The photocyclization of arylpropynes

MISSING IMAGE

Material Information

Title:
The photocyclization of arylpropynes
Physical Description:
ix, 111 leaves. : ill. ; 28 cm.
Language:
English
Creator:
Kulig, Martin Joseph, 1945-
Publication Date:

Subjects

Subjects / Keywords:
Arylpropynes   ( lcsh )
Ring formation (Chemistry)   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Bibliography: leaves 108-110.
Statement of Responsibility:
By Martin Joseph Kling.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000577574
notis - ADA5269
oclc - 13990188
System ID:
AA00003955:00001


This item is only available as the following downloads:


Full Text





















THE PHOTOCYCLIZATION OF ARYLPROPYNES


By



MARTIN JCSF2P 77.rI














A DISSERTATION PRESENTED TO THE GRAiDUATE
COUNCIL OF THE UNIVERSITY OF FLO.RIDA IN PARTIAL.
FULFILLMENT OF THE REQUIT..IEE:lTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY









UNIVERSITY Of ,L'II9A
1973








































DEDICATION


To My Farancs















ACKNOWLEDGQ1ENTS


It is with appreciation that the writer expresses his gratitude to

Dr. Merle Battiste for his influence and direction in the planning and

completion of this project. He would also like to thank Dr. J. Deyrup,

Dr. R. Isler, Dr. W. Person and Dr. P. Tarrant for their guidance and

interest. Finally the writer would like to express his gratitude for

the help he received from his friends Jim Hcrvath, Bud Mihal, Stan

Weller, Pete Wentz and Mike Williams.





















TABLE OF CONTENTS


ACKNOWLEDGMENTS . .


LIST OF TABLES . .


LIST OF FIGURES .. .....


Chapter


I INTRODUCTION .


II SYNTHESIS AND MASS SPECTRA


III RESULTS AND DISCUSSION .


IV EXPERIMENTAL .


BIBLIOGRAPHY . .


BIOGRAPHICAL SKETCH .


Page


. . iii


. . v


. . vi


. . .


. . .



. . .


. . .


. . .
















LIST OF TABLES


RELATIVE PRODUCT DISTRIBUTION IN THE IRRADIATION
OF TETRAPHENYLPROPYNE . .

PRINCIPAL FRAGMENT IONS IN THE MASS SPECTRA OF
(20), (21), AND (22) . .

PRINCIPAL FRAGMENT IONS IN THE MASS SPECTFA OF
(35), (36), AND (26) . .

FRAGMENTATION ION (m/e 267/265) RATIO .

PRINCIPAL FRAGMENT TONS IN THE MASS SPECTRA OF
(43), (44), AND (45) . .

RELATIVE PRODUCT DISTRIBUTION IN THE IRRADIATION
OF (42) . . .

ULTRAVIOLET tiSORPTION OF REACTIVE PROPYNES .


Table

I


II


III


IV

V


VI


VII


Page


. 5


7


. 18

. 26


. 29


. 49

. 50















LIST OF FIGURES


Figure Page

1 NMR SPECTRUM OF THE PHOTOPRODUCT FROM (32) 90

2 NMR SPECTRUM OF AUTHENTIC TRANS-1,2,3-TRIPHENYL-
CYCLOPROPANE (48) . .... .91

3 NMR SPECTRUM OF TRANS-1,2,3-TRIPHENYL-1-CHLORO-
CYCLOPROPANE (47) . .... 92

4 MR SPECTRUM OF 3-METHYL-1,2,3-TRIPHENYLCYCLO-
PROPENE (36) . .... .93

5 NMR SPECTRUM OF 3-METHYL-1,2-DIPHENYLTEDENE (49). 94

6 NMR SPECTRUM OF THE DIMER FROM 3-MZTITL-1,2-DI-
PHENYLINDENE (50) . .... 95

7 NMR SPECTRUM OF THE THERMOLYSIS PRODUCTS FROM 3-
METHYL-1,2,3-RIPHENYLCYCLGOROPENE (36) 96

8 NMR SPECTRUM OF THE THERMO.LYSIS PRODUCTS .7ROM 3-
PHENYLETHYNYL-1,2,3-TRIPHENYLCYCLOPROPENE (41). .. 97

9 NMR SPECTRUM OF STANDARD SAMPLES OF (45), (42),
AND (44) . . .98

10 NMR SPECTRUM OF THE PHOTOPRODUCT FROM 3-P-
ANISYL-1,3,3-TRIPHENYLPROPYNE (42) . 99

11 NMR SPECTRUM OF THE PHOTOPRODUCT FROM 1-P-ANISYL-
3,3,3-TRIPHENYLPROPYNE (43) . .. 100

12 NMR SPECTRUM OF THE PHOTOPRODUCTS FROM 3-P-
ANISYL-1,2,3-TRIPHENYLCYCLOPROPENE (44) .. 101

13 NKR SPECTRUM OF THE PHOTOPRODUCT FROM 1-P-
ANISYL-2,3,3-TRIPHENYLCYCLOPROPENE (45)) ..... 102

14 NMR SPECTRUM OF O-ANISOYL-O-BENZOYLBENZENE (59) 103

15 UV SPECTRUM OF TETRAPHENYLPROPYNE (20) ..... 104

16 UV SPECTRUM OF 4,4-DIMETHYL-1,1,1--TRIPHENYL-2-
PENTYNE (27). . . 105









Figure Page

17 UV SPECTRUM OF 9-PHENYL-9-PHENYLETHYNYLFLUORENE (40) 106

18 UV SPECTRUM OF 1-a-NAPTHYL-3,3,3-TRIPHENYL-
PROPYNE (54) .............. .... .... 107


vii
















Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial
Fulfillment of the Requirements for the Degree of Doctor of Philosophy




THE PHOTOCYCLIZATION OF ARYLPROPYNES



By


Martin Joseph Kulig

March, 1973


Chairman: Dr. Merle Battiste
Major Department: Chemistry

The primary interest of this study was the feasibility of preparing

ary.cyclopropenes from the irradiation of arylpropynes. Previous work

indicated that cyclopropanes were formed by the irradiation of olefins

in a di-r-methane rearrangement. An added complication wouLd be the

photochemical liability of the cyclopropenes themselves, since photo--

isomerizaticn of arylcyclopropenes to indenes has been previously

reported.

A secondary purpose of this study was an investigation into the

mass spectral behavior of these arylpropynes since there existed a real

possibility for interconversion between the propynes and their analogous

cyclopropenes on electron impact. A similar relationship between the

mass spectra of arylcyclopropenes and their analogous indenes has been

previously observed. It was therefore of interest to compare the nmss


viii









spectra of the propynes with their analogous cyclopropenes in an attempt

to gain insight into the photochemical behavior of the arylpropynes. In

general, the mass spectra of three out of four arylpropynes correlated

very closely to their analogous arylcyclopropenes.

Four different groups of arylpropynes were investigated. Group I

contained those propynes with the general formula (C6H5)3CCECR, while

group II had the general formula C6H5RR'CCGCC6H5. Group III comprised

a miscellaneous series of arylpropynes and related derivatives such as

an organometallic propyne, allene, etc. were contained in this group.

Group IV constituted tetraarylpropynes designed to investigate the

migratory apptitudes of differently substituted aryl groups.

Tetraphenylpropyne and closely related derivatives indeed photo-

cyclize to tetraarylcyclopropenes but the primary conclusion from the

study was that photochemical di-7-methane rearrangement of arylpropynes

is not as general as that found for 1,4 dienes. Unlike arylpropenes, the

arylpropynes investigated photocyclized only when the excitation was

shown to directly involve the propyne chromophore, and this was only

possible with extended conjugation.

As a further conclusion to the investigation of the photochemical

reactivity of arylpropynes, it became apparent that subtle steric and

electronic factors play an extremely important part in photocyclization

reactions.















CHAPTER I

INTRODUCTION

Previous work has shown that arylsubstituted propenes photo-
1,2
cyclize to cyclopropanes. Irradiation of 3,3,3-triphenylpropene

(1) at 2537 A led to the formation of 1,1,2-triphenylcyclopropane (2).


(C6H5 3C /H

H > 11


hv
_-___9


H
C di


It was also found that cis (4) and trans (5) 1,2-diphenyl-

cyclopropane were formed by irradiation of 1,3-diphenylpropene (3).


C6H5C / h
=C ------9
H- tC6H5


C6H5

H 6H5 +


(4)


Although the above rearrangements involve phenyl migration,

a hydrogen atom migration was observed in the: formation of


-1-








l,1-diphenylcyclopoppane (7) from the irradiation of 3,3-diphenyl-

propene (6).


(C6H5) 2CII HH
H>=-


H
C6115
H 6H5

(7)


The rearrangement appeared to be quite general in nature, although

some propenes studied [1,2,3-triphenylpropene (8) and 1,1,2,3-tetra-

phenylpropene (9)] failed to yield any cyclic products.


C6H5CH2, /H

C6H5' C6H5


hv
--9t----4


H5

I 6H,
C6H5


C695


C6H5CH2 /C6H5
c=c
C6H5 6rH5


hv
- \w ---


6H5


The thermal and photochemical rearrangements of cyclopropenes to
3-8 3
indenes have also been reported. Kristinieson isolated


hv
-----







1,2,3-trimethylindene (11) from the thfiel-y-si& of 1,2,3-triimethyl-3-

phenylcyciopropene (10).




CH' H H
3
CH CH
CH3
C 3 C 3 ll/ 3

(10) (11) 3
&
Griffin reported the formation of 1,2-and 2,3-dimethylindene

(13 and 14) from the irradiation of 1,2-dimethyl-3-phenylcyclo-
/L531
propene (12) at -2-47- A




H 6 H H CH3

hv C \ 3c + >H3

3 (1_22) 3 C a-CH3 I
S() (13) H 3 (14) H

5
Battiste postulaLed a diradical intermediate in the thermal

rearrangement of some tetraarylcyclopropenes (15 a-c) to tri-

arylindenes (16 a-c).







GH- H 6 //I K) 6 H5
SC6, 6H5 6H5R- C6 6
(15) (16)



(a) R=phenyl
(b) R=p-anisyl
(c) R=mesityl








While work was being performed on analogous cyclopropene rearrange-

ments, a similar diradical intermediate was proposed in a photo-

chemical rearrangement involving a substituted acetylene. Wilson and
9
Huhtanen irradiated methyl 4,4,4-triphenylbut-2-ynoate (17) and

isolated methyl 13-H-indeno-(1,2-l)phenanthrene-13-carboxylate (19).

A diradical was proposed as a possible intermediate leading to

l-carbomethoxyl-2,3-diphenylindene (18) which undergoes dehydrocycl-

ization to the indenophenanthrene.


(C6 5)3CC=CC02CH3 h [(C6H5)2CC6H5C=CO2CH3


(17)



$6Y5

H C62CH53 / 2

H CO CC3
2 3 H CO2CH3
(18) ((19)
This interesting reaction suggested the possibility of cyclopropenes

as intermediates in the irradiation of appropriately substituted

acetylenes. It was decided to investigate the feasibility of preparing

cyclopropenes in this manner, and the initial compound of interest was

1,3,3,3-tetraphenylpropyne (20). If cyclopropenes could be formed, the

scope of the reaction would then be investigated. Synthesis and

irradiation of (20) afforded 46 percent of tetraphenylcyclopropene (21).

Longer irradiation times led to formation of 1,2,3-triphenylindene (22)

and 13-phenyl-13-H-(1,2-1)indenophenanthr ne (23).












C6H. C6H5
(C6H, ) CCC6 55
6 5-6 5-- > &

(20) C6H5 C6H5

(21)


H C 6H

(22)


-3-I

H C6H5
(2365
(23)


The relative amounts of (21), (22), and (23) were detera ned by inte-

gration of the mir specLrum of the irradiation mixture. A multiple

for the ontho protons on the l,2-phanyl substituents of (21) could be

easily recognized, and the formation of products as a function of

time is shown in Table I.


RELATIVE PRODUCT DISTRIBUTION

Time (hrs) 21

1 70
4 46
24 25


TABLE I

IN THE IRRADIATION OF TETRAPHENYLPROPYNE

22 23

trace trace
35 9
42 19


With the initial success obtained in the irradiation of (20),

attention was focused on the scope of this new photocyclization

reaction. Becau-se of the similarity of uthe rass spectra of (20)





-6-


and (21), the existence of a correlation between the mass spectra of
11-13
various acetylenes and their photochemistry was also of interest.

Analogous acetylenes which were investigated were grouped into

four categories. Group I contained those acetylenes which had the

general formula (C6H5)3CC-CR (R=methyl, t-butyl, phenyl, and

carbomethoxyl). Acetylenes having the general formula C6 HR'RCCCC6 H

(R,R'=methyl, phenyl, etc.) were placed in Group II. The third group

of acetylenes did not have a common structure, and this group consisted

of analogous allenes, organomettalic acetylenes [(C6H5) 3MC=CC H5,

(M=Si, Sn, etc.) ], and other unique acetylenes. Group IV was com-

posed of tetraarylacetylenes (C6Hs)2ArCCGCAr' which would be

investigated with regard to the migratory apptitudes of the various

aryl substituents.

Therefore the majority o: work attempted was the synthesis,

analysis of mass spectra, and irradiaLion of the various acetylenes.















CHAPTER II

SYNTHESIS AND MASS SPECTRA

Groups (C6H5)3CC-CR

The first compound investigated, 1,3,3,3-tetraphenylpropyne (20)
14
(R= phenyl), was synthesized using a modification of the Wieland
15
and Kloss method. Triphenylchloromethane was added to a freshly

prepared solution of phenylacetylene magnesium bromide and refluxed

overnight. After chromatography, (20) was obtained as a pure white

solid in 52 percent yield. Because of the striking similarity of

the mass spectra 1,2,3,3-tetraphenylcyclopropene (21) and 1,2,3-
16
triphenvlindene (22), its thermal and photochemical rearrangement
5,17
product, the mass spectrum of 1,3,3,3-tetraphe.iylpropyne (2o)

was also carefully analyzed. Inspection of the mass spectrum of (20)

indicated an intense molecular ion (m/e 344) as well as the fragment

ions corresponding very closely in mass as well as relative abundances

(Table II) to those of (21) and (22).

TABLE II

PRINCIPAL FRAGMENT IONS IN THE MASS SPECTRA OF (20), (21) AND (22)
(RELATIVE INTENSITY)

Fragment (20) (21) (22)

(M+I) 345 32 30 30
(M) 344 100 100 100
(M-I) 343 9 13 15
(M-76) 268 10 13 6
(M-77) 267 44 50 35
(M-78) 266 9 13 8
(M-79) 265 29 38 33
(M-169) 165 22 21 7


-7-











Halton and Battiste postulated a rearrangement of (21) to (22)
16
occurring on electron impact. A common molecular ion or a rapid

equilibrium between the molecular ions of (21) and (22) was suggested.
+ +

C H5 C6H5 C6H5
S----- \ C H
4----- 6 \ 5

6H5 6H56

m/e(21) C6H /e(22)

+' C6H5








16
the ratio of (m/e 344) to (m/e 343) to be 6.5 to 1 at 70 ev. This

indicates the loss of hydrogen is still preferred to that of deuterium

in the indene. Some scrambling could have occurred on electron
18
impact, therefore the result is not conclusive that hydrogen is lost

from a position other than 1. Another interesting labelling experiment

with 3-pentadeutero-1,2,3-triphenylcyclopropene, (25) indicated that the

loss of phenyl (or pentadeuterophenyl) did not occur exclusively from

the 3 position. A ratio of the fragment ions (m/e 272) (loss of phenyl)

to (m/e 267) (loss of pentadeuterophenyl) was 2.5 to 1 at 70 ev and

3.2 to 1 at 20 ev. One explanation is a symmetrical molecular ion

(where x is pentadeuterophenyl 25 percent of the time) or a rapid

migration of phenyl (or pentadeuterophenyl) after loss of a phenyl

radical from the 1 or 2 position.











C H5 C6D C6H5 C6
e.65 1 C 6H1 + X*


C6HH5 C6H5 6H5 C6H5 6H5
L L
S 6D5 65 6D5 C6H5
CF C-D C C 1 C-H
e + C6H5*

C6H C6 C6H 6H C6 C6D


Although a partial rearrangement at least of (21) to (22) was

postulated on electron impact, a closer inspection of the spectra

revealed a memory affect. The [4n+-2(n-0) r electron] triphenyl-

cyclopropenium cation (m/e 267) would be expected to be more stable

than the [4n(n=l) r electron diphenylindenyl cation (m/e 267) in

solution. Also, the abundance of fragmentation to [ m/e 267(50)] and

[m/e 265(38)] from (21) is greater than that of [m/e 267(35)] and

[ m/e 265(33)] from (22). The memory effect suggested that in the

fragmentation of (21) and (22) there was leakage of cyclopropenyl to

indenyl (or vica versa) ions but the fragment ions of the initial ring

system were always preferred.

A further inspection of the close similarity between the mass

spectra of (20), (21), and (22) reveals that the relative abundances

for fragment ions from (20) are in closer accord with those of (21).

This close resemblance suggests partial rearrangement of (20) to (21)

may be occurring on electron impact. In fact, a common ion or a rapid

equilibrium between (20) and (21) may be formed. A phenyl migration

must be invoked on electron impact in either case. The molecular ion





-10-


from (20) might also have contributions from the triphenylindenyl

structure, but because of fragmentation abundances, it would appear

to resemble the cyclopropenyL type more closely. This would provide

a facile entry into the cyclopropenium (m/e 267) ion and a contribution
from the diphenylphenylethynl cation (m/e 267) although suggested
19
before does not appear too favorable.


C H C H +
C6 5 6 5 H +
(20) e.1..(C ) C1j +
(20) --- C 3CCCC 5E 6 5
m/e 344 IC 6H C6H
\m/e 344 6H5

(C6H5)2CC-CC6H5] -
m/e 267 / C6H5

C6 H567'



-CH3. C6H- 6H5
--'--~-- m/e 267


S-C14H11'


m/e 252 C '
SO m/e 265

H
m/e 165
The second compound in Group I investigated was 1,1,1-triphenyl-

2-butyne (R=methyl) (26). It was prepared in 26 percent yield from

the reaction of triphenylchloromethane with 1-propynyl magnesium

bromide, and identified by its nmr spectrum T [8.03(3,s) and 2.79(15,

a) mass spectrum [m/e 282(100) ], and elemental analysis.





-11-


The mass spectrum indicated loss of a methyl radical [m/e 267(92)]

as a primary process. Since the most stable fragment ion at (m/e 267)

would be the triphenylcyclopropenium cation, a partial rearrangement

to the molecular ion of methyltriphenylcyclopropene is suggested.

A phenyl migration must be invoked in formation of the cyclopropene

molecular ion. In order to form the more preferred (m/e 267) ion from

the cyclopropene (m/e 282) ion a similar argument to that of the

3-pentadeutero-l,2,3-triphenylcyclopropene mass spectrum would have to

be invoked. Since the formation of the alkynyl cation seems unlikely,

there are two possibilities for formation of the (m/e 267) triphenyl-

cyclopropenium cation: (i) loss of methyl radical from the cyclo-

propenyl (m/e 282) ion with concerted phenyl migration, (ii) loss of

methyl radical from alkynyl (m/e 282) ion with concerted migration of

two phenyl radicals.

The first process would involve two discreet phenyl radical

migrations whereas the second would involve a methyl radical removal

and simultaneous shifts of two phenyl radicals. It would appear

preferable to invoke the first scheme.






(26)e----[(C6H5)3CCGCCH3] ---[ c _c6 6H5 + CH3.




C6H5
e282 C6HH C6H5
(26) ^(C 6H 5 3rC CCH3 3] K / + CH3'
m/e 232 L C6H-








From labeling studies of 3-peatadeutero-1,2,3-tripher.yi-
16
cyclopropene, it was observed that loss of a phenyl radical was

not restricted to the 3 position.


C6H5 C6H5


m/e 282 C65 CH3
J.-C6H4




-H~~h2 ~ -
SH o 6 5 C6 5 6 5J L m/e 205 3
m/e -H2 m/e 257 /-CH -H2
-CH. -H 2

SC6 5 C3





m/e 189 J m/e 265 m/e 203
SC8104 /-C3H2





00 H
m/e 2 L m/e 165
In the fragmentation of (26) the preponderance of methyl radical

loss to phenyl radical loss (3.5 to 1) could reflect the stability of

the triphenylcyclopropenium cation (m/e 267) formed, or ease of the loss

of methyl radical compared to phenyl. Each of the fragment ions could

in turn lose a hydrogen molecule to form the appropriate phenanthrene

fragment (m/e 265 and m/e 203) although more direct routes are more




-13-


prominent, and loss of CsH4and C3H2 from the respective phenan-

threnes would lead to the observed fluorene fragment at (m/e 165).

A third compound studied was 1,l,1-triphenyl-4,4-dimethyl-2-

pentyne (27) (R=t-butyl). The desired compound was synthesized in

33 percent yield from the reaction of triphenylchloromethane with

3,3-dimethylbutynyl magnesium bromide and was identified by the nmr

spectrum T [8.80(9,s) and 2.75(15,s)], the mass spectrum (m/e 324),

and its elemental analysis.

The mass spectrum of (27) was somewhat different than that of

(26) in that the molecular ion (m/e 324) was present in only 2 percent

relative abundance with only the ions at [ m/e 268(43), 267(100) and

91(11)] being larger than 10 percent relative abundance. The

formation of the stable ion at (m/e 267) and the apparent relative

easy loss of a tert-butyl radical appear to be the strong driving

forces in the fragmentation. Again, as seen previously, rearrangement

to a cyclopropenyl molecular ion may be involved. The argument depends

on two discreet phenyl migration steps or two synchronous phenyl

migrations with the loss of a tert-butyl radical as indicated below.


[(C 6H 3CC-CC(CH3] L

m/e 324 i5 3

2e.i.


CH+ (CH33C
C .5 5j 33


SC6H5 + (CH3)3C'

(27) [(C6H5) 3CCECC(CH 3)3--

m/e 324 C6H5 C6H5





-14-


in the mass spectrum of (27), the ratio of (m/e 267) (loss of

phenyl radical) to (m/e 247) (loss of tert-butyl radical) is 50 to 1

which reflects the ease of formation of the corresponding cyclopropenium

ions as well as the formation of the tert-butyl free radical compared

to the phenyl radical. It is very possible that the stability of the

tert-butyl radical is the main reason for the small relative abundance

of the molecular ion.


(27) -.-e (C6H5)3 CCC(CH3)3l=--


m/e 324


-C H1


m/e 91


C6H5
+






m/e 265


+



C(CH 6H5
3 3 6 5




C6j


C(CH H5
33 6 5

m/e 247


1











H
m/e 165


Another compound intended for photochemical study was methyl

3,3,3-triphenylpropynecarboxylate (R=carbomethoxyl). Several


-C8H4
------




-15-


attempts to synthesize methyl 4,4,4-triphenylbut-2-enoate (30) were

performed with the desired olefin finally being prepared in 48 percent
20
yield from the Wittig reaction of triphenylacetaldehyde (28) and

diethylcarbomethoxymethylphosphonate (29).21 The compound was identi-

fied from its nmr spectrum T [2.02(I,d), 2.82(15,m), 4.33(1,d), and

6.27(3,s)], C=O stretch at 1720 cm- in the ir spectrum, mass spectrum

(m/e 328), and elemental analysis. Although bromination and subse-

quent dehydrobromination was expected to yield the desired alkyne, a

colorless solid mp 196-1970 with a singlet at T (3.55 and a multiple

at 2.70) in the nmr spectrum indicated the expected dibromide was not

formed. The solid was later characterized by its lactone C=O stretch
-l
at 1750 cm- in the ir spectrum, its mass spectrum (m/e 312), and its

elemental analysis as 3,4,4-triphenylbut-2-enoic acid lactone (31).

Related phenyl migration had been noted before in brominations of
22,23
arylpropenes, but this seemed to be a novel example for the

formation of a lactone without bromine in the final molecule.






(C5) 3C H Br2 (C6H5 2 H (C6H5)ZC+ Br
CC )sC>< --2 C (6H5 / P\
H/ =COCH3 6 OCH3 C j5 OCH3

(30)


C6H5 Er C6H5 (Br C6H5


C6H CH3Br- -- C6H C5
C6H C6H5 C5
(31)





-16-


Other more direct attempts to synthesize the alkyne were

conducted using a variety of bases sodium = hydride, n-butyl lithum,

ethyl magnesium bromide, etc.) and various propargyl derivatives

(pyranyl ether, diethyl acetal, etc.) without success. A final

unsuccessful attempt at bromination of 3,3,3-tripheny.v-l-propyne (1)

resulted in an uncharacterizable tar and also some rearranged

dibromide (2,3,3-triphenylallyl bromide).

Group II C6H5R'RCCCC6H5

This group of acetylenes was the most widely investigated. The

first two compounds synthesized were 1,3,3-triphenyl-L-propyne (32)

(R=phenyl, R'=H) and 1,3-diphenyl-l-propyne (R=R'=H) (33). Both were
14
synthesized by a modification of the procedure of Herriot and

obtained in yields of 45 percent for (32) and 31 percent for (33).

The reaction consisted of employing the appropriate halomethane [di-

phenylbromcmethane for (32) and benzyl chloride for (33) ] with

phenylacetylene magnesium bromide.

One complication arose in the purification of 1,3,3-triphenyl-

propyne (32). The initial slightly yellow solid product became even

more yellow after chromatography. A rearrangement to the correspond-
24, 25
ing allene was taking place on the activated alumina. The alumina

was deactivated by treating it with ethyl acetate and heating it to 1600

for at least 5 hours. Use of the deactivated alumina resulted in pure

(32) being obtained.

Another compound cf interest was 1,1,3-triphenyl-2-propyn-l-ol

(R=phenyl, R'=OH) (34). Since the diphenylphenylethyn-yl cation

could be generated in concentrated sulfuric acid, its -photochemistry





-17-


could be studied. The desired alkyne was prepared in good yield from

the reported reaction of phenylacetylene magnesium bromide and
26
benzophenone. The mass spectrum of (34) was not of the general

type in that a large relative abundance of fragment ion m/e 267 was

lacking. Alcoholic compounds in general show loss of a hydrogen atom

as a predominant process on electron impact, and the diphenylphenyl-

ethynyl cation (or triphenylcyclopropenium) isomer was not being pro-

duced on electron impact.



e.1. (C6H5 + -H
(34) --. (C6H5)2COHC-CC6 H I5 H (CgH5)2COCCC6H5 +


m/e 284 m/e 283


-OH- _-OH.






(C6H52CC-CC 6H5 --

m/e 267 C6H "C6-5


m/e 267





The next compound of interest, 3-methyl-1,3,3-triphenyl-

propyne (35) (R=phenyl, R'=methyl), was prepared by the reaction

of 1,1-diphenyl-l-chloroethane with phenylacetylene magnesium

bromide in 14 percent yield. The compound was identified by its

nmr spectrum I [7.92(1,s) and 2.60(5,m)], mass spectrum









-18-

[m/e 282(78)], and its elemental analysis. The possible photo-

chemical rearrangement product of (35) was also investigated at this

time. The desired compound, 3-methyl-l,2,3-triphenylcyclopropene (36)
27
was synthesized in 67 percent yield by the method of Breslow and Dowd.

The thermal and photochemical behavior of (36) was also of interest

to us.

Because of the close similarity of previous cyclopropenes and

analogous alkynes, the mass spectra of (35) and (36) were carefully

compared. The other isomer (26) previously investigated is also

included in Table III.

TABLE III

PRINCIPAL FRAGMENT IONS IN THE MASS SPECTRA OF (35), (36), AND (26)
(RELATIVE INTENSITY)

Fragment (35) (36) (26)

(M) 282 78 100 100
(M-14) 268 22 17 18
(M-15) 267 100 73 92
(M-17) 265 21 20 12
(M-30) 252 12 12 10
(M-77) 205 16 10 29
(M-79) 203 11 11 18
(M-80) 202 10. 10 14
(M-91) 191 11 11 11
(M--93) 189 11 5 9
(M-117) 165 18 5 12
(M-156) 126 10 13 9

One can see the similarity in all three systems although the

resemblance between (36) and (26) appears closer in the higher

molecular weight region (above 252). However, in the lower molecular

weight region (below 205) (35) and (36) are closer in accord with

respect to relative abundances.




-19-


The well known stability of the cyclopropenium ion would again

appear to be the deciding factor. In the case of (26) the lack of

rearrangement to the cyclopropenyl molecular ion would result in a

very unlikely ion being formed by simple cleavage of a methyl radical,

whereas (35) could possibly form an alkynyl ion which could perhaps

rearrange as easily to the cyclopropenyl ion.


(26) e.


H


m/e 165


m/e 265


(C6H5)2CH3CCCCC H

(35)
e.i.

[(C6H5) 2CH3CCC6H5 ]
m/e 282

(C H5 CCEC H

m/e 267
2


m/e 203


The complete fragmentation pattern of (35) and (36) lead to

similar ions as seen previously in these systems. The phenanthene

m/e (265,203) and fluorenyl type ions (m/e 165) are seen again,





-20-



and while it is difficult to conclude what amount of rearrangement to

the cyclopropenyl molecular ion if any has occurred in (35), the close

resemblance of the spectra of (35) and (36) below (m/e 267) indicates

the strong possibility of this rearrangement.

An interesting compound 3-methyl-l,3-diphenyl-l-butyne

(R=R'=methyl) (37) was prepared by the reaction of cumyl chloride

with phenylacetylacetylene magnesium bromide in 77 percent yield. The

clear liquid was identified by its nmr spectrum T[2.65(10,m) and

8.23(6,s) ] mass spectrum (m/e 220), and its elemental analysis.

The mass spectrum was unusual in that two distinctly different

processes appeared to be taking place. The most predominant feature

of the spectrum was the base peak at (m/e 205) which could correspond

to the methyldiphenylcyclopropenium cation, but the abundant peak at

[m/e 119(21)] indicated another type cleavage (previously unseen) was

also taking place. In this spectrum, the equilibrium between the

alkynyl molecular ion and the cyclopropenyl molecular ion is probably

not largely in favor of the cyclopropene as seen before. The ion of

(m/e 119) would result from loss of phenylethynyl radical from the

alkynyl molecular ion while the base peak would result from loss of

methyl radical from either the cyclopropenyl molecular ion or the alkynyl

molecular ion. This seems to be a good example to substantiate some of

the previous alkynes. Other tertiary cations from analogous systems

would be at least as stable as the dimethylphenyl cation and the

phenylethynyl radical would remain similar in stability if not exactly

the same.

Therefore, lack of this type of cleavage heretofore could indicate

a favorable equilibrium towards the cyclopropenyl molecular ion




-21-


previous alkynes mentioned.


+
(37) Ce.i. C6H5 (C3)2CC=CC6H5 C
m/e 220


-C6H5CEC'


C6H5 (C3)2C+
m/e 119


m/e 205


CH 3H3

C6H5 C6H5


m/e 220
-CH3


C6H5CH3C CCC6H5
m/e 205


I I


L^4


m/e 91


- +


j-C6H-5


CH3

CH3
CO-16 3


m/e 143


m/e 203


Group III


This group of compounds was not closely related in structure, but





-22-


was also not applicable to the two previous groups. One of the com-

pounds synthesized was tetraphenylallene (38) because of its isomeric

relationship to tetraphenylpropyne (20) and irradiation could possibly

lead to analogous products. Tetraphenylallene (38) was synthesized
28
by the procedure of Vorlander and Siebert. The mass spectrum of

(38) was very similar to that of (20), (21), and (22), although some-

what closer to that of the indene (22). The peak at [m/e 343(11) 1

would seem itself indicative of at least partial rearrangement on

electron impact.


C6H5 C6H5
e.i. 3> e 3 +

\6 //I C6H 5 H C H5
(38) I1 CC6H



C6H5),C=C=C(C6H5)2] m/e 344 m/e 343C6






C6H5 C6H5 C6 5 C6H5
SL m/e 267
e 344
m/e 265 -CH3

H-








L /c 165H
m-/ \1-



m/e H65 C -( }


m/e 252





-23-


In fact both indenyl and cyclopropenyl molecular ions are notably

formed. Since, as seen before, the fragmentation pattern indicates a

leakage over to the cyclopropenyl type fragments at [m/e 267(35) and

m/e 265(24)]. A comparison of the fragmentation abundances particularly

the 267:265 ratio (1.46 to 1) reveals that allene (38) probably

rearranges initially or primarily to the indene molecular ion on

electron impact. The reasons for the difference between isomers

(20) and (38) are not appreciated at this writing.

The silicon analog of (20), phenylethynyltriphenylsilane (39) was
29
prepared by a scheme similar to that of Eaborn and Walton utilizing

lithium phenylacetylide and chlorotriphenylsilane.

The mass spectrum of (39) was quite similar to that of its carbon

analog. A major fragment peak at (m/e 283) corresponds to loss of

pbenyl radical. Another competing process seems to be cleavage of

the phenylethynyl radical producing the triphenylsilyl cation (m/e 259).

Although the major fragmentation appears to be loss of phenyl, one

cannot predict the structure of the (m/e 283) ion since previous chemi-

cal attempts to synthesize silicon analogs to cyclopropenes have
30
failed, and these compounds are not available for comparative study.

The third compound of Group III, 9-phenyl-9-phenylethynyfluorene

(40) was prepared in 43 percent yield from 9-phenyl-9-chlorofluorene

and phenylacetylene magnesium bromide. Identification was made by the

analysis of the nmr spectrum, i [ 2.71(l,m) and 2.30(8,m) ], the mass

spectrum [m/e 312(100)], the ultraviolet spectrum, and its elemental

analysis.

The mass spectrum of (40) was indicative of a rearrangement tak-

ing place on electron impact. There are two feasible rearrangements





-24-


(phenyl migration or fluorenyl migration). However, the large

relative abundances at m/e 341(34) and m/e 339(18) seem to indicate

a rearrangement to an indenyl type molecular ion. The possibility of

leakage back to the cyclopropenyl type fragment ions remains favorable

as in previous examples. This fragmentation appears to be analogous to

the stability effect seen in tetraphenylcyclopropene. Therefore, al-

though the molecular ion is largely indenyl in nature, the fragmentation

ion (m/e 265) is probably largely cyclopropenyl in nature.



(40) {O O O

C6H5 PCC6H5 H
m/e 342 c--6H C6H5
SCH5 6H5 m/e 342
C6H5 + / -H*


O 6 5*/ O m// 341cX-
C 60 55 +
//--^,\~-- //-H/y^. "
0 --- -C6H 65 C6H5l 23

n/e C6HS 0 0








m/e 265 m/e 339
The last compound investigated in this group was

3-phenylethynyl-l,2,3-triphenylcyclopropene (41). It was a very

interesting compound in that two types of photolytic rearrangements

seemed possible; (a) formation of a spiro compound, (b) typical

cyclopropene to indenyl rearrangement. The desired cyclopropene

(41) was prepared by addition of triphenylcyclopropenium bromide to





-25-


a solution of phenylacetylene magnesium bromide and identified by

its nmr spectrum T (2.17 to 2.90,m), mass spectrum (m/e 368), ultra-

violet spectrum, and elemental analysis.

The mass spectrum was not very informative as the major process

was simply loss of phenylethynyl radical.




C6H rCCG H5 C6H5 C=CC6H
66"9 C66C6H5
e.i.
-7'
C65 --5 CH5 C6H 5
(41) 6 5 C6H5
m/e 368


-C8H5CH
C65 -C8 7



6H5 C6H5
S-CH3 m/e 267 H2




C6H5








m/e 252 m/e 265

At this point it was of interest to compare the (m/e 267) ion

versus (m/e 265) ion ratio in some of the previous compounds. It can

be easily seen that the (m/e 265) ion is of almost equal importance

in the mass spectrum of (20), (21), and (22), but there is little

conversion of (m/e 267) to (m/e 265) as seen in triphenylcyclopropenium







-26-


TABLE IV

FRAGMENTATION ION (m/e 267/265) RATIO



C6H CH5 1.3


C6H5 C6 5
(C6H5)3CC=CC6H5 1.5
CH
65

Q\6H5 1.0


(C6H5)2CH3CCECC6H5 4.8






C6H5 C6H5 7.1

CH3 C6 5


C6H5 C6H5 3.7

(C6H5)3CC=CCH3 7.7

(C6H5) 3CCCC(CH3)3 16.7



C6H5 CCC6CH5


C6H5 C6H5 6.3





-27-


31
chloride. Tne high ratios of the remaining compounds indicate

that the preferred fragmentation path is toward the (m/e 267) ion.

Group IV (C6H5)2ArCCECAr'

The final series of compounds synthesized were a group that could

serve as a test of migratory apptitudes upon irradiation. The initial

compound prepared was 3-p-anisyl-l,3,3-triphenylpropyne (42). It was

synthesized from p-anisyldiphenylchloromethane and phenylacetylene

magnesium bromide and identified by its nmr spectrum T [6.29(3,s) and

2.90(19,m)], mass spectrum (m/e 374), and its elemental analysis.

The mass spectrum was quite unusual in that the major fragmen-

tation appeared to be the loss of phenylethynyl radical. The main

driving force must be the stability of the (m/e 273) ion. The

absence of the [m/e 267(1)] indicates that the expected rearrangement

to the cyclopropene is not taking place on electron impact. The

fragmentation pattern is plainly quite different from the tetraphenyl

analog.

The loss of phenylethynyl radical in the fragmentation was seen

before in 3-methyl-l,3-diphenyl-l-butyne (37), but not to such an

exclusive extent. There was no indication for formation of triphenyl-

cyclopropenium cation [m/e 267(1) ] and little if any anisyldiphenyl-

cyclopropenium cation [m/e 296(7)]. The different fragmentation

pattern must be due to the additional stabilization of the fragment

cations by the anisyl moiety. In this specific instance there doesn't

seem to be any indication for formation of cyclopropenyl molecular ion

on electron impact.




-28-


C61, OCH3

(C6H5) 2 CCC6H5

m/e 374


C6H5 C6H40CH3


C6H5 C6H5
m/e 374


-C CC-C6H4

-C6H30CH3

40CH3 C6H5

5C6H5 -
5 H


m/e 273


m/e 167


2-H2


D +P


m/e 165


OCH3



H


m/e 195


The second compound studied at this time was 1-p-anisyl-3,3,3-

triphenyl-l-propyne (43). It was synthesized from triphenylchloro-

methane and p-anisylacetylene magnesium bromide and identified by its

nmr spectrum T [6.29(3,s) and 2.37(19,m)], mass spectrum (m/e 374), and

elemental analysis.

Two related cyclopropenes were also prepared and investigated.


e.2)


m/e 197


H2





-29-


The reported reaction of p-anisyl magnesium bromide with triphenyl-

cyclopropenium bromide6 yielded 3-p-anisyl-1,2,3-triphenylcyclopropene

(44). Another isomer was obtained by reaction of phenyl magnesium

bromide with l-p-anisyl-2,3-triphenylcyclopropenium bromide. A mixture

of the two possible cyclororopenes in a ratio of (95/5) resulted which

was separated by column chromotography to give the major component

l-p-anisyl-2,3,3-triphenylcyclopropene (45). Identification of

structure was made by analysis of the nmr spectrum r [6.13(3,s) and

2.38(19,m)], mass spectrum (m/e 374), ultraviolet spectrum, and

elemental analysis.

The mass spectra of (43), (44), and (45) were quite similar and

completely different than (42). Although the (267/265) ratio is less

than seen before, the similarity of (43), (44), and (45) is shown

below in Table V.


PRINCIPAL FRAGMENT IONS


Fragment

(M)
(M-l)
(M-14)
(M-15)
(M-31)
(M-76)
(M-77)
(M-93)
(M-107)
(M-109)
(M-120)
(M-121)
(M-122)
(M-209)


TABLE V

IN THE MASS SPECTRA OF (43), (44), AND (45)
(RELATIVE INTENSITY)

(43) (44) (45)

100 100 100
8 4 6
4 3 11
14 11 27
6 8 5
10 8 10
40 28 31
4 10 5
6 6 9
10 16 20
7 11 12
11 18 16
14 24 18
5 7 16


Both cyclopropenes (44) and (45) exhibit larger relative abundances





-30-


in the smaller molecular weight ions (below 250), but there seems to

be close agreement in the other relative abundances in all three

compounds. It is interesting that the (m/e 267) versus (m/e 265)

ratio is less than one in all three cases. It seems reasonable to

invoke a rearrangement of (43) to a cyclopropene type molecular ion

on electron impact. The spectrum of (43) is very similar to that of

its analog (20) except for the (267/265) ratio. This propyne (43)

evidently is not showing the type of fragmentation as in the other

isomer (42) and is apparently following alkynyl-cyclopropenyl type

molecular rearrangement. Although the analogous indenyl compounds

were unavailable, the small abundance of (m/e 343) ions for (44) and

(45) might indicate little, if any, indene formation on electron

impact.

+ -S+ 5C5 +
+ C6H5 C6H5 C6HS C6H5

(43)---[(C6H5) 3CC-CC6H40CH31

m/e 374 L m/e 374 /e
SC \i ~-CH3' e 359
S 461140C3
C6H40CH3 C H5 C H5



C6HS 6H51 j C6 6H5 e
Sm/e 297 m/e 267 (1) (

-H 2 3 m/e 265


C H40CH3






i)Ur,-, L nm/e 165






-31-


As a conclusion to the analysis of the mass spectra of the various

propynes, different types of fragmentation were observed. In some

instances, it appeared very probable that a rearrangement to a cyclo-

propenyl type ion was occurring on electron impact. In other cases,

it was not as clear as to whether the rearrangement was occurring to a

large extent, and in some cases, there was no evidence for any

rearrangement at all with fragmentation of the alkynyl moiety as the

major process.

As an insight into the photochemical behavior of the propynes, a

comparison of the mass spectra of the propynes with the mass spectra

of their analogous cyclopropenes showed very close agreement in three

out of four cases.














CHAPTER III


PHOTOCYCLIZATTON STUDIES OF ARYL SUBSTITUTED FROPYNES
RESULTS AND D:SCUSSIN

Group I (C6H5)3CCzCR

The first compound of Group I studied, 1,3,3,3-tetraphenylpro-

pyne (20), (R=phenyl) has already been briefly mentioned in the intro-

duction. The ultraviolet spectrum of (20) in cyclohexane consisted

of two maxima at 259,5(e=21,100) and 244(25,000) nm respectively. The

large extinction coefficients confirmed that the transitions are

u-ndcubtedly of the allowed 7-7* type indicative of conjugated phenyl-

propyne excitation [benzene maxima at 254(E=200) nrm It was noticed

that in cyclohexane the irradiation was most efficient while employi.-g

the 2537 AO lamps, although somewhat slower rearrangement was also

observed in benzene. The photocyclization proved to be quite rapid

with a small scale sample (100 milligrams or less) and in these cases

Lhe reaction was usually complete in about one hour. The reaction was

monitored by either ultraviolet spectra [tetraphenylcyclopropene maxima

at 280(e=33,800), 305(21,400), 318(22,300) and 335(20,400) nm ]or nmr

spectra (integration of the multiple for the ortho protons on the

1,2-phenyl substituencs in the cyclopropene product).

Attempts to sensitize the cyclization reaction with benzophenone,

acetophenone, or acetone at both 3100 and 3500 A0 were unsuccessful,

which suggested that the initial process was probably singlet in nature.

1he secondary reaction (formation of indene) could then possibly be a
10
triplet process. In the originally postulated mechanism,





-33-


a singlet diradical was the initially formed species which in turn

undergoes a 1,2-phenyl migration yielding a second diradical inter-

mediate that closes to product. On further inspection, the mechanism
32
could be formally viewed as a di-n-methane type of rearrangement.



C6H5
Sh r phenyl |
(20) (C6H5)3CC=CC6H5 in 2C=CC6H5







C6H5 C6H5 C6H5
-- mi gration (C5H C










5C6H5


C6H5 C6H5 C65
(2) H C65
(22)



Zimmerman states that acylic and monocyclic di-n-methanes rearrange

via a singlet process and the entire rearrangement can be viewed as a
33
concerted process. However, it does not seem probable that (20) can

concertedly rearrange to (21) because of the improper geometric

relationship between the appropriate orbitals. The reported photo-
1
cyclization of 3,3,3-triphenyl-l-propene (1) to 1,1,2-triphenyl-

cyclopropane could be proceeding by three mechanisms: (1) excitation

of the alkene chromophore, (2) triplet sensitization by solvent, or

(3) excitation of the benzene chromophore. The first possibility can

be ruled out because of the large amount of energy needed, and the
33
second seems unlikely in view of other acyclic data. Therefore using





-34-


(20) (C6H5)3CC=CC6H5





C6HC5CC65 CCH5 C6H5 C6 g




CC6H5 C65 6H5
(21)





(1) as a model for Group I compounds, the excitation of the phenyl

moeity would be the initial step. In Group I the excited phenyl moeity

can then attack another phenyl moeity or the alkynyl moeity in the

second step. The former yields an intermediate which would simply

radiate or fluoresce back to starting material and the latter leads

to product. The feasibility of this mechanism for Group I compounds

hinges on the different R groups. The different R groups would have

no effect on the reactivity if this mechanism were operating, because

the geometry between the alkynyl and phenyl groups would remain the

same.

However, after inspecting the irradiation of Group I as a whole,

it is apparent that the R group does play an important part in the

reactivity of the molecule. In order for the photocyclization to

proceed, the excitation must be initially involved with the alkynyl

moeity and only in cases where this excitation is favorable will the

reaction proceed.





-35-


The second compound investigated was (R=Fethyl) (26). The

ultraviolet spectrum of 1,l,l-triphenyl-2-butyne (26) in cyclohexane

consisted of a single maximum at 257 (E=200) nm.

Solutions of (26) in cyclohexane were irradiated at 2537, 3100,

and 3500 Ao. However, there was no evidence of product formation by

nmr spectra. Another sample was irradiated using the Hanovia 450 W lamp,

but again no products were detected by glpc. One explanation for the

lack of phenyl rearrangement is that the propyne moeity was not directly

excited, rather only the isolated benzene chromophore was excited,

and this excitation does not lead to product formation.

A third system studied was l,l,l-triphenyl-4,4-dimethyl-2-pentyne

(27). The ultraviolet spectrum of (27) consisted of maxima at

266(E=510), 260(725), and 253.5(635) nm in isoctane. Again, the low

extinction coefficients indicate the isolated benzene chromophore as

the moeity undergoing excitation.

Irradiation of solutions of (27) in cyclohexane at 2537, 3100,

and 3500 A resulted in no detectable reaction when monitored by nmr

or glpc. Only starting material could be recovered when using the

Hanovia 450 W lamp. An explanation similar to that of (26) could

possibly be employed, due to the similar ultraviolet spectra. The

photolytic energy promotes excitation in the isolated benzene moieties.

Both (26) and (27) behave similarly on irradiation, with decomposition

of starting material after two or three hours but without any detectable

product formation. Apparently these types of propynes are photo-

lytically decomposing, although Wilson and Huhntanen's analog
9
(R=carbomethoxyl) undergoes rearrangement. However, the excited

chromophore in their case is obviously the a,3 unsaturated carbonyl.





-36-


(26) + (27) > No Reaction








(C6H5) 3CC0CCO2CH3 hv

(17) H CO CH
2 3
(19)

Group II (C H )RR'CC-CC H
65 65
The first compound investigated was 1,3,3-triphenyl-l-propyne

(32) (R=phenyl, R'=H). The ultraviolet spectrum of (32) consisted

of maxima at 242(e=24,500), 253(21,800) and 279(425) nm in cyclo-
24
hexane.

A solution of (32) in cyclohexane was irradiated at 2537 AO

for six hours and monitored by the disappearance of the benzyl proton T

(4.88) in the nmr spectrum. Three components of the resulting yellow

oil were separated by preparative glpc and identified by comparison

with authentic compounds. The mixture was found to contain diphenyl-

methane, benzoic acid, and benzophenone. Ostensibly, these products

could have arisen due to the presence of oxygen during the irradiation

and future irradiations were thoroughly purged and degassed with

argon.

A second irradiation was performed on a degassed solution of (32)

in cyclohexane at 2537 A and monitored by glpc. The irradiation

was ceased after four hours when only approxi:iaLcly 11 percenii of (32)





-37-


remained and one major product was estimated at consisting of approx-

imately 70 percent of the reaction mixture. The yellow oil which could

not be crystallized was purified by employing preparative glpc. The

nmr spectrum (CDC13) consisted of a singlet at T [7.20 (3) and a multi-

plet from 2.5 to 3.3 (15)]. The mass spectrum showed a parent ion at

(m/e of 270) which corresponded to an increase of two units in the

molecular weight with respect to the starting material.

At first, the product was tentatively assigned as 1,1-diphenyl-
34
indane (46) due to the similarity in their nmr spectra, but the

preparation of authentic 1,1-diphenylindane (46) (by Wolf-Kishner
35
reduction of the appropriate diphenylindanone) and comparison of

the glpc retention times indicated that the two compounds were

not identical.

The product was positively identified by comparison of the

nmr spectra (Fig. 1,2), the mass spectra, and the glpc retention times

(single and mixed), with authentic trans-1,2,3-triphenylcyclopropane

(48). The authentic sample was prepared from trans-1,2,3-triphenyl-l-
27
chlorocyclopropane (47) by an adaptation of Breslow's original method

(Fig. 3). A previous attempt (using excess potassium-t-butoxide) re-

sulted in the formation of triphenylcyclopropene in about a 10

percent yield.



C6H5 C6 5
------ y5
CH H --H
C6H H C6H5CHC12 (1) Mg H

6 5 KO-t-Bu C6H5-' C6H5(2) H3C- C6H5 6H5
(47) (48)





-38-


This interesting reaction was indeed photocyclizing but with

addition of a hydrogen molecule somewhere along the reaction sequence.

It wculd be difficult to assign a definite mechanism without further
1
study, but Griffin has shown the relative ease of rearrangement of

analogous olefins and this might be a reasonable path for the reaction

after preliminary addition of hydrogen to the propyne. Once the hy-

drogen molecule is added, the mechanism can be viewed as a concerted
33
di-T7-methane type. However, both hydrogen atoms may not necessarily

be added in the first step, and phenyl migration after addition of a

hydrogen atom might lead to an allylic radical which would photolyt-

ically close to a cyclopropyl radical. The second hydrogen atom could

then be added. Interestingly, the glpc analysis of the irradiation does

not indicate an increase of one species early in the reaction with sub-

sequent decrease as the cyclopropane product is formed as should be the

case with the first mechanism.


(C6H5)2CH > H

H C6H5


IIH2


(32)
+ H

C6H5


C6H 65 H65 C66H5
6 5g~


C H

6 5
.- H

(48)


C6H5
H" "H


C6g5 C6 5
(48)





-39-


The next compound of interest in this series, 1,3-diphenyl-l-

propyne (33) (R=R'=H) exhibited ultraviolet maxima at 240(e=25,200),

251(22,900), 272(1,050), 279(800), 305(460), and 326(370) nm in
25
ethanol.

Irradiation of (33) in cyclohexane at 2537 AO resulted in disap-

pearance of the starting material after about five hours, but the

nmr spectrum of the reaction mixture was discouraging in that no

detectable signal was present for either a cyclopropene or cyclopro-

pane type product. Only brown oils could be recovered on work-up.

The propyne was indeed labile under photolytic conditions, but attempts

to isolate any distinguishable products were unsuccessful.

Another compound of interest was 1,1,3-triphenyl-2-propyn-l-ol

(34) (R=phenyl, R'=OH). The cation obtained from treatment of (34)

with strong acid was of primary interest, and was generated by addition

of a methylene chloride solution of (34) to concentrated sulfuric
36
acid. The resulting deep crimson solution exhibited visible absorp-

tion at 512(E=33,000) and 447(28,400) nm. The sulfuric acid solution

was irradiated at 3100 AO and a decrease of the original maxima occur-

red along with the appearance of a new maximum at 466 nm. The desired

triphenylcyclopropenium cation product should absorb in the 330 nm

region of the ultraviolet spectrum. However, a photolytic rearrange-

ment was definitely occurring, since irradiation of a sulfuric acid

solution of (34) in a nmr tube resulted in a different nmr spectrum

after two hours. Unfortunately, the lack of evidence for the formation

of a cyclopropenium cation was obvious and further investigation of

the product was discontinued.

Photolytic investigation of 3-methyl-l,3,3-triphenylpropyne (35)





-40-


(R=phenyl, R'=methyl) was also performed. The ultraviolet spectrum

of (35) in cyclohexane exhibited maxima at 253(e=24,000), and

242.5(26,000) nm. A solution of (35) in cyclohexane was irradiated

at 2537 Ao and monitored by glpc. Consumption of the starting material

was complete in one hour, and at least twelve products were observed

by glpc analysis without any predominating. Irradiation in benzene

solution was also performed at 2537 Ao, but no products were detected

by glpc after nineteen hours. The cyclohexane was most likely serving

as a hydrogen donor in enabling (35) to be photochemically labile, as

was the case with (32). The large number of products could possibly

mean three or four major pathways and further investigation was

discontinued.

The irradiation of 1,2,3-triphenyl-3-methylcyclopropene (36)

(Fig. 4) was also studied at this time. The ultraviolet spectrum of

(36) in isoctane exhibited maxima at 328(E=25,000), 312(30,000) and

227(31,000) nm.

A solution of (36) in cyclohexane was irradiated at 3500 AO for

twenty-seven hours, and a light yellow oil remained after removal of

solvent. Chromatography yielded two products which constituted approx-

imately 72 percent and 22 percent of the product mixture. The more

abundant product was a white solid (mp 89-900) which exhibited a nmr

spectrum (CDC13) (Fig. 5) that consisted of a three proton quartet at

T [7.80(J=2cps) ], a poorly resolved one proton doublet at 5.44 (J=2cps),

and a fourteen proton multiple from 2.55 to 3.05. The product was

identified as 3-methyl-l,2-diphenylindene (49).

The second product was a white solid (mip 303-3050) and the nmr

spectrum (CDC13) (Fig. 6) consisted of a three proton singlet at t





-41-


(8.07), a one proton singlet at 5.02, a two proton multiple centered at

3.95, a ten proton symmetrical multiple from 2.90 to 3.40, and a two

proton multiple centered at 2.50. The mass spectrum showed a molecular

ion at (m/e 564) and the remainder of the spectrum resembled the spectrum

of the cyclopropene very closely. Because of analogy with similar sys-
37
tems, along with the interpretation of the nmr spectrum, the second

product was postulated as the head to tail dimer of 3-methyl-l,2-

diphenylindene (50). For further evidence that the second product was

indeed a dimer of the indene, a solution of (49) was irradiated at

3500 AO for twenty-two hours and an nmr spectrum (CDCI3) showed the

presence of the identical compound obtained from irradiation of (36).

The same white solid (mp 304-3050) was obtained on work-up.

In conjunction with the irradiation of (36), a thermolysis was

also investigated. A sealed tube containing (36) was immersed in a

silicon oil bath at 235-2400 for four hours. The nmr spectrum (CDCI3)

(Fig. 7) of the resulting yellow oil showed two products present;

3-methyl-l,2-diphenylindene (49), 55 percent, and 1-methyl-2,3-

diphenylindene (51), 45 percent.



C6H5 H3 CH3
hv hv
S----- 6H5 Dimer

C65 H 6H5 (50)
(36)
(49)


CH 3 C6H5
C6 3 [C 6-5 + OC6H5


C6H5 C6H5 H7-'C6 5 I 1H3
(36) (49) (51)




-42-


Another analogous compound studied was 3-methyl-l,3-diphenyl-l-

butyne (37) (R=R'=methyl). The ultraviolet spectra of (37) in cyclo-

hexane exhibited maxima at 252(E=27,600) and 245(26,400) nm.

A solution of (37) in cyclohexane was irradiated at 2537 A and

the appearance of a single major product was noted from glpc analysis.

The major component represented 54 percent of the reaction mixture,

and was purified by preparatory glpc. The clear liquid exhibited an

nmr spectrum (CDC13) which consisted of a two proton singlet at T

(7.65), a six proton singlet at 9.01, and a ten proton singlet at

2.81. The mass spectrum showed the molecular ion at (m/e 220) (addi-

tion of hydrogen molecule), which was also confirmed by the correct

elemental analysis obtained for C17H18. By analogy with (32), the com-

pound was tentatively assigned as trans-1,2-diphenyl-3,3-dimethylcyclo-

propane (52).

A number of methods were employed attempting to synthesize the

unknown cyclopropane (52). Insertion of phenylchlorocarbene (generated

from benzal chloride and potassium-t-butoxide) into R,P-dimethylsty-

rene, and irradiation of B,B-dimethylstyrene with phenyldiazomethane

were both unsuccessful.

A solution of (37) in hexane was hydrogenated using 5%-palladium-

on-barium sulfate as the catalyst. The hydrogenation was slow and

finally terminated after twelve hours. The olefin was purified by

preparatory glpc, and the clear liquid had an nmr spectrum (CDC13)

which consisted of a six proton singlet at T (8.64), a one proton

doublet at 4.08(J=12 cps), a one proton doublet centered at 3.45(J=

12 cps), and a ten proton syLire.trical multiple from 2.50 to 3.20. The

compound was assigned as cis-l,3-diphenyl-3-methyl--l-butene (53)





-43-


because of the coupling constants and method of hydrogenation. Ir-

radiation of (53) in cyclohexane at 2537 AO yielded an oil which had an

identical nmr spectrum to that of the photoproduct of (37) and comparison

of the retention times of both photoproducts from (53) and (37)

showed them to be identical. The photoproduct of (37) was therefore

postulated as trans-l,2-diphenyl-3,3--dimethylcyclopropane (52).



C6H5(CH3)2CC=CC6H5
(37)
hv

G"3
66 5 H
H '"C 6H5


W ^ S, ^s-s .^(52)
C6H5(CH3 ) C2C C6H5 (5.2)

H H

(53)


That concludes the investigation of compounds which belong

to Group II, and every member of the class was photochemically labile

although cyclization took place in only two cases, (32) and (37).

Group III

The third group of compounds investigated were not grouped be-

cause of any common feature but rather because they were not suitable

for placement in any of the other groups. Tetraphenyllene (38) was

investigated because of the reactivity of its isomer (20). The ultra-

violet spectrum of (38) exhibited a long wave length maxima at 265(s=
38
28,800) in tetrahydrofuran.

Solutions of (38) in cyclohexane were irradiated at 2537, 3100,




-44-


3500 Ao, and also with the Hanovia 450 W lamp for several hours

without any indication of any products. Some starting material could

be recovered even after several hours. A possible explanation for the

lack of rearrangement could be that unlike (20) the expected diradical

from irradiation would gain no stabilization by a further phenyl shift

in the second step. Also, one cannot formally invoke the di-r-methane

mechanism because the diradical would consist of a tertiary radical and

a vinylic radical, and after a phenyl migration the diradicals are still

tertiary and vinylic in nature. Although the second diradical could

merely close to product, the generation of the second diradical involves

a phenyl migration and there doesn't appear to be any driving force for

this migration, as well as the unfavorable geometry for this phenyl shift.






(38) ----=c-c i =C-
C6Hs 6H5 C6 6H5


CH C H
6 5 6 5


A 6H
C65 C6H5
(21)

The irradiation of phenylethynyltriphenylsilane (39) was also

investigated. The ultraviolet spectrum of (39) in cyclohexane

exhibited maxima at 262(c=32,300), 251(35,500), 239(14,600) and

224(29,500) nm. Solutions of (39) were irradiated at 2537, 3100,

and 3500 Ao, and also with the Hanovia 450 W lamp without any sign of





-45-


product formation by nmr spectra. Analysis of an irradiated cyclohexane

solution of (38) at specific time intervals by glpc indicated a small amount

of product being formed after ten hours, but attempts to isolate or

characterize this compound were unsuccessful. The product had an

extremely long retention time and could easily have been polymeric.

Other attempts to prepare organometallic cyclopropene analogs have
30
failed. Silicon is known not to form i-bonds and the hybridization
2
postulated for cyclopropene a-bonds is much closer to sp hybridization
3
than sp hybridization.

A compound analogous to (20) was studied next. The compound

investigated was 9-phenyl-9-phenylethynylfluorene (40). It exhibited

maxima in isooctane at 306(e=7,800), 294(3,700), 256(34,000), 247.5

(35,000), and 230(33,000) nm in the ultraviolet spectrum. Solutions

of (40) in cyclohexane were irradiated at 2537, 3100, and 3500 A0 and

also with the Hanovia 450 W lamp without any indication of reaction by

nmr spectrum.

It is difficult to rationalize the lack of reactivity of (40)

because of its close resemblance to (20). One possible explanation

might be that the radiant energy is being absorbed by the fluorenyl

chromophrome and is localized in that moiety. The necessary diradical

formation at the propyne moiety would therefore be halted since as

indicated by the ultraviolet spectrum, the fluorenyl group represents a

lower energy sink than the phenylethynyl chromophore.

Due to the lack of rearrangement of (40) another analog of (20)

was prepared at this time. Napthylacetylene (prepared from aceto-
39
napthone) was added to a freshly prepared solution of ethyl magnesium

bromide. The solution was refluxed foi eight hours, and a solution of





-46-


triphenylchloromethane in ether was added. The resulting mixture was

refluxed for fourteen hours and worked up as usual. Chromotography

over Merck alumina yielded 38 percent of colorless crystals mp 144-145.

The nmr spectrum (CDC13) consisted of a one proton mru.tiplet at T (1.70),

a three proton multiple centered at 2.22, and an eighteen proton sym-

metrical multiple centered at 2.68. The mass spectrum showed major

peaks at [m/e(%) ] 394(100), 317(63), 315(35), and 165(36), and the

correct elemental analysis was obtained for C34H22 sutporting the

l-a-napthyl-3,3,3-triphenylpropyne (54) structure.

An inspection of the mass spectrum indicated that on electron impact

a rearrangement toward an indenyl type molecular ion was taking place.

The relatively high abundance of (P-l) ions [m/e 393(,6)] has been in-

dicative of an indenyl moiety rather than a cyclopropenyl moiety, and

the indenyl type of molecular ion would be contributing largely in the

molecular ion. However, the fragment ions are most Likely cyclopropenyl

in nature as seen before.

oH + Hf
lH7 10H7
(54) e6i. H 6H5 C6

-C6H5. / C6H5 6H5
Sm/e 394 \ I. m/(a 393
r C___ T7 I CH, T


~H2


m/e 315


-




-47-


The ultraviolet spectrum of (54) in isoctane exhibited maxima at

319.5(e-13,600), 300(16,900), 288(11,800), and 228.5(71,300) nm.

Solutions of (54) in cyclohexane were irradiated at 2537, 3100 and

3500 AO and monitored by nmr and glpc with no indication of reaction.

Another attempt using the Hanovia 450 W lamp resulted in only recovery

of some starting material with again no indication of any cyclopropene

or indene formation.

Instead of reacting like (20) the napthyl analog was quite stable

like (38) with respect to irradiation. Originally the purpose of syn-

thesizing (54) was to examine what effect a slight change in the phenyl-

propyne chromophore would have on the photochemical reactivity of the

acetylenic molecule. However, the napthyl moiety may be achieving the

same effect as the fluorenyl moiety in the case of (40).

If the napthyl ring (or fluorenyl ring) in (40) acts as a radiant

sink, then it is virtually impossible to pump the required energy into

the triple bond since intramolecular energy transfer would also occur

rapidly and in the direction of the lowest energy site.

The last compound investigated that belongs to Group III was

3-phenylethynyl-l,2,3-triphenylcyclopropene (41). The ultraviolet

spectra of (41) in cyclohexane exhibited maxima at 324(E=22,700),

297(36,000), 292(37,000), 261(41,000) and 226(41,500) nm.

Irradiations were carried out at 2537 and 3500 AO in cyclohexane

with little success. A brown-red viscous oil was usually obtained. An

irradiation at 3100 A showed a singlet in the nmr at T 4.43 which could

be an indenyl proton but attempts to purify this oil failed. Thermolysis

of (41) was attempted at 175-1800 for one hour but yielded another red

oil with a similar nmr spectrum (Fig. 8) to the irradiation product.





-48-


Again attempted purification of this product failed. One possible

explanation for the behavior of (41) is shown below. After re-

arrangement to the indene a 1,5-hydrogen shift would afford the

allenylfulvene derivative (41a) which would be very reactive and

possibly polymerizes under the above conditions.

H CC6H5

C6H5 C

(4l) --> (O T-c -- O 611

C6H5 6H5
(41a)




Polymers




Group IV (C6H5)2ArCC2CAr'

The final series of compounds investigated were the group in which

aryl migratory apptitudes could be observed. The initial member of this

group was 3-p-anisyl-l,3,3-triphenylpropyne (42). The ultraviolet spec-

trum of (42) in cyclohexane exhibited maxima at 256(E=28,600), 245.5

(30,400), and 234(28,400) nm. The irradiation of (42) was very facile

in cyclohexane at 2537 AO and irradiation of dilute (100 mg) solutions

were generally complete in one hour with two major products being detect-

ed in the nmr spectra. Because of the isomeric relationship of the

products, careful spectra were taken of standard samples in benzene

(Fig. 9). The methoxyl protons appeared at 195.5, 198 and 199 cps for

(45), (42), and (44). From careful integration of the product mixture,





-49-


an indication of the migratory apptitudes of the phenyl radical and

the anisyl radical could be determined for the irradiation of (42).

The data indicates that the phenyl radical migrates preferentially

with respect to the anisyl radical (statistically 2.0/1.0) unless (45)

is more reactive than (44), and the irradiation data suggests that

(44) is actually more reactive than (45). This is most likely due to

the extra stabilization rendered by the anisyl group to the developing

radical character at the 3 position. These results are shown in Table

VI (relative error 1%).


C H4OCH3 hv

(C6H5)2CC CC6H5 ---

(42)


C6H5 C6H4OCH3



C6H5 C6HS5


C6H5 C6H5



C6H5 C6H140CH3


(44)

TABLE *VI

RELATIVE PRODUCT DISTRIBUTION IN THE

Time(hrs) (42)(%) (44) (45)

0.25 67 22 11
Run 1 0.50 46 38 16
0.75 27 52 21
1.00 19 58 23

2.00 22 53 25
Run 2 4.00 11 48 28
6.00 8 46 26


(45)


IRRADIATION

Other

0
0
0
0

0
13
20


A larger scale reaction was attempted and irradiated for five hourS.

Chromatography of the product mixture resulted in recovery of (9%) of

a colorless solid, mp 161-163 with ultraviolet absorption maxi:a

at 334, 312.5, 300(sh), and 230 nm and nmr spectra (Fig. 10). This solid


OF (42)

(44/45)

2.00
2.33
2.42
2.48

2.46
1.74
1.77





-50-


was pure 3-p-anisyl-1,2,3-triphenylcyclopropcne (44). An earlier

fraction contained the other cyclopropene isomer (45) (ultraviolet

absorption at 343.5, 326.5, 310(sh), 240(sh), and 220 nm). This

fraction could not be completely purified (small amount of starting

material present). Other minor products were detected in the irradi-

ation mixture which probably were resulting from secondary reactions

of the cyclopropenes.

An investigation into the other isomer, l-p-anisyl-3,3,3-tri-

phenylpropyne (43) revealed that this rearrangement was much slower

than (42). The ultraviolet spectra of (43) in cyclohexane exhibited

maxima at 265(E=35,000), 258(sh) and 255(36,800) nm. One reason for

the longer reaction time could be the shift in the ultraviolet spectrum

(Table VII).


TABLE VII

ULTRAVIOLET ABSORPTION OF REACTIVE PROPYNES

(20) (42) (43)

259.5(E=21,100) 256(e=28,600) 265(c=35,000)
244(25,000) 245.5(30,400) 255(36,800)
234(28,400)



A comparison of (20), (42), and (43) shows that both (20) and (42)

have similar absorptions maxima while the bands for (43) are batho-

chromically shifted about 10 nm. A further inspection into ultra-

violet spectra of (32) and (37) shows that their ultraviolet spectra

also showed this double maxima at 2553 and 2452 nm, These

compounds also underwent rezarr-Igement. The ultraviolet spectra of

compounds such as (40) and (54) exhibited longer wave length maxima





-51-


and failed to photocyclize. This specific type of absorption apparent-

ly is critical for the rearrangement, since it now seems clear that the

radiant energy must be absorbed by the propyne moeity to insure cy-

clization.

A larger scale irradiation of (43) in cyclohexane yielded a small
o
amount of a colorless solid, mp 174-176 The compound was identified

as pure 1-p-anisyl-2,3,3-triphenylcyclopropene (45) (Fig. 11).





C6H5 A6H5
hv hv
(C6H5)3CCECC6H40CH3 --- L Indenes
C6H5 C6H4OCH3
(43) (5)









The cyclopropene analogs were also investigated. The ultraviolet

spectra of 3-p-anisyl-l,2,3-triphenylcyclopropene (44) in cyclohexane

exhibited maxima of 334(e=15,800), 312(22,000), 300(sh) and 230(28,500)
6
nm. This was an efficient way of deciding which cyclopropene isomer

was present because the other isomer had a quite different ultraviolet

spectrum. The ultraviolet spectra of (45) was bathochromically shifted

about 10 nm because of extended conjugation with the anisyl group in the

1 position.

A degassed cyclohexane solution of (44) was irradiated at 3500 A

for twenty-four hours. Evaporation of solvent afforded a yellow oil

(Fig. 12) which contained two maia products along with some minor coim-





-52-


ponents. Chromatography failed to separate these two products,

but one product is postulated as 3-p-anisyl-l,2-diphenylindene (55)

(methoxyl peak enhancement in nmr spectra on addition of authentic sam-

ple). The other is presumably 6-methoxyl-l,2,3-triphenylindene (56)
17
(comparison of methoxyl spike in nmr spectra). The previously proposed

photorearrangement mechanism would predict these two isomers.







C HOCH C H

h O HO O6 3 65
C6H5 6H5 6 CH30-
C6H5 H H C6H5
(44) (55) (56)








The final compound studied was l-p-anisyl-2-3,3-triphenylcyclopro-

pene (45). The ultraviolet spectrum of (45) in cyclohexane exhibited

maxima at 343.5(e=24,700), 326.5(27,200), 310(sh), 240(sh), and 220

(35,000) nm.

A degassed solution of (45) in cyclohexane was irradiated at
0
3500 A for twenty-four hours. Evaporation of solvent gave a yellow

oil which contained one major product and some impurities (Fig. 13).

Chromatography yielded the major product, a yellow oil which failed to

crystallize. By analogy with earlier work, the oil was either l-p-

anisyl-2-3-diphenylindene (57) or 2-p-anisyl--l,3-diphenylindene (58).

The yellow oil was oxidized with chromic anhydride in acetic




-53-


acid. A colorless solid, mp 131-1330, (Fig. 14) was obtained which

corresponded to o-anisoyl-o-benzoyl benzene (59). Therefore the

product from the irradiation was postulated as l-p-anisyl-2,3-diphenyl-

indene (57).




C6H5 C6H CH5 C6H5


C66 H5 6 60CH
C6H 6H40C6H40CH3CH3 40CH
H 6H4OCH3 H C6H5
(45) (57) (58)


Oxidation


% "OCH3




(59)






In this case there is preferential photorearrangement to a single

indene product whereas statistical cleavage of the cyclopropene ring

bonds would decree two products in approximately equal amount. Since

product (58) was not detected in the irradiation of (45), the major bond

breaking occurs between the carbon 1,3-bond and not between the carbon

2,3-bond. Apparently, the anisyl group stabilizes the incipient

radical much more than the phenyl group.

In conclusion it can be said that the photocyclization reaction of





--54--


propynes on irradiation is not general in nature. Of the four various

groups studied some were reactive (Group IV and Group II), but with

complications (addition of hydrogen in Group II). Group III compounds

did not photocyclize and its members were more or less stable to the

irradiation conditions. Group I had some reactive members (17) and

(20) but the other compounds didn't rearrange, and it is evident that

the di-r-methane photocyclization rearrangement of arylpropynes is

very selective with respect to the electronic, and steric character

of the molecule. Unlike the di-T-methane rearrangements involving

olefins, the radiant energy must be involved directly with the

propyne moiety, and without the appropriate extended conjugation,

photocyclization does not appear to occur.
















CHAPTER IV

EXPERIMENTAL

General

Melting points were determined on either a Thomas-Hoover uni-melt

or a Laboratory Devices mel-temp capillary melting point apparatus.

All boiling and melting points are uncorrected. Elemental analysis

were determined by either Galbraith Laboratories, Inc., Knoxville,

Tennessee, or Atlantic MicroLab, Inc., Atlanta, Georgia. The analy-

tical vapor phase chromatography (glpc) was performed with an Aero-

graph Hy-Fi 600-D instrument equipped with hydrogen flame ionization

detector. The preparative vapor phase chromatography was performed with

either a Hewlett-Packard Model 700 or a Varian aerograph model 90-D,

equipped with thermal conductivity detector.

Spectra

Infrared spectra were recorded on either a Beckman IR-10 or a

Perkin-Elmer Infracord spectrophotometer. Ultraviolet and visible

spectra were performed on a Cary 14 recording spectrophotometer.

Nuclear magnetic resonance spectra were determined on a Varian A-60 A

instrument. Mass spectra were determined on a Perkin-Hitachi RMU-6E

instrument.

Photochemical Reactions

Most photochemical reactions were performed in an inert atmosphere

or degassed completely before irradiation. A solution was degassed by

refluxing for at least one hour while purging the vessel wiLt argon.


-55-








The irradiation were performed in either a Rayonet photochemical

ie:ict-or (Southern New England Ultraviolet Co., Middletown, Conn.)

or with a Hanovia 450 W high pressure mercury vapor lamp (Hanovia

Lamp Div., Newark, N.J.). The Rayonet reactor was equipped with

sixteen ultraviolet lamps. The lamps available are RPR 2537 (35 watts),

RIR 3000 (21 vatts), and RPR 3500 (24 watts) A.

The Hanovia lamp radiates a more continuous spectrum (3660-2224)

than the Rayonet and various filter sleeves are available (vycor

7910, coreK 9700, pyrex 7740). The Rayonet reactor had both quartz

and pyre. vessels available for use.

Tetraphenaylpropyre (20)
15
The procedure employed was similar to that of Wieland and Kloss.

A solution of ethyl magnesium bromide was prepared from 1.3 grams

(0.054 g-atom) of magnesium. 3.7 grams (0.034 mole) of ethyl bromide,

and 100 ml of anhydrous ether. Phenylacetylene (5.1 gracs, 0.050 mole)

was added and the resulting solution was reflu:-xe for five hours. A

solution of triphenylchloromethane (12.8 grams, 0.046 mole) in 100 ml

of anhydrous ether was added dropwise over a one hour period. The

solution was then refluxed for twelve additional hours. The product

was obtained by adding a dilute hydrochloric acid solution, separating

the layers, and extracting the water layer with ether (2 x 100 ral).

The extracts were combined and dried over magnesiurM sulfate. A

yellow solid (9.5 grams) was obtained on crystallization (hexane-etner)

of the crude residue. This solid was dissolved in hexane-ether (90/10)

and chromatogiaphed over Merck alumina yielding 8.8 grams (52%) cf
15
colorless crystals, mp 137-138 (Lit. mp 138-139).




-57-


1,-.,1 .- Tri p l _--butne (26)

A solution of ethyl magnesium bromide was prepared from 1.3 grams

(0.054 g--atom) of magnesium, 5.7 gr-as (0.054 mole) of ethyl bromide,

and 100 ml of anhydrous ether. Methylacetylene (2.0 grams, 0.050 mole)

was collected in an adjoining flask. The methylacetylene was then al-

lowed to volatilize into the flask containing the echyl magnesi.um

bromide solution. Employing a dry-ice condeasor, the solution was then

re lxed for eight hours. A solution of triphenylc!loromethane (12.8

grams, 0.046 mole) in 100 m] of anhydrous ether was added dropwise over

a one hour period. The solution was then refluxed for twelve additional

hours. The prod--.t was obtained by adding a dilute hydrochloric acid

solution, separating the layers, and extracting the water layer with e-

tht=z (2 x 100 ml). The extrac:ts were co:ibinedl and dried over magnes-

ium s':. fte. A cride yellcw-orange solid (4.6 =grams was cbtain-'d. o1

crystallization (hnxane-etihe The solid was dissolved in he .:ae-
2 4,2.3
ether (90/10) ard chromatographed over deactivated alumina yield-

ing 3.5 grams (26%) of colorless crystals map 142-1430. The rnmr spec-

trum (CDC13) consisted of a three proton singlet at i 8.03 and a fif-

teen proton multiple centered at 2.79. The mass spectrum (70 ev)

showed peaks at ~/e (Rel intensity) 284(100), 266(18), 267(92), 265(12),

252(10), 205(29), 203(18), 202(13), 196(11), 165(12), 77(8). The ir

(KBr) showed absorption bands at 1600(m), 1490(s), 1445(s), 1185(m),

-1
1070(m), 1035(m), 755(s), 700(s), 640(m), 505(m) cm The uv

(isoociana) exhibited a single maximum at 257(c=785) nm.

Anal. Clcd. for C22 18: C, 93.58; H, 6.42
Found: C, 93.44; H, 6.57





-58-


4-Dimethyl-l,1 i -trihenvl- -pen n (27)

A solution of ethyl magnesium bromide was prepared from 1.3 grams

(0.054 g-atom) of magnesium, 5.7 grams (0.054 mole) of ethyl bromide,

and 100 ml of anhydrous ether. Tertiary-butyl acetylene (4.1 grams;

0.050 mcle) was added and the resulting solution was refluxed for

twenty hours. A solution of triphenylchloromethane (12.8 grams; 0.046

mole) in 100 ml of anhydrous ether was then added dropwise over a one

hour period. The solution was then refluxed for twenty-four additional

hours. The product was obtained by adding a dilute hydrochloric acid

solution, separating the layers, and extracting the water layer with

ether (2 x 100 ml). The extracts were combined and dried over

magnesium sulfate. A crude, slightly yellow solid (6.2 grams) was

obtained on crystallization (hexane-ether). The solid was dissolved

in hexane-ether (90/10) and chromatographed over Merck alumina,

yielding 5.4 grams (33%) of colorless needles, mp 134-1350. The nmr

spectrum (CDC13) consisted of a six proton singlet at T 8.80 and a

ten proton singlet at 2.75. The mass spectrum (70 ev) showed peaks at

m/e (Rel intensity) 324(3), 309(6), 269(10), 268(43), 267(100),

265(6), 215(6), 165(7), 91(11). The ir (KBr) showed absorption bands

at 2235(vw), 1945(w), 1805(w), 1600(m), 1035(m), 895(w), 765(s),

730(m), 700(s), 645(m), 550(m), 515(m) cm-1. The uv (isooctane) ex-

hibited maxima at 266(E=510), 260(725), and 253.5(635) nm (Fig. 16).

Anal. Calcd. for C25H24: C, 92.54; H, 7.46
Found: C, 92.41; H, 7.58

Triphenylacetaldehyde (28)

The desired aldehyde was prepared according to the procedure of
20
A. Cope, P. Trumbull, and E. TruL.bull. Cliro:uitography and fractional





-59-


crystallization yielded pure triphenylacetaidehyde, mp 104-1050
40
(Lit. mp 104-1050). The nmr spectrum (CDC13) consisted of a one proton

singlet at T -0.33 and a fifteen proton singlet at 2.70.

Attempted Reaction of Triphenylacetaldehyde (28) and Carbomethoxy-
methylenetriphenylphosphorane

Triphenylacetaldehyde (28) (1.00 gram, 0.036 mole) and carbo-
41
methoxymethylenetriphenylphosphorane (1.15 gram, 0.036 mole)

were dissolved in 75 ml of N,N-dimethylformamide and heated

at 850 for twenty-eight hours. Water (100 ml) was added and the water-

formamide solution was extracted with ether (2 x 100 ml). After drying

the ether solution over magnesium sulfate, chromatography over alumina

yielded no Wittig product and only 0.23 grams of triphenylacetaldehyde.

Attempted Preparation oc Diethyl carbomeethoxmethylphosohonate (29)
42
A modification of the Arbuzov reaction was carried out using
43
the sodium salt of diethyl hydrogen phosphite. Adding methyl

bromoacetate to the sodium salt (in situ) resulted in a clear liquid

bp 108-117 1.0 mm, which contained six components by glpc analysis.

Further attempts to fractionally distill the mixture failed.

Diethyl carbomethoxymethylphosphonate (29)

Triethyl phosphite (which was stirred with sodium for four days

and distilled from the same flask) and freshly distilled methyl bromoace-

tace were used to prepare the desired phosphonate ester (29). Fractional

distillation yielded pure diethyl carbomethoxymethylphosphonate,
42
bp 102-1040 1.5 mm (Lit. bp 103-1050 1.5 mm).

Reaction of Triphenylacetaldehyde (28) and Diethyl carbomethoxymethyl-
phosphonate (29)
21
Using a procedure sLmilar to that of W. Wads-worth and W. Emmons,




-60-


sodium hydride (0.490 grams, 11.4 mmole, 55% suspension) was added to

200 ml of freshly distilled 1,2-dimethoxyethane. This mixture was

stirred slowly for thirty minutes under a stream of nitrogen. Diethyl

carbomethoxylphosphonate (29) (2.38 grams, 11.4 mmole) was then

added dropwise over a period of one hour. The gray solution was

stirred at room temperature for an additional hour, and the addition

of triphenylacetaldehyde (28) (3.10 grams, 11.4 mmole) over a period of

five minutes caused a color change to orange. The solution was then

refluxed for fourteen hours. The product was recovered by addition

of a five fold excess of water and extraction with ether (3 x 100 ml).

After drying over magnesium sulfate, the ether solution was evaporated,

yielding 2.1 grams of yellow crystals. Chromatography over deactivated
24,25
alumina yielded 1.8 grams (48%) of trans-methyl-4,4,4-triphenyl-
44
but-2-enoate (30) mp 104-1050. The ethyl ester has been reported.

The nmr spectrum (CDC13) consisted of a one proton doublet at T 4.33(J=

16 cps) a one proton doublet at 2.02(J=16 cps), a three proton singlet

at 6.27, and a fifteen proton multiple centered at 2.82. The mass

spectrum (70 ev) showed peaks m/e (Rel intensity) at 328(0.7), 313(4),

297(8), 269(49), 268(100), 267(10), 254(12), 243(14), 192(13), 191(67),

165(31), 105(16), 91(20), 77(8). The ir (KBr) showed absorption bands at

1720(vs), 1640(s), 1595(m), 1490(s), 1445(s), 1370(w), 1300(vs), 1210(s),

1175(s), 1095(w), 1035(m), 995(m), 760(s), 700(vs), 635(m), 600(m),
-1
530(w) cm The uv (cyclohexane) exhibited maxima at 325(e=370)

and 260(3,180) nm.

Anal. Calcd. for C23H2002 : C, 84.12; H, 6.14
Found : C, 84.07; H, 6.31





-61-


Attempt ted z .:.in. Li& i of iMethyl 4,4,4-triphenylbut- 2-enoate C3) in
Carbon Tetrachloride

The ester (30) (0.600 grams, 1.8 mmole) was dissolved in 10 ml

of carbon tetrachloride and 0.300 grams (1.8 mmole) of bromine in

5 ml of carbon tetrachloride was added. The color did not fade so

the reaction was allowed to proceed for fourteen hours. Evaporation

of solvent, yielded 0.452 grams of starting material.

Attempted Bromination of Methyl 4,4,4-triphenylbut-2-enoate (30) in
Acetic Acid

The ester (30) (0.300 grams; 0.9 mmole) was dissolved in 10 ml of

glacial acetic acid and 0.150 grams (0.9 mmole) of bromine in 5 ml

of acetic acid was added. The solution was stirred for four days at

room temperature and then the excess acetic acid was evaporated

leaving a yellow solid. Crystallization from ethanol yielded 0.157

grams (55%) of colorless needles mp 196-197W. The ir (KBr) showed ab-

sorption bands at 1750(s), 1625(m), 1495(m), 1450(s), 1240(s), 1275(s),

1210(s), 1190(m), 995(s), 940(s) cm-1. The nmr spectrum (CDC13) con-

sisted of a one proton singlet at T 3.55 and a fifteen proton broad

singlet at 2.70. The mass spectrum (70 ev) showed peaks at m/e (Rel.

intensity) 312(57), 284(12), 207(16), 183(23), 165(12), 104(64),

102(100), 77(27). The uv (95% ethanol) spectrum exhibited a single

maximum at 272(e=12,600) nm. The structure suggested by this data

is 3,4,4-triphenylbut-2-enoic acid lactone (31).

Anal. Calcd. for C22H1602 : C, 84.59; H, 5.16
Found : C, 84.72; H, 5.14

Preparation of 3,3,3-Triphenyl-l-propene (1)
45
Using the method of R. Greenwald, M. Chaykovsky, and E.J. Corey

sodium hydride (0.320 grams, 7.5 mmole) was added to 20 ml of dry di-

methylsulfoxide. After evacuating and flushing with nitrogen three times,







the. mixt l :as 1-' ~t o ne. Lour. The bLh-.-.- een solution

was cooled and 3.02 ::.s (7.5 mn1 'e) of .ie thtyl riphenylphosphoniumr
i 46 .-,
iodide4 wa added. 1e solutio i :was heaced at for thirty minutes

during wh.ichj time it turned dark ora-nge. Triphenylacet;l,,-r./de (28)

(2.00 gra.:;, 7.35 n t ) was then added and the solution was heated at

75" with sL: -in; for wehnty-four hours. The product was obtained by add-

ing 200 r.l of water, extracting wiLh pentane (3 .: 100 ml), and dryi;nl over
24,25
magnesium sulfate. Chlro ato!a-: over deactivated alumina2 yielded

1.1 grams (55%) of authentic 3,3,3-triphe:y-l->-propene (1), mo 77-780

(Lit.1 mp 75).

Attempted :cmination of 3, i 3-Tri-heny1-l*-proiene (1)

The 3,3,3-triphenyl-l-propene (1) (1.1 grams, 4.1 rmo.le) was dissolved

in 25 ml of carbon tetrachloride, and 0.65 gram (5.1 mmole) of bromine

in 10 mil of carton tetr-chloride was added. After stirring the solution

for four days, the solvent was evaporated leaving a black tar which

could not be crystallized or characterized.

Atte.rtcd Pr,_r.ar ti,:,n :. : t-t -',,'.. --trJ,.--1 r: lhbi -2-enate (1i7

Sodium hydride (2.1 grams, 0.074 mole) was added to 150 ml of

freshly distilled 1,2-dirmethoxyethane, and methyl propiolate (4.2 grams,

0.050 movie) was added dropwise to the stirred mixture. After one hour,

13.3 gramss (0.043 mole) of triphenylchloromethane was added and the

solution was heated at 80' for a period of 24 hours. Water was then

added and the water solution was ex-racted with ether (2 x 100 ml).

After drying over magnesium sulfate, the ether solution was concen-

trated and chromatographed over Merck alumina. Only 7.1 grams of

crude tripht:,,'lchloromethane could be recovered.




-63-


Attempted Prep:ration of Mcthyl 4,4 4-triphenylbut-2-ynoate (17)

A solution of ethyl magnesium bromide was generated from 1.3 grams

(0.054 g-atom) of magnesium, 5.7 grams (0.054 mole) of ethyl bromide and

100 ml of anhydrous ether. The ethyl magnesium bromide solution was

then added dropwise (transferred with a syringe) into an ice-cooled

solution of 4.0 grams (0.048 mole) of methyl propiolate in 50 ml of

anhydrous ether. The solution was stirred for one hour and triphenyl-

chloromethane (12.8 grams, 0.046 mole) was added. After refluxing

for 24 hours, the solution was worked up by adding water, extracting

with ether (2 x 100 ml), and drying the extracts over magnesium sul-

fate. Chromatography over alumina yielded only 6.9 grams of triphenyl-

chloromethane.

Attempted Reaction of Triphenylchloromethane and Lithium Acetylide

Lithium acetylide commercialy available) (3.1 grams, 0.033 mole)

was added to 50 ml of dimethylsulfoxide. Triphenylchloromethane

(9.3 grams, 0.033 mole) was then added and the solution stirred at

room temperature for twenty-four hours. Work up yielded only 6.3 grams

of triplhenylchloromethane.

Attempted Preparation of Methyl 4,4,4-triphenylbut-2-ynoate (17)

Methyl propiolate (4.0 grams, 0.048 mole) was added to 200 ml

of anhydrous ether. After cooling the solution with an ice bath,

30 ml (0.048 mole, 15% solution) of commercial n-butyl lithium was

added via a syringe. The resulting solution was stirred for two hours

at the ice-bath temperature, and then triphenylchloromethane (13.3 grams,

0.048 mole) was added. The reaction was stirred at reflux for thirty-

six hours. After work up as usual, only triphenylchloromethane (6.5 grams)

could be recovered.





-64-


Attempted Preparation of 4,4,4-triphenalbut-2-yn-l-ol

A solution of ethyl magnesium bromide was prepared from 1.3 grams

(0.054 g-atom) magnesium, 5.7 grams (0.054 mole) of ethyl bromide and
47
100 ml of anhydrous ether. The pyranyl ether adduct of propargyl

alcohol (7.0 grams, 0.050 mole) was then added and the resulting

solution was stirred for two hours. Triphenylchloromethane (13.3 grams,

0.048 mole) was then added and the solution was stirred at room temper-

ature for two days. Work up in the usual manner resulted in recovery of

3.7 grams of triphenylchloromethane.

Attempted Preparation of Methyl 4,4,4-triphenylbut-2-ynoate (17j

A solution of ethyl magnesium bromide was prepared from 1.3 grams

(0.054 g-atom) of magnesium, 5.7 grams (0.054 mole) of ethyl bromide and

100 ml of anhydrous ether. After cooling with an ice bath, propional-
48
dehyde diethylacetal (6.4 grams, 0.050 mole) in 100 ml of anhydrous

ether was added. The solution was stirred for six hours at ice bath

temperature and triphenylchloromethane (12.8 grams, 0.046 mole) was

then added. The solution was stirred for eighteen hours at room

temperature and then an additional twelve hours at reflux. Water was

added, the water layer extracted with ether, and the extracts were

dried over magnesium sulfate. Only triphenylchloromethane (4.7 grams)

could be recovered.

1,3,3-Triphenylpropyne (32)
14
The procedure employed was similar to that of Herriot. A solution

of ethyl magnesium bromide was prepared from 1.3 grams (0.054 g-atom) of

magnesium, 5.7 grams (0.054 mole) of ethyl bromide, and 100 ml of

anhydrous ether. Phenylacetylene (5.1 grams, 0.050 mole) was added and

the resulting solution was refluxed for five hours. A solution of




-65-


bromodiphenylmethane (11.8 grams, 0.048 mole) in 100 ml of anhydrous ether

was added dropwise over a one hour period. The solution was then re-

fluxed for twelve additional hours. The product was obtained by adding a

dilute hydrochloric acid solution, separating the layers, and extract-

ing the water layer with ether (2 x 100 ml). The extracts were combined

and dried over magnesium sulfate. A crude, slightly yellow solid (7.2

grams) was obtained on crystallization (hexane-ether) of the residue in an

increase in yellow coloration. Chromatography over deactivated Merck
24, 25
alumina yielded 5.8 grams (45%) of colorless needles, mp 78-790
15
(lit. mp 78-790). The compound yellowed on exposure to light and re-

arranged to the allene on activated alumina. The inmr spectrum (CDC13)

consisted of a one proton singlet at T 4.88 and a fifteen proton sym-

metrical multiple centered at 2.72.

1, 3-DiphenyloroDone (33)
14
The procedure employed was similar to chat of Herriot. A solu-

tion of ethyl magnesium bromide was prepared from 1.3 grams (0.054

g-atom) of magnesium, 5.7 grams (0.054 mole) of ethyl bromide, and 100 ml

of anhydrous ether. Phenylacetylene (5.1 grams, 0.050 mole) was added

and the resulting solution was refluxed for five hours. A solution of

benzyl chloride (6.2 grams, 0.048 mole) in 100 ml of anhydrous ether

was then added dropwise over a one hour period. A catalytic amount

(about 0.5 grams) of both anhydrous cuprous chloride and anhydrous

cupric chloride was also added and the resulting mixture was refluxed

for three days. The product was obtained by adding a dilute hydro-

chloric acid solution, separating the layers, and extracting the

water layer with ether (2 x 100 ml). The extracts were combined and









dried over mavnesiitm sulfate. The solvent was evaporated and the

resulting dark oil was fractionally distilled, yielding 2.8 grams
49
(11%) of a clear liquid, bp 141-144 3.0 nm (Lit. bp 130-1350

2.0 mm). The rmr spectrum (CDC13) consisted of a one proton singlet

at T 6.23 and a five p;oton multiple centered at 2.65.

Preparation of 1,1,3-Tr ihnyl-2-pon-l-ol (34)

A solution of ethyl magnesium bromide was prepared from 2.4 grams

(0.10 g-atom) of magnesium, 1.0.8 grams (0.10 mole) of ethyl bromide and

100 ml of arnhyrous ether. Phenylace.tylene (10.2 grams, 0.10 mole) was

then added dr)pwise over a one hour period. The resulting solution

was si:irred for five hours at reflux, then benzophenone (17.0

grams, 0.093 mole) was added to the solution. Work up in the

usual manner yielded 15.8 grams (60%) of pure 1,1,3-triphenyl-2-propyn-
50
i-cl, p 8-L-82 (Lit. mp 820).

Preparation of 1..3. -Trifhenv1-1-butvne (35)

A solution of ethyl magnesium bromide was prepared from 2.4 grams

(0.10 g-atom) of magnesium, 10.8 grams (0.10 mole) of ethyl bromide, and

100 ml of anhydrous ether. Phenylacetylene (9.1 grams, 0.09 mole) was

then added and the solution was refluxed for five hours. A solution of

1,1-diphenyl-- chloroethane (prepared from 1, -diphenyl-l-ethanol but
51
not purified) (17.3 grams, 0.08 mole) in 100 ml of anhydrous ether

was then added dropwise over a one hour period. A white solid pre-

cipitated out imLediately and the resulting solution was refiuxed for

twelve hours. Work up resulted in a yellow oil which appeared to be a

mixture of a-phenylstyrene and the desired acetylene (nnr). Distillation

of the a-phenylstyrene and chromatography resulted in another yellow oil

but soaIe --p ciylstyrenea persistcd an.d further dJisti nationn of the


-66-




-67-


a-phenylstyrene was attempted and the residual oil rechromatographed over

silica gel yielding 4.3 grams of yellow crystals. Recrystallization

yielded 3.1 grams (14%) of colorless crystals, mp 62-630. The nmr

spectrum (CDC13) consisted of a three proton singlet at T 7.92 and a

fifteen proton multiple from 2.35 to 2.85. The mass spectrum (70 ev)

showed peaks at m/e (Rel intensity) 282(78), 268(23), 267(100),

265(21), 252(12), 203(16), 203(11), 191(11), 189(11), 165(18), 126(10),

77(10). The ir (KBr) showed absorption bands at 2235(vw), 1595(s),
-i
780(m), 760(s), 745(i), 695(vs), 630(m), 590(m), 444(m), 515(m) cm-1

The uv (cyclohexane) exhibited maxima at 253(E=24,000), and

242.5(26,000) nm.

Anal. Calcd for C22H18 : C, 93.58; H, 6.42
Found : C, 93.35; h, 6.56

Preparation of 3-Meti;1-1 .2,3-triphenvycyclopronene (36)

The desired conrmound was prepared by the procedure of R. Breslow
27
and P. Dowd. A solution of methyl magnesium iodide was prepared from

0.96 grams (0.040 g-atom) of magnesium, 5.7 grams (0.040 mole) of methyl

iodide and 100 ml of anhydrous ether. Triphenylcyclopropenium bromide

(4.0 grams, 0.011 mole) was then added and the solution was stirred for

fifteen minutes. Methanol (5 ml) was added to quench any excess grig-

nard and work up as usual yielded a yellow oil. Chromatography over

Merck alumina yielded 2.1 grams (67%) of colorless crystals mp 96-970
27
(Lit. mp 95.5-97.5'3). The nmr spectrum (CDC1 ) consisted of a three

proton singlet at T 3.10 and a fifteen proton multiple from 2.25 to

3.00. The mass spectrum (70 ev) showed peaks at m/e (Rel intensity)

282(100), 268(17), 267(73), 265(20), 251(12), 205(10), 203(11),

202(11), 196(11), 126(13), 77(3). The ir (KBr) shoved absorption at






-68-

1815(m), 1600(m), 1490(s), 1445(s), 1030(m), 930(m), 789(m), 760(s) ,

750(s), 740(m), 700(s), 685(s), 535(m), 480(m) cm-1. The reported uv

spectrum (cyclohexane)27 has maxima at 328(e=25,000), 312(30,000), and

227(31,000) rm.

Preparation of 1,3-Diphenyl3-methyl-l-butyne (37)

A solution of ethyl magnesium bromide was prepared from 2.64 grams

(0.11 g-atom) of magnesium, 12.0 grams (0.11 mole) of ethyl bromide, and

100 ml of anhydrous ether. Phenylacetylene (10.2 grams, 0.10 mole) was

then added and the solution was refluxed for five hours. A solution

of cumyl chloride52 (13.5 grams, 0.09 mole) in 100 ml of anhydrous

ether was then added dropwise over a period of one hour. A white solid

precipitated as the cumy] chloride was added. The mixture was refluxed

for fourteen hours and worked up as usual. Chromatography yielded a

light yellow oil which appeared to be the desired acetylene (nmr) and

fractional distillation yielded 15.3 grams (77%) of a clear liquid, bp

130-1340 0.5-1.0 mm. The nmr spectrum (CDC13) consisted of a six

proton singlet at T 8.23 and a ten proton multiple from 2.25 to 2.90.

The mass spectrum (70 ev) showed peaks at m/e (Rel intensity) 220(38),

206(17), 205(100), 204(10), 203(12), 202(11), 127(17), 119(21), 91(13),

77(17). The ir (neat) showed absorption bands at 1600(m), 1490(m),

1445(m), 1295(m), 1035(m), 760(s), 700(s), 565(m) cm-1. The uv

(cyclohexane) exhibited maxima 252(e=27,600) and 245(26,400) nm

Anal. Calcd. for C17H16 : C, 92.68; H, 7.32

Found : C, 92.56; H, 7.34

Preparation of Tetraphenylallene (38)

Tetraphenylallene was prepared according to the scheme of D.
28
Vorlander and C. Siebert." After refluxing 1,1,3,3-tetraphenyl-2-

bromo-1-propene with alcoholic potassium hydroxide, water was added





-69-


and the water solution was extracted with hexane (3 x 100 ml).

Crystallization from hexane-ether (90/10) yielded authentic tetra-
28
phenylallene, mp 164-1650 (Lit. mp 164-1650).

Preparation of Phenylethynyltriphenysilane (39)
29
Using a procedure similar to Eaborn and Walton, 26 ml (0.04 mole)

of commercial n-butyl lithium was added to a solution of (4.1 grams,

0.04 mole) of phenylacetylene in 100 ml of ether-pentane (4/1). A white

precipitate formed immediately and the mixture was stirred for two hours.

Triphenylchlorosilane (8.0 grams, 0.028 mole) was added along with 100

ml of dry benzene. The ether was distilled from the reaction and the

benzene solution was refluxed for three hours, set aside overnight, and

hydrolyzed with saturated aqueous ammonium chloride. The benzene layer

was separated and dried over magnesium sulfate. Crystallization from

methanol-hexane (10/90) yielded authentic phenylethynyltriphenylsilane
53
(5.8 grams, 55%), mp 101-102 (Lit. mp 100-1010).

Preparation of 9-Phenyl-9-phenylethynylfluorene (40)

A solution of ethyl magnesium bromide was prepared from 1.0 gram

(0.041 g-atom) of magnesium, 4.4 grams (0.041 mole) of ethyl bromide, and

100 ml of anhydrous ether. Phenylacetylene (3.1 grams, 0.030 mole) was

added and the solution was refluxed for five hours. A solution of
54
9-phenyl-9-chlorofluorene (5.5 grams, 0.020 mole) in 50 ml of

anhydrous ether was then added and the resulting solution was refluxed

for twenty-four hours. Water was then added, and the layers were separated,

the water layer extracted with ether (2 x 100 ml), and the extracts

were dried over magnesium sulfate. Crystallization from hexane-ether

(90/10) yielded 4.2 grais of yellow crystals. Chromatography over

Merck alumina yielded 3.3 (43%) grams of colorless needles mp 134-1350





-70-


The nmr spectrum (CDC13) consisted of a two proton multiplet centered

at T 2.71 and a sixteen proton multiple centered at 2.30. The mass

spectrum (70 ev) showed peaks at m/e (Rel intensity) 343(28), 342(100),

341(34), 339(18), 267(2), 266(11), 265(52), 263(13), 163(7). The ir

(KBr) showed absorption bands at 1600(m), 1490(s), 1445(s), 1155(w),

1035(w), 920(w), 765(s), 745(s), 730(s), 695(s), 645(m), 575(m),
-1
535(m), 420(m) cm The uv (isooctane) exhibited maxima 306(e=7,800),

294(6,700), 271(sh), 256(34,600), 247.5(35,000), and 230(33,000) nm

(Fig. 17).

Anal. Calcd. for C H : C, 94.70; H, 5.30
27 18
Found : C, 94.59; H, 5.36

Preparation of 3-Phenylethynyl-l,2,3-triphenylcyclopropene (41)

A solution of ethyl magnesium bromide was prepared from 0.60 grams

(0.025 g-atom) of magnesium, 2.71 grams (0.025 mole) of ethyl bromide, and

50 ml of anhydrous ether. Phenylacetylene (2.1 grams, 0.020 mole) was

then added and the solution was refluxed for five hours. Triphenylcyclo-

propenium bromide (1.71 grais, 0.005 mole) was then added and the solution

was stirred for fifteen minutes. Methanol (3 ml) was added to quench

the excess grignard reagent and the solution worked up as usual. Chrom-

atography over Merck alumina yielded 1.24 grams (67%) of colorless

crystals, mp 152-1530. The nmr spectrum (CDC13) consisted of a complex

multiple from T 2.17 to 2.90. The mass spectrum (70 ev) showed peaks

at m/e (Rel intensity) 368(1), 296(23), 268(25), 267(100), 265(16),

252(10), 203(10), 165(8), 107(19), 94(14), 86(17), 84(26), 77(4). The

ir (KBr) showed absorption bands at 2210(w), 1835(w), 1600(m), 1490(s),

1445(s), 1310(m), 1070(m), 1025(m), 920(m), 785(s), 755(s), 735(s),

685(vs), 600(m), 540(m), 495(o) cm-1. The uv (cyclohexane) exhibited





-71-


n!cxi- at 324(-=22,700), 297(36,400), 292(37,000), 261(41,000),

234(42,500). and 226(41,500) nm.

Anal. Calcd. for C29H20 : C, 94.53; H, 5.47
Found : C, 94.60; H, 5.39

Preparation of l-a-Napthyl-3,3,3-triphenylpropyne (54)

A solution of ethylmagnesium bromide was prepared from 1.44 grams

(0.060 g-atcm) of magnesium, 6.54 grams (0.060 mole) of ethyl bromide, and

100 ml of anhydrous ether, then a-napthylacetylene (prepared from aceto-
39
napthcne) (7.65 grams, 0.050 mole) was added and the resulting

solution was refluxed for eight hours. A solution of triphenylchloromethane

(12.8 grams, 0-048 mole) in 100 ml of anhydrous ether was added drop-

wise over a one hour period. The solution was refluxed for fourteen

hours and worked up as usual. Chromatcgraphy over Merck alumina

(twice) finally yielded 7.4 grams (38%) of colorless crystals np 144-'

145 The nmr spectrum (CDC13) consisted of a one proton multinlet

at T 1.70, a three proton multiple centered at 2.22, and an eighteen

proton symmetrical multiple centered at 2.68. The mass spectrum

(70 ev) showed peaks at m/e (Rel intensity) 394(100), 393(16), 372(13),

318(18), 317(63), 316(21), 315(35), 313(16), 302(13), 291(12), 289(14),

265(9), 244(12), 239(23), 215(12), 182(14), 167(20), 165(36), 158(iA),!

107(12), 105(26), 95(29), 85(12), 77(18). The ir (KBr) showed absorp-

tion bands at 1600(m), 1590(m), 1490(s), 1445(s), 1395(m), 1185(m),

1080(w), 1035(m), 805(s), 780(s), 760(s), 700(s), 640(m), 585(m),
-1
565(m), 515(w), 460((w) cm-. The uv (isooctane) exhibited maxima at

319.5(e=13,600), 300(16,900), 288(11,800), and 228.5(71,300) nm (Fig. 18).

Anal. Calcd. for C34H22 : C, 94.85; H, 5.15
Found : C, 94.73; H, 5.25





-72-


Preparation of Diphenvl-p-anisvlcarbinol

A solution of p-anisyl magnesium bromide was prepared from 6.0

grams (0.25 g-atom) of magnesium, 46.8 grams (0.25 mole) of p-bromoan-

isole, and 100 ml of anhydrous ether. A solution of benzophenone (27.3

grams, 0.15 mole) in 100 ml of anhydrous ether was added over a one

hour period. The solution was stirred for an additional four hours.

Work up in the usual manner yielded a yellow oil (29.3 grams, 71%)
55
which would not crystallize. The oil was used in the preparation of

chloride.

Preparation of Diphenyl-p-anisylchloromethane

The yellow oil (29.3 grams) from the previous experiment was dis-

solved in 150 ml of anhydrous benzene. Anhydrous hydrogen chloride was

bubbled through the solution for one hour. Evaporation of the solvent

and crystallization of the yellow oil from hexane-ether (90/10) yielded
56
26.7 grams (83%) of colorless crystals, mp 121-1220 (Lit. a~ 122-1230).

Preparation of 3-p-Anisvl-l,3,3-tripheniylropvne (42)

A solution of ethyl magnesium bromide was prepared from 1.7 grams

(0.070 g-atom) of magnesium, 7.6 grams (0.070 mole) of ethyl bromide, and

100 ml of anhydrous ether. Phenylacetylene (6.1 grams, 0.060 mole) was

added and the resulting solution was refluxed for five hours. A solution

of diphenyl-p-anisylchloromethane (14.7 grams, 0.050 mole) in 100 nil of

anhydrous ether was added over a one hour period. The solution was re-

fluxed for fourteen hours and worked up as usual. A yellow oil was ob-

tained which would not crystallize, however, chromatography over Merck

alumina afforded 15.9 grams (85%) of colorless needles, mp 108-109.

The nmr spectrum (CDC13)consisted of a three proton singlet at T 6.29

and a nineteen proton multiple from 2.40-3.40. The mass spectrum (70 ev)





-73-


showed peaks at m/e (Rel intensity) 374(20), 296(8), 273(100), 272(94),

242(25), 198(13), 197(79), 196(10), 195(16), 176(22), 175(53), 153(17),

152(19), 77(11). The ir (KBr) showed absorption bands at 1600(m), 1505(s),

1490(s), 1465(m), 1445(s), 1295(m), 1245(s), 1185(s), 1030(s), 920(w),
-1
835(s), 760(s), 750(s), 695(s), 640(m), 595(m), 555(m), 535(w) cm The

uv (cyclohexane) exhibited maxima at 256(e=28,600), 245.5(30,400), and

234(28,400) nm.

Anal. Calcd. for C28H220 : C, 89.81; H, 5.92
Found : C, 89.70; H, 5.89

Preparation of 1-p-Anisyl-3,3,3-triphenyl-l-propyne (43)

A solution of ethyl magnesium bromide was prepared from 1.44 grams

(0.060 g-atom) of magnesium, 6.54 grams (0.060 mole) of ethyl bromide, and
57
100 ml of anhydrous ether. P-Anisylazatylene (6.60 grams, 0.050 mole)

was added and the resulting solution was refluxed for five hours. A solution

of triphenylchloromethane (11.12 grams, 0.040 mole) in 100 ml of an-

hydrous ether was added dropwise over a one hour period. The solution

was then refluxed overnight and worked up in the usual manner. Chroma-

tography over Merck alumina yielded 9.0 grams (60%) of colorless crystals,

mp 144-145 The nmr spectrum (CDC13) consisted of a three proton sing-

let at 1 6.13 and a nineteen proton complex multiple from 2.48 to 3.28.

The mass spectrum (70 ev) showed peaks at m/e (Rel intensity) 374(100),

359(14), 343(6), 297(40), 267(6), 265(10), 253(11), 252(13). The ir

(KBr) showed absorption bands at 1605(m), 1510(s), 1490(m), 1450(m),

1290(m), 1250(s), 1175(m), 1030(m), 840(m), 765(m), 700(s), 640(m),
-1
550(m) cm The uv (cyclohexane) exhibited maxima at 265(e-35,000),

258(sh), and 255(36,800) nm.

Anal. Calc. for C28H220 : C, 89.81; H, 5.92
Found : C, 89.72; H, 6.04





-74-


Preparation of 3-p-Anisyl-l,23-triphenIcycloDropene (44)

A solution of p-anisyl magnesium bromide was prepared from 0.48

grams (0.020 g-atom) of magnesium, 3.74 grams (0.020 mole) of p-bromoan-

isole, and 50 ml of anhydrous ether. Triphenylcyclopropenium bromide

(1.70 grams, 0.005 mole) was then added slowly, and the resulting solu-

tion was stirred for fifteen minutes. Methanol (5 ml) was used to

quench the excess Grignard reagent and then 50 ml of water was added.

The solution was then extracted with ether 2 x 50 ml, and the ex-

tracts were combined and dried over magnesium sulfate. Chromatography

over Merck alumina yielded 1.4 grams (75%) of colorless crystals mp 163.5-
6
164.50 (Lit. mp 162-1630). The nmr spectrum (CDC 3) consisted of a

three proton singlet at T 6,29 and a nineteen proton complex multiple

from 2.15 to 3.32. The mass spectrum (70 ev) showed peaks at m/e (Rel

intensity) 374(100), 359(11), 297(28), 281(10), 265(16), 254(11), 253(18),

252(23), 148(9). The ir (KBr) showed absorption bands at 1815(w),1610(m),

1595(m), 1510(s), 1490(s), 1465(m), 1445(s), 1295(m), 1245(s), 1180(m),

1030(m), 835(m), 785(m), 755(s), 705(s), 690(s), 595(m), 565(w), 545(w),
-1
em The uv (cyclohexane) exhibited maxima at 334(e=15,800), 312(22,800),

300(sh), and 230(28,500) nm.

Anal. Calcd. for C28 H 220 : C, 89.21; H, 5.92
Found : C, 89.35; H, 5.82

Reaction of Phenvl Magnesium Bromide with Diphenyl-p-anisylcycloprop enyl
Bromide

A solution of phenyl magnesium bromide was prepared from 0.60 grams

(0.025 g-atcm) of magnesium, 4.0 grams (0.025 mole) of bromobenzene, and

50 ml of anhydrous ether. Diphenyl-p-anisylcyclopropenium bromide (3.1
58
grams, 0.008 mole, available from F. Haupt) was then added. The solution





-75-


was stirred for ten minutes and then quenched with 3 ml of methanol.

Water (50 ml) was then added and the layers were separated. The water

layer was extracted with ether (2 x 50 ml), and the extracts were com-

bined and dried over magnesium sulfate. An initial nmr spectrum indi-

cated both isomers were present (95/5 relative amounts) and chromatography

over Merck alumina yielded 1.1 grams (36%) of colorless crystals mp 176.5-'

177.5 The nmr spectrum (CDC13) consisted of a three proton singlet

at T 6.25 and a nineteen proton complex multiple from 2.21 to 3.22.

The mass spectrum (70 ev) showed peaks at m/e (Rel intensity) 374(100),

360(11), 359(32), 298(16), 297(31), 265(20), 254(12), 253(16), 252(18),

226(10), 165(16), 155(21), 148(39), 141(29), 126(10), 105(16), 91(11),

77(14). The ir (KBr) showed absorption bands at 1810(w), 1600(s),

1570(m), 1500(s), 1490(m), 1465(m), 1445(s), 1355(m), 1250(s), 1170(s),

1075(m), 1035(s), 830(s), 780(m), 760(m), 740(m), 725(m), 700(s), 685(m),
-1
620(m), 575(m), 515(m) cm The uv (cyclohexane) exhibited maxima

at 343.5 (e=24,700), 326.5(27,200), 310(sh), 240(sh), and 220(35,000)

nm. The compound was identified as l-p-anisyl-2,3,3-triphenylcyclopropene

(45).

Anal. Calcd. for C28H220 : C, 89.21; H, 5.92
Found : C, 89.22; H, 5.87

Photolysis of Tetraphenylpropyne (20)

The photolysis of tetraphenylpropyne (in cyclohexane, 2537 A ) for

short periods of time (3.4 hours) yields tetraphenylcyclopropene (46%).10

Irradiation for longer periods (24 hours) results in the formation of

1,2,3-triphenylindene (42%) and 13-phenyl-13H-indeno-(1,2-1) phenanthrene

(19%). Attempts to sensitize the rearrangement with acetone, aceto-

phenone, and benzophenone at 3100 and 3500 A0 resulted in lack of a





-76-


rearrangement.

Photolysis of l,l,l-Triphenyl-2-butyne (26)

A solution containing 0.290 grams (1.0 mmole) of 1,1,1-triphenyl-

2-butyne in 100 ml of cyclohexane (argon purged) was irradiated at

2537 AO. The solution was periodically checked by glpc for any in-

dication of a reaction. After three hours, the starting material

remained unchanged, and the solution was then irradiated with the

450-W Hanovia Lamp for an additional five hours without any indi-

cation of a reaction (glpc and nmr). Some starting material (0.236

grams) was recovered.

Photolysis of l,l,l-Triphenyl-4,4-dimethyl-2-pentyne (27)

A solution containing 0.323 grams (1.0 mmole) of 1,1,1-tri-

phenyl-4,4-dimethyl-2-pentyne in 100 ml of cyclohexane (argon

purged) was irradiated at 2537 A. The solution was periodically check-

ed by glpc for any indication of product formation. After seven hours,

the starting material remained unchanged, and the solution was then

irradiated with the 450-W Hanovia Lamp for an additional five hours with-

out any indication of reaction (glpc and nmr). Some starting material

(0.257 grams) was recovered.

Photolysis of 1,3,3-Triphenyl-l-propyne (32)

A solution containing 0.844 grams (3.0 mmole) of 1,3,3,-

triphenyl-l-propyne in 800 ml of cyclohexane was irradiated at 2537 AO

(large prep reactor). The photolysis was monitored by nmr at specific

time intervals. After approximately six hours, the photolysis was

stopped and removal of solvent afforded a light brown oil. The oil was

analyzed by glpc and showed a number of components present. Using prep-

aratory glpc [(;" x 5') 15% FFAP on 60/80 Chromosorb W column, inj.





-77-


255 det. 2500, col. 2150] four major products were separated and

three were identified by ir, nmr and mass spectral comparison to

authentic compounds. The reaction mixture contained diphenylmethane,

benzoic acid, and benzophenone. The products ostensibly resulted from

oxidative cleavage of the original propyne. Further irradiations were

either degassed completely or performed in an argon purged atmosphere.

Photolysis of 1,3,3-Triphenyl-l-propyne (Degassed)

A degassed solution containing 0.336 grams (1.25 mmole) of pure

1,3,3-triphenyl-l-propyne in 125 ml of cyclohexane was irradiated at

2537 Ao. The reaction was monitored by glpc [(1/8" x 5') 7% SE 30

on 60/80 Chromosorb W column, inj. 2650, det. 2500, col. 2350] at

specific time intervals. After three hours of irradiation, the

reaction mixture contained some starting material (18%), three minor

products (20%), and one major product (62%). The photolysis was ter-

minated after four hours. The reaction mixture now contained start-

ing material (11%), three minor products (20%), and one major product

(69%). Evaporation of solvent yielded a yellow oil. Preparative gas-

liquid chromatography [(0" x 5') 15% FFAP on 60/80 Chromosorb W column,

inj. 2500, det. 2500, col. 2250 ]yielded 0.076 grams (85% pure by glpc)

of a slightly yellow oil which would not crystallize. The nmr spectrum

(CDCl3) consisted of a singlet at T 7.20 and a multiple from 2.50 to

3.31 (Fig. 1). The mass spectrum (70 ev) showed a parent ion at (m/e

270). The major product was originally suspected to be 1,1-diphenyl-
34
indane (46) because of the similarity of the nmr spectra but a comp-

arison between the authentic compound (46) and the irradiation product

gave different glpc retention times. The photoproduct was finally

identified as trans-1,2,3-triphenylcyclopropane (48) by comparison of





-78-


their nm~- .:~"tra (Fig. 2,3), mass spectra, and identical retention

times on glpc (mixture of photoproduct and authentic sample gave one

S','.,t,-:icai peak).

rll'r *- .t_ '. Of ], 7 T' ^ i '. liir-' .ie ( ,6")

The desired compound was prepared by the method of W.H. Starnes

using the TWolf-Kishner reduction of the diphenylindanone. Pure
59
l,l-d, i,,.:.i-indane (46), mp 69-70 (Lit. mp 67-68) was obtained.

Pr-:' T, i.-' : o Tr- -i -- 1,2, -Tr- .-benr.-1-c1hl.0C-L.'cl,,prro p O,1c (47)

The procedure employed was mentioned by not described by R. Breslow
27
and P. Dowd. Trans-stilbene (3.6 grams, 20.0 mmole) and potassium-

t-butoxide (4.5 grams, 40.0 mole) were dissolved In 400 ml of 1dr'

benzene and the solution was stirred rapidly with a mechanical stirrer.

A solution of benzal chloride (6.5 grams, 40.0 1-mole) in 400 ml of dry

benzene as added over a period of onr hour. The 'v. ry viscous sol iion

was then refluxed with stirring for five hours. Water was added (200 ml)

and the benzene layer was separated. The water layer was extracted with

ether (2 x 50 ml) and the extracts were combined with the benzene

layer and dried over mancsiumr. sulfate. The brown oil was chromato-

graphed over silica gel yielding 2.1 grams of trans-stilbene and 0.35

gram (14%) of a light yellow solid (Fig. 4) vhich appeared to be

trans-1,2,3-triphE, nl-l-chlorocyclopropane (47). Earlier attempts

(using a three-fold excess potassium-t-butoxide) to prepare the

chlorocyclopropane resulted in the formation of triphenylcyclopropene

in 10% yield (identified by nmr comparison with authentic triphenyl-

cyclopropene).

Prep-:L rr- ion of Trans-12,3-Tri hanv ycycloro ne (48)

The authentic trans-1,2,3-triphenylcycloproc eLi was prepared by a








method s.iLuilar -o tr1,t of i, i.'wd and -:. rsi.ow. 'ie trans-,2,3-

triphbcnyl-l-chlorocycloprop. .- ((47) (0.3"72 gra:ns, '.0J2 mole) ,;as

placed in a flash along wit-h 1 0 mlt of a-. ,:']rous ethei., and 0.72

(0.030 g-atom) of wag-esium :. small aou::1t (0.5 7il) cf 1. 2-diblomo-

ethace was used to initiate the reaction jhlich was wLJLmd also by a

heat lamp. After the reaction started the solution was refluxed for

one hour, and a dilute acid solution was added. A brown oil was ob-

tained which was chromatograpihed over silica gel yielding 0.262 grams

(80%) of authentic trans-1,2,3-tripheny]cyclo-.'-pop:ni,. (48), mp 65-66
60
(Li:. mp 630).

Phort.o- .<1 .,* 1,3-L ._,.. ,'l-1.-_, -'. :. (' o2)

A solution containing 0.100 grams (5.5 mmole) of 1,3-diphenyl-l-

propyne in 50 ml of cyclohexane was irradiated at 2537 AC for five

hours. Most of th- starti-g rT-terial (78%) had disappeared (n:mn) but no

products could be isolated or characterized.
Phou!,;. j: j- _.j 1, ?. Iw.-" -.. -?-ijj. jprn-!-: l (3A) i.i lfiric,_ Ac-;d

A solution of 0.012 grams (0.041 mmole) of l,l13-triphenyl-2-

propyn-1-ol (34) in 1 ml of methylene chloride was added to 100 ml of

rapidly stirring concentrated sulfuric acid. A deep red solution was

immediately formed which exhibited visible maxima at 512(c=33,000) and

447(28,400) nm. The solution was then irradiated at 3100 A and the

reaction was monitored by uv and visible spectra at specific time inter-

vals. After four hours the original maxima had disappeared and a new

maximum appeared at 466, but no other maxima were between 240 and 600 nm.

Another irradiation was attempted using nmr to monitor the

reaction. A solution containing 0.0545 grams (0.19 mmole) of 1,1,3-

triLh eri;l-2-propyn-l-ol (34) in 0.2 ml of carbon tetrachloride was





-80-


added with rapid stirring to 0.5 ml of concentrated sulfuric acid. The

solution was placed in an nmr tube (tetramethylammonium fluoborate as

internal standard) and irradiated at 3100 AO. The nmr spectrum was taken

at specific time intervals up to three and one half hours. The spectrum

was definitely changed but both the uv and nmr spectra indicated the ab-

sence of the triphenylcyclopropenium cation.

Photolysis of 1,3,3-Triphenyl-l-butyne (35)

A solution of 0.0557 grams (0.21 mmole) of (35) in 100 ml of cyclo-

hexane (argon purged) was irradiated at 2537 A0. The photolysis was

monitored by nmr and glpc which indicated formation of at least twelve

products after only one hour. Another solution was irradiated in ben-

zene for nineteen hours without any indication of any major new pro-

duct formation (glpc). Most of the products were most likely initiated

by reduction in cyclohexane. No further attempts to isolate or charac-

terize the products were attempted.

Photolysis of 3-Methyl-l,2,3-triphenylcvclcpropene (36)

A degassed solution containing 0.2020 grams (0.71 mmole) of 3-methyl-

1,2,3-triphenylcyclopropene (36) (Fig. 4) in 100 ml of cyclohexane was

irradiated at 3500 A for twenty seven hours. Solvent was removed and

the nmr spectra indicated complete absence of starting material. The

yellow oil was chromatographed over silica gel yielding a yellow oil

(0.1458 grams) and a slightly yellow solid (0.0437 grams). Crystal-

lization of the yellow oil from (95%) ethanol yielded 0.0938 grams (46%)

of a very slightly yellow solid. The nmr spectra (CDC13) consisted of a

three proton quartet at T 7.80(J=2 cps), a very poorly resolved one pro-

ton quartet at 5.44(J=2 cps), and fourteen proton multiple from 2.55

to 3.05 (Fig. 5). The product was identified by its nmr and its mp




-81-


61
89-90 (Lit. mp 910) as 3-methyl-1,2-diphenylindene (49). The

second fraction was recrystallized from (95%) ethanol yielding

0.0326 grams of a colorless solid mp 304-3050. The nmr spectrum (CDC13)

(Fig. 6) consisted of a three proton singlet at T 8.07, a one proton

singlet at 5.02, a two proton multiple centered at 3.95, a ten

proton symmetrical multiple from 2.90 to 3.40, and a two proton

multiple centered at 2.50. The mass spectrum showed a parent peak at

564 (<1%) while the remainder of the spectrum closely resembled that of

the starting material (36). The compound is postulated as the head to

tail dimer of 3-methyl-1,2-diphenylindene (50).

Photolysis of 3-Methyl-l,2-diphenylindene (49)

A degassed solution containing 0.048 grams (0.17 mmole) of

3-methyl-1,2-diphenylindene (49) in 100 ml of cyclohexane was irradiated

at 3500 A for twenty-two hours. After removal of solvent, the nmr

spectrum (CDC13) indicated a mixture of 3--methyl-l,2-diphenylindene (49)

and the same product from the irradiation of 3-methyl-1,2,3-triphenyl-

cyclopropene (36). Chromatography over silica gel yielded some starting

material (a yellow oil, 0.035 grams) and 0.007 grams of a colorless

solid mp 303-304.50 (50).

Thermolysis of 3-Methyl-1,2,3-triphenylcyclopropene (36)

After sealing 0.0427 grams (0.15 mmole) of 3-methyl-1,2,3-

triphenylcyclopropene (36) in a glass tube at 0.5 mm, the tube was

immersed in a silicone oil bath at 235-2400 for four hours. The tube

was cooled and broken yielding a yellow oil which contained a mixture

of 3-methyl-1,2-diphenylindene (49) (55%) and l-methyl-2,3-diphenyl-
62
indene (51) (45%), its thermal isomer (Fig. 7).

Photolysis of 1,3-Diphclnyl-3--uethyl-l-butyne (37)




-82- r-



A solution of 0.062 grams (0.28 mmole) of 1,3-diphenyl-3-methyl

1-butyne in 100 ml of cyclohexane (argon purged) was irradiated at 2537

A The photolysis was monitored by glpc [(1/8" x 5') 5% FFAP on 60/80

Chromosorb W column, inj. 220 det. 220 col. 18001. After one hour

the starting material was completely consumed and one large product peak

had appeared. Attempts to irradiate the compound in benzene resulted in

no reaction.

A large scale irradiation was performed on 0.31Q grams of starting

material in 100 ml of cyclohexane at 2537 A After eight hours all of

the starting material had been consumed (glpc). The yellow oil contained

approximately 54% of one component along with some i:ipurities from the

starting material. The main product was obtained from prep. glpc (0.065

grams of a clear liquid) [(4" x 5') 10% FFAP on 60/80 Chromosorb W

column, inj. 220 det. 220 col. 1850 ]. The nmr spectrum (CDC13),

Fig. 8, consisted of a two proton singlet at T 7.65, a six proton

singlet at 9.01, and a ten proton broad singlet at Z.S1. The mass

spectrum (70 ev) showed peaks at m/e (Rel intensity) 222(86), 207(75),

179(23), 178(28), 165(13), 131(73), 130(14), 129(10G:), 128(22), 115(23),

105(22), 91(65), 77(20). The ir (neat) showed absorption bands at

1605(m), 1500(m), 1450(m), 1380(m), 1120(m), 1030(mO 800(m), 760(m),
-1
730(m), 700(s) cm The compound was identified as trans-1,2-diphenyl-3,3-

dimethylcyclopropane (52).

Anal. Calcd. for C17H-8 : C, 91.84; H, 8.16
Found : C, 91.74; H, 8.17

Attempted Preparation of Trans-1,2-Diphenyl--3,3-dinmethylcyclopropane (52)

Using an analogous method employed in the preparation of





-83-


triphenylcyclopropane, 2.0 grans (15.0 mmole) of 3,B-dimethyl styrene

and 4.5 grams (40.0 mmole) of potassium-t-butoxide were dissolved in

500 ml of dry benzene and stirred rapidly. A solution of benzal

chloride 6.5 grams (40.0 mmole) in 200 ml of dry benzene was added

dropwise over a one hour period. The solution was then refluxed for

five hours. Work-up in the usual manner yielded a brown oil which did

not appear to be the desired compound (nmr). No attempts were made to

chaacterize the oil.

Another attempt to prepare the desired cyclopropane was performed

by irradiating a mixture of B,B-dimethylstyrene and phenyldiazomathane

at 3500 AO. There was no indication of cyclopropane formation (nmr).

Hydrogenation of 1,3-Diphenyl-3-methyl-l-butyne (37)

A catalytic amount (0.10 grams) of 5% Pd/BaSO4 in 50 ml of hexane

was added to 2.20 grams of 1,3-diphenyl-3-methyl-l-butyne (37). The

stirred mixture was then hydrogenated for twelve hours. Filtration and

subsequent work-up yielded a yellow oil (2.31 grams). There was appa-

rently some starting material still present (nmr) but an analytical

glpc indicated the possibility of separating the product from the start-

ing material. Using prep. glpc [(" x 5') 15% FFAP on 60/80 Chromosorb

W column, inj. 2100, det. 2100, col. 1800], a small amount of olefin

(0.0423 grams) was collected. The nmr spectrum (CDC13) consisted of a

six proton singlet at I 8.64, a one doublet centered at 4.08(J=12 cps),

a one proton doublet centered at 3.45(J=12 cps), and a ten proton

symmetrical multiple from 2.50 to 3.20. The product was postulated as

cis-1,3-diphenyl-3-methyl-]-butene (53).

Photolysis of Cis-l,3-Diphenvl-3-methvi-l-butene (53)

The olefin (53) (0.0423 grams) was added to 50 ml of cyclohexane





-84-


and irradiated at 2537 AO(argon purged) for one hour. Removal of

solvent left a yellow oil which had an identical nmr spectrum to that

of the propyne (37) photoproduct. A comparison of the retention times

[(1/8" x 5') 5% FFAP on 60/80 Chromosorb W column, inj. 2150, det. 2200,

col. 1800 ]showed identical retention times (separate and combined) for

the two photoproducts.

Photolysis of Tetraphenylallene (38)

A solution containing 0.0651 grams (0.19 mmole) of tetraphenyl-

allene in 50 ml of cyclohexane (argon purged) was irradiated at 2537 AO.

The reaction was monitored by nmr at specific time intervals, but after

fourteen hours there was still no apparent reaction.

A solution containing 0.1235 grams (0.36 mmole) of tetraphenyl-

allene in 100 ml of cyclohexane (nitrogen purged) was irradiated with

Hanovia-450 W lamp and monitored by nmr at specific time intervals.

After eight hours no apparent reaction could be detected and only

0.0833 grams of starting material were recovered.

Photolysis of Phenylethynyltriphenylsilane (39)

A solution containing 0.1114 grams (0.33 mmole) of phenylethynyl-

triphenylsilane in 100 ml of cyclohexane (argon purged) was irradiated

at 2537 Ao. The reaction was monitored by glpc at specific time

intervals. After seven hours of irradiation no products were observed,

but after ten hours a small amount of product appeared. Attempts to

characterize this highly retained product were unsuccessful.

Photolysis of 9-Phenyl-9-phenylethynylfluorene (40)

A solution of 0.0589 grams (0.17 mmole) of 9-phenyl-9-phenyl-

ethynylfluorene in 100 ml of cyclohexane (argon purged) was irradiated

at 2537 A. The photolysis was monitored by nmr at specific time









intervals. After fifteen hours, there was no indication of a reaction

according to the nmr spectrum. Another solution was irradiated at 3100

A for twenty-four hours with no apparent reaction.

A solution of 0.1270 grains (0.36 mmole) of (40) in 100 ml of cyclo-

hexane (nitrogen purged) was irradiated with the Hanovia-450 W lamp for

ten hours. Only 0.0721 grams of starting material could be recovered.

Photolysis of l-a-Napthyl-3,3,3-triphenyl-l-propyne (54)

A solution containing 0.0763 grams (0.19 mmole) of l-a-napthyl-3,3,

3-triphenyl-l-propyne in 100 ml of cyclohexane (argon purged) was irrad-

iated with the 2537 A0 lamps and monitored by nmr at specific time

intervals. After eleven hours, there was still no indication of a

reaction. Other irradiations at 3100 and 3500 Ao also failed to produce

any indication of a reaction.

A solution containing 0.0835 grams (0.21 mole) of (54) in 100 ml

of cyclohexane (nitrogen purged) was irradiated with the Hanovia-450 W

lamp for eleven hours, and only starting material (0.0476 grams) was

recovered.

Photolysis of 3-Phenylethynyl-1,2,3-triphenylcyclopropene (41)

A degassed solution containing 0.2035 grams (0.55 mmole) of 3-

phenylethynyl-l,2,3-triphenylcyclopropene in 100 ml of cyclohexane was

irradiated at 3500 AO for twenty-four hours. Evaporation of solvent

yielded a red-brown oil which exhibited just a very broad ill defined

aromatic signal in the nmr spectrum.

Another irradiation was carried out at 3100 Ao on a solution of

(41) (0.1831 grams in 100 ml of cyclohexane) and monitored by nmr at

specific time intervals. After eighteen hours, a small signal at T

4.93 appeared in the nnir spectrum along with the broad aromatic region.


-85-





-86--


Attempts to isolate any products from the irradiation failed.

Thermolysis of 3-Phenylethyny]-1,2,3-triphenylcyclopropene (41)

After sealing 0.1545 grams (0.42 mmole) of 3-phenylethynyl

1,2,3-triphenylcyclopropene in a glass tube at 0.5 mm the tube was

immersed in a silicon oil bath at 175-1800. After one hour, the tube

was cooled and a red-brown oil had formed. The nmr spectrum (CDCl3) con-

sisted of a singlet at T 4.88 and a complex aromatic multiple from 1.95-

2.95 (Fig. 8). Attempts to purify and isolate this product by chromato-

graphy failed to yield any solid.

Photolysis of 3-p-Anisyl-l,3,3-triphenyl-l-propyne (42)

A solution of 0.0507 grams (0.136 mmole) of 3-p-anisyl-l,3,3-tri-

phenyl-l-propyne in 100 ml of cyclohexane (argon purged) was irradiated

at 2537 A0. At fifteen minute time intervals the photolysis was stopped

and the nmr spectrum was taken. The suspected products, as pure com-

pounds, were previously combined (in varying amounts) and examined by

nmr (Fig. 9 ) for comparative analysis. The methoxy proton signals

could be separated using benzene as a solvent in the nmr spectrum, with

the protons appearing at 195.5, 198, and 199 cps for (45), (42) and (44)

respectively. After fifteen minutes there was approximately 22% of

3-p-anisyl-1,2,3-triphenylcyclopropene, 11% of l-p-anisyl-2,3,3-tri-

phenylcyclopropene, and 67% of starting material. The photolysis was

very fast and after sixty minutes there remained only 19% of starting

material and 58% of the 3-p-anisyl-l,2,3-triphenylcyclopropene along

with 23% of l-p-anisyl-2,3,3-triphenylcyclopropene. Irradiations were

carried out on other samples for longer time periods and after four or

five hours other products (indenes) started to appear in greater amounts.

A solution of 0.2873 grams (0.770 mmole) of (42) was dissolved in









100 i11 of cyclohLxane. The solution was. riegasseC for one hour- and ir-

radi atc at 2537 A for five hours. After evaporaL.ion of solvent, a

yelo:'.:; oil vwas obtained. The nmr spectrum indicaLed that both cyclo-

propencs and ,tarting material were pres-nt with very little, if any,

indenes. The oil was chromatographed over Merck alumina elitting with

ben..er c-hexanc (3-97). Fractions of 50 !:ml were collected. Nmr ex-

amination of the first product containing fraction indicated that it

contained both l-p-anisyl-2,3,3-criphenylcyclopropene (45) and start-

ing material (42). A qualitative uv (cyclohexane) indicated maxima at

343, 326, 310(sh), 256, 245, and 233 nm. Attempts to fractionally

crystallize this fraction failed. Later fractions contained both cyclo-

propenes and starting material. One of the final fractions appeared

(nmr) to contain only cyclopropene products. Fractional crystallization of

this -ate:rial f .rom. b en-h exan e yi lded 0.02 1 grams of colorless cr y-
6
stalls, mp 161-1630 (Lit. mp 162-1630). The uv (qualitative, cyclo-

hexane) showed absorption at 334, 312.5, 300(sh), and 230 no while the

nmr indicated pure 3-p-anisyl-l,2,3-triphenylcyclopropene (44) (Fig. 10).

Photolysis of l-p-Anisyl-3,3,3-triphenyl-l-propyne (43)

A solution of 0.0543 grams (0.145 mmole) of l-p-anisyl-3,3,3-triphenyl-

l-propyne in 100 ml of cyclohexane was irradiated at 2537 Ao (argon purg-

ed). The photolysis was monitored by nmr at specific time intervals.

The formation of l-p-anisyl-2,3,3-triphenyicyclopropene (45) was indi-

cated; however, the other products were being produced at these same

time intervals as this photolysis was much slower than the other pro-

pyne isocer (nmr).

A degassed solution of 0.2473 grams (0.665 mmole) of l-p-anisyl-

2,3,3--triphenyl-l-propyne in 100 ml of cyclohexane was irradiated at


-S7-





-88-


2537 A and monitored by nmr. After eight hours there was approximately

56% of starting material, 28% of l-p-anisyl-2,3,3-triphenylcyclopropene,

and 16% of other products. The photolysis was continued for an ad-

ditional seventeen hours after which there was approximately (nmr) 27%

of starting material, 19% of l-p-anisyl-2,3,3-triphenylcyclopropene, and

54% of other products. Removal of solvent at this point afforded a

yellow oil which was chromatographed over Merck alumina eluting with

benzene-hexane (5/95). Fractions of 50 ml were collected with the early

fractions containing mostly starting material along with some cyclopro-

pene products. An intermediate fraction contained mostly l-anisyl-2,3,3-

triphenylcyclopropene. The later fractions contained mixtures of pro-

ducts (probably indenes). Fractional crystallization (with a seed crystal)

from benzene-hexane yielded 0.0173 grams of colorless crystals mp 174-

1760. A qualitative uv (cyclohexane) exhibited maxima at 343, 326,

31(sh), and 226 nm, while the nmr indicated pure l-p-anisyl-2,3,3-

triphenylcyclooropene (45) (Fig. 11).

Photolysis of 3-p-Anisyl-l,2,3-triphenylcyclopropene (44)

A solution of 0.1776 grams (0.475 mmole) of 3-anisyl-l,2,3-tri-

phenylcyclopropene in 100 ml of cyclohexane was irradiated at 3500 AO

for twenty-four hours. Evaporation of solvent left a yellow oil (0.1420

grams) which contained two main products (nmr) and some impurities. Chrom-

atography over Merck alumina failed to separate these two products and

attempts to crystallize the oil failed (Fig. 12). One of the products

was tentatively identified as 3-p-anisyl-l,2-diphenylindene, (peak

enhancement in nmr by adding authentic 3-p-anisyl-l,2-diphenylindene (55)

while the other is most probably 6-methoxyl-1,2,3-triphenylindene (56)

(comparison of methoxy spike with that of 6-methoxyl-1,2,3-triphenyl-





-89-


indene in the nmr spectrum).

Photolysis of l-p-Anisyl-2,3,3-triphenylcyclopropene (45)

A degassed solution of l-p-anisyl-2,3,3-triphenylcyclopropene

(0.1863 grams, 0.50 mmole) in 100 ml of cyclohexane was irradiated at 3500

A for twenty-four hours. Evaporation of solvent left a yellow oil

which appeared to be (nmr) (Fig. 13) largely one indene and a small

amount of impurities. The oil was chromatographed over silica gel and

eluted with hexane. A pale yellow oil (0.1395 grams) was isolated from

one of the first fractions. The nmr spectrum (CDC13) consisted of

a three proton singlet at T 6.33, a one proton singlet at 4.97, and a

complex (approximately 20 proton) multiple from 2.55 to 3.40. By
5
analogy with earlier work the compound was believed to be either -

anisyl-1,2-diphenylindene (57) or 2-p-anisyl-l,3-diphenylindene (58).

Oxidation of the oil by chromic anhydride (0.130 grams) in 5 ml of

acetic acid afforded a yellow oil which was identified as o-anisoyl-o-

benzoylbenzene (59). Crystallization from 95% ethanol yielded 0.0023
63
grams of colorless crystals mp 131-1330 (Lit. mp 133-1350). The nmr

spectrum (CDC13) consisted of a three proton singlet at T 6.18 and a

thirteen proton multiple from 2.25 to 2.20. The mass spectrum (70 ev)

showed peaks at m/e (Rel intensity) 326(100), 289(10), 288(18), 240(22),

239(96), 211(16), 210(13), 209(55), 193(11), 152(24), 135(85), 105(32),

92(18), 77(49). The ir (KBr) showed absorption bands at 1655(vs), 1595(vs),

1315(s), 1265(vs), 1185(m), 1155(s), 1035(m), 940(s), 840(m), 775(m),

655(m), 600(m) cm-
655(m), 600(m) cm







-90-


3
'-I-- ;---~--~--E----------






-91-


o






_ |



03
0 ":
Hv


ac
o
0
C,]
U


44
C-,

0 M
, .- -
H
E-4
erl


" 0
!i! i
'.1 [


CM
-4-
Jj z
H
0





C-,
K ^ M




-5-- (e)-- --5





cc
~I __f








_____ ___-0 J4




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ELCI2E920_FTE991 INGEST_TIME 2012-03-08T17:35:11Z PACKAGE AA00003955_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES