close

Вход

Забыли?

вход по аккаунту

?

Direct Observation of the Dissociation Equilibrium of Tropylium Salts.

код для вставкиСкачать
at our disposal, which is just enough to produce either
singlet (E:c,,,,, = 84 kcal/mol) or triplet acetone (E~c,,,,,
= 78 kcal/mol). This accords with recent results that after
taking up enough activation energy to yield an “activated
complex”, (/) needs no further addition of energy to form
singlet or triplet acetone”! Our results now to hand indicate, however, that it may be possible to obtain considerably
more luminescent singlet acetone on decomposition of ( 1 )
in the vapor phase than is possible in solution.
Received: July 20, 1973 [Z 892 IE]
German version: Angew. Chem. 85, 822 (1973)
[I] F . M K u p r a , Pure Appl. Chem. 24, 61 1 (1970).
[2] N. J . Turro and P. Lerhtken, J. Amer. Chem. SOC.94, 2886 (1972).
[3] J . C. Dulton and N . J . Xrrro, Annu. Rev. Phys. Chem. 21, 499
(1970).
[4] N. J . Turro and P . Lerhtken, J. Amer. Chem. SOC. 95, 264 (1973).
[5] H . E . O’Neol and W H . Richurdsun, J. Amer. Chem. SOC.92, 6553
(1970); 93, 1828 (1971); T Wilson and A. P. Schaap, ihid. 93, 4126 (1971):
W H . Richardson, M . 8. Yelcingron, and H . E . O N e u l , ;bid. 94, 1619
( 1972).
[6] a ) DSC calorimeter, type C P C 600, Arion, Grenoble; b) combustion
calorimeter, Peters KG, Berlin 21
[7] At 395 K about 1 mg of the acetone formed is in the gas phase,
as can be cdlculated from vapor pressure formulas [X] and the container
volume.
[8] L u n d o l t - B o r ~ s t r i n - B o t h - S ~ h ePhysikalisch-chemische
~,~;
Tabellen. 5th
Edit., Hauptwerk 11, Berlin 1936.
[9] H - C . Steinrnetzer. A. Yektu, and N . J . Turro, J. Amer. Chem. Soc.,
in press. We thank Prof. N J . Turro for this information.
Direct Observation of the Dissociation Equilibrium
of Tropylium Salts[**]
By Horst Kessler and Axel Walterp]
Dedicated to Professor Karl Winnacker on the occasion of
his 70th birthday
Mechanistic studies of the SN1reaction have shown that
dissociation to “free” cdrbenium ions occurs by way of
an intimate ion pair and a solvent-separated ion pair, and
that a steady state concentration of cations is established[’].
The definition of intermediates makes sense only if such
species are separated by an energy barrier[’]. Ions (ion
pairs) and nonionic molecules should be observable
alongside each other by NMR spectroscopy if the time
scale of the detection methodf3] permits their detection
and if their free energies are comparable. Hitherto, to
the best of our knowledge, dissociable compounds have
been observed either in nonionic or in dissociated form,
but not side by side.
We have found a tropylium salt that exists in solution
in dissociated and undissociated forms together, these
forms being separated by relatively high energy barriers.
The NMR spectrum of cycloheptatrienyl isothiocyanate
(I)[‘] in CDCI3 or ether at -4O‘C is the usual one for
7-substituted cycloheptatrienes [6=6.8 (t, 3-H, 4-H), 6.4
(m, 2-H, 5-H), 5.6 (dd, 1-H, 6-H), and 4.1 (t, 7-H, J = 5 Hz)],
but when the temperature is raised to ca. +2O-C it is
reversibly broadened and all the signals coalesce, indicating
migration of the NCS group around the whole sevenmembered ring”]. In acetonitrile, ( / ) dissolves only as
t r o p y h m isothiocyanate (singlet at 6 = 9.2), but surprisingly both forms are observed together in CD,CN/CDCl,
(1:3) at -40°C (Fig. 1).
[*] Prof. Dr. H. Kessler and Dipl.-Chem. A Walter
Institut fur Organische Chemie der Universitat
6 Frankfurt (Main), Sdndhofstrasse 2 (Germany)
[**I This work was supported by the Deutsche Forschungsgemeinschaft
and the Fonds der Chemischen Industrie.
AMJPW C h m i .
internut. Edit. / V i l . I2 ( 1 9 7 3 ) / Nu 9
Fig. I . IH-NMR spectrum of tropylium isothiocyanate in CD,CN/CDCI,
(1:3) at -4O’C; B, C : forms B and C of ( 1 ) ; numerals are numbers of
H atoms in A.
n
B
I])
C
The equilibrium between the ionized form (B, C) and the
nonionized form (A) was demonstrated by irradiation at
the tropylium frequency at - 1 5 T , which led to saturation
of the whole cycloheptatriene spectrum[’].
The intensity ratio of A: (B,C) can be modified by addition
of tropylium tetrafluoroborate or by change in the composition of the solvent, but not by alteration of the concentration of the solute in the solvent mixture. The tropylium
salt must therefore exist in this solvent mixture preponderantly as ion pairs B. With increasing amount of chloroform
in the (originally pure) acetonitrile solvent, the position
of the tropylium signal shifts to higher field and thus
indicates increasing pairing of the ions. This also follows
from the kinetics. Broadening of the tropylium signal in
the solvent mixture is independent of the concentration
(first-order reaction ; AG:,, = 12 kcal/mol[*]). The barrier
thus lies between the ion pair B and the cycloheptatriene
A. A possible reason for its size may perhaps be seen in
the aromatic tropylium system which has to be distorted
on approach of the anion[’01.
Tropylium halides cannot be dissolved in apolar solvents
at concentrations detectable by NMR spectroscopy: in
acetonitrile only the dissociated form can be detected.
On the other hand, 7-cyanocycloheptatriene exists completely undissociated and there is no migration of the cyanide group. Nevertheless, equilibria between 7-substituted
cycloheptatriene and their corresponding ions were observed, just as for the isothiocyanate, with the anions N,:
ONOe,and NCOe[’zl. Therefore those anions represent an
intermediate position and permit a study of ion recombinat i o n ~ [ ’ that
~ ] will contribute to our understanding of the
S,1 mechanism. The migration of these anions around the
cycloheptatriene ring can therefore be explained by ionization. To what extent other mechanisms (e. g. [1,5]- or
[3,3]-sigmatropic shifts) are involved remains to be investigated.
Received: July 19, 1973 [Z 891 IE]
German version: Angew. Chem. 85, 821 (1973)
[ I ] R . A . Snern, Accounts Chem. Res. 6, 46 (1973), and literature cited
therein.
121 R . Huisgen, Angew. Chem. 82, 783 (1970); Angew. Chem. internat.
Edit. Y, 751 (1970).
773
[ 3 ] H. Kessler, Angew. Chem. 82, 237 (1970); Angew. Chem. internat.
Edit. 9, 219 (1970).
[4] Prepared by reaction of tropylium tetrafluoroborate with sodium
thiocyanate. The IR spectrum of its ethereal solution shows a strong band
at 2060 cm-' characteristic of R-NCS compounds.
[ S ] The same effect was observed previously with tropylium azide [6].
[6] D. S. Wulfman, L. Durham, and C. E . Wulfman, Chem. Ind. (London)
1962,859.
[ 7 ] R. A . Hoffman and S. Forst%, Progr. NMR Spectrosc. 1, 15 (1966).
[8] Determined from the broadening of the tropylium signal according
to the approximation k = xbn (bn =line broadening by exchange) [9].
[9] I . 0.Sutherland, Annu. Rep. NMR Spectrosc. 4. 71 (1971).
[lo] A high barrier was also observed for dissociation of Meisenheimer
complexes [ I I]. Thesealsocorrespond to the system: aromatic compound
tx-.
[ I I] P . Cawny and H . Zollinger, Helv. Chim. Acta SO, 861 11967).
[12] Solvent: SOJCDCI, for NY,CD,CN for ONO- and NCO'.
[I31 C. D. Ritchie, Accounts Chem. Res. 5 , 348 (1972).
Influences of a-and fl-AlkylGroups on the Rearrangement of3-Butenyl Grignard Reagents: Astable Primary Cyclopropylmethyl Anion['*]
By Adalbert Maercker, Paul Giithlein, and Hermann WittrnayrC'1
Investigation of the rearrangement ( 3 a ) + ( 4 ) + ( 3 b )
with X=CI in tetrahydrofuran was only partly successful.
Certainly the tertiary Grignard reagent (3 a ) rearranges
quantitatively (half-reaction time ca. 30 h at 70°C) and
the cyclopropylmethyl compound ( 4 ) (ca. 0.07 %) can be
unambiguously detected but the equilibrium lies more than
99.9% on the side of the primary 3-butenyl Grignard
reagent (3 b ) (determined by gas chromatography after
esterification of the carboxylation products by diazomethane). Thence it follows that the two geminal methyl groups
d o in fact powerfully stabilize the three-membered ring,
yet the Thorpe-Ingold effect alone does not suffice to shift
the equilibrium in favor of the cyclopropylmethyl compound ( 4 ) .
Dedicated to Professor Gerhard Hesse on the occasion of
his 65th birthday
By isotopic labeling experiments Roberts et al.['l demonstrated that the a- and 0-carbon atoms of 3-butenyl
Grignard reagents gradually interchange positions:
( I a ) $ ( ] b ) (half-reaction time ca. 30 h at 27 T).A cyclopropylmethyl Grignard reagent (2) is assumed as an intermediate but its presence in the equilibrium mixture could
not be detected. If (2) is prepared at low temperatures
by an independent route, it rearranges quantitatively to
( I ) already at -24°C with a half-reaction time of 2h['].
The reason for this is the large ring strain in the cyclopropane (27 kcal/mol), which also explains why (2) cannot
be detected in the equilibrium ( I a)+(2)+(1 b ) : the
energy difference between ( 1 ) and (2) was estimated to
be 7 kcal/mol[21.
it
f l HC./ i H 2
+ BrMg-CHp
The energy difference between (2) and ( I ) can in principle
be reduced by stabilizing the three-membered ring of (2)
or by destabilizing the 3-butenyl Grignard reagent ( I ) .
The best chance of success lies in the combination of
these two possibilities.
We hoped to stabilize the cyclopropane ring by introducing
geminal alkyl groups (Thorpe-Ingold effect)I3 '1.
A possible method of destabilizing the open-chain
Grignard reagent (I) appeared to be the introduction
of alkyl groups on the r-carbon atom, i. e. at the carbanionic
center (cf. Ref. [7]).
[*] Priv.-Doz. Dr. A. Maercker, Dr. P. Guthlein, and H. Wittmayr
lnstitut fur Organische Chemie der Univcrsitiit Erlangen-Nurnberg
852 Erlangen, Henkestrasse 42 (Germany)
[**I This work was supported by the Deutsche Forschungsgemeinschaft
and the Fonds der Chemischen Industrie.
774
Complete success, however, followed introduction of four
methyl groups. In the equilibrium mixture ( 5 ) + ( 6 )
(X=CI) there is only one open-chain compound since the
x- and the 0-carbon atoms are equally substituted. This
tertiary Grignard reagent ( 5 ) rearranges already at room
temperature in tetrahydrofuran (concn. O . W . 4 mol/l)
to give more than 99.9% of the cyclic compound (6)
which, unlike ( 4 ) , cannot pass into a primary 3-butenyl
Grignard reagent by ring opening.
In (6), which is also accessible directly from (6) with
C1 in place of CIMg, we have the first completely stable
primary cyclopropylmethyl Grignard reagent [I'H-NMR :
r-methylene doublet at T = 10.9 (.
=I
7 Hz)]. Except for the
2,2,3,3-tetramethylcyclopropylmethyl anion (7), all the
stable cyclopropylmethyl anions previously known were
substituted at the carbanionic center, e. g. the benzhydryl
(S)['], the phosphane oxide (9)19] and the vinyl anion
( 10)' 0'.
i 9)
110)
Kinetic measurements (NMR and gas chromatography)
gave a half-reaction time of ca. 6 h for the equilibration
( 5 ) = ( 6 ) at 23"C, so the rearrangement is appreciably
faster than that of ( 3 a ) . Remarkable also is that preparation of ( 5 ) and (6) is possible only when X =C1. Reaction
Anguw. Chem. Inlrrnaf. Edit. 1 Vol. 12 (1973)
1 No. 9
Документ
Категория
Без категории
Просмотров
1
Размер файла
236 Кб
Теги
salt, tropylium, observations, direct, dissociation, equilibrium
1/--страниц
Пожаловаться на содержимое документа