ABSTRACT

SUBSTITUENT EFFECTS ON THE TYPE-II PHOTOREACTION 0F PHENYL KETONES

By

Allen Edward Kemppainen

Substituent effects on the reactivity towards y-hydrogen abstraction
in the type-II photoelimination of alkyl phenyl ketones were measured for
a wide range of substituents on the y-position. Inductive effects at the
y-carbon were also measured for 5- and e-substituents and for many of
these ketones the solvent effects on the quantum yields were investigated.
The data obtained by quenching the excited triplet states with an effi-
cient triplet quencher were treated using the Stern-Volmer relationship
to determine triplet lifetimes. In addition the quantum yields and trip-
let lifetimes for a wide variety of ring substituted alkyl phenyl ketones

were determined.

The following results were obtained: (1) The relative reactivities
of the type-II photochemical process compare favorably with those deter-
mined for hydrogen abstraction by tert-butoxy radicals. (2) A correla-
tion exists between the 01 of the substituent and the reactivity of the
y-hydrogen towards abstraction by the phenyl ketone triplet. A p of -2.0
was found for substituents on the 5-carbon. (3) If p for the y-carbon
is taken as -4.4 then contributions from inductive and radical stabiliz-
ing effects on the excited state reactivities can be quantitatively sep-
arated. (4) It was generally observed that the type-II photoproducts

accounted for virtually all of the reaction. A few exceptions were noted

in cases where the substituent was a good photoreducing or otherwise re-
active group. (5) No correlation exists between the excited state re-
activities and the type-II quantum yields (oII) for the alkyl phenyl ke—
tones tested. (6) Electron withdrawing ring substituents activate the
phenyl carbonyl triplet, and electron donating groups deactivate it.
However, in most cases the deactivation is much larger than can be ac-

counted for by the inductive effect alone.

The results are consistent with a mechanism which involves a 1,4-
biradical intermediate. This can accomodate the lack of correlation be-
tween the quantum yields and the excited state reactivities since the
type-II quantum yield can be expressed as the product of two probabili-
ties, the quantum yield of biradical formation (aBR) and the probability

of product formation from the biradical (op).

in = ¢BR¢p

The results with 6-substituents also establish that the reactivities
do correlate with OI substituent constants. The estimated p constant
for the y-carbon indicates that it is highly sensitive to inductive ef-
fects. The results obtained for ring substituted alkyl phenyl ketones
clearly show that the excited state reactivity is influenced by the in-
ductive effect and the nature of the triplet. The evidence can be inter-
preted as supporting a thermal equilibrium of reactive n,«* and unreac-
tive n,n* triplet states in which the reaction occurs solely from the

n,«* triplet.

SUBSTITUENT EFFECTS ON THE
TYPE-II PHOTOREACTION
0F PHENYL KETONES

by

Allen Edward Kemppainen

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Chemistry

1971

- , 1/,
F ‘ LJ

’7
1;.

J

DEDICATION

The writing of this thesis is dedicated to Dean Tom King, Dean of
Students at Michigan State University during my undergraduate years and
those of my brother and sisters. During those early years of decision
and indecision his influence on me was greater than that of any other
person. His friendship, encouragement, and advice were and always will

be deeply treasured.

ii

ACKNOWLEDGEMENTS

First and foremost I thank Professor Peter J. Wagner for his
guidance and support during the years of "research and requirements"
in quest of the Ph. D. His patience with the "old man" of the group
in many instances during my tenure is greatly appreciated. Our many
discussions, whether political, philosophical, or scientific in
nature have been very enjoyable. Gratitude for their teachings and
assistance which were always appreciated is extended to the faculty
members of the chemistry department at Michigan State. I also mention
my fellow graduate students at Michigan State, and especially those of

the Wagner Group, for many enjoyable associations over the years.

Acknowledgement is hereby made for the financial support in the

form of teaching assistantships from the department of chemistry, and

research grants from the National Science Foundation.

111

TABLE OF CONTENTS

SUBSTITUENT EFFECTS ON THE TYPE-II PHOTOREACTION OF PHENYL KETONES

page
INTRODUCTION ....................... l
1. Historical Notes ................ l

2. Discovery of the Type-II Photoelimination . . . . l
3. Later Developments Involving the Type-II Process 4
a. Identification of cyclobutanols ...... 4
b. Energy transfer and kinetic studies 5
c. Characterization of the excited state 8
d. Related work in photoreduction ....... ll

4. Recent Studies on the Effects of Solvents and
Substituents on the Type-II Photoreaction . . . . 14

a. MeChanistic implications .......... l4

5. Effects of Ring Substituents on the Reactivity
and Excited State ................ l7
6. Direction of the Research Effort ........ 21

a. General contributions to the area of

molecular photochemistry .......... 2l
b. Specific goals ............... 2l
7. Practical Significance ............. 23
a. Degradation of polymers .......... 24
b. Synthesis of four-membered ring compounds 24

c. Use as a model for determining substituent
effects .................. 25

iv

TABLE OF CONTENTS (Continued)

8. Definition of Terms ...............

a. Reactivity .................

b. Non-radiative decay ............

c. Quantum yield ...............

RESULTS .........................
l. General Explanation of Data ...........

a. Absolute type-II quantum yields ......

b. Disappearance and cyclobutanol quantum yields

c. Solvent effects on the quantum yields

d. Stern-Volmer quenching slopes .......
e. Miscellaneous data .............
2. Tabulated Results ................
a. y-Substituted alkyl phenyl ketones .....

b. Ring substituted alkyl phenyl ketones
c. Solvent studies ..............
d. Variation of ketone concentration .....

3. Justification of Data and Controls Relative to
Photochemical Data ...............

a. Identification of cyclobutanols ......
b. Detecting quenching impurities in the ketone

c. Does the type-II quantum yield vary signifi-

cantly with percent conversion? ......
d. The problem of product quenching ......
e. Solution viscosity .............

page
25
25
25
25

27
27
27
27
28
28
28
29
29
3O
35
38

4O
4O
41

41
43
46

TABLE OF CONTENTS (Continued)

DISCUSSION ........................

l.

6.

Quantum Yields of Side-Chain Substituted Ketones

a. ¢dis’ a test for material balance .....
b. A search for steric effects on ¢II .....
c. Steric effects on (ocyc) ..........
d. Solvent effects on ¢II ...........
e. ¢II of o-substituted ketones ...... ‘. .
f. Competitive triplet deactivation ......
Substituent Effects on the y-Position ......
a. Variations with C-H bond strength .....
b. Nonalkyl side chain substituents ......
c. Comments on relative selectivities .....
d. Quantitative relationships .........
Support for a Biradical Mechanism ........

Additional Solvent Effects ...........
a. Changes in qu ...............
b. An estimation of biradical lifetime

Effects of Ring Substituents on Triplet Reactivities

a. Change in the nature of the triplet

b. The effects of meta substituents ......
c. Qualification of "reactivity" .......
d. Powerful deactivating substituents .....
e. Ortho substituents .............
Summary .....................
a. Conclusions ................

vi

page
48

48
51
53
54
56
56
57
57
6l
64
65
69
71
71
73

76
78
8O
82
84
86
86

TABLE OF CONTENTS (Continued)

b. Significant observations ..........
7. Indications for Further Research ........
a. Photolysis of y-chlorobutyrophenone
b. Studies on the cyclobutanols for meta—
methoxyvalerophenone ............
c. Wavelength studies .............
d. The biphenyl ketone ............
e. The qu's from product quenching ......
EXPERIMENTAL. PART ONE. CHEMICALS ...........
l. Preparation and Purification of the Phenyl Ketones
Used in Photolyses ...............
a. Methods of preparation ...........
b. Methbds of purification ..........
c. Criteria of purity .............

2. Purification of Solvents and Other Compounds

a.

b.

Benzene ..................
Methanol ..................
tert-Butyl alcohol .............
Acetonitrile ................
2,5-Dimethyl-2,4-hexadiene .........
Piperylene .................
Internal standards .............
Pyridine ..................
Ethyl acetate ...............

vii

page
87
88

88

89
9O
91
91

93

93
93

100
110
111
111
112
112
113
113
113
114
114
114

TABLE OF CONTENTS (Continued)

EXPERIMENTAL. PART TWO. TECHNIQUES ...........
l. Preparation of Photolysis Samples ........

a. Photolysis solutions ............

b. Degassing procedure ............

c. A typical run ...............

2. Photolysis Procedure ..............

3. Procedure for Estimation of Ketone Disappearance
and Cyclobutanols ................

4. Analysis Procedure ...............
a. Instruments ................
b. Conditions .................
c. VPC trace .................
5. Area-Mole Response Ratios for Internal Standards
6. Controls in Experimental Procedures .......
a. Degassing procedures ............
b. Volumetric glassware ............
c. Photolysis tubes ..............
d. Cleaning of glassware ...........

e. Accuracy and reliability ..........

BIBLIOGRAPHY (Literature cited) .............

APPENDICES ........................

APPENDIX A. PART 1. EXPERIMENTAL QUENCHING RUNS FOR
DETERMINING STERN-VOLMER DIAGRAMS FOR KETONES:

page
115
115
115
115
116
117

118
119
119
120
120
120
124
124
124
126
126
126

129

135

135

TABLE OF CONTENTS (Continued)

page
APPENDIX A. PART 2. EXPERIMENTAL QUENCHING RUNS FOR
DETERMINING STERN-VOLMER DIAGRAMS FOR KETONES:
94°
CHZCHZCHZCH3 .............. 152
APPENDIX A. PART 3. EXPERIMENTAL DATA FOR DETERMINING
DISAPPEARANCE AND CYCLOBUTANOL QUANTUM YIELDS . . 163

APPENDIX A. PART 4. SOLVENT STUDIES ON THE PHOTO-
CHEMICAL BEHAVIOR OF SUBSTITUTED PHENYL KETONES 165

APPENDIX A. PART 5. MISCELLANEOUS DATA ON QUANTUM
YIELDS ..................... 170

APPENDIX B. COMPUTER PROGRAM FOR DETERMINING LEAST
SQUARES SLOPES OF STERN-VOLMER DIAGRAMS . . . . 172

APPENDIX C. IDENTIFYING SPECTRAL CHARACTERISTICS OF
PREPARED PHENYL KETONES ............ 174

ix

TABLE

TABLE

TABLE
TABLE

TABLE
TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

I.

II.

III.
IV.

VI.

VII.

VIII.

IX.

XI.

XII.

XIII.

XIV.

LIST OF TABLES

Triplet State Reactivities and Quantum
Yields of Phenyl Ketones: 40
~ @4
CH2'R

Triplet State Reactivities and Quantum
Yields of Phenyl Ketones:@ ,/O
R

C\
CHZCHZCHZCH3

Data for Solvent Studies on Selected Ketones

page

31

33

36

Viscosity Measurements for Benzene, Valerophenone,

and Pentadecanophenone ............

Maximum ¢II in Alcohol Solvents .......

Comparison of Relative Reactivities of Carbonyl

Triplet Species and tert-Butoxy Radicals . . .

Substituent Effects on Hydrogen Abstraction by
tert-Butoxy Radicals .............

Effects on the Type-II Intramolecular Hydrogen
of Side-Chain Substituents on Alkyl Phenyl
Ketones ......... . ..........

Comparison of Experimental Reactivities to
Calculated Reactivities from GI and
Experimental p Value .............

Effects of Solvents on qu ..........

Comparison of Initial Slopes [khs] with
Type-II Quantum Yields ............

Summary of 1/1 [x l07] sec‘1 from Table II
for Ring Substituents on Valerophenone . . . .

Estimation of kr and kd for Several Ring
Substituted Alkyl Phenyl Ketones .......

Type-II Quantum Yields for Highly Deactivated
Alkyl Phenyl Ketones .............

47
55

58

62

63

68

72

75

79

82

83

LIST OF TABLES (Continued)

page
TABLE XV. Comparison of qu Values from Product

Quench1ng and Diene Quench1ng ....... 92
TABLE XVI. Physical Data for Synthesized Ketones . . . 101
TABLE XVII. Purification of Purchased Ketones ..... 107
TABLE XVIII.Analytical Conditions ........... 121
TABLE XIX. Special Analytical Conditions ....... 122
TABLE XX. Standard-Product Molar Response Ratios . . . 125
TABLE XXI. Effect of Silicone Stopcock Grease on

Photolysis of Valerophenone ........ 127

xi

Figure

Figure

Figure

Figure

Figure
Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

10.

11.

12.

13.

14.

15.

LIST OF FIGURES

page
Modified Jablonski Diagram .......... 10
Effect of Ring Substituents on Triplet
Energy Levels ................ 19

Relationship of Type-II Quantum Yield to
Ketone Concentration for Several Ketones . . . 39

Product Formation Per Unit Time versus

Irradiation Time ............... 42
1/qu versus Average Quencher Concentration 44
1/4,II versus Average Quencher Concentration 45

Stern-Volmer Quenching Slopes of Alkyl
Phenyl Ketones ................ 60

Hammett Plot for 6- and e-Substituted
Alkyl Phenyl Ketones ............. 66

Schematic Representation of the Mechanism of
the Photochemical Process in the Type-II
Photoreaction ................ 7O

Hydrogen Bonding to the Biradical
Intermediate in Alcohol Solvents ....... 73

Stern-Volmer Type Treatment of Increasing
Quantum Yields upon Addition of Alcohol Cosolvent 74

Log of the 1/1 x 10'7 sec'1 of Ring Substituted
Phenyl Ketones versus am or op ........ 77

Stern-Volmer Quenching Slopes for ortho-
Methoxyvalerophenone ............. 85

Stern-Volmer Quenching Slopes for meta-
Methoxyvalerophenone ............. 89

VPC Chart for Analysis of Hexanophenone . . . 123

xii

INTRODUCTION

1. Historical Notes.

 

For at least the last two centuries men of a scientific bent have
been very curious about the nature of the interaction of light with
matter. To the naturalist the dependence of plant life on sunlight is
axiomaticl. To the physical scientist, however, the greatest challenge
lies in the testing and interpretation of photochemical phenomena.
Major credit for the early stages in the formulation of this area
should be given to Grothaus and Draper2 who in the early 1800's devel-
oped what is now referred to as the first law of photochemistry: Only
the light which is absorbed by a molecule can produce photochemical
change, Draper used the term "tithonic rays"3 to describe a component
of light which produced photochemical change in certain compounds, yet,
if "ultraviolet rays" were to be substituted in its place, one would

find his observations and conclusions to be quite accurate.

2. Discovery of the Type-II Photoelimination.

During the early part of the twentieth century extensive studies
were done on the photodecarbonylation of carbonyl compounds”’5’5. Much
of this work consisted of physical measurements of the kinetics of gas
phase reactions and the results are often cited in basic physical
chemistry textbooks7’8. While conducting such studies on the photolysis
of various ketones in the gas phase, Norrish and co-workers discovered
that when the ketone or aldehyde contained y-hydrogens a new type of
reaction took place in addition to the decarbonylation9’10. Analyses
of the product mixtures showed that a cleavage between the carbon-
carbon bond which was a-B to the carbonyl took place yielding a lower

molecular weight carbonyl compound and an olefin. Norrish classified

them as type-I and type-II reactions (See Equation 1). It was also
noted that the latter pathway is preferred. Subsequently it was found

that the type-II photoreaction readily occurs in hydrocarbon solventsll.

tMPe'I >. co + R-CHZCHZCH-R' + RH +

 

 

Rll
CH2=CH-CHR'R"
0 in
R-E-CHZCHZ H-R‘ "” (Eq. 1)
(1) R = R' ‘ CH3’ t e-II " S
R" = H «9» > R-c-CH3 + H=CR'R"
(2) R = CHZCHZCHB, R' = R" = H
(3) R = H, R' = R" = CH

3

A number of ketones and aldehydes were irradiated in isooctane and in
medicinal paraffin (at 70 to 100°C) and furnished interesting results.
The type-II process was found to be relatively unaffected by the sol-
vent or the temperature changes whereas the type-I process, which was
entirely suppressed at room temperature, gave high yields at elevated
temperatures. Also, the production of small saturated hydrocarbons in
the type-I reaction was accompanied by a corresponding unsaturation in
the solvent. These facts led to the conclusion that the type-I reaction
with ketones involved production of free radicals and the type-II reac-
tion involved some type of concerted mechanismlz. The case in which
the aldehydes photolyzed to give carbon monoxide and the corresponding
saturated hydrocarbon with little or no unsaturation in the solvent was

considered to be a type-I reaction without production of radicals.

The common differentiation made today is that of the primary process of
the excited carbonyl: In the type-I reaction the bond between the car-
bonyl carbon and the o-carbon is cleaved forming two radicals; in the
type-II reaction a y-hydrogen is abstracted by the excited carbonyl,
resulting in the cleavage of the a-B bond. Approximately ten years
later Noyes and coworkers13 restudied the photolysis of methyl butyl
ketone in the gas phase and essentially confirmed Norrish's results. A
product ratio of 1:1, acetone to propylene, with a quantum yield (¢) of
about 0.5 was found. Since the quantum yield was relatively unchanged
from 25° to 300°C, and considering Norrish's results in hydrocarbon
solvents in which the quantum yield was nearly the same, Noyes suggest-
ed that possibly a cyclic 6-membered ring-like form of the ketone exis-
ted in solution involving hydrogen bonding with the y-hydrogens, and
that this was responsible for the type-II reaction (See Equation 2). A
similar argument had been given earlier by Rice and Tellerl“ in their
treatment of the theory of least motion in elementary free radical re-
actions. Attempts were made to strengthen this argument using dueter-

ated ketone515 , the best test being that made by Srinivasan16 who used

 

/H
9" \?H"CH3 H ?H
' hv . CH -C\ 3
c”3'C\ /CH2 7 3 \CH CH =CH “‘3‘" 2)
CH2 2 2

2-hexanone-5,5-dz which produced acetone-d1 as a photolysis product.
Some plain acetone which was also formed was attributed to exchange of
the enol-intermediate with hydroxy groups bound to the walls of the

photolysis cell. 2-Hexanone produced a similar amount of acetone-d1

when photolyzed in a cell preconditioned with 020 vapor. This enol-
intermediate, which was first invoked to explain the isotope exchange
upon photolysislS, was later confirmed by actual observation of its
infra-red absorption during the gas phase photolysis of 2-pentanone17.
It was found to be a transient species with a half-life of about 3.3
minutes and the rate of appearance of acetone absorption corresponded

to the rate of decay of the enol-species.

3. Later Developments Involving the Type-II Process.

a. Identification of cyclobutanols. By 1960 the type-II photo-
elimination had begun to lose its status as a side reaction and was
being studied in its own right. Yang and Yang18 had recently identi-
fied another photoproduct in the photolysis of 2-pentanone, 2-octanone
and 2-nonanone as the cyclobutanol. This was an important clue to the
mechanism of the reaction. It was later found that a-methoxy and a-
ethoxyacetophenone as well as 1-methoxy-3,3-dimethyl-2-butanone readily
formed the oxetanoll9. LaCount and Griffin20 later photolyzed valero-
phenone, y-phenylbuterphenone and a-benzyloxyacetophenone, separated
the gj§_and tran§_isomers of the cyclobutanols by chromatography and
identified each by nmr spectroscopy. They found that the trag§_isomer
was preferred in each case. To account for these results a mechanism

proceeding through a 1,4-biradical was proposedle’21 (Equations 10 & ll).

 

3* i”
R- -CH2CH2CH2-R > R- -CH2CH2CH-R (Eq. 10)

H o
. ll -
-CH CH CH-R A> 3 + CH2‘CHR

R 2 2
H (Eq. 11)

 

o
1
-C

b. Energy transfer and kinetic studies. Classical energy trans-
fer studies had established the fact that direct transfer of excitation
occurred in matrix or solution. Terenin and Ermolaev observed that
triplet energy from carbonyl compounds was transferred to species which
had lower triplet energy levelszz. The same phenomenon was reported by
Backstrom and Sandros23 for the irradiation of biacetyl. They discov-
ered furtherZ“ that besides being able to quench the phosphorescence of
biacetyl with triplet quenchers, they could sensitize its phosphores-
cence with benzophenone which has a higher triplet energy. Ben20phen-
one is known to convert efficiently to the triplet state since only its
phosphorescence emission can be observed. By varying the biacetyl con-
centration the mean lifetime of the benzophenone triplet was found to

be 1.9 x 10'5

second. In studies on the sensitized gj§;trgfl§_isomer-
ization of piperylene, Hammond and coworkers25 found that the diene was
an efficient quencher of the triplet state of benzophenone and aceto-
phenone. It was also used in later more detailed studies on gj§;trgg§_
isomerization525. Wagner and Hammond21 used piperylene to quench the
reaction of 2-hexanone and 2-pentanone, and Dougherty27 used it to

quench 2-octanone. Their results were similar: A considerable portion

of the reaction was rapidly quenched by addition of small concentrations

of piperylene, after which the quantum yield leveled off to a constant
value and seemed unaffected even by very high quencher concentrations.
This was good evidence for reaction from two excited states, a quench-
able triplet and a non-quenchable singlet. Wagner and Hammond also
compared the relative ease of quenching of 2-pentanone and 2-hexanone.
They found the latter, with the secondary y-hydrogens, much more diffi-
cult to quench. This was in agreement with previous work by Ausloos28
and also could be compared to studies by Walling and co-worker529-32 who
found that relative rates of hydrogen abstraction by alkoxy radicals
were 4 to 8 times faster from secondary hydrogens than from primary
hydrogens. It was also shown that the reactivity of benzophenone trip-
lets towards the C-H bond strength in hydrogen abstractions followed
quite closely the reactivity of the alkoxy radicals33i3“. If any dif-
ferences were to be noted, the triplet abstraction showed somewhat
greater selectivity for the more reactive hydrogens. Wagner and
Hammond35 later performed quenching experiments on butyrophenone and
valerophenone. This was the first direct comparison of the two and it
was predicted and found that the secondary y-hydrogens of valerophenone
were more reactive (less sensitive to quenching) than the primary hy-
drogens of butyrophenone. The data was treated by plotting the quantum
yield without quencher (o0) over the quantum yield with quencher (6)
versus the quencher concentration [Q], as is described by the Stern-

Volmer23’35 equation (Equation 12).

99 = 1 + kq[o]T (Eq. 12)

The classical method of deriving this equation is by using a steady-

state treatment developed by Stern and Volmer over 50 years ago. How-
ever, by using a modernistic stochastic formulation, one can arrive at
the same relationship36. The Stern-Volmer relationship can also be de-
rived by using a simplified treatment based on the definition of the
quantum yields of the various processes that take place and the derived
mechanism of the reaction37. The sum of the quantum yields of all of

the processes taking place is defined as unity (Equation 13).

¢total = ¢1 + ¢2 + ¢3 T P4 + ' ' + ¢ = 1 (Eq. 13)

The quantum yield of each individual process of the triplet state can
then be expressed as the ratio of the rate of that process to the sum
of the rates of all of the processes. For sake of brevity, if the uni-
molecular rate constant for biradical formation from the triplet is
defined as kr’ the unimolecular rate constant for all other deactivat-
ing processes as kd, and the bimolecular quenching rate constant as kq,
then the quantum yield with no quencher present can be expressed as in

Equation 14.
(pa = -—r———- = k‘l’ (Eq. 14)

The quantum yield with quencher added then would become:

k

__ 1‘
‘P ' kr + kd +j(q[0] (Eq. 15)

 

Then dividing Equation 14 by Equation 15 one arrives at the Stern-

Volmer relationship (Equation 16).

 

 

$9 = kr/(kr + kd) = kr + kd + kq[9]
¢
kr/(kr + kd + kq[Q] kr + kd
(Eq. l6)
k + k k [Q] k [Q]
= r d + __.g.__ : 1 + __g_—
kr + kd k + kd k + kd

In the special case where kd is very small, l/kr can be equated with

the average triplet lifetime (1). The quenching rate constant, kq, has

been calculated using a simplified Debye formula (Equation 17) under

the assumption that the rate of quenching is diffusion controlled and

thus inversely proportional to the viscosity of the solventZ“. Although

Wagner and Hammond35 estimated that the rate of quenching was only about
8RT

kdiff = -§665;———-liters/(mole x sec) (Eq. 17)

half that predicted by the Debye equation, the notion that quenching was
diffusion controlled was still popularly assumed38. The careful work of
Wagner and Kochevar39 has now shown that only in relatively viscous sol-
vents is the quenching rate diffusion controlled, and as the solvent
becomes less viscous the deviation from the Debye relationship becomes
greater. It was also found that a modified Debye equation (Equation 18)

was more representative of the experimentally determined values39’“°.

kdiff = -§%%5;--liters/(mole x sec) (Eq. 18)

c. Characterization of the excited state. The photochemical

behavior of carbonyl compounds is rather obviously related to their
excited state(s). In his early work Norrish9 alluded to an "upper
level" of the reacting ketone, however, little was known of its nature
at the time. Phosphorescence and fluorescence emissions of carbonyl
compounds have been studied for many years“1 and provide critical in-
formation in this area. The actual nature of phosphorescence, which
was recognized more than a century and a half ago“2, wasn't untangled
until the 1940's when the extensive studies of Lewis and co-workers‘+3
made clear the distinction between fluorescence and phosphorescence:
The former resulting from a transition between singlet excited states
to singlet ground state and the latter resulting from a spin-forbidden
triplet to singlet transition. They also observed the triplet-triplet
absorptions”” in fluorescein which was irradiated sufficiently to pro-
mote approximately 80% of the molecules into the triplet state, predic-
ted and observed the extremely weak singlet-triplet absorption bands“5
of a number of compounds which phosphoresced, and suggested that the
molecule in the triplet state should have a measurable paramagnetic
susceptibility”3. The concept of a "meta-stable" state, first proposed
by Jablonski“6 to account for the longer lifetime phosphorescence,
proved to be essentially correct and led to the identification of the
triplet state (See Figure 1). Yet the characterization of the triplet
state in the photoreactions of carbonyl compounds was not accomplished
without difficulties. The hydrogen abstracting species in the photo-
reduction of benzophenone was first considered to be a "diradical" by
Backstrbm“7 but as knowledge in this area advanced he redesignated it
as the triplet“3. A controversy developed over whether the type-II

cleavage of the dialkyl ketones occurred from the excited singlet or

10

or triplet state. It was first proposed that 2-hexanone“9, 2-pentanoneso
and a number of alkyl aldehydes and ketones51 reacted from the excited
singlet state. Other evidence strongly suggested that the triplet
state was the one involved in going to type-II productSZ-S“. Wagner
and Hammond21 later showed that 2-hexanone and 2-pentanone reacted from
both the excited singlet and triplet states and determined that early
difficulties were due in part to the differences in reactivity of the
two ketones. They found that the relative rate of reaction was much
greater for the singlet than for the triplet and the reaction of the
secondary hydrogens was faster than the primary hydrogens. Another
aspect of the excited state arises when the chromophore contains an

atom with non-bonding electrons. It is then possible to have either an

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T2
51 T _
7::---mfi —: 07011133232222.)
_—-_- - (Phosphorescence)
So
— >J
4

 

 

Figure 1. Modified Jablonski Diagram. Diagram is designed to show a
case in which the upper singlet ($1) and the ground state singlet (SO)
have the same internuclear relationships and the 0- 0 bands represent
the most probable transitions. ISC represents intersystem crossing to
the triplet level (T1), which corresponds to Jablonski' s "metastable
state. " Information taken from Reference 2, pages 274 and 285.

11

n,n* or a n,n* lowest energy triplet state. This consideration becomes
important in explaining the large effects on relative reactivities of a
ring substituent on phenyl ketones, yet it is not completely understood.
An up to date account has recently been published in a review by Wagner
and Hammond“°. Further discussion on this topic is presented in the

section on the effects of ring substituents.

d. Related work in photoreduction. Paralleling the developments
in photoeliminations another area of photochemistry of carbonyl com—
pounds, that of photoreduction, developed in a quite independent fash-
ion. Although it had been known that a reaction took place when an
alcoholic solution of benzophenone was exposed to sunlight, the pro-
ducts were not known until Ciamician and Silber55 correctly identified
them as the benzpinacol and the corresponding aldehyde or ketone (See

Equation 19). Acetophenone was also photolyzed and found to pinacolize,

R' HO OH

0
2@C'-R] + CH3CH-0H “" @C- C©+ g-R'
R = CH3 or Q

R' = H or Alkyl

although more slowly than benzophenone. Several alcohols were also
tried and it was found that while photoreductions occurred readily in
primary and secondary alcohols, only very little reaction took place in
tert-butyl alcoholSS. During the next several years benzophenone was
photoreduced in a number of media including hydrocarbon557, aliphatic

acids and etherssa, and ester559. Complex byproducts often resulted

12

such as resins or addition products. The photoreduction of acetone in
a number of alcohols was also studied50. Here the principal product
appeared to be the 1,2-glycol corresponding to a 1:1 adduct of the
acetone and the alcohol, although some isopropyl alcohol and other pro-
ducts, depending on the alcohol used, were noted. The study of the
photoreduction of benzophenone was continued by Cohen61 who used a num-
ber of various alcohols as reactants*. His findings corroborate those
of earlier works in most cases. He seemed perplexed, however, that
cinnamyl alcohol would not photoreduce benzophenone although allyl and
benzyl alcohol readily did. Present knowledge of energy transfer would
predict that the excited benzophenone would transfer its energy to the
styryl group. The investigations62 continued in this direction until
the late 1930's when studies were begun into the mechanism of the
photoreduction. Concluding that the hydroxylic hydrogen of the alcohol
was not involved in the reaction, Weizmann, Bergmann and Hirshberg63
claimed that the first step during the irradiation was the activation
of the carbonyl to a "diradical form." Splitting of a C-H bond on the
alcohol followed, resulting in two radicals which later dimerize (Equa-

tion 20). This conclusion was based partly on the results obtained

0' $H3
©C-CH3 + R'R"-CH-0H ____, ©C-OH + R'R"-C-0H (Eq. 20)

with optically active l-phenylethanol. It was found that acetophenone
pinacols formed with the optically active alcohols were inactive, and

also that the unreacted alcohol retained its original activity.

 

* Among his observations is one that "water is a strong negative cata-
lyst" in the photoreduction of benzophenone.

13

Bergmann and HirshbergGR later reported substituent effects on photo-
reduction. They noted that ring substituents on benzophenone or aceto-
phenone nearly always impair the reactivity. A naphthyl or biphenyl
substituted for a phenyl group in benzophenone stops pinacolization.
The same result is seen when a para-methoxy group is placed on the ring
of acetophenone, or when a naphthyl is substituted for the phenyl ring.
The benzophenone photoreduction was further studied by Pitts and co-
worker565 who found that oxygen dissolved in the photolysis solution
inhibits (quenches) the photoreduction. Quantum yields for samples
saturated with oxygen, air, and samples degassed on a vacuum line
varied from 0.00 to 0.50 to 0.95 respectively. Relatively large con-
centrations of olefins were also found to inhibit beaninacol formation.
As a result of his and Hammond's66 work, Pitts concluded that the sole
reacting species of benzophenone was the triplet. Further information
on the mechanism was obtained by irradiating benzophenone in optically
active sec-butyl alcohol. As was found previously with acetophenone63
the unreacted alcohol retained its original activity. This rules out

the disproportionation of the radical species formed (Equation 21) as

0

ll
3CH2CCH3

3 (Eq. 21)

CH3CH2C-OH + CH3CH2C-OH -—%——+- CH3CH2CHOH + CH
H

well as the reverse of the abstraction step (Equation 23). Also, since
the isolated wavelength of 3660A readily produces photoreduction, the
initial step has to be excitation of the carbonyl as the alcohol is

transparent in this region. Pitts proposed the following mechanism:

14

 

.2C=0 h" A. ©12C=0* (Eq. 22)

©12C=o* + CH3CHzgfl-OH -———-> ©2C-OH + CH3CHzgl-I-10H (Eq. 23)
3 3

.2C=0 + CHBCHZE-OH ——-—-> ©J2C-OH + CH3CHZEEO (Eq. 24)

H3 3

2{@]26-0H1 + @3241" '©]2 (Eq. 25)

 

The mechanism of benzophenone photoreduction is still under investiga-
tion and some modifications have been proposed to account for a small
amount of the mixed pinacol found57. Although the notion of hydrogen
abstraction by the excited carbonyl was considered quite early in the
studies of photoreduction63, it wasn't until twenty years later that

the argument was applied to the type-II photoreaction.

4. Recent Studies on the Effects of Solvents and Substituents on the

Type-II Photoreaction.

 

Since the mid-1960's the type-II photoreaction has generated a
considerable amount of interest. Recently aquired data have been inval-

uable in resolving several of the early problems.

a. Mechanistic implications. Wagner and Hammond35 proposed the

 

following mechanistic scheme for the steps involved in the type-II pro-

CESS:

1K —“-"—-> 1511* ..—_—» ‘K* (Eq. 26)

o l

15

19* —1-+ PRODUCTS (Eq. 27)
“(1* Lise» 3Kn* ———-> 319* (Eq. 28)
3K1* -'£3—+ PRODUCTS (Eq. 29)
319* —kd—~v 11<o (Eq. 30)
3K]* + Q —5q——+ 1KO + Q* (Eq. 31)

1

In the above scheme the ground state ketone is represented by Ko’

excited singlet at all possible levels by 1Km*, and the lowest energy

1

excited singlet by K]*. The triplet ketone at all possible levels of

3Kn* and the lowest triplet level by 3K]*.

excitation is represented by
Non-radiative decay from the excited singlet, which is not included in
the scheme, was considered to be negligible. Their results further
showed that the quantum yields of valerophenone, butyrophenone, 2-hexa-
none, and 2-pentanone in no way reflected the differences in reactivi-
ties between the ketones with secondary and primary y-hydrogens. The
quantum yields, which ranged from 0.4 to 0.5, also indicated that con-
siderable inefficiency was involved in going to product. As one possi-
bility to account for this a reverse hydrogen transfer was suggested
which would provide a radiationless mechanism for return to the ground
state of the ketone. Support for this theory comes from Wagner's“:69
investigations on solvent effects on type-II quantum yields for a num-
ber of ketones. Polar, hydrogen bonding solvents raise the ¢dis to
near unity, and since triplet lifetimes don't change much68 this indi-
cates that the inefficiency in non-polar solvents must be due to the
biradical. The results are consistent with a 1,4-biradical intermed-

iate in which the abstracted hydrogen is prevented from returning to

16

its original location by solvation or hydrogen bonding to the solvent.
The original mechanistic scheme can be modified to account for this by

replacing Equation 29 with the following:

3K1 lb» B.R. (Eq. 32)

B.R. —"p—+ PRODUCTS (Eq. 33)
k- 1

B.R. ._r_. K (Eq. 34)

Here B.R. represents the biradical intermediate. Kelso70 recently
found that optically active y-methylhexanophenone racemizes several
times faster than it forms type-II products. This verifies that a re-
vertible radical site is formed at the y-carbon. Results of prelimin-
ary investigations further emphasized the discrepancies between triplet
state reactivity and quantum yield. It is proposed by Wagner69 that
the type-II quantum yield can be expressed as the product of the fol-
lowing probabilities:

k k
4’11 ‘ ¢isc¢BR¢P = ¢isc(E‘_ITd)(F_‘ET ) “‘1' 35)
r p '1”

Intersystem crossing quantum yields which have been measured are uni-
formly near unity for unsubstituted phenyl alkyl ketones71, pyridyl
alkyl ketones73, and ortho-, meta-, and para-methoxyvalerophenones7‘1 so
¢isc is commonly taken as one unless there are indications to the con-
trary. Wagners proposal69 on the inhibition of revertible hydrogen
transfer in polar solvents provides a key to determining the specific

rate constants in Equation 35. Since k-r becomes negligible in polar

17

solvents op becomes 1, and therefore:

1‘
411(a1c0h01) 2 WW d) - 4BR (Eq. 36)

All that is needed to determine kr is the triplet lifetime, 1 =

l/(kr + kd), which when substituted into Equation 36 gives:

kr = (I/T)¢Il(alcohol) (Eq. 37)

Note that in the special case where kd is very small compared to kr and
¢II(alcohol) is very close to 1, kr can be equated to 1/1. Using the

above relationship Wagner and Schott75 have calculated kr, kd, and ¢p
for a number of substituted phenyl alkyl ketones in which ¢BR is signi-

ficantly different from 1.

5. Effects of Ring Substituents on Reactivity and Excited State.

As mentioned before substituent effects were first noticed in the
studies on benzophenone photoreduction by Bergmann and HirshbergG“.
Later, studies by Porter and co-workers76 were also done on the photo-
reductions of substituted benzophenones. In these cases the quantum
yields were considered to be a measure of reactivity towards inter-
molecular hydrogen abstraction*. Substituents such as fluorine or bro-
mine had little effect on the quantum yield of photoreduction. A

phenyl group on one of the rings reduced the quantum yield about one

 

* Note that there may be some justification for this in the case of
intermolecular hydrogen abstraction since the probability of reverse
hydrogen transfer would appear to be much smaller than in the case of
intramolecular hydrogen abstraction.

18

order of magnitude or wiped it out completely. Hammond and co-workers66
also found that the quantum yields of photoreduction of di-para-methoxy-
and di-para-cyanobenzophenone were one fifth and one third respectively
of that of benzophenone. Porter explained his results by proposing
three possible types of triplet states in the following order of reac-

tivity: n,«* > n,n* > C-T (Charge-Transfer).

3- 3+
@f‘l "i=0 ”WOT-0‘
R R
n,n* 141:_ C-T

 

Ben20phenone and the halo-substituted benzophenones then reacted from a
lowest n,n* state which was characterized by an electron deficient oxy-
gen. A phenyl substituent lowered the n,n* triplet below the n,«*, so
Porter felt that the reaction in these cases must occur from the n,n*
triplet. An inconsistency in his interpretation arises where he consid-
ers the lowest triplet state in para-methoxybenzophenone to be n,n* in
nature, yet the quantum yield of photoreduction is about the same as for
benzophenone. The unreactive nature of the amino- and hydroxyl- phenyl
ketones was attributed to a charge-transfer state, which was stabilized
in the polar solvent, isopropanol. When the amino- and hydroxy- benzo-
phenones were irradiated in cyclohexane a small amount of photoreduction

took place lending support to the argument.

The first study of the effect of phenyl substituents on the type-II
photoreaction was that of Pitts and co-workers77 in 1966. They photo-
lyzed a number of ortho- and para- substituted butyrophenones at 3130A

in alcohol and hydrocarbon solvents. The idea of a charge-transfer

19

state was discounted in favor of an explanation which considered two
effects produced by substitution at the para- position: (1) The substit-
uent could inductively influence the reactivity of the excited carbonyl,
and (2) it could alter the electronic structure of the molecule such as
to change the nature of the reacting species. High quantum yields were
found for para— substituted chloro-, acetoxy-, methyl-, and fluorobutyr-
ophenones which were considered to have an n,n* lowest triplet state.

A para-methoxy group lowered the quantum yield somewhat and para-aminO-,
para-hydroxy-, and para-acetamido- groups completely eliminated the
type—II photoproducts. However, all four substituents presumably caused
the n,n* triplet to fall below the n,n* triplet in energy, making the
excited ketone unreactive. The difficulties in this situation arise
from attempting to correlate quantum yield and reactivity. Pitts also
quenched the type-II reaction of butyrophenone using piperylene and ob-
tained a linear Stern-Volmer plot. However, this was cited only as evi-
dence that the reaction takes place entirely from the triplet state and

no attempt was made to ascertain the triplet lifetime.

A simple explanation of the reason for rearrangement of triplet
energy levels is that the perturbation of the n-electron cloud of the
molecule by the substituents raises the energy of the ground state n-or-

bitals above the level of the non-bonding orbitals (Figure 2). The

 

 

 

 

 

 

1T
0 o
E @-il.-CH2CH2CH3 X©C-CH2CH2CH3

Figure‘2. Effect of Ring SUbstTtuents onTTriplet Energy LeveTs.

 

 

20

nature of this interchange, which is too complex for discussion here,
may also involve the transfer of energy from one level to another by
vibronic coupling78 and/or as recently has been prOposed by Wagner75,
thermal equilibrium between the two types of lowest triplet states. It
should be pointed out here that a change in environment of the ketone
molecule may also cause a switch in the triplet energy levels. Lamola79
has shown that in a non-polar hydrocarbon glass acetophenone has a low-
est n,n* triplet state. When polar hydrogen bonding solvents are used
the lowest energy triplet is n,n* in nature. Spectroscopic studies by
Kearns and Case80 revealed both n,«* and n,n* triplets for several ring
substituted acetophenones and indicated that they are quite close in
energy. In studies on the photoreduction of substituted acetophenones
Yang and co-workers73.81 have correlated the reactivities of substituted
acetophenones with their spectroscopic properties. Trifluoromethylace-
tophenone and acetophenone which are most reactive have n,n* lowest
triplets while para-methyl- and 3,4-dimethylacetophenone have n,n* low-
est triplets. Yang attributes the reactivity of the ketones with n,n*

lowest triplets to vibronic coupling between the two states79.

Recently Schott7‘+ has found that the initial enhancement of ¢II for
para-methoxyvalerophenone in benzene upon addition of tert-butyl alcohol
is reversed by continued addition and in high concentrations of alcohol
falls below what it was in pure benzene. This is interpreted as the re-
sult of an increasing separation between the higher n,n* and lower n,n*
triplets as the solvent medium becomes more polar. This data is consis-
tent with a small amount of n,n* triplet as the reacting species in

equilibrium with n,n* triplet. Schott7“ has measured the reactivity of

21

several y-substituted para-methoxyvalerophenones and found that the
relative reactivities are the same as those for the valerophenones with
the same y-substituents. Since a n,n* triplet would not be expected to
Show the same substituent effects as an electron deficient n,n* triplet,
the theory that the reaction occurs from a small concentration of n,«*

triplet is strongly supported.

6. Direction of the Research Effort.

a. General contributions to the area of molecular photochemistry.
That the interest in molecular photochemistry has sky-rocketed in the
last 10 to 12 years can be easily verified by thumbing through the in-
dexes of the basic journals. However, much of what is known is qualita-
tive in nature and very little is known of the processes occurring after
excitation of the reacting compound. Thus the necessary task of col-
lecting data, considered by some to be mundane, must be performed to
provide a basis for rules and correlations. Hopefully this project will
contribute to this basic area of physical photochemistry. Also, an at-
tempt will be made to demonstrate the utility of the type-II photoelim-
ination. In all save one35 of the investigations which pre-date this
work the product yields or quantum yields were the basis of correlations
and comparisons. The type-II process has the distinct advantage of
measuring the specific rate constant of y-hydrogen abstraction and of
detecting subtle changes in inductive effects on hydrogen abstraction

because of its high selectivity.

b. Specific goals. The purpose of this research project was to

further investigate the photochemistry of the type-II process in phenyl

22

alkyl ketones and extend the work of Wagner and Hammond35 in determining
structure-activity relationships. The phenyl alkyl system was chosen
because of its versatility in providing numerous possible variations in

1 1
©C-CH2CH2CHz-Y x

the basic model:

ortho, meta or para
substituents.

.<
ll

alkyl, phenyl or
heteroatom substit-
uents.

Substituents on the alkyl chain would be chiefly on the y-position,
however, other positions such as B, 6, or e could also be substituted.
Also, since the type—II photoreaction for the phenyl alkyl ketones oc-
curs entirely from the triplet state35a55 there would be no complication
from singlet products which possibly proceed by way of a different mech-

anismez. Following are some of the specific areas to be investigated:

(1) Even from the small number of examples used Wagner and
Hammond concluded that the quantum yield was a poor indication of excit-
ed state reactivity. Sufficient data could be obtained to prove or dis-

prove this generalization.

(2) The y-alkyl position will be substituted with a variety of
groups to determine the effects on the rate of hydrogen abstraction.
Expected effects would be those due to C-H bond strength, steric inter-
actions, stabilization of the radical generated, and inductive effects
on the y-carbon. Since these effects are known for free radical hydro-
gen abstractions, the results would be a test of a mechanism involving a

1,4-biradical intermediate.

23

(3) The phenyl ring will be substituted with as large a variety of
substituents as is conveniently possible without changing the y-hydrogens
on the alkyl chain. This would measure the changes in reactivity of the
triplet carbonyl to abstraction of the y-hydrogens. Electron donating
and withdrawing substituents should produce opposite effects if the na-
ture of the reacting Species remains unchanged. Effects of meta-substi-
tuents are as of yet uncharacterized as are those of ortho-substituents
with only a few exceptions77383. Some substituents which could alter
the nature of the lowest triplet state by stabilizing the n,n* level

will also be studied.

(4) Solvent effects on the quantum yields of selected ketones will
be studied for information on the efficiency of the biradical intermed-
iate in going to products. This is an important test in that it indi-
cates whether 1/T is a good representation of kr' It was previously in-
dicated that if 1/1 is not a good measure of kr’ then kr can be found by
the relationship in Equation 37. There is also the possibility that
changing the solvent will affect the nature of the lowest triplet state.
Therefore, any changes noted in the reactivity when changing solvents

will have to be carefully analyzed.

7. Practical Significance.

The classical, seemingly unavoidable, question of "does your re-
search have any practical significance?“ is usually answered in one of
two ways: (1) A rather defiant "no" followed by a quick change of sub-
ject, or (2) a half-apologetic account of the importance of scientific

curiosity. It may surprise some that the type-II photoelimination may

24

have application of a practical nature or utility other than that of
being a scientific curiosity. Following are applications of some impor-

tance:

a. Degradation of polymers. The author well remembers being in-
volved in a project to standardize the determination of the thickness of
a uv-absorbing coating that was applied to polymer film intended for
outdoor useB“. The polymer film showed progressive degradation upon ex-
posure to sunlight even for moderate lengths of time. Since the mechan-
ism for the degradation was not known the only alternative was to filter
out the uv light. A review published recently concludes that the photo-
degradation of several polymers is due for the most part to the type-II
photochemical cleavage85. Even in cases where no original carbonyl
groups are present, oxidations at some points in the chain during aging
or impurities in the monomer could produce them in the polymer in suf-

ficient amounts to cause photochemical degradation.

b. Synthesis of four-membered ring compounds. The ring closure
product of the 1,4—biradical occurring during the photolysis of the
phenyl alkyl and dialkyl ketones is obtained in yields ranging from poor
to good. Even in the cases where the yields are low this is probably
the most efficient means of synthesizing this highly strained ring sys-
temle. The products are in the form of cyclobutanols or oxetanols. An-
other synthetic possibility is the terminal olefinic product produced in
the type-II cleavage. 'For example, l-tridecene was formed during the

photolysis of pentadecanophenone.

25

c. Use as a model for determining substituent effects. Although
this aspect of the photochemistry of phenyl ketones is to be explored
in this project, this may prove to be an important application for the

type-II photoreaction in the future.

8. Definition of Terms.
The terms defined below will generally be used throughout the text

without further explanation.

a. Reactivity. This term has been used in various ways in the
past literature75s77a86 although several attempts have been made to
clearly define its meaning35s33’37’71. As used in this work, reactivity
refers to the facility with which the excited carbonyl abstracts a y-hy-
drogen and is expressed as a rate constant determined by quenching ex-

periments.

b. Non-radiative decay. A means of decay of the excited triplet
(of which quenching is an example) which competes with the hydrogen ab-
straction. Although its nature is not well understood in many cases its

rate constant, kd, can be determined.

c. Quantum yield. Quantum yields are determined by quantitative
measurements of product formation or ketone disappearance. The defini-
tion of a specific quantum yield is the number of moles of a specific
product formed per unit volume divided by the number of Einsteins (6=
one mole of photons) absorbed per unit volume by the reacting compound83.

The specific quantum yields measured in this work are:

26

¢II The quantum yield of type-II cleavage product. In this

work the carbonyl cleavage product was measured.
¢dis = Quantum yield of disappearance of parent ketone.
¢cyc = Quantum yield of cyclobutanol formation.
Also, the following probabilities are expressed in the notation of

quantum yields:

O = Quantum yield of intersystem crossing from excited singlet

to excited triplet state.

¢BR = The probability of formation of 1,4-biradical from the ex-
cited triplet state.
6 = Probability of the 1,4-biradical going on to products.

RESULTS

1. General Explanation of Data.

The ketones with the desired substituents were prepared or purchas-
ed (See Experimental Section, Part 1, Chemicals) and purified to meet the
stated criteria. Benzene, which was used as the solvent, and other com-
pounds used were also carefully purified before use. Solutions of the
ketones and an internal standard were irradiated with a medium pressure
mercury lamp (3130A) for a predetermined length of time and then analyzed
for product formation or ketone disappearance (See Experimental Section,
Part 2, Techniques). Some of the specific characteristics of the data

are described below.

a. Absolute type-II quantum_yields. The type-II quantum yields for
the ketones studied were determined by concurrently irradiating degassed
solutions of the ketone and an actinometer (usually valerophenone). The
quantity actually measured was the appearance of acetophenone product or
disappearance of parent ketone. Since the ¢II for 0.10M valerophenone in
benzene has been measured to be 0.3371, all of the quantum yields report-
ed are relative to valerophenone at this value. The product to standard
ratios of the photolyzed ketone and actinometer solutions were measured

by gas chromatography (VPC) (See Experimental Section, Part 2).

b. Disappearance and cyclobutanol quantumpyields. Disappearance
quantum yields (Odis) were determined by measuring a product to standard
ratio of the parent ketone before and after photolysis of the sample sol-
ution. The concentrations in this case had to be quite accurately known
($0.001M) so that the number of moles of ketone actually disappearing

could be accurately calculated. The same general conditions and methods

27

28

were used to determine the quantum yield for the presumed cyclobutanols.
In this case, however, they were measured by appearance of products in
the photolyzed samples and taken as a percentage of the parent ketone in

the unphotolyzed solution. (See Experimental Procedures, Part 2)

c. Solvent effects on thegguantum yields. The effect of varying
the solvent on the type-II quantum yield was determined for several ke-
tones. This was usually done by measuring the type-II quantum yield of
the ketone upon gradually increasing the concentration of tert-butyl al-
cohol. Another method used was to substitute another solvent entirely

or to modify the benzene by addition of a fixed amount of a co-solvent.

d. Stern-Volmer quenching slgpes. The Stern-Volmer quenching
slopes for the ketones studied were Obtained by photolyzing them in solu-
tions containing varying amounts of quencher (See Experimental Section,
Part 2) and then plotting the ratio of the type-II quantum yield without
quencher over the value with quencher versus the quencher concentration.
The intercept on the vertical axis should be 1 and the slope represents
the value of k r. The quencher most commomly used in this study was

q
2,5-dimethyl-2,4-hexadiene.

e. Miscellaneous data. Other important data which were needed or
desired, such as concentration effects and quantum yields at different
percent conversion, are represented graphically. The raw data is con—
tained in Appendix A, Part 5. Results of viscosity measurements on ke-
tone solutions are treated in the section on justification of experimen-

tal results.

29

2. Tabulated Results.

The experimental results obtained from the kinetic studies and
quantum yield studies of alkyl-substituted phenyl ketones are tabulated
in Table 1, those for the ring-substituted phenyl ketones in Table II

and the information pertaining to solvent effect studies in Table III.

a. y-Substituted alkyl pheny ketones. The values presented for
aII and kr in Table I are, with a few exceptions, averages of two or more
separate determinations. The variation shown is the actual experimental
spread found for the determinations. For example, if two values for aII
were measured as 0.34 and 0.36, the aII in the table would be 0.35 1 .01.
The kr values were determined from the least squares analysis (See Appen-
dix B) of Stern-Volmer slopes for each individual run (Appendix A). The
ketones with only one quenching run performed on them are: (1) Butyro-
phenone, y-methoxy- and y-phenylbutyrophenone, the values of which agreed
Closely with previously determined values71; (2) pentadecanophenone, of
which only sufficient ketone was available for one normal run; and (3)
e-cyano- and e-chlorohexanophenone and B-phenylbutyrophenone which were
not checked because of time limitation. The data for pentadecanophenone
seems reasonable based on data for similar ketones. An earlier run with
B-phenylbutyrophenone using an excess amount of quencher had indicated
that the qu value was quite low (=5), so the Single repeat run seems to
be valid. It would have been desirable to repeat the measurements for
e-cyano- and e-chlorohexanophenone as the values are quite revealing.
The e-cyanohexanophenone was suspected of having an impurity and after
one attempt at further purification the aII for 0.10M ketone increased

from 0.24 to 0.28. The quenching lepe can then be assumed to be larger

30

by a like percent. The prepared nonanophenone used in this work showed

a sizable impurity (220%) in the mass spectrum which was larger by 14
atomic mass units. The octyl bromide used to prepare this ketone showed
the same characteristic, as did a commercially obtained sample of nonano-
phenone. Analysis of the ketone by VPC showed only one Sharp peak. It
is believed that the impurity is the one higher homolog, decanophenone,
and that no change in the photochemistry of nonanophenone occurs by its
presence. It is felt that the values in Table I are quite good and

should suit the purpose of this study.

The disappearance and cyclobutanol quantum yields were the result of
a single run on each compound, the main purpose being a qualitative check
to determine whether the type-II process was the major reaction. It is
noted that a-carbomethoxyvalerophenone has no values here as the parent
ketone would not come off the VPC column. It is also noted that some of
the ketones produce a quencher upon type-II cleavage. The method of
handling this was to photolyze the ketone solutions to several different
percent conversions and then extrapolate back to zero conversion for aII
and kqt. This will be explained further in the section on justification

of experimental results. Other pertinent information relative to the

data is included as referenced notes to the table.

b. Ringpsubstituted alkyl phenyl ketones. The type-II quantum

 

yields and Stern-Volmer quenching slopes for a number of ortho-, meta-,
and para-substituted valerophenones are tabulated in Table II. The var-
iation indicated is the actual spread in experimental values and each

result is the average of two or more separate determinations. Each

31

TABLE I. Triplet-State Reactivities and Quantum Yields of Phenyl

Ketones:
‘<::*C‘CH2 -R ‘ k5

k7?

 

 

 

 

Ketone (R = ) qua’M-l [2433]] ¢II ¢dis (¢cyc)

(Primary Hydrogens)

-CH2CH3 568 0.88 0.35:.01C 0.45 0.033
-CH(CH3)2 240:3c 2.1 0.35:.01 0.41 0.040
-C(CH3)3 81:5 5.2 0.19:.005 0.21 0.00

(Secondary Hydrogens)

-CH2CHZCH3 41.0:.05 12.2 0.33:.01 0.43 0.075
-CH2CH2CH2CH3 38:1 13 0.30:.01 0.35 0.075
-CH2CH2 CH(CH3)2 27:1 18.5 0.25: 01 0.35 0.074
-CH 2CH 2C(CH 3)3 24:.05 21 0.24:.01 0.34 0.094
-(CH2)6CH 3 32:3 15 0.25:.01 0.33 0.078
-(CH2)]2CH3 28 18 0.25:.00 0.31 0.025
(Tertiary Hydrogen)

-CH2CH(CH3)2 10.2:.5 49 0.25:.01 0.34 0.020
(Benzyl and Allyl Hydrogen)

-CHZCH24<::> (12.3)8 (§;)e 0.50d 0.50d 0.055d
-CH2CH2CH=CH2 11.3d 44 0.25d 0.33d 0.042d
a k r is the Stern-Volmer quenching slope; k = 5 x 109 M'lsec"].

q q

kr is taken to equal 1/1 for the ketones in this table.

b

C The error cited here is the actual range of values obtained for the
ketone in two or more runs. For standard deviations of separate runs
see Appendix A.

d Extrapolated to zero conversion to correct for product quenching.

e Corrected value from known percent lowering of all.

TABLE 1., Continued.

(Ketone (R = )

k 1,14"1

(y-Heteroatom Substituents)

-CH2CH20CH3

3)2

-CH2CH2C1

-CH2CH2COOCH3

-CH CH

2 2CN

0.60:.1
7.8:.5

10.5

13.0:.8

180:4

490d

1330d

(6-Heteroatom Substituents)

-CH2CH2CH2C0@

-CH2CH2CH2COOH

-CH2CH2CH2C1

55:3

131:3
189:2
230:5

516:14

(e-Heteroatom Substituents)

-CH2CH2CH2CH2C1

-CH2CH2CH2CH2CN

(a and B Substituents)

-CH«<:>»CH3

-0CH3

88.5

74.5
(87)e

32

 

kr

[x103] ¢

sec J II
830 0.025:.001
54 0.23:.01
48 0.20

38.5 0.31:.04
2.8 0.090:.005
1.02 0.50d
0.38 0.32d

9.2 0.34:.03
3.8 0.51:.03
2.5 0.56:.01
2.2 0.58:.02
0.97 0.48:.02
5.7 0.44

5.7 0.24
(5.7)e (0 28)e
100 0.0019
185 0 54:.02

¢dis

 

cyc
0.058 0.002
0.37 0.089
0.28 0.045
0.42 0.00
0.34 0.003
0.51d 0.00
0.72d 0.00
0.34 0.055
0.79 0.18

not obtained

0.81
0.57

0.54
0.35

0.018
0.94

0.085
0.045

0.012
0.00

0.000
0.31

 

33

TABLE II. Triplet-State Reactivities and Quantum Yields of Phenyl

Ketones: (§:>»C49
R ‘CH CH CH CH

 

 

 

 

 

2 2 2 3
1/T
ortho-CF3 38.0 : .4c 13.2 0.20 : .01
meta-CF3 15.5 : 1 32.2 0.23 : .01
para-CF3 l8 : l 28 0.26 : .01
ortho-F 34.7 : .3 14.4 0.33 : .01
meta-F 28 : 3 18 0.27 : .01
para-F 34 : 1 14.7 0.36 i .00
ortho-Cl 141 : 9 3.5 0.45 : .03
meta-Cl 32.0 : 2.5 15.6 0.33 : .02
para-C1 135 : 3 3.7 0.29 : .02
meta-CH3 128 : 4 3.9 0.34 : .02
para-CH3 272 : 1 1.84 0.39 : .00
a k r is the Stern-Volmer quenching slope; k = 5 x 109 M'lsec'].

q q

b all determined by the appearance of the substituted acetophenone.

c Error where cited is the actual range of values in two or more runs.
For standard deviations of each individual run see Appendix A.

34

TABLE II., Continued.

 

 

 

 

UT

-1
Ketone (R = ) qu, M [x 107 sec-1] ¢II
ortho-OCH3 variable 0.20 : .05d
meta-OCH3 320 : 30d 0.013 : .003d
para-OCH3 2250 : 50 0.22 0.13 : .01
para-SCH3 ---- ---- 0.000
para-OH ---- ---- <0.002
[Alkyl portion = y-methylvalerophenone]
meta-OCH3 (200)e ---- 0.030 : .001
para-OCH3 865 i 22 0.58 0.19 i .02
para-Q ---- ---- 0.0002

 

d These determinations have wide variations and should be regarded as

estimates. The k T values for these ketones were determined
graphically and ngt by least squares analysis.

e This value is from one run which is considered to be the most
reliable of a series of runs.

35

individual slope (with a few exceptions) was determined by a least
squares computer program on the CDC-6500 (See Appendices A and B). The
results which were not subjected to least squares analysis were those
for ortho- and meta-methoxyvalerophenone and meta-methoxy-y-methylval-
erophenone. These ketones proved quite difficult to work with and ap-
parently are quite sensitive to a number of variables, some of which are
unknown. The values indicated are averages of the most reliable runs
and the variation indicates the Spread between the values. Also, the
para-phenyl- and parahydroxyvalerophenone had such low quantum yields
that determining a quenching slope was not practicable. The disappear-
ance and cyclobutanol quantum yield studies are not presented for the
ring substituted ketones, due in part to the practical limitation of
time and to other difficulties (such as possible alternate photoprocess-
es) which might occur with these ketones. This aspect will also be
treated in the section on indications for further study. Other informa-

tion pertaining to the table is included as footnotes to Table II.

c. Solvent studies. In the latter stages of this project it be-

 

came obvious that information available from solvent studies was needed.
Although all ketones could not be studied, those with O-Substituents
which were of a critical nature in determining inductive effects for
substituents, several others which had an unusually low all, and a num-
ber of the more common ketones were studied. Table III contains the da-
ta for these runs. In the first portion of the table values of all with
increasing tert-butyl alcohol are shown; the raw data are available in
Appendix A, Part 4. The important point here is to show the highest

value achieved upon addition of alcohol. One aspect of the data on

36

.cmpoe o.m coppmcucmucou pocoupo quzmuugmu u

.meos o.m cowpmcpcmucou posou—m quzm1pcmu
a

n

.mcmwvmch-¢.m-cmumewuum.m m? com: cmnucwzc .maoFm mcwgucmac cmEFo>ucgmpm mg» mp » x m

 

om

09x

 

m¢.o ----
omoo.o ----
ma.o mu.o
mm.o mm.o
mm.o oo.o
---- Ne.o
mm.o 3m.o
500.2
hm.o Am.o
8~.o om.o
em.o mm.o
zo.m zo.m

 

mm.o Nu.o
mo.o mm.o
no.0 Fm.o
nmm.o Fm.o
nvm.o mm.o
nvm.o en.o
nvm.o mm.o
nme.o mm.o
nwm.o en.o
zo.m zo._

fixesm-u :83: HHSH

ow.o
5m.o
mFoo.o
_o.o
mm.o
m¢.o
mm.o
om.o
mm.o
8N.o
mp.o
Km.o
HHS

 

mcmNcmm mcwuvcxa zm.o + mcocmgaocm—m>
mcmNcmm wcmuanpzo~cmnwo-e._
mcmncmm acocmgaogxnanpacmsa-m

mcmncmm mcocmsaocwFm>>xogumsoncm018

mcm~cmm mcocmsaocmpm>ogo_guuc
mcmNcmm mcoconaocmpm>ocmau-o
mcm~com mcocmzaocxuznAxocuazu>
mcmNcmm «cocogaocmxmx
mcm~cmm mcocmgqocmpm>
acm~cmm mcocmzaocmpm>Fxgu521>
w=m~cmm mcocmgqogxuznszume_a-m.m
acm~cmm mcocmzqocmpm>omH

 

ucm>Fom vcmucmpm accumumcump zvoo.o
.u:m>—om cw zop.o .mzopmx

 

.mmcouox vmuumpwm :o mmwvspm pcm>pom Low mama .HHH m4m<h

37

.cwcucmsc mm com: acmpmspcam: .umm: ugmvcmpm accumumucmg

u

 

-111 ooow
omo.o own
oF.o m._

mm.o _.Hm.¢

m.o me
p.— cop

 

zo.m

mm.o
vooo.o
Poo.o
Foo.wmoo.o
mo.o
Po.HmN.o
vo.wmw.o
eo.HNm.o

 

zo._ HHS

Hzozm-p ;o_z HHGH

_o=5;aaz
_ocmsaaz
Loam:
_ocmgpaz
a_2:6w=ouau<
_o=8;uaz
a__ca_=oaau<

Focaspaz

uzm>~om

.cmacwucou

umcocmnaocmPm>_»:umz-mcma

auwsocm Fsgpm
mcocmnaocxuanocwEmPaspmewo1>

muwgopguocv»:
mcocmsaogxuznocPEmFxcpmewo-»
ococmgaocxuanocwsmngumepou>
acocmcaocxuaao:Pampxzumswou>
mcocmsaocmpm>

mcocmgaogmpm>

 

ucmucoum accumumgpmu zeoo.o
.pcm>Pom cw zo_.o .mcopmx

.HHH m4m<k

38

solvent effects that is a bit disconcerting is the case of the 6-substi-
tuted valerophenones where the aII drops off after reaching a maximum at
low alcohol concentration. This could indicate a small quenching impur-
ity in the tert-butyl alcohol or the alcohol itself might possibly be
affecting all. The values are the result of a single run in most cases,
some values were checked, however, and this is indicated by a Spread in
values. Other pertinent information is included as footnotes to the

table.

d. Variation of ketone concentration. For sake of consistency all
of the photochemical data were determined with 0.10 molar ketone solu-
tions. For the results to be reliable at least 99% of the light should
be absorbed by the solution and it was found for several of the ketones
that were checked that the absorbance was well over 2 in the vicinity of
3130A. The aII at this concentration is raised somewhat because the ke-
tone increases the polarity of the solvent medium analogous to a polar
co-solvent59. To compare quantum yields in a non-polar medium they
would have to be extrapolated to zero ketone concentration. Results
with valerophenone upon addition of ethyl acetate were similar to those
with increasing ketone concentration. The aII's for 0.05M, 0.10M, and
0.20M valerophenone solutions were 0.31, 0.33, and 0.35 respectively71.
When 0.10M and 0.20M ethyl acetate were added to 0.10M valerophenone the
aII was 0.35 and 0.37 respectively. Pentadecanophenone (Figure 3, line
B) has a gradual increase in aII with ketone concentration which resem-
bles quite closely that of valerophenone. In the case of ketones with a
polar functional group on the alkyl chain additional enhancement should

be seen. The a-benzoyl group clearly does this in the

39

 

 

0.7 ‘ /F
’r'G C F'

0.61 7”

0.5 i

 

 

0.1 J

 

 

 

' t W—

0 5 10 _215 20 25
Ketone concentration, molar [x 10 ]

Figure 3. Relationship of Type-II Quantum Yield to Ketone Con-
centration for Several Ketones. [A] e,s-Dimethylbutyrophenone,
[B] Pentadecanoghenone, [C] 1,4-Dibenzoylbutane, [D] -Hydrox§-

buterphenone, E] v-Carbomethoxybutyrophenone, and [F] & [F'
5-Carbomethoxyvalerophenone.

1,4-dibenzoylbutane (C) and in cases where limited data is available:
y-hydroxybutyrophenone (D), y-carbomethoxybutyrophenone (E), and a-carbo-
methoxyvalerophenone (F) and (F') the trend is recognizable. However,
the 8,B-dimethylbutyrophenone (A), which has reactive primary hydrogens,

Shows very little increase with concentration. Similar behavior has been

40

observed for butyrophenone by others‘?’0 so this is accepted as character-
istic of these compounds. This test was also used to determine whether
quenching impurities were present in the ketone being studied and will

be further discussed in the section on juStification of data.

3. Justification of Data and Controls Relative to Photochemical Data.

a. Identification of cyclobutanols. The term "presumed cyclobutan-
015" is used in this study and the symbol (acyc) is enclosed in brackets
because the products here presumed to be the cyclobutanols were not ac-
tually separated and identified. There is substantial evidence, however,
to support the assumption that the product(s), where claimed, is the
cyclobutanol(s). First, it is definitely a photoproduct appearing in
significant amounts upon photolysis of most of the ketones (See page 4),
and has a retention time slightly less than the parent ketone on a polar
VPC column (Figure 15). Also, where gj§_and tran§_isomers of the cyclo-
butanols are expected, two peaks are observed in the VPC analyses (Ap-
pendix A, Part 4) in most cases. As mentioned previously the cyclobutan-
ol from butyrophenone and the gj§_and trag§_cyclobutanols from valero-
phenone, y-phenylbutyrophenone, and a-benzyloxyacetophenone have been
separated and each characterized by its nmr spectrumzo. Recently Turro
and Lewis89 separated and identified the oxetanols of a-methoxyacetophen-
one and several other a-alkoxyacetophenones by their nmr spectra. In the
above cases, therefore, the assumption that the additional photoproduct
observed is the cyclobutanol(s) has a strong basis and for the remainder
of the ketones the photoproduct(s) analogous to that of the above men-
tioned examples is assumed to be the cyclobutanol(s). Although many of

the ring substituted ketones were observed to have photOproductS which

41

were very likely the cyclobutanols, none are reported here.

D. Detectingpquenching impurities in the ketone. Increasing the

 

concentration of ketone would also increase the concentration of any
quenching impurity which is present in the ketone and would therefore
decrease the quantum yield. Most of the ketones represented in Figure

3 were tested for quenching impurities and the results indicated that no
Significant amount of such were present. The e-cyanohexanophenone which
was suspected of having a small amount of quenching impurity did not
show as rapid an increase in ¢II as would be expected for a ketone with
a polar functional group on the alkyl chain. The increase with concen-
tration is even slightly less than that for pentadecanophenone, so it is

likely that an impurity is offsetting the normal concentration effect.

Another important observation is that the good linearity Of the
lines A, B, and C indicates that within an experimental range of a per-
cent or two all of the light is being absorbed by the ketone solution.
What happens when this is no longer true is seen in the cases of y-car-
bomethoxybutyrophenone (E) and 6-carbomethoxyvalerophenone (F') at the
0.02M ketone concentrations. The apparent quantum yield is below the

extrapolated line since only 85-90% of the 3130A light is absorbed.

c. Does the type-II quantum yield varypsignificantly with percent
conversion? With respect to the measurement of the amount of products
formed upon photolysis of the ketone solutions, it was assumed that for
conversions of low percentage (up to 8%) the 911 would remain constant
throughout the duration of the run. This assumption is especially criti-

cal since the quantum yields Of all ketones were determined relative to

42

 

 

 

 

 

 

 

a n D
10 V u =fi=®=c
f
4.:
3
U 91
O
s.
a.
\.
z 39’
5 \\
:8 \
l— \\\
,2 \ \\
o \171 \
I'- ~"\ \ \ ‘
'40 R.) \\ \ ‘1
3 \\ 1" \ \
X A n A \\‘<:-7
'T V U W“ (9 3
$- 6'l ‘\\
S ‘1
x QT):
33
A
'6 V G 47—“ A
E 5.
0r-
,—
F
'I-
z I I I I j I ' '
0 30 60 90 120

Minutes irradiation +

fjgg§g_fl, Product Formation Per Unit Time versus Irradiation Time.

[A] & [B] 0.10M Valerophenone, [C] 0.10M & [D] 0.05M 5-Carbometh-
oxyvalerophenone, [E] 0.10M & [F] 0.05M v-Carbomethoxybuterphenone.

valerophenone by simultaneous irradiation. The only practical means of
checking this would be to irradiate the ketone solution for increasing
lengths of time and measure the product yield per unit time. This was
done in the early stages of this work for several ketones. Here another
assumption, a technical one, that the output of the mercury lamp would
not vary significantly for Short irradiation periods was made. For long-

er periods of irradiation of 10 or 20 hours the output noticeably drOps

43

Off due to mineral deposits in the cooling jacket of the lamp. Because
of its importance as an actinometer valerophenone was tested several
times; two typical runs are shown in Figure 4, lines (A) and (B). The
millimoles of product per hour per liter is very consistent at various
percent conversions, which in the case of (B) is in excess of 12%. The
very slight fading noticed is undoubtedly due to slightly decreasing

lamp output. The o-carbomethoxyvalerophenone, whether 0.05M or 0.10M,
had a very consistent rate of product formation out to 15% conversion for
the 0.10M ketone (C) and 30% conversion for the 0.05M ketone (0). Also,
the photolysis of valerophenone has been carried out to over 75% conver-
sion with no apparent decrease in a1190. The relationships shown in Fig-
ure 4, (A), (B), (C), and (0) could have occurred only if the assumptions
made were good ones, SO for low percent conversions the aII is considered
to be constant. This is not so if a quencher happens to be produced in
the photolysis, as can be seen for y-carbomethoxybutyrophenone (E) and
(F). In cases where product-quenching occurs a separate treatment is

used.

d. The problem ofproductgguenching;_ Four of the ketones listed in
Table l produce an olefin fragment which acts as a quencher; these are
y-cyano-, y-vinyl-, y-carbomethoxy-, and y-phenylbutyrophenone. As more
product is produced the type-II quantum yield is progressively decreased,
as illustrated for y-carbomethoxybutyrophenone in Figure 4, (E) and (F).

This "added" quencher decreases the k I value in a quenching run because

q
the "unquenched" samples actually are quenched. In order to have a more
accurate value for qu the y-cyano-, y-vinyl-, and y-carbomethoxybutyro-

phenone were photolyzed to several different percent conversions each

 

 

 

 

B
2.5-
A
2.04
mg 1.4- C
x d
p
XO-
2 ..
K
0.8-
0 1 '2 3 4 5 5

Average product-quencher concentration, molar [x 10'3]

figgrg_§, l/k 1 versus Average Quencher Concentration.

[A] y-Cyanobutyrophenone, [B] y-Carbomethoxybutyrophenone, and
[C]]y-¥inylbutyrophenone (vertical scale expanded by a factor
of 00 .

and the reciprocal k r values were plotted against one half of the pro-

duct concentration (gee Figure 5). The intercept at zero product concen-
tration is taken as the best estimate of the actual qu. The method for
finding aII is to plot l/aII versus the average quencher concentration
(one half of the product concentration). The intercepts were determined
graphically using the known data points (See Figure 6). Only one qu

value, at 5.6% conversion, was found for y-phenylbutyrophenone. From the

plot of l/aII versus one half the product concentration aII at 5.6%

45

 

7.0- A

l-a

C)
C)
C)
C)
C)
C)
O

 

 

 

T 1' I I V I

0 l 2 3 4 5 _3 6
Average product-quencher concentration, molar [x 10 ]

Ejggrg_§, 1/aII versus Average Quencher Concentration.

[A] y-Cyanobutyrophenone, [B] v-Carbomethoxybutyrophenone,
C] y-Phenylbutyrophenone, and [D] v-Vinylbutyrophenone.

conversion is found to be 0.46. By multiplying by the same proportion
needed to arrive at the extrapolated aII of 0.50, qu would then be 13.3
at zero conversion. From the Stern-Volmer relationship (Equation 16) it
is seen that l/aII is a linear function of kq [OJ/Poll with an intercept
of l/aoII. Since qu is directly related to l/aII it should also have a
linear plot versus the average product concentration with an intercept
of 1/quo (Figure 5). When extrapolated back to zero average product-
quencher concentration the best values for the qu's were found to be:
y-vinylbutyrophenone = 11.3; y-carbomethoxybutyrophenone = 490; y-cy-
anobuterphenone = 1330. Further discussion on product quenching appears

in the Discussion Section, Part 7, Indications for Further Research.

46

The disappearance quantum yields for these ketones are estimated by
extending the plot in Figure 6. The average quencher concentration was
taken as one half the ketone which disappeared (less the contribution
from cyclobutanols). Then aII was found for this average quencher con-
centration and a ratio of extrapolated aII to product-quenched aII was

found. The extrapolated value for adis was then found by the following:

aII(Extrp)
911(EXP)

 

qdis(Extrp) = x adis(Exp) (Eq. 38)

Cyclobutanol quantum yields, where shown, were found in the same manner.

e. Solution viscosity. For the ketones with secondary hydrogens

 

in Table I it is noted that as the chain length increases the kq: value
decreases. This could be caused by a change in r, or perhaps a change in
kq. If the solution viscosity significantly increased with some ketones
kq may be noticeably affected, since from the Debye equation (Equation
17) it is seen that kq is inversely proportional to the viscosity coef-
ficient n- Tests were made with pure benzene, 0.10M valerophenone and
0.004M tetradecane in benzene, and 0.10M pentadecanophenone and 0.004M
tetradecane in benzene, the latter having the longest alkyl chain as
well as the highest molecular weight of the ketones measured. An Ost-
wald viscometer was used and the liquid flow was timed with a stopwatch.
The comparative times are reported as they are proportional to the vis-
cosity coefficient91. Although n for the pentadecanophenone solution is
larger by about 4.7% this cannot account for the decrease in qu which

is about 32% lower than that of valerophenone. Based on the comparison

in Table IV changes in the viscosities of the sample solutions due to

47

TABLE IV. Viscosity Measurements for Benzene, Valerophenone, and

 

 

Pentadecanophenone.
Average time of
Sample Solution 4 determinations
Pure benzene 2 min 47.8 sec

0.10M Valerophenone, 0.004M tetradecane in benzene 2 min 52.3 sec

0.10M Pentadecanophenone, 0.004M tetradecane in 3 min 00.3 sec
benzene

 

the various ketones are considered to have a relatively minor effect on

the reactivity measurements.

DISCUSSION

1. Quantum Yields of Side-Chain Substituted Ketones.

As was previously mentioned the quantum yields of product formation
or ketone disappearance are the only measureable quantities and all re-
lationships are derived from them. Table I lists the quantum yields
found for the ketones with alkyl substituents. For virtually all of
these ketones aII is determined by the behavior of the biradical, as

shall be shown in the discussion of solvent effects.

a. 9115’ a test for material balance. With only a few of the ke-
tones studied has it been previously shown that the type-II reaction ac-
counts for practically all of the products. In order to establish whe-
ther aII (plus cyclobutanols) is quantitative an independent determina-
tion was made of the quantum yield of disappearance. The quantum yield

of cyclobutanol formation (a ) was also determined independent of all.

cyc
The condition required is:

4’dis = 4’II + (¢cyc) (Eq. 39)

The ketones in Table I with hydrocarbon side-chain substituents all ful-
fill this condition quite well. Small discrepancies are seen for butyro-
phenone and y-methylvalerophenone where the adis appears larger than can
be accounted for by experimental error (:4% reliability, See Experimen-
tal Section). Intermolecular hydrogen abstraction may be competing in
the case of butyrophenone as photoreduction products have been reported
by others77. This is probably not the case for y-methylvalerophenone,
however, as the tertiary hydrogen reacts about 60 times faster than the
primary hydrogens of butyrophenone. It is possible that a coupling pro-

duct may be formed here since both radical sites are quite stable, but

48

49

until more is known about the biradical lifetimes much of their behavior

will be a matter of speculation. The low apparent (a ) for pentadec-

cyc
anophenone compared to the other ketones with secondary hydrogens also
deserves comment. It is difficult to attribute this to any steric ef-
fect since nonanophenone has a (¢cyc) three times larger. Also, only
one peak was apparent in the VPC analysis for the cyclobutanol (Appendix
A, Part 3), but this may be an analytical shortcoming. At the tempera-
tures required for parent ketone analysis nonanophenone also appeared to

have only one cyclobutanol peak, but at lower temperature two overlap-

ping peaks could be seen.

Several of the ketones in the second half of Table I which have
heteroatom substituents do not show good material balance. For example,
the all accounts for only 50% of the disappearance for y-dimethylamino-
butyrophenone, 75% for y-hydroxybutyrophenone, and less than 30% for
y-chlorobutyrophenone. The difference for y-dimethylaminobutyrophenone
can be rationalized as due to intermolecular photoreduction. Amines are
known to photoreduce triplet carbonyl species92 and valerophenone has
been found to have a adis of 0.64 with 0.80M triethylamine present in
benzene72. Small but Significant amounts of photoreduction may also ex-
plain the difference for v-hydroxybutyrophenone Since there effectively
is 0.10 molar secondary alcohol present in the solution. A few percent
competing intermolecular hydrogen abstraction could account for this
discrepancy. Special nOte is given to the fact that y-, 6-, and :-
chloro- substituted ketones all do not satisfy Equation 39. It is temp-
ting to attach some special Significance to the chlorine in these cases.
When y-chlorobutyrophenone was photolyzed in benzene an additional pho-

toproduct was produced in good yield which had approximately twice the

50

retention time on the VPC as the parent ketone. The degassed solutions
turned yellow during irradiation but became colorless upon exposure to
air. Although no conclusions can be drawn at this point, the observa-
tions suggest that a higher molecular weight photOproduct, possibly a
dimer, is also formed and that it apparently is air sensitive. If this
can be attributed to the presence of an alkyl chloride substituent, it
would seem logical to assume that the 6- and e-chloro-ketones would be
similarly affected but to a lesser degree since the chlorines are fur-
ther from the reaction site. This situation deserves further study to
determine whether chlorine substituents do in fact play some role in the

photoreaction.

For the B-phenylbutyrophenone aII is an order of magnitude smaller
than adis but even in the latter case the quantum yield is extremely
small. Such small quantum yields require long irradiation periods and
it would not be surprising if small amounts of photoreduction were to
take place in the solution. The importance of adis here is that it ver-
ifies the unreactive nature of this ketone. Possible reasons for this
will be forwarded later on in the discussion. Finally, the y-cyanobuty-
rophenone also falls into this category, however, the comparison must be
treated cautiously. In this case the efficient quencher, acrylonitrile,
is produced during the reaction and the extrapolated slope (21300) indi-
cates that the reaction is very sensitive to quenching. In order for
the extrapolated adis to be accurate no processes competing with aII can
occur, or if they do, they must be quenched with the same efficiency as
the type-II process. Since there is no basis for making this assumption

the material balance for y-cyanobutyrophenone cannot be evaluated. The

51

other ketones which have product quenching are much more reactive and
therefore any differences in product quenching between aII and adis

would be smaller. This supposition agrees with the data. An overall
view of the results of adis shows that the type-II reaction accounts for
virtually all of the photoreaction except in special cases where good
hydrogen donating substituents are present, and apparently where reactive

heteroatom substituents, such as chlorine, are present.

b. A search for steric effects on all. After noting the consider-
able variations of aII caused by substituent groups on the y-carbon, it
became necessary to establish whether or not steric factors were invol-
ved. The usual low efficiency of these ketones can be attributed to re-
vertible hydrogen transfer, however, the dimethylamino group on the y-
carbon lowered aII by an order of magnitude to 0.025. It seemed possible
here that the substituent may have reached a critical size which inter-
fered with hydrogen abstraction. To test this possibility ketones with
alkyl groups of increasing size on the y-position, yet all having second-
ary y-hydrogens of the same reactivity, were measured. The results from

this test (See Table I) are summarized below:

all: 0.33 0.30 0.25 0.26 0.25 0.24

 

The tendency of aII is to decrease, however, the effect is seen to be
gradual. The isopropyl group, which is about the same size as the di-
methylamino group, causes no drastic reduction in all, and neither does

the tert-butyl group. It would also appear that for long straight-alkyl

52

chains 211 reaches a minimum of about 0.25. It is apparent from this
that little steric hindrance from the y-substituent is involved, or at
least the critical size has not been reached. It can also be inferred
that the geometry required for type-II cleavage is quite close to that
achieved in the psuedo 6-membered ring formed upon hydrogen abstraction.
If a substantial amount of rotation or movement of the alkyl side chain
through the solvent cage were required it would seem that going from a
methyl to a tert-butyl group would produce a larger effect on aII than
the approximate 25% drop observed. The trend noted as the size of the
group increases can be rationalized. It was prOposed by Wagner71 that an
optimum geometry for type-II cleavage of the biradical would be one in
which the carbon atoms would be coplanar and in which maximum overlap be-
tween the p-orbitals of the biradical and the breaking a-B bond is pre-
sent. As the biradical is formed it would tend to be pushed out of co-
planarity by the eclipsing interactions of the hydrogens on the a and B
carbons. This argument has also been used by Lewis and Hilliard93 in
their work on cyclization versus cleavage of substituted butyrophenones.
Another source of poor overlap is possible in the case of large alkyl
substituents on the y-carbon. There may be sufficient interaction be-
tween such a group and the phenyl or hydroxyl group to skew the p-orbital

30

H [OH R
_ \ /R "r _c 0:7 _C/OH fie?“
C\ — c/C\R —_‘* /E};RL—+ \ --,C: Q
11' 1-1 I: each 11' 11ch H H
-C 0“ + ORR (E 40)
_ q.
\CHZ HZC/

S'lightly out of the plane of the carbon atoms. Both situations would

53

necessitate slight rotation to attain optimum overlap for type-II clea-
vage and this ability to rotate may be affected by increasing substituent
size on the y-carbon. The effect of increased eclipsing interactions is
quite evident with two B-methyl groups as the aII for 8,8-dimethylbutyro-
phenone drops to 0.19.

c. Steric effects on (a ). The fact that 8,8-dimethylbutyrophen-

 

cyc
one yields a negligible amount of cyclobutanol can be attributed to a

steric effect on the formation of the cyclobutane ring. The two methyl
groups on the B-carbon would end up on the 3-position of the ring and

would interfere with either the hydroxyl or phenyl group (Equation 41).

OH CH
@C )3 ———> “0 CH3 (Eq. 41)
CH3 CH3

This effect was independently found by Lewis and Hilliard93.

It was surprising that no cyclobutanol could be detected for the y-
hydroxybutyrophenone since there would seem to be no steric problems. A
unique characteristic of this cyclobutanol is that it is a 1,2-glycol and
it is possible that its formation involves unusual interactions. It is
also possible that the glycol either decomposes during VPC analysis or is
held up for an unusually long time on the column. No cyclobutanols were
observed for y-carbomethoxy-, y-cyano-, or e-cyano-ketones, however, this
may be due to analytical conditions so their absence is not conclusive.

The large enhancement of (a ) for a-methoxyacetophenone is almost

cyc
certainly due to relief of ring strain by having an oxygen as one of the

atoms in the oxetane ring. Turro and Lewis89 found large acyc's for

54

several a-alkoxyacetophenones which they studied. Also, the high ¢II
for this compound may result from a faster mode of type-II cleavage

(Equation 42).

0H 0H OH
I - /
ochre —~ @R;;_é”2 ——» @sz («”2
H (Eq. 42)

d. Solvent effects on 911' ¢II can be influenced by either com-

 

petitive decay of the triplet (quenching, for example) or by return of
the biradical to form the ground state ketone. If the biradical is in-
tercepted for some reason, ¢II would be lowered but ¢dis or kr would not
be affected. The prOposed modification of the original mechanism add-
ed the steps:

Considering the expression for the type-II quantum yield, ¢II = ¢BR¢p’
the importance of Wagner's69 proposal becomes obvious. When the maximum
¢II is found in alcohol solvents it can be assumed that all the biradi-
cal is going on to product, so ¢p c l. This allows ¢BR to be determined
by Equation 36 and kr can be found by Equation 37. In Table I the kr
values were assumed to equal l/r. For this to be true ¢II in alcohol
should be near unity to insure that the large majority of triplet is pro-
ceeding to biradical. Table V summarizes the results from Table III for
those ketones which were tested. Taking (acyc) in benzene into consid-
eration and using a "rule of thumb" that a combination of ¢II(alcohol)

55

TABLE V. Maximum aII in Alcohol Sovents.

 

 

 

 

Adjusted
Maximum 4’11 (¢cyc) k x 107 sec-1

Ketone in Alcohol from Table I r
Isovalerophenone 0.88 0.04 none
3,3-Dimethylbutyrophenone 0.76 0.00 nonea
Hexanophenone 0.86 0.08 none
Valerophenone 1.00 0.08 none
y-Methylvalerophenone 0.87 0.02 none
y-Hydroxybutyrophenone 0.72 0.00 nonea
a-Cyanovalerophenone 0.7l 0.05 0.74
6-Chlorovalerophenone 0.69 0.09 l.7
6-Carbomethoxyvalerophenone 0.72 0.l8 none
l,4-Dibenzoylbutane 0.43 0.06 4.5
y-Dimethylaminobutyrophenone 0.25 0.002 208

 

a aII(alcohol) appears to extrapolate to higher values for these ketones
(See Results Section).

plus (a ) less than 0.90 requires correction according to the relation-

c c
ship in Equation 37, an adjusted kr can be calculated. The a-substitu-
ents were specifically measured because of their usefulness in determin-
ing the inductive effects on the reactivity of the y-hydrogens. Except
for the l,4-dibenzoylbutane fairly significant increases in aII were no-
ted in alcohol, however, it is not certain that the maximum had been
reached in all cases (See Results Section on solvent effects). The sol-

vent effect on qu will be treated later on in the discussion on that

topic.

56

e. all of 6-substituted ketones. With the exception again of the
l,4-dibenzoylbutane, the aII of the 5-substituted valerophenones is no-
ticeably higher than those with plain hydrocarbon side chains. This in-
crease by roughly a factor of 2 could conceivably be due in part to in-
tramolecular hydrogen bonding which would assist in overcoming some of
the eclipsing interactions. This explanation is weakened somewhat by
fact that chloro and cyano groups are poor hydrogen bonders. Another
possibility is that the inductive effect of the substitutent tends to
polarize the radical center and retard the revertible hydrogen transfer.
The l,4-dibenzoylbutane behaves anomalously as can be seen by aII and
¢dis' The low value of 0.43 for aII in alcohol indicates that another
mode of decay of the triplet is occurring with roughly equal efficiency
as the hydrogen abstraction. The cause of this is not understood, how-
ever, it is possible that an energy transfer through space occurs to the
other benzoyl group with reduction in the efficiency of y-hydrogen ab-

straction.

f. Competitive triplet deactivation. Two ketones with very small
-¢II'S deserve to be considered separately. y-Dimethylaminobutyrophenone
(aII = 0.025) and B-phenylbutyrophenone (aII = 0.002) both were quenched
only slightly with diene and thus assumed to have quite reactive y-hydro-
gens. In a separate study72 evidence was found to support a charge-
transfer intermediate which competes with y-hydrogen abstraction for the
triplet by a 20:1 ratio (Equation 44). The l0-fold increase in aII when
y-dimethylaminobutyrophenone was photolyzed in methanol may in part be
attributed to hydrogen bonding of the solvent to the nitrogen making the
charge transfer more difficult. The B-phenylbutyrophenone is highly

57

Q N(CH3)_2_—» C» p “5103132”—+ ()0 N(CH
' H2 H2
35.-3 2...»: 2.7.4;

deactivated with a aII in alcohol of only 0.0026. The overwhelming ma-

3)2

(Eq. 44)

jority of triplet decays in preference to abstracting a y-hydrogen.
Kelso9“ has found that B,y-diphenylbutyrophenone has a aII in benzene of
0.ll and a aII(alcohol) of 0.l9. These are low compared to y-phenylbu-
tyrophenone which has a aII in benzene of 0.50 and a aII(alcohol) of
0.90. An excimer-complex with overlap of the keto- and B-phenyl rings

is cited as a possible mode of triplet decay in this case.

2. Substituent Effects on the y-Position.

a. Variations with C-H bond strength. One of the first results
obtained was the confirmation of the predicted order of reactivity of the
y-hydrogen towards abstraction by the carbonyl triplet. It was found to
parallel the already well established series according to the strength of
the C-H bond being broken95. Table VI compares the relative reactivities
for the different types of C-H bonds tested in this investigation with
previous work using other hydrogen abstracting species. The ratios shown
for the reactivity towards benzophenone triplet were calculated from val-
ues determined using toluene as the standard (=l). ,Also, some difficulty
was reported in obtaining a reliable tertiary to primary reactivity ratio.
If the ratio, (16:8), for two types of secondary hydrogens towards inter-
molecular hydrogen abstraction at 40°C are good values (there is no rea-

son to assume they are not), then a similar ratio applied to the case of

58

TABLE VI. Comparison of Relative Reactivities of Carbonyl Triplet

Species and tert-Butoxy Radicals.

 

 

 

 

 

 

 

Triplet Species (22°C) tert-Butoxp Radicals per-H
Phenyl Ketone Benzo- Intermolec~ Intra-
phenonea ularb molecularc

Bond kr f1107 Per Per

Type sec Mole Hyd. per Hyd. (40°C)(20°C) (40°C)

1° C-H 0.83d 1 1 1 ' 1 1 1

2° C-H 12.2 14 21 --- 8 9 9

(2° C-H) --- -- -- 50e 16e -- --

3° C-H 49 56 168 300 44 53 47

a

b

C

d

e

Reference 34. Hydrogen donors were 2,3-dimethylbutane and n-butane
except for (e) which was cyclohexane.

Reference 29. Same conditions as in Reference 34.

Reference 32. Long chain tert-butoxy radicals were used for intra-
molecular hydrogen abstraction.

This kr obtained for butyrophenone may be slightly high. Earlier val-
ues obtained were 0.75 x 107 sec-1 by Wagner71 and 0.67 x l07 sec-1

by Pittsloo.

Values are for cyclohexane, see Reference 29.

 

benzophenone would give a relative reactivity ratio of l:25 for the pri-

mary to penultimate secondary hydrogen. The analogy is not perfect but

it is good. The relative reactivities found for the type-II intramolecu-

lar hydrogen abstraction are in the same direction as those reported for

59

intra- and intermolecular hydrogen abstraction by the tert-butoxy radi-
cal, and they exhibit an increasing selectivity of 3° > 2° > l° as was
reported for the case of benzophenone triplet hydrogen abstraction. One
of the underlying reasons for the differences in reactivity of the vari-
ous types of hydrogen can be looked at as due to the inductive effect of
the methyl groups on the C-H bond undergoing attack by an electron de-

ficient species. When substitution is made at a position one carbon re-

@” \ H @c/l: ”wa @3061“ ”\EHECH
___ / ' +CH3 \ ___ / 3
k. .101. 11%.?“ f1. {1.

Increasing reactivity

 

v

moved from the carbon undergoing hydrogen abstraction the effect is di-
minished. A graphic comparison of the qu values found for a number of
phenyl alkyl ketones is shown in Figure 7. The Stern-Volmer sl0pes form
distinguishable groups for the ketones with primary hydrogens [A], [B],
and [C]; secondary hydrogens [0] through [J]; and tertiary hydrogen
[M]. The effects of B-methyl groups on the kr for primary hydrogens can
be seen by the ratios l : 2.4 : 7.l for butyrophenone, isovalerophenone,
and B,B-dimethylbutyrophenone respectively. When statistical correction
is made for the number of hydrogens the ratios are l : l.2 : 2.4. The
analogous case for secondary hydrogens would be valerophenone, hexano-
phenone (l 6-CH3), and 6-methylhexanophenone (2 6-CH3's) for which the
relative reactivity ratios are l : l.06 : l.5. Although the difference
is small it does appear that the inductive effect influences the less
reactive primary hydrogens to a greater extent than it does the second-

ary hydrogens. The changes in kr on increasing alkyl substitution are

60

 

 

 

 

 

 

A B
3.4‘ C
o
E
2.8 F
H
' .1
2.2‘
1.
<1
' K
L
l.6‘ M
1.0
b i '2 5 '4 B '6

Quencher concentration, molar [x l0'2]

Figure 7. Stern-Volmer Quenching Slopes of Alkyl Phenyl Ketones.

[A] But rophenone, [B] Isovalerophenone, [CE 3,3-Dimethylbutyrophen-
one, [D Valerophenone, [E] Hexanophenone, F] Nonanophenone, [G]
Pentadecanophenone, [H] o-Methylhexanophenone, [J] 6,6-Dimethylhex-
anophenone, [K] y-Phenylbutyrophenone, [L] y-Vinylbutyrophenone,

and [M] y-Methylvalerophenone.

indeed rather small for the secondary hydrogens but a trend is apparent
in the reactivities which qualitatively is the order of inductive ability
of the alkyl groups95: CH

-CH3 < -CH2CH3 8 l8

3

I
< -C H < -C14H30 < -CH-CH < - -

61

No such comparison is available for the more reactive tertiary hydrogen
[M] but any effect of substitution on the adjacent carbons would probab-
ly be small. Its reactivity is seen to be quite similar to hydrogens
next to a radical stabilizing group [K] or [L]. Radical stabilization
at the tertiary center most likely plays some role in the increased re-

activity of the tertiary hydrogen.

b. Nonalkyl side chain substituents. When unsaturated groups, he-
teroatoms, or groups containing heteroatoms are substituted at the y-pOS-
ition the effects are more complex. Some of the substituents affect the
reactivity of the hydrogen being abstracted by both inductive and radical
stabilizing effects. Work by Walling and co-workers30s32.97 with tert-
butoxy radicals shows the influence of a heteroatom substituent on the
relative amounts of hydrogen abstraction as one progresses down the car-
bon chain from the substituent (Table VII). Substituent effects on the
relative reactivities towards abstraction by benzophenone triplet have
been looked at for a number of hydrogen donor compounds of which only
one, benzyl hydrogen, can be compared to the present work. The results
vary considerably. Padwa found a ratio of 4.6 : l for the relative re-
activity per hydrogen for the a-hydrogens of ethylbenzene compared to
toluene93. When this is compared to Walling's data3” the relative re-
activity per hydrogen for secondary a-benzyl hydrogen is 460 times larger
than for primary alkyl hydrogen. The two sets of data are not consis-
tent, however, as Padwa found a relative reactivity per hydrogen of ter-
tiary hydrogen to toluene of 1.26 : l, while Walling obtained a 3 : l
ratio. Padwa98 also has shown that the triplet states of propiophenone

and acetophenone are similar to benzophenone triplet in their reactivity

62

TABLE VII. Substituent Effects on Hydrogen Abstraction by tert-Butoxy

 

 

 

Radicals.
Relative Reactivity per Hydrogena at 40°C
(Carbons numbered from substituent)
Substituent (H-donor) Ci'“ C2"“ c3"” C4"“
Chlorob (cu3cnzcnzcuzc1) 5.1 4.6 10.2 2.4d
Cyanob (CHBCHZCHZCN) 0.67 1.3 0.67d

C o
Phenyl ,o c (CH3CH2<::>» 45

. c _

Vinyl (CH3CHZCH-CH2) 6l
Alkoxyc,0°C (CHBCHZOEt) 78
Alkylc, 0°C (CHBCHz-R) 13

 

a Compared to the primary hydrogens of 2,3-dimethylbutane.
b See Reference 30.
c See Reference 97. Ratios at 0°C are slightly higher than at 40°C.

d These values only are for the primary hydrogens at the end of the

chain, other results are for secondary hydrogens.

towards a number of types of hydrogen98. Table VIII contains analogous
values found for the type-II intramolecular hydrogen abstraction. When
the effects on the reactivities by the first six substituents are compar-
ed to those found for hydrogen abstraction by tert-butoxy radicals in
Table VII the similarities are striking. The ratios for the chloro and
cyano substituents are nearly equal for the two methods, and for the
other four substituents they are larger by a factor of about l.4 for the
type-II process. Also, the relative reactivities progressing down the

side chain are almost identical for the cases where data are available.

63

TABLE VIII. Effects on the Type-II Intramolecular Hydrogen Abstraction
of Side-Chain Substituents on Alkyl Phenyl Ketones.

Relative Reactivity per Hydrogen at 22°C
(Carbons numbered from substituent)

 

Substituent 01-H C2-H C3-H C5'H
[Side-chain position] _[1]__ _[p]__ _[§]__ ______
Chloro 4.8 3.8 9.7

Cyano 0.55 1.65 [9.7]a

Phenyl 65

Vinyl 75

Methoxy 104

Alkyl 21 22 --- 27
Carbomethoxy 1.7 6.5

Carboxy --- 4.4

Benzoyl --- 15.7

Hydroxy 66

Dimethylamino 1420

 

a This position cannot be compared to the tert-butoxy radical abstrac-
tion for which butyronitrile having 03 primary hydrogens was used.

These results strongly suggest that the mechanism for the type-II photo-
chamical reaction involves a hydrogen abstraction by a species similar

to tert-butoxy radicals. It is also apparent that this step is influenc-
ed by inductive and radical stabilizing effects. The remainder of Table
VIII lists the effects of substituents for which no data was found for
comparison. An interesting observation is that the type-II process can

discern between the subtle differences of a carboxylic acid and a

54

carboxylate ester. It should also be pointed out that all of the com-
pounds in Tables VII and VIII involve secondary hydrogen (except where
indicated) which in part accounts for the good comparisons. Existing
data for radical abstraction of primary hydrogen which was sought for

comparison is poor and incomplete.

0f major concern in the work with tert-butoxy radicals and benzo-
phenone triplets were the difficulties in obtaining reliable relative
reactivities. Since they had to be determined by product yields several
inherent difficulties arose. The product yields depend on two steps in
the cited cases: (1) Abstraction of hydrogen from donor compound by the
tert-butoxy radical or benzophenone triplet, and (2) abstraction of a
chlorine from a chlorine donor by the alkyl radical to give product.
Although yields were high the process for benzophenone was shown not to
be quantitative and in some cases significant multiple substitution was
noted3”. This causes problems in determining large ratios accurately
where minute amounts of one product is formed. The reaction with tert-
alkoxy radicals is complicated by B-scission of the radicals31 and also
by secondary reactions with the products97. In comparison the type-II
process is relatively clean with very little secondary reaction. Being
an intramolecular process many of the variables are removed and the en-

tire system can be kept constant except for the substituent.

c. Comments on relative selectivities. The notion of selectivity
of a radical towards abstraction of hydrogen has been rationalized by

invoking their relative reactivities95, thus for the halogens:

65

Increasing selectivity-+
F' Cl‘ Br‘ I’
+- Increasing reactivity
In comparing the present work with that of tert-butoxy radicals and ben-

zophenone triplets, the following order is indicated for selectivity:

Qualitatively this should be the inverse order of their reactivities.

d. Quantitative relationships. Some correlations between photo-
chemical reactivities and substituent effects have been attempted. A
respectable Hammett plot was obtained for the relative reactivities of
substituted toluenes towards benzophenone triplet3“. A similar attempt
with the type-II photoreaction of para-substituted butyrophenones was
not as successful because the 0+ values were plotted against all rather
than the reactivity77. It is obvious from the kr values in Table I that
the y-position is affected by both inductive and radical stabilizing ef-
fects and that no reasonable correlations are possible. In order to
eliminate the effect of radical stabilization the 6-substituted ketones
were measured. A Hammett plot of the log of the relative reactivities
(using kr x 10'7) versus the °I values99 for the substituents one methyl-
ene group away from the reaction center is shown in Figure 8. In obtain-
ing the best slope for this plot the corrected kr values for a-chloro-
and a-cyanovalerophenone were used (Table V). The correction improves
the correlation for 6-chlorovalerophenone but the value for 6-cyanoval-

erophenone is shifted slightly from the best line. The reaction constant

66

 

 

 

 

 

0.9 .
[\l—W -
l

2
x
x3.

._. 0.6‘
In
.2
.t’
o;

:3 .
U
E

a, 0.3

.2
4.:
f6
75
a:
U!
3

0 .

-0.1‘

-01 0 01 0.3 0 5 0 7
“’1

Figure 8. Hammett Plot for 6- and e-Substituted Alkyl
Phenyl Ketones. Corrected kr values used f0 5-chloro
and 6-cyan0 ketones. CD= 6-substituents; é;s= e-sub-
stituents.

(p) found from the slope is -2 for the 6-carbon. This relationship was
further tested by measuring the kr for e-chloro- and s-cyanohexanophen-
one to see whether they would correlate with a °I calculated by the fol-

lowing equation (Equation 45) for an additional interposed methylene

67

oI(X-CH2-) = 0.4501(X-)99 (Eq. 45)

group. Excellent agreement was found for the e-chlorohexanophenone and
a fairly good one in the case of e-cyanohexanophenone considering the
known direction of the error (See Results Section). Equation 45 can
also be used to calculate p for the y-position since the effective 01
would be l/0.45 times greater, or conversely, the p would be larger by
the same factor if 0 were held constant. The p for the y-position is
calculated to be -4.4, a large negative value, indicating it is very sen-
sitive to substitution and the reaction is enhanced by electron donating
substituents. Knowledge of p for the y-position now allows a calcula-
tion of the expected reactivity for each of the y-substituents. It
should be possible by comparing the expected and experimental reactivi-
ties to determine the contribution of radical stabilization to the over-
all reactivity. This data is compared in Table IX. The data admittedly
could stand a bit of polish, however, the results are still highly in-
formative. Compared to a methyl group the carboxylate ester group shows
little additional stabilization of the radical and the phenyl and cyano
groups are about equal in their effects. A surprising factor is the ap-
parent high stabilizing ability of the substituents with nonbonding elec-

trons. The order, in fact, seems to be that of their basicities:

-0CH > -OH > -C1

3

The y-dimethylamino group is omitted from the table because of uncertain-
ty over the actual kr value. Although aII for y-dimethylaminobutyr0phen-
one has been determined to be 0.25 in methanol (Table III), it is felt

that hydrogen bonding by the solvent to the amino group may be a

68

TABLE IX. Comparison of Experimental Reactivities to Calculated Reac-

tivities from “I and Experimental p Value.

b Relativec Apparent

a Relative Experimental Stabilization
y:Substituent 0I O” Reactivity Reactivity, Factor

 

 

Methyl 0.00 0.00 1.00 1.00 1.00
Phenyl 0.10 -0.44 0.36 3.1 8.6
Methoxy 0.25 -l.10 0.080 5.00 62
Hydroxy 0.25 -1.10 0.080 3.16 40
Carbomethoxy 0.30 -l.32 0.048 0.081 1.7
Chloro 0.47 -2.07 0.0085 0.23 27
Cyano 0.56 -2.46 0.0035 0.031 8.9

 

a up = log(Expected Relative Reactivity); p= -4.4.
b Due to inductive effect only, found from op.

c Relative to valerophenone taken as 1.

significant factor. The complications introduced by this occurrence are
(1) a change in the inductive effect of the dimethylamino group, (2) a
decrease in the ability of the triplet to decay via charge transfer, and
(3) a probable decrease in the radical stabilizing ability of the di-
methylamino group. The degree of influence of each of these three fac-
tors is not known. If the comparison in Table IX is applied to the di-
methylamino group (aI = 0.10) using the value of aII(a1cohol) = 0.25 in
Equation 37, then the "apparent stabilization factor" is 47. This value
is in the vicinity of those of other substituents with nonbonding elec-

trons .

69

An unresolved problem also exists for the y-chlorobutyrophenone as
the aII can account for less than one third of the ¢dis‘ Whether this
has an effect on the apparent kr is not yet known but it should be re-
called that its behavior is quite similar to that of hydrogen abstraction
by tertiary alkoxy radicals. The “I for the benzoyl group is not common-
ly found in reference texts, however, that for the acetyl group is listed
as 0.2899. Using Equation 37 the kr for 1,4-dibenzoy1butane is found to
be 4.5 x 107 sec". Finding the antilog of this value on the Hammett
plot in Figure 8 yields a “I for the benzoyl group of 0.26. It would a1-
so be possible using this method to determine the inductive effect of a
double bond by making 5-viny1valerophenone and measuring its relative
reactivity. This quantitative treatment has provided, probably for the

first time, a means of separating out the inductive and radical stabiliz-

ing effects on hydrogen abstraction.

3. Sppport for a Biradical Mechanism.

Probably the strongest argument for the intermediacy of a 1,4-bi-
radical in the type-II photochemical process has been the lack of evi-
dence to the contrary. The current work by Kelso7° with the optically
active ketone provides the best direct evidence since racemization of the
optically active y-carbon in recovered parent ketone can only be explain-

ed by the occurrence of an sp2

center at this position. The present work
provides strong evidence for reinforcement of this mechanistic route.
First it can be pointed out that the substituent effects on the relative
reactivities of the type-II photoreaction closely resemble those for hy-
drogen by alkoxy radicals. In order for the type-II reactivities to ex-

hibit such similar behavior the step involved must also be one of

70

1 3
K1—'* K1

\k k
kd [BR] -JL———+ Products
hv 1. Type-II

2. Cyclobutanols

1K0/ "‘ 3. Other?

Figure 9. Schematic Representation of the Mechanism of the Photochemi-

cal Processes in the Type-II Photoreaction.

hydrogen abstraction leading to a biradical. This comparison was made
for several substituent groups of alkyl and heteroatom make-up and the
relative reactivities were always found to be in good agreement (See
Tables VI and VII). A second indication supporting a biradical inter-
mediate, as was pointed out by Wagner59, is that as a hydrogen bonding
cosolvent is added in increasing increments the initial effect is to in-
crease all. The opportunity for hydrogen bonding to solvent impedes the
hydrogen from reverting to the radical site and effectively increases
the radical lifetime. This results in larger aII's which is found ex-
perimentally for a number of ketones studied (Table III). A third rea-
son, which often tends to be underestimated, is that including the birad-
ical step into the mechanism allows a logical explanation of the experi-
mental observations that quantum yields and reactivities do not corre-
late. From the schematic representation of the mechanism in Figure 9
several important relationships can be traced. The quantum yield of bi-
radical formation from the triplet is dependent on the relative values
of kr and kd. Thus kr may vary considerably yet aBR could be large pro-

viding kd is small in comparison to kr' It can also be seen that

71

observed quantum yields from the biradical will depend on relative values
of kp and k-r’ so even highly reactive ketones could have low quantum

yields. The relative amounts of aII and a may also vary and there may

c c
be competition from other processes such asya-hydrogen abstraction,
coupling,_photoreduction, etc. If the biradical step were to be omitted
from the scheme all changes in kr would have to be accompanied by corre-
sponding changes in kd to account for the quantum yields. This explana-
tion is less than satisfactory as it would require large variations in
kd for ketones with the same phenyl ketone chromophore structure. From

the large number of ketones studied the observed behavior best fits a

biradical mechanism.

4. Additional Solvent Effects.‘

a. Changes in k Besides the effect on aII which was discussed

‘1'.
in a previous sectionq6ther differences are noted in Table III. The
measured slope changes with solvent, an effect which can usually be at-
tributed to changes in kq39. Comparison to a more commonly used ketone
such as valerophenone is helpful when interpreting data for those ketones
which for solubility problems must be run in methanol. Table X lists the
qu's for valerophenone, y-dimethylaminobutyrophenone, and y-dimethylami-
nobutyrophenone hydrochloride in three solvents. The qu for valerophen-
one goes up in methanol by a factor of 2.5, but for y-dimethylaminobutyr-
ophenone it increases about 7.5 times. The increases in acetonitrile,
about 1.7 and 2.5 times respectively, are not quite as far apart. An ex-
planation for the result in methanol is that the reactivity of the y-hy-
drogens of y-dimethylaminobutyrophenone is being affected by hydrogen

bonding of solvent to the amine nitrogen. By converting the

72

TABLE X. Effect of Solvents on k T.

 

 

9
k T in k T in qu in all in
Ketone Benzene Acetonitrile Methanol Methanol
Valerophenone 41 68 100 1.00
y-Dimethylaminobutyrophenone 0.6 1.5 4.5 0.25
y-Dimethylaminobutyrophenone
Hydrochloride --- --- 720 0.009

 

y-dimethylaminobutyrophenone to the hydrochloride salt the y-hydrogens
are strongly deactivated as evidenced by a quenching slope of 720 with a
aII(alcohol) of 0.009. For y-dimethylaminobutyrophenone hydrochloride
both disappearance and type-II quantum yields show that the process is
extremely inefficient. y-Dimethylaminobutyrophenone ethyl bromide was
even less efficient with a aII of 0.0004 in methanol. The reason for
such low efficiency is not yet understood, especially when considering
that the opportunity for charge-transfer is eliminated (Equation 34).
In control experiments neither 0.08M KBr nor 0.08M (CH3)3NCZH5Br had any
effect on the quantum yield of valerophenone in methanol. Neither should
the size of the y-substituents produce any serious steric problems. A
possibility to account for the low aII's could be that establishing a
full positive charge next to the y-carbon introduces serious repulsive
interactions with the triplet carbonyl, or that deactivation in these
cases is so strong that other means of triplet decay overwhelm y-hydro-
gen abstraction. Also, since photoreduction appears to occur in prefer-
ence to y-hydrogen abstraction in these cases, it is possible that small
amounts of quenching products in the methanol may have a devastating ef-

fect on qu and all. A more careful look at the behavior of these

73

ketones is needed, however, the results do give an indication of the

deactivating properties of the quaternary amine group.

An interesting result was obtained when the benzene solvent was mod-
ified by making it 0.60 molar with pyridine. The aII increased to 0.80,
which is as high or higher than with equimolar tert-butyl alcohol, but
qu was unaffected (Table III). This is in good agreement with the
proposal69 that the inefficiency in the type-II process in benzene is

due to the biradical.

b. An estimation of biradical lifetime. It was previously mention-
ed that knowedge of the biradical lifetime in solution would be of help
in explaining some of the photochemical behavior of the phenyl alkyl ke-
tones. 0ne obvious benefit in having a reliable measure of biradical
lifetime would be that of determining whether anomolous behavior such as
extremely low quantum yields due to the biradical. Perhaps a relative
measure of this lifetime can be acheived by a kinetic treatment of the
dependence of aII on small increments of alcohol added to the benzene
solution. The biradical can be visualized as being affected as in Fig-

ure 10. In order to adapt the data to a SterneVolmer type diagram each

 

 

H OH
. . _ p ,+ I _
[@{Hfiagwa 0a."; £21013
k _r kh(ROH) 0H. .3 R /
{CD-ct”: $121— -CH3 [@4th 2_dfiH- -CH3]

Figure 10. Hydrogen Bonding to the Biradical Intermediate in Alcohol

Solvents.

74

 

 

 

 

 

0 l 2 3 4 5
Concentration of tert-butyl alcohol, molar

Figure 11. Stern-Volmer Type Treatment of Increasing
guantum Yields upon Addition of Alcohol Cosolvent.
A] Valerophenone, [B] Isovalerophenone, [C] Hexanophenone,
D] y-Methylvalerophenone, [E] y-Hydroxybutyrophenone, and
F] 3,3-Dimethylbutyrophenone.

side of the equation in the normal treatment is subtracted from one.

k k
1-410 1 - k—L— = —k——+£k—- (Eq. 46)
p -Y‘ p -T‘
kP + kh(ROH)

1-a = 1 - = " (Eq. 47)
A kp + kh(ROH) + k_r kp + kh(ROH) + k_r

 

 

Then dividing l-ae by l-aA the expression becomes:

kh(ROH)

1:22. = ________ =
1_¢A l + kp + k-r 1 + kh(ROH)B (Eq. 48)

75

TABLE XI. Comparison of Initial Slopes [khB] with Type-II Quantum

 

 

 

Yields.

Ketone ::;t1a] 3:115ivea ¢II

Valerophenone 1.6 6.4 0.33
Isovalerophenone 1.45 5.8 0.36
Hexanophenone 1.15 4.6 0.30
y-Methylvalerophenone 0.75 3.0 0.25
3,B-Dimethylbutyrophenone 0.25 1.0 0.19
y-Hydroxybutyrophenone 0.33 1.3 0.35

 

a The smallest khs was merely taken as l for an easier comparison.

Here 3 = l/(kp + k-r) which is proportional to the average biradical
lifetime in solution, and kh would be the effective rate of hydrogen

bond formation. By plotting Equation 48 a value for kha is found from
the initial slope. As long as all other factors remain constant khB
should not change very much and the slope would represent a relative
average lifetime of the biradical. In Figure 11 data for several ketones
were plotted in this manner and the results are summarized in Table XI.
With the exception of y-hydroxybutyrophenone the trend of longer relative
biradical lifetimes to larger aII is fair. In the case of y-hydroxybu-
tyrophenone it is possible that intramolecular hydrogen bonding inter-
feres with kh and alters the kbs. The data could be improved by using
smaller alcohol concentrations to determine the initial slope. This

would give a more linear and accurate relationship of aII to increasing

76

alcohol concentration.

5. Effects of RingSubstituents on Triplet Reactivities.

A major area of investigation was the effect of ring substituents
on the reactivity of the triplet towards y-hydrogen abstraction. The be-
havior was first investigated for the methoxy and methyl groups, both
electron donating in nature to the phenyl ring. As seen in Table II the
1/T decreases considerably for these compounds indicating that the reac-
tivity has decreased. The effect is in the right direction for the con-
cept of an electron deficient reactive n,h* triplet being deactivated by

electron donating ring substituents (Equation 49). It was predicted that

03+
// 5 I?
X C C - +—-——+ -C
[ C ‘CHz-éng CH3] [x C %”2'C{':2 H3J(Eq. 49)

electron withdrawing groups on the ring would increase the reactivity of
the triplet. This was verified for the trifluoromethylvalerophenones,
especially for the meta and para isomers (Table II). These results para-
llel those found by Wagner and Capen73 for the n-butyl pyridyl ketones in
which the n,h* triplet is acitvated by the inductive effect of the nitro-
gen in the ring. A closer examination of the data for the ketones mea-
sured in Table 11 indicates that a simple inductive effect does not ac-

count for the observations.

a. Change in the nature of the triplet. If the 1/t's for the meth-
oxy and methyl substituted valerophenones are compared to those for the
trifluoromethylvalerophenones and valerophenone, it appears that the val-
ues in the former cases are much too large. A good indication that the

inductive effect does not account entirely for their behavior can be seen

77

 

 

 

 

1.5 .. ,6m-CF3
/
SD p-CF3
/
/
/
// @m-F
/
1.2 1 / Gill-C1
/
/
/®'H
/
/
,.|. /
/
§ 0.9 .
N
12 1
X
a.-
>
31 0.6 « Gm-CH3
_J
0.3 - o
p-CH3
-0.5 -0.3 -011 0 0.1 013 0.5
[om 0P op]

-1

Eigg[g_1§: Log of the 1/T x 10'7 sec of Ring Substi-

tuted Phenyl Ketones versus 0m or °p'

1'1 Figure 12 in which am or o are plotted against the log of the 1/T

P
values for several substituents. A poor relationship exists between the
various substituents. If the meta- and para-trifluoromethyl substituents
are assumed to have only an inductive effect and their log 1/T x 10'7

va7lJes are taken to be on a line with that of valerophenone, then the

78

values for the meta- and para-methyl substituents plotted against am and
op fall far below the line (Figure 12). Meta- and para-methoxy substi-
tuents have an even larger deviation. This behavior is explained by the
rearrangement of triplet energy 1evels“°,75977 under the influence of the
substituent so that the nonreactive h,h* triplet is lower in energy than
the n,n* triplet. The photoreaction which then occurs may do so from an
equilibrium concentration of n,«* triplet. According to Yang's”,81 re-
sults it would be expected that para-methoxye, para-methyl-, and possibly
para-chloro-, and para-fluorovalerophenone would have h,n* lowest energy
triplets. The analogous meta and ortho substituted ketones may also have
u,n* lowest triplet states, however, less is known about these compounds
and much of their character must be inferred from comparison to the para
substituted ketones. From the data in Table II it is seen that para-me-
thoxy-, para-methy1-, and para-chlorovalerophenones are strongly deacti-
vated, the behavior associated with a h,n* triplet. The fluoro substi-
tuents are interesting in that they have almost no effect on the reacti-
vity or the ¢II’ Pitts and co-workers100 reported the same effect on
fluoro substituted butyrophenones. There seems to be only a small induc-
tive effect operating with very little, if any, effect on the nature of
the triplet. This may be characteristic of the nature of fluorine sub-
stituents on an unsaturated system; to influence the sigma bonds of the
molecule without affecting the h-systemlol. The high reactivity of the
ring substituted fluoro-valerophenones indicates that the lowest triplet

state must be n,«* in nature.

b. The effects of meta substituents. An interesting observation
of the 1/1 values is that all of the meta substituted ketones have great-

er reactivities, or alternatively, are not deactivated as much as the

79

TABLE XII. Summary of 1/1 [x 107] see”1 from Table II for Ring Substi-

tuents on Valerophenone.

 

Substituent 9r_thg_ meL m__
Trifluoromethyl 13.2 32.2 28
Fluoro 14.4 18 14.7
Chloro 3.5 15.6 3.7
Methyl --- 3.9 1.84
Methoxy --- (1.6)a 0.22

Methoxy; Y-Methyl -- (2.5)a 0.58

Valerophenone = 12.2.

 

a Estimated values, See Table II.

corresponding para or ortho substituted ketones. Two possible reasons
can be given for this: (1) A more effective inductive ability from the
meta position, or (2) a decreased ability of a meta substituent to effect
an interchange in lowest triplet levels. Both effects may operate to
varying degrees. Also, the ortho substituted ketones may not be accomo-
dated by this explanation. A qualitative explanation of the data in
Table XII is as follows: For the trifluoromethyl and fluoro substituents
the triplet energy levels are relatively unaffected and the substituents
activate the excited state, the trifluoromethyl groups more so than the
fluoro and the meta substituents slightly more than the para. The ortho
trifluoromethyl group appears to be anamolous in this situation as it
behaves as if it were less activating than its para isomer by a factor of

two. For the chloro ketones the meta substituent exerts a small

80

activating inductive effect but the ortho and para isomers appear to be
deactivating the triplet significantly, possibly causing a shift in the
nature of the triplet. The meta- and para-methyl groups even further de-
activate the triplet, the para more so than the meta, and the meta- and
para-methoxy substituents greatly deactivate the triplet towards y-hydro-
gen abstraction. This is probably due to the nature of the triplet in
the latter cases as para-methyl and para-methoxy substituents are believ-

ed to cause the lowest triplet to shift to fl,fl*78.

c. Qualification of "reactivity". The behavior of the ring substi-
tuted ketones has been compared using 1/r as a measure of reactivity.
Data needed for determining the actual values of kr are lacking for most
of the ketones in Table II. Wagner and Schott7s have found that for
para-methoxy- and para-chlorovalerophenone and para-methoxy-y-methylval-
erophenone the adis's in alcohol are 0.26, 0.80, and 0.67 respectively.
Para-methylvalerophenone was found to have a aII of 0.88 in methanol in
which case kr can be approximated by 1/T. For the others kr must be cal-
culated using Equation 37. An inspection of the aII's in Table 11 re-
veals that except for the methoxy substituents not much change occurs for
those ketones considered so far. The trifluoromethyl ketones have aII's
which are lower by about 20-40%, and the ortho-chlorovalerophenone has a
aII about 30% higher, but the rest vary only slightly from valerophenone.
If the assumption is made that the probability for cleavage of the alkyl
chain from the biradical does not change much, the implication is that
the ratio of reverse hydrogen transfer also stays about the same when the
ring is substituted. If this were the case the l/r values for trifluoro-

methyl, fluoro, chloro, and methyl substituted valerophenones would be

Bl

_ good measures of their kr‘s. For para- and meta-methoxy substituents,
however, all's of 0.l3 and 0.0l3 indicate'that other processes must be
competing with y-hydrogen abstraction, especially for the meta substi-
tuent. Using the assumption just mentioned and the relation for valero-

phenone in Equation 50 from which ¢br can be found to be 0.33, values of

¢II = ¢br¢p = 0.33 (Eq. 50)

¢II(alcohol) = l = ¢br (Eq. 51)

kr and kd for meta- and paramethoxyvalerophenone can be estimated.

4’11

kr
= ——X°-333 " 673—37:

¢ (Eq. 52)
11 kr + kd r

Likewise, since y-methylvalerophenone has a ¢II of 0.25 and a ¢II(alco-
hol) of 0.87, the relationship for meta- and para-methoxy-y-methylvalero-

phenone would be:

¢
- r I - II
¢II ' kr + de 0'25 ’ kr ' 0.25 x 1 (Eq. 53)

Using the values found for r, where l/r = kr + kd, and substituting into
Equations 52 and 53 gives the results shown in Table XIII. It is inte-
resting that for those ketones measured the ¢br found this way is some-
what higher in value, but not drastically different, from ¢BR measured
from ¢II in alcohol solvents. These results probably give a significant
comparison of kd's for meta- and para-methoxy ketones, which in the case
of the meta-methoxy substituents are an order of magnitude larger than
for the para-methoxy ketones. The kr's for the meta- and para-methoxy
substituents are seen to slightly different in each case. These results

must of course be treated cautiously as the assumption made for ap may

82

TABLE XIII. Estimation of kr and kd for Several Ring Substituted Alkyl

Phenyl Ketones.

 

 

 

. kr kd
Sglzfiogflgggne £3b53_ d’II(a]C) [x107jlsec"1 [x107]sec']
para-Methyl l.2 0.88 l.84 ---
meta-Methoxy 0.039 --- 0.062 l.538
para-Methoxy 0.39 0.26 0.086 0.l34
meta-Methoxy; y-Methyl 0.l2 --- 0.30 2.2
para-Methoxy; y-Methyl 0.76 0.67 0.44 0.l4

 

not always be true (as can be seen for several ketones in Table II
with aII's larger than 0.33). Also, abr was determined from the aII
values which makes no allowance for cyclobutanols. Results from solvent

studies would be a great help in interpreting the data.

d. Powerful deactivating,substituents. Three of the ring substi-
tuted ketones in Table II, para-thiomethoxy- and para-hydroxyvalerophen-
one and para-phenyl-y-methylvalerophenone were found to be extremely un-
reactive. The ¢II and ¢dis for para-thiomethoxyvalerophenone were both
found to be zero indicating that no triplet is reacting. The possibility
that the methyl phenyl sulfide is quenching the reaction can be eliminat-
ed on the basis of investigations which show that phenyl ketones with a
sulfur in the alkyl chain readily indergo the type-II photoreactionloz.
Also, alkyl sulfides have been found to be inefficient quencher5103. In
previous examples where the quantum yields were on the order of 0.0l or

less low aBR's were found indicating that kr was very low or that kd was

83

larger than usual. This must also be the situation for the three ke-
tones in question. The very low aII's indicate that the triplet states
are n,n* in nature. In terms of an equilibrium concentration the equi-
librium would be displaced far towards the n,n* triplet in these cases.
Good supporting evidence for a n,n* triplet in para-phenyl-y-methyl- and
para-thiomethoxyvalerophenone also comes from phosphorescence studies71

which show the characteristic long lifetimes of the n,n* triplets79.

These results could also be consistent with Porter's76 concept of a
charge-transfer triplet. The quantum yields for these ketones are sum—
marized in Table XIV. There is a certain logic to the argument that a
correlation exists between the reactivity and the ability to stabilize a
charge separation in these ketones. A difficulty may arize in the de-
finition and connotation of the term "charge-transfer." Perhaps the a-
bility to support a charge separation (charge-transfer?) greatly affects
the rate of nonradiative decay81 leading to the low observed quantum

yields.

 

TABLE XIV. Type-II Quantum Yields for Highly Deactivated Phenyl Alkyl

 

 

 

 

Ketones.
Substituted ¢ ¢
Valerophenones II dis Solvent
para-Thiomethoxy 0.00 0.00 benzene
para-Hydroxy 0.002 ---- benzene
para-Methoxy 0.l3 ---- benzene

(Unsubstituted) 0.33 0.43 benzene

84

e. Ortho substituents. The ortho-chloro and ortho-methoxy substi-
tuents on valerophenone are seen to increase the aII's significantly
(Table II). An explanation for this is difficult at this stage. There
is the possibility that intramolecular "solvation" can occur with the
biradical and thereby increase ap (Equation 54). The fact that the or-

tho-chloro- and ortho-methoxyvalerophenone, which may hydrogen bond with

64:” Hz-szw- -CH3 ‘——— *6?“ icfcw -3CH

2 3(Eq. 54)

their nonbonding electrons, have higher aII's than the meta and para
isomers supports this. Ortho-trifluoromethylvalerophenone, which would
not be expected to hydrogen bond, has a lower aII than its meta or para
isomers. Ortho-fluorovalerophenone presents an ambiguous situation

since its aII is smaller than its para but larger than its meta isomer.

The ortho-trifluoromethyl group has already been mentioned as de-
creasing the apparent reactivity of the ketone triplet when it would be
expected to activate it by induction. Wagner and Capen73 found the same
relative effect with the 2-pyridyl butyl ketone. It may be that electro-
static or field effects occur involving the excited carbonyl in these
cases; the nature of these effects is not known. A clue to the impor-
tance of intramolecular salvation of the biradical may be found by study-
ing the relative effects of added increments of a hydrogen bonding cosol-
vent on aII of the ortho, meta, and para isomers of the ketones in ques-
tion. The ortho-methoxyvalerophenone deserves special mention because of
its variable behavior. Attempts to obtain a reliable measure of I were

hampered by large fluctuations in results from seemingly consistent

85

 

4.2 J

 

 

 

 

1

o 1 2 3 _34 5

]

Figure 13. Stern-Volmer Quenching Slopes for ortho-Meth-
oxyvalerophenone. [A] & [C], two similar runs photolyzed
at 3130A; [B] photolyzed at 3660A.

Quencher concentration, molar [x 10

techniques. The Stern-Volmer plots for the quenching runs in three in-
stances are shown in Figure 13. Some of the runs exhibited high aII's
(=0.25) and had a curved or "bent" graph as in line A of Figure l3. In

a kinetic analysis, which is beyond the scope of this work, Wagnerlo“
demonstrated that the case of two reacting excited states, one being more
reactive (less easily quenched) would behave in such a fashion. Yang”5
has also found similar behavior in the quenching of photocycloaddition

reactions and attributes the behavior to two reacting excited states.

86

Wagner's and Hammond's35 original work on quenching of dialkyl ketones
which react from both the singlet and triplet excited states also have
similar plots. The difference in the case of ortho-methoxyvalerophenone
is that both excited states would be triplets. A quenching run was also
performed on ortho-methoxyvalerophenone at 3660A (line B, Figure 13) with
the result that the slope was linear out to a higher quenching ratio.

The indication is that the relative amounts of the two excited states is
sensitive to the energy of the light absorbed. Some of the runs had con-

siderably lower k 1's as in line C, Figure 13, and at times some scatter

q
in the points. It is quite obvious that this ketone is extremely sensi-
tive to other variables, possibly the efficiency of degassing, efficiency
of the filtration of the light source, small amounts of contaminants in
the solvent, etc. Further study with these precautions in mind is war-

rented.

6. Summary

a. Conclusions. From the data presented in this work several con-
clusions of basic importance concerning the type-II photoreaction can be
made.

(1) The mechanism of the type-II photochemical reaction must in-
volve a l,4-biradical intermediate. Evidence for this is seen in the
similarity of the substituent effects on the relative reactivities to-
wards hydrogen abstraction for the phenyl ketone triplets and tert-alkoxy
radicals. Additional support comes from the observed solvent effects on
the quantum yields and the fact that the quantum yields show no correla-
tion to the measured reactivities.

(2) It has been established that the reactivity towards y-hydrogen

87

abstraction correlates with °I substituent constants. Using the tech-
nique of determining the inductive effect of a substituent when removed

by one methylene group from the reaction center, a reaction constant, p,
can be calculated for the y-carbon. This made possible a unique deter-
mination: The separation of the inductive and radical stabilizing effects
of substituents on the reactivity towards y-hydrogen abstraction.

(3) Substituents on the ring of alkyl phenyl ketones affect the
triplet state reactivity in two principal ways, by induction and by al-
tering the nature of the excited triplet. The latter effect is manifest-
ed by changes of much greater magnitude than the former. The large range
of reactivities found for the various ring substituted valerophenones is
consistent with the concept of a thermal equilibrium between n,n* and

n,n* excited triplet states.

b. Significant observations.

(l) The type-II quantum yield (including cyclobutanols) was found
to account quantitatively for the reaction products of alkyl phenyl ke-
tones substituted on the alkyl chain except in cases where the substi-
tuent is a good photoreducing, or otherwise reactive, group.

(2) The steric effects at the y-posttion on ¢II are very small over
a fairly large range of substituent sizes. There is evidence, however,
‘that a moderate effect on aII may occur from the eclipsing repulsions of
:3 and 3 hydrogens or substituents in the planar transition state neces-
sary for type-II cleavage.

(3) The formation of cyclobutanols is very sensitive to substitu-
tents which would sterically interfere on the cyclobutane ring.

(4) Intramolecular salvation of the biradical may account for

88

increased aII‘s in certain alkyl phenyl ketones which are substituted be-
yond the y—carbon or on the ortho ring position with polar substituents.

(5) The aII can be drastically affected by special avenues of trip-
let decay, such as charge-transfer and excimer-complex formation.

(6) The overall utility of the type-II photoreaction and its many
advantages over other techniques makes it a valuable tool in physical or—
ganic chemistry. In the area of measuring substituent effects on reacti-
vity, the type-II process measures the specific rate constant for hydro-
gen abstraction whereas the other methods depend on analyses of product
yields. In determining oI substituent constants the type-II process has
the advantage of its high selectivity, being able to discriminate between
subtle changes in the substituent. The remainder of the substrate does
not change for the various substituents. Also, from the known effects of
hydrogen bonding solvents on the biradical, a method of measuring the
average lifetime of the biradical is proposed. The results of this work
show that the type-II photoreaction can be used in a variety of ways to

obtain basic structure-activity relationships.

7. Indications for Further Research.

At times it would seem that finding the answer to one question would
generate two or three more. Several of the unresolved problems which are
“substantial enough to be an interesting research project are mentioned

here.

a. Photolysis of y—chlorobutyrophenone. The additional photoprod-
uct which was consistently observed in the y-chlorobutyrophenone solu-
tions deserves to be isolated and identified. This would provide infor-

mation on the process competing with the type-II photoreaction. It might

89

 

4.0 ‘

-e-|-e-

l.0 ‘

 

 

 

I V I

o 1' 2 3 _34
Quencher concentration, molar [x 10 ]

Eigugg_lfi: Stern-Volmer Quenching Slopes for

meta-Methoxyvalerophenone. [A] Type-II product,
[B] presumed cyclobutanols.

also be determined whether the triplet or biradical is responsible for

the low all.

b. Studies on the cyclobutanols for meta-methoxyvalerophenone.
When a sample of meta-methoxyvalerophenone (which was very pure so that
no interfering peaks occurred in the VPC analysis) was photolyzed, the
products presumed to be the cyclobutanols were found to have about three
times the quantum yield and were quenched about ten times faster (Figure
l4) than the meta-methoxyacetophenone (Appendix A, Part 2). This behav-

ior was not noted for any of the other ketones, however, the

90

meta-methoxy- -methylvalerophenone which may behave similarly contains
large interfering impurities. It is possible that these products are not

cyclobutanols but some other compounds such as ring adducts (Equation 55).

CH3 0CH3
°©“ .. c" (bf;
‘CH2 -CH2 6:;H 2 -——* H2 (Eq. 55)
(EH-H2 QH-CH2
CH3

In order to test this the photoproducts must be isolated and identified.
The difference in sensitivity implies that the two photoproducts occur
from different excited states. It might be noted that if the excited
state of the ketone has some double bond character between the carbonyl
carbon and the phenyl ring, then two forms of the l,4-biradical can be

drawn. This may be of significance as one form may favor the formation

0:6:C<CHZ fig” CH3 ——5 mgficnz a?” CH3

0CH3 (Eq. 56)

of cyclobutanols and/or be quenched at a different rate.

c. Wavelength studies. The results of the photolysis of ortho-
methoxyvalerophenone indicate that this ketone is sensitive to the wave-
length of the irradiating light. This is interesting because in this
case the reaction may be occurring from two excited triplet states which
are close enough in energy to be affected by changes in the wavelength of
irradiating light. This should be investigated further as it would pro-

vide information on the nature of the reacting excited states.

91

d. The biphenyl ketone. The results with para-phenyl-y-methylval-
erophenone cannot be explained in the same terms as those for the alkyl
phenyl ketones. The aII has been postulated as being extremely low be-
cause of the n,n* nature of the triplet. However, when an attempt was
made to quench the small amount of reaction using 2,5-dimethyl-2,4-hexa-
diene the aII increased. At 0.0] molar quencher the quantum yield was
four times larger. Also, 0.10 molar benzophenone or 0.05 molar triphen-
ylene decreased the quantum yield slightly. When the biphenyl ketone
was photolyzed in 1.0 molar tert-butyl alcohol the aII decreased, and in
8.0 molar tert-butyl alcohol a white precipitate was formed. It should
also be mentioned that when the solutions in the sample tubes were held
next to a mercury lamp which was filtered for 3660A light they seemed to
fluoresce intensely. This may be a clue to its behavior as the triplet
may be decaying rapidly by phosphorescence. Measurements of the phos—

phorescence quantum yield would be helpful in explaining this behavior.

 

e. The qu's from product quenching. The ketones with carbonyl or
unsaturated groups on the y-carbon produce their own quencher during the
type-II process and this required special treatment to arrive at qu (See
Results Section). A kinetic relationship has been worked out by Wagner
and Kelso7°, and independently by Zweig and co-worker5123, which indi-
cates that if the reciprocal of the quantum yield is plotted against the
average quencher concentration (one half the percent conversion minus

the cyclobutanols), the slope divided by the intercept should equal the
qu (Equation 57).

a-a.
[OJ/2

|-'

 

II
x
H

(Eq. 57)

'6-

o q

92

TABLE XV. Comparison of qu Values from ProductQuenching and Diene

 

 

 

Quenching.
Substituted qu qu
Butyrophenone [Product quenching] [Diene quenching]
y-Vinyl 42 ll
y-Phenyl 33 l3
y-Cyano 403 934
y-Carbomethoxy 79 470

 

When this calculation was carried out the agreement with the qu found

by quenching with diene was not very good (Table XV). The most surpris-
ing result is that for y-vinylbutyrophenone, which produces butadiene.

It would be expected to have about the same value as the qu for added
diene quenching. The cases of the other quenching products cannot be
directly compared as the relative effectiveness of quenching compared

to the diene is not known (except for methyl methacrylate which was

found to be about l0% as efficient as the diene)71. The product-quench-
ing for the y-carbomethoxybutyrophenone does show less quenching by a
factor of 5, which is a fair comparison. It may be that a portion of

the quenching is neglected in the common treatment, and that is the trip-
let quenching of the excited acetophenone (product) which may receive
energy by energy transfer or absorption. This would cause a drop in the
aII due to the carbonyl product in addition to that due to the olefin
product. Studies to detect and characterize this effect would be impor-
tant and probably help in the explanation of the discrepancies of product

quenching.

EXPERIMENTAL

PART ONE. CHEMICALS

1. Preparation and Purification of the Phenyl Ketones used in Photo-

lyses.

a. Methods of preparation. Several methods of preparation were

 

used depending on the availability of the starting materials and the de-
sirable characteristics of the reaction. The general methods are listed
below with the ketones prepared by that method. The reagents used are
given with the source in parentheses; special preparative procedures or
conditions are treated in more detail. The physical properties for the
prepared ketones are listed in Table XVI. The structure for each ketone
was verified by its infra-red spectrum, and in most cases also by its
nmr and/or mass spectrum (See Appendix C). The purchased ketones were
purified according to the listed methods, the physical properties and

suppliers are listed in Table XVII.

Method I. The alkyl Grignard1°6 was prepared and benzonitrile or a
substituted benzonitrile was added to it. In a typical preparation 0.10
moles of the alkyl Grignard was prepared from the alkyl bromide and mag-
nesium turnings in dry ether and 0.095 moles of the nitrile was added
dropwise. After a short reflux period the product was worked up by using
approximately three times the stoiciometric amount of hydrochloric acid
over ice cubes to insure the formation of the imine salt. The cold aq-
ueous layer was quickly separated and placed on a steam bath for l t0'2
'hours to hydrolyze the imine salt. The resulting ketone was extracted
with ether and dried over anhydrous magnesium sulfate, the ether then

was removed and the ketone purified. An advantage to this method is

93

94

that the bulk of the organic impurities are eliminated in the first
ether layer after formation of the imine salt. The following ketones

were prepared by this method:

a-Methylhexanophenone. Made from benzonitrile (Eastman Organic
Chemicals) and 4-methyl-l-bromopentane (bp = 95-98°C at l46mm Hg). The
latter was made from 4-methyl-l-pentanol (Chemical Samples Co.) and

phosphorous tribromide.

a,a-Dimethylhexanophenone. Made from benzonitrile and 4.4-dimethyl-
l-bromopentane (bp = lll-113°C at 160mm Hg). The latter was prepared
from 4,4-dimethyl-l-pentene (Chemical Samples Co.) by oxidative borohy-
dration* and reaction with phosphorous tribromide as follows:

BH 'THF PBr

- 3 ‘ 3
(CH3)3CCH2CH-CH2 H 0 0H]? (CH3)BCCH2CH2CH20H -———»-(CH
2 2’

 

3)3C(CH2)3Br

Nonanophenone. Made from benzonitrile and repurified student-pre-
pared octyl bromide. The octyl alcohol (Matheson Coleman & Bell) used
to prepare the octyl bromide contains a significant amount of what ap-

pears to be nonyl alcohol.

IeMethoxybutyrophenone. Made from benzonitrile and 3-methoxy-l-
bromopropane which was prepared from 3-methoxy-l-propanol (Pfaltz and

Bauer).

 

* The author is indebted to Prof. Michael W. Rathke, Michigan State Uni-
versity, for his assistance in the procedure for this step in the
preparation.

95

meta-Methoxyvalerophenone. Made from l-bromobutane (Matheson Cole-
man & Bell) and meta-methoxybenzonitrile which was prepared from meta-
cresol (Aldrich Chemical Co.) by the following well known procedures:
oH 0CH3 0CH3

0 PC]
©c~3° am 2% . @411 ——-»S
(CH: )2 S04 ,H 0

 

 

 

2 (mp = 106°C)
CH CH
0 3 ,/o NH4 0H 3
©c\c1 ©sz11 "_5_’* '9"
cold, conc. distill p = 97.5-98°C at 5mm
(recrystallized Hg)

from EtOH)
*(See A. Oppenheim and S. Pfaff, Berichte, 885 (l875)).

meta- and ortho-Chlorovalerophenone. Made from l-bromobutane and

the meta- and ortho-chlorobenzonitriles (Aldrich).

meta- and ortho-Fluorovalerophenone. Made from l-bromobutane and

the meta- and ortho-fluorobenzonitriles (Columbia Organic Chemicals).

ortho-, meta-, and para-Trifluoromethylvalerophenones. Made from
l-bromobutane and the ortho-, meta-, and para-trifluoromethylbenzoni-

triles (Pierce Chemical Co.).

meta-Methylvalerophenone. Made from l-bromobutane and meta-tolu-

nitrile (Eastman).

y-Methoxyvalerophenone. Made from benzonitrile and l-bromo-3-meth-
oxybutane which was made from 3-methoxy-l-butanol (Matheson Coleman &

Bell) and phosphorous tribromide.

96

y—Vinylbutyrophenone. Made from 5-bromo-l-pentene (Chemical Sam-

ples Co.) and benzonitrile.

Method II. Phenyl magnesium bromide was prepared107,108 and the
aliphatic nitrile or substituted nitrile was added to it. The stoichio-
metric amounts and work-up procedure were the same as in Method 1. The
resulting yields were generally lower when this method was used and a
biphenyl coupling by-product was formed in some cases. Most of this was
eliminated in the ether layer during work-up but any that carried over
had to be carefully removed by recrystallization. The following ketones

were made by Method II:

y-Methylvalergphenone. Made from 4-methylvaleronitrile (K & K Lab-

oratories) and bromobenzene (Fisher Scientific Co.).

1:Chlorobutyrophenone. Made from 4-chlorobutyronitrile (Aldrich)

and bromobenzene.

y-Dimethylaminobutyrophenone. Made from 4-dimethylaminobutyroni-

trile (Columbia) and bromobenzene.

a-ChlorovalerOphenone. Made from 5-chlorovaleronitrile (Aldrich)

 

and bromobenzene.

meta-Methoxyvalerophenone. Made from meta-bromoanisole (Eastman)

 

and valeronitrile (K & K Labs). Several impurities were present in the

ketone made by this method.

97

ortho-Methoxyvalerophenone. Made from ortho-bromoanisole (Eastman)

and valeronitrile.

meta- and para-methoxy-y-methylvalerophenone. Made from meta- and

para-bromoanisole (Eastman) and 4-methylvaleronitrile (K & K Labs).

para-Thiomethoxyvalerophenone. Made from 4-bromothioanisole (Al-

drich) and valeronitrile.

e-Chlorohexanophenone. Made from bromobenzene and 6-chlorocapro-

nitrile (Columbia).

Method III. The cadmium reagent was prepared from the correspond-

 

ing Grignard reagent and anhydrous cadmium chloride109. In a typical
preparation 0.10 moles of the Grignard reagent was prepared and anhy-
drous cadmium chloride (0.105 moles for phenyl Grignards and 0.055 moles
for alkyl Grignards) was added. The ether solvent was replaced with
benzene and the acid chloride was then added dropwise after which the
solution was refluxed for l to 2 hours. Work-up proceded by pouring the
reaction mixture into ice water and acidifying with 10% sulfuric acid
solution. The ketone was then extracted, dried and purified. The fol-

lowing ketones were prepared by Method III:

6-Cyanovalerophenone. Made from bromobenzene and 5-cyanovaleryl
chloride. The latter was prepared from adipic acid monomethyl ester

(Aldrich) as follows:

98

 

 

0 6g NH 0H 0
\ 4 A \
H0>C(CH2)4 CH3 conc. ’ NH4+'0’b(CH2)4éNH2
”gt CflHz SOCl2 in benzene 0§
(CH2)4 reflux 24 hours ‘9 Cl"c(CH2)4CN
(N. E. found= 2140i5) (bp = 132-133°C at lOmm Hg)

para-Methoxyvalerophenone. Made from l-bromobutane and para-ani-

 

soyl chloride which was made from para-anisic acid (Aldrich) and phos-

phorous pentachloride.

meta-Methoxyvalerophenone. Made from l-bromobutane and meta-ani-

 

soyl chloride prepared as shown in Method I.

meta- and para-methoxy-y-methylvalerophenones. Made from 3-methyl-
l-bromobutane (Aldrich) and the meta- and para-anisoyl chlorides as

above.

Method IV. The corresponding aliphatic acid chloride was dissolved
in a large excess (10 to 15 fold) of pure benzene and l.l moles of anhy-
drous aluminum chloride per mole of acid chloride was added110. The re-
action mixture was protected with a drying tube and allowed to stir over-
night in the hood. The ketone was obtained by pouring the mixture into
ice water, acidifying to clear the solution, and extracting with ether.

The following ketones were made by method IV:

y:Methylvalerophenone. Made from 4-methylvaleryl chloride (East-

 

Inan) and benzene (Fisher).

99

Bjfi-Dimethylbutyrophenone. Made from tert-butyl acetyl chloride

(Aldrich) and benzene.

para-Phenyl-y-methylvalerophenone. Made from 4-methylvaleryl

chloride and biphenyl (Eastman) in benzene solution.

Method V. Occasionally the alkyl substituents on the phenyl ke-
tone can be altered. If the carbonyl group must be protected the ketal
is made and the required operation performed on the ketal (for example,
substitution111 or reductionllz). The following ketones were made by

substitution or alteration of the alkyl group on the phenyl ketone:

y-Hydroxybutyrophenone. Student-prepared B-benzoyl propionic acid,
made from succinic anhydride (Fisher) and benzene via the Friedel-Crafts
reaction, was repurified by recrystallization from chloroform and con-
verted to the ethyl ester (bp = l58°C at 8mm Hg). The ketal of ethyl
B-benzoylpropionate was made by refluxing with ethylene glycol (Fisher)
and a trace of benzene sulfonic acid in a benzene solution. The water
was removed via a Dean-Stark trap, the ketal-ester was isolated, and the
ester was reduced with lithium aluminum hydride in ether. The procedure
followed was that of Ward112; the only changes were that the work-up of
the reduced ketal was accomplished with a dilute sodium bisulfate solu-

tion and the hydrolysis of the ketal with a 1% sulfuric acid solution.

y-Cyanobutyrophenone. The ketal of y-chlorobutyrophenone (prepared
by Method II) was prepared as in the case of ethyl s-benzoylpropionate

and reacted with sodium cyanide in dimethylsulfoxide (Matheson

100

Coleman & Bell) at 85-90°C111. The reaction mixture was then diluted

with an equal volume of water and the ketone extracted with pentane.

e-Cyanohexanophenone. Made from e-chlorohexanophenone (prepared by

Method II) by reacting with sodium cyanide in dimethyl sulfoxide.

y-Dimethylaminobutyrgphenone hydrochloride. A solution of y-di-
methylaminobutyrophenone (made by method II) in benzene was vigorously
stirred while a slow stream of anhydrous hydrogen chloride was passed
over the surface of the solution. The crystals were filtered off and

purified.

y-Dimethylaminobutyrophenone ethyl bromide. A solution of y-dl-
methylaminobutyrophenone and a 3 to 4 fold excess of ethyl bromide in
pentane was allowed to sit, with occasional swirling, for 4 days. The

crystals were filtered off and purified.

Method VI. In two cases the alkyl ester was prepared by reacting
the corresponding acid with diazomethane. The diazomethane was prepared
from N-methyl-N-nitrosourea and base113 and was added to an ether solu-
tion of the acid until decolorization no longer took place. The ketones
prepared this way were y-carbomethoxybutyrophenone from 4-benzoylbutyric
acid, and y-carbomethoxyvalerophenone from 5-benzoylvaleric acid (both

acids from Aldrich).

b. Methods of purification. Purification of the ketones often

proved to be the most difficult step. The method which most

lOl

.Aoump .om Aymacmcv mm .a .e& mmw .mZmz gmcm a Emsu
mam .Q<Q:H as» ha umuaoum mpsmwmz uwsopm mo apnea ummpmp as» scam uwumpaupmo mew: mucmwm: gmpzumpoa ask a

 

 

.au_aopgaaapab Rom om_.eo_ Nom_=opu
congmu saga um~w~pmumxgumm > tango: ooomumm u as acocmnqogxuanxxoguazu»
.ua__wpmwuat mm m_
ace vagaacmopasocgu caspoo RAM a: 55m be cope_-oep cam.m_~ o I u
.Pocwgums soc; awNPppmumALomm H conpmz a: 55m um oomepumep u an mcocmcaocmcoz
.copua_FPmeuaL ace .sgamaa Roe a: 52m.“ an UOF¢F «Fm.¢o~ oo~=¢_o
-oumsoccu cszpoo .cowumFPFHmwo H conga: a: ssm.m um oomcpuoep u an mcocmsaocmxm;pazumswoue.c
nee omm.om_ omP=m_u
.=o_ua__wbm_o H uogbaz a: ssm.m um gamms-mm_ u an acocagaocaxagpsgpaz-a
gem
.cowa >H conga: u.m.o- u as
-mp__umwvmg ucm .mcmpcmn soc; up up
sac» ace Pocmgpas saga “map; fiom-em a: sea an u.mm-¢m mmm.m~_ o z u
cowp-_~_mpmsguaa .cowuap_wam_o HH aogbaz a: sam.m pm moNNF u an acocagaoaapmstguaz-»
.eowamp.wpm_uaa new .xgaaam amm a: saw an oomo_-hop mm~.m~_ om~=~_u
-ouasoagu =s=_ou .co_pappwpm_a >H vogpaz a: EEPN pa Oom~_-hmp u an acocaza02su=n_xgpaswo-u.m
.eowpa0wcwaza u_aa> a mawuaamata _auwaxmm «Seawaz Lapauapoz \apseLoa
we vosumz .cowumcmamcm
to noguaz agapag

.mmcoumx um~wmmsg=>w so» mama _muwngm

.H>x m4m<h

102

.coHumguHHH
mcwgzu mcmgcma szz ummch

.Hocmguw
musHomam scam umNHHHmumngmm

 

uem
.mcmucma :H «no»

1mg on muHEoga
Hxspa uauu<

.m=m~=on
:_ ocoumx

op Ho: uauu<

 

.NHo.o can» mmmH .mchEmL ammuom
HHHL=QE_ HHaam .=o_paH_HpmHo HH coca»:
.gmcmmz .a .a

.Lo an umgmamgm

.coHamHHHumHo H cocumz

xwm

Locum: .o .a .Ln

an umgwamgm

.coHHmHHHHmHa H cosumz

Rom

.coHpmHHHpmHo HH conga:

mom

.coHumHHHumHa H oozpmz
coHumuHmnga uHmH>a

Ho nogumz .coHumgmamga

Ho uosuwz

9: 25m an OOHHH

a: ago an oomm_-NmH u an

"OD

a: saw an oom.HNH-HNH u an
m: sam.m an QOHHH-0._

a: 55m an oomNH-NN_ u an

9: 55H.H an oceoH-moH u an

 

mmHugmnoga HmuHm>:¢

oem.ocm aszNNISHu
cases 33
mcocwcqogauznocHEmegpmeHou>

mmu.-~ HoozmHINHU
mvHLonuogu»:
acocmgaogzganocHsmszumeHon>

¢H~._mp ZOHHINHU
mcocmzqogzgaaocHamezpmeHau>

mmN.~m_ NooHINHu
mcocmcaoLmHm>xxosum21>

mm~.¢H_ OHHINHu
mcocmsaogxuaoncw>u>

omm.~mH HQOHHIQHQ
mcocmgaogxnzaogoHsuu>

m-.mHH NoeHIHHu
mcocmcaogzuznzxozuozu>

 

agmHaz L~_=aa_o:1HuH=eaoa

acouax

103

.mcmxm; eoLH Homuasmppmv
cowumNHHHmungums van :5:
-Hou Lama-pco;m co coHpmHHHamHo

.mcmucma soc» can» ucm Hmonpv
mcmxm; Eogm :oHpmuHHHmumxgumm

.mcHoLmHH cmzu mogsuxHe gmzpm
-wcwucma soLH :oHumNHHHmpmxgumm

.umsHHaam can mqungu
-mguop congmu sexy can» .mgau
-xHE gmgpm EamHomea1Hocmnpm
sag» umgHH :oHum~HHHmumxgumm

.Lacmmz .a .a .ao Ha uaHaHtaa
.mcmxm; Eogw :oHHmNHHHmpmxgumm

.mgszHE
gmgum EzmHomeauHocmgum EogH
:oHHmNHHHmumxgumL .coHumHHHumHo

.coHpmHHHpmHa

zomHzmHu

 

 

cowumuwawaza
ac nospaz

xmm ooem1m.mm u as Ho~.Ho~
> uogpmz a: saw an ooumHumoH n ma acocwsaocmxmcocmxuum
Ham SOH.oHN HQOmHINHQ
HH nozpmz Dom.Hm-Hm u as mcocmgaocmxmzocoH;Q-a
Hmm mow.o- momHzmHo
H> cogumz ooomumm u as mcocmznogmHa>xxogpmeongmuuo
Hem ¢m~.HmH zomHINHo
HHH cozpmz ooHH1mo u as ococmgaoLmHm>ocmxu10
.memwz .w .a mH
.Lo Ha mam: mom.omH ‘Huo IHHU
HH cospmz 00Hm1m¢ u as wcocmgaogmHm>ogonuuo
How comm u as mH~.mHH onHzHHu
> nocumz m: EEN pm gomaH1mmH u an mcocwgaogxgznocmxuur
Rom goap-m_ u as mm~.ao~ moapzmHo
H> vospmz 0: SEN um oom.HmH-HmH u an mcocmsaogxuanxxozumsongmuur
u—mH> a mmHuLwaaLa Hmonxga ugmHmz LmHaumHoz \uH32gou
.comumgmamga

Co uogpaz

accumx

104

.eoHBHHHHBHHQ

.ucmmmga HquzaeH
chamcmoumeoggu m>Hmemqmgg
an unsaap<v .coHBHHHHHmHo

.coHumHHHumHumg
ucm .Hocmgpme can mcmpcma saga
:oHumNHHkungumg .coHpmHHHumHo

.coHumHHHumHumg vcm
Hgamgmoumsogsu .Hocmcpme new
mcmucma eogm coHumNHHHmumALumm

.mcmzuo
Low :oHpmHHHumHumg mxzaogmou
-meogcu “Hocmgums can acmucma
scam coHpm~HHHmpngumm .cho
coHHMHHHHmHa - SuzuOLa H uosaa:

.Hocm
-gums ucm «amazon EoLH coHumN
-HHHmumxgumg new .coHpmHHHHmHa

 

coHpmuHaHaaa
Ho waspaz

 

m—IHH

 

Hmm aoa.omH Hue u

H voguaz a: ago an oomHH-o_H u an acocagaoaaH~>ocoH;o-o
.xoggao Rom

HHH aozaaz N mH mH

new a: see an u,~¢_-o¢H mm~.oo~ o x 0

HH uocpaz a: esm.~ an u._m_-om_ u an aeocagaoaaHo>Hagpae-»-xxo;paz-s
mom

Rom a: see an ooamH-omH mmm.mo~ o = 0

HH vocumz m: EEm.m um oommH u an «cocmznogmHn>Ha:pmsu>-axozpmzia

Hes www.mmH NomHINHQ

HH vogue: m: EEK um oomeH u an mcocmgaocmHm>zxosuwzio
HHH-¢~
HHH nocuaz
Hum

«Hm 9: ssm.a pa oommH-HmH mm~.~m_ o z u

H nogumz a: saw no oommHuemH u an mcocmcaoLmHo>xxospm215

.xoeaam amm www.mmH NoonmHo

HHH noguaz a: saw an gameH u an aeocagno2aHassxogpaz-a

uHmH> a mmHuLmnoLa Hmumega usmHm: LmHaumHoz \szsgou
.coHpmgmamgm

Ho uogpaz acoaax

105

.memmz .a .a
.29 an uaHHcha .coHHmHHHHmHo

.memmz .w .a
.Lo Hg uaHHHLHH .coHHaHHHHmHo

.cmcmmz .w .a
.LH Hg uaHHHLHH .coHHHHHHHmHo

.Hmcmm: .c .m
.Lo Ha umHHHezm .coHpmHHHpmHo

.memmz .w .H
.Lo Ha uaHHHtaa .coHHmHHHHmHo

 

.gmcmmz .c
.a .Lo Hg mum:
H aogHaz

.memmz .a
.H .Lo Ha ace:
H uogHaz

.memm: .6
.a .Lo Ha one:
H uogHaz

.gmcmmz .a
.H .Lo Ha one:
H conga:

.Hmcmmz .w
.H .29 Ha mum:
H nonumz

.pumnoca goHcom

 

a mo mHHgm:

. .m an age:

.mcmxm: so»; :oHum~HHHmungumm H uocumz
:oHHmuHHnga unH> a

we wanna: .coHumcwnmga

Ho uoguaz

as H-.om~ «HomHINHH
an meocmgaogwHm>Hanumsogo=HHthna

QOmMI¢m
a: EEoH Ha HomHH-mHH

HN~.om~ maomHINHu
an mcocmcnogmHm>HanumsogoaHmth15

a: EEHH Hm QOHHH-HHH

H-.om~ mHomszHo

m: EEHH an ooNNHuHNH an ococmzaonHm>HznumsogoaHHHHhuo

HHN.omH HomHzHHu

m: sEoH um QOeHH1mHH u an acocmgqogmHm>ogoaHmus
¢H~.omH HomeHHH

a: ssoH Hm uomoH u an mcocmgaogon>ogoaHmuo

mmo.mmH HuomHzHHu

 

 

oommumm u as mcocmgqogmHm>ogoH5u15
mmwugmaoga HmuHnga agmww: LmHaumHoz \szsLom
acoumx

106

 

.Hocmgpa aHHHOHHH soc» coHHHN HHH HomH-HH u as Hmm.~m~ oomzmHu
-HHHmHmngaL new coHHmHHHHmHo >H vogHaz a: 55m Hm HoemH-mmH u an acocagaocaHm>HH=Hae-»-HHcaga-a
.acmacaa EOLH coHHHN HHH Hon-mH u as mm.mo~ momHzmHu
-HHHmHmHgoaL new coHHaHHHHmHo HH noeHaz a: 55m Hm HommH-mmH u an acocagaoeaHm>ongHasoH;H-H
HHH mm~.HHH oonmHu
.coHHHHHHHmHo H cogHaz a: EEoH Ha oomNH u an meccaHHOLaHa>HHHHaz-E
:oHumuHHnga uHmH> m mmHuLmqoLg HmuHmaza unmwmz LmqumHoz \anexom

Ho uosumz .coHumLmamgm

Ho voguaz

mcoumx

107

.Hmumz
eaHHHHmHu EOLH coHHHNHHHHHmngam

.mcmxmc EoLH
mmsHu ngm>mm cmNHHprmxgumm

.wHaEwm
mo pcaoEm HHmEm can zquzq
;m_; we mmsmumn coHumuHHHgsg oz

.mcmxm; soHH :oHHmNHHHmHngumm

.coHHHHHHHmHo

.coHpmHHHHmHv
new azqwgmoumsocsu HHocmnams new
mcmucmq scam :oHHMNHHHmHmngmm

.coHumHHHpmHv
ucm qumgmoumeogcu HHocmgums new
mcmucma scam :oHHmNHHHmHmngmm

 

coHHaaHHHtaa Ho nonpaz

.mocoumx vmmmsugzm Ho coHpmuHHHgam

HomH-m.HH u as
o.mm-m.¢m u as
u.m~-m~ u as
u.m~-m.m~ u as

a: 52m
Ha u.~.ooH-ooH u an

a: as

m.oH Hm oomoH-moH u an
a: eem.m Hm Homo 1 Ha

 

mmHuconmga Hmunxmm

xcmasou
HmuHEmzu :uHLuH<

acmasou
HmoHemcu gquuH<

mHmuHamgu
uchmgo cmsammm

H¢~aeou
HmuHEmso :uchH<

mHmuHemzu
uHcmmgo cmsumam

mHmuHemgo
uHcmmgo casummm

acmasou
HmuHEmgo gquvH<

 

LmHHaqsm

.HH>x u4m<h

mm~.oom moaHINHu
uHu< uHLmHm>on~cmmum

mom.¢- omeoHo
acocmnaogxuanxcmsm-»

ama.~om oemzHHU
acocmcaocmuwumucma

mm~.mHH omHzmHu
acocmsaocmxmz

¢-.~mH QHHIHHU
mcocmsaogmHm>omH

HN~.NHH oeHzHHu
wcocmsaogmHo>

HHH.maH ouonHo
mcocosqogxuam

 

uzmHm: LmHauwHoz1wuH=ELom

mcoamx

108

.ogauxHE gmspm EamHoEuma1HoEmsum
musHomnE soEH EoHHmNHHHmHmHEumm

.mcmxm; soc» coHHENHHHmumxgomm

.mcmwa Eogw EoHHENHHHmpmHEomm

.mcmxw; Eogm :oHHmNHHHmpmxgomm

.mgzust Emsum EamHoEumH1Hocmzum

mquomnm Eogw :oHHmNHHHmumxgumm

.HHHHHHHHHHHH

.mgaust Locum EzmHoEHmauHocmsum
mquomam EoEH EoHam~HHHmungumm

o.mo-¢o u HE

Hoo~-HH u HE

u.m.o~-e~ u HE

Hom.Hm-Hm u HE
Hom.NH-m.HH u HE

H: EEm.N HH HooH-Hm 1 HH
UOHHH-HHH u HE

mHmustcu
oHcmmEo Emsummm

gmzmm
EEH NHHHEH

Emamm
HHH NHHHEH

mHmuHEmgu
uHcmmEo mHnsaHou

xdmasoo
HmuHsmzu EuHEuH<

Acmqeou
HEUHEmEu guHEuH<

zcmasoo
HmoHemsu EuHEuH<

m-.mHH NHEHEHHQ
mEoEmEHoEmHE>onEuH11H

HHN.EHH HHHENHH
mEoEmEHoEmHm>chpm21H

EH~.omH HomHEHHH
acocmEHoEmHE>oEo=HH1H

HHE.HHH HuomHEHHH
acoEoEHoEmHm>oEoH5uua

mom.¢- HHHEHHU
mzocmcaogxuanHacmgHum
HHH.EHH NHEHEHU

mcocmgaoumumxxocpm215

oEm.HH~ NomHEEHH
acmuanHHUNanHo-¢.H

 

EHHHHHHEHEHH EH EHEHaz

mmeLumogm HmuHmaza

EmHHnnmw

pcmHoz EmHauoHoz \onsgoH

mcopmx

109

efficiently improved the product was used. Quite frequently several

methods were used to purify one ketone.

Method A. Distillation at reduced pressure was the most commonly
used purification step. Reduced pressure was necessary for the com-
pounds in this general class because of their high boiling points. The
usual range of pressure used was 5 to 20 mm Hg. Distillations were made
through a 25 cm vacuum jacketed vigreaux column or a l2 cm microware
vacuum jacketed vigreaux column, depending on the amount of crude ketone
available. The acceptable product cut was usually taken when the ther-
mometer reading at the top of the column came within 3 degrees of the

thermometer reading at the bottom of the column.

Method B. Recrystallization was found to be a very effective means
of removing small amounts of impurities and producing a very pure pro-
duct. Both polar and non-polar, and sometimes mixed solvents were used.
The last traces of solvent was removed in a vacuum dessicator. Most
common solvents used were: pentane, hexane, methanol, absolute ethanol,
distilled water, carbon tetrachloride, petroleum ether, and petroleum

ether-ethanol mixtures.

Method C. Some liquid ketones were passed neat through a small
column of alumina while close watch was maintained so that any colored
bands formed did not pass through into the product.- In some cases this
seemed to be the only way that small amounts of colored impurities could
be removed from the ketone (i.e. valerophenone). A neutral alumina of

fairly high activity was used. (If the activity was in doubt the

110

alumina was dried for several days at 120°C to insure activity. It is
felt that this treatment also removes small amounts of moisture from the

ketone.

Method 0. Sublimation was used on occasion when other methods
could not purify the ketone to desired stadards and when the physical
properties of the ketone allowed for efficient sublimation. A vacuum
sublimation was used at a pressure of about 0.01mm Hg with gradual warm-
ing until sublimation took place. The collection tube was generally

cooled with a dry ice-isopropyl alcohol mixture.

c. Criteria of purity. It is very important that the ketones used
are of the utmost purity since in some cases even the smallest amount of
quencher (0.0001 molar) changes the photochemical behavior. Also, small
amounts of polar impurities such as water or alcohols may affect the
data. Besides the usual physical characteristics of small melting or
boiling ranges and a colorless or white appearance, all ketones were
checked closely using vapor phase chromatography. Only a few ketones
had impurities amounting to 0.10% or more of the peak area of the pro-
duct ketone. Assuming the impurities were of approximately the same
molecular weight as the ketones, the final 0.10 molar solution would
have 0.0001 molar impurity present. Most of the ketones were very pure
having barely detectable impurities or amounts less than 0.01%. As a
final check, if an impurity was suspected to be causing quenching and
lowering of the quantum yield, a photolysis run was made at varying
concentrations of ketone. If no decrease in product with increasing ke-

tone concentration was noted, it was assumed that no quencher was

111

present in the ketone.

2. Purification of Solvents and Other COmpounds.

The purity of the solvents is a very critical consideration in re-
lation to the photochemical behavior of the ketones to be photolyzed in
them. It would be possible for some impurity to be a triplet quencher,
a light absorber, or a species more polar than the solvent itself, all
of which would alter the values obtained in the experimental runs. This
is especially true where the ketone is of low reactivity and therefore
more subject to quenching, etc. The solvents, quenchers, and standards
used were all purified by some means and an evaluation of their purity

made. Methods of purification of the compounds used are listed below.

a. Benzene. Benzene was the solvent of choice for the photolysis
of the phenyl ketones. It is transparent in the region of absorption
studied, it is quite non-reactive photochemically in this region, the
hydrogens are not easily abstracted, and it is fairly easy to purify to
a high degree. Thiophene free, 99 mole % benzene supplied by Fisher
Scientific Co. was further purified by stirring over concentrated sulfur-
ic acid (5% by volume) for several days. The sulfuric acid layer was
then removed via a separatory funnel and another portion added and stir-
ring was continued for a like period of time. This was repeated 3 or 4
times or until the sulfuric acid no longer turned yellow after stirring.
The benzene was then stirred over a dilute (1M) KDH solution (=10% by
volume) for one day and then dried over 4 mesh calcium chloride for a
day. The benzene was finally distilled from P205 (about 10 gr/liter of

benzene) through a 45 cm column packed with glass helices. A reflux

112

ratio of 10:1 or larger was maintained at the distilling head and approx-
imately 10% of the benzene was discarded as the forerun and 10% remained
in the pot. The boiling point was 79.8:0.2°C, uncorrected. The benzene
treated in this manner seemed quite satisfactory, no discoloration occur-
red even on extended irradiations. 0n injection into the VPC under normal
conditions a small impurity precedes the benzene peak on the VPC chart
and a very small one, which is also apparent on the manufacturers VPC
strip, comes off on the tail of the benzene. These are apparently inert

and in no way interfere with the analysis.

b. Methanol. Methanol supplied by Fisher Scientific Co. was fur-
ther purified by adding approximately 1 gram of magnesium shavings per
liter of methanol and distilling through a 45 cm glass helice packed col-
umn. A reflux ratio of 10:1 or greater was maintained at the distilling
head and a middle fraction of approximately 60% was collected. The boil-
ing point was 64.5:0.2°C, uncorrected. The methanol purified as above
was stored in a clean, dry bottle and kept tightly capped. For reliable
results it was found that the methanol should be purified on the day it
is to be used, or at least not more than one or two days prior to use.

Methanol stored for more than a week was found to be unsatisfactory.

c. tert-Butyl alcohol. The tert-butyl alcohol used was supplied by
the J. T. Baker Co. and further purified by treatment with clean, fresh-
ly cut metallic sodium, about one gram per liter of alcohol. The sodium
did not react until refluxing temperature was reached, indicating that
the alcohol was quite dry. The tert-butyl alcohol was distilled through

a 45 cm glass helice packed column at a reflux ratio of 10:1 or larger,

113

and a middle fraction of about 60% was taken for use and stored in a

tightly capped bottle. The boiling point was 82.0:0.2°C, uncorrected.

d. Acetonitrile. Acetonitrile supplied by Fisher Scientific Co.
was further purified by D. J. Buchekll“ by distillation from potassium
permanganate. Distillation was done through a 45 cm glass helice packed
column and approximately 10% was discarded as forerun and 10% remained

in the pot. The boiling point was 81.5:0.2°C, uncorrected.

e. 2,5-0imethyl-2,4-hexadiene. This most commonly used quencher
was obtained from Aldrich Chemical Co. and was purified by first distil-
ling through a 25 cm vigreaux column, collecting a 60% middle fraction
with a boiling range of 40.0 to 40.5°C at 20mm Hg. The collected mater-
ial was then recrystallized from itself by cooling until partially fro-
zen and decanting the unfrozen portion of the liquid. The recrystalli-
zation was repeated. The impurities in the commercial diene were found
to be significantly reduced by the above procedure. The 2.5-dimethyl-
2,4-hexadiene from another supplier, Chemical Samples Co., was found to
be of somewhat better purity than that obtained from Aldrich. Upon
standing on the shelf near the freezing compartment in the refrigerator,
large crystals would sublime to the top of the bottle. These crystals

were scraped out and used on occasion without further purification.

f. Piperylene. Commercially obtained piperylene from Aldrich
Chemical Co. which was merely distilled as a means of purification was
found to cause a reddish-brown color in benzene solutions upon extended

irradiation. By passing the piperylene through a 5 inch layer of

114

neutral alumina and then redistilling, extended irradiations could be
made without discoloration. Distillation was made through a 25 cm vig-

reaux column and a 60% middle fraction was collected for use.

9, Internal standards. The internal standards used were all high

 

molecular weight alkanes which were further purified by stirring over
concentrated sulfuric acid until the acid would no longer discolor. The
alkane was then rinsed with a dilute base solution, dried over calcium
chloride and distilled at a reduced pressure. In the cases where the
alkanes are solids at room temperature, the final pruification was by
recrystallization from absolute ethanol. The standards used in this pro-

ject were all purified by Prof. P. J. Wagner and are listed below.

 

 

Standard Supplier bp or mp
Tetradecane (C14) Columbia Organic Chem. 119-120°C at 10mm Hg
Pentadecane (C15) Columbia Organic Chem. 132°C at 10mm Hg
Hexadecane (C16) Aldrich Chemical Co. 146°C at 10mm Hg
Heptadecane (C17) Aldrich Chemical Co. 158°C at 8mm Hg
0ctadecane (C18) Aldrich Chemical Co. mp = 29-30°C

Eicosane (020) Matheson Coleman & Bell mp 35-35.5°C

h. Pyridine. The commercial pyridine supplied by Fisher Scienti-

 

fic Co. was distilled through a 25 cm vigreaux column and a middle frac-
tion of about 60% was collected for use. The boiling range was

ll4.5i0.5°C, uncorrected.

i. Ethyl acetate. The ethyl acetate was distilled as above, and a

 

middle fraction (60%) boiling at 77:0.2°C was collected for use.

mmim.rmmmws

1. Preparation of Photolysis Samples.

a. Photolysis solutions. Stock solutions of the ketones were pre-
pared by weighing out the required amount into a volumetric flask, then
pipetting into the flask the predetermined amount of an internal stand-
ard solution and then filling to volume with solvent. Individual flasks
for the quenching runs were made up by pipetting an equivalent amount of
the stock ketone-standard solution into numbered volumetric flasks, add-
ing the required amount of a standard quencher solution and filling to
volume with solvent. The solutions were then injected into pyrex photo-
lysis tubes using a 5 ml hypodermic syringe with a 4 inch needle, fill-
ing each tube uniformly with 2.8 ml. The photolysis tubes were prepared
from selected culture tubes by heating the neck of the tube to the soft-

ening point and drawing it out approximately 4 inches.

When solvent effects were measured the same procedure was followed
except that increments of a standard tert-butyl alcohol solution were
added instead of quencher. In some cases quenching runs were made in
solvents other than benzene. The procedure was the same except for sub-

stitution of the solvent.

b. Degassing_procedure. The ketone solutions were degassed using
a process similar to that used in earlier studies35’115. The sample
tubes were attached to a vacuum line over No. 00 one-hole rubber stop-
pers on individual stopcocks. The solutions were frozen in liquid nitro-

3

gen and a vacuum of l x 10' mm of mercury (or less) was applied for

115

116

several minutes. The samples were then allowed to thaw and the cycle
was repeated. After the third freezing and evacuation the tubes were

sealed off with a gas-oxygen torch.

c. A typical run. The procedures for making up the solutions and
the photolysis samples are illustrated in the following run performed on

hexanophenone.

(l) A 0.10M stock tetradecane standard solution was made up by
weighing out 0.9920 gr of tetradecane into a 50 m1 volumetric flask and

filling to volume with benzene.

(2) A stock 0.10M quencher solution was made up by weighing out
2.7550 gr of 2,5-dimethy1-2,4-hexadiene into a 250 m1 volumetric flask

and filling to volume with benzene.

(3) A stock 10.0M tert-butyl alcohol solution was made by weighing
out 18.531 gr of tert-butyl alcohol into a 25 ml volumetric flask and

filling to volume with benzene.

(4) The stock ketone solution was made by weighing 2.2033 gr of
hexanophenone into a 25 ml volumetric flask and pipetting in 5 ml of

stock 0.10M tetradecane standard solution.

(5) Individual sample flasks were made up by pipetting 2 ml of the
stock ketone-standard solution into a 10 ml volumetric flask, adding the

desired amount of quencher solution or alcohol solution, and filling to

117

volume with benzene. The run was labeled as follows:

Quenching run [0.10M hexanophenone, 0.004M tetradecane standard].

 

 

Quencher Quencher
Flask # added Concentration
X - 1 0 0
X - 1' 0 0
X - 2 1 m1 0.010M
X - 3 2 ml 0.020M
X - 4 3 ml 0.030M
X - 5 4 m1 0.040M

Solvent study [0.10M hexanophenone, 0.004M tetradecane standard].

 

 

 

tert-Butyl tert-Butyl alcohol
Flask # alcohol added Concentration
X - 6 0.5 ml 0.5M
X - 7 1.0 ml 1.0M
X - 8 2.0 ml 2.0M
X - 9 5.0 ml 5.0M
X - 10 8.0 ml 8.0M

The solutions were then placed in photolysis tubes for degassing as pre-
viously described. For the analyses see Appendix A, Part 1, Hexanophen-

one, Run 2, and Appendix A, Part 4, Hexanophenone.

2. Photolysis Procedure.

 

The sample tubes were irradiated in a merry-go-round apparatus spe-
cially designed to give each tube an equivalent amount of light, and to
allow for filtration of the light for a specific mercury emission band.
The apparatus is described in detail by Moses, Liu, and Monroe115. The
light source used was a Hanovia medium pressure 450 watt mercury lamp
which was held in a water cooled quartz probe. The probe was inserted
into a cylindrical pyrex tube containing a filter solution and of a dia-

meter to allow a one centimeter pathlength through the filter solution.

118

All photolyses were done at 3130A (unless another wavelength is speci-
fied) using a 0.002 molar potassium chromate; 1% potassium carbonate
filter solution116,117. The samples were generally photolyzed until from
3 to 6% of the original ketone (0.10M solution) was converted to product.
Occasionally the conversion was outside of these limits if the conditions
warranted it (such as extremely high or low quantum yields), and occa-
sionally actinometer tubes were photolyzed to 8 or 10% conversion, but
only after determining that the quantum yield was not affected by doing

so (See section on justification of results).

3. Procedure for Estimation of Ketone Disappearance and Cyclobutanols.

 

The procedure for determining the quantum yields of ketone disap-
pearance and cyclobutanol formation follows quite closely that already
described with some minor changes. The ketone-standard solution was made
up and a 2.8 ml sample was degassed, sealed, and photolyzed as before.
Then the original ketone disappearance was measured as a ratio of ketone
to standard. The proposed cyclobutanols were measured as a product to
standard ratio. The standard had to be carefully chosen so it would not
interfere with either the parent or product ketones or the cyclobutanols.
Analyses were done by gas chromatography with varying of the conditions
for maximum efficiency in separation of the components. The principal
differences in procedure followed from that of Section 1 were: (1) The
stock solutions were made up by weighing out the ketone and internal
standard to a minimum accuracy of 10.0005 grams. This was necessary be-
cause actual moles of disappearance had to be determined from the percent
disappearance of the parent ketone. Also, the presumed cyclobutanol(s)

formed were measured as a percent appearance of cyclobutanol compared to

119

original ketone, assuming that the detector sensitivity towards the cy-
clobutanol was the same as for the parent ketone. It is felt that any
actual difference here would not be very large. (2) Photolyses were
carried out to 10 to 25% conversion in order to achieve a more measure-
able difference in the parent ketone peaks with VPC. This causes no
serious problems except in cases where the ketone produces a quencher
upon photolysis. (3) The analysis of these samples had to be carried
out at somewhat more extreme conditions in order to get reasonable ex-
perimental accuracy. Although the parent ketone should not be affected
by these higher temperatures (Appendix A, Part 3), there is some possi-
bility that the cyclobutanols may be decomposed. (4) In a few cases no
cyclobutanol(s) was seen in the VPC analysis and it was thought possible
that it may be coincident with the parent ketone. In such cases another
analysis was made using a column of slightly higher polarity (5% QF-l
and 1.5% Carbowax 20M). This column was checked using a photolyzed val-
erophenone solution and was found to hold up the cyclobutanols longer
relative to the parent ketone than the standard column of 5% QF-l and

1% Carbowax 20M. If no cyclobutanol(s) was seen using both columns it

was assumed none was present.

4. Analysis Procedure.

 

a. Instruments. All of the analyses for product formation and ke-
tone disappearance were obtained by gas chromatography. The instruments
used all had flame ionization detectors and in general had similar char-
acteristics. Response to the standard-product mixtures used were found
to be the same on the three instruments employed. Each of the instru-

ments was prepared for on-column injection of the samples so that high

120

injection port temperatures for "flashing" the samples were not needed.
Nitrogen or helium was used as a carrier gas and flow rates were adjust-
ed to manufacturers recommendations. The data was quantitatively re-
corded on strip chart recorders equipped with DISC area integrators.

The samples were injected with a Hamilton microliter syringe (#7101) us-
ing two or three 0.3 to 0.5 microliter shots per sample. The syringe

was rinsed 20 times with clean-solvent before analyzing another sample.

b. Conditions. The analytical conditions providing the optimum

 

time per single analysis (about 3 to 5 minutes) and the greatest sensi-
tivity to the samples being analyzed were adopted. The following condi-
tions are arranged in sets and will be referred to as such elsewhere.

Since the most common standard-product combination is tetradecane-aceto-
phenone, the conditions used for its analysis are referred to as stand-
ard sets of conditions (Std Set). Special conditions denote changes to
analyze other standard-product combinations (e.g. octadecane-meta-meth-
oxyacetophenone). The sets of conditions are given in Table XVIII with

special conditions in Table XIX.

c. VPC trace. An exact copy of a VPC trace of the analysis of

hexanophenone is shown in Figure 15.

5. Area-Mole Response Ratios for Internal Standards.

The use of a photochemically inert internal standard in the photo-
lysis solution permits the normalization of the injection sample sizes.
The error involved in attempting to use the exact amount in each VPC in-

jection is thereby eliminated. For quantitative work or when using an

121

w Lo .¢.N
oH

HHHHH .xHEHH<
HooHH-HHH

Hana EzHgmzc<

:HE\HE om
cHE\HE ow

umHHngaumEa .Emmoguwz

H HHHHHEHEEH HHHHH EH
EHN XHEHHEHH HH H H-HH HH

Eacpssz
nucH :pm\H x pH 0

H-HHH HHHHE HH-H= EHHEHHEHE

 

HHH Ham HEHHEHHH

m Lo .¢.~
H

goomHuonH
oommHuomH
oooHHumoH

xcmu HHH ummmmgaeou
EHE\HE om

EHE\HE om

cmHHHeaqumLa .Emmoequ

a geomosoenu om\om co
zom xmzonemu nH a Hume we

EEEHEHHH
socH ;Hm\H x aw NH

oomH
mmHgmm zamgmogw< EHHLH>

 

HH Ham HEHHEHHH

m so .e.m
H

uoomHuonH
ooovHummH
uHmHHuoHH

asza EHHEm3c<

EHEHHE HN
HHEHHE Hm
EHHHHE

H HEHHHEHLEH HHHHH EH
EHN XHEHHEHH HH H H-HH NH

EHEHEEHH
HHEH EHHHH x HE NH

AcsaHou Hmsov com
mmpemm namemoem< EHHEH>

 

H Ham HLHHEHHH

.HEHHHHHEHH HHHHHHHHEE .HHH>x HHH<E

HEoHumscmuu<
"mmcmz
.HHEHHHHH
"saucepan
HugoH HoHumnEH
HEw>o
.mmgzpmemgeoh
”musaom EH< msHHH
"Emmogca:
"new EwHeemo
.mmume onH

”new gmngmu

HmcpxumH cszHoo

”EszHou

upcoEEEHmcH

 

 

TABLE XIX.
Spec1al
Set Same as:
Ia Standard Set I
Ib Standard Set I
Ic Standard Set I
Id Standard Set I
Ie Standard Set I
If Standard Set I
19 Standard Set I
Ih Standard Set I
IIa Standard Set II
IIb Standard Set II
IIc Standard Set 11
IId Standard Set II
IIIa Standard Set 111
IIIb Standard Set III
IIIc Standard Set III
IIId Special Set IIIc
IIIe Special Set IIIc
IIIf Special Set IIIc
1119 Standard Set III

122

Special Analytical Conditions.

 

Except for:

Oven Temperature = 120-125°C
Oven Temperature = 130°C
Oven Temperature = 140°C
Oven Temperature = 150°C
Oven Temperature = 155°C
Oven Temperature = 170°C

Oven Temperature = 175°C

Column. 8 ft x 1/8th inch aluminum
Column packing. 4% QF- -1 & 1. 5% Carbowax
20M on 60/80 Chromosorb G.

Oven Temperature = 120°C
Oven Temperature = 125-130°C
Oven Temperature = 135°C
Oven Temperature = 140°C
Oven Temperature = l40-145°C
Oven Temperature = 165°C

Column: 6 ft x l/8th inch stainless steel
Column packing: 5% SE-30 on 60/80
Chromosorb N

Oven Temperature = 195°C

Injector Port Temperature- = 220°Ca

Oven Temperature = 210°C
Oven Temperature = 170°C
Oven Temperature = 185°C

Column: 10 ft x l/8th inch aluminum
Column packing: 10% Carbowax TPA on
60/80 Chromosorb w

Oven Temperature = 200°C a
Injection Port Temperature = 250°C

 

a These are approximate temperatures since there is no direct tempera-
ture measurement on the Hi-Fi GOO-C.

123

.mzocmgaoumu< Auv .acmumumeumh Amy .m=m~=wm A<v

 

.mcocmsaocmxm: Amy new .mPocmuzaopoau.mHm new mcmgp on
.mcocmnaocmxm: mo mwmapmc< Low usage oa>

 

 

Amy

 

 

“my

 

 

 

Auv

 

a:

.12 .233“.
+. GEE.

 

A<V

J.

+ asuodsau

 

 

 

 

 

124

actinometer which produces a product different from the ketone under in-
vestigation it is necessary to know the response ratio per mole for the
standard-product combinations. These ratios have been determined as

area/mole (Std) to area/mole (Prod) for a number of standard-product

combinations by weighing out the appropriate amounts of each and dilut-
ing with benzene; then making 6 or more determinations of product/stan-
dard ratios by VPC. The average value of the ratios and the molar con-
centrations of the standard and product were then used to calculate the

molar response ratios (Table XX).

Illustrating the use of the molar response (MR) ratios the quantum
yield for a ketone with a different standard-product combination from

the actinometer is calculated from a series of ratios as follows:

Prod

 

 

 

§f3-[Ketone] Std conc[Ket] MR ratio [Ket]
°II = -Prod x x x °IIEA°tJ
§{3—-[Act] Std conc[ActJ MR ratio [Act]

6. Controls in Experimental Procedures.

3

a. Degassing_procedures. A pressure of 1 x 10' mm Hg was consid-

ered adequate for degassing all of the ketone solutions. Under optimum

conditions 1 x 10'4

mm Hg could be obtained. The adequacy of the former
pressure for reducing the dissolved oxygen to a non-interfering level
has been demonstrated for the case of ketones with extremely long trip-

let lifetimeslle.

b. Volumetricglassware. All volumetric pipets and flasks used in

making up solutions and dilutions were class-A volumetric ware. The

125

TABLE XX. Standard-Product Molar Response Ratios.

Area/mole($td) Analytical

 

 

 

 

 

Standard/Product Area/molelProd): Conditions
Tetradecane/Acetophenone 2.0 Std Set II
Tetradecane/m-Trifluoromethyl- a

acetophenone 1.85 Std Set 111
Tetradecane/m-Fluoroacetophenone 2.0b Std Set I
Pentadecane/m-Methylacetophenone 2.0 Special Set IIa
Pentadecane/p-Methylacetophenone 2.0 Special Set IIa
Hexadecane/p-Chloroacetophenone 2.3C Special Set IIb
Hexadecane/m-Chloroacetophenone 2.3 Special Set IIb
Hexadecane/Acetophenone 2.3 Std Set I
Heptadecane/m-Methoxyacetophenone 2.3d Special Set IIb
Octadecane/p-Methoxyacetophenone 2.4e Special Set IIIa

Specialoget IId

Octadecane/p-Chloroacetophenone 2.1 Special Set IIc
Eicosane/p-Phenylacetophenone 1.5 Special Set IIIc

 

a Also used for ortho- & para-trifluoromethylacetophenone determinations.

b Used also for ortho- and para-fluoroacetophenone determinations.

c Used also for ortho-chloroacetophenone determination.
d

Used also for ortho-methoxyacetophenone determination.

e Used also for meta-methoxyacetophenone determination.

126

least accurate piece of volumetric equipment was the 5 m1 hypodermic
syringe used to inject 2.8 mls of solution into the photolysis tubes.

Here the consistency is dependent on technique.

c. Photolysis tubes. Corning brand pyrex 13 x 100 mm straight lip

 

culture tubes were used for all determinations. In early tests Exax
brand tubes were tried, however, the results were erratic. One problem
is that Exax glass transmits only about 60% of the light at 3130A. The
pyrex tubes were culled to exclude any with outside diameters not within
the 13 to 13.1 mm range. Also the lip was carefully examined for any
deformity which would preclude a good seal with the rubber stoppers on
the vacuum line. The following paragraph describes the cleaning of the

tubes.

d. Cleaning of glassware. Since very small amounts of impurities
in the solution could affect the photochemical behavior of the ketones
in some cases, all of the glassware used was subjected to a ritualistic
cleaning. First it was soaked in a commercial glass cleaner solution
over a steam bath for a day or two, then rinsed with hot tap water 4
times, with distilled water 3 times, and allowed to drain dry upside
down. When completely dry the glassware was placed in an oven at 120°C
for l or 2 days. There was some concern that silicone stopcock grease
may be contaminating the solutions and quenching the photoreaction of
the ketones. Silicone grease was tested and found not to be a quencher.

(See Table XXI)

e. Accuracy and reliability. It is interesting to estimate the

 

127

TABLE XXI. Effect of Silicone Stopcock Grease on Photolysis of Valero-

 

phenone.
Valerophenone, 0.10M in benzene, Prod/Std
0.004M tetradecane standard. Ratio
Actinometer tubes (no silicone grease) 6.80
Contains 0.20 gr/10 mls of silicone grease 6.64

 

maximum possible error, excluding gross accidental errors, and compare
this to the general spread of the data. A qualitative method for esti-
mating the maximum error is to total the percent uncertainties of each
inidvidual operation119. This was done as follows for a "normal" run
which is typical of practically all the ketones studied. Each operation
for which the tolerances are known is listed.

(1) Uncertainty in standard concentration (e.g. tetradecane)

 

120

(a) Weighing on Sartorius substitution balance, %
approximately 0.500010.0005 grams ' 0.10

(b) Class A volumetric flask, 25 ml, stock sol'n. 0.12

(c; Class A 2 ml pipet, to ketone solution 0.30

d Class A 10 ml volumetric flask 0.20

(e) Class A 2 m1 pipet, to sample flask 0.30

(f) Class A 10 m1 volumetric flask 0.20

(2) Uncertainty in ketone concentration (Not critical for ¢II)

(3) Uncertainty in filling photolysis tubes,
5 ml hypodermic syringe, 2.8:0.02 mls 0.70

(4) Uncertainty in light absorption (greater than
99.9% absorption for 1 cm cell) for selected
photolysis tubes, outside diameter 1.305:0.005 cm 0.10

(5) Uncertainty in area of irradiation ports of
merry-go-round apparatus (machined to 0.0001 inch
tolerance) 0.00

128

(6) Uncertainty in VPC analysis, based on range of
Prod/Std ratios for a large number of
determinations ‘ 2.0*

 

 

Total uncertainty in results (%) 4.0

When the actinometer is subjected to essentially the same treatment
as above and also considered, the maximum possible error is 8%. The
spread in the values of ¢II in Tables I and 11 generally fall into the
3 to 4% range and the percent deviation, if it were calculated, would
even be less. In the quenched samples, which involve two more pipeting
operations, all but a few of the individual slopes have a standard de-
viation of less than 5% of the value of the slope. This comparison
shows that the technique is sufficiently good to cause much of the un-
certainty to cancel out. This could be especially true in comparing ac-
tinometer and sample tubes. Reliability of the measurements appears to
be on the order of 14%, except in the cases of the ortho- and meta-meth-

oxyvalerophenones which are apparently subject to very subtle variations.

 

* This was the maximum deviation from the average that was acceptable,

analyses were repeated until satisfactory in cases where deviations were
larger. In the qreat majority of the analyses two or three consecutive

injections varied by less than 1%.

BIBLIOGRAPHY

Literature cited:

10.
11.
12.
13.

14.
15.

16.
17.

18.
19.
20.
21.

Wilfred W. Robbins, T. Elliot Weier and C. Ralph Stocking, Botan ,
3rd Ed., John Wiley & Sons, New York (1964), pp. 56-58; 183- .

Jack G. Calvert and James N. Pitts, Jr., Photochemistry, John Wiley
& Sons, New York (1966), p. 19.

 

John William Draper, A Textbook on Chemistry, 6th Ed., Harper and
Brothers, New York (1850), pp. 90-94.

W. A. Noyes, Jr., G. B. Porter and J. E. Jolley, Chemical Reviews,
56, 49 (1956).

W. Davis, Jr., Chemical Reviews, 49, 201 (1947).

W. A. Noyes, Jr., and P. A. Leighton, Photochemistry_of Gases,
Reinhold, New York (1941).

Walter J. Moore, Physical Chemistry, 2nd Ed., Prentice-Hall, Inc.,
New Jersey (1955), pp. 605-606.

Farrington Daniels and Robert A. Alberty, Ph sical Chemistr , 2nd
Ed., John Wiley & Sons, New York (1963), pp. 666-668.

R. G. W. Norrish and M. E. S. Appleyard, J. Chem. Soc., 1934, 874.

 

C. H. Bamford and R. G. W. Norrish, J. Chem. Soc., 1935, 1504.
C. H. Bamford and R. G. W. Norrish, J. Chem. Soc., 1938, 1531.
Ibid., p. 1532.

W. Davis, Jr., and W. A. Noyes, Jr., J. Am. Chem. Soc., 69, 2153
(1947).

F. 0. Rice and E. Teller, J. Chem. Phys., 6, 497 (1938).

. R. McNesby and A. S. Gordon, J. Am. Chem. Soc., 89, 261 (1958);
. Ausloos and E. Murad, J. Am. Chem. Soc., 89, 5929 (1958).

. Srinivasan, J. Am. Chem. Soc., 81, 5062 (1959).

J

p

R

G. R. McMillan, J. G. Calvert and J. N. Pitts, Jr., J. Am. Chem.
Soc., 86, 3602 (1964).
N

p

R

p

. C. Yang and D. H. Yang, J. Am. Chem. Soc., 89, 2913 (1958).

. Yates and A. a. Szabo, Tet. Let. #9, 485 (1965).

. B. LaCount and C. E. Griffin, Tet. Let. #9, 1549 (1965).

. J. Wagner and G. S. Hammond, J. Am. Chem. Soc., 81, 4009 (1965).

129

22.
23.

24.
25.

26.

27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.

38.
39.
40.

41.
42.

43.

44.

130

A. Terenin and V. Ermolaev, Trans. Faraday Soc., 68, 1042 (1956).

H.9Lé)J. Backstrdm and K. Sandros, Acta Chem. Scand., 18, 823
l 5 .

H. L. J. Backstrfim and K. Sandros, Acta Chem. Scand., 18, 48 (1960).

G. S. Hammond, P. A. Leermakers and N. J. Turro, J. Am. Chem. Soc.,
88, 2396 (1961).

G. S. Hammond, J. Saltiel, A. A. Lamola, N. J. Turro, J. S.
Bradshaw, D. 0. Cowan, R. C. Counsell, V. Vogt and C. Calton, J.
Am. Chem. Soc., 86, 3196 (1964).
T. J. Dougherty, J. Am. Chem. Soc., 81, 4011 (1965).
P. Ausloos, J. Phys. Chem., 66, 1616 (1961).
C. Walling and B. B. Jacknow, J. Am. Chem. Soc., 88, 6108 (1960).
C. Walling and B. B. Jacknow, J. Am. Chem. Soc., 88, 6113 (1960).
C. Walling and A. Padwa, J. Am. Chem. Soc., 86, 1593 (1963).
Ibid., p. 1597.
. Walling and M. J. Gibian, J. Am. Chem. Soc., 86, 3902 (1964).
. Walling and M. J. Gibian, J. Am. Chem. Soc., 81, 3361 (1965).

C
C
P. J. Wagner and G. S. Hammond, J. Am. Chem. Soc., 88, 1245 (1966).
J. I. Steinfeld, Acc. Chem. Res., 8, 313 (1970).

P

. J. Wagner, Lecture Notes: Chemistry 956, Organic Photochemistry,
Winter term, 1967, Michigan State University.

N. J. Turro, Chem. & Engr. News, #20, 66, 84 (1967).
P. J. Wagner and I. Kochevar, J. Am. Chem. Soc., 99, 2232 (1953).

P. J. Wagner and G. 5. Hammond, in Advances in Photochemistry,
Vol. 5, Interscience, New York (1968).

 

See Reference 2, pp. 273-293 and leading references.

J. L. Comstock (M. 0.), Elements of_Chemistry, 19th Ed., Robinson,
Pratt & Co., New York (1836), pp. 45-48.

G. N. Lewis and M. Kasha, J. Am. Chem. Soc., 66, 2100 (1944) and
previous included references.

G. N. Lewis, D. Lipkin and T. T. Magel, J. Am. Chem. Soc., 68,
3005 (1941).

 

45.
46.
47.
48.

49.
50.

51.
52.
53.
54.
55.
56.
57.

58.

59.
60.

61.
62.
63.

64.
65.

66.

67.

131

G. N. Lewis and M. Kasha, J. Am. Chem. Soc., 61, 994 (1945).
See Reference 43, p. 2101, and Reference 2, pp. 286-287.
See Reference 2, p. 283.

H. L. J. Backstrbm, A. Steneryr and P. Perlmann, Acta Chem. Scand.,
12, 8 (1958).

V. Brunet and W. A. Noyes, Jr., Bull. Soc. Chim. France, 121 (1958).

Ji9%B)Michael and W. A. Noyes, Jr. , J. Am. Chem. Soc. , 85,1027

. Borrell, J. Am. Chem. Soc., 86, 3157 (1964).

. P. Borkowski and P. Ausloos, J. Phys. Chem., 66, 2257 (1961).
. Ausloos and R. E. Rebbert, J. Am. Chem. Soc., 86, 4513 (1964).
E. Rebbert and P. Ausloos, J. Am. Chem. Soc., 86, 4803 (1964).
. Ciamician and P. Silber, Chem. Ber., 88, 2911 (1900).

. Ciamician and P. Silber, Chem. Ber., 86, 1530 (1901).

1Paterno and G. Chieffi, Chem. Zentr. , I, 1161 (1909); II, 195
1909

. Paterno and G. Chieffi, Chem. Zentr., I, 552 (1911 ;
. Ciamician and P. Silber, Chem. Ber., 68, 194 (1915 .

F1 mm :l'l'l m G) W 'U 70 'U

. Paterno and G. Forli-Forti, Chem. Zentr., I, 552 (1911).

G. Ciamician and P. Silber, Chem. Ber. , 44,1280 (1911). Also
previous included references by authors.

W. 0. Cohen, Rec. Trav. Chim., 88, 243 (1920).
A. Schonberg and A. Mustafa, Chem. Rev., 68, 181 (1947).

C. Weizmann, E. Bergmann and Y. Hirshberg, J. Am. Chem. Soc., 68,
1530 (1938).

F. Bergmann and Y. Hirshberg, J. Am. Chem. Soc., 66, 1429 (1943).

J. N. Pitts, Jr. , R. L. Letsinger, R. P. Taylor, J. M. Patterson,
G. Recktenwald and R. B. Martin, J. Am. Chem. Soc. , 81_, 1068 (1959).

G. S. Hammond, W. P. Baker and W. M. Moore, J. Am. Chem. Soc., 88,
2795 (1961).

S. A. Weiner, J. Am. Chem. Soc., 88, 485 (1971).

68.
69.
70.

71.

72.

73.
74.
75.
76.

77.

78.
79.
80.
81.

82.
83.

84.

85.

86.

87.

88.
89.

132

P. J. Wagner, Tet. Let. #18, 1753 (1967).
P. J. Wagner, J. Am. Chem. Soc., 88, 5898 (1967).

P. J. Wagner and P. A. Kelso, Unpublished results, Michigan State
University, 1971.

Prof. P. J. Wagner, Personal Communication, Results of Preliminary
Work, Michigan State University, 1967-1971.

P. J.)Wagner and A. E. Kemppainen, J. Am. Chem. Soc., 81, 3085
1969 .

P. J. Wagner and G. Capen, Molecular Photochemistry, 1, 173 (1969).
Dr. H. N. Schott, Ph. D. Thesis, Michigan State University, 1970.
P. J. Wagner and H. N. Schott, J. Am. Chem. Soc., 81, 5383 (1969).

G. Porter and P. Suppan, Trans. Faraday Soc., 61, 1664 (1965).
Also previous references by authors included therein.

E. J. Baum, J. K. S. Wan and J. N. Pitts, Jr., J. Am. Chem. Soc.,
88, 2652 (1966).

N. C. Yang and R. L. Dusenbery, J. Am. Chem. Soc., 88, 5899 (1968).
. A. Lamola, J. Chem. Phys., 61, 4810 (1967).
R

. Kearns and W. A. Case, J. Am. Chem. Soc., 88, 5087 (1966).

20>

C. Yang, D. S. McClure, S. L. Murov, J. J. Houser and R.
Dusenbery, J. Am. Chem. Soc. , 89, 5466 (1967).

P. J. Wagner, Tet. Let. #52, 5385 (1968).

E. F. Zwicker, L. I. Grossweiner and N. C. Yang, J. Am. Chem. Soc.,
86, 2671 (1963).

A. E. Kemppainen, DuPont Technical Report (Monthly Summary for June
1960). pages 3-5 (July 1950).

J. E. Guillet, J. Dhanraj, F. J. Golemba and G. H. Hartly, Adv. in
Chem. Series, No. 85, Am. Chem. Soc. , Washington, D. C. (198 8).

S. G. Cohen, N. Stein and H. M. Chao, J. Am. Chem. Soc. , 90, 521
(1968).

6. Porter, Reactivity of the Photoexcited Molecule, Interscience
Publishers, New York (1967), p. 80.

 

See Reference 2, page 15 and pages 587 through 590.
F. D. Lewis and N. J. Turro, J. Am. Chem. Soc., 88, 311 (1970).

90.

91.

92.

93.
94.

95.
96.

97.
98.
99.

100.

101.

102.

103.
104.
105.

106.

107.

108.

109.

133

Dr. Irene E. Kochevar, Personal Communication, Michigan State
University, 1970.

F. Daniels, J. H. Mathews, J. W. Williams, P. Bender and R. A.
Alberty, Experimental Physical Chemistry, 5th Ed., McGraw-Hill,
New York (1956), p. 61.

S. G. Cohen and 8. Green, J. Am. Chem. Soc., 91, 6824 (1969);
S. G. Cohen and N. Stein, J. Am. Chem. Soc., 81, 3690 (1969).
F. D. Lewis and 1. A. Hilliard, J. Am. Chem. Soc., 22, 6672 (1970).

P. J. Wagner, P. A. Kelso, A. E. Kemppainen, A. Haug and D. R.
Graber, Molecular Photochemistry, 2, 81 (1970).

W. A. Pryor, Free Radicals, McGraw-Hill, New York (1966), p. 154.

 

R. O. C. Norman and R. Taylor, Electrophillic Substitution in
Benzenoid Compounds, Elsevier, New York (1965), p. 11.

 

C. Walling and M. J. Mintz, J. Am. Chem. Soc., 88, 1515 (1967).
A. Padwa, Tet. Let. #46, 3465 (1964).

E. M. Kosower, An Introduction to Physical Organic Chemistry, John
Wiley & Sons, New York (1968), pp. 49-50.

 

J. N. Pitts, D. R. Burley, J. C. Mani and A. D. Broadbent, J. Am.
Chem. Soc., 88, 5902 (1968).

C. R. Brundle, M. B. Robin, W. Jolley, P. Finn, R. Pearson and J.
Hollander, Chem. & Engr. News, 18, (#41), 44 (1970).

M. C. Caserio, W. Lauer and T. Novinson, J. Am. Chem. Soc., 88,
6082 (1970).

J. Guttenplan and S. G. Cohen, Chem. Comm., 247 (1969).

P. J. Wagner, Energy Transfer Kinetics in Solution, in press.

 

N. C. Yang, R. Loeschen and 0. Mitchell, J. Am. Chem. Soc., 88,
5465 (1957).

C. R. Noller, Chemistr of Dr anic Com ounds, W. B. Saunders Com-
pany, Philadelphia (1951), PP. 103, I96 6 522.

C. R. Houser, W. J. Humphlett and M. J. Weiss, J. Am. Chem. Soc.,
70, 426 (1948).

L. F. Fieser, Ex eriments in Or anic Chemistr , 3rd Ed., D. C.
Heath and Co., Boston (1957), Pp. 77-81; 269-270.

J. Cason and F. S. Prout, Organic Synthesis, Coll. Vol. III, p. 64.

 

110

111.
112.
113.
114.
115.
116.

117.
118.

119.
120.

121.

122.

123.

124.

134

. J. Cason and H. Rapoport, Laboratory Text in Organic Chemistpy,

 

2nd Ed., Prentice-Hall, Inc., New Jersey (1962), pp. 435-440.

L. Freidman and H. Schechter, J. Org. Chem., 86, 877 (1960).

H. R. Ward and P. D. Sherman, J. Am. Chem. Soc., 88, 3816 (1968).
F. Arndt, Organic Synthesis, Coll. Vol. II, p. 165.

 

Dr. David J. Buchek, Ph. D. Thesis, Michigan State University, 1969.
See Reference 52, p. 3216.

F. G. Moses, R. S. H. Liu and B. M. Monroe, Molecular Photochemis-
try, 1, 245 (1969).

See Reference 2, p. 732.

Personal communication, Dr. Richard Zepp, Postdoctoral Fellow,
Michigan State University, 1969-1971.

See Reference 91, p. 325.

Kimax Laboratory Glassware Catalog SC-300, April 1964, pp. 113 and
174.

D. D. McCracken, A Guide to Fortran Programming, John Wiley & Sons,
Inc., New York (1961).

 

L. G. Parrat, Probability and Experimental Errors in Science, John
Wiley & Sons, New York (1961), p. 129.

 

K. R.)Huffman, C. E. Kuhn and A. Zweig, J. Am. Chem. Soc., 88, 599
1970 .

R. M. Silverstein and G. C. Bassler, Spectrometric Identification
of Organic Compounds, John Wiley & Sons, New York (1966), pp. 56-57.

 

 

APPENDICES

APPENDIX A. PART 1. EXPERIMENTAL QUENCHING RUNS FOR DETERMINING

/o

 

 

 

 

 

, /
STERN-VOLMER DIAGRAMS FOR KETONES. @0054,
R = CHZCH3
Butyrophenone, 0.10M in benzene, 0.004M tetradecane standarda.
Run No. 1 Run No. 2
Quencherb Prod/StdC pp Quencher Prod/Std 29
Conc. M Ratio 9 Conc.(M) Ratio 9
0 0.831 1.00 0 0.797
0.001 0.523 1.59 Act 0.740
0.002 0.388 2.14
0.003 0.336 (2.48)9 Anal Cond Std Set I
0.004 0.254 3.28
Actd 0.82

L Sq Slope (o)e = 568 (3)
Intercept (o) = 1.01 (.01)

Anal Cond Std Set If

 

The following conditions apply to all quenching runs unless
specifically restated:

a An internal standard is used for analytical purposes, standard is
named in heading. See Experimental Procedures.

b The quencher used is 2,5-dimethy1-2,4-hexadiene unless otherwise

specified.
c The product to standard ratios given are averages of two or more

actual measurements. See Experimental Procedures.

d All actinometers are 0.10M valerophenone, 0.004M tetradecane in

benzene unless otherwise indicated. 611 is taken to be 0.33.

e The least squares slope, intercept, and standard deviations were
calculated on the CDC-6500 computer at the MSU computer center. See
Appendix B for computer program used.

f Analytical conditions are tabulated in the Experimental Section.
9 This point omitted because it deviated by twice the standard deviation.

135

136

 

 

R = CH(CH3)2
Isovalerophenone, 0.10M in benzene, 0.004M tetradecane standard.
Run No. 1 Run No. 2
Quencher Prod/Std 6_o Quencher Prod/Std 6_o
Conc. (M) Ratio 6 Conc. Ratio 6
0 1.23 1.00 0 0.868 1.00
0.002 0.817 1.50 0.002 0.579 1.49
0.004 0.625 1.99 0.004 0.452 1.91
0.006 0.495 2.53 0.006 0.340 2. 54
0.008 0.424 2.91 0.008 0.303 2.85
Act none 0 0.862
Act 0.758
0 a 0.565
Act 0.536 L Sq Slope 20) = 237 (9)
Intercept o) = 1.01 (.05)
L Sq Slope (o ) = 43 2(5)
Intercept (o ) = 2(. 02) Anal Cond Std Set I

Anal Cond Std Set 11

 

R = C(CH3)3

B,B-Dimethylbugyrgphenone, 0.10M in benzene, 0.004M tetradecane
standard.

 

 

Run No. 1 Run No. 2
Quencher Prod/Std 6, Quencher Prod/Std 6_O
Conc. (M) Ratio 6 Cone. (M) Ratio 6
0 0.587 1.00 0 0.942 1.00
0.010 0.330 1.81 0.010 0.589 1.60
0.020 0.229 2.62 0.020 0.388 2.42
0.030 0.165 3.62 0.030 0.298 3.16
0 0.609 0.040 0.234 4. 02
Act 1.05 0 0.938
Act 1.68
L Sq Slope (o ) = 2.3)
Intercept (o) = .04) L Sq Slope (a; = 76.0 (1.9
Intercept (o = 0.92 .05

Anal Cond Std Set I

a This actinometer only is 0.10M butyrophenone and 0.004M tetradecane

Anal Cond Std Set I

 

in benzene; 911 = 0.35.

137

 

 

 

 

R = CHZCHZCH3

Valergphenone, 0.10M in benzene, 0.004M tetradecane standard.

Run No. 1 Run No. 2

Quencher Prod/Std 6, Quencher Prod/Std 6o
Conc. (M) Ratio 6 Conc. (M) Ratio 9
0 0.99 1.00 0 0.833 1.00
0.010 0.729 1.36 0.010 0.604 1.40
0.020 0.557 1.77 0.020 0.465 1.81
0.030 0.443 2.24 0.030 0.375 2.25
0.040 0.382 2.59 0.040 0.318 2.65
0 1.052 O 0.853

Act none Act none

L Sq Slope (o) = 40.5 .7)
Intercept (0) = 0.98 (.02)

Anal Cond Std Set I

L Sq Slope (o) = 4 5 3
Intercept (o) = 0.99 (.0

Anal Cond Std Set I

 

R = CHZCHZCHZCHB

Hexanophenone, 0.10M in benzene, 0.004M tetradecane standard.

 

Run No. l

Quencher Prod/Std 60
Cone. M Ratio 6

0 1.76 1.00
0.010 1.245 1.40
0.020 0.970 1.80
0.030 0.813 2.14
0.040 0.670 2.60
0 1.71

Act 1.88

L Sq Slope (o) = 39.4 (.71)
Intercept (o) = 1 00 (.02)

Anal Cond Std Set I

 

Run No. 2

Quencher Prod/Std 6,
Conc. (M) Ratio 6
0 0.886 1.00
0.010 0.648 1.39
0.020 0.539 1.67
0.030 0.422 2.13
0.040 0.361 2.49
0 0.911

Act 0.99

L Sq Slope (o) = 37.2 (1.1)
Intercept (o) = 0.99 (.03)

Anal Cond Std Set I

138

 

 

 

 

R = CHZCH2 CH(CH3)2

6-Methylhexanophenone, 0.10M in benzene, 0.004M tetradecane standard.
Run No. 1 Run No. 2

Quencher Prod/Std 6, Quencher Prod/Std 6_o
Conc. (M) Ratio 6 Conc. (M) Ratio 6
0 1.44 1.00 0 1.20 1.00
0.010 1.08 1.32 0.010 0.953 1.25
0.020 0.917 1.55 0.020 0.779 1.52
0.030 0.779 1.83 0.030 0.654 1.82
0.040 0.672 2.12 0.040 0.592 2.01
0 1.41 0 1.18

Act 1.88 Act 1.63

L Sq Slope (o) = 27.5 0.5) L Sq Slope (o) = M5 9(0.7)
Intercept o) = .01) Intercept (o) = NO (.01)

Anal Cond Std Set I

Anal Cond Std Set I

 

R = CH CH ZC(CH3)3

2

6,6-Dimethylhexanophenone, 0.10M in benzene, 0.004M tetradecane standard.

 

 

 

Run No. 1 Run No. 2
Quencher Prod/Std 6o Quencher Prod/Std 6,
Cone. (M) Ratio 6 Conc. (M) Ratio 6
0 1.34 1.00 0 0.893 1.00
0.010 1.08 1.24 0.010 0.695 1.27
0.020 0.912 1.47 0.020 0.590 1.50
0.030 0.785 1. 71 0.030 0.510 1.74
0.040 0.680 1. 97 0 0.878
0 1.34 Act 1.19
Act 1.88

L Sq Slope a) = 24.5 0.4)
L Sq Slope (o) = 24.1 (0.2) Intercept o = 1.01 .01
Intercept (o) = 1. 00 (.005)

Anal Cond Std Set I
Anal Cond Std Set I

139

 

 

 

R = (CH2)6CH3
Nonanophenone, 0.10M in benzene, 0.004M tetradecane standard.
Run No. 1 Run No. 2
Quencher Prod/Std 6, Quencher Prod/Std 6o
Conc. M Ratio 6 Cone. (M) Ratio 6
0 1.48 1.00 0 0.753 1.00
0.010 1.17 1.265 0.010 0.553 1.36
0.020 0.912 1.60 0.020 0.448 1.68
0.030 0.754 1.93 0.040 0.320 2.36
0.040 0.687 2.13 0 0.751
0 1.44 Act 1.05
Act 1.88
L Sq Slope (o) = 33.8 (0.3)
Intercept (o) = 1.01 (.01)

L Sq Slope (a) = 29.3 (1.0)
Intercept o) = 1.00 .006)

Anal Cond Std Set I
Anal Cond Std Set I

 

R = (CH2)12C”3 R = CHZCH(CH3)2

Pentadecanophenone, 0.10M in
benzeng, 0.004M tetradecane

y-Methylvalerophenone, 0.10M in
benzene,,0.004M tetradecane

 

 

 

 

 

standard. standard.
Quencher Prod/Std 60 Run No. 1
Cone. (M) Ratio 6
Quencher Prod/Std 6o
0 1.69 1.00 Conc. M Ratio 6
0.020 1.06 1.56
0.040 0.783 2.11 0 3.73 1.00
0.060 0.630 2.62 0.020 2.95 1.19
0.080 0.508 3.24 0.060 2.16 1.62
O 1.61 0.080 1.89 1.86
Act 2.07 0 a 3.27
Act 4.85
L Sq Slope (o) = 27.7 (0.3)
Intercept (o) = 1.00 (.02) L Sq Slope (a) = 10.75 (.2)
Intercept (o = 0.99 (.02)

Anal Cond Std Set I
Anal Cond Std Set III

 

a This actinometer only was 0.10M butyrophenone, 0.004M tetradecane
standard, 6II = 0.35.

140

 

 

R = CHZCH(CH3)2
I-MethylvalerOphenone, 0.10M in benzene, 0.004M tetradecane standard.
Run No. 2 Run No. 3a
Quencher Prod/Std 6o Quencher Prod/Std 60
Cone. M Ratio 6 Conc.(M) Ratio 6
O 0.762 1.00 0 0.241 1.00
0.020 0.616 1.24 0.020 0.196 1.22
0.040 0.557 1.37 0.040 0.173 1.38
0.060 0.478 1.59 0.060 0.149 1. 61
Act 0.991 0. 080 0.130 1. 84
0 b 0.236
0 0.367 Act 0.646
Act 0.463
L Sq Slope (o) = 10.35 (0.3)
L Sq Slope (a) = 9.5 (0.5) Intercept (o) = 1.00 (.01)
Intercept o = 1.02 (.02)

Anal Cond Std Set III
Anal Cond Std Set III

 

=CH ZCHZ4<::>

y-Phenylbutyrophenone, 0.10M in benzene,,0.004M tetradecane standard.

 

 

Quencher Prod/Std 99 6II versus per cent conversion.
Conc. M Ratio 6 Prod/Std %
- 6
o 0.731 1.00 Rat1o Conv. Act II
8'833 8'23? ('2; 0.215 1.72 0.144 0.492
0 060 0'418 1 74 0.395 3.16 0.275 0.474
0' 0'722 0.585 4.68 0.416 0.464
Act 0'515 0.722 5.77 0.516 0.462
° 0.872 6.97 0.644 0.448
1.07 8.55 0.795 0.444
L Sq 51°99 )°( = ‘2'3 '4) 1 18 9 45 0 904 0.431
Intercept o = 0.98 .015) 1.38 11 1 1.08 0.422

Anal Cond Std Set I

a Standard concentration was 0.008M for this run.

b

Anal Cond Std Set I

 

This actinometer only was 0.10M butyrophenone, 0.004M tetradecane
standard, 611 = 0.35.

141

R = CHZCHZCH=CH2

y-Vinylbutyrpphenone, 0.10M in benzene, 0.004M tetradecane standard.

Run No. 1 (2.91% conv.) Run No. 2 (5.82% conv.)

 

Quencher Prod/Std 6, Quencher Prod/Std 6,
Conc. M Ratio 6 Cone. M Ratio 6
0 0.352 1.00 0 0.729 1.00
0.010 0.333 1.10 0.010 0.645 1.12
0.020 0.305 1.20 0.020 0.622 1.17
0.030 0.279 1.31 0.030 0.581 1.25
0.040 0.263 1.39 0.040 0.527 1.38
0 0.376 0 0.724

Act 0.490 Act 1.05

L Sq Slope (o) = 9.9 (.2) L Sq Slope (a) = 8.9 (.6)
Intercept (o) = 1.00 (.00) Intercept (o) = 1.01 (.01)

Anal Cond Std Set I

Anal Cond Std Set I

 

Run No. 3 (12.1% conv.)

Quencher Prod/Std 6,
Conc. (M) Ratio 6

 

0 1.52 1.00
0.010 1.40 1.08
0.020 1.37 1.10
0.030 1.26 1.20
0.040 1.17 1.29
0 1.50

Act 2.40

L Sq Slope (a) = 7.0 (.6)
Intercept o = 0.99 (.01)

Anal Cond Std Set I

 

142

R = CHZCH2N(CH3)2

1;Dimethy1aminobutyrophenone, 0.10M in benzene, 0.004M tetradecane

 

 

 

 

 

standard.
Run No. 1 Run No. 2a
Quencher Prod/Std 6_o Quencher Prod/Std 60
Conc. M Ratio 6 Conc. M Ratio 6
0 0.310 1.00 0 0.446 1.00
0.20 0.276 1.12 > 0.40 0.358 1.24
0.60 0.241 1.29 0.80 0.298 1.49
1.00 0.191 1.62 1.20 0.259 1.72
Act 3.69 0 0.442
Act 5.66
L Sq Slope (o) = 0.60 (.04
Intercept (o = 0.99 (.03 L Sq Slope (o) = 0.60 (.005)
Intercept (o) = 1.00 (.003)
Anal Cond Std Set II
Anal Cond Std Set II
Run No. 3b Run No. 4C
Quencher Prod/Std 6, Quencher Prod/Std 6,
Conc. (M) Ratio 6 Conc. M Ratio 6
0 0.831 1.00 0 0.428 1.00
0.20 0.762 1.08 1.00 0.282 1.53
0.40 0.606 1.35 0 0.434
0.80 0.534 1.55 Act 5.63
1.20 0.444 1.85
0 0.806
Act 5.66 Slope = 0.53
L Sq S10pe a) = 0.71 (.04 Anal Cond Std Set II
Intercept o = 0.99 .03

Anal Cond Std Set II

 

a The quencher used in this run was piperylene.

b The internal standard in this run was 0.002M hexadecane.

An impurity

in the high concentration of 2.5-dimethyl-2,4-hexadiene interfered

with the dteradecane during analysis.

C This run had a single quenched sample using biphenyl as a quencher.

143

R = CH CHZOCH

 

 

2 3
y:Methoxybutyrophenone, 0.10M in benzeng,p0.004M tetradecane standard.
Run No. 1 Run No. 2
Quencher Prod/Std 6, Quencher Prod/Std 6_o
Conc. M Ratio 6 Conc. M Ratio 6
0 0.737 1.00 0 0.814 1.00
0.020 0.585 1.23 0.020 0.664 1.19
0.040 0.554 1.30 0.040 0.609 1.30
0.060 0.458 1.59 0.060 0.543 1.45
0.080 0.438 1.65 0.080 0.491 1.61
0 0.706 0 0.840
Act 1.01 Act 1.19
L Sq Slope (o) = 8.3 (.75) L Sq Slope (o) = 7.4 (.24)
Intercept (o) = 1.02 (.04) Intercept (a) = 1.01 (.01)

Anal Cond Std Set I Anal Cond Std Set I

 

R = CHZCH(0CH3)CH3

y-Methoxyvalerophenone, 0.10M in benzeng, 0.004M tetradecane standard.

Run No. 1

Quencher Prod/Std 6o
Conc. (M) Ratio 6
0 0.842 1.00
0.020 0.711 1.19
0.040 0.600 1.41
0.060 0.537 1.58
0.080 0.455 1.86
0 0.852

Act 1.44

L Sq Slope (o) = 10.5 (.4)
Intercept (o) = 0.99 (.02)

Anal Cond Std Set I

144

 

 

R = CHZCHZOH

y-Hydroxybutyrophenone, 0.10M in benzene, 0.004M tetradecane standard.
Run No. 1 Run No. 2

Quencher Prod/Std 6, Quencher Prod/Std 60
Cone. M Ratio 6 Conc. M Ratio 6
0 1.33 1.00 0 1.295 1.00
0.010 1.15 1.15 0.020 1.055 1.25
0.020 1.00 1.32 0.030 0.957 1.38
0.030 0.94 1.41 0.040 0.886 1.49
0.040 0.85 1.56 0.050 0.824 1.61
0 1.31 0 1.335

Act 1.24 Act 1.63

L Sq Slope (a) = 13.8 (.5) L Sq Slope (o) = 12.2 (.1)
Intercept o = 1.01 .01) Intercept (o) = 1.00 (.004)

Anal Cond Std Set I Anal Cond Std Set I

 

 

R = CHZCH2C1
y;Ch10robutyrophenone, 0.10M in benzene, 0.004M tetradecane standard.
Run No. 1 Run No. 2
Quencher Prod/Std 6_o Quencher Prod/Std 6,
Conc. Ratio 6 Conc. M Ratio 6
0 0.252 1.00 0 0.278 1.00
0.002 0.183 1.34 0.002 0.210 1.30
0.004 0.142 1.72 0.004 0.154 1.76
0.006 0.119 2.05 0.006 0.133 2.04
0 a 0.237 0.008 0.114 2.38
Act 1.01 0.008 0.108 2.52
0 a 0.266
L Sq Slope (o) = 176.5 (2.3) Act 1.01
Intercept o) = 1.00 (.01)
L Sq Slope (a) = 183 (7)
Anal Cond Std Set III Intercept (o = 0.98 (.04)

 

 

Anal Cond Std Set III

 

a These actinometers only are 0.10M butyrophenone, 0.004M tetradecane

in benzene;

4’11

= 0.35.

145

R = CHZCHZCDOCH3

y-Carbomethogybutyrophenone, 0.10M in benzene, 0.004M tetradecane
standard.

Run No. 1 (9.2% conv.) Run No. 2a (2.10% conv.)

 

Quencher Prod/Std 6_o Quencher Prod/Std 6o

Conc. Ratio 6 Conc. (M) Ratio 6

0 1.15 1.00 0 1.05 1.00

0.002 0.680 1.69 0.001 0.721 1.46

0.004 0.478 2.40 0.002 0.569 1.85

0.006 0.366 3.14 0.003 0.466 2.25

0.008 0.289 3.98 0.004 0.362 2.90

0.008 0.290 3.96 0 1.05

0 1.15 Act 0.78

0 1.16

Act 0.97 L Sq Slope (a) = 459 (21)
Intercept (o = 0.97 (.05)

L Sq Slope (o) = 373 (5)

Intercept (o) = 0.96 (.03) Anal Cond Std Set 111

Anal Cond Std Set III

 

b

Run No. 3 Run No. 46

6II versus per cent conversion 6II versus per cent conversion

 

 

Prod/Std % Prod/Std %

 

 

 

 

Ratio Conv. Act ¢II Ratio Conv. Act ¢II
2.10 4.20 0.777 0.444 1.05 2.10 0.777 0.444
3.55 7.10 1.57 0.375 1.85 3.70 1.57 1 0.389
7.12 14.24 3.19 0.369 3.70 7.40 3.19 0.384
9.48 18.96 4.69 0.334 4.92 9.84 4.69 0.347
11.39 22.78 6.17 0.305 Anal Cond Std Set III

Anal Cond Std Set III

a Tetradecane concentration 0.001M.

b

 

Ketone concentration 0.050M and tetradecane concentration 0.0005M.

146

R = CHZCHZCN

y-Cyanobutyrophenone, 0.10M in benzene, 0.004M tetradecane standard.

Run No. 1 (3.6% conv.) Run No. 2 (4.84% conv.)

 

Quencher Prod/Std 6, Quencher Prod/Std 6o

Conc. M Ratio 6 Conc. M Ratio 9L

0 0.437 1.00 0 0.598 1.00

0.001 0.305 1.46 0.001 0.418 1.44

0.002 0.203 2.19 0.002 0.340 .78

0.003 0.190 2.35 0.003 0.260 2.32

0.004 0.152 2.93 0.004 0.216 2.79

0.008 0.081 5.50 0 0.608

0 0.440 Act 1.25

0 0.457

Act 1.01 L Sq Slope (o) = 446 (14)
Intercept (o) = 0.97 (.03)

L Sq Slope (a) = 558 (25)

Intercept o = 0.90 (.10) Anal Cond Std Set I

Anal Cond Std Set I

 

Run No. 3 (2.02% conv.) Run No. 4

 

 

 

Quencher Prod/Std 9w 611 versus per cent conversion
Conc. (M) Ratio 6
Prod/Std %
0 0.252 1.00 . ¢
0,001 0.150 1.68 Rat1o Conv. Act II
0.002 0.105 2.40
0,003 0.078 3.20 0.187 1.50 0.252 0.245
ACt 0°373 0.317 2.53 0.505 0.207
L Sq Slope (o) = 732 (14)
Intercept (O) = 0.97 (.03) 0.437 3.50 0.736 0.196
Ana] Cond Std Set I 0.538 4.30 1.02 0.174
Run No. 1a 0.145
Run No. 2 0.159
Run No. 3 0.223

Anal Cond Std Set I

 

a The actinometer value in this run is suspected of being slightly

large.

147

R = cuzcuzcnzco©

1,4-Dibenzoy1butang, 0.10M in benzene, 0.004M tetradecane standard.

 

 

Run No. 1 Run No. 2

Quencher Prod/Std 6, Quencher Prod/Std 6,
Cone. (M) Ratio 6 Conc. M Ratio 6
0 1.35 1.00 0 1.21 1.00
0.010 0.739 1.81 0.010 0.860 1.40
0.020 0.673 1.99 0.020 0.606 1.98
0.030 0.522 2.57 0.030 0.463 2.59
0.040 0.417 3.20 0.040 0.367 3.30
0 1.33 0 1.19

Act 1.28 Act 1.08

0 1.50a L Sq Slope (o) = 57 .9 (2.5)
Act 1.60 Intercept (o) = 0. 90 (.06)
L Sq Slope (o) = 51.6 (3.8) Anal Conc Std Set I
Intercept (o) = 1.08 (.09)

Anal Conc Std Set I

 

R= CHZCHZCH2 3

a-Carbomethoxyvalerophenone, 0.10M in benzene,y0.004M tetradecane
standard.

COOCH

 

 

Run No. 1 Run No. 2b

Quencher Prod/Std 6, Quencher Prod/Std 6,
Conc. (M) Ratio 6 Conc. M Ratio 6
0 1.87 1.00 0 1.25 1.00
0.002 1.49 1.26 0.001 1.15 1.09
0.004 1.27 1.48 0.002 1.04 1.21
0.006 1.03 1.825 0.003 0.93 1.35
0.008 0.921 2.04 0.004 0.81 1.54
0.008 0.947 1.99 Act 0.659

0 1.89

0 1.88 L Sq Slope (o) = 134 (8)
Act 0.97 Intercept (o) = 0.97 (.02)
L Sq Slope (o) = 128.5 (4.2) Anal Cond Std Set III

Intercept (0) =1.00 (.01)
Anal Cond Std Set III

 

a Separate determination.

b Tetradecane concentration 0.001M, irradiated 15 minutes.

148

 

 

 

R= CHZCHZCHZCOOH
6-Carb05yvalerophenone, 0.10M in benzene, 0.004M tetradecane standard.
Run No. 1a Run No. 2
Quencher Prod/Std 6, Quencher Prod/Std 6_o
Conc. M Ratio 6 Cone. (M) Ratio 6
0 0.528 1.00 0 2.55 1.00
0.002 0.375 1.39 0.002 1.88 1.36
0.004 0.305 1.71 0.004 1.44 1. 78
0.006 0.258 2.02 0.006 1.21 2.12
0.008 0.206 2. 53 0.008 1.01 2. 53
0.008 0.207 2. 51 0 2.57
0 b 0.514 Act 1.50
Act 0.646

L Sq Slope (o) = 191 (2)
L Sq Slope (a 0) =6) Intercept (o) = 0 99 ( 01)
Intercept ) = 9( 03)

Anal Cond Std Set I
Anal Cond Std Set III

 

R = CHZCHZCHZC1

a-Chlorovalerophenone, 0.10M in benzene, 0.004M tetradecane standard.

Run No. 1 Run No. 2

Quencher Prod/Std 6, Quencher Prod/Std 6,
Cone. M Ratio 6 Conc. (M) Ratio 6
0 3.44 1.00 0 2.52 1.00
0.010 1.12 3.07 0.002 1.75 1.44
0.020 0.627 5.48 0.004 1.35 1.87
0.030 0.428 8.04 0.006 1.06 2.38
0.040 0.316 (10. 90) 0.008 0.90 2.80
0 3. 43 0 2.61

Act 1. 98 Act 1.44

L Sq Slope (o) = 9(5) L Sq Slope (o) = 27 (3)0
Intercept (o) = 9(.02) Intercept (a) = M9 (.0 1)

Anal Cond Std Set I

 

Anal Cond Std Set I

 

a Tetradecane concentration of 0.008M.

b This actinometer only was butyrophenone, 0.10M in benzene, 0.004M
tetradecane standard, 6II = 0.35.

d-annovalerophenone, 0.10M in benzene, 0.004M tetradecane standard.

 

 

 

R = CHZCHZCHZCN

Run No. 1

Quencher Prod/Std 6_o
Conc.(fl) Ratio 6
0 1.785 1.00
0.010 0.274 6.50
0.020 0.153 11.6
0 1.765

Act 1.25

L Sq Slope (a) = 530 (7)
Intercept o = 1.07 (.09)

Anal Cond Std Set I

 

Run No. 2

Quencher Prod/Std 6_O
Conc. (M) Ratio 6
0 1.69 1.00
0.001 1.15 1.47
0.002 0.905 1.87
0.003 0.678 2.49
0.004 0.563 3.00
0 1.70

Act 1.23

L Sq Slope (o) = 502 (16)
Intercept (o) = 0.96 (.04)

Anal Cond Std Set I

 

R = CHZCHZCHZCHZCT

e-Chlorohexanophenone, 0.10M in

 

benzene, 0.004M tetradecane
standard.

R = CHZCHZCHZCHZCN

e-Cyanohexanophenone,,0.10M
benzene, 0.004M tetradecane

 

in

 

 

 

Quencher Prod/Std 6,
Conc. (M) Ratio 6
0 2.02 1.00
0.002 1.71 1.17
0.004 1.47 1.35
0.006 1.29 1.54
0.008 1.17 1.70
0 1.96

Act 1.50

L Sq Slope (a) = 88.5 (1)
Intercept (o) = 1.00 (.005)

Anal Cond Std Set I

L Sq Slope (o)
Intercept (o)

74.5 (2.6
1.01 ( 1

Anal Cond Std Set I

 

93.98M-

Quencher Prod/Std 6,
Conc.(M) Ratio 6

0 1.13 1.00
0.002 0.963 1.15
0.004 0.848 1.31
0.006 0.753 1.48
0.008 0.704 1.58
0 1.09

Act 1.50

150

= CH-(©-CH

B-Phenylbutyrophenone, 0.10M in benzene, 0.004M tetradecane standard.

 

Run No. 1 (Irradiated 36 hrs) Run No. 2

Quencher Prod/Std 6_o Quencher Prod/Std 6,
Conc. M Ratio 6 Conc. M Ratio 6
0 0.188 1.00 0 0.162

0.10 0.118 1.55 Act 11.7

0.20 0.0917 2.00 .

0.30 0.0759 2.42 Anal Cond Std Set I

0.30 0.0729 2.52

0.40 0.0593 3.08

Act 0.89/hour

L Sq Slope (o ) = 5.04 (.15)

Intercept (o) = 1.00 (.04)

Anal Cond Std Set I

 

 

 

 

R = OCH3
a-Methoxyacetoppenone, 0.10M in benzene, 0.001M tetradecane standard.
Run No. 1 Run No. 2a
Quencher Prod/Std 6, Quencher Prod/Std 6,
Conc. M Ratio 6 Conc. (M) Ratio 6
0 5.27 1.00 0 0.881 1.00
0.040 4.65 1.133 0.040 0.763 1.166
0.080 4.02 1.31 0.080 0.716 1. 24
0.120 3.81 1.38 0.120 0.667 1. 333
0.160 3.56 1.48 0.160 0.640 1. 39
Act 0.794 0 b 0.897
Act 0.685
L Sq Slope (o )= 2(. 2)
Intercept (o ) = 2(. 02) L Sq Slope (a) = 2. 38 (.21)
Intercept (o = L 04 (.02

Anal Cond Std Set III
Anal Cond Std Set III

 

a Tetradecane standard concentration 0.005M.

b Actinometers are 0.10M butyrophenone, 0.004M tetradecane standard in

benzene; 6II = 0.35.

151

R = OCH3

a-Methoxyacetophenone,(0.10M in benzene, 0.004M tetradecane standard.

Run No. 3 (Irradiated 1 hr) Run No. 4 (Irradiated 2 hrs)

 

 

 

 

Quencher Prod/Std 6_o Quencher Prod/Std 6_o
Conc._(M) Ratio 6 Conc. (M) Ratio 6
0 0.715 1.00 0 1.38 1.00
0.020 0.660 1.07 0.020 1.19 1.13
0.040 0.606 1.17 0.040 1.09 1.23
0.060 0.569 1.24 0.060 1.02 1.32
0.080 0.559 1.27 0.080 1.00 1.34
0.10 0.525 1.35 0.10 0.942 1.42
0.20 0.470 1.51 0.20 0.792 1.69
0.30 0.402 1.76 0.30 0.716 1.87
0.40 0.338 2.09 0.40 0.622 2.15
0 a 0.698 0 0.130

Act 0.433 Act none

Anal Cond Std Set III Anal Cond Std Set 111

 

Run No. 5 (Irradiated 3 hrs) Initial (0.020 to 0.080M

quencher) and Final (0.10 to

 

 

 

Quencher Prod/Std 6_o 0.40M quencher) Slopes at
Conc. (M) Ratio 6 various Irradiation Times.
0 2.02 1.00 L Sq Slope Intercept
0.020 1.64 1.20
0.040 1.67 1.18 (1 hour) (0) o
0.060 1.50 1.31 Initial: 3.55 (.26) l.01(.01)
0.080 1.32 1.49 Final: 2.59 (.13) l 02(.03)
0.10 1.29 1.53
0.20 1.17 1.68 (2 hours)
0.30 1.04 1.89 Initial: 4.35 (.45) l.03(.02)
0.40 0.926 2.13 Final: 2.75 (.18) 1.08(.04)
0 1.91
Act none (3 hours)

Initial: 5.45 .75 1.02 .04)
Anal Cond Std Set III Final: 2.62 .28 1.12 .07

 

a Butyrophenone actinometer, 0.10M in benzene, 0.004M tetradecane

standard; 611 = 0.35.

152

APPENDIX A. PART 2. EXPERIMENTAL QUENCHING RUNS FOR DETERMINING

STERN- VOLMER DIAGRAMS FOR KETONES: C\0
Mz:3>/ CH ZCHZCHZCH3
R = O‘CF3

ortho-Trifluoromethylvalerophenone, 0.10M in benzene,(0.004M tetra-
decane standard.

 

 

Run No. 1 Run No. 2

Quencher Prod/Std 6, Quencher Prod/Std 6o
Conc. M Ratio _6 Conc. (M) Ratio 6
0 0.556 1.00 0 0.859 1.00
0.020 0.297 1.86 0.020 0.463 1.85
0.040 0.215 2.58 0.040 0.328 2.61
0.060 0.172 3.22 0.060 0.255 2.61
0.080 0.136 4.08 0.080 0.210 4.08
0 0.554 0 0.854

Act 0.89 Act 1.29

L Sq Slope (o ) = 6(. 8) L Sq Slope (o ) = .5)
Intercept (o ) = 4(. 04) Intercept (o) = .03)

Anal Cond Std Set 11

Anal Cond Std Set 11

 

R = m-CF3

meta-Trifluoromethijalerophenone, 0.10M in benzene, 0.004M tetradecane
standarHI

 

 

Run No. 1 Run No. 2

Quencher Prod/Std 6, Quencher Prod/Std 6,
Conc. (M) Ratio 6 Conc. (M) Ratio 6
0 0.637 1.00 0 1.03 1.00
0.020 0.500 1.28 0.020 0.79 1.29
0.040 0.396 1.61 0.040 0.64 1.59
0.060 0.330 1.93 0.060 0.547 1.87
0.080 0.286 2.23 0.080 0.452 2.26
0 0.638 0 1.00

Act 0.89 Act 1.29

L Sq Slope (o) = 15 .6 .2) L Sq Slope (o) = .4)
Intercept (o) = 0. 99 .01) Intercept (o) = .02)

Anal Cond Std Set II

Anal Cond Std Set II

153

R = p-CF3

para-Trifluoromethylvalerophenone, 0.10M in benzene..0.004M tetra-
decane standard.

 

 

 

Run No. 1 Run No. 2

Quencher Prod/Std 99 Quencher Prod/Std $9
Conc. Ratio ¢ Conc. Ratio 4
0 0.724 1.00 0 1.08 1.00
0.020 0.546 1.315 0.020 0.797 1.36
0.040 0.436 1.66 0.040 0.634 1.71
0.060 0.356 2.02 0.060 0.513 2.11
0.080 0.306 2.35 0.080 0.427 2.53
0 0.712 0 1.09

Act 0.89 Act 1.23

L Sq Slope (o) = 0(. 2) L Sq SlOpe (o) = 19 9.0 (.3)
Intercept (o) = 9(. 01) Intercept (o) = 0.998 (.02)

Anal Cond Std Set II

Anal Cond Std Set I

 

ortho-F1uorovalerophenonel(0.10M in benzene, 0.004M tetradecane

 

 

R = o-F

standard.

Run No. l

Quencher Prod/Std .gp
Conc. M Ratio o
0 1.40 1.00
0.002 1.27 1.07
0.004 1.20 1.13
0.008 1.06 1.28
0 1.33

Act 1.32

L Sq Slope (o) = 4.9 (.7)
Intercept (o) = .00 (.00)

Anal Cond Special Set Ih

 

Run No. 2

Quencher Prod/Std $9
Conc. M Ratio ¢
0 1.37 1 00
0.010 1.02 1.34
0.020 0.82 1.67
0.030 0.650 1 2.10
0.040 0.585 2 34
0 1.38

Act 1.37

L Sq Slope (o) = 34.4 (1.1)
Intercept (o) = .00 (.03)

Anal Cond Special Set Ih

R = m-F

meta-Fluorovalerophenone, 0.10M in benzene, 0.004M tetradecane standard.

 

 

Run No. 1

Quencher Prod/Std 99
Cone. (M) Ratio o
0 1.12 1.00
0.010 0.848 1.32
0.040 0.538 2.08
0 1.12

Act 1.32

L Sq Slope (o) = 26.6 E. )
0) =1.02

Intercept (
Anal Cond Std Set I

R = p-F

para-Fluorovalerophenone, 0.10M in benzene, 0.004M tetradecane standard.

.02)

 

Run No. 2

Quencher Prod/Std $9
Conc. (M) Ratio ¢
0 1.07 1.00
0.010 0.828 1.29
0.020 0.677 1.58
0.030 0.544 1.97
0.040 0.492 2.17
0 1.06

Act 1.37

L Sq Slope (o) = 30.2 (1.1)
Intercept (o) = 1.00 (.03)

Anal Cond Std Set I

 

 

 

 

Run No. 1

Quencher Prod/Std $9
Conc. (M) Ratio o

0 1.15 1.00
0.002 1.06 1.08
0.004 1.00 1.15
0.008 0.91 1.27
0 1.16

Act 1.32

L Sq Slope E0; = 33.4 (1.2

Intercept o = 1.01 .05

Anal Cond Std Set I

 

Run No. 2

Quencher Prod/Std g9
Conc. (M) Ratio ¢
0 1.02 1.00
0.010 0.754 1.37
0.020 0.592 1.74
0.030 0.523 1.97
0.040 0.422 2.44
0 1.03

Act 1.23

L Sq Slope (a) = 34.8 (1.4)
Intercept (o) = 1.01 (.03)

Anal Cond Std Set I

 

155

R = o-C1

ortho-Chlorovalerophenone, 0.10M in benzene, 0.002M 0ctadecane standard.

 

 

 

Run No. 1 Run No. 2
Quencher Prod/Std 99 Quencher Prod/Std 39
Cone. (MD Ratio o Conc. (M) Ratio o
0 2.11 1.00 0 3.47 1.00
0.002 1.66 1.27 0.002 2.69 1.34
0.004 1.32 1.60 0.004 2.25 1.54
0.006 1.13 1.87 0.006 1.94 1.79
0.008 0.96 2.20 0.008 1.66 2.09
0 2.10 Act 1.46
Act 0.792

L Sq Slope (o) =
L Sq SlOpe £0; = 150 (2) Intercept (o) = P03 (. 02)
Intercept o = 0.99 (.01)

Anal Cond Special Set IIc
Anal Cond Special Set IIc

 

 

 

Run No. 3a
Quencher Prod/Std $9
Conc.¥(M) Ratio 9
0 1.00 1.00
0.002 0.768 1.30
0.004 0.620 1.61
0.006 0.536 1.87
0.008 0.466 2.15
0 1.00

- Act 0.78
L Sq Slope (o) = 2)
Intercept (o) = l(. 01)

Anal Cond Special Set IIb

 

a Standard used in this run was 0.004M hexadecane.

R = m-Cl

meta-Chlorovalerophenone, 0.10M in benzene, 0.002M 0ctadecane standard.

Run No. 1

Quencher Prod/Std 99
Conc. (M) Ratio 9
0 2.02 1.00
0.002 1.89 1.07
0.004 1.74 1.16
0.006 1.66 1.22
0.008 1.59 1.27
0 2.02

Act 1.46

L Sq Slope (a) = 34.5 1.6)
Intercept o = 1.01 .01

Anal Cond Special Set IIc

Run No. 2a

Quencher

Gone.

0
0.004
0.008
0.012
0.016
0
Act

L Sq Slope
Intercept

131

Prod/Std $9

Ratio

0.519
0.467
0.425
0.387
0.358
0.542
0.78

__$L___

1.00
1.14
1.25
1.37
1.48

29.8 (.6)
1.01 .01)

Anal Cond Special Set IIb

 

 

 

 

R = p-Cl

para-Chlorovalerophenone, 0.10M in benzene, 0.002M 0ctadecane standard.
Run No. 1 Run No. 2a

Quencher Prod/Std 99 Quencher Prod/Std $9
Conc. (M) Ratio g Conc. (M) Ratio 9

0 2.27 1.00 0 0.636 1.00
0.002 1.80 1.26 0.002 0.499 1.26
0.004 1.47 1.54 0.004 0.415 1.53
0.006 1.24 1.83 0.006 0.351 1.79
0.008 1.08 2.10 0 0.619

Act 1.46 Act 0.78

L Sq Slope (a) = 138 (1) L Sq Slope (a) = 132 (l)
Intercept (o) = 0.99 (.01) Intercept (o) = 1.00 (.002)

Anal Cond Special Set IIc

Anal Cond Special Set IIb

 

a Standard used in these runs was 0.004M hexadecane.

157

 

 

 

R = m-CH3

meta-Methylvalerophenone, 0.10M in benzene,)0.004M pentadecane standard.
Run No. 1a Run No. 2

Quencher Prod/Std 99 Quencher Prod/Std 39
Conc. (M) Ratio 9 Conc. (M) Ratio (g
0 0.795 1.00 0 1.43 1.00
0.002 0.659 1.20 0.002 1.13 1.25
0.004 0.551 1.44 0.004 0.934 1.51
0.006 0.505 1.59 0.006 0.810 1.74
0.008 0.440 1.80 0.008 0.704 2.00
0 0.788 0 1.39

Act 1.04 Act 1.41

L Sq Slope (o ) = 99 .5 (2.7) L Sq Slope (o) = 124 (1)
Intercept (o ) = l. 01 (.01) Intercept (o) = 1.00 (.005)

Anal Cond Special Set IIa

Anal Cond Special Set Ia

 

 

 

Run No. 3b Run No. 4

Quencher Prod/Std $9 Quencher Prod/Std $9

Conc. (M), Ratio 9 Conc. Ratio o

0 1.435 1.00 0 1.20 1.00

0.002 1.09 1.33 0.010 0.505 2.30

0.004 0.936 1.55 0.020 0.324 3.58

0.006 0.802 1.81 0 1.13

0.008 0.698 2.08 Act 1.27

0 1.455

Act 1.27 L Sq Slope (a) = 129 (1)
Intercept (o) = 1 .00 0(. 00)

L Sq Slope (a) = 132 (3)

Intercept o = 1.03 (.02) Anal Cond Special Set Ia

Anal Cond Special Set Ia

 

a The slope and quantum yield in this run are suspected of being
slightly low.

b The quantum yield in this run is much too high, possibly due to an

erroneous quencher concentration. This would not negate the
quenching study.

158

 

 

R = p-CH3

para-Methylvalergphenone, 0.10M in benzene, 0.004M pentadecane standard.
Run No. 1 Run No. 2

Quencher Prod/Std 3p Quencher Prod/Std 3p
Conc. M Ratio 9 Cone. (M)_ Ratio ¢
0 1.11 1.00 0 1.52 1.00
0.002 0.722 1.55 0.001 1.19 1.26
0.004 0.545 2.06 0.002 0.971 1.55
0.006 0.432 2.60 0.003 0.822 1.83
0.008 0.350 3.20 0.004 0.725 2.07
0 1.12 0 1.48

Act 1.04 Act 1.41

L Sq Slope (o) = 273 (3) L Sq Slope (o) = 271 (4)
Intercept (o) = 0.99 (.02) Intercept (a) = 1.00 (.01)

Anal Cond Special Set IIa

Anal Cond Special Set 118

 

R = P©S Y'CH3

para-Phenyl-y-methy1va1erophenone, 0.10M in benzene,(0.00lM eicosane

standard.

Quencher Prod/Std 99

Cone. M Ratio 9

Quencher Prod/Std $9

Conc. M Ratio 9

0 0.249 1.00 (0.0125M ketone, 0.0005M std)
0.005 0.884 0 275

0.010 1.00 0.243 0 0.393

0.020 0.09 0.223 0 0.439

0 0.238 0.10M szha 0.350

Act 81.3

1.0M t-BuOH 0.200

0.10M szhb 0 340
0.05M Trph 0.346
0.05M Trph 0.361

(0.05M ketone, 0.002M std) Act 81.3
0 0.113
0 0.118 Anal Cond Special Set IIIc

(0.025M ketone, 0.001M std)
0 0.214

0.10M szh 0.186

0.05M Trph 0.201

Act 81.3

0 0.114
0.10M szh 0.097
0.05M Trph 0.094
Act 35.9

 

a Ben20phenone

b Triphenylene

R = p-OCH3

para-Methoxyvalerophenone, 0.10M in benzene,)0.002M 0ctadecane

 

 

 

 

standard.

Run No. 1 Run No. 2

Quencher Prod/Std 99 Quencher Prod/Std $9
Conc. M Ratio 0 Conc. (M) Ratio 0
0 2.49 1.00 0 4.16 1.00
0.0005 1.15 2.14 0.0002 2.81 1.47
0.0010 0.805 3.06 0.0004 2.14 1.93
0.0015 0.564 4. 32 0.0006 1.82 2. 27
0.0020 0.483 5. 68 0.0008 1.48 2. 79
0 2.43 0 4.10

Act 3.85 Act 6.50

L Sq Slope (o ) = 8(71) L Sq Slope (o) = 0 (55)
Intercept (o ) = 3(. 09) Intercept (o) = 2 (.03)

Anal Cond Special Set IId

R = p-OCH

Anal Cond Special Set Ie

 

3; Y'CH3

para-Methoxy-y-methylvalergphenone,i0.10M in benzene, 0.002M 0ctadecane

 

standard.
Run No. l

Quencher
Conc. M

0
0.001
0.002
0.003
0.004
0
Act

Prod/Std
Ratio

4.35
2.29
1.61
1.29

L Sq Slope (o)

Intercept

(a) ;

—'oo

99
¢

1. 00
1. 90
2. 70
3. 37
4.48

 

43 (28)
.00 (.07)

Anal Cond Special Set IId

Run No. 2

Quencher
Conc. M

0
0.001
0.002
0.003
Act

Prod/Std
Ratio

2.08
1.086
0.741
0.569
2.47

L Sq Slope (o) =
Intercept (o) =

99
¢

1.00
1.92
2.81
3.66

 

8)
2( 02)

Anal Cond Special Set 18

160

R = m-OCH3

meta-Methoxyvalerophenone, 0.10M in benzene, 0.002M heptadecane
standard.

 

 

 

Run No. 1 Run No. 2
Quencher Prod/Std 99 Quencher Prod/Std 99
Conc. (M) Ratio 9 Conc. (M) Ratio 0
0 0.770 1.00 0 0.416 1.00
0.0005 0.652 1.18 0.0004 0.369 1.13
0.0010 0.569 1.35 0.0030 0.227 1.83
0.0015 0.514 1.49 a
0.0020 0.485 1.58 0 0.684 1.00a
0 0.765 0.0004 0.682 1.01a
Act 8.81 0.0030 0.370 1.87
Act none

Anal Cond Special Set IIb

Anal Cond Special Set IIb

 

 

 

 

Run No. 3 Run No. 4b

Quencher Prod/Std 99 Quencher Prod/Stdc $9
Conc. (M) Ratio 0 Conc.)(M) Ratio 0

0 0.798 1.00 0 0.516 1.00
0.001 0.735 1.06 0.001 0.484 1.07
0.002 0.532 1.47 0.002 0.398 1.30
0.003 0.380 2.05 0.003 0.323 1.60
0.004 0.313 2.49 0.004 0.303 1.71
0 0.761 0 0.519

Act 14.7 Act 15.8

Anal Cond Special Set Ib

Anal Cond Special Set Ib

 

a Ketone prepared via cadmium reagent method, other cases ketone

prepared via Grignard reagent.

b Quantum yield is too low in this run, probably due to contaminated
stock standard solution.

C Cyclobutanol areas were also measured for this run, the values for
the six samples are in order: (Prod/Std), (00/0); (1.56), (1.00);
0.535), (2.91); (0.328), (4.76); (0.238), (6.56); (0.19), (8.28);

1.69), (1.00).

161

R = o-OCH3

ortho-Methoxyvalerophenone, 0.10M in benzene, 0.004M heptadecane

 

 

 

 

 

 

standard.
Run No. 1 Run No. 2a
Quencher Prod/Std 99 Quencher Prod/Std 99
Conc.(M) Ratio 9 Conc. (M) Ratio 9
0 2.43 1.00 0 2.96 1.00
0.001 1.14 2.13 0.001 1.23 2.36
0.002 1.00 2.43 0.002 1.01 2.87
0.003 0.90 2.70 0.003 0.894 3.24
0.004 0.83 2.94 0.004 0.793 3.66
0 2.43 0 2.83
Act 3.54 0.0005 1.85 1.57

0.0002 2.40 1.21
Anal Cond Special Set IIb Act 1.99

Anal Cond Special Set IIb
Run No. 3a Run No. 4a,c
Quencher Prod/Std 99 Quencher Prod/Std $9
Conc. (M) Ratio 9 Conc. (M) Ratio 9
0 2.21 1.00 0 2.42 1.00
0.0002 2.02 1.10 0.0002 2.02 1.15
0.0004 1.72 1.29 0.0004 1.74 1.34
0.0006 1.70 1.31 0.0006 1.50 1.55
0.002 1.00 2.21 0.001 1.28 1.82
0.003 0.91 2.43 0.002 0.88 2.65
0.004 0.69b 3.20 0.003 0.77 3.03
0 3.62 0.004 0.64 3.64
Act 2.65 0 ~ 2.25

Act none

Anal Cond Special Set Ib
Anal Cond Special Set Ib

 

a Standard concentration of 0.002M.

b This value for unquenched ketone gave 011 of 0.25.

C This series irradiated at 3660A.

162

R = o-OCH3

ortho-Meth05yyalerophenone, 0.10M in benzene, 0.004M heptadecane
standard. .

 

Run No. 5 Run No. 6
Quencher Prod/Std 99 Quencher Prod/Std 99
Cone. (M)_ Ratio 9 Conc. M Ratio 9
0 0.722 1.00 0 2.20 1.00
0.0002 0.671 1.076 0.0002 1.90 1.16
0.0004 0.627 1.15 » 0.0004 1.82 1.21
0.0006 0.602 1.20 0.0006 1.65 1.33
0.001 0.557 1.296 0.001 1.75 1.26?
0.002 0.370 1.95 0.002 1.57 1.40
0.003 0.359 2.01 0.003 1.54 1.43
0.004 0.270 2.59 0.004 1.47 1.56
Act 1.48 0 2.19 ‘
Act 3.82

Anal Cond Special Set Ib
Anal Cond Special Set Ib

 

R = p-SCH R = m-OCH y-CH

33 3
meta-Methoxy-y-methylvalero‘
Ehenone, 0.10M in benzene,
0.002M heptadecane standard.

3

para-Thiomethoxyvalerophenone,
0.10M inTbenzene, 0.002M
0ctadecane standard.

 

 

 

Quencher Prod/Std gm Quencher Prod/Std $9
Conc. M Ratio 0 Conc. M Ratio 0
0 0.00 ---- 0 1.54 1.00
Act 1.92 (Irrad. 3 hrs) 0.0005 1.37 1.12
0.0010 1.30 1.18
Anal Cond Special Set IId 0.0015 1.18 1.30
0.0020 1.10 1.39
0 1.52
R = OH Act 8.81
0 1.14
para-Hydroxyvalerophenone, 0.10M Act 7 54

 

in benzene, 0.001M heptadecane
§£§fl駧§3 Anal Cond Special Set IIb
Relative area for 0.5 microliters:

Irrad. ketone sol'n 2

0.05M product sol'n 275
Act (Prod/Std) ratio = 12.1

Anal Cond Special Set 1119

163

.cmmm wn upzou memsomw mace» ucm.mwm ou mcwuconmmccou .mxmmq mcwaampcm>o o3» AuooNFV
mmcaumcmuemu cmzop um .Lm>m3o; .ucmcu ua> mgu co xmma mco xpco mg on nmcmmaam mews» mwcaumcmaEmu gunmen u<

a

.mwmxpmcm mcwczu mung» ua> mg» no um>swmno mpocmuanopuau togammca mo Luggac mg» co$ museum A*v mg» m

 

 

 

 

 

 

 

 

0N 38m _eedddm N mmN.o N_.e NP.N NN.N mpg mNmo.o decededoeexdepseudz-8
UHHN Nam Fawddam N N_N.o ON.N we.N oo.e N_u memo.o deepaeFAONeenwo-e._
LN gem _eedddm 8:8: No.4 Nm.m _m.m Neg mewo.o decedgdoesa=n_seeea-m
LN New _ewddam N mm_.o ON.m mN.N 8N.m N_u memo.o decedeaoexu=e_aeded-»
8N “am _eeddam N NNN.o No.4 ON.N mm.m N_u Nmo.o decededoeeeoz
8N paw _aeddam N NmN.o ee.e .N._ me.N o_u mo_.o decededossusnaxoeuez-»
8N pew _ecddam _ oeN.c em.P N_.F mm.. o_u mop.o deoeeeaopddezxoeudz-e
8N New _8_dddm N mN.o ON.N ON.N oe.N e_u FNF.o decedeaoeeXdz
8N “am Newdddm _ omo.o ON.N mm.N mm.m mpg NFP.o decededoed_espsepdz-s
8N dam P8_ddam med: ON.N om.N 8N.m mpg mdmo.o decededoexusa_Nepdewa-m.m
UN Sam _e_ddam N ep.o ON.N mm._ e_.m mpg mp_.o decedeaoed_e>omN
N dam 83m 8:0: ON.N No._ 8N.” New oo_.o decedeaoesusnaxoeusz-»

8N New _ewddam N Nm_.c ON.m mo._ NN.N N_u Nmmo.o decedeadsdfie>
8H new _ecddam F _mo.o ee.e 88., No.m N_u No_.o deoeeedoesusm
eeou _mee N epm\eoea pd< Asev eeee<-deoedm No.o azv.deou deepdg

a m_oeee=no_dxu ”o_uem eNm\deopd¥ 88m
.moNNN> z=ez<ao soz<e=moeu>u oz< Nuz<N<Naa<mNa quzNzNNeNo Non <e<o N<ezuzNNNaxm .m NN<N .< xHazmda<

164

.ZNO0.0 u cowpme—Hcmucou ULMUcoum m

 

 

 

 

 

 

eNH dam _8_wdam meme Nae” u mnmm wnmm ew_w oofl.o decedeaoedPessmoeueeoeee-d
UN 38m _ewdedm N Nmo.o oe.N om.N eo.m N_u oop.o deoeeeaoesu=n_aew>-»
LNHN pew Fecddam deoe oe.m mo.~ mm.N N_u oop.o decededoeeXdeoeesu-d
LNNH 38m Pdcddam F NNo.o oe.m om.F ON.N N_u oop.o decedeaoeeXdeoeo_eu-d
NH dam _eeddam N op.o oN.m NN.F mm.N mpg ONo_.o decedeaosd_e>oedao-e
NH “em _ewdddm _ Ne.o oN.m No._ mm.N N_u N_F_.o decededoed_assxoeueeonedu-©
dNNH New Pe_ddam N e_.o ON.N Nm.o Fm._ _N_u Nmop.o decedeaoedpesosopeu-8
dNHH “em Pe_dddm eeoe ON.N om._ mm.N New mmo_.o decededoexusexxoeudeoaeeo-»
Acowumcwssmpmc ocv mcocmgaocmfim>xxoncmuic
eNNN paw _ewddam P __.o ON.N mm.N Nm.e N_u oNNo.o decededoeaueeepeda
8N gem _eeddam deoe Np.e me.N _N.N N_u mNoN.o decedeaossuanoeeau->
8H Hem Paeddam _ Noo.o NF.e 8N._ om.p N_u emo_.o decededosNuseoeopeu-»
nHNH pew _ewdddm _ mNo.o NN.N_ me.N NN.N N_u ooe.o decedeaosseeeoePse_»;pde_a-»
8H Hem _ewdeam _ me~.o mo.e Na._ NN.N N_o eop.o deoeeedoedPM>Nxoeedz->
8N New _e_ddam N mom.o No.8 me._ NN.N N_u memo.o decededoeexdeesegeeeo-e.8
eeou Nee< N eym\eoed pd< A>ev sdpc<-deoeem zNo.o sz.deoo deepdx
"mpoeeusao_dzu "cream u¢m\deoud¥ cum
amazfiezoo .N eN<a .< xHozmaa<

165

APPENDIX A. PART 4. SOLVENT STUDIES ON THE PHOTOCHEMICAL BEHAVIOR OF
SUBSTITUTED PHENYL KETONES.

5,3-Dimethy1butyrophenone, 0.10M
in benzene, 0.004M tetradecane

Isovalerophenone,,0.10M in benzene,
0.004M tetradecane standard.

 

 

 

a standard.
t-BuOH Prod/Std
Conc. M Ratio ¢II t-BuOH Prod/Std
Conc. (M), Ratio ¢II
0 0 868 0.376 -
0.50 1.46 0.635 0 0.445 0.195
1.00 1.71 0.744 0.50 0.646 0.285 _s.
2.00 1.88 0.816 1.00 0.806 0.352
5.00 2.03 0.881 2.00 1.02 0.445
8.00 1.93 0.836 5.00 1.37 0.598
0 b 0.862 8.00 1.73 0.755 *-
Act 0 758 0 0.449
Act 0.758

 

Anal Cond Std Set I

 

Anal Cond Std Set I

 

IeMethylvalerophenone, 0.10M

in

 

benzene,)0.004M tetradecane

 

 

Valerophenone, 0.10M in benzene)

0.004M tetradecane standard.

 

 

 

standardi

t-BuOH Prod/Std t-BuOH Prod/Std

Conc. (M) Ratio ¢II Conc. M Ratio ¢II
0 0.883 0.26 0 0.99 0.33

0.50 1.61 0.48 0.50 1.77 0.585
1.00 1.93 0.576 1.00 2.19 0.742
2.00 2.47 0.735 2.00 2.48 0.840
5.00 2.92 0.87 6.00 3.02 1.00

8.00 2.92 0.87 Act none

0 0.862

Act 1.10 Anal Cond Std Set I

Anal Cond Std Set I

 

a Tart-butyl alcohol was added in increments to the benzene solutions.

b Actinometers are 0.10M valerophenone, 0.004M tetradecane standard in
benzene unless otherwise indicated.

Hexanophenone, 0.10M in benzene,
0.004M tetradecane standard.

t-BuOH Prod/Std

 

Conc. M Ratio ¢II
0 0.886 0.30
0.50 1.64 0.547
1.00 2.02 0.675
2.00 2.21 0.738
5.00 2.53 0.840
8.00 2.56 0.856
0 0.911

Act 0.99

Anal Cond Std Set I

166

yeflydroxybutyrophenone, 0.10M in
benzene, 0.004M tetradecane

 

standard}

t-BuOH Prod/Std

Conc. (M) Ratio ¢II
0 1.33 0.35

0.50 1.70 0.45

1.00 1.93 0.51

2.00 2.11 0.56

5.00 2.70 0.72

0 1.31

Act 1.24

Anal Cond Std Set I

 

o-Chlorovalerophenone, 0.10M
benzene, 0.004M5tetradecane
standard.

t-BuOH Prod/Std

Conc. M Ratio ¢II
0 1.51 0.60
1.00 1.68 0.67
3.00 1.70 0.68
5.00 1.72 0.69
8.00 1.33 0.53
0 1.53

Act 0.829

Anal Cond Std Set II

o-Cyanovalerophenone, 0.10M in
benzene, 0.004M tetradecane
standard.

 

 

t-BuOH Prod/Std

 

Conc. M Ratio ¢II
0 1.27 0.50

1.00 1.79 0.71

3.00 1.53 0.61

3.00 1.60 0.64

5.00 1.50 0.60

8.00 1.34 0.53

0 1.28

Act 0.829

Anal Cond Std Set 11

 

o-Carbomethoxyyalerophenone,
0.10M in benzene, 0.004M
tetradecane standard:

 

 

 

Prod/Std

Conc M Ratio d’11
0 1.44 0.58
1.00 1.82 0.72
3.00 1.94 0.77
5.00 1.88 0.75
8.00 1.84 0.73
0 1.46

Act 0.829

Anal Cond Std

B-Phenylbutyrophenone, 0.10M in
benzene, 0.004M tetradecane
standard.

Prod/Std

 

t-BuOH

Conc. M Ratio ¢II
0 0.044

8.0 0.0604

Act none

Anal Cond Std Set I

 

167

Valerophenone, 0.10M in solvent, 0.004M tetradecane standard.

Run No. 1 (benzene solvent

Run No. 2 (benzene solvent)
plus 0.60M pyridine added)

 

 

 

EtAca Prod/Std
Quencher Prod/Std 99 Conc. (M) Ratio ¢II
Conc. (M) Ratio 9

O 0.916 0.33
0 1.12 1.00 0 0.912 0.33
0.010 0.840 1.32 0.10 0.971 0.35
0.020 0.655 1.69 0.10 0.982 0.35
0.030 0.535 2.07 0.20 1.038 0.375
0.040 0.432 2.56 0.20 1.021 0.37
0 1.10
Act 0.460 Ana1 Cond Std Set I

Ana1 Cond Std Set I

 

Run No. 3 (methanol solvent) Run No. 4 (acetonitrile solvent)

 

 

 

Quencher Prod/Std $9 Quencher Prod/Std $9
Conc. (M), Ratio 0 Conc. (M) Ratio 9
0 1.64 1.00 0 1.40 1.00
0.020 0.618 2.59 0.020 0.621 2.26
0.040 0.315 5.08 0.040 0.382 3.67
0.060 0.208 7.68 0.040 0.378 3.71
0 1.56 0.060 0.274 5.10
Act none 0.060 0.275 5.09
0 1.40
0 1.96 Act none
Act 0.70
0 2.12
0 1.42 Act 0.785
Act 0.486
0 1.16
0 1.48 Act 0.478
Act 0.554

Ana1 Cond Std Set I

a Ethyl acetate used in solvent study.

Anal Cond Std Set I

 

168

y-Dimethylaminobutyrophenone, 0.10M in solvent, 0.004M tetradecane

stanaard.

Run No. 1 (methanol solvent)

Quencher
Conc. ( )

0
0.10
0.20
0.30
0.40
Act

Prod/Std op

Ratio 9
1.12 1.00
0.844 1.33
0.60 1.87
0.497 2.26
0.386 2.90
1.56

Anal Cond Std Set I

Run No. 2 (acetonitrile solvent)

 

 

Quencher Prod/Std g.

Conc. (M)_ Ratio 0

0 0.552 1.00

0.10 0.465 1.165
0.20 0.417 1.30

0.30 0.375 1.445
0.40 0.337 1.61

0 0.542

Act 3.70

Anal Cond Std Set I

 

y-Dimethylaminobutyrophenone Hydrochloride, 0.10M in solvent, 0.004M
tetradecane standard (except where indicated).

Run No. 1 (methanol solvent)

Quencher
Conc. ( )

0
0.001
0.005
0

0
Act

Prod/Std $9

Ratio 9

0.253 1.00
0.140 1.78
0.056 4.5

0.245

0.248

9.59

Anal Cond Std Set Ia

Run No. 2b (water solvent, no
standard used).

 

 

Quencher Prod/Std $9
Conc. (M) Ratio 0
0 0.13
Act 33.5

Ana1 Cond Std Set I3

 

a The amine hydrochlorides were analyzed by regenerating the amine by
adding a small amount of fine potassium carbonate, shaking for one
minute, and analyzing the sample within ten minutes.

b

The product to standard ratio was estimated by taking accurate 0.5

microliter shots of the sample and the actinometer.

169

Disa earance of y-Dimethylaminobutyrophenone, 0.10M in given solvent,
0 0235 0ctadecane standard.

 

(Prod/Std Ratio) Moles Moles
Solvent Before After (hv) disapp'd Act ¢dis
methanol 3.04 2.40 0.021 0.074 0.28
acetonitrile 3.77 3.49 0.0074 0.074 0.10

[y-Dimethylaminobutyrophenone hydrochloride, same conditions as above]
methanol 2.62 1.11 ‘ 0.058 0.812 0.071

water 2.30 2.29 0.00 0.812 0.00

Anal Cond Special Set IIIb

 

 

 

 

 

 

 

 

 

para-Methylvalerophenone, 0.10M 1,4-Dibenzoy1butane, 0.10M in
in methanol, 0.004M pentadecane Benzene, 0.004M tetradecane
standard. standard.

Quenchera Prod/Std op t-BuOH Prod/Std

Conc. (M) Ratio 6 Cone. (M) Ratio ¢II
0 0.740 1.00 8.0 2.16 0.43
0.0002 0.525 1.40 Act 1.66

0.0004 0.419 1.75

0.0010 0.240 3.06 Anal Cond Std Set 11

0.0020 0.133 5.52

0 0.727

Act 1.19

y-Dimethylaminobutyrophenone
Ethyl Bromide, 0.10M in methanol,
0.004M tetradecane standard.

Anal Cond Special Set Ia

Quencher Prod/Std 99
Cone. (M) Ratio 0

0 0.014
Act 12.5
0 0.024
Act 19.9

Ana1 Cond Std Set II

 

a Naphthalene used as quencher.

170

APPENDIX A. PART 5. MISCELLANEOUS DATA ON QUANTUM YIELDS.
a. Type-II Quantum Yields versus Ketone Concentration.

Ketone, in benzene solvent, Std [Prod/Std Ratios]

 

 

 

 

tetradecane standard. Conc. 0.02M 0.05M 0.10M 0.20M Act
8,8-Dimethy1butyrophenone 0.004 ---- 0.92 0.90 0.92 1.64
Pentadecanophenone 0.004 ---- 1.22 1.31 1.43 1.64
1,4-Dibenzoy1butane 0.004 ---- 1.31 1.50 1.74 1.64
y-Hydroxybutyrophenone 0 004 ---- 2.26a 1.33 ---- 1.24
e-Cyanohexanophenone 0.004 ---- 1.36 1.40 1.52 1.66
o-Carbomethoxyvalerophenone 0.001 2.47 5.48b 2.74 ---- 1.44d
b d

y-Carbomethoxybutyrophenone 0 001 1.60c 3.56 1.85 ---- 1.57

 

b. Type-II Quantum Yield versus Irradiation Time(Conversi0n)?

Ketone, 0.10M in benzene, Prod/Std Ratios for Irradiation Times:
0.001M tetradecane standard. 15 min 30 min 62 min 90 min 120 min

y-Carbomethoxybutyrophenone 1.05 1.85 3.70 4.92 ----
("f) (2.10) (3.55) (7.12) (9.48) (11.39)

Valerophenone 0.78 1.57 3.19 4.69 6.17
15 min 33 min 65 min 90 min

6-Carbomethoxyva1erophenone 1.25 2.74 5.50 7.46
("f) (2.57) (5.49) (11.26) (15.6)

Valerophenone 0.659 1.44 2.78

 

a Standard concentration 0.002M.
b

d

Standard concentration 0.0005M. c Standard concentration 0.0012M.
These actinometers also 0.001M in standard; others are 0.004M.

e Other measurements of ¢II versus conversion are included in the data
with y-phenyl, y-vinyl, y-cyano, and y-carbomethoxybutyrophenone.
f Ketone 0.05M and standard 0.0005M, to correct to 0.001M standard

divide by 2.

 

171

c. Viscosity Measurementsa on Solvent and Solutions.

 

 

 

 

Valerophenone, PentadecanOphenone,
{33'} gggggggd 3:1311‘31292ifim’ 3233183129???“
1 2 min 49.0 sec 2 min 51.6 sec 3 min 01.0 sec
2 2 48.7 2 52.0 3 00.0
3 2 48.5 2 52.9 3 00.1
4 2 48.7 2 52.7 3 00.0 8}
Ave 2 48.7 2 52.3 3 00.3

 

d. Effects of Miscellaneous Materials on the Type-II Quantum Yield.

Valerophenone, 0.10M in methanol, 0.004M tetradecane standard.

 

 

Material added: Prod/Std Ratio
none 1.465

none 1.50

0.080M KBr 1.47

0.020M Et4N*8r' 1.48

0.040M “ 1.51

0.080M “ 1.49

0.20M Et3N 0.756

0.40M “ 0.486

0.80M " 0.258

8 Measurements were made with an Ostwald viscometer at 25°C.

 

172

APPENDIX B. COMPUTER PROGRAM FOR DETERMINING LEAST SQUARES SLOPES 0F

 

STERN-VOLMER DIAGRAMS.a

This Fortran IV program121 is a straightforward determination of
the least squares values122 of Stern-Volmer slopes. One useful feature
0f the program is that the line is not constrained to going through one
for the unquenched sample so the experimental intercept and its standard
deviation can be used for diagnostic purposes. Instead, the value 1.0
at zero quencher concentration is used as one point on the line, giving
it 20 to 25% weight in most cases. The data per sample is limited to
one card of 16 five-place columns (4 digits plus decimal). Calculations

were done on the CDC-6500 computer at Michigan State University.

 

Program Kemp(Input, Output, Tape60=Input, Tape61=0utput)

C Ref L. G. Parratt Probability and Exp Errors in Science, p129
Dimension Phi(16), Quen(16), Ident(10)
C Read Number of Pairs of Data Points on Card
1 Read (60,2) IN
2 Format (15)

IF (IN.GT.100) Go to 99
Read (60,3) Ident
Nrite (61,33) Ident

3 Format 10A8

33 Format 10A8

Read (60,6)((Phi(I),Quen(I)), I=1, IN)

6 Format (16F5.3)

C Compute Sums of Phi(I), Quen(I), and of Products

SUMXX=0.0
SUMY=0.0
SUMX=0.0
SUMXY=0.0
SUM=0.0

 

a This program was designed by Dr. Ellister MacDonald, post-doctoral
fellow at Michigan State University, 1969-1971.

173

APPENDIX B. Continued.

 

Do 10 I=1,IN
SUMXX=SUMXX+Quen(I)*Quen(1)
SUMXY=SUMXY+Phi(I)*Quen(I)
SUMY=SUMY+Phi(I)
SUMX=SUMX+Quen(I)
SUM=SUM+1.0
10 Continue
C Now Compute Average X and Y
. AVY=SUMY/SUM
AVX=SUMX/SUM
C Now Form Deviations in X and Y
DiffY=0.0
DiffX=0.0
DFSQX=0.0
DD=0.0
Do 11 K=1,IN
DiffY=Phi(K)-AVY
DiffX=Quen(K)-AVX
DD=DD+DiffY*DiffX
DFSQX=DFSQX+((Quen(K)-AVX)**2)
11 Continue
C Now Form A and B
A=(AVY*SUMXX-AVX*SUMXY)/DFSQX
B=DD/DFSQX
C Now Form E.S.D f S on Y
SOMY=0
Do 12 L=1,IN
Z=Phi(L)-A-B*Quen(L)
SOMY=SOMY+Z*Z
C Print Deviations From L.S.Line
Nrite(61,44) Phi(L), Quen(L), Z
44 Format(2F8.4,F7.3)
12 Continue
SY=SQRT(SOMY/SUM)
C Now Compute E.S.D of A
SIMX=0.0
SIMXX=0.0
Do 13 M=1,IN
SIMX=Quen(M)*Quen(M)+SIMX
13 Continue
SIMXX=SUM*SIMX-(SUMX*SUMX)
SAA=SQRT(SIMX/SIMXX)
SA=SAA*SY
C Now Compute E.S.D of B
SBB=SQRT(SUM/SIMXX)
SB=SBB*SY
write(61,14) A, SA, B, SB
14 Format(2(2F10.6,4X))
Go to 1
99 Call Exit
END

174

 

 

 

 

 

 

c
Mass

Spectrum
parent

peak

m/e

176

176

190

204

218 +
(232=ZO%)

APPENDIX C. IDENTIFYING SPECTRAL CHARACTERISTICS OF PREPARED KETONES.
Ketone IR Spectruma nmr Spectrumb
Phenyl ketone S = singlet
Key carbonyl stretch D = doublet
T = triplet
Substituted M = multiplet
phenyl bands Q = quintet
Other bands 5 = 5 protons
labeled P
VS = very strong
etc.
8,8-Dimethy1butyro- 1685 cm" vs 5PM at 2.16 8 2.651
phenone 1600 )5 2pS at 7.231
1450 9pS at 8.961
y-Methylvalero- 1685 cm-1 vs spM at 2.15 8 2.661
phenone 1600 IS 2pT at 7.161
1450 gpg at 8.411
p at 9.101
o-Methylhexanophenone 1685 cm'1 VS SpM at 2.18 & 2.671
1600 2 T at 7.181
1450 }5 53M at 8.1-8.91
6pD at 9.111
6,6-Dimethy1hexano- 1685 cm-1 vs SpM at 2.18 8 2.651
phenone 1600 }S ZpT at 7.191
1450 4PM at 8.2-8.91
9P5 at 9.11
Nonanophenone 1685 cm’1 VS
1600 }S
1450

 

a Taken on a Perkin Elmer 237B Infrared Spectrometer. Substituted phenyl
bands were generally doublets of which the strongest band is indicated.
Assignments were made according to chart in Reference 124.

b

Taken on a Varian A-60 nmr spectrometer.

c The author is indebted to Mrs. R. L. Guile for the mass spectra taken
on a Hitachi-Perkin Elmer RMU-6 Mass Spectrometer.

APPENDIX C. (Continued)
Ketone

y-Hydroxybutyro-
phenone

Y-Methoxybutyrophenone

Y-Chlorobutyrophenone

Y-Dimethylaminobutyro-
phenone

Y-Carbomethoxybutyro-
phenone

Y-Cyanobutyrophenone

175

IR Spectrum

nmr Spectrum

 

1685 cm-1 vs
1600
1450 }S

3300-3500 vs
(O-H)

1685 cm-1 vs
1600
1450 )5
1120 vs
1680 cm'1 VS
1600
1450 }5
1685 cm-1 vs
1595
1445 }5
2770 vs
(NCHz-H)
1685 cm-1 vs
1600
1450 }S
1735 vs
- =0
( )
3...,
1685 cm'1 VS
1600
1450 }5
2240 M

(CN)

SpM at
1ps at
2pT at
2pM at

2P" at

59" at
2pT at
2pT at
2pQ at

2.16 & 2.71
6.021
6.421
7.021
8.141

2.16 & 2.601
6.42T
6.951
7.881

Mass
§EE§JEEEE

164

178

182

191

206

173

176

APPENDIX C. (Continued)

 

 

Mass
Ketone IR Spectrum nmr Spectrum Spectrum
y-Methoxyvalerophenone 1685 cm“ V5 192
1600 }S
1450
1130 S
y-Vinylbutyrophenone 1685 cm" vs 5PM at 2.1 8 2.61 174
1600 15 1pM at 3.8-4.451
1450 2PM at 4.8-5.1T
2pT at 7.08T
1640 M 4P” at 7.8-8.31
(C=C)
o-Chlorovalerophenone 1685 cm'] VS 5PM at 2.14 & 2.601 197
1595 }S 2P1 at 6.501
1445 ZpT at 7.081
4PM at 8.181
o-Cyanovalerophenone 1690 cm'1 VS 187
1600 }S
1450
2240 M
(CN)
o-Carbomethoxyvalero- 1685 cm'1 VS 220
phenone 1600 }S
1450
1735 VS
( CH3OC=O)
e-Chlorohexanophenone 1685 cm'1 VS 210
1595 }S
1445
e-Cyanohexanophenone 1685 cm'1 VS 201
1600 }S
1450
2240 M

(CN)

APPENDIX C. (Continued)

Ketone

 

p-Methoxyvalero-
phenone

m-Methoxyvalero-
phenone

o-Methoxyvalero-
phenone

p-Methoxy-y-methyl-
valerophenone

m-Methoxy-y-methyl-
valerophenone

m-Chlorovalerophenone

177

IR Spectrum

1680 cm 1 vs
1600 vs
1510,1460}S

1420

1260 VS
(o-OCH3)

1680 cm" vs
1600 vs
1480,1470}S
1420

1265 VS
(o-OCH3)

1680 cm']
1600,1490
1470,1440

1250-1280 VS
(0-0CH3)

VS
}S

1680 cm" vs
1600 vs
1505,1455}s
1415

1260 V5
(o-OCH3)

1680 cm"1
1600,1490
1480,1430

VS
15

1250-1290 VS

(6-0CH3)
1685 cm'1 vs
1570 15

1420

nmr Spectrum

 

Mass
Spectrum

196

APPENDIX C. (Continued)

Ketone

178

IR Spectrum

 

o-Chlorovalerophenone

m-Fluorovalerophenone

o-Fluorovalerophenone

p-Trifluoromethyl-
valerophenone

m-Trifluoromethyl-
valerophenone

o-Trifluoromethyl-
valerophenone

m-Methylvalerophenone

1

1700 cm' vs

1590

1430 }5

1685 cm‘1 vs

1590

1440 1V5

1260 vs
(9-F)

1685 cm" vs

1610

1480,1450}VS

1210 vs
(9-F)

1695 cm'] VS

1580 w

1410 s

1130,1165 VS
(0:3)

1

1695 cm' VS
1610
1440 }5

1130,1160 VS

(CF31
1710 cm'] VS
1580
1450 15
1125,1160 VS
(CF31
1685 cm"1 VS
1600

1465 13

nmr Spectrum

 

Mass
Spectrum

196

180

180

230

230

230

179

APPENDIX C. (Continued)

 

 

 

Ketone IR Spectrum nmr Spectrum
p-Thiomethoxyvalero- 1680 cm"1 VS
phenone 1590 VS
1465,1435}M
1400
p-Phenyl-y-methyl- 1680 cm'] VS
valerophenone 1605 V3
Valerophenone (for 1690 cm"1 VS 5PM at 2.13 & 2.631
comparison) 1600 }S 2pT at 7.161
1450 4PM at 8.1-8.91

3pT at 9.1T

Mass
Spectrum

162

111111111

M. IIII||I1
m m71
V ”o
1 m3