close

Вход

Забыли?

вход по аккаунту

?

Electron Pyrolysis a New Method of Organic Microanalysis.

код для вставкиСкачать
ANGEWANDTE CHEMIE
International Edit’
VOLUME 5 - N U M B E R 9
AUGUST I966
PAGES 751 -856
Electron Pyrolysis, a New Method of Organic Microanalysis
BY PROF. H. SCHILDICNECHT
BASED ON STUDIES BY DR. F. ENZMANN, K. GESSNER, K. PENZIEN, F. ROMER,
AND DR. 0. VOLKERT
ORGANISCH-CHEMISCHES INSTITUT, UNIVERSITAT HEIDELBERG (GERMANY)
Organic compounds incubuted with tritiated water of specific activity 5 Cilml are specifically Ji-agmented. The J-active fragments can be detected in very small quantities by chrcmatographic separation. Chemical identification of these fragments provides information
about the original compounds, as has been shown by experiments with amino acids, esters,
lactones, amides, quinones, terpenes, nitrogen-sulfur heterocycles, and natirrul products.
1. Introduction
“The object and the chief aim of all chemistry is skilfully
to break down substances into their components, to
discover their properties, and to combine them in
various ways”. With these words, Scheele placed analysis in the forefront of all the chemist’s activities.
The earliest method used to decompose organic substances was simply to heat them, i.e. to pyrolyse them
Several methods are now known to supply the energy
leading to cleavage of chemical bonds. Thus we observed
that P-resorcylic acid is decarboxylated not only when
heated at 150°C, but also at room temperature in
tritiated water. We called this process Elektronenbrenzen
(“electron pyrolysis”) [I]; it was not known at that time
that often not only the P-radiation, but also the water
itself or its radiolysis products, such as hydrated electrons or OH-radicals [21, can be of use in the fragmentation. However, we were aware of the fact that the
unavoidable labeling with tritium greatly facilitates
detection of fragments formed only in traces, and is
often the only means of detection.
Succinic acid also is decarboxylated on electron pyrolysis, forming tritium-labeled propionic acid [31. Decarboxylation of mandelic acid yielded benzaldehyde
[l] H . Schildknecht, Kerntechnik 6, 249 (1964). We suggest the
application of the German term Elektronenbrenzen in English as
well, rather than “electron pyrolysis” or “fragmentation by
electron impact”.
[2] Basic Mechanisms in the Radiation Chemistry of Aqueous
Media, Radiat. Res. Suppl. 4 , 1-217 (1964).
[3] H . Schildknecht and 0 . Volkert, Z . analyt. Chem. 216, 91
(1966).
Angew. Chem. internat. Edit.
Vol. 5 (1966) / No. 9
as a result of the oxidation of the initially formed benzyl
alcohol by the oxidizing components of the tritiated
water. The formation of n-hexanal from n-hexanol and
of cyclohexanone from cyclohexanol in tritiated water
confirmed this assumption. All carbonyl compounds
could be detected as active 2,4-dinitrophenylhydrazones.
At the beginning of our investigations we could thus
already define electron pyrolysis as the breakdown of
organic molecules into smaller, readily detectable
complementary fragments involving reduction, oxidation, and labeling.
It was still necessary to determine whether it is possible
to establish fragmentation rules by means of which a
structure could be suggested on the basis of the fragments obtained. This question can now be partly
answered in the affirmative, though many investigations
will still be necessary to make electron pyrolysis a
generally useful method for the elucidation of the
structures of organic compounds.
2. Experimental Procedure
Samples varying in size from a few pg to several mg are
sealed in glass tubes some 5 cm long and having an internal
diameter of 3 mm, together with 2.5 t o 50 mg of tritiated
water of specific activity 5 Ci/ml. The solution or suspension
is left at or above room temperature for several days to a
few weeks, until the contents of the tube become discolored.
The products are separated chromatographically.
In thin-layer chromatography, the fractions are recorded by
means of a thin-layer scanner, and in gas chromatography
by a flow detector. This gives activity curves which resemble
spectra, the bands of which can often be assigned to chemi-
75 1
cally identifiable products. We have, therefore, included the
thin-layer chromatographic distribution of the fragments in
each figure showing "electron pyrolysis spectra". This shows
which fragments were detected chemically and which were
detected only on the basis of their radioactivity or of their
fluorescence. The tritiated fragments were also identified in
every case by comparative thin-layer chromatography on
Kieselgel G plates, using at least three different solvent
systems. The active spots were eluted from the plate and
chromatographed once more, and in most cases the ultraviolet and infrared absorption spectra of the eluted compound were determined.
Electron pyrolysis with tritium gas is carried out as follows:
of pyruvic acid; however, (11) may also be formed
directly from (10). As can be seen from the thin-layer
radio-chromatogram (Fig. l), the main product apart
As a precaution the tritium is bound to uranium powder.
Uranium hydride is decomposed at 450 "C and 10-3 mm Hg
to hydrogen and uranium powder. The tritium gas (3 Ci,
97.7 %) is adsorbed on the latter at room temperature; it is
regenerated as needed at 450 "C. The tritium is pumped into
a reaction vessel ( 5 ml) containing the substance to be
analysed. The powdered sample is continuously stirred in the
dark with a glass-coated magnetic wire.
3. Electron Pyrolysis of Known Compounds
3.1. Electron Pyrolysis of Amino Acidsf41
In the electron pyrolysis of glycine ( I ) we obtained
only two products, formed by simple cleavage of bonds
with no secondary reactions. In the paper radiochromatogram we found the ammonium salt of acetic
acid, which was used as an eluent, and in another
chromatogram acetic acid (2). Here and in subsequent
electron pyrolysis schemes, the bonds which are broken
are indicated by dotted lines. No indication is given
that part of the hydrogen in all compounds is replaced
NH,
0 .
t
H3N*CHz-COO'
J
(3)
F-
HOOC- H CHz-COOH
OH
(6)
(I)
J.
--t
CH3-COOH
I
(2)
Fig. 1. Thin-layer radio-chromatogram of the solution obtained on
electron pyrolysis of 5 yg of alanine (7). Solvent system: ethanol/NH,/
H20 (80:4: 16 v/v).
1, alanine ( 7 ) ; 2, lactic acid (10) ; 3, propionic acid ( 8 )
pyruvic acid
( 1 1 ) ; 4, alanine ethyl ester.
+
from (11) was propionic acid (8). The cleavage of the
C-N bond is understandable in view of the fact that
in amino acids in solution, the distance between the
protonated N atom and the C atom is 1.50 A, as
compared with the normal value of 1.32 8, [61. Thus the
C-N bond is weakened in the zwitterion. If the compound is not in the form of the zwitterion, as in the gas
HOOC -CHz-CHz- COOH
(5)
HO - CHz- COOH
(4)
by tritium. This labeling is a direct result of the saturation of the CH2-CO2H radical with tritium atoms.
O H (OT) radicals also can react with CH2-C02H, as
is shown by the formation of glycolic acid (4) and
malic acid (6). In agreement with literature reports [51 on
the radiolysis of organic compounds in water with
y-rays, the formation of succinic acid ( 5 ) can be explained only on the basis of intermediate free-radical
fragments.
In view of these results on glycine, we were not surprised
to find lactic acid (10) and pyruvic acid (11) on chromatographic analysis of the solution obtained on electron pyrolysis of alanine (7). According to Garrison 151,
OH radicals are also involved in the radiolytic formation
Fig. 2. Thin-layer radio-chromatogram of the solution obtained on
electron pyrolysis of 5 yg of aspartic acid (1.2). Solvent system: ethanol/
NH, (7:3 vlv).
1 , malic acid ( 6 ) ; 2, aspartic acid (12) ; 3, succinic acid ( 5 ) ; 4, 0-aminopropionic acid (13) : 5, alanine (7) : 6, propionic acid ( 8 ) ; 7, pyruvic
acid (II) ; 8, diethyl aspartate.
Diploma Thesis, Universitat Heidelherg, 1966.
Radiat. Res. Supp. 4, 158, (1961).
[6] R. B. Corey and J . Donohue, J. Amer. chem. SOC. 72, 2899
(1950).
[4] K . Gepner,
[ 5 ] W . M . Garrison,
7 52
Angew. Chem. internat. Edit.
VoI. 5 (1966) 1 No. 9
phase in the mass spectrometer, cleavage occurs preferentially in the next-weaker C-CO bond [71. However,
decarboxylation with formation of ethylamine (9) can
also be observed to a minor extent in solution.
thin-layer radiochromatogram (Fig. 3), relatively many
primary cleavage products are formed.
Decarboxylation is particularly pronounced in the
electron pyrolysis of aspartic acid (12). As can be seen
from the corresponding thin-layer radio-chromatogram
(Fig. 2), alanine (7), P-aminopropionic acid (13), and
propionic acid (8) could be unambiguously detected
even when (12) was simply allowed to stand at room
temperature with tritiated water. This result emphasizes
clearly the specific action of the tritiated water.
,
The products again contained succinic acid (5), formed
by reductive deamination, and propionic acid (8),
formed from the succinic acid, while malic acid (6) and
pyruvic acid (11) were formed by oxidation; (8) may
have been formed also from (7).
HOOC - CH,- CHOH- COOH
HOOC -cH,- CH,-COOH
f
7
(6)
(s)
HOOC -cH,- CH+COO@ CH,-CH,-COOH
GH3
J
CH3- CO - COOH
(12)
.1
"OOC
- CHz- F H 2
(13)
(11)
%H3
I
t
CH3- CH - COOo
GH3(7)
In the electron pyrolysis of proline (14), the ring was
opened in a particularly simple manner, and ( I 4 ) was
degraded to valeric acid (16). Activity measurements
showed that the yields were lower than in the pyrolysis
of alanine (7). The compounds containing 5-membered
rings (as is also shown in mass spectrometryL81) are
probably more staole than open-chain compounds, or
else chain degradation may be preceded by hydrolytic
ring cleavage. Thus tryptophan (21) is degraded slightly
to indole (18) when simply heated with water. The
electron pyrolysis of (21) also leads to the formation of
(It?), which, when itself pyrolized, gives anthranilic
acid (20), probably via (19).
r
1
,
I
I'
f
1
0
1,
cm
12
-8
[153351
Fig. 3. Thin-layer radio-chromatogram of the solution obtained o n
electron pyrolysis of 10 v g of tryptophan (211. Solvent system: butanoll
NHJ ( 7 : 3 v/v).
I , starting peak; 2, glycine ( I j ; 3, alanine ( 7 j ; 4, tryptophan ( 2 1 ) ;
5, 3-indolyl-$propionic acid (23) ; 6, tryptanline (241 ; 7, indole (18) ;
8, skatole 1221.
Radiolysis of (21) with X-raysr9l also yields several
cleavage products resulting from decarboxylation and
changes in the carbon skeleton. No deamination is
found, as on mass spectrometry rsl; the strongest mass
peak occurs at mle = 130, suggesting that fragmentation
tends to stop at the skatole (22) stage. The same result
is found in electron pyrolysis, where the greatest
activity, apart from the band for the starting material
(21), was obtained for the skatole fraction. The complementary fragment, glycine ( I ) , was also found. The
structure fif the starting material can, however, be
deduced from the fragments (18) and (7), as well as
from the above fragments.
+
(7)
+
J
H,C-CH-COO@
@AH3
CH,- CH,- COOH
H
(23)
0
H,N - CH,- COO@
(1)
As in the case of tryptophan (21), we obtained indole
( I s ) and skatole (22) as well as acetic acid in the
electron pyrolysis of the growth hormone 3-indoleacetic acid.
3.2. Electron Pyrolysis of Esters, Lactones,
and AmideW"
This fragmentation is quite unimportant in the electron
pyrolysis of tryptophan (21), since according to the
[71 G. Junk and H . Svec, J. Amer. chern. Soc. 85, 839 (1963).
[81 K . Biernunn: Mass Spectrometry. McGraw-Hill, New York
1962, p p . 260-296.
Angew. Chem. internat. Edit.
/ Vol. 5 (1966) / No. 9
The principal product in the electron pyrolysis of
benzyl benzozte was benzoic acid which, like the former,
was strongly labeled, but which could not be detected
191 G. Peter and B. Rujewsky, Z. Naturforsch. 186, 1 1 0 (1963).
1101 H . Schildknecht and F. Enzmann, 2. analyt. Chem., in press.
753
chemically on the thin layer either with Bromocresol
Purple or with Bromophenol Blue. p-Hydroxybenzoic
acid was formed in addition, though this was naturally
absent when tritium gas was used instead of tritiated
water; benzaldehyde, on the other hand, was formed
in both cases. Owing to the high radioactivities of the
fragments, it was possible to characterize as little as
0.1 pg of benzyl benzoate.
I
1
1000 -
t
0
12
8
L
2
0
.......
.. ...
i
:
3.. 0 Q
,
(153351.
Fig. 5. Thin-layer radio-chromatogram of the solution obtained o n
electron pyrolysis of 5 vg of dibenzyl succinate in tritiated water with
added UO2. Solvent system: ethanol/ammonia/water (100: 16: 12 v/v).
1, succinic acid; 2, propionic acid; 3, dibenzyl succinate; 4, benzyl
propionate.
Fig. 4. Thin-layer radio-chromatogram of the electron pyrolysis prodCHZOH
ucts of 4 v g of benzyt benzoate in tritium gas. Solvent system: benzene/ I
CHOH
dioxane/glacial acetic acid ( 9 0 : 2 5 : 4 v/v).
I
1, benzoic acid; 2, benzaldehyde; 3. benzyl benzoate.
CHzOH
-
(29)
In the formulation of the radiolytic reaction, it is
conceivable that the 0-benzyl bond is broken, in
agreement with considerations by McLachlaii [11,121 and
with Emery's mass-spectrometric results [131, but this
does not explain the high activity of the benzoic
acid. It could be due to labeling in the phenyl nucleus,
which could in turn point to the presence of a resonancestabilized benzoyl radical. The radical combines with
hydrogen to form benzaldehyde, which is partly oxidized
to benzoic acid by the oxygen in the tritium gas.
The electron pyrolysis of phenyl salicylate in tritiated
water or tritium gas yields salicylic acid, and phenol
has also been detected. The acid is evidently not formed
by hydrolysis, but by nucleophilic displacement at the
C=O carbon atom by thermal electrons.
Attempted electron pyrolysis of dibenzyl succinate
under the above conditions was unsuccessful at fist.
The radio-chromatogram shown in Figure 5 was obtained only after the addition of uranium dioxide
powder to the reaction mixture. The ester was then
degraded, via succinic acid, to propionic acid or benzyl
propionate. The benzyl group again gave benzaldehyde,
which could be detected as the 2,4-dinitrophenylhydrazone.
The large number of fragments formed in the electron
pyrolysis of tributyrin (25) with tritiated water can be
seen from Figure 6. A surprising feature is that the
butyric acid residue is degraded primarily to the acetic
acid residue, as is shown in the following scheme.
[ll] A . M . McLachlan, J . Amer. chem. SOC.82, 3309 (1960).
[I21 A . M . McLachlan, J. org. Chemistry 29, 1598 (1964).
[13] E. M . Emery, Analytic. Chem. 32, 1495 (1960).
754
2
A
0
4
cm
- 8,
12
I
Fig. 6. Thin-layer radio-chromatobram of the solution obtained on
electron pyrolysis 01 10
of tributyrin ( 25) . Solvent system: benzene/
dioxane (90: 25 v/v).
I , monobutyrin (26) ; 2, butyric acid (28) ; 3, acetic acid ( 2 ) ;4, dibutyrin
(27) ; 5, triacetin (30); 6, tributyrin ( 25) .
Though we were unable to distinguish chromatographically
between 1,3- and l,Zdibutyrin, it may be assumed from the
secondary products that the electron pyrolysis of (25) yields
1,3-dibutyrin (27). Glycerol (29) also could be separated
from the other components in a second chromatogram. The
Aiigew. Chem. internat. Edit. i Vol. 5 (1966)
i No. 9
radio-chromatogram in this case contains practically no
activity maximum, since the labile tritium is almost completely exchanged during the development of the chromatogram. However, glycerol can be made visible with K M n 0 4 /
HzS04.
y-Butyrolactone (31) is only partially hydrolysed by
water to hydroxybutyric acid (32), and only on prolonged heating. With tritiated water, however, (32) is
formed even at room temperature and in higher yields.
In addition to (32), propionic acid (8) is formed. A C3
fragment is formed also in the mass spectrometry of
y-butyrolactone, and is recorded as the C3H40e
ion [141.
every case. This agrees with Vermeil's findings [181,
according to which hydroquinone is formed as a principal product on y-irradiation of p-benzoquinone in
water. The formation of hydroxyquinones, which react
further to humic acids, can be avoided if benzene,
toluene, or ethanol is added to the tritiated water as a
solubilizing agent. Moreover, when an organic solvent
is present, the temperature can be raised from 40 to
80°C without danger of formation of large quantities
of polymers during the electron pyrolysis.
4fter electron pyrolysis of n-propyl-p-benzoquinone,
thin-layer chromatography revealed the presence of
toluquinone (37) in addition to the expected hydroquinones. Benzoquinone was detected only as hydroquinone. After electron pyrolysis of thymoquinone (39),
only half of the active pyrolysis product could be
stained on the thin layer by spray reagents (Fig. 7). The
!n order to avoid hydrolysis, sulfanilamide (33) was
incuoated for 10 days with 3 Ci of tritium gas. The
fragmentation led to aniline (34), ammonia, and probably sulfur dioxide.
3.3. Electron Pyrolysis of Quinonescl5J
0
The behavior of quinones on electron pyrolysis was of
interest to us, because the defe;lsive substances of
arthropods
and also of plants [171, which we were
studying, quite often are p-benzoquinones. Particularly
when only a few pg of such a compound are available
for analysis, the nature and position of side chains in
alkylated p-benzoquinones can be found only with great
effort. We therefore applied the new fragmentation
procedure to quinones of this type. Already in the first
example studied, ethyl-p-benzoquinone (35), we Gbserved that the ethyl group is split off in stages; benzoquinone and toluquinone were found by thin-layer
chromatography, an3 tritium-labeled ethane by radiogas chromatography.
'
cm--
8
12
Fig. 7. Thin-layer radio-chromatogram of the solution obtained by
electron pyrolysis of 40 yg of 2-isopropyl-5-methyl-p-benzoquinone
(39).
Solvent system: petroleum ether (60-70 "C)/ethyl acetate (6:4 v/v).
1, humic acids; 2, dithymoquinone; 3, toluhydroquinone; 4, 2,5-dimethylhydroquinone; 5, thymohydroquinone; 6, toluquinone (37) ;
7, 2,5-dimethyl-p-henzoquinone(40) ; 8, thymoquinone (39).
presence of toluquinone (37) and 2,s-dimethylquinone
(40) showed that the isopropyl side chain had been
split off and fragmented. 2-Ethyl-5-methyl-p-benzoquinone must have been formed as intermediate, the
ethyl group then being further degraded, as described
above. The methyl group, 011 the other hand, appears
to be more stable, as was shown by the electron pyrolysis
of toluquinone (37) itself. Thus degradation occurs
mainly at the longer and possibly branched side chains.
In the electron pyrolysis of phenyl-p-benzoquinone,
even the phenyl group is readily split off, with formation
of benzoquinone.
In addition to the tritium-labeled quinones (35), (36),
and (37), labeled hydroquinones are also formed in
[14] L. Friedman and F. A. Long, J . Amer. chem. SOC.75, 2832
(1953).
[I 51 F. Romer, Diploma Thesis, Universitat Heidelberg, 1966.
[16] H . Schildknecht, Angew. Chem. 75, 762 (1963); Angew.
Chem. internat. Edit. 3, 73 (1964).
[17] Nachr. Chem. Techn. 12, 177 (1964).
Angew. Chem. internot. Edit. Vol. 5 (1966)
! No.
9
[18] C. Vermeil, G . Roqurt, and L. Salomon, J.
60, [5], 659 (1963).
Chim. physique
755
In the study of the defense substances of arthropods we
had encountered methoxy-p- benzoquinones
which
we also examined by electron pyrolysis. We found that
the methoxy group is readily eliminated. This is particularly significant in the elucidation of the structures of
complex methoxyquinones, which can be converted by
electron pyrolysis into simpler, and hence in most cases
known, alkqlated p-benzoquinones. Not only niethoxyp-benzoquinone (42), but also 2,6-dimethoxy-p-benzoquinone (41) gave labeled benzoquinone (36), again
together with the corresponding hydroquiriones, when
treated with tritium (cf. Fig. 8).
3.4. Electron Pyrolysis of Menthol and Thymol[*ol
We used menthol (45) purified by zone melting. The
starting band in Figure 9 is due to adsorbed tritiated
water and hydroxylated m e n t h d Band 2 was assigned
to cis-3-methylcyclohexanol (49), which we prepared
by reduction of 3-methylcyclohexanone with LiAlH4.
The trans-isomer (50) was also formed in the electron
I
0
'
I
''
cm-
12
8
' ' 1 ' 1
I
Fig. 9. Thin-layer radio-chromatogram of the solution obtained on
electron pyrolysis of 15 &g of menthol (451. Solvent system: chloroformlpetroleum ether (60--70 "C)/ethanol (60: 37: 3 viv).
1, hydroxylated menthol; 2 , cis-3-methylcyclohexanol ( 4 9 ) ; 3, Iranr-2cis-5-dimethylcyclohexanol 1 5 1 ) ; 4, menthol ( 4 5 ) ; 5 , neomenthol (461 ;
6, nienthone 147) ; 7, 3,7-dimethyloctanal 152).
'
pyrolysis, but with weaker labeling. Band 3 corresponds
to tra1zs-2-cis-5-dimethylcyclohexanol
(51). The formation of this isomer during the electron pyrolysis is in
agreement with Hiickel's findings 1211, according to which
(51) is the predominant species in equilibrium mixtures
with its isomers. It corresponds to menthol in its configuration. Owing to the equality of the Rf values of
menthol and isomenthol, band 4, and of neomenthol
and neoisomenthol, band 5, it was not possible to assign
these bands to a particular isomer. A second chromatogram (solvent system : methylene chloride) showed the
7
0
cm-
8
12
Fig. 8. Thin-layer radio-chromatogram of the solution obtained o n
electron pyrolysis of 50 yg of 2,6-dimethoxy-p-benzoquinone(431.
Solvent system: chloroformlpetroleum ether (60-70"C)Iethanol
(60:37:4 vlv).
I , polymer; 2, hydroquinone; 3, rnethoxyhydroquinone; 4, 2,6-dimethoxyhydroquinone; 5 , 2,6-din.ethoxy-p-benzoquinone(411 ; 6, methoxyp-benzoquinone ( 4 2 ) ; 7, p-benzoquinone (361 ; 8, second solvent front.
In the electron pyrolysis of the isomers 2-methoxy5-methyl-p-benzoquinone (43) and 2-methoxy-6-methylp-benzoquinone (44), only toluquinone (37) is found in
addition to benzoquinone. Thus the methoxy group is
removed more readily than the methyl group.
/
ISO)
(36)
(51)
(37)
[201 H.Schildknecht and K.Penzien,Z.analyt.Chern.
219,102(1966).
[19] H . Schildknecht and K. H . Weis, Z. Naturforsch. 166, 810
(1961).
756
[21] W . Hiickel, H. Feltkamp, and S. Geiger, Liebigs Ann. Chem.
637, 1 (1950).
Angew. Chem. internat. Edit.
Vol. 5 (1966) / No. 9
presence of isomenthone (48), menthone (47), and 3,7dimethyloctanal(52) among the products of the electron
pyrclysis of menthol.
As in the fragmentation of thymoquinone (39), the isopropyl
group of menthol (45), too, is split off. This cleavage is
probably favored by the high stability of the isopropyl radical
presumably formed. The fact that no removal of the methyl
group attached to the ring in menthol was observed supports
this interpretation. It is still remarkable that 2,5-dimethylcyclohexanol (51) is found, but no 2-ethyl-5-methylcyclohexanol,
since it is unlikely that the cleavage of two C-C bonds takes
place simultaneously. Thus Hentz [*21 found, inter alia,
benzene, ethylbenzene, and toluene in the 6oCo-y-radiolysis
of isopropylbenzene. The 3,7-dimethyloctanal (52) obtained
from menthol in tritiated water is probably formed because
the bond between the C atom carrying the hydroxyl group
and that carrying the isopropyl group is the weakest of the
C-C bonds in the ring. The dissociation energy of a C-CO
bond is smaller than that of a C-C bond. As far as the
mechanism is concerned, the dealkylation is best explained
by the attack of a solvated electron. However, the oxidation,
reduction, and ring cleavage reactions may be initiated by
the other radiolysis products of water, particularly by the
OH radicals.
For R3 = H, the presence of the 5-membered ring in
(61) can be readily demonstrated by deamination, if
detection of the amino group by the formation of a
Schiff base is not sufficient in itself. However, this proof
is no longer possible when R3 is an organic group.
For R3 = CO-CsHs, Sato [241 formulated a 6-membered
ring, since compounds of this type are soluble in alkali
and this property can be explained by thiol formation.
However, Ege [251 found that solubility in alkali is not a
criterion for the 6-membered cyclic structure. The
5-membered cyclic structure could not be proved either
chemically or spectroscopically (NMR, ultraviolet, infrared) in the case of 3-anilino-4,5-dimethyl-2H-l,3thiazol-2-one (62).
The results of the electron pyrolysis of thymol (53)
agreed closely with the fragmentation observed in the
case of menthol. Of particular interest to the nitural
-
&H
&,, 6oH
... ...
4
1
OH
(56)
t
products chemist may be the fact that the isopropyl
group again is successively and preferentially split off.
4. Elucidation of the Structures of Thiazoles and
Thiadiazines by Electron Pyrolysis 1231
Tne derivatives (59) obtained from cr-halogenoketones
(57) and N-substituted hydrazides Jf dithiocarbonic
acid (58) can undergo ring closure to form either a
3,4-dihydro-2 H-l,3,4-thiadiazine-2-thione(60) or a
1,3-thiazole-2-thione (61).
Rl-CO-CHBr-R'
+
HS-CS-NH-NH-R3
(57)
1
(58)
P3
J(59)
The electron pyrolysis of (62) should lead to the formation of aniline. In fact, the two complementary
fragments aniline and 4,5-dimethyl-2H-l,3-thiazol-2one (6.3) could be detected both chemically and with the
thin-layer scanner in the thin-layer radio-chromatogram
of the solution obtained on electron pyrolysis of 20 pg
of (62) (cf. Fig. 10).
.-c
.
E
m
c
3
u
0
c
I
1
0
I
I
1,
crn
I
-8
1
12
Fig. 10. Thin-layer radio-chromatogram of the solution obtained o n
electron pyrolysis of 20 vg of ( 6 2 ) . Solvent system: CHCII/CH,OH
(95: 5 v/v).
I , uncertain; 2, 4,5-dimethyl-2H-l,3-thiazol-2-one
(63) ; 3, aniline;
4, 3-anilino-4,5-dimethyl-2H-1,3-thiazol-2-one
(62).
Electron pyrolysis of the compounds (64) and (65) also
confirmed the 5-membered cyclic structure established
by Ege, since (66) and (67) were detected in the thinlayer chromatogram. Similarly, the electron pyrolysis of
3-benzoylamido-4-phenyl-2
H-l,3-thiazol-2-thione (68)
I
NH - R3
Angew. Chem. internat. Edit. / Vol. 5 (1966)
[22] R. R. Hentz, J . physic. Chem. 66, 1622 (1962).
[23] G. Ege, K . GeJner, and H . Schildknecht, unpublished.
[24] T. Sato and M . Ohta, J . pharmac. SOC. Japan (Yakugakuzasshi) 75, 1535 (1955); Chem. Abstr. 50, 10727 (1956).
[25] G. Ege, unpublished.
/ No. 9
757
(64), R
'
: C,H5; R2: H; R3: HN-CO-C,H,
(65). R' = R2: CH3;
R3: HN-CO-C,H5
with methanol as a solubilizing agent led to the formation of benzamide (70) and 4-phenyl-2 H-l,3-thiazol2-thione (69).
H5c6&
NH - CO - C6H5
HZN-CO-C6H5
(70)
--*
H
H
In the electron pyrolysis of thiazoles and thiadiazines,
the mode of fragmentation remarkably often led to
conclusive structure proofs, conclusive in that the
thiadiazines do not rearrange to the corresponding
5-membered compounds, but are degraded into as yet
unidentified fragments. It is likely that during the
electron pyrolysis the hydrolytic action of water is superposed on the specific fragmenting action of the hydrated
electron.
In agreement with the observations made in the electron
pyrolysis of other substituted p-benzoquinones, the
alkyl side chain fortunately is broken down only to such
an extent that the positions of the substituents on the
quinone nucleus of primin are fixed by the remaining
methyl group and by the methoxy group still present.
The alkyl group is satisfactorily characterized by the
detection, by radio-gas chromatography, of n-butane
among the volatile components.
It was mentioned at the beginning that, hopefully,
structures can be proposed by properly arranging the
electron-pyrolysis fragments like the pieces of a jig-saw
puzzle. This hope may yet be realized. Certainly, the
identification of the fragments of the arthropod alkaloid
glomerine [281 has required much more time than did
the rational combination to give the proposed structure [291. Glomerine, a 1,2-dimethylquinazol-4-one(741,
was fragmented by electron pyrolysis to anthranilic
acid (20), N-methylanthranilic acid (76), N-methylanthranilamide (75), and acetic acid (cf. Fig. 11).
5. Elucidation of the Structures of Natural
Products by Electron Pyrolysis
The results obtained with model substances encouraged
us to use electron pyrolysis in order to confirm structures proposed for unknown natural products.
For example, we had found by a comprehensive spectroscopic analysis that the poison of the primrose,
which Karrer had named primin, is a methoxy-n-pentylp-benzoquinone [261. Several synthetic quinones had to
be used as reference substances in order to show that
primin is 2-methoxy-6-n-pent5 1-p-benzoquinone(71) [27J.
We later found that the two possible isomers can be
easily distinguished by electron pyrolysis, since the
known compounds 2-methoxy-5- and 2-methoxy-6methyl-p-benzoquinones, (43) and (44), are formed as
fragments. This is shown in the following scheme:
8
(44)
[26] H . Schildknecht, I. Bayer, and H . Schmidt, Z. Naturforsch.
in press.
[27] H. Schildknecht and H. Schmidt, Z. Naturforsch., in press.
758
I
I
0
L
rm
-8
12
Fig. 1 1 . Thin-layer radio-chromatogram of the solution obtained on
electron pyrolysis of 5 yg of glomerine (74). Solvent system: petroleum
ether (60-70 'C)/ether/pyridine (1: 1 : 1 v/v).
1 , unidentified; 2, anthranilic acid ( 2 0 ) ; 3, unidentified; 4, acetic acid
(2) ; 5, N-methylanthranilamide (75) ; 6 , N-methylanthranilic acid (76) ;
start, glomerine (74).
The synthesis of glomerine was a reversal of the fragmentation, since anthranilic acid was converted via (76) into (75),
which in turn was converted into (74) by means of acetic
acid. The biosynthesis of glomerine probably also starts from
anthranilic acid.
The composite proanthocyanidins [301 are condensation
or dehydrogenation products of two flavonoids, a monomolecular proanthocyanidin, and a catechol. The two
halves of the molecule may be linked either by an ether
oxygen, by a C-C linkage, or by a ketal linkage. In
order to clarify the structure, therefore, Weinges and
Ebert 1311 examined model substances corresponding to
the composite proanthocyanidins with a C-C linkage,
and attempted to achieve an informative cleavage by
[28] H . Schildknecht, K . H . Weis, W . F. Wenneis, and (1. Maschwitr, Z . Naturforsch. 21b, 121 (1966).
[29] H. Schildknecht and W. F. Wenneis, Z . Naturforsch. 21b,
5 5 2 (1966).
[30] K . Freudenberg and K . Weinges, Tetrahedron Letters 1961,
267.
[31] W . Ebert, Diploma Thesis, Universitat Heidelberg, 1965.
Angew. Chem. internat. Edit.
/ Vol. 5 (1966) / No. 9
0
fore be concluded that the interaction leading to fragmentation occurs between the molecule and an energytransfer agent. the energy of which is comparable to the
bond energy. Whether the energy is transferred via H
atoms, OH radicals, electrons, radiation, or even excited
water molecules is of no importance to the above
argument. Nevertheless, efforts will. be made to find a
mechanistic interpretation of electron pyrolysis, simply
to permit the prediction of the course of fragmentation.
The growing understanding of the radiolysis of organic
compounds in aqueous solution with X- and y-rays is
particularly valuable in this connection.
0
I
CH3
(75)
( 74)
1
chemical methods. In the presence of acids, however,
2,4,6,3’,4‘-pentamethoxydiphenylmethane(77) was broken down to phloroglucinol trimethyl ether (78) and
veratrole (79),with loss of the central methylene group.
Consequently, the substitution position in the veratrole
residue could not be established in this way. In the
thin-layer radio-chromatogram of the solution obtained
on electron pyrolysis of (77), on the other hand, one
fraction could be satisfactorily assigned to homoveratrole (80).
( 78)
- te -BREN2
CH3
.
CH,
I
uLn3
OCH3
(80)
(77)
OCH,
(79)
6. Prospects
In all chemical methods of structural elucidation it
should be known which bonds in the unknown molecule
are attacked by the reagent. I n reactions with tritiated
water, the attack usually takes place on the weakest
bonds, and C-N, C-C, C-0, and C-H bonds can be
broken oxidatively and reductively. Of great advantage
in this case is that part or all of the alkyl side chain is
removed from the basic structure, e.g. a quinone nucleus, so that infmmation is obtained not only about
the nature of the side chain, but also about the substitution site.
The selectivity observed in the fragmentation by electron pyrolysis is surprising, since the P-decay energy of
tritium is higher by a factor of l o 3 than the dissociation
energies of the above-mentioned bonds. It must there-
Angew. Chem. internat. Edit.
VoI. 5 (1966) No.9
We believe that solvated electrons may be assumed to
be particularly important in the fragmentations observed by us. Consequently, the fragmentation in the
mass spectrometer often can be used for comparison.
However, whereas only one half of the complementary
fragments is recorded in mass spectrometry, both parts
can be expected to be detected in electron pyrolysis.
Further, the fragments formed in electron pyrolysis are
“frozen” by saturation with H (or T) atoms or OH (OT)
radicals and by reaction with the tritiated water, whereas
the primary radicals formed in the mass spectrometer
are stabilized intramolecularly, with further fragmentation. This often leads to a spectrum with a large number
of peaks. Moreover, it is not always possible to distinguish between isomeric ions by mass spectrometry,
whereas after electron pyrolysis the isomeric fragments
can be separated by chromatography and identified.
Concerning the future of radio-gas chromatography, we
hope that it will not only become possible to detect all
fragments as a result of the development of more
sensitive radiation detectors, but also that these hill
often be sufficiently well characterized by their retention
times alone. The long incubation periods ranging fram
a few days to some weeks will then no longer be necessary if tritiated water having a specific activiiy of
25 Ci/ml instead of 5 Ci/ml is used. This requires particularly careful work, which can be guaranteed better in
gas chromatography than in thin-layer chromatography.
These experiments, sometimes not entirely without danger,
and the unconventional and, to an organic chemist,
somewhat unfamiliar techniques required great diligence
and idealism on the part of my doctorate candidates, for
which I express my sincere thanks. The investigations
would not have been possible without the generous support
of the Ministeriuni fur wissenschaftliche Forschung the
Deutsche Forschungsgemeinschaft, and the Fonds der
ChenTischen Industrie. W e are grateful to Prof. W . Huckel
and to the Dragoco- Werke for the donation of chemicals.
Received: June 14th, 1966
[A 533 I € ]
German version: Angew. Chem. 78, 481 (1966)
Translated by Express Translation Service, London
759
Документ
Категория
Без категории
Просмотров
0
Размер файла
751 Кб
Теги
organiz, method, pyrolysis, electro, new, microanalysis
1/--страниц
Пожаловаться на содержимое документа