close

Вход

Забыли?

вход по аккаунту

?

Spectroscopic Properties of Organic Azo Compounds.

код для вставкиСкачать
[I097 W B. Hammond and N . J. flrrro, J. Amer. Chem. SOC.88, 7880
(1966).
[ I 321 Ch. Riickardt, E. Merz, B. Freudenberg, H . J. Opgenorth, C. C. Tau,
F. Walls, J . Padilla, P. Joseph-Nathan, F. G i r d , and J . Romo,
Tetrahedron Lett. 1965, 1577.
[Ill]
H . U . Wagner and R . Gompper, Tetrahedron Lett. 1970, 2819;
1971,4061,4065.
[I121 H . 0. Huismnn, Angew. Chem. 82, 511 (1971); Angew. Chem.
internat. Edit. 10, 450 (1971).
[I131 W H . Urri, P . Szeczi, C. Ikoku, and D. W Moore, J. Amer. Chem.
SOC.86, 2224 (1964).
[ I 3 3 1 R. R. Schmidt, W Schneider, J . K a r g , and U . Burkert, Chem.
Ber. 105. 1634 (1972).
11341 R . C. Petterson, private communication; see also Ref. [135].
[I351 R . C. Pefterson, T G. Poendle, G. W Lin, and J . P . Mykytka,
J. Org. Chem., in press.
[I361 R . R . Schmidt and H . Vatter, Tetrahedron Lett. 1971,1925.
[I371 Th. Kauffmann, H . Berg, and E. Koppelmann, Angew. Chem. 82,
396 (1970); Angew. Chem. internat. Edit. 9, 380 (1970).
11381 Th. Kaufmann, H . Berg, E. Ludorff; and A . Woltermann, Angew.
Chem. 82,986 (1970); Angew. Chem. internat. Edit. 9,960 (1970).
[I391 Th. K a u f m n n , D. Berger, B. Scheerer, and A: Woltermann,
Angew. Chem. 82, 986 (1970); Angew. Chem. internat. Edit. 9, 960
(1970).
[I401 Th. Kauffmann and R. Eidenschink, Angew. Chem. 83,794(1971);
Angew. Chem. internat. Edit. 19,739 (1971).
[I411 D. Martin, Z. Chem. 7,123 (1967).
[I421 G. Wittig and E. Knauss, Chem. Ber. 91, 895 (1958).
[I431 C. F. Huebner and E. M . Donoghue, J. Org. Chem. 33,1678 (1968).
[I441 W T Ford, R . Radue, and J . A . Walker, Chem. Commun. 1970,
[IIO]
[I141 S . Hiinig, Helv. Chim. Acta 54,1721 (1971).
[I151 L. A . Paquette, Chem. Commun. 1971,1076, and literature cited
therein.
11161 D. E . Heitmeier, J . 7: Hortensrine, j r . , and A . P . Gray, J. Org.
Chem. 36,1449 (1971).
[I171 K . Griesbaum, W Naegele, and G . G. Wanless, J. Amer. Chem.
SOC.87, 3151 (1965).
[I181 R . Gompper and G. Seybold, Angew. Chem. 80, 804 (1968);
Angew. Chem. internat. Edit. 7, 824 (1968).
[I191 A . E. Lodder, H . M . Buck, and L. J . Oosterhoff; Rev. Trav. Chim.
Pays-Bas 89,1229 (1970).
[I201 J . March: Advanced Organic Chemistry: Reactions, Mechanisms, and Structure. McGraw-Hill, New York 1968, p. 563.
11211 G. A . Olah, J . M . Bollinger, and J . Brinick, J. Amer. Chem. SOC.
90, 6988 (1968).
[I221 G. Heublein, Z. Chem. 9, 281 (1969).
[123] G. M . Beverly and D. R . Hogg, J. Chem. SOC.B 1971,175.
[124] H . C. Clark and L. E. Mauzer, Chem. Commun. 1971, 387
[I251 G. A. Olah, J. Amer. Chem. SOC. 94,808 (1972).
[126] R . G. Pearson, Chem. Eng. News 48, Nr. 41, p. 66 (1970).
[127] D. F. De Tar, Org. Reactions 9,409 (1957).
[I281 W D.Zahler and R . Huisgen, Chem. Ber. 96, 765 (1963).
[I291 For other hypotheses see Ref. [133].
[130] R. R. Schmidt and W Schneider, Lecture at the Chemiedozententagung, Hamburg, March 1971.
[I311 R. R . Schmidt and ct! Schneider, Tetrahedron Lett. 1970, 5095.
and R . Werner, Chem. SOC.Spec. Publ. 1970 (24), 51.
966.
11451 W T Ford, J. Org. Chem. 36,3979 (1971).
[146] ?: L. Cairns,J . C . Sauer, and W K . Wilkinson, J. Amer. Chem.
SOC.74, 3989 (1952).
[147] R . R . Schmidt, Angew. Chem. 83, 622 (1971); Angew. Chem.
internat. Edit. 10, 572 (1971).
11481 U.M . Kempe, T K . Das Gupta, K . Blatt, P . Gygax, 5. Felix, and
A . Esckenmoser, Helv. chim. Acta 55, 2187 (1972).
[I491 J . Marchand-Brynaert and L. Ghosez, J. Amer. Chem. SOC.94,
2870 (1972).
[I501 P . Eberhardand R . Huisgen, J.Amer. Chem. Soc.94,1345(1972);
R . Huisgen and P. Eberhard, ibid. 94, 1346 (1972).
[15l] U . Sckollkopf; Angew. Chem. 82, 795 (1970); Angew. Chem.
internat. Edit. 9, 763 (1970).
[152] E. J . Corey and S . W Walinsky, J. Amer. Chem. SOC.94, 8932
(1972).
Spectroscopic Properties of Organic Azo Compounds
By Hermann Rau[*]
Azo compounds are characterized by a low-energy ‘(n,a*) electronic state, which gives
rise to a weak band at long wavelengths in the absorption spectrum. Coupling of this
state with other low-lying molecular states is responsible for the high intensity of the
“azo band” compared to other forbidden transitions, for its lack of vibrational structure,
and for the inability of nearly all organic azo compounds to fluoresce.
1. Introduction
Although the azo group -N=N- has given its name
to a large class of organic dyes, spectroscopists have shown
comparatively little interest in this atomic grouping. The
reason may lie in the effects of two rules characterizing
the spectroscopic behavior of the azo compounds:
1. the azo group has a “structure-destroying effect”[’];
2. azo compounds d o not emit.
[*I
Doz. Dr. H. Rau
Abteilung fur Physikalische Chemie der Universitat
Hohenheirn (LH)
7 Stuttgart 70, Postfach 106 (Germany)
224
The information that can be gleaned from absorption and
emission spectra is therefore very limited.
By searching for exceptions to these rules, however, it
is possible to clarify the spectroscopic properties in molecular terms. This involves the extensive use of theoretical
considerations, and because of the shortage of experimental
data the conclusions are necessarily somewhat speculative.
Only one field of the spectroscopy of azo compounds
has been investigated in depth, i. e. their photochemistry,
in particular the cis-trans isomerization of aromatic azo
compounds and the photodecomposition of aliphatic azo
compounds. These two processes will be considered in
Angew Chem. internat. Edit.
/ Vol. 12 (1973)
No. 3
the following discussion only insofar as they are relevant
to the coupling of the lowest excited states.
This leads to an energy level diagram as shown in Figure
1bin which electronic states are combined with vibrational
states vo,vl,v2..., the individual properties of the states
being retained.
2. Fundamentals
2.1.3. Nonstationary States
2.1. Molecular States
2.1.1. Pure Stationary States
It is known from spectroscopic experiments that a molecular system can absorb or emit radiation only in characteristic quanta hv, and that in so doing it passes from one
energy state into another (Fig. la). The spin state must
remain the same in this process (So+S,), any change in
multiplicity (So-Tl) being forbidden.
The independence of the coordinates is keally only an
approximation, which is satisfied to a greater or lesser
degree by any real system. Electronic and vibrational states
may be coupled among themselves and with one another.
Since the molecular states have the properties of oscillators,
interaction between two states leads to two new states,
one of higher energy and another of lower energy (Fig.
1 c). The splitting A E depends on the extent of coupling.
I
a1
V
C1
E S dl
-
,
,
Y2
I--
Iv, ---'
w3=q w2
+
0
Fig. 1. a) Stationary states; b) Born-Oppenheimer states; c) coupled classical oscillators; d) coupled molecular states.
These states are characterized in that the molecular system
does not spend only a short time in them. They are described by standing matter waves, the wave function containing all the information about the state. The spectroscopist is interested in energy, symmetry, and lifetime. A state
in which the molecule remains is a pure stationary state.
2.1.2. Quasistationary States
States observed by spectroscopic means cannot be pure
stationary states, for spectroscopy registers only transitions. However, the states that can be detected spectroscopically are normally so long-lived that they may be
regarded as quasistationary and described by a single wave
function.
The Born-Oppenheimer (BO) functions are particularly
suitable for describing such states. In the BO approximation the internal coordinates of the molecule are resolved
into three independent sets corresponding to electronic
motion, vibration, and rotation. For molecules in solution,
the rotational structure can be observed only in very rare
cases; we shall therefore disregard rotation. The complete
function of the vibronic state can then be represented
in the BO approximation as the product of the electronic
and vibrational functions.
In wave-mechanical formalism, coupled states are described by mixing of wave functions (Fig. 1d):
The mixing coefficient h is the ratio of the coupling energy
between the states JI1 and $ 2 under the interaction V
to the energy difference E , - E , . The BO approximation is
valid as long as k < 1.
Another property of coupled states can be illustrated by
the behavior of two identical coupled pendulums. If one
pendulum is set in motion, the other soon starts to swing,
and takes over the energy of motion to such a degree
that the first pendulum comes to rest, whereupon the
process starts anew with the roles reversed. A molecular
system moves to and fro between coupled states, so that
the individual state is no longer stationary.
The time tprequired by the system to return to a certain
state depends on the intensity of coupling and on the
complexity of the system. The weaker the coupling and
the greater the number of coupled states, the longer the
system will take to return to its original state, i.e. the
greater is tpl21.
2.2. The Azo Group
where Y is the set of electronic coordinates and q the
set of vibrational coordinates.
Angew. Chem. internat. Edit.
1 Vol. 12 ( 1 9 7 3 ) No. 3
The azo group consists of two nitrogen atoms linked by
a double bond, each of these atoms carrying a non-bonding
225
pair of electrons. The chemical and photochemical stability
clooctene ( 4 ) , pyrazolines such as ( 5 ) , ( 6 ) , and ( 7 ) , or
o-diazines such as pyridazine (8), phthalazine ( 9 ) , or benof azo compounds increases with increasing resonance
and inductive interaction with the rest of the r n o l e c ~ l e ~ ~ ~ .zo[c]cinnoline (10).
Azo compounds are characterized spectroscopically by
These compounds naturally cannot isomerize. Only a few
a low-energy '(n,n*) state, which is manifested as a weak
molecules are known which contain a trans azo group
absorption band at long wavelengths (e.g . Fig. 2).
1
20m
lDOO0
in a ring system (e.g. (11)[13] and (I.2)['"]), and only
few spectroscopic data are presently available for the cyclic
trans azo compounds.
LO lo3
20
Fig. 2. Absorption of trans-azobenzene in methylcyclohexane/isopentane
(3: I):-atroorn
temperature,-----at
77°K; -.-.-in polyethylene
film at room temperature.
The bond lengths in the azo group are approximately
equal in aromatic and in aliphatic azo compounds. The
N=N distance is 1.23 in azobenzene (I)'"] and in azomethane ( 2 ) I 5 ] , and the length of the N-C bond is 1.41
in trans-azobenzene[6], 1.46 in ~is-azobenzene[~],
and
1.47.k in az~methane[~]
(this is practically the length of
a resonance-free C-N bondl8I).
A
A
A
3
\
H
N
I1
N
H
f 13)
(3)
(4)
The bond angles at the nitrogen atom are close to 120"[5.61.
This geometry results in cis-trans isomerism (cf. ( I ) and
(2) with (3) and (4)). The trans form is always the more
stable of the two isomers. In the case of azobenzene, the
energy differenceis 12 kcal/m01[~]and the activation energy
for the cis-trans transition is 23 kcal/mol in solution['01
and 31 kcal/mol in the solid["l, so that the two isomers
can be separated at room temperature. For azomethane,
this is possible only at low temperatures"*1. Aromatic
cis azo compounds cannot be planar because of the steric
hindrance of the o-hydrogen atoms. cis-Azobenzene has
a planar azo group, but the phenyl rings are parallel to
each other and are twisted by 56" out of the plane of
the azo group[71.
The azo group in the cis configuration can also be part
of an aliphatic or aromatic ring system, as in 2,3-diazabicyclo[2.2.1]-2-heptene ( 3 ) , the corresponding diazabicy226
An important factor in the spectroscopy of cis and trans
azo compounds is that they belong to different symmetry
groups; the cis compounds have CZYsymmetry and the
trans compounds CZh.For aliphatic azo compounds, one
need only consider the local symmetryfIs1(e.g. for ( 3 ) ) .
3. Spectral Characteristics of Azo Compounds
3.1. Position of the n + z * Band in the Absorption Spectrum
The position of the n+n* band depends on the groups
attached to either end of the azo group[3'. It appears
at about 28000cm-' for aliphatic azo compounds (Table
1)and at 22000 to 23000cm- for the aromatic representatives. Surprisingly, the position of the n - m * band is practically unaffected by substitution in aromatic azo compounds, whereas the n+a* bands in the spectrum are
shifted considerably (Table 2). This is particularly noticeable for the trans diarylazo compounds in Table 3.
Angew. Chem. internat. Edit. J Vol. I 2 ( 1 9 7 3 ) J
NO.3
Table 1. n +x* bands and fluorescence yields of aliphatic azo compounds
..,C
Compound
E,,,,,~
Structure
Yield
,
28 000
28 000
13
17
none
none
0
trans-(CH,),CN=N F HO C H 3
'CH(CH,),
28 OOO
19
none
0
(11)
28 000
28400
27 OOO
29 500
31 000
28 OOO
30 500
30 500
25 900
60
240
150
400
280
250
77°K: weak
none
sharp
sharp
weak
none
weak
0
trans-CH,N=NCH
trans-(CH ,),CHN=NCH(CH
3)2
cis-CH,N=NCH,
(4)
(3)
(5)
(6)
(7)
3,5-Diphenylpyrazoline
(13)
0.55
0.1 5
0.01
0
z 0.01
290
770
3.3. Vibrational Structure of the Electronic Bands
Table 2. Spectra of substituted trans-azobenzenes p-R-C6HdN=NC6Hs.
long-wave x +x*-band
log E,,,
[cm- '3
Vmax
H
N(CH,),I
CH3
OCH,
NO2
N(CH&
N=NC6Hs
22300
22500
22200
22500
22000
21500 [a]
22000
2.61
2.71
2.83
2.8
2.83
31600
31600
30000
29000
30000
24500
27900
-
3.4
Ref.
[ XI
[cm-'1
3.3.1. Vibrational Structure of the n+n* Band
Ref.
Vibrational structures of electronic bands are normally
most pronounced in gas spectra, and the n-m* bands
usually have a more sharply defined structure than the
n+n*
However, the n+n* band of gaseous
azobenzenerZg1,
and also the long-wave bands of the azoalkanes, are without ~ t r u c t u r e r ' ~ .on
~ ~the
~ ; other hand,
the structures of the n+n* band are extremely sharp in
(3) and (4jf201(Fig. 3a).
P I
4.35
4.3
4.37
4.31
4.35
4.36
4.6
~ 3 1
~ 4 1
r251
~ 7 1
1251
C261
[a] Calculated.
Table 3. Spectra of trans-diarylazo compounds, after [27]
Compound
Ph-N=N-Ph
Ph-N=N-1-Napht
Ph-N=N-2-Napht
1-Napht-N=N-1-Napht
l-Napht-N=N-2-Napht
2-Napht-N=N-2-Napht
n
+
22500
22100
22400
21 600
22000
22 500
I st x
x*-band
2.65
2.95
2.96
3.17
3.18
3.18
28500
26000
26000
23500
25000
25500
4
x*-band
31 500
27000
27000[a]
25000
26300
25500 [a]
4.35
4.08
3.8[a]
4.22
4.29
4.15 [a]
[a] Shoulders in the ascent side of the second x + x* band.
3.2. Intensity of the n+z* Band in the Absorption Spectrum
The intensities of the n+n* bands of substituted azobenzenes are also only slightly affected by the nature of the
substituent (Table 2); for the diarylazo compounds in Table
3, the intensity of the n+n* band depends only on the
nature of the substituent, and not on its position in the
aromatic ring.
The intensity of the n+x* band of trans-azobenzene decreases with falling temperature (Fig. 2).
The intensities are generally greater for the cis azo compounds than for the trans isomers (Tables 1, 2, and 4).
The oscillator strength of cis-azobenzene is of an order
of magnitude that is normally restricted to n+n* transitions.
Angew. Chem. internat. Edit.
1 Vol. 12 (1973) / N o . 3
Structured n+x* bands have not been found in solution
for trans-azobenzene or cis-azobenzene.Even low-temperature spectra (Figs. 2 and 4) and the spectrum of a solution
in polyethylene show only continuous n+n* bandsczg1.
Attempts to influence the shape of the n+n* band by
fixation of the molecule on surfaces have also been unsuccessfulf3'1.
The solution spectra of the acyclic trans azo and cis azo
aliphatic compounds (Table 1) also contain only unstructured bands (Fig. 3b)[I6.''I. In contrast, the cyclic aliphatic
azo compounds ( 3 ) and ( 4 ) in apolar solvents exhibit
very sharp structures (Fig. 3 b)cZo.321. Distinct structures
are observed in the low-temperature spectra of the pyrazolines ( 5 ) and (7), whereas (6) gives a smooth n+n*
band even under these conditions (Fig. 5)I1'].
227
-Absorotion
Emission
I
:on
bl
&
I
25
T Ccm-'1
131
=rn
/N
, -CH
3 1I
:H3-N
/
/
I21
/
I
4'
25
laszs3]
03
B,
-
-----)
Owing to its solubility properties, a low-temperature spectrum can be recorded for the cyclic trans azo compound
( J l ) only in polar solvents. In ethanol and ether-ethanolethyl acetate mixtures, a shoulder (Fig. 6)r361appears on
the n - m * band at low temperatures.
35 lo3
30
[crn-l'l
35 lo3
The o-diazines (8), ( 9 ) , and (10) give distinctly structured
n+x* bands in apolar solvents, even at room temperaturer33 - 351
0
m
30
Fig. 6. Absorption of (11).
t
h
10
n
Excitation
Fig. 3. a) Gas spectra of ( 4 ) [ZO]; b) absorption of __ ( 3 ) in
isooctane [32], --- frans-(2) in hexane.
The crystal spectrum of azobenzene (transmission at 298
and 77°K and reflection at room temperature) shows only
a continuous n+x band'"'. A hint of a structure was
found in the low-temperature spectrum of azobenzene in
a stilbene host crystal at 77°K(371,while the n+n* band
in the spectrum of azobenzene in a bibenzyl crystal at
20°K is unstructured'381.
3.3.2. Vibrational Structure of the Low-Energy K+X* Band
il
The n+x* bands behave quite differently from the n+x*
bands with regard to their vibrational structure. The gas
spectrum of azobenzene contains a sequence of vibrational
bands separated by intervals of 220cm- i139.401 and of
1300-1400cm-
i
The x+x* band of trans-azobenzene shows a hint of structure in apolar solvents, and this structure becomes sharp
for solutions in polyethylene or in solid glasses at 77°K
(Fig. 2)r291.
Fig. 4. Absorption ofcis-azobenzene - - ~ -and
- o,o'-azodiphenylmethane
___ m methylcyclohexane/isopentaneat 77°C.
I
161
151
The x+x* band of cis-azobenzene shows no structure
even at 77 oK[201,whereas the bridged o,o'-azodiphenylmethane ( J 4 ) , in addition to a large red shift and an
increase in intensity, gives a structured x+x* band with
intervals of 1400-1600cmbetween the maxima (Fig.
4, Table 4)[l9].
The x+x* bands of o-diazines are strongly structured;
the aromatic character of these compounds is dominant
in this respect.
171
3.4. Emission of Azo Compounds
3.4.1. trans Azo Compounds
At present hardly any''] acyclic trans azo compounds are
known that exhibit emission, even under conditions that
normally promote such behavior, such as deuteration,
Fig. 5. Absorption and fluorescence of (7 ) ~ - - - and
- (5)
and absorption of (6) ~. - .-.
228
~
p] Some exceptions are being investigated at present. The reason for
the emission of these compounds seems to be the exchange of the l(n,x*)and '(x,rr*)-states on the energy scale [31, 411
Angew. Chem. internat. Edit.
Vol. I 2 (1973)
1 No 3
incorporation in a vitreous matrix, incorporation in a
solid solution, low temperatures, etc.
4. Electronic States of the Azo Compounds
The cyclic trans azo compound ( I 1 ) does not fluoresce,
even at low t e r n p e r a t ~ r e d ~either
~ ] , in solution or in the
crystalline state where the azo groups are planar[4z! In
ethanolic solution (12) fluorescences weakly at room temperature and fairly strongly at low temperatures. Closer
investigations lead, however, to the conclusion that it is
not the azo form of the molecule that is the emitting
species but probably some other tautomeric form[361.
4.1. (x,n*)States
3.4.2. cis Azo Compounds
The acyclic aliphatic or aromatic azo compounds in the
cis configuration d o not emit, nor d o the nonplanar molecules o,o‘-azodiphenylmethane ( 1 4 ) and o,o’-azobibenZ Y I ~ which
’ ~ ~ , are related to cis-azobenzene.
Among the cycloaliphatic compounds in Table 1, the rigid
bicyclic molecules have higher fluorescence quantum yields
than the planar pyrazolines. Diazacyclohexene ( I 3) does
not fluoresce.
The bands of the azo compounds that are attributable
to x-m* transitions are very similar to those of the analogous hydrocarbons. Useful methods based on molecular
orbital (MO) theory are available for calculation of the
x orbitals and z+x* transitions.
One approach using the M O model proceeds according
to the “molecules in molecule” (MIM) concept starting
from the assumption that the azo compounds consist of
more or less strongly coupled electronic subunits; azobenzene, for example, is regarded as two benzene systems
and the azo g r o ~ p ‘ ’ ~This
] . approach is favored by the
molecular geometry (Section 2.2). It can be seen from
Figure 7 that there is only a small difference between
the MO and the MIM energy level diagrams in the simple
Hiickel-MO approximation.
I =
3.4.3. o-Diazines
I
Emission is the rule rather than the exception for these
compounds. Pyridazine (S)[331 and benzo[c]cinnoline
( ZO)[351 fluoresce, and phthalazine (9)[43.441 phosphoresces.
3.4.4. Protonated Azo Compounds
Azo compounds dissolve readily in strong acids with asymmetric p r ~ t o n a t i o n [461
~ ~and
. concomitant disappearance
of the n+x* band in the absorption spectrum. Almost
without exception, the protonated azo compounds fluorescef4’]. Without the low-lying ‘(n,x*) state, however,
the protonated molecules are spectroscopically no longer
azo compounds.
3.4.5. uydroxy Azo Compounds
Another class of azo compounds that can fluoresce
embraces the 0-and p-hydroxy derivatives. However, this
fluorescence is due to the tautomeric hydrazone form[48.491.
3.5. Position of the Triplet States in Azo Compounds
Triplet states have not so far been observed for any normal
azo compound by the usual spectroscopic methods.
’I it was posFor p y r i d a ~ i n e [and
~ ~ ~benzo[~]cinnoline[~
sible to identify the lowest triplet states, from the crystal
absorption spectrum at 4.2 “K, as the 3(n,x*) and the 3 ( x , ~ * )
states, respectively; phthalazine
In addition to spectroscopic measurements, photosensitized reactions can also provide information about the
position of the triplet levels. Azobenzene can undergo
photosensitized isomerization, thus revealing that the triplet state lies below 21000 m-1[521.Azoalkanes quench
the phosphorescence of acetone and b i a ~ e t y l [ ~the
~ ] ;singlet-triplet difference is 3000 to 4 0 c m - ’ . A similar split
of about 2 5 0 0 m - ’ is found for (3)[541.These results
are of the expected order of magnitude[301.
Angew. Chem. internat. Edit.
Vol. I 2 ( 1 9 7 3 ) 1 N o . 3
Fig. 7. Energy level diagram of azobenzene in the Hiickel MO and
MIM approximations.
cis Azo compounds are mostly nonplanar. When this is
taken into account, it is possible to simulate the observed
solution spectra[25.561. The effect of steric f a ~ t o r s ~ ~591
~.~’on the n+n* bands is clear from Figure 4.
4.2. (n,n*) States
4.2.1. M O Model
The n orbitals of the azo group have a symmetry property
of the o-bond skeleton: unlike the x orbitals, they are
symmetrical on reflection in the plane of the molecule.
If two n orbitals are situated close enough together, as
in the azo group, M O theory predicts that they can combine
in the same way as the pz orbitals of the C atoms in
conjugated molecules. The two n orbitals then split up
into a group orbital having a higher energy and one having
a lower energy (Figs. I d and 8). The functions of the
new states, which have definite symmetry properties that
differ in the trans and in the cis isomers, can be expressed
in this method as a linear combination of the original
n orbitals. Since the group orbitals n, and n, are both
doubly occupied, they contribute practically nothing to
the bond energy.
229
A first theoretical approach has been made to find a model
for the n+z* excitation that attempts to combine the
elements of the M O and exciton models[651. However,
it has not yet been adapted for azo compounds.
1
n ---lnl-n21
a
4.2.3. Ab-inirio Calculations
-v
1
n ---lnltn21
-vT
I
Fig. 8. The n orbitals of the trans and cis azo compounds in the MO
model.
The magnitude ofthe n,-n,
splitting has been the subject
of a series of theoretical studies within the M O concept.
Small values[“] and values of the order of loo00
cm- 1[25.37.611
have been calculated for aromatic azo compounds and o-diazines. An ab-initio calculation even gives
3 0 0 0 0 m - ’ for diimine HN=NHr30J.
In the M O model, the selection rules for trans and for
cis azo compounds are different. For trans-azobenzene,
the low-energy n+n* band belongs to a forbidden transition, whereas the transition to the higher ‘(n,,n*) state,
with the polarization direction perpendicular to the plane
of the molecule, is allowed. The situation is reversed for
cis-azobenzene; the transition to the lowest l(na,n*) state
is allowed, again with the polarization direction perpendicular to the plane of the molecule.
In the MIM modification the n,-n,
on symmetry grounds.
splitting is retained
The semiempirical M O and exciton models start with
the assumption that the n orbitals are localized on the
nitrogen atoms. This basis appears uncertain from ab-initio
calculations. Calculations by various methods for pyrazine
lead to the conclusion that the highest occupied n orbital
of the ground state has a large share in ISorbitals of
other atoms[“. ‘’I. A geometry-dependent participation
of the s electrons of the hydrogen in the n orbitals of
the azo group is also found from calculations on diimine[30!
This result throws much greater doubt on the applicability
of the exciton model than on that of the M O model,
since the symmetry properties of the group orbitals at
least are retained in the latter case.
4.2.4. Comparison of the Models with Experiment
The correct model must first reproduce correctly the following properties of the spectra of azo compounds: similar
energy of the ‘(n,n*) states in most of the azo compounds
(Tables 1-3), splitting of the n orbitals, and the selection
rules.
A constant position of the n+n* bands in the spectrum
is compatible with the exciton and the MIM models. The
exciton transitions are localized on the nitrogen atom,
and they are little influenced by the groups at either end
of the azo moiety. A critical comparison of exciton and
MO calculations favors the M O theory, since this agrees
better with the experimental results[‘*!
4.2.2. Exciton Model
In the MIM modification of the M O model, the n* orbital
is a perturbed orbital of the azo group (Fig. 7). On substitution in the aromatic subsystem, therefore, the lowest unoccupied a* orbital is displaced relatively little, whereas the
highest occupied n orbital, an orbital of the aromatic
subunits, is sensitive to substituents. Thus the n+n* band
isdisplaced on substitution in the phenyl groups of azobenzene or on replacing the benzene by naphthalene moieties,
whereas the n+n* band is not. The n+z* transition is
localized in the azo group, i.e. it is not a charge transfer
transition. However, the n+n* band is displaced if the
In molecules such as the o-diazines and the azo compounds,
substituent on the phenyl group also carries azo groups,
two such exciton centers are sufficiently close together
as in bisazobenzene, trisazobenzene, tetrakisazobenzene,
for the interaction of the equally large transition moments
etc., since the n orbitals of the individual azo groups can
M , and Mb to lead to splitting of the exciton ~ t a t e s I ~ ’ . ~ ~ ’ . then interact with one another.
Even when M is small, the splitting can be considerable
The strange similarity of the intensities (Table 3) of the
in azo compounds and o-diazines because of the short
n+n* bands of trans-1,l’- and -2,2’-azonaphthalene (symdistance dabbetween the centers in a molecule, according
metry CZh)and trans-l,2’-azonaphthalene(symmetry C,)
to1641
can also be reconciled with the MIM model. The intensity
of the n+n* band is determined only by the nature of
the residues on the azo group, and not by the total symmeIn the exciton model, the selection rules for cis and trans
try of the molecule, as long as the configuration of the
azo group itself is not altered. If the configuration is altered,
azo compounds are the same. The position of the transition
the probability of transition is greatly increased in the
moments leads to the selection rules for “card-packed
dimers”; the low-energy exciton transition is always forcis isomer (Table 1). These selection rules provide the
641.
bidden, and the other transition is
strongest argument in favor of the M O model.
When an electron localized in an n orbital of a hetero
atom is excited into a n* orbital, a positive hole remains
behind on the hetero atom, and this must influence the
negative charge in the delocalized n* orbital. The charge
distributions in the (n,n*) and (n,n*) states will therefore
be fundamentally different. If the attraction between the
electron and the hole is very strong, the (n,n*) state may
be described essentially by an electron-hole pair, i. e. an
exciton[‘’l.
230
Angew. Chrm. internat. Edit. 1 Vol. I 2 ( 1 9 7 3 )
/ No. 3
Experimental checking of the splitting of the n orbitals
as required by the models is hindered by the fact that
only the one low energy n+n* band can be satisfactorily
assigned in the azo compounds. Numerous attempts have
been made to identify weak bands in the spectra of aromatic
and aliphatic azo compounds as the second n+n* transition” 6 . 69.70J, but these assignments can hardly be regarded
as final. However, the splitting of the n orbitals of the
ground state is clearly recognizable in the photoelectron
spectrum[71],where it has a magnitude of 26000cm-’
the rnolecule[l7. 29.371. This rules out any possibility that
this n+n* transition is allowed, and hence aIso rules out
the applicability of the exciton theory.
O n the other hand, measurements on diazines point t o
several weak transitions in the region of the long-wave
n+n*
7 2 . 731. However, the results are partly contradictory, and the identification of these transitions has
not been verifiedLg4!
An interaction of this type must influence both of the
coupled states. This can be seen directly on comparison
of the vibrational bands of the 7 c - m transition in the
low-temperature spectrum of stilbene and of azobenzene
in the bibenzyl host lattice. The bands are about 15 times
as broad in the case of a ~ o b e n z e n e ~The
~ ~effect
!
of coupling
can also be demonstrated by the influence of increasing
deviations from a planar molecular structure on the
allowed n+n* transition of the cis azo compounds. Table
4 shows the correlations between molecular geometry and
vibrational structure, intensity of the n+n* band, and
fluorescence yield.
4.3. Interactions of the Lowest l(n,n*) State
with Other Molecular States
Table 4. Correlation between molecular geometry and spectroscopic
properties of cis azo compounds.
Molecule
n+n*-band
structure
4.3.1. Vibrational Coupling with the Lowest ‘(n,n*) State
According to the M O theory, which we now favor, the
n+n* transition is forbidden in the trans azo compounds.
The observed intensity is nevertheless unusually high, and
for trans-azobenzene it is actually higher than for the
closely related benzo[c]cinnoline, for which the transition
is allowed. The reason for this must be that the ‘(n,n*)
state is not pure (Section 2.1) but is coupled with other
states. Such coupling is possible, e. g. through a deformation
ofthe molecular field of a certain symmetry by corresponding vibrations or deviations from the planar arrangement
of the atoms.
The transition moment Mg0, which corresponds to the
intensity of the transition from the vibrationless electronic
ground state (ebo).(p8) to the p-th vibrational state of the
electronically excited state i (8i0).q$),
is
E,,
pronounced 350
500 cm-
’
n+n*-band
structure
E,
Fluorescence
yield [%]
sharp
1500
500, 1250 cm-’
0.7
RT: none
77°K: weak
1400 cm-’
-
(10)
970
N=N.
_.
(14)
9900
Benzo[c]cinnoline is actually an azo compound rather
than a diaza-aromatic compound. The absorption and
emission properties[’*] can be satisfactorily described by
the MIM model.
4.3.2. Interaction with Other States
and taking into account the vibrational
V, is the operator of the perturbation of the nuclear field
by the vibration p, q,, is its normal coordinate, and Eio)
is the position of the pure electronic level i in the energy
level diagram. If the transition to the state is electronically
forbidden, rnb =0, and the observed intensity must be
“stolen” from states j whose dois large.
The integrals conceal the selection rules, which can be
determined by the use of the character tables; the transition
is allowed only if the sign and magnitude of the integrand
are unchanged by any symmetry operations that are possible for the molecule in question.
For the trans azo compounds, coupling between the l(n,n*)
state and the neighboring allowed ‘(x,n*) state is possible
through an out-of-plane deformation vibration. This brings
the polarization direction of the n+n* transition into
the direction of the transition moment m$.n*). Both bands
of trans-azobenzene are in fact polarized in the plane of
Angew. Chem. Internat. Edit. / Vol. 12 ( 1 9 7 3 )
No. 3
There is no doubt that the intensity of the n+n* band
in the trans azo compounds is “stolen” from the n+n*
transition. However, this does not necessarily lead to loss
of the vibrational structure, as is shown by the sharply
structured benzene band at 40000 cm-
’.
The overlap integral of the vibrational functions, the socalled Franck-Condon (FC) factors in equation (3), differs
from zero only for vibrations of the same symmetry. It
is therefore expected that only totally symmetrical vibrations p determine the structure of the electronic bands.
This is observed for the benzene band, and for trans-azobenzene one also finds the totally symmetrical N-N
stretching vibration at 14GQcm- in the n+n* band and
the totally symmetrical in-plane deformation vibration at
220 cm- 1[761. The n - m * band, however, is structureless.
A band form of this type may be due to a quasi-continuum
or to a true level broadening.
A quasi-continuum is observed when the distance between
discrete bands is smaller than their half-width. This is
possible e. g. for polyatomic molecules of low symmetry,
where very many vibrations may be stimulated with the
231
electronic transition. The molecules of azobenzene or azomethane on the other hand, are too small and their symmetry too high for this to occur.
True level broadening is to be expected if, according to
the Heisenberg uncertainty relation
A l x A E = 5.25 x
The most extensively used model of radiationless deactivat i ~ n [ ’ ~ - ~is’ based
]
on coupling of the lattice vibrations
of the solvent Jrf, whose number is very large, with the
vibrationless excited electronic state +’, which is assumed
to lose its energy non-radiatively (Fig. 10).
[c m -ls ]
the true lifetime T = A of
~ an excited vibronic state is sufficiently small. Line widths of 220 and 1500cm- correspond
to lifetimes of 2.5 x 10- l 4 and 3 x 10- l 5 s respectively.
The intrinsic lifetime of a state, T,,, which is determined
by the rate of the resonance fluorescence, can be calculated
from the intensity of the corresponding absorption
band[??],and is to= 5 x 10- s for the ‘(n,n*) state of transazobenzene. This lifetime may be shortened by radiationless processes, which are faster than the emission of light.
However, a radiationless transition between two states
is possible only if the states are coupled.
’
The degree to which the states retain their individuality
and can be identified in the spectrum is dependent upon
the strength of coupling. If the energy of a system capable
of vibrating is removed by very tight coupling before the
characteristic vibrational motion can develop, the actual
lifetime drops to below one vibrational period, and the
individuality of this vibration is no longer recognizable.
In molecular systems this means that the BO approximation is no longer valid. One interpretation of the lack
of structure of the n+x* band of azo compounds would
be the assumption of particularly tight coupling of the
vibrations in the ‘(n,a*) state. However, this is opposed
by theoretical considerations[731and the spectroscopic
findings.
The fact that azo compounds d o not fluoresce indicates
that the coupling of electronic states is responsible for
the shape of the n - m * band of the azo compounds. The
processes that are important to the deactivation of an
excited molecule are outlined in Figure 9.
Reaction
cont tnuurn
Fig. 10. Coupling in radiationless deactivation.
If only the initial state +’=e,.cp; and the final state
=€lo.
q$ (Fig. lo), a highly excited vibrational state of
the electronic gi-ound state, were coupled by a low interaction energy j3,the molecular system would oscillate between
the two states with a frequency of
+“
where H’ is the coupling operator. However, the final
state is in interaction cc with the environment, i.e. with
the thermal vibrations of the solvent. According to spectroscopic findings, cc is much greater than !3, and the states
+“’ ere. coupled with the final state are very numerous.
The energy is very rapidly partitioned from the “hot”
final state
over so many low excited vibrations of
the solvent that the chance of an accumulation of energy
to regenerate the final state $” and reversion to the initial
state is extremely small ( T ~ =a).
+’I,
+”
+’
The BO approximation is assumed to be valid in Figure
10.Thegreater the coupling j3, the faster is the radiationless
deactivation. This is true even when the vibronic states
$’ and
have largely lost their identity, and one can
therefore hardly expect a vibrational structure in the n+n*
band.
+”
There must thus be at least one vibronic state in the
azo compounds at approximately the same energy as the
‘(n,a*) state, with which the l(n,n*) state can interact.
Fig. 9. Deactivation processes
Transitions in the energy level system can be characterized
by first order rate constants; k, and k,, are the constants
for fluorescence and for phosphorescence respectively, krC
is the constant for vibrational relaxation within a multiplicity, and k,s and kphot are the constants for the multiplicity
change and for photochemistry.
232
The first possibility would be a highly excited vibrational
state of the ground state, as shown in Figure 10. However,
the values of the F C factors, which can lie between 1
and lo-’’ in the various cases[791, decrease roughly
exponentially with the difference in the vibrational quantum numbers in the initial and in the final state[761.Owing
to the large energy difference +’-So (Fig. lo), the FC
factors are very small. Deuteration of the molecule, which
causes the FC factors to decrease further and thus normally
increases the fluorescence quantum yield[63.821, has no
influence on the band shape and fluorescence in the case
of azobenzene. This points to a small difference between
the coupling vibrationless electronic statedS3!
Angew. Chem. internat. Edit. / Val. 12 (1973)
No. 3
Optimal coupling would be offered by a singlet state S,
(Fig. 9) in the vicinity of the ‘(n,n*) state. There need
not be any measurable absorption corresponding to this
state. However, there is no evidence of such a state for
azo compounds, though evidence is found in the low-temperature spectra of the o-diazines[62].The existence and
identity of such a singlet state has yet to be confirmed.
The spectroscopic behavior of the azo compounds could
also be explained by tight coupling of the ‘(n,n*) state
with a triplet state T I (Fig. 9). The prohibition of intersystem crossing would have to be overcome in this case.
The 3(n,n*)state is situated only slightly below the ‘(n,n*)
state, and this leads to favorable FC factors. The coupling
between singlet and triplet states corresponds to the spinorbit interaction of the electrons, and the coupling energy
is
The spin-orbit coupling operator H, o. has the symmetry
properties of a rotation[M4];there are consequently also
symmetry-dependent selection rules for the radiationless
S-T transitions. Coupling between similar states, e. g.
‘(n,n*) and 3(n,n*) states, is forbidden in cis azo compounds, but allowed in the trans isomers[441.On the other
hand, transitions between singlet and triplet states of different kinds are allowed. However, this electronic selection
rule is strictly valid only if the configuration of the molecule
is planar. Structured n+n* absorption bands and n+n*
fluorescence can therefore be expected only for planar
cis compounds where deactivation uiu the ’(n,n*) state
is prevented, and this is in agreement with the experimental
findings. The more resistant the skeleton to out-of-plane
deformation vibrations, the sharper is the structure and
the higher the fluorescence yield (Tables 1 and 4).
However, the energy of the 3(n,n*) state can be lower
than that of the ‘(n,n*) state for aromatic compounds.
The probability of fluorescence would then be small. It
is in fact found that benzo[c]cinnoline, for which the lowest
triplet state is regarded as a 3(n,n*) state on the basis
of vibrational analysis[”], has a fluorescence quantum
yield of only 0.7%[75.851.
A possible objection lies in the question whether the mixing
of states having different multiplicities can be sufficiently
strong. Coupling is usually determined experimentally from
quantum yield measurements; no answer can be given
for the azo compounds concerned, owing to the absence
of emission. However, the spin-orbit coupling depends
on the gradient of the nuclear field, and because of their
s character, the n electrons are much more exposed to
this field than n electrons[8’! This coupling is also favored
by any deviations from planarity of the molecule. A pathway uia the triplet system is accepted for the photoisomerization152.5 3.861
A final possibility is the coupling of the lowest excited
state with a dissociative molecular state. This is undoubtedly an important case for the aliphatic azo compounds.
Both the vibrational structure and the fluorescence are
more pronounced for the more photostable chloromethylenepyrazoline f5)[871 than for the more labile ketopyrazoAngew. Chrm. internat. Edit.
/
Vol. 12 (1973)
/ No. 3
line (6). The weak shoulder observed for the n+n* a b s o r p
tion band of ( 1 1 ) could also be due to the very much
higher photostability of this compound in relation to rransazomethanel8*I.
The azobenzenes, and under certain conditions azoalkaneslM2.
861, are photostable apart from cis-trans isomerism. However, the molecule is not necessarily destroyed
by a photoreaction. A transition of the excited molecule
into a different form, e.g. a more extended form as an
intermediate of the isomerization due to inversion at the
nitrogen atom[s9.901 before the totally symmetrical deformation vibration can become fully developed, may also
be described by coupling of the ‘(n,n*) state of the angular
molecule with a state of the molecule having the transition
geometry. The FC factors in equation ( 5 ) will become
increasingly favorable with increasing difference in geometry between the coupling states.
The quantum yields of the isomerization of azobenzene
are high[”], and the kinetic analysis calls for intermediate
states[’l.”2! The decrease in the quantum yield of the
trans-cis i s ~ m e r i z a t i o n [ with
~ ~ ] falling temperature does
not necessarily mean that the intermediate state is not
reached, as long as the step that requires an activation
energy takes place from this state.
Vibrational structure and fluorescence are therefore to
be expected only for azo compounds that are stabilized
in the trans or cis form. This is also largely in agreement
with experimental results.
The last two possibilities are not mutually exclusive. The
state that couples with the ‘(n,a*) state is probably a
triplet state with modified molecular symmetry. If the existence of such a state is accepted, the spectroscopic and
photochemical results as a whole can be logically explained,
particularly the kinetic analysis of the cis-trans isomerizat i ~ n [ ~and
’ ] the sensitization of this reactionls4.581.
5. Conclusion
The absorption spectrum of the azo compounds is best
explained with the aid of an M O model in which the
molecules are assumed to be built u p from coupled electronic subunits. The electronic system of the azo group
is significantly perturbed in the aromatic compounds, but
the n+n* transitions are determined primarily by the
local symmetry of the azo group, and not by the molecular
geometry .
The ‘(n,n*) states of the azo compounds exhibit unusual
coupling properties. By vibrational coupling, the n - m *
transition, which is forbidden for the trans isomers, “steals”
its high intensity, which may exceed that of the corresponding allowed transition in a closely related cis molecule,
from the neighboring allowed n+n* transition. The inability to fluoresce and the absence of a vibrational structure
in the n+n* band for most azo compounds is due to
the tight coupling of the ‘(n,n*) state with a somewhat
lower-lying molecular state, which is determined by symmetry-dependent selection rules for electronic states and
by geometric factors. The identity of this state has not
233
yet been established; however, the experimental data point
to a triplet state, probably with a molecular geometry
that is strongIy modified in relation to the configuration
of the ground state.
Received: February 16, 1972 [A 929 IE]
German version: Angew. Chem. 85,248 (1973)
Translated by Express Translation Service, London
[l] H . A. Staabr Einfuhrung in die theoretische organische Chemie.
Verlag Chemie, Weinherm 1962, p. 305; Chem. Ber. 90, 1320 (1957).
[2] W Siebrand, J. Luminescence 112, 122 (1970).
[3] H. Bock, G. Rudolph, E. Baltin, and J . Kroner, Angew. Chem. 77,
469 (1965); Angew. Chem. internat. Edit. 4, 457 (1965).
[4] Landolt-Bornstein: Zahlenwerte und Funktionen aus Naturwissenschaften und Technik. Springer Verlag, Berlin 1955, Vol. 1, Part 4, p.
404.
[5] H. Boersch, Monatsh. Chem. 65, 331 (1935).
[6] J . M . Robertson, J . Chem. SOC.1939,232.
[7] G. C . Hampson and J . M . Robertson, J. Chem. SOC.1941,409.
[8] F. KIagesr Lehrbuch der organischen Chemie. de Gruyter, Berlin
1957, Vol. 2, p. 542.
[9] E. Miiller: Neuere Anschauungen der organischen Chemie. SpringerVerlag, Berlin 1957, p. 389.
[lo] G. S. Hartley, J. Chem. SOC.1938, 633.
[I I] M . Tsuda and K . Kuratani, Bull. Chem. SOC. Japan 37, 1284 (1964).
[I21 R. F . H u t o n and C . Steel, J. Amer. Chem. SOC.86, 745 (1964).
1131 S. Hiinig and J . Cramer, Angew. Chem. 80, 1000 (1968); Angew.
Chem. internat. Edit. 7,943 (1968).
[I41 P. Skrabal and H. Zolfinger, Helv. Chim. Acta 54, 1069 (1971).
[I51 J . R. Platt, J . Chem. Phys. 19, 101 (1951).
[I61 G. Kortiim and H . Rau, Ber. Bunsenges. Phys. Chem. 68,973 (1964).
1171 M . 8.Robin and W T Simpson, J . Chem. Phys. 36, 580 (1962).
[I81 H. Rau, J. Luminescence I j 2 , 191 (1970)
[I91 H. Rau, unpublished.
[20] B. S. Solomon, 7: F . n o m a s , and C . Steel, J. Amer. Chem. SOC.
90, 2249 ( I 968).
[21] C . G. Ouerberger and J . Anselme, J. Amer. Chem. SOC. 84, 869
( 1962).
1221 DMS-UV-Atlas. Verlag Chemle, Weinheim 1967, Vol. 111, Sp. C
715.
[23] Landolt-Bornsteinr Zahlenwerte und Funktionen aus Naturwissenschaften und Technik. Springer-Verlag, Berlin 1951, Vol. 1, Part 3, p.
310.
[24] P. P. Birnbaum and D. W G. Style, Trans Faraday SOC.50, 1192
(1954).
[25] G. Hohlneicher and W. Sanger in E. D. Bergmann and B. Pullman:
Quantum Aspects of Heterocyclic Compounds m Chemistry and Biochemistry. Proceedings of an International Symposium held In Jerusalem
March 31-April 4, 1969. Academic Press, New York 1970, p. 193.
[26] H. Dahn and H. v. Castel-Mur, Helv. Chim. Acta 36, 638 (1953).
[27] J . Schulze, F. Gerson, J . N . Murrefl, and E. Heilbronner, Helv.
Chim. Acta 44, 428 (1961).
[28] M . Kasha, Discuss. Faraday SOC.9, 14 (1950).
[29] H . Rau and G. Kortiim, Ber. Bunsenges. Phys. Chem. 71,644 (1967).
[30] M . 5. Robin, R. R. Hart, and N . A. Kuebler, J . Amer. Chem.
SOC.89, 1564 (1967)
[31] H . Rau, Ber. Bunsenges. Phys. Chem. 75, 1347 (1971).
[32] S. G. Cohen, R. Zand, and C. Steel, J. Amer. Chem. SOC.83, 2895
(I 96 1).
[33] 5. J . Cohm, H. Baba, and L. Goodman, J. Chem. Phys. 43, 2902
(1965).
[34] R. C . Hirt, F. T King, and J . C. Cauagnol, J. Chem. Phys. 25,
574.
[35] E. Lippert and W Voss, 2. Phys. Chem. N F Frankfurt 31, 321
(1962).
[36] H. Rau, unpublished.
[37] R . M . Hochstrasser and S. K . Lower, J . Chem. Phys. 36, 3505
( 1962).
[ i X ] R ff. Di-(.i, a n d D S t f < C l i i n , . I Chcm I'll\\. 36. 2726 t19621.
[39] Y. Kanda, Mem. Fac. Sci. Kyushu Univ. Ser. C Cbem. I , 189 (1950).
[40] S. rrrl~l~li~~ll
i l n d 1. I~I</II. _I L I~CI~I. Soc. .lapan 6 4 . 521 (lv-I.31
[41] H . Bule and H . Rau, unpublished.
234
[42] R. Allmann and I . Kawada, Angew. Chem. 80, 1001 (1968); Angew.
Chem. internat. Edit. 7, 944 (1968).
1431 R. Miiller and F . Ddrr, Ber. Bunsenges. Phys. Chem. 63, 1150
(1959); E. C. Lim and J . Stanislaus, J. Chem. Phys. 53, 2096 (1970).
[44] H. Rau, Ber. Bunsenges. Phys. Chem. 72, 408 (1968).
[45] E. Haselbach and E. Heilbronner, Tetrahedron Lett. 1967, 4531.
[46] F. Gerson, E. Heilbronner, A . van Veen, and B. M . Wepster, Helv.
Chim. Acta 43, 1889 (1960).
[47] H . Rau, Ber. Bunsenges. Phys. Chem. 71, 48 (1967).
[48] H. Rau, Ber. Bunsenges. Phys. Chem. 72, 637 (1968).
[49] G. Gabor, t: Frei, D. Gegiou, M Kaganowitch, and E. Fischer,
Isr. J. Chem. 5 , 193 (1967).
[50] R. M . Hockstrasser and C . Marzzacco, J. Chem. Phys. 46, 4155
( 1 967).
[Sl] R. M Hochstrasser and C. Marzzacco, J. Chem. Phys. 45. 4681
(1966).
[52] E. Fischer, J. Amer. Chem. SOC.90, 796 (1968).
[53] R. E. Rebbert and P. Ausloos, J. Amer. Chem. SOC. 87, 1847 (1965).
[54] P. S. Engel, J . Amer. Chem. SOC.89, 5731 (1967).
[55] H . Suzuki: Electronic Absorption Spectra and Geometry of Organic
Molecules. Academic Press, New York 1967, Chap. 19, 20, 23.
[56] D. L. Beveridge and H. H. Jo@,
J . Arner. Chem. SOC.88, 1948
( 1966).
[57] M . Ashraf El-Bayoumi and F. M . Abdel Halim, J . Chem. Phys.
48, 2536 (1968).
[58] W H. Laarhouen, R. J . F . Nivard, and E. Hauinga, Rec. Trav.
Chim. Pays-Bas 79, 1153 (1960).
[59] H . Suzuki: Electronic Absorption Spectra and Geometry of Organic
Molecules. Academic Press, New York 1967, Chap. 14, 15, 18.
[60] H. H. Jafli, Si-Jung Yeh, and R. W Gardner, J . Mol. Spectry.
2, 120 (1958).
[61] S. F . Mason, J . Chem. SOC.1959, 1240.
[62] M . A. El Sayed and G. W Robinson, Mol. Phys. 4, 273 (1961);
J. Chem. Phys. 34, 1840 (1961); 35, 1896 (1961).
[63] 5. J . Coken and L. Goodman, J. Chem. Phys. 46, 713 (1967).
[64] M . Kasha, H. R. Rawles, and M . Ashraf El-Bayoumi, Pure Appl.
Chem. 11, 371 (1965).
[65] R. A. Harris and L. M . Falicou, J . Chem. Phys. 55, 2931 (1971).
(661 E. Clementi, J. Chem. Phys. 46, 4737 (1967).
[67] M . Hackmeper and J . L . Whitten, J . Chem. Phys. 54, 3739 (1971).
[68] D . R. Kearns and M . Ashraf El-Boyoumi, J . Chem. Phys. 38, 1508
( I 963).
[69] J . E. Parkin and K . K.' lnnes, J. Mol. Spectry. 15, 407 (1965).
[70] M . 5. Robin, H . Basch, N . A. Kuebler, K . Wiberg, and G. B. Ellison,
J . Chem. Phys 51,45 (1969).
[71] E. Haselbach and E . Heilbronner, Helv. Chim. Acta 53, 684 (1970).
[72] R. M . Hochstrasser and C . Marzzacco, J. Chem. Phys. 48, 4079
(1968).
[73] R. M . Hochsrrasser and C . Marzzacco, J. Chem. Phys. 49, 971
( 1968).
[74] E. g. R. M . Hochsrrasser Molecular Aspects of Symmetry. Benjamin,
New York 1966, Chap. 18.
[75] E. M . Kreidler, Dissertation, Universitiit Stuttgart 1966.
[76] H. Srammreich, Experientia 6, 225 (1950).
1771 7h. FBrsterr Fluoreszenz organischer Verbindungen. Vandenhoek
& Rupprecht, Gottingen 1951, paragraph 32.
'
[78] G. W Robinson and R. P. Frosch, J . Chem. Phys. 37, 1962 (1962);
38, 1184 (1963).
[79] G. R. Hunt, E . F . McCoy, and I . G. Ross, Austr. J . Chem. 15,
591 (1962); J P. Byrne, E. F. McCoy, and f. G. Ross, ibid. 18, 1589
(1965).
[80] M . Bixon and J . Jortner, J. Chem. Phys. 48, 715 (1968); 50, 3284,
4061 (1969); J . Jortner and R. S. Berry, ibid. 48, 2757 (1968).
[$I] 5. R. Henry and M . Kasha, Ann. Rev. Phys. Chem. 19, 161 (1968).
[82] G. D. Johnson, L. M . Logan, and I . G. Ross, J. Mol. Spectry.
14, 198 (1964).
1837 E. C. Lim and J . D. Laposa, J . Chem. Phys. 41, 3257 (1964).
[84] D . S. McClure, J. Chem. Phys. 17, 665 (1949); S. P. McGlynn,
7: Azumi, and M . Kinoshira. Molecular Spectroscopy of the Triplet
State. Prentice-Hall, Englewood ClifTs, N. J. 1969, p. 191.
[85] H. Rau and H . Bisle, Ber. Bunsenges. Phys. Chem., in press.
[86] 1. 1. Abram, G. S. Milne, B. S. Solomon, and C . Steel, J. Amer.
Chem. SOC.91, 1220 (1969).
Angew. Chem. internat. Edit. j Voi. 12 (1973) / No. 3
[87] A. C. Day, personal communication
[88] H . Rau, unpublished.
[89] D. Gegiou, K . A. Muszkat, and E. Fzscher, J. Amer. Chern. SOC.
90, 12 (1968).
[90] Z . Simon, Rev. Roumaine Chim. 21, 35 (1966)
[91] H . Sfegemeper, J. Phys. Chem. 66, 2555 (1962); S. Yamashira, H.
Ono, and 0. Toyama, Bull. Chem. SOC.Japan 35, 1849 (1962).
[92] U . B. Hezel, Dissertation, Universitat Tiibingen 1966.
[93] S . Malkrn and E . Fischer, J. Phys. Chern. 66, 2482 (1962)
[94] R. M . Hochstrasser and D . A. Wiersma, J. Chem. Phys. 56, 528
(1972).
C 0 M M U N I CAT10N S
Onium Salts Containing Carbonylmetalate
Anions"]
goes desilylation even on reaction with the phosphorussilyl ylides''].
By Wolfgang MaEisch[*]
The structures of the products, which are pale yellow salts,
sensitive to oxygen, can be proved by analysis and NMR
and IR spectroscopy[51.Their behavior towards organometallic bases in transylidation and deprotonation processes [(a) and (b), respectively] is comparable with that of
onium
Ylide chemistry has so far provided hardly any information
regarding the reactivity of ylide systems toward transition
metal complexes. While looking for metalate reagents
suitable for the synthesis of metal-metalloid bonds,
we have found that an acid-base reaction according to the
following scheme occurs on treatment of carbonylmetal
hydrides ( I ) with alkylidene or imino derivatives of phosphorus, arsenic or sulfur.
+
[ ( CH3),PI0[M(CO)3- n-C5H51@
M = M o ,W
Of the onium salts studied so far, a few are extremely reactive. Their use for transfer of the complex metalate ion,
is particwith formation of metal-metalloid bonds"'
ularly advantageous in apolar solvents, for with some catTable 1. Properties of some onium carbonylmetalates [3]
Starting
ylide
Cation
246-248
246-249
266-269
94-98
115-117
107-109
98-101
107-109
125- I28
139- 142
131- I33
116-119
120-122
173-176
125-128
< I85
1895, 1758
1898, 1760
1888. 1750
1892, 1765
1889, 1761
1887, 1764
1894, 1768
1890. 1762
1893. 1763
1887, 1759
1894, I759
1892, I767
IXXX, 1762
1892, I752
1885. 1752
1893. 1755
[a] Closed capillary (N2).
All the compounds melt with decomposition.
[b] Measured as Nujol suspension.
With ylides of slight basicity, such as the compounds
(CH,),S(O)CHSi(CH,), or (CH3)3PNSi(CH3)3,
the addition is followed by loss of the trimethylsilyl group, so that
silyl-free onium salts result. The most strongly acidic representative of the carbonyl hydride derivatives of VI b
transition metals, namely, HCr(CO),-x-C,H,, also under[*] Dr. W. Malisch
Institut fur Anorganische Chemie der Universitat
87 Wurzburg, Landwehr (Germany)
Angew. Chem. internat. Edit. 1 Vol. 12 (1973)
1No. 3
[A]efCr(C0)3-n-C5H5J0
+
RX
<
R-Cr(C0)3-n-C5H5
+
[Alo[Xlo
A = N a , (C2H5),P
R
X
Vro
M. p. ("Cj
(CH.d&i
CI(CH,)HSi
CI,HSi
Br
CI
CI
1992, 1924, I897
2012, 1944, 1913
2027, 1962, 1932
Dec. at room temp.
47-49
119 122
235
Документ
Категория
Без категории
Просмотров
4
Размер файла
1 090 Кб
Теги
spectroscopy, properties, azo, compounds, organiz
1/--страниц
Пожаловаться на содержимое документа