]'@. However, if the metal atoms
d o not possess sufficient electrons to fill the 2 b and especially
the 3c levels, a much reduced v x 2 frequency is observed compared to the mononuclear complex [ReCl(N2)(PMe2Ph)4]
( v y 2 = 1925 cm-I); examples include [ReCl(PMe,Ph)4-NrCrC13(THF)2]( v N ~ =1875cm- I ) , in which the Re' and Cr"'
centers can contribute a maximum of nine electrons, and
[ReC1(PMe~Ph)-NN-TaCIs] ( V Y ? = 1695),in which the metal
atoms can contribute only six electrons for occupation of
the levels.
This effect is particularly well illustrated in [ReCl(PMe2Ph)4N2-MoC14(OMe)]. This complex, containing one more electron, exhibits a V V ? frequency of 1660cm-'. In accordance
with the low V N ? frequency, X-ray structure analysis shows
the long N-N distance of 1.21 A for the Re-N-N-Mo
unit 1' 7hl.
Figure 5 also illustrates the difficulty of reducing the N,
unit in such complexes. The electrons taken up first fill the
2 b and 3e levels and thereby strengthen the N r bond. This
is clearly no longer the case when a lowering of the fourfold
symmetry resolves the degeneracy of the e levels. Such may
be the case in the Fe and Ti complexes mentioned in Section
5.4.
Most of the N z complexes isolated so far obey the inert
gas rule and are diamagnetic. Exceptions include
[ C O N A P P ~ ~and
) ~ ] [ { ( C s H ~ ) z T i i ~ N las
] , well as adducts
of Lewis acids to [ReCI( N ,)( PMe zPh)J].
5. Chemical Reactions of Dinitrogen Complexes
increasing thermal lability of the N2 complex; the residual
coordinatively unsaturated complexes can be stabilized by
the addition of other ligands, r . ~HzO,
.
NHJ, pyridine, PPh3,
or solvent molecules. In the absence of such donors, the resultingcomplexes are frequently suitable as highly reactive catalysts
for the polymerization of olefins. These Nz complexes thus
bear some resemblance to aliphatic diazo compounds which
yield the reactive carbenes on loss of Nz.
5.1. Oxidation
N 2 complexes are particularly sensitive to oxidizing agents.
Strong oxidizing agents are therefore frequently used for the
quantitative determination of NZIh8j.
If the N2 complexes are very stable they can sometimes be
oxidized by one unit without loss of the N z 1igandl''l:
The oxidized complexes are less stable than the starting compounds, in agreement with the increase of 80-100cm-' in
the vx2 frequency which indicates a weakened M-N bond.
Some complexes undergo reversible electrochemical oxidation,
0.9.the [ ( N H ~ ) ~ R U - N ~ - - O S ( N H ~ ) S ] ~
ion
' (0.I V, rotating
platinum electrode) to the ion with a fivefold positive
charge[")l. Generally, however, the oxidized complexes
undergo rapid secondary reactions with loss of the N 2 ligand.
5.2. Displacement or Exchange Reactions
Very stable N z complexes can undergo displacement reactions
in which the M-N, unit is retained. These include the diazotization of the exceptionally stable [Os(NH.3)sN2]" ion to
the [ O S ( N H & ( N ~ ) ~ ] ~ion[321.
'
(The [ O S ( N H ~ ) ~ N ~ion
]~'
is so stable that it resists several hours' boiling in conc. HCI.)
The Re--N, unit in [ReCI(N,)(PMe,Ph),] is also very stable
and is still retained on the displacement of all the phosphane
ligands" 81.
More numerous, however,are the reactions resultingin displacement of the N, ligand. Among them. the reversible exchange
reactions of the N r ligand with Hz and water [eqs. (7) and
(8)respectively] are of significance for the course of N2 assimilation. Ammonia too-the final stage of N 2 assimilation --can
displace the Nz group in dinitrogen complexes, sometimes
even reversibly:
[ R L I ~ N H . J ) - N ~ ] ' "+ N H ,
The discovery of the [ ( N H J ) ~ R U N ~ ] complex
"
nurtured the
hope that a pathway for definite reduction of the N r molecule
under mild conditions had been found, a hope that was not
at first fulfilled.
Attempts to reduce the complex-bound N2 ligand usually
results in its liberation. This reaction becomes easier with
Anquw. Chrm. internal. Edit.
1 Vol. 13 ( 1 9 7 4 ) f No.
10
+
[ C O H ( N L ) I P P ~ J+~ ]N H J
[ C O H ( N H . ~ ) I P P ~ . % IN
~ ]I
+
[ R U ( N H ~ ) ( , ]+
' ~ Nz
(19)1-"hl
(20)
The exchange reactions of the Ru system have been very
thoroughly studied, especially by thermochemical methods.
The heat of reaction A H 7 - 18.350.9kcal/mol found for eq.
(8) clearly indicates the strong tendency to form the
[Ru(NH3)5N2I2' ion. The greater stability of the analogous
CO compound is again evident from the heat of reaction['":
645
+
[ R U ( N H ~ ) S H ~ O ] ~CO
"
A H = -36.1 ? 1.6 kcalirnol
[ R u I N H ~ ) s C O ] ~ *+ H 2 0
terized, they certainly contain an N-N unit. Solvolysis of
these complexes results in evolution of N2 and in partial
reduction of the N 2 ligand to hydrazine and ammonia. Thus
Furthermore, the N2 ligand in [Ru(NH3)5Nz12@can rotate
a
red complex isolated from the (PPh3)2FeC1s/isoin the complex, the metal-bound and the terminal N atoms
C3H7MgC1/Nz/ether
system has been postulated to have the
interchanging places. This is deduced from the isomerization
structure:
of [ ( N H ~ ) S R U - ~ ~ N S ~ Nto] *[ @
(NH~)~RU-'~N='~N]~@,
which proceeds at 25°C in aqueous solution with a half-life
OEtz
OEtz
I
I
of ca. 2 hand is approximately 45 times faster than the exchange
(Ph3P)zF
I e-N- N-F e( P P h , )
reaction with H z O according to eq. (8). Thus isomerization
H
proceeds without dissociation of the N 2 Iigand; the dihapto
structure (2) of the M-Nz group in Scheme 1 is postulated
(If this structure is correct then the complex does not possess
as a transition state1"]. However, recent studies of the Ru-N
fourfold symmetry, in contrast to the Ru and Re complexes
cast doubt upon this hypothesis, since the basic
discussed in Section 4.2.) Solvolysis of this complex with
assignment of the VR"-N vibration to the Ru-N2 unit could
HCl/ether at -50°C yields 10% of hydrazine alongside
be wrong. Monohapto-dihapto bonding isomerism for the N2
decomposition products and N21421.
ligand was recently demonstrated by "N-NMR and IR specA deep blue complex of formula [{(C5H5)2Ti}2Nz], for which
troscopy in [(C5Me5)2TiNz]at -61 oC172b!
structure ( S a ) in Scheme 1 is postulated, can be isolated
+
121)
5.3. Lewis Base Reactions
As explained in Section 4.2, the terminal N atom in the
M-N-N
unit is probably negatively polarized. Mononucleat
N Zcomplexes can therefore react as Lewis bases. This leads
to symmetrical N z complexes
[ I N H ~ ) ~ R u N z ] "+ [ ( N H ~ ) S R U I H ~ O ) +
]~"
[ ( N H J ) ~ R u - N ~ - R u ( N H ~ ) ~+] H
~ ~2 0
(22)
AH = - 6.7 kcal/rnol
from the (C,H,),TiCI/CH,MgI/N, system. Solvolysis of this
complex with methanolic HCl yields mainly dinitrogen; however, hydrazine and ammonia also apbear as minor product S[39*'.
While these experiments show that it is possible in principle
to reduce complexed dinitrogen, they have the disadvantage
that with the compounds mentioned it is not possible to
convert a definite Nz complex into a definite N,H, complex
( x = l or 2, y = 1, 2, 3, or 4). This was first accomplished
by the reaction of well-defined bis-(dinitrogen) complexes of
molybdenum and tungsten with concentrated HC11741:
as well as asymmetric N Z complexes:
[ ( N H ~ ) ~ R U N ~+
] "[ I N H ~ ) J O S I H ~ ~ +
)]~"
[ I N H J ) ~ R u - N ~ - O S ( N H ~ ) ~ ]+~ "H 2 0
[(PMe2Ph)aCIReN2] + [C~CIIITHF),]
-
[(PMe,Ph),CIRe--N2---CrCIl(THF)2]
+ TH F
(23)
(24)
According to eq. (24) a series of asymmetric N2 complexes
can be prepared in which the CrCl,(THF), grouping is replaced by TiCI,(THF),, PF,, AlEt,, TaCI,, NbCI,, and many
other Lewis acids'401.
In principle, the same reactions are observed for N2-bridged
complexes as for terminal N2 complexes. As a further reaction
type dissociation can yield terminal N 2 complexes, e. g . according to eq. (25):
One N 2 ligand is eliminated, the central metal is oxidized,
and the other N z ligand is reduced to an N2H2 moiety.
The authors favor a diimine complex of the type
H\
M-N-N,
H
for the M-NzH2 group. However, IR and 'H-NMR
spectroscopic data d o not rule out a metal hydrazido structure
M=N-NH,[*~].
The formation of N-C bonds in the reaction of N 2 complexes
withe. g . acyl chlorides had been observed previously[z0b-'Odl.
A metal-hydrazido structure has been confirmed for
[W(N*H2)Cl(diphos)2]BPh4 by X-ray structure analysis. The
observed relatively short N-N distance of 1.37 suggests
bond as in
partial double bond character of the M--N-N
[W2N-.-NH2]@ 1831,
A
5.4. Reduction
As mentioned at the beginning of Section 5 , most attempted
reductions of N z complexes result in a loss of the N z unit.
However, this is not always the case: [CoN,(PEt,Ph),], for
instance, can be reduced to an anionic N, complex :
[CoNdPEt2Ph),]
+ Na/Hg
4
Na[N2Co(PE12Ph)i]
(26)
Neither loss nor reduction of the N 2 ligand occur in this
reaction.
Greater interest attaches to the reactions of some complexes
isolated as solids from nonaqueous systems of the ZieglerNatta type. Although they have not yet been definitely charac646
6. Dinitrogen Complexes and Other Potential Intermediates of N z Assimilation
Coordination of dinitrogen to a metal was suspected to be
the decisive step in N 2 assimilation long before the isolation
of active nitrogenase extracts from natural sources and the
synthesis of stable N2 complexes. Biochemical studies have
contributed greatly to our understanding of N2 assimilation,
but it has not been possible to isolate any intermediates.
Even today, all we know for certain is that the process
Angew. Chem. infernat. Edit.
/
Vol. 13 ( 1 9 7 4 )
/ No. 10
starts with atmospheric nitrogen and yields ammonia. Recent
years have seen the synthesis of model substances which enable
the chemist to develop precise and experimentally verifiable
concepts concerning the reduction of dinitrogen to ammonia.
the reduced forms of the N 2 molecule possibly existing in
deprotonated form, e. g. diimine as monoanion [:N=NHI0.
Which of the steps postulated in the scheme have so far
been verified in vitru?
It is unlikely that the N2 molecule takes up six electrons
and six protons and decomposes to ammonia in a single
step; a stepwise reduction is much more probable. Study
of the intermediates therefore acquires considerable significance because such information is indispensable for the simulation of N2 assimilation in vitru and for the development of
technical catalysts for reduction of gaseous dinitrogen, e. g.
to hydrazine under normal conditions.
1) Coordination of gaseous dinitrogen under physiological
conditions is possible [ e . g. eqs. (7)-( lo)]. In the complexes
isolated so far the dinitrogen is invariably bound in the end-on
and never in the side-on mode.
2) Displacement of the end-product NH3 by Nz can be reproduced e. g. according to eq. (1 9).
A series of mechanisms has been proposed in recent years
for the reduction of dinitrogen with nitrogenase~['~!In accordance with biochemical data, the most likely alternatives postulate two-electron reduction steps and unaminously assume
that the dinitrogen is reduced to ammonia via diimine and
hydrazine (Scheme 2)
4) In Scheme 2, diimine complexes are postulated as intermediate stages. Until recently diimine was only known as an
unstable intermediate, e. g. in organic hydrogenations with
hydrazine, or as the decomposition product of HN3, N 2 H 4
or alkali metal tosylhydrazides at low temperatures (ca.
- 196°C)~77-80!Lately, however, it has been stabilized under
normal
conditions
by
complexation
in
[transN ~ H z { C ~ H ~ M ~ ( Cand
O ) [cis-NzH~{Cr(C0)~}~].
~}~]
1
H\
L .M;N-N;(
- Nli
-+
H Ic
L,M-NH~ +
3) Hydrazine complexes are well known. In addition, hydrazine is formed in the reduction of gaseous nitrogen with V2'
salts in alkaline solution[76' and by the hydrolysis of the
Fe-N2 and Ti-Nz complexes mentioned in Section 5.4.
The most conclusive argument for hydrazine as an intermediate, however, is probably the synthesis of the
[MCI z(N2Hz)(dipho~)2]
(M = Mo, W) complexes according
to eq. (27).
/
These complexes are formed by the oxidation of hydrazine
complexes'81~821:
H
M'L,)
H
H~N-(M'L,)
4
NH,
Scheme 2. Intermediates of N2 reduction, greatly simplified
According to this scheme, the first step in assimilation involves
coordination of dinitrogen to one or two metal centers (the
actual number has not yet been established). They may be
the same or different (e. g. M = M' = M o or M = Mo, M'= Fe)
and in the resting phase, z.e. in the absence of dinitrogen,
they probably exist as aquo complexes. Presumably, the
oxidation state of the metal centers changes during reduction
of the dinitrogen, e. g.
L,M"
+
L , M ' o r L,M"
The trans and cis structures of the diimine in the manganese
and chromium complexes respectively are readily deduced
from their IR spectra. The IR spectrum of [cisN2H2{Cr(C0)5)2 1 is particularly clear. Since the typical
absorptions of the Cr(C0)s groups are confined to the 21CK1800 and 800-350cm- ' regions, bands outside these regions
can be safely assigned to the N2H2 ligand vibrations. From
group theory considerations, three IR-active vibrations
v,?(NH), G,,(HNNH), and r(HNNH) are anticipated for trunsN,H,, but five IR-active vibrations for.cis-diimine [v,, and
v,(NH), 6,- and F,(HNNH), and v(N=N)]
40r'-7qJ
20
0
4000
3500
3000
limabsi
2500
2000 2000
-
1800
1600
1400
1200
1000
800
600
600
400
0
LI [cm-11
Figure 6. IR spectra (KBr) of [NLHI{CT(CO),)~],_..[ N 2 D z ( ( C r ( C 0 ) ~ ) , ] .and ["N2H2{Cr(COs)2] (only those regions that
change compared with the spectrum of the N,H, complex).
Angew. Chem. infernat. Edit. / Val. 13 ( 1 9 7 4 )
N o . I0
647
A glance at the IR spectrum of [N2H2{Cr(C0)5)2](Fig. 6)
immediately shows that in the regions involved (4000-2100
and 1800--800cm-') there are five absorptions, which in
itself points to a cis structure for N ~ H zDeuterium
.
and "Nlabeling confirm that the absorptions are in fact due to N z H 2
vibrations and allow assignment of the absorption to vibrational forms of the diirnine. Thus, it is seen that e.y. the
absorptionat 1415cm-' canonly bedue to the N-N stretching vibration, which, however, might be coupled to the Cr-N
stretching vibrations[841.
What is extremely surprising in the synthesis of the diimine
complexes is the fact that the tungsten-diimine complex also
occurs in the cis form, although it is best prepared from
the mononuclear hydrazine
[lCO),WN,H,]
+ H,OL
Cuzo THF
Na,SO, O'C
141 a ) International Union of Pure a n d Applied Chemistry: Nomenclature
of Inorganic Chemistry, Butterworths, London 1970: b) K . Jorius, Angew.
Chem. 85. 1051 (1973): Angew. ('hem. internat. Edlt. 12. 997 (1973): C.
Ki-iigcrand Y - H f i u I , Angew. Chem. 8 5 . 105 I (1973):Angew. Chem. internat.
Edit 12, 998 (1973)
[5] G. C. Dohiriaon. R. Muson. C. B. Roherrsoii, R. L:go. F . Corifi, D. Monrlii,
S. Ciwini. a n d F . Bor~uti,Chem. Commun. lY67. 739
161 M.: H . Knoth. J. Amer. Chem. SOC. 88, 935 (1966).
[7] J . K . Biirdert, M . A. Gruhuni. and J. J . 7iirncr. J. C. S Dalton Trans.
1972. 1620.
[8] H. Huhrr. E. P. Kurirlig, M. Moskorits, a n d G. A. O m . J . Amer Chem.
Soc. Y.5. 332 ( 1 973).
191 A. J . Ri,.\i, J. Organometal. Chem. 40. C 7 6 (1972).
[ 101 D. .Si,llmurin and C. Mor.se/, Z. Naturforsch. 27 h, 465 ( 1972).
[I I ] D. S~/J?ILI~IJI
and G Alui.srl. Z. Naturforsch. 27h. 718 (1972).
[I21 M . Hidui, K . Timiiiuri, Y C'ihirlu, a n d A . Miwrio, Chem. Commun.
1969. 1392: 7 A. Giwrgi, a n d C. D. Srrhold, Inorg. Chem. 13. 2544, 2548
(1973).
[c~s-NLH, /W(CO),/.,]
(30)
+by-products
This could signify that in the complex cis-diimine is thermodynamically more stable than trans-diimine. A decision as
to which of the two isomers is more stable in the free state
has so far been impossible, both experimentally and by calculation[8h].The preferential formation of the cis-diimine complexes in the case of chromium and tungsten gains importance
by the fact that reduction of substrates by nitrogenases most
probably also proceeds stereospecifically in the cis position:
thus, the reduction ofC,D, yields exclusively cis-deuterioethylene. As, furthermore, the diimine complexes undergo redox
reactions -under mild conditions in which N2, N2H4, and
NH3complexes are formed, they are suitable models for studying the intermediate stages in Nr assimilation; just as the
likely synchronous abstraction of two H atoms converts the
binuclear chromium-hydrazine complex into the diimine
complex, so the synchronous addition of two hydrogen atoms
to a binuclear dinitrogen complex could reduce the latter
to a cis-diimine complex.
7. Outlook for the Future
The first seven years of N z complex chemistry were marked
by the search for new N z complexes and the failure to reduce
the dinitrogen bound in them. Very recently a new development has begun to emerge. Now that it has been demonstrated
experimentally that complex-bound dinitrogen can be reduced
under relatively mild conditions and all the reduction states
of the N z molecule have been identified, the prospects of
finding a catalyst system allowing in vitro reproduction of
the processes occurring In N 2 assimilation under mild conditions appear brighter than ever before.
[I,]
M . Arc,stu and A. S u ( c o , Garz. Chim. Ital. 102. 755 (1972)-
[I41 M . L. H. G r ( w a n d W E . S i l w r t h o r n , J. C. S. Dalton Trans. lY73,
30 I.
[ 151 B. Bcll. J C1iuii. a n d G. J Lriglr, Chem. Cornmun. iY70, 842.
[I61 D. Si4lrnuriii, Angew. Chem. 83. 1017 (1971): Angcw. Chem. internat.
Edit. 10. 919 (1971)
[I71 D. Sdlmorin. J. Organometal. Chem. 36, C27 (1972).
[I81 J. Cllrrft.J . R. Dihwrth, a n d G J. Leigh. Chcm. Commun. 1Y69, 687:
cf. also M . E. EOlj a n d A . P Gin.\h(n~, J. Amer. Chem. SOC. Y5, 2042
( 1973)
[I91 J . Cliutt, J. R. Diiworth, a n d G J . Leigh. J. Orgmometal. Chem. 21,
P49 (1970)
[20] a ) J. Chutt. J. R. Diiiwrrh. H . P. Gitnz, G. J. Li,igh, a n d J . R. Sunrlvrs.
Chem. C o m m u n . lY70. 90; b ) J . Cllurt, J . R. Dilrvorrh, a n d G. J Leigh.
J . C . S. Dalton T r a n s 1973, 612: c ) J. Chutr, G. A. Hrufh, a n d G. J . Lrigli,
J . C. S. Chem. Comm. 3972. 444; d j J. Organometal. Chem. 57, C 6 7 (1973).
[21] A Suiio a n d M .
Arcsro,
Chem. C o m m u n . lY(i8, 1223
[22] G. M B u i r u ~ ) f trM.
. J. Muj.a. a n d B E. Prutrr. Cheni. C o m m u n . IY6Y.
5x9.
[23] W H. Kriorh, J Amer. Chem. Soc. YO, 7172 ( 1968).
[24] P. G. Doiiglri.~,R D. Fdt/iurti. a n d H. G. Mrrzger. Chem. C o m m u n .
lY70. 889.
[25] A. D. Allcri. F . Bortoriilc!, R 0. Hurri\,
S r n o f f , J. Amcr. Chcm. Soc. KY. 5595 (1967).
V P Reimulu, a n d C. 1.:
[26] L. A . P. Kurie-Mugiiire, P. F . Shrriiluri. F. Busolo, and R G. Prurson,
J. Amer. Chem. SOC.YO, 5295 (1968).
[27] Y. G. Borodko, A. K . Shiioiw. and A. E . Shilor, Dokl. Akad Nauk
SSSR 176, 1297 (1967).
[2X] B. 5~11,J . Churt, a n d C. J . Leigh, Chem Commun. lY70, 576.
[29] J . Cliufr. G. J. Leigh, and R. L. Richurds, J. Chem. Soc. A 1970, 2243.
[30] J . Chorr. D. P Melri1le. a n d R. L. Richariis, J Chem. SOC. A 1Y71,
895
[31] A D. Allrri a n d J. R. Srrrrris, Chcm. C o m m u n . 1967. 1147.
[32] H. A. Schrir/cygi,r, J . N . A r m o r , and 11. Tiiiihe, J Amer. Chem. SOC.
YO. 3263 ( 1968).
[33] G. Sp&r a n d /.. M u r k o , lnorg Chim. Acta 3. I26 (1969).
[34] a ) A. Suwo a n d M. R&sr, Chem. Commun lY6Y. 471 : b) A. Yumurnoro.
Kuu:iiiiii,, L. S f i r . a n d S. Ikedu. ihrd lY67, 73.
S.
[2] A. D. .4//rri and C. V Si~iio//, Chem. C o m m u n . 1Y65. 621.
[35] M . Arrstrr, C. F . Nohili,, M . R o . w a n d A. Sulco, Chem. Cornmun.
IY71. 781.
[36] L. Y Ukhiri. A . Y Shi~rrsor,a n d M. L. Khirlrkr1, Isv. Akad Nauk
SSSR, Ser. Khim lY67. 957
[37] J . 1'. Co//miri, M . Kiiboru, F . D Vustine, J . Y Sun, a n d J. CV K u n g .
J. Amer. Chem. SOC.YO. 5430 (1968).
[38] S. C. Srirusrui~ua n d M. Bigorgne, J. Organometal. Chem. I # , P30
1 1969)
[39] a ) I'u G. Borodko, / N . IrleIu. L M. Kuchupiria, S. 1. S u h z h o . A.
K Shiloca, a n d A E Shilov, J. C . S. Chem. Comm. IY72. I17X: b) J. E.
Bercuw. R. H Murrii-/I, L. C. Brll, a n d H. H . Bririt:ingrr, J. Amer. Chem
SOC. 94, 1219 (1972): c) E . €. IWI Turnelm,W Crcrnei.. N . Klorntschi, a n d
J . S. Miller, J. C . S. Chem. Comm. IY72, 481; d j cf. also J . H . Ewhrn,
J Organometal. Chem. 57, 159 (1973).
[40] J . Churi. J. R. Dilnorth, G. J . Leigh. and R. L. Richurds, Chcm. Cornmun.
IY70, 955: cf. also J Churr, R. 11. Cruhrrw, E . A. J r f f r r i . a n d R. L. Rii.hurds.
J. C. S. Dalton Trans. 1973, 1167.
[3] H . Borreis. Arch. Mikrobiol. I . 333 (1930).
[41] W. E. Siicrrrhorn. Chem. C o m m u n . IY71. 1310
Received. May 23. 1973
Supplemented: December 14, 1973, March 18, 1974, a n d June 19. 1974
[A 1000 I€]
G e r m a n version: Angew. Chem. 56.692 (1974)
Translated by Express Translation Service, London
[ I ] Selection of earlier reviews: a ) A. D. A l l m . R. 0. Harris, B. R. Lor.s<hrr.
J . R Ster-cri.\, a n d R. N . W / l i t r / < , j ~Chem.
,
Rev. 73, I I (1973): b) J. Churr
a n d G J. LcrgIi, Chem Soc. Rev. 1 , 121 (1972). c) J . Chutr a n d R. L.
Richards in J . R. Postgurr- T h e Chemistry and Biochemistry of Nitrogen
Fixation. Plenum Press. London 1971, p 57: d j G. Henrici-Olirc' a n d S
O/rw< Angew. Chem. # I . 679 { 1969): Angew. Chem. internat. Edit. X, 650
(1969): e ) Yu C; Borodko a n d A . E. Shrlor, Russ. Chem. Rev. 38, 355 (19691.
648
Angew. Chem. infernaf. Edit.
/ Vol. 13 (1974 j 1 No. 10
[42] Yu G. Borodko. M . 0. BroUman, L . M . Kachaplna, A . E . Shilot, and
L . Yu. Ukhin, Chem. Commun. 1971, 1185.
[43] 0 . F. Ifarrison, E. Weisshrryrr, and H . Taube, Science 159, 320 (1968).
[44] C. Creutz and H . Tauhe, lnorg. Chem. I @ , 2664 (1971).
[45] A . D. Allen, Section Lecture, XIII’h Intern. Con[. Coord. Chem., Zakopane 1970.
1461 R. M . Magnusson and H . Taube, J. Amer. Chem. Soc. 94, 7213 (1972).
[47] P. W Jolly and K . Jonas, J. Organometal. Chem. 33, 109 (1971).
[48] Tables of Interatomic Distances etc., Special Publications No. I 1 and
No. 18 of the Chemical Society, London 1958, 1965.
1491 F. Borfomley and S. C. Nyhurg. Acta Crystallogr. B 24, 1289 (1968).
[50] J . E. Fwgu.\.son, J L. L o w , and W T Robinson, Inorg. Chem. 1 1 ,
I662 ( I 972).
[51] B. R. D a i s and J . A . Ibers, Inorg. Chem. 9, 2768 (1970)
[52] B R. Daris and J . A Ibers, Inorg. Chem. 10, 578 (1971).
1531 B . R. Ducis, N. C. Poyne, and J . A . Ibers, Inorg. Chem. 8, 2719 (1969).
[54] T Cichida, E Cchida, M . Hidai, and T Kodomo, Bull. Chem Soc.
Jap. 44. 2883 (1971).
[55] J . M . Treitel, M . T Flood, R . E . March, and H . B. Gr aj , J. Amer.
Chem. SOC.91. 6512 11969).
[56] R. Nast and H. Kohl, Chem Ber. 97,207 (1964).
[ 5 7 ] C . Hrrzherg Molecular Spectra and Molecular Structure 1. 2nd Edit.
van Nostrand, Inc.. Princeton 1955. p 551.
[58] M . A . I . A/-Johourr, 0. P. May, and D . W Xirner, J. Chem. SOC.
1965, 616: D W Turner and D . P . M a y , J . Chem. Phys. 45, 471 (1966).
[59] R. S. M u l l i k m , Can. J. Chem. 36, 10 (1958).
[60] C. Edmisron and K . Ruedmhrrg. J. Chem. Phys. 43, S97 (1965).
[61] K . G. Cauitoti, R. L. D e Kock, and R . f Fenske, J Amer. Chem
Soc. 92, 5 I 5 ( 1970).
[62] G. J . Lpigh, J . N . Murrrl, W Brrmser,and W G. Procror, Chem. Commun.
1970. 1661.
[63] P. Finn and W L. Jolly, Inorg. Chem. 1 1 , 1434 (1972).
[64] J . Chart, D. P. Melrilli,, and R. L . Richards, J Chem. Soc. A 1969.
2841.
[65] A . D. Allen, 1 Eliades. R. 0. Harris. and !4 P. Rrinsalir, Can. J. Chem.
4 7 . 1605 11969).
[66] D. J . Darensboirr.g, Inorg. Chem. 10. 2399 (1971).
1671 a ) J . Charr. R . C. Fay, and R. L. Ridiard.\. J. Chcm. Soc. 4 1Y71.
702, b) M . M e r c e r . R . H . Crabtree. and R L . Rirliards. J. C. S. Chem
IY73.XO8.
[6R] D. E. H o r r i s o ~ iand H . Tairhe. J Amer. Chem. SOC. XY. 5706 (1967).
[69] J Chart, C. J . Lrigh, and R . L. Ricliurd\. J. Chem. Soc. A IY70, 2243.
[70] a ) C. M . Elson, J . Gulms. and J . A . Page. Can. J . Chem. 4Y. 207
(1971); b) A . Yamumoto, S. Kota:irrrie, L. S. Pu. and S. lkcdu. J. Amer.
Chem. Soc. 89, 3071 (1967).
[71] J N. Armor and H . Tairhc,, J. Amer. Chem. SOC.YZ, 2560 (1970).
1721 a ) M . W Bee. S. F. A . K e r r k , and D. B Poiwii. J. C. S. Chem. Comm.
1972. 767; b) J . E . Bn.cmt. E . Rosenhrrg, and J . D. Roberts. J. Amcr. Chcm
SOC.96, 612 (1974).
[73] G. D. Wart, J. Amer. Chem. Soc. 94, 7351 (1972).
[74] J . Chart, G . 4 . H r o i h , and R. L. Riiliurds. J. C. S . (‘hem. Comm.
1972, 1010.
[75] Cf. [ic], pp. 89, 90, and 101 as well as [ I b].
[76] A . Shiloc, N . Denisor. 0. Eliirior, N. S l i i r r u l o ~ .N. Shiri u l o i ~ i .and A.
Shilow, Nature 231, 460 (1971): cf. also G . N. S(.hruir:er, P. A. Doviriwij~.
R. H . Frazier, and G . W K i
, J . Amer. Chem. Soc. 94, 7378 (1972).
[77] S N . Forirr and R. L. Hud.>on. J. Chem. Phys. ZX, 719 (1958).
[78] K. Rosengren and G. C Pinienfel. J . Chem. Phys. 43. 507 (1965).
[79] A . Fomberri. Can. J. Phys. 46, 1005 (1968).
[80] N . Wihrrg, H . Bat.hhirhrr, and G. Fis<bcr, Angew. Chem. 84, 8x9 (1972):
Angew. Chem. internat. Edit. I I , 829 (1972).
[ X I ] D. Sellnmtln, J. Organometal. Chem. 44, C 4 6 (1972).
Et&I/. J . Organomctal. Chcni. 4Y. C 22
[82] D. Sellmarin, A. Brand!, and R .
( 1973).
[83] G. A. H ~ a f h R.
, M u s o ~ i ,and K . M. Thoniir, J. Amer Chcm Soc
96, 259 ( I 974).
[84] D. Sellmann, A . Brundl, and R . Endell, Angew. Chem. 85. 1122 (1973):
Angew. c h e m . internat. Edit. 12, 1019 (1973).
I851 D. Sellniann, A . Brondl, and R. Endell. Angew Chem. 85, 1121 (1973).
Angew. Chem internat. Edit. 12, 1019 (1973)
[86] Cf. J . A l s f r r and L . A . B i m e i i r . J Amcr. Chcm. Soc. 84. 1261 (1967).
L. J . Schoad and H . B. Kiiiso., J. Phys. Chem. 73. 1901 (1969): L. R~~duri.
W J . Hehre, and J A . Pople. J. Amer. Chem. Soc. 93. 2x9 (1971)
The Helicenes
By Richard Henri Martin[*]
“Helicene” is the name introduce1 by Newman in 1955, to describe the benzologues of phenanthrene in which the extra ortho-condensed rings give rise to a (regular) cylindrical helix. The
pioneer work of Newman in this field cannot be overemphasized; his brillant synthesis and
resolution of [6]helicene, achieved eighteen years ago, will remain as a landmark, for it opened
the way to the study of a fascinating class of synthetic molecules. In the following review,
an attempt is made to summarize the present state of our knowledge in this rapidly expanding
field.
1. Introduction
compounds is due to the unique combination of these three
properties in a single molecule.
The helicenes are characterized by a helical structure made
up of ortho-condensed aromatic rings, by the presence of
a powerful inherently chiral chromophore, and by the possibility of interactions ( e . g . electronic interactions) between overlapping aromatic rings. The scientific interest raised by these
We shall deal first with the nomenclature of these compounds.
The “class name” is “helicene”; “carbohelicenes” contain only
carbon atoms in the skeleton, “heterohelicenes” at least one
hetero-aromatic ring. In this review helicenes that are made
up of benzene rings only are referred to as “all-benzenehelicenes”[**].“Double helicenes” are helicenes containing
[‘I
Editorial nore: The expression “benro-heliccnes“ is unsuitablc. sincc thc
prefix “benzo“ denotes, according to I U P A C rule A-21 4. the annelation of ;I
benzene ring The expressions “benzoid helicenes” or “henzenoid heliccncs”
could be misunderstood for the terms “benzoid“ and “benzenoid” are ~ i s u a l l ~
used in the sense of “benrene-like“ andlor “aromatic”.
[**I
Prof. R H. Martin D. Phil.
Service d e Chimie Organique. Fac. Sc.
Uriiversite Libre de Bruxelles
Av. F. D. Roosevelt, 50
8-1050 Briixelles (Belgium)
Angew. Chem. internat. Edit.
Vol. 13 (l974)
1 No.
10
649
Документ
Категория
Без категории
Просмотров
0
Размер файла
1 094 Кб
Теги
synthesis, properties, metali, dinitrogen, transitional, complexes, significance
1/--страниц
Пожаловаться на содержимое документа