The thermolysis of gem-difluorospiropentane, the kinetic effect of fluorine on the homodienyl hydrogen shift, and prepar...


Material Information

The thermolysis of gem-difluorospiropentane, the kinetic effect of fluorine on the homodienyl hydrogen shift, and preparation and reactions of difluorodiiodomethane
Physical Description:
Elsheimer, Seth Robert
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 30570222
oclc - 11902141
System ID:

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
        Page vii
        Page viii
    Chapter 1. General introduction and theory
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Chapter 2. Thermolysis of 1, 1-dideutero-2, 2-difluoro-spiropentane
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
    Chapter 3. The kinetic effect of fluorine on the homodienyl hydrogen shift
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
    Chapter 4. Difluorodiiodomethane: Preparation, properties, and potential uses
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
    Chapter 5. Experimental
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
    Chapter 6. General conclusions
        Page 88
    Appendix A. Kinetic data
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
    Appendix B. Selected spectra
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
    Biographical sketch
        Page 150
        Page 151
        Page 152
        Page 153
Full Text







To Harvey and Joanne Hall
without whom this would not have been possible.


No acknowledgement can ever adequate ly express my gratitude to Professor William R. Dolbier, Jr., for his leadership, encouragement and friendship during the course of this research. His unrelenting enthusiasm and patience make him an effective educator and an outstanding scientist. It has been a pleasure to be a part of his

research team.

I would also like to thank Dr. Simmon Sellers and

Dheya Al-Fekri for the helpful instruction in the use of the kinetics apparatus, and Jim Rocca for running and helping to interpret many of the high field NMR spectra. Thanks go to Randy Winchester and Gene E. wicks for their helpful discussions and comradery.

Finally, it is with my deepest appreciation that I acknowledge the inestimable contributions of Ms. Dale Midgette who offered moral support and long hours of editorial assistance.



ACKNOWLEDGEMENTS ------------------------------------------iii

ABSTRACT --------------------------------------------------- vi



Fluorine as a Substituent ------ -------------------- 1
Fluorine as a Substituent on Alkenes -------------- 4
Fluorine as a Substituent on Cyclopropane ------- 9

2 THERMOLYSIS OF 1,1-DIDEUTERO-2,2-DIFLUOROSPIROPENTANE -------------------------------------- 13

Introduction ---------------------------------------- 13
Results and Discussion ----------------------------- 17

HOMODIENYL HYDROGEN SHIFT -------------------------23

Introduction ---------------------------------------- 23
Thermolysis of cis,cis-l-Fluoro-2-methyl3-vinylcyclopropane------------------------------- 25
Results and Discussion ----------------------------- 27
Thermolysis of endo- and exo-8-Fluorobicyclo[5.1.0]oct-2-ene --------------------------- 35
Results and Discussion ----------------------------- 36

AND POTENTIAL USES -------------------------------- 46

Introduction ---------------------------------------- 46
Results and Discussion ----------------------------- 49


5 EXPERIMENTAL -------------------------------------- 59

General -------------------------------------------- 59
Preparation of l,1-Dideutero-2,2-difluorospiropentane 19 ---------------------------------- 59
Thermolysis of- 1,l-Dideutero-2,2-difluorospiropentane 19 ---------------------------------- 61
Comparative RaEe for Thermolysis of
1,1-Difluorospiropentane 12 --------------------- 62
Preparation of cis,cis-1-Fluoro-2-methyl3-vinylcyclopropane 31 -------------------------- 62
Thermolysis of cis,cis-1-Fluoro-2-methyl3-vinylcyclopropane 31 -------------------------- 66
Preparation of endo- and exo-8-Fluorobicyclo[5.1.0]oct-2-ene -------------------------- 67
Thermolysis of endo- and exo-9-Fluorobicyclo[5.1.0]oct-2-ene -------------------------- 71
Reaction of CHI with HgF2 ----------------------- 73
Preparation of FI2 ----- ----------------- 74
Photolyses of CF --------------------------- 75
Attempted Thermal Reaction of CF2 I2 with
2,3-Dimethyl-2-butene 54 ------------- 81
Simmons-Smith Reactions f CF I --------------- 81
Free-Radical Additions of CF 2to olefins----- 82
2 2
Zinc Reductions of 1,l-Difluoro-l,3diiodoheptane 61 ------------------------------ 87

6 GENERAL CONCLUSIONS ------------------------------- 88


A KINETIC DATA ----- --------------------------------- 89

B SELECTED SPECTRA -----------------------------------102

REFERENCES ------------------------------------------------145

BIOGRAPHICAL SKETCH ---------------------------------------150


Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE THERMOLYSIS OF gem-DIFLUOROSPIROPENTANE,


December 1984

Chairman: William R. Dolbier, Jr. Major Department: Chemistry

An isotopic labeling experiment showed that 1,1-difluorospiropentane thermally rearranges to 2,2-difluoromethylenecyclobutane via two competing mechanisms which are analogous to the rearrangement of the parent hydrocarbon. One mechanism involves initial cleavage of the C 1-C2 bond and the other involves initial C4-C5 bond cleavage. The former pathway predominates by 3 to 1. Difluorocarbene addition to (dideuteromethylene)cyclopropane gave l,1-Dideutero-2,2-difluorospiropentane. Gas phase thermolysis at 339.60C (k = 2.363+

0.025 x 10-5 s ) produced a mixture of the three possible gem-dideuterated isomers of 2,2-difluoromethylenecyclobutane.


The relative amounts of these were determined by 1H and 19F NMR spectroscopy. The small secondary isotope effect observed (kH/kD = 1.15+0.02) is consistent with the major reaction path involving initial C1-C2 bond homolysis.

Cis,cis-l-fluoro-2-methyl-3-vinylcyclopropane was

prepared in 35% yield by treatment of cis-l,3-pentadiene with CHFI2 and diethylzinc. Its gas phase thermolysis at 152.8 171.40C produced a mixture of trans- and cis-1,3,5hexatriene the latter of which equilibrated with 1,3-cylcohexadiene under the reaction conditions. Rate constants for this first order, unimolecular process were measured at 6 temperatures and activation parameters were calculated (Ea=27.6+1.2 kcal/mol, Log A = 10.3+0.6). These results are consistent with an initial 1,5-homodienyl hydrogen shift followed by rapid HF elimination.

Treatment of 1,3-cycloheptadiene with CHFI2 and diethylzinc afforded endo- and exo-8-fluorobicyclo[5.1.0]oct-2-ene. Gas phase thermolysis of either isomer produced an approximately 2:1 ratio of 1-fluoro- and 7-fluoro-l,3-cyclooctadiene. Some preliminary kinetic studies were carried out. Possible mechanisms which account for these unexpected products are presented.

Difluorodiiodomethane, which was previously available

only in small amounts via difficult and low yield procedures, was prepared in practical quantities by treatment of CI4


with HgF2. As a difluorocarbene synthon, CF2 12 was used in photochemical and Simmons-Smith cyclopropanations. Free-radical additions of CF2 2 to alkenes gave a,a-difluoro-a,y-diiodo adducts in good to excellent yields.



Fluorine as a Substituent

Fluorine is unique among substituents in its effects on organic systems. The high electronegativity, unshared electron pairs, and small size of fluorine make it a potent a-withdrawer and a potential -donor while at the same time having minimum steric requirements. A brief description of these properties is presented here, but the interested reader is directed to the excellent reviews of Smart '2 for a more comprehensive treatment. Electronegativity

Fluorine is the most electronegative of the elements. The electronegativity demands of fluorine result in a C-F bond that is rich in p-character as carbon gives up its
less tightly held p electrons. Fluorinated organic molecules often display changes in geometry reflecting changes in hybridization relative to their hydrocarbon analogues. Consider, for example, the enlarged HCH angle of 113.70 for CH2F2 compared with the idealized 109.50 of CH4.



Non-bonding Electrons

The presence of threenonr-bonded pairs of electrons on

a fluorine make it a potential 7-donating substituent. This Property is nicely illustrated by the observations that fluorine directly bound to a carbocation center stabilizes
the carbenium ion. This is best explained as conjugative interaction of fluorine's lone pairs wIith the vacant p-orbital of the carbenium ion, (1)

9C-F C=F (1)
/ /

The 7-donor capability of fluorine has also been used to explain the observed trends in bond dissociation energies of the halomethanes (Table 1-1).6 Note that as halogen content increases, Do(C-X) decreases for iodo-, bromo- and chloromethanes. This trend is reversed for fluoromethanes and spans a wider range of energies.


Table 1-1. Bond Dissociation Energies, D0(C-X) of


F 109.9+1 119.5b 127.5b 130.5+3

Cl 84.6+0.2 80.1 77.6 73.1+1.8

Br 70.9+0.3 64c 62c 56.2+1.8

I 57.2+0.3 51.3c 45.7c

aAll data from ref. 7 unless otherwise noted. bRef. 8.
CRef. 9.


Pauling suggested that contributions from "double-bond, no-bond" resonance forms (2) account for these results. As n increases from 0 to 4 for CH(4-n) Fn, a greater number of such forms can be drawn. Ab initio calculations seem to support this description.10'11

7-C-F F-C=Fe (2)
i I


With reward to fluorine's size, two points should be made. Excluding hydrogen, fluorine has the smallest


Van der Waals radius of any substituent atom or group.12 This means that the steric requirements for a fluorine are minimal. Fluorine's small size also makes it well suited for overlap of its atomic orbitals with those of carbon. This results in a strong C-F bond. In fact, of all the elements, fluorine forms the strongest single bonds to carbon.

Fluorine as a Substituent on Alkenes

The effects of vinylic fluoro substituents have

frequently been misunderstood and misrepresented. Contrary to what was once believed,13 monofluoro and gem-difluoro substituents stabilize alkenes thermodynamically. The equilibration of fluoropropenes shows that a single fluorine is thermodynamically favored on the C=C bond by about

3 kcal/mol.14


C=C C=C + C=C
1 2 3

AHr, kcal/mol

1 2 -3.34

1 3 -2.68

2 + 3 +0.64


Equation (3),which is useful for calculating substituent 15
effects relative to CH3, seems to indicate that F is similar to CH3 in its ability to stabilize a double bond. A CH3 is considered to stabilize a C=C bond by about

3 kcal/mole.16


CHF + CH2=CH-CH3 CH3-CH + CH2=CHF (3)


AHr = 0.0+1.0 kcal/mole

The increased thermodynamic stability of monofluoro

olefins is most easily explained by the resonance contribution depicted in equation (4) in which the non-bonding electrons of fluorine back-bond into the CC bond. This 7r
description is supported by ab initio calculations.11,17,18

q) F

C=C + GC-C (4)

Geminal difluoro substitution on a C=C bond is also thermodynamically stabilizing. The 7-bond dissociation
energies for ethylene, gem-difluoroethylene, and tetrafluoroethlylene are shown in Table 1-2.


Table 1-2. 7-Bond Dissociation Energies.

Do kcal/mol

CH2=CH2 59.1+2.0a

CH2=CF2 62.1+1.0b

CF2=CF2 52.3+2.0c

aRef. 19.
bRef. 20.
cRef. 21.

It appears that the 7-bond of CH2=CF2 is stabilized relative to CH2=CH2 and to CF2=CF2. Consider also the AHr for the isodesmic reaction (5).

CF2=CF2 + CH2=CH2 + 2 CH2=CF2 (5)

AH = -14.6 kcal/mol

DO(CF2=CF2) + D(C1H2=CH2) 2D (CH2=CF2 ) (6)
iF2 2T T t'22 -()
-14.2 kcal/mol

The similarity in the value of equation (6) to the

AH of equation (5) implies that the thermodynamic stability
of CH2=CF2 relative to CH2=CH2 and CF2=CF2 is due almost entirely to stabilization of the r-bond.

At first these conclusions may seem inconsistent with some of the thermodynamic studies of gem-difluoro systems. The Cope rearrangement of 4 to 5 (equation (7)), for example,


has a AHr of -5.1+0.6 kcal/mol, and the heat of hydror
genation of gem-difluoroethylene is 6.2 kcal/mol more exothermic than that of ethylene (equations (8) and (9)).


F2 F2 -5.1+0.614 (7)

4 5

CH2=H2 + H2 CH3 CH3 -32.6+0.3 (8)

CH2=CF2 + H 2 CH CHF -38.8+2.8 (9)

These observations are not, however, a result of the destabilization of the alkene, but rather, a strong stabilization of the allylic and alkylic positions by gemdifluoro substitution. A useful concept here is that of "Incremental geminal stabilization"14 (IGSTAB) which is defined as the increase in thermodynamic stabilization of a geminally substituted system, per substituent, relative to the respective monosubstituted system. Isodesmic equations are used to determine these quantities and, as shown below, IGSTAB values are strongly system-dependent.


AH ,kcal/mol IGSTAB, kcal/mol

2CH3CH2F CH3CHF2 + CH3CH -6.6 (10)

2CH2=CHF CH2=CF2 + CH2=CH2 -2.5 (11)

2CH2=CHCH2F CH2=CH2CHF2 + CH2=CH-CH -6.6 (12)

Notice that the IGSTAB for the alkene is negative but those of the alkyl and allyl systems are even more negative. The driving force to convert vinyl CF2 to alkyl or allyl is the greater stability of the latter two groups, i.e. larger IGSTAB, and not destabilization of the gem-difluoro olefins.

The difference in vinylic and alkylic IGSTAB has been

rationalized using the double-bond, no-bond resonance model.14

C ++ C


6a 6b

C=C ++ C=C

7a 7b


A resonance contributor such as 6b would be expected to provide greater stabilization to 6 than 7b can for 7.* Analogous contributing structures cannot be drawn for the monofluoro alkyl and vinyl systems.

Fluorine as a Substituent on Cyclopropane

The advantages of cyclopropane ring systems for the study of substituent effects are well known.22 How might the presence of fluorine substituents influence a cyclorpropane?

a-Withdrawing effects

Consider first the inductive effects. The high electronegativity of fluorine causes its bonds to carbon to be rich in p-character and leaves the remaining bonds poorer in p-character.3 This should result in an increased ring
strain as the unfluorinated sites are rehybridized to larger bond angles.

This description is neatly consistent with the

observations for gem-difluorocyclopropanes. The structure of l,l-difluorocyclopropane (Table 1-3) has been determined
by mirowae stdies24
by microwave studies24 and shows the C(l)-C(2) bond lengths to be shorter and the C(2)-C(3) bond length to be longer than the bonds of the parent cyclopropane.25 One way to

It may be helpful to recognize that the bonded portions of 6b and 7b are isoelectronic with a ketone and a ketene respectively.
For purposes of this discussion the term "strain" is used to denote the total net destabilization. The reader is reminded that this may have several components.23


visualize this effect is to recognize that the "bent" bonds of the cyclopropane ring are even more bent with gem-difluoro substitution on carbon.26 One would predict a qualitatively similar result, although smaller, for fluorocyclopropane (Table 1-3). The structure of C-C3H5F is not known experimentally but has been calculated.27

Table 1-3. Bond Lengths.
0 o
r[C(1)-C(2) ],A r[C(2)-C(3)],A

> 1.510a 1.510a

> F 1.494-1.497b 1.527-1.528b

> F2 1.464c 1.533c
aRef. 25.
bRef. 27.
cRef. 24.

As has been noted elsewhere,1,22,23 bond lengths cannot necessarily be correlated with bond strengths. Kinetic studies of gem-difluorocyclopropanes22,28,29 indicate a general weakening of the C-C bonds compared to the corresponding hydrocarbons. Homolysis of the C(1)-C(2) bond is enhanced by 0 to 2 kcal/mol while C(2)-C(3) cleavage is enhanced by 8 to 10 kcal/mol.

There are currently no reports in the literature of a single fluorine's kinetic effect on cyclopropane opening. 7r-Donating effects

A description of 7-acceptor influence on cyclopropane was offered by Hoffmann30 and extended by Gt'hther31 to include '-donors such as fluorine. Here one employs the familiar Walsh model of cyclopropane bonding in which the carbons are sp2 hybridized. One of the sp2 hybrid orbitals from each carbon is directed into the ring and the remaining two are used to form a-bonds to the hydrogens. The remaining p-orbitals, which are perpendicular to the CH2 planes, overlap to form the "bent" C-C bonds. The Walsh molecular orbitals of the ring can be qualitatively represented as shown on the left of Fiaure 1-1.

In order to assess the effect of -donors on this

model, one considers the HOMO and LUMO of the interacting fragments. The LUMO of the cyclopropane is 4 and the HOMO of a fluorine fragment is one of its three degenerate non-bonding atomic orbitals. As the fluorine donates electron density to the cyclopropane, the antibonding interactions for all C-C bonds increases. The implication here is that T-donors should destabilize the cyclopropane ring and weaken all the ring bonds.

Consistent with this model's inability to account for

the observed shorter C(l)-C(2) bond lengths in 1,1-difluorocyclopropane is the recent report from Clark et al.33 in


which they challenge the assumptions upon which the above model is based. These authors conclude that the e-withdrawing effect of a fluorine far exceeds its w-donor capability in cyclopropane systems.




92 3

Figure 1-1. The Effect of 7-Donors and -Acceptors on the Walsh Orbitals of Cyclopropane.



Thermolyses of spiropentane systems have been studied extensively and the mechanism for the rearrangement of the parent hydrocarbon 8 is well understood. Initial cleavage of a peripheral bond34 occurs reversibly35 to produce diradical 9. Whether 9 then undergoes a second cleavage to give diradical 10 or goes directly to methylenecyclobutane 11 is not completely clear, but 11 has been shown 36
to automerize via 10.36 Activation parameters for the rearrangement of 8 to 11 have been determined.37

H2 CH2

8 9 10 11



In order to further investigate the kinetic effects of gem-difluoro substitution in cyclopropane systems, 1,1-difluorospiropentane, 12, was studied by Dolbier et al.38 Thermolysis of 12 gave 2,2-difluoromethylenecyclobutane 13 as the only product. Control experiments demonstrated that 10.5% of the observed product 13 was formed via (difluoromethylene)cyclobutane 14 which rapidly isomerized to 13 under the reaction conditions.


12CF 2 13 2



Activation parameters for the thermolysis of 12 are listed in Table 2-1 along with those for the parent hydrocarbon.

Table 2-1. Activation Parameters for the Thermolysis of Spiropentane and 1,l-Difluorospiropentane.

Eakcal/mol Log A Krel(3400C)

8 57.60.6 15.9+0.04 1.00

12 2 58.0+0.5 16.1+0.2 1.14


The modest rate enhancement of 12 relative to 8

(Ea and Log A equal within the experimental uncertainties) is consistent with the two mechanistic pathways depicted in Figure 2-1. These differ by which of the two peripheral bonds cleaves initially.

Mechanism a

F2 ,CF2 F2

12 15 16


14 13
Mechanism b

F2 H2 IF2 C 2

2 CH
12 17 CH2 2

Figure 2-1. Pathways for Rearrangement of 1,1-Difluorospiropentane to 2,2-Difluoromethylenecyclobutane.


In order to determine how much 13, if any, is formed through the competition of mechanism a with mechanism b, an isotopic labeling study was devised.

Thermolysis of the dideuterated compound 19 can reveal, through the distribution of the deuteriums in the product, if there is significant competition by

mechanism b (Figure 2-2)

Mechanism a CF
19 20 21

CD 2 C

F2 D2

23 22


D 2 F 2 F 2
27 24

Mechanism b

> ~ F2 CH2>KIF2 F2
F 2

D2 H 2
19 2 25 CH2 26 CD2

Figure 2-2. Deuterium Distribution According to Mechanism.


Results and Discussion

Synthesis of 19 was achieved through difluorocarbene addition39 to (dideuteromethylene)cyclopropane.40

> =2CD 2 CF3HgPh 2

Thermolysis of 19 produced a mixture of the three possible gem-dideuterated 2,2-difluoromethylenecyclobutanes 23, 24 and 27. The identities and relative amounts of these were determiined by comparison of the 1H and 19F NMR spectra of the mixture with those of the undeuterated material.

The 1H NMR spectrum of 13 consists of two unresolved multiplets at 62.43-2.76 which were assigned to the C(3) and C(4) protons, and two vinylic multiplets at 65.15 and 65.49. The ratio of vinyl to non-vinyl protons is 1:2. The 1H NMR spectrum of the thermolysis products from 19 shows the same upfield multiplets at 62.41-2.76 but the vinyl signals are completely resolved into two quintets of doublets at 65.15 and 65.47. This pattern is explainable assuming that the allylic F and H coupling constants are each 3Hz. The smaller splitting, J=l.l Hz, is due to geminal coupling. The ratio of vinyl to non-vinyl protons is 1.0:2.1 which indicates that 23 is present in 37%.


S.S 4r+ 13

27 H<

24 D2 Z H


Figure 2-3. Comparison of the Vinylic H NMR Signals for 13 and the mixture 23, 24, and 27.


The ratios within these vinylic quintets are 1.0:7.6: 16.6:8.1:1.1 which is consistent with superimposition of a triplet from 24 and a quintet from 27, having identical chemical shift and coupling. Figure 2-3 shows a comparison of the vinylic regions of the IH NMR spectra for the undeterated 13 and the deuterated product mixture of 23, 24 and 27.

The 19F NMR spectrum of the undeuterated 13 shows a complex multiplet at 94.49 with a large vicinal HF coupling to the C(3) protons of 11.8 Hz. The mixture of thermolysis products from 19 shows not only the multiplet described above, but also a second, narrower multiplet at 94.85 which lacks vicinal HF coupling. This new signal is due to the presence of 27 which, in addition to possessing no H vicinal to the fluorines, displays a deuterium isotope effect on the 19F chemical shift. The magnitude and direction of this shift (0.36 ppm upfield) are consistent with those previously observed.41'42 The ratio of low to high field multiplets is 2.8:1.0, indicating 27 comprises 26% of the mixture. The 19F NMR spectra for the undeuterated 13 and the mixture of deuterated 23, 24 and 27 are compared in Figure 2-4. Knowing the amounts of 23 and 27 present to be 37% and 26% respectively, one then arrives, by subtraction, at 37% for the amount of the remaining isomer 24.


24 23 27 c 'S

Figure 2-4. Comparison of the 19F NMR Spectrum of 13 to that of the Mixture 23, 24, and 27.


The ratio 37:37:26 for 23:24:27 is consistent with the complete thermal equilibration of 23 and 24 which one would expect under the conditions used for thermolysis of 12. 36Since products 23 and 24 are formed via mechanism a, and 27 is formed via mechanism b, the product ratio indicates a competition between the two proposed pathways in which a is favored over b by about 3 to 1.

a b

>< 2 D2 D Ej

D2 2 2f

12 23 24 27

37% 37% 26%

The product ratios, which were established by 1H and 19F
NMR integrations have a reliability of q,+2%.


Samples of 19 and the undeuterated compound 12 were

thermolyzed under identical conditions to allow a comparison of their rates (Table 2-2). Rates were determined both by following disappearance of starting material versus an internal standard or alternatively, monitoring product formation. Greater precision was obtained for the latter method and those results were regarded as more reliable.

Table 2-2. Thermal Isomerization of 12 and 19 at 339.60C.
kx10 5, S -1 k x 10 5, s
(internal standard) (product)

12 3.7147+0.0345 2.7156+0.0196

19 3.3683+0.0316 2.3627+0.0246
D 2
The secondary kinetic isotope effect observed (k H/k =

1.15+0.02) is consistent with the majority of the reaction proceeding through initial C(1)-C(2) bond homolysis.*

The results from the thermolyses of 12 and 19 can

now take their place among those for other gem-difluorocyclopropanes. The observation of a very slight rate enhancement for cleavage of the C(l)-C(2) bond is consistent with earlier findings.22

In a similar but lower E rearrangement, the thermal isomerization f dideuterobis(cyclopropylidene) showed a kH/kD = 1.21.4


In contrast to the wealth of information on gemdifluorocyclopropanes, the studies of the monofluoro derivatives are quite sparce, and these provide information on thermodynamics, rather than kinetics. Paquette's observation of the fluorosemibullvalene equilibrium qualitatively showed that a monofluoro substituent is favored thermodynamically on a double bond 43
rather than a cyclopropane.43

A similar but quantitative study by Dolbier and Burkholder44 supports this result with the observed equilibrium between the isomeric monofluorinated methylenecyclopropanes.




AHr = -2.60+0.06 kcal/mol

A particular problem in the study of monofluoro

systems is that of HF elimination. A monofluoro compound shows a much greater tendency to eliminate IHF than does the corresponding gem-difluoro compound. Consider
for example the study by Dolbier and Medinger14 in which the attempted Cope rearrangement of 3-fluoro1,5-hexadiene could not compete with the rapid formation of HF elimination products.






Thermolysis of cis,cis-l-Fluoro-2methyl-3.-vinylcyclopropane

The 1,5 homodienyl hydrogen shift (equation (13)) was selected as a rearrangement in which the kinetic influence of a monofluoro substituent on a cyclopropane might be probed.

/ 2
4 1= (13)

The mechanistic details of the homodienyl shift have been presented elsewhere.4547 Due to the low activation energy for this concerted process the potential problem of HF elimination would be minimized. This system is additionally attractive since kinetic studies on the 45
parent hydrocarbon 29,45 and the gem-difluorinated
compound 30,29 are available for comparison.




F2 F2


A stereochemical requirement for the homodienyl

shift is that the carbon bearing the migrating hydrogen be on the same side of the cyclopropane ring plane as the alkene (i.e. cis). Given this restriction, there exist two possible isomeric l-fluoro-2-methyl-3-vinylcyclopropanes which should be accessible from formal addition of fluorocarbene to cis-l,3-pentadiene (14).


+ IF (14)


Excluding diadducts, there are four isomeric products one would expect from this synthetic strategy; two regioisomers each of which has two stereoisomers.


31 32 33 34

Results and Discussion

Employing a variation of the method by Nishimura
and Furukawa, cis-l,3-pentadiene was heated with CHFI2 in the presence of diethylzinc (15).

CHFI2 31 33
Et2Zn F

PhCH3 +
34h F

35% 21%

+ 25% diadducts


In agreement with earlier observations, the predominant mode of addition of the organometallic carbenoid produced cyclopropanes with the fluorine cis to the other groups on the ring. None of the trans monoadducts, 32 and 34 were observed under these conditions.

Gas phase, thermal isomerization of 31 carried out in a well-conditioned pyrex vessel with the results shown in equation (16). Yield was determined by direct GC analysis


6" (16)

of the gaseous product mixture compared to an internal standard. Products were identified by their 1H, 13C (and their absence of 19F) NMR spectra. Rates of thermal isomerization were measured at six temperatures. Data obtained are given in Table 3-1.

Table 3-1. Rates of Thermal Isomerization for 31.

T,0C k x 104,s152.8 1.210+0.030

157.1 1.740+0.043

160.4 2.300+0.063

165.1 2.978+0.082

168.1 3.612+0,183

171.4 5.064+0.046


An Arrhenius plot of the rate data gave a good straight line and the frequency factor and energy of activation were calculated by the method of linear least squares. Activation parameters calculated from rates based on product formation agreed, within experimental error, with those calculated from rates based upon the disappearance of starting material versus an internal standard. Greatest precision was obtained from the latter method, and it is those data which are used throughout this chapter.

Table 3-2 lists activation parameters and relative rates for the unfluorinated, cis-monofluorinated, and gem-difluorinated species.

Table 3-2. Activation Parameters for 29, 30, and 31.

Ea Log A T10C AH3 AS AG (166.20C) Krel at
kcal/mol kcal/mol cal/mol.!K kcal/mol (166.20c)

29a' 30.9+,0.4 10.9+0.2 166.2- 29.9 -11.7 35.0 (1)
(31.2) (11.0)

31 27.6+1.2 10.3+0.6 152.8- 26.7 -14.4 33.0 11.2

30 c
23.3+0.5 10.3+0.3 52.6- 22.7 -13.5 28.6 1540

bFrom a linear least squares treatment of the rate data in ref. 45.
Numbers in parentheses are those reported in ref. 45.
CRef. 29.


These results are most easily explained by the following mechanistic scheme. Initially the 1,5-homodienyl hydrogen shift takes place in a concerted fashion to produce cis-3-fluoro-1,4-pentadiene 32. This bisallylic fluoride then rapidly loses HF to give trans- and cis- 1,3,5-hexatriene, the latter of which equilibrates with 1,3-cyclohexadiene under these conditions.



From the data in Table 3-2 one can see that a cismonofluoro substituent lowers the activation energy for the homodienyl shift by 3.3 kcal/mol and a second fluorine (gem-difluoro) lowers it by an additional 4.3 kcal/mol. At 166.21C the relative rate of thermolysis is increased only 11-fold by a single fluorine substituent, while an additional 130-fold rate enhancement is observed for 30.


A rationalization for this non-linearity of substituent effect can be found in the following model. Consider the parameters and 0 shown in Figure 3-1.


Figure 3-1. Substituent Bond Angles Relative to the Cyclopropane Ring Plane.

These are defined as the angle between the cyclopropane ring plane and the C-X or C-y bond respectively. When x and y are both F, one expects that and e will be equal. The bonds to fluorine will be rich in p-character and the resulting rehybridization around C(l) must manifest itself through additional ring strain as the C(2)-C(l)-C(3) valence angle is opened outward.


Now consider the monofluoro case (x = F, y = H). As before the bond to fluorine is richer in p-characterand causes a rehybridization around C(l) resulting in an increase in all the angles not involving the C-F bond.* The already strained cyclopropane bonds cannot accommodate the loss of p-character as well as the C(l)-H bond can. One would expect a flattening of the C(l)-H bond relative to the cyclopropane ring, i.e., an increase in 6 and a decrease in . The important point here is that the diminution of p-character in the bonds of C(l) to nonfluorine atoms can be partially compensated by something other than ring strain. By contrast, the gem-difluoro species (x = y = F) does not have this option.

The observed non-additivity of rate enhancement by monofluoro and gem-difluoro substituents also can be explained using a transition state argument. Although the exact nature of the activated complex is not known, it can be reasonably assumed to be some structural hybrid of the reactant and the product. Recall that the IGSTAB values discussed in Chapter 1 are strongly system-dependent. Incremental geminal stabilization by fluorine follows the trend alkyl = allyl >> cyclopropyl > vinyl.49 As the reaction

It may be helpful to consider the (absurd but enlightening) limiting cas where the C-F bond is pure p. This would result in sp hybridization around C(l) making C(l), C(2), C(3) and y essentially coplanar with the C-F bond perpendicular to the plane.


proceeds from the cyclopropane to the bisallylic product, it passes through an activated complex which can be represented as 33. This species would be expected to

x y

33 a. x =y = F

b. x = H, y = F

have an IGSTAB for fluorine somewhere between cyclopropyl and allyl. One would predict a lowering of the transition state for 33a relative to that of 33b. Double-bond, nobond resonance forms (equation (17)), which have no counterparts for the monofluoro species, can be used to rationalize the system-dependence of IGSTAB.

E e P E 9 9

-~ I I(17)


Thermolysis of endo- and exo8-Fluorobicyclo [5.1.0] oct-2-ene

In a related study, the thermolyses of 34 and 35

were examined. The hydrocarbon 36, and the gem-difluoroanalogue 37 have been examined previously.


34 35

36 K = 37
-- eq
E = 38.6+0.6 kcal/mol Log A = 13.3

F F2

37 38
-4 -1
k(152.50)=4.47x10 s


The hydrocarbon undergoes a reversible homodienyl shift for which the activation and equilibrium parameters were
determined.50 The difluoro species 37 was found to rearrange cleanly to 3,3-difluoro-l,4-cyclooctadiene 38 and a rate constant was determined for a single temperature.29

Results and Discussion

Reaction of 1,3-cycloheptadiene with CHFI2 and

diethylzinc yielded the products shown in Table 3-3. The compounds were identified by standard spectral methods.

~Et2Zn + CHFI2 Et2Zn


Table 3-3. Reaction Products from 1,3-Cycloheptatriene, CHFI2 and Diethylzinc.

% Yield

F 71

35 F



39 F



40 F


Thermolysis of 34 was carried out at three temperatures. The results are given in Table 3-4. Interestingly, neither 41 nor the HF elimination product 42 was observed in the product mixture.


41 42

A single rate was determined for the exo isomer 35 which thermolyzed to give two products with the same GC retention times as 43 and 44 and which appeared in a similar ratio.


Table 3-4. Rate and Product Data for the Thermolysis of 34 and 35.


34 43 44

4 -1L
T, 0C 43/44 kx10 ,s-1

190.0 1.84 1.796+0.059

181.0 1.72 1.199+0.041

175.2 1.54 0.745+0.027

173.8 1.57 0.194+0.045

34: Ea 24.1+3.9 kcal/mol, Log A = 7.63+1.90
-5 -1
-k(calculated for 173.80C) = 7.0x00 s


The observed products are difficult to account for

mechanistically (Figure 3-2). Assuming that 34 undergoes the expected homodienyl shift, the primary product formed would be the 1,4-diene 41. There is, however, no obvious mechanism for going from a 1,4-cyclooctadiene to a 1,3-cyclooctadiene. Products 43 and 44 can interconvert with one another (and also the other two isomeric monofluoro 1,3-cyclooctadienes) via 1,5-hydrogen shifts. The problem, then, is reduced to finding a mechanistic entry into 1,3-cyclooctadiene systems from 34.


34a 34b 41

1,5 H-Shifts


43 44 45 46

Figure 3-2. Possible Intermediates in the Thermal Rearrangement of 34 to 43 and 44.


A possible conversion of 41 to 45 could involve a

1,3-fluorine shift. Similar 1,3-fluorine shifts, although never unambiguously demonstrated, have been proposed and strongly implicated for the rearrangements of more highly fluorinated compounds. Gas phase thermolysis of 47, for example, has been shown to give 48 as the major product. Control experiments seem to have ruled out the possibility of surface-catalyzed or fluoride ion-promoted reactions.51

F2 F F

2 H 380-4000C FO H
F2 F2

47 48


An alternative mechanism, which does not require the intermediacy of 41, can be visualized by first considering the simplified systems shown in equations (18) and (19). The usual homodienyl hydrogen shift involves, in the simplest case, a methyl vinyl cyclopropane. The cyclopropane bond is in the 2,3 or "internal" position of the 1,5 system

(18). Equation (19) shows a different system in which the migrating cyclopropane bond is at the 4,5 or "terminal" position. Neither of these molecules can undergo the reaction depicted for the other. However, a species such as 34 meets the structural requirements for either mechanism.

3 2
internal (18)
Li\ CH 2


3 H terminal (19)



Figure 3-3 shows an additional factor which may help account for the unusual products. In order to yield the observed product with cis double bonds in the ring, the usual internal homodienyl shift is required to proceed via the "saddle" conformation 34b47 in which the crossring Van der Waals interactions are unfavorable. Molecular models show that a lower energy conformer 34a has good geometrical disposition for the terminal homodienyl shift and can give a cis,cis-l,3-cyclooctadiene. A product formally derivable from such a terminal homodienyl shift


34a 34b


Figure 3-3. Conformations for Internal and Terminal Homodienyl Shifts.


has been observed previously. Dolbier and Sellers reported 4,4-difluoro-l,3-hexadiene 49 as a thermolysis product of 50.

50 49

It is not obvious why the internal homodienyl pathway should be favored for both the hydrocarbon50 and the gemdifluoro compound29 while the terminal pathway is implicated for 34 and 35. Table 3-5 compares the activation parameters and rates for these systems. Such a comparison must, however, be approached cautiously due to the large uncertainties in the activation parameters determined for 34 and the suggestion of an alternative mechanism for that species.


Table 3-5. Activation Parameters and Relative Rates for Thermal Isomerization of 34, 36 and 37.

Ea = 38.6+0.6 kcal/mol Ea

/-Log A = 13.3 36

F Ea = 24.1+3.9 kcal/mol
Log A = 7.6+1.9 34

F kl52.50 = 4.47x10-4 s-1


ID # k(152.50),s krel

36 3.25 x 10-7 1

34 1.68 x 10-5 52

37 4.47 x 10-4 1400


Fluorodiiodomethane, CHFI2, has been shown to be a useful reagent for the preparation of monofluorocyclo48
propanes. Hine et al. reported obtaining CHFI2 in 32% yield from the reaction of neat iodoform heated to 1200C and treated with solid mercuric fluoride followed 52
by distillation of the product at reduced pressure.

2CHI3 + HgF2 2CHFI2 + HgI2

It was thought that a variation of this method might produce better results. An apparatus which would allow the reaction to be run at 1201C and under reduced pressure provided the possibility for continuous removal of the product as it was formed, thus minimizing the formation of CHF2 I and CHF3.

As expected, this procedure gave CHFI2 in an improved yield of 47%. Also found were the unexpected products diiodomethane and difluorodiiodomethane.



H5F 2

CHI3 1200 CHFI2 + CH2 2 + CF2 2

47% 6% 6%

That the two minor products were formed in an essentially aprotic medium and appear in equal amounts suggests that they are secondary products resulting from the disproportionation of CHFI2.

2CHFI2 + CH2 I2 + CF2 12

Interest in CF2 12 arose from its possible synthetic utility (Figure 4-1). Besides having potential as a difluorocarbene precursor, CF212 was expected to undergo free radical additions to olefins to produce adducts 51 which in turn might be reduced to primary gem-difluorides 52 or reductively coupled to yield gem-difluorocyclopropanes 53.


F 2

>=< + CF 2I2




51 52

Figure 4-1. CF 2I2 as a Synthetic Reagent.

A literature search on CF 2I2 revealed another

surprising fact. Contrary to what its simple formula would suggest, CF2I2 has previously been available in only small amounts via difficult and low-yield procedures involving the y-irradiation of CF3I 53,54 or the addition of difluorocarbene to molecular iodine.55-58 Clearly any routine use of CF 2I2 requires a more practical source.


Results and Discussion


As expected from analogy to the preparation of

CHiFI 2 from CHI 31'5 treatment of the readily available CI 459,60 with HgF 2 produced CF 2,2 in fair yield.

25-110 0C
CI 4 + HgF 2 27% CF2 12 + HIq2

An attempt to carry out this transformation using a 9:1 molar ration of KF/HgF 2 gave unsatisfactory results. Properties

Difluorodiiodomethane is a pale yellow liquid which at room temperature or upon exposure to light quickly acquires the characteristic burgundy color of I 2-containing solutions. It boils at 39-410C (80 torr) and has a vapor pressure of 28 torr at 251C. Although refrigeration is recommended, only minor decomposition (<2%) is observed when CF 2 12 is stored at room temperature in a pyrex flask for several weeks. It is stable indefinitely at -781C and is miscible with most organic solvents. Detailed spectral data are given in the experimental chapter.



Photolysis of CF 2I2 in the presence of 2,3-dimethyl2-butene 54 under various conditions (see experimental) gave 55 and 56. In the absence of olefins, photolysis of CF212 yields 57 and 12.


hv 55 56


hv 57

These observations are most easily rationalized through the processes shown in Figure 4-2.




I -i \-/5

_ F 2
:CF2 A


Figure 4-2. Mechanisms of CF 2I2 Photolysis.

Initial photodissociation of CF2I2 produces a difluoroiodomethyl radical and an iodine atom. The CF2I then has several possible fates. It can be captured by the olefin to give a radical intermediate 58 which then loses a hydrogen atom to give 56. It can lose a second iodine atom to produce difluorocarbene which is then trapped by the olefin to give 55. It can couple to another CF2I to give 57. Finally, it can couple to an iodine atom to regenerate the starting material.


The highest yields of 55 obtained photochemically

were through the use of Hg or NaOH(aq) as iodine scavengers.

CF 2 12 F2 C
.... h-v S +/

Hg 25% 29%

Analogous reactions have been reported for CH212 and
CHF 12.

Less satisfactory results were obtained when 1-hexene

was reacted under these conditions. Only a trace amount of the desired cyclopropane 60 was detected.

CF 21

59 60 2



Difluorodiiodomethane is not a good thermal source of difluorocarbene. When a sample of CF 212 was heated at 870C for 32h with excess 54, no significant reaction was observed and the CF 212 was recovered in >85%. Simmons-Smith reactions

When CF 2I2 was used in place of CH212 for a modified Simmons-Smith reaction,63 54 was converted to 55 in a very modest yield (3% by NMR). A recent report that SimmonsSmith cyclopropanations with CH212 are facilitated by ultrasonic irradiation64 suggested that a similar enhancement might be observed for CF212. When the reaction of 54 with CF 212 and Zn was carried out in a sonicator, the yield was improved more than 12-fold.

CF212/Zn 2 38%


54 55


Free-radical additions

As expected, CF212 proved to be an excellent reagent for free-radical chain additions to alkenes. Its addition to 1-hexene and methyl propenoate proceeded smoothly in 73% and 47% isolated yields respectively. These reactions have analogy in the addition of CF2Br2 to alkenes65 but are cleaner in that telomerization is not a significant side reaction.

(BzO) 2


59 61


62 63

Similar reactions of CF 212 with internal olefins such as trans-4-octene and cyclohexene were also quite successful. However, its attempted addition to 54 resulted in the exclusive formation of 56, likely via diversion of the expected radical intermediate 58 by H atom abstraction.


C2 2

(BzO) CH3 (CH2 2CHICH (CH2 2CH3
64 84% 65 CF 2 I

2 2

(BzO) 2 CF2 F2

66 65% 67 68

(67/68 = 70:30)

CF2 2

(BzO) 2 54 95% 56

Reduction of radical adducts

With the radical adducts 51 so readily accessible,
it seemed worthwhile to devise methods for their conversion to primary gem-difluorides 52 or gem-difluorocyclopropanes 53. Analogous reductions of alkyl iodides to alkanes66 and intramolecular reductive couplings of a,y-diiodides to cyclopropanes67-69 are known.


Treatment of 1,1-difluoro-l,3-diiodoheptane 61 with Zn in refluxing ethanol gave 1,1-difluoroheptane 69 and l-butyl-2,2-difluorocyclopropane 60 along with a variety of mostly unidentified side products. A cleaner reaction occurs when it is carried out in THF and the relative amount of 69 produced is enhanced.

69 60

F2 _____ _F2 Others

61 .> 36% 23% 41%

ZV\k CF2I + Zn

S52% 25% 23%

(Relative yields)

Reductive couplings of the type 61 + 60 are especially interesting and useful in that they provide a route into gem-difluorocyclopropanes that does not require difluorocarbene addition. Consider, for example, ester 70.



CO 2 CH 3 CO2CH3 + 2 CO2CH3

70 62

One would retrosynthetically view this as having come from CF2 addition to methyl propenoate 62. Unfortunately, all of the usual methods for formal CF2 addition to alkenes are notoriously unsuccessful with electron deficient olefins.1,70 An alternative route to 70 would involve radical addition of CF 212 to 62 followed by reductive coupling of the resulting diiodide 63.

CO 2CH3 F2

62 63 70



Difluorodiiodomethane is now available in quantities that make it a practical laboratory reagent. It can be produced in fair yield by the treatment of CI4 with HgF2.

As a difluorocarbene synthon, CF2 12 has been used for

photochemical and Simmons-Smith cyclopropanations. Although the generality of these reactions has not been verified, fair yields of gem -difluorocyclopropane 55 have been achieved with alkene 54.

Difluorodiiodomethane undergoes facile free-radical additions to alkenes. Isolated yields of 47 84% are obtainable from the addition of CF2 12 to various monosubstituted and disubstituted olefins.

8 F

54 55



NMR spectra were recorded on a Nicolet NT-300, a Jeol FX-100, a Varian XL-100 or a Varian EM 360L spectrometer. 1H and 13C chemical shifts (6) are reported as ppm downfield of internal TMS. 19F chemical shifts (f) are reported in ppm upfield of internal CFC13. All NMR spectra were obtained at ambient temperature in CDC13 unless otherwise specified. Where 19F NMR yields are reported, a predetermined amount of m-bromobenzotrifluoride (Columbia Chemical Co.) was used as an internal standard. Infrared spectra were recorded on a PerkinElmer 283B spectrophotometer from films of neat liquids between KBr plates. Mass spectra and exact masses were determined on an AEI-MS 50 spectrometer at 70 eV and are reported as m/q(% rel. base)fragment. Unless otherwise specified all photolyses were carried out in a Rayonet photochemical reactor with 350 nm source lamps. Preparation of 1,1-Dideutero-2,2-difluorospiropentane, 19 Into a small carius tube were placed 1.16g of CF3HgPh 39 (3.34 mmol)., 1.58g of Nal (10.5 mmol), 0.18g of (dideuteromethylene)cyclopropane40 (3.21 mmol), about 1.5 ml of benzene (freshly distilled from CaH2 onto molecular 59


sieves), and a few mg of Bu4N IG. The tube was sealed under vacuum and suspended in a preheated oil bath equipped with a mechanical stirrer. The tube was situated such that the blades of the stirring propeller could vibrate against it and provide agitation for the reacting mixture. After 48h of heating at 85-88*C, the tube was opened and all volatile material was vacuum transferred to a small flask. The product 19 was isolated by preparative GC (10% tricresylphosphate, 800C, 15' x 0.25", He flow = 60 nL/min) in 40% yield. 1H NMR: 6 1.15 (m)

19F NMR: 135.69 (m)

13C NMR: 6 114.56 (t,J = 285.2 Hz, CF2)

5 15.3 (m, CD2)

6 27.6 (v weak, tentatively identified as spiro C)*

6 6.47 (s, C3 and C4)

IR: 700, 740, 960, 1060, 1210, 1255, 1450, 1550, 3025, 3100 cmi MS: High resolution mean of 11 scans

M= 106.05571, std dv + 0.00248 (23.4 ppm)

Calculated C5H4F2D2 = 106.05631

dev = -0.00060(-5.7 ppm)

compund 13
Model compound 71 shows a spiro C resonance at 5 25.1. Henryk Koroniak, unpublished results, 1982, Univ. of Fl.

71 = 0/F


Thermolysis of 1, l-Dideutero-.2,2-difluorospiropropane 19

The thermolysis of 19 was carried out in a wellconditioned, spherical, pyrex vessel of approximately 200 mL capacity, submerged in a stirred, thermostated, molten salt bath (eutectic mixture 50/50 by weight of NaNO2 and KNO 3, mp = 1500). The temperature was measured using a chromel-alumel thermocouple in conjunction with a Tinsley type 3387E potentiometer. The thermocouple was immersed in a well next to the thermolysis vessel. The temperaturewas controlled to +0.10C using a Hallikainen (now TOTCO) Instrument Thermotrol proportional controller with a model 1256A platinum resistance probe. Product identification was done by standard spectroscopic methods on material from preparative runs in which 40 torr of 19 were expanded into the pyrolysis vessel through a contiguous vaccum line and thermolyzed for 5 to 6 halflives. A mixture of 23, 24 and 27 was purified by preparative GC (10% ODPN, 20' x 0.25", 350, He flow = 30 mL/min) Compounds 23, 24, and 27 were identified by their NMR spectra which were similar to those of the undeuterated 13 (See discussion in text). For kinetics, a mixture of approximately 10 torr of 19 and the internal standard pentane, was stirred for several hours in an isolated section of the vacuum line and checked by GC for homogeneity before expanding the mixture into the thermolysis vessel.


The kinetic run was sampled at convenient intervals by removing small fractions of the pyrolysis mixture by expansion through a small section of the vacuum line into a gas storage bulb and diluting with argon. This sample bulb was removed and used to make multiple GC injections through a gas sampling valve. A HewlettPackard 5710A gas chromatograph with a flame ionization detector was used in conjunction with an HP-3380S integrator for the anlaysis. Baseline resolution of peaks was observed. Each point in the rate constant is an average of at least 3 GC runs. The rate constant plot contained 6 points and was derived by a linear least squares analysis of the experimental data yielding a correlation coefficient of 0.9998.

Comparative Rate for Thermolysis of
1,!7-Difluorospiropentane 12

For comparison purposes, a thermolysis of the

undeuterated material 12 was carried out in the manner described above at the same temperature on the same day. Samples of the reaction mixture taken at six times gave a plot with a correlation coefficient of 0.99989. Preparation of cis,cis-l-Fluoro-.2-methyl-3-vinylcyclopropane 31

The following procedure was patterned after that of

Nishimura and Furukawa.48 A 50-mL 3-necked flask was equipped with a nitrogen inlet stopcock, a dry ice condenser, a magnetic stirring bar, and a rubber septum. The outlet


of the condenser was connected to a mineral oil, U-tube bubbler. The glassware used for the apparatus was dried in an 1201C oven and assembled hot, then cooled to ambient temperature under a slow flow of dry nitrogen. The nitrogen 71
flow was discontinued and, with standard syringe technique, the following reagents were placed in the reaction flask: 20 mL 1.6 M diethylzinc in toluene (32 mmol, Aldrich, Cautior! extremely pyrophoric), 4.0 mL cis-l,3-pentadiene (40 mole, Fluka), and 2.7 rL CHFI252 (8.5g, 30 mmol). The condenser was filled with dry ice and the reaction mixture was magnetically stirred. Reaction progress was monitored by anaerobically syringing out a small sample of the mixture into a septum-capped NMR tube and examining it by 19F NMR for disappearance of CHFI2 ( i02, d, J=48 Hz). The sample could then be placed back into the reaction mixture. No reaction was observed to occur at room temperature after 16h. The reaction mixture was heated to 940C for 15h, and an additional

1 mL of neat diethylzinc72 was added by syringe followed by another 19h of heating. The reaction flask was cooled to 00C and the mixturewas slowly quenched (caution!) with approximately 30 mL of 0.75 M HCI and transferred to a separatory funnel. The organic layerwa's washed thrice with 0.75 M HCI and thrice with distilled water then dried over CaCI2. The resulting solution was examined by 19F NMR and yields were determined by integration. Samples of the products were isolated by preparative PC (20% Diisodecylphthalate, 10' x 0.25", 730, 30 mL/min He flow)


31, 35%
H NMR 6 4.63 (dt, J=65.5 Hz, J-5.8 Hz, 1H, FC-H)

6 1.12 (m, 3H, CH3)

6 1.48 1.60 (m, 1H, allylic)

6 1.02 1.16 (m, 1H, MeC-H)

6 5.:66(dddd, J=17.2 Hz, J=10.4 Hz, J=9.3 Hz,

J=1 Hz, 1H, internal vinyl)

5 5.24 (dd, J=17.2 Hz, J=2.1 Hz, 1H, terminal

vinyl cis to cyclopropyl)

6 5.15 (dd, J=10.4 Hz, J=2.1 Hz, 1H, terminal

vinyl trans to cyclopropyl). 19F NMR:4237.1 (dt, J=65.6 Hz, J=7 Hz) 1C NMR:6130.6 (d, J=8.5 Hz, internal olefinic) 6117.0 (s, terminal olefinic)

6 74.6 (d, J=221 Hz, CHF)

6 22.7 (d, J=9.8 Hz, allylic)

6 15.6 (d, J=11.0 Hz, C-Me)

6 5.8 (d, J=8.6 Hz, CH3)

IR: 285(wk), 330(wk), 710(wk), 730, 780, 860, 905, 970, 990,

1015, 1045, 1075, 1115(wk), 1145, 1180, 1230, 1330,

1390, 1460, 1635, 1695, 2880, 2940, 2960, 2990, 3010,
3040, 3085 cm 1 MS: 100 (30.2) M

85 (100) M+-CH3

79 (10.6) 72 (17.3)

65 (18.8) C5H5


59 (37.7) C2H2F

53 (17.4) 51 (11.4) 46 (9.3) 41 (20.7)

39 (46.2) C3H3 27 (20.7) C2 H3

33, 21%
H NMR 6 4.65 (dtd, J=66 Hz, J=7 Hz, J= 4 Hz, FC-H)

6 4.8 5.9 (m, vinylic) 19
19F NMR #223.5 (ddddd, J=65.0 Hz, J=23 Hz, J=12.3 Hz, J=5.0 Hz, J=1.1 Hz)

At least four isomeric diadducts which were not separable by GC were also observed (24%). The chemical shifts and coupling patterns from the 19F NMR spectrumof a mixture of these were consistent with the presence of approximately 11-12% each of the syn, syn diadducts 72 and 73 and approximately 1% each of the syn, anti diadducts 74 and 75 (Table 5-1).



72 73 74 75


Table 5-1. 19F NMR Data from Mixture of Diadducts.

Description % rel

207.0 dtd, J=64.0 Hz, J=21,4 Hz, J=10.5 Hz 1

208.0 dtd, J=64.2 Hz, J=21.7 Hz, J=10.0 Hz 2

224.7 m, at least 2 different F's, unresolved, 47

syn environment for each i.e., dt

J 65 Hz, J=10 Hz

239.0 m, several different F's, unresolved, 50

syn environment for each i.e., dt

J=65 Hz, J=10 Hz

Thermolysis of. cis,Cis-l-F.luoro-2-methyl-3-vinylcyclopropane 31

The gas phase thermolyses of 31 were carried out as was described for that of 19 above except the internal standard used was methylcyclohexane and GC analyses were performed using a Hewlett-Packard 5790A series gas chromatograph with gas injection system and flame ionization detector in conjunction with a Hewlett-Packard 3390A integrator. The column was 10' x 0.125", 10% B,B-oxydipropionitrile on chromosorp WHP at 600C, with 5.5 mL/min N2 flow. A preparative run showed, assuming uniform detector response, 99% material


balance by internal standard after 7.2 half-lives. A sample of the gaseous reaction mixture was taken directly from the thermolysis vessel without condensing and examined by GC/MS. Only C6H8 products, starting material, and internal standard coud be detected. One preparative run was condensed directly into anNMR tube, which was then sealed at -1960C. The sample was maintained at -500C or below until NMR analysis was complete. Only elimination products were observed. These were identified by their 1H and 13C NMR spectra.

Preparation of endo- and exo-8-Fluorobicyclo[5.1.0]oct-2-ene, 34 and 35.

A procedure similar to that described above for the preparation of 31 was used. Intoan oxygen-free apparatus were syringed 2.8 ml of 1,3-cycloheptadiene (26 mmol, Aldrich), 12.5 ml of 1.6 M diethylzinc in toluene (20 mmol, Aldrich)
and 5.0g of CHFI52 (17.5 mmol). The magnetically stirred mixture was heated at a pot temperature of 75-900C for 43h. In addition to 8% unreacted CHFI2, the following products were obtained by preparative GC (20% carbowax, 20' x 0.25", 135cC, 60 mL/min He flow)

34, 71%
H NMR: 6 4.50 (dt, J=66 Hz, J=6 Hz, 1H, FC-H)

6 4.30 4.75 (m, 2H, vinylic)

6 0.85 -2.42 (m, 8H)


19F NMR: 234 (dt, J=66.9 Hz, J=6.7 Hz)

C NMR:6 132.4 (s, C3)

6 122.2 (d, J=5.1 Hz, C2) 6 73.7 (d, J=220.9 Hz, C3)

6 29.0 (s, C4) 6 23.7 (s, C5)

6 19.6 (d, J=7.0 Hz, C6) 6 18.3 (d, J=9.6 Hz, C1)

6 18.2 (d, J=10.9 Hz, C7)

IR: 585, 690, 710, 880, 1005, 1040, 1115, 1125, 1195,
1225, 1410, 1450, 1665, 2870, 2940, 3005 cm MS: M+ 126.0845 std dev = + 0.00065 (+ 5.2 ppm)

mean of 10 scans

Calculated for C8H11F = 126.08448

dev = 0.00003 (+ 0.2 ppm)

126 (15.9) M+

111 (21.0) 98 (17.5) 97 (37.2) 93 (57.1) 91 (24.0) 85 (19.0) 79 (26.1) 67 (11.8) 65 (11.7) 59 (13.4) 39 (23.0)


35, 9%
H NMR: 6 4.22 (dt, J=64.6 Hz, H=2.2 Hz, 1H, FC-H)

6 5.32 5.58 (m, 1H, C2-H)

6 5.62 5.84 (d br s, 1H, C3-H)

6 1.0 2.15 (br unresolved m) 19
19F NMR:c199.6 (dt, J=64.8 Hz, J=44 Hz) 13C NMR:6130.6 (d, J=3.1 Hz) 5125.4 (d, J=0.7 Hz)

6 79.7 (d, J=231.1 Hz, C8)

6 30.4 (s, C4) 6 26.5 (s, C5)

6 26.1 (d, J=9.0 Hz, C1) 6 25.0 (d, J=1.7 Hz, C6)

6 24.2 (d, J=11.2 Hz, C7)

IR: 690, 710, 1105, 1380, 1450, 1640, 2860, 2920, 2960,
3010 cm1

MS: M 126.08473 std. dev. = + 0.00065 (+ 1.5 ppm)

mean of 9 scans

Calculated CH F = 126.08448

dev. = -0.00025 (-2.0 ppm)

126(5.7)M+, 67 (100,base) C5H7, and many others


39, 7% H F
a /


C 2

IH NMR: 6 4.59 (dt, J=66.2 Hz, J=6.2 Hz, 2H, Ha)

6 1.58 1.80 (br s, 6H, H's on C4 and C 5)

6 0.91 1.03 (m, 2H, Hb) 6 0.68 0.81 (m, 2H, H c)
19F NMR:c237.4 (dt, J=66.2 Hz, J=5.9 Hz) 13C NMR:6 73.0 (d, J=223.4 Hz, C1)

6 23.2 (s, C5)

6 17.9 (d, J=8.5 Hz, C4)

6 15.0 (d, J=11.0 Hz, C3)

5 10.6 (dd, J=10.4 Hz, J=7.9 Hz, C2)

IR: 530, 565, 605, 615, 740, 775, 785, 835, 930, 990,

1025, 1070, 1140, 1190, 1220, 1360, 1405, 1455, 2870,

2930, 2960, 3010, 3050 cm1

MS: M 158.0910 std. dev. = + 0.0018 (14.5 ppm)

mean of 13 scans

Calculated for C9HI2F2 = 158.09070

dev. = +0.00029 (+ 1.9 ppm)

86 (100) base peak


40, 5%

This material was not isolated by preparative GC, but was observed in the reaction mixture by NMR 19F NMR 6237.3 (dt, J=65.7 Hz, J=6 Hz, endo F)

6203.5 (dt, J=64 Hz, J=21 Hz, exo F)

Thermolyses of endo- and exo-8-Fluorobicydlo[5.1.] o]ct-2-ene

The procedure and apparatus were the same as those described for the thermolysis of 19 except that because of the low vapor pressures of 34 and 35, the sample bulbs had to be heated to about 50'C in order to maintain good reproducibility of GC injections. The sampling bulbs were wrapped with Thermolyne, BriskHeat- silicon rubber embedded flexible electrical heating tape and covered with a blanket of plaster of Paris. All portions of the sampling area, kinetics line, and gas injection system were maintained at about 50'C. A maximum vapor pressure of 4 to 5 torr could be obtained and was used for all kinetic and preparative runs. One preparative run was carried out on 34 with

5 torr of nonane present as an internal standard and, assuming uniform defect response, >95% material balance was observed after 7.5 half-lives. Several preparative runs were performed consecutively and condensed into a small flask. The resulting products were isolated by preparative GC (20% dinonylphthalate, 1350, 20' x 0.25", 60 mL/min He flow).


1H NMR: 8 5.78 (m, 1H, C2-H)

6 5.61 (dt, J=10.8 Hz, J=7.7 Hz, 1H, C 4-H) 8 5.46 (dd, J=21.0 Hz, J=4.7 Hz, 1H, C3-H)

5 2.34 (m, 2H, C8-H) 5 2.24 (m, 2H, C5-H)

6 1.67 (m, 2H) 5 1.51 (m, 2H)
13C NMR:S 161.3 (d, J=252.5 Hz, C1)

5 130.8 (d, J=2.0 Hz, C4)

5 122.9 (d, J=12.4 Hz, C3) 5 102.9 (d, J=27.1 Hz, C2)

5 30.0 (d, J=28.5, C8)

5 26.8 (s, C5) 5 24.8 (s, C6)

5 20.7 (d, J=4.9 Hz, C7) 19F NMR:4 90.9 (m) 44
IH NMR: 5 6.00 (dd, J=11.1 Hz, J=4.1 Hz, 1H)

5 5.77 (dd, J=12.0 Hz, J=4.] Hz, 1H)

5 5.72 5.55 (m, 2H) 5 2.78 1.42 (m, 6H) 13C NMR:S 131.5 (s, C3)

5 129.9 (s, C2)

5 125.6 (d, J=10.3 Hz, C4)

5 125.1 (s, C1)


6 90.4 (d, J=169.5 Hz, C6)

6 33.2 (d, J=22.8 Hz, C5) 6 28.2 (d, J=21.2 Hz, C7)

6 23.7 (d, J=8.0 Hz, C8)
F NMR: 172 (m)

Reaction of CHI3 with HgF2

Into two solid additiontubes were placed a total of 99.1 g of HgF2 (0.415 mol, Alfa Products). These were inserted into a 500-mL 3-necked flask containing a large stirring magnet and 349.3 g of solid CHI 3(0.8872 mol), and equipped with a distillation head. The system was evacuated to 50 torr, and the reaction flask was heated over an oil bath at approximately 1400C until nearly all the CHI3 had melted. The temperature was decreased to 1201C and the solid HgF2 was added gradually to the magnetically stirring CHI3 melt over a 40 minute period. A dark burgundy-colored distillate was collected at 600 to 800 in a 200-mL receiving flask which was cooled to 00. Approximately 60 mL of distillate (154q) were collected. This material was washed twice with aqueous Na2S203 and then fractionally vacuum distilled to produce lllg of CHFI2 (47%, BP=55-65', 50 torr) and 6% each of CF212(BP=29-33,50 torr), and CH212(BP=>750, 50 torr).


Preparation of CF 2I2

Into a 500-mL three-necked flask equipped with

distillation head and vacuum Hershberg stirrer were placed
142.4g of C14 (0.274 mol) and 69.6g of HgF2 (0.292 mol). The flask was closed off with a 125-mL pressure-equalizing addition funnel containing 100 mL of 1,2-dichlorobenzene.* The system was evacuated to 50 torr through a -78'C trap, and the dichlorobenzene was added slowly (exothermic as reactions begin!) until the heterogeneous system was wet enough to stir easily (ca. 80 mL). (In an alternative procedure a solid addition tube was used to slowly add HgF2 to a stirring solution of C14) Heating at 1001C produced a dark purple distillate which came over at 50-81'C (50 torr) into a receiving flask cooled to 01C. Distillation ceased after about 45 min. The trap contents and distillate were combined with those from a previous run (0.270 mol of CI4 and 0.273 mol of HgF2) and distilled under vacuum: yield 44.8g (27%); bp 39-41C (80 torr) Small amounts of crude CF212 can be purified (i.e., freed from 12 ) as needed by Other inert, high-boiling solvents (e.g. perfluorodecaline or n-C F 7SO 2F) can also be used.

Higher isolated yields can be obtained by claiming the
small amount of CF 2I2 remaining in the reaction mixture. This can be done by slowly pulling a vacuum above the stirring mixture at room temperature through a -781 trap followed by vacuum distillation of the trap contents.


trap to trap distillation under high vacuum through a 0,

-40, -78, and -1960C trap system. Pure CF21 2 is found in the -780C trap. It has a vapor pressure of 28 torr at


19F NMR:4 -18.6 (s) 13C NMR:S 2.29 (t, J=378 Hz) MS: 304 (28) M

285 (3.6) CF2I

254 (24) 1 2

177 (100) CF2I 158 (3.0) CFI

127 (36) I

50 (2.2) CF2
IR: 720, 1045, 1095 cm-1 Photolyses of CF2 2 Photolysis of CF2 12 with 2,3-Dimethyl-2-butene 54 and Hg

Into a small Carius tube were weighed 0.30g of CF2 12 (0.99 mmol), 1.33g of 2,3-dimethyl-2-butene (16 mmol), and

0.87g of Hg (4.4 mmol). The reaction mixture was thoroughly degassed, and the tube was sealed under vacuum. The mixture was photolyzed for 23h with occasional shaking. Afterward the tube was opened, and all volatile material was vacuum transferred and analyzed by 19F NMR. The product mixture contained 25% l,l-difluoro-2,2,3,3-tetramethylcyclopropane 55 and 29% 4,4-difluoro-4-iodo-2,3,3-trimethyl-l-butene 56. Pure samples of each were isolated by preparative gas chromatography(20% dinonyl phthalate on Chromosorb WHP, 20' x 0.25", 110C).



The 1H NMR (6 1.1, t, J=2 Hz) and the 19F NMR (0149) spectra were identical with those reported for 55. 73, 56
1H NMR: 6 5.09 (br s, 1H, vinylic)

6 5.05 (br s, 1H, vinylic)

6 1.85 (dd, J=3.3 Hz, J=1.5 Hz, 3H, allylic)

6 1.29 (t, J=0.8 Hz, 6H, methyls on C3) 19F NMR:4 41.5 (br s) 13C NMR:6 145.5 (br s, C2)

6 116.3 (s, C1)

6 114.5 (t, J=322 Hz, C4)

6 52.7 (t, J=15.9 Hz C3)

6 23.1 (t, J=3.7 Hz, methyls on C3)

6 21.3 (t, J=3.1 Hz, allylic methyl on C2) IR: 575, 770, 860, 925, 1070(s), 1215, 1385, 1450,
1640(w), 2960 cm1
MS: M = 259.98822 std. dev. = + 0.0027 (10.1 ppm)
Calculated for C7H 11F2I = 259.98736

dev. = 0.00086 (+ 3.3 ppm)

260 (0.1) M+

133 (71.3) M+-I

128 (10.7) 127 (8.0)

113 (21.2)

97 (15.4)


91 (29.1) 85 (11.0)

83 (16.7) M+-CF 2 1

77 (26.3) 73 (33.2) 70 (13.4) 69 (22.0) 67 (16.7) 65 (28.8) 61 (14.1) 57 (35.4) 55 (37.5) 53 (17.8) 51 (11.1)

43 (32.0) C 3 H 7 41 (100) C 3 H 5

39 (46.3) 29 (13.6) 28 (15.2) 27 (20.9)

Photolysis of CF 2 1 2 and 54 without Hg

In the absence of mercury, a photolysis performed as described above showed only 40% conversion of CF 2 1 2 and gave a similar product ratio.


Gas Phase Photolysis of CF 2I2 and 54

Into a 570 mL vycor photolysis bulb were vacuum

transferred 0.74 mmol of CF 2I2 and 2.1 mmol of 49. The system was thoroughly degassed and then the bulb was suspended in a Rayonet photolysis apparatus equipped with 254 nm source lamps. Almost immediately the reaction mixture darkened with 12 coloration. Photolysis was continued for 4h. All volatile material was vacuum transferred into a small plum-bottom flask and examined by 19F NMR: 19% 46, 14% 45 and two unidentified products 6% 103.9(s) and 2% 96.2(s).

Photolysis of Neat CF 2I2

Liquid CF2I2 was degassed and a sample was allowed to expand into an evacuated vycor photolysis bulb until its maximum vapor pressure at room temperature was obtained (28 torr). The bulb was closed off and the CF 212 was photolyzed at 254 nm as described above. Within minutes, 12 was seen crystallizing on the walls of the bulb. Afer 2h of photolysis, all volatile material was vacuum transferred to a small flask and examined by 19F NMR which showed ICF 2CF 2I (57, 453.2) and CF 2I2 to be the only 19F-containing species present in other than trace amounts. 57/CF212 1 0/1.


Photolysis of CF I and 54 at Elevated Temperature
2 2

A solution of 0.79g of CF212 (2.56 mmol) and 0.81g of 54 (9.6 mmol) was refluxed and photolyzed for 21h under a blanket of dry N2. A Sears 150W, 120V projector flood lamp was used as a light source mounted about 8 cm from the reaction flask. The 19F NMR spectrum of the crude reaction mixture showed 56 to be the major 19F-containing species in addition to small amounts of other unidentified products. No 55 or unreacted CF 212 was observed. Photolysis of CF 212 with 54 and Aqueous NaOH

Into a 25 mL round-bottom flask were placed 0.56g of

54 (6.7 mmol), 15 mL CH2Cl2 1.16g of CF212 (3.82 mmol) and

5 mL of 5M aqueous NaOH. The magnetically stirred reaction flask was equipped with a reflux condenser and the mixture was photolyzed for 23h. The two-phase liquid system was separated and the organic phase was dried over CaCl2 then distilled to remove most of the CH2Cl 2' Absolute yields were not determined; however, 19F NMR analysis showed four 19F-containing species (two unidentified) in the following relative amounts: 54% 56, 30% 55, 9% 80(s), 7% 38(s). Photolysis of CF I2 with 1-Hexene and Hg

Into a small carius tube were weighed 0.36g of CF 212 (1.2 mmol), 0.90g of 1-hexene (10.7 mmol) and 0.59g of mercury (2.9 mmol). The mixture was degassed and the tube was sealed under vacuum at -196C. After thawing, the


mixture was photolyzed and mechanically agitated for 18h. The tube was opened and the volatile material was examined by 19F NMR. The reaction yield was too low to be determined by NMR integration. A trace amount of l-butyl-2,2-difluorocyclopropane 60 was detected and identified by its Fourier transformed 19F NMR spectrum. AB quartet with further splitting midpoint $136.7, JAB=156.3 Hz, Ap=16.8 J(F H )=12.4 Hz aa
J(FbHb)-J(FbHc)=13.5 Hz


a c



Some trans, vicinal H-F couplings are also visible (1-3 Hz). The 19F NMR spectrum of 60 closely resembled that of model compound 76.



Attempted Thermal Reaction of CF2 I2 and 2,3Dimethyl-2-butene 54.

Into a small Carius tube were placed 0575g of CF 2I2 (1.89 mmol), and 0.817g of 54 (9.73 mmol). The solution was degassed and the tube was sealed under vacuum at -196'. After it was allowed to warm to ambient temperature, the tube was placed in a preheated oil bath and kept at 83-871C for 32h. Analysis by GC and NMR showed the only significant new component in the mixture was 2,3-dimethyl-l-butene which resulted from the I 2-catalyzed isomerization of 54.

Simmons-Smith Reactions of CF 2I2 Reaction of CF2 I2 with 54, Zn, and CuCl

The following procedure was patterned after that of 63
Rawson and Harrison. A 25-mL, three-necked flask was equipped with a spiral condenser, a rubber septum, an addition funnel, and two stopcocks for the inlet and outlet of N2* A solution of 0.463g of CF212 (1.16 mmol) in 1 mL of anhydrous diethyl ether was placed into the addition funnel. Into the reaction flask were placed 0.219g of Zn dust (3.35 mmol), 75
0.322g of freshly prepared CuCl, and 2 mL of anhydrous diethyl ether. The magnetically stirred mixture was warmed to gentle reflux for 15 minutes whereupon 0.5 mL of 54 (4.2 mmol, Aldrich) was added by syringe. Heating was temporarily discontinued while the CF 2I2 solution was added dropwise over 10 minutes. Reflux heating was resumed for 4.5h. All volatile material was vacuum transferred to a small flask. Yield of 55 was 3% by 19F NMR.


Reaction of CF 2I2 and 54 with Zn and Ultrasound

Into a 10-mL flask were weighed 0.36g of Zn dust

(5.5 mmol), 0.43g of 1,2-dimethoxyethane, and 0.56g of 54. The flask was equipped with a spiral condenser, nitrogen inlet and outlet stopcocks and an addition funnel containing a solution of 0.96g of CF 2I2 (3.16 mmol) in 0.66g of 1,2-dimethoxyethane. The system was purged with a slow flow of dry N2 and the reaction vessel was mounted inside a sonic cleaner (Bransonic 2 117V, 60 Hz, 40W, Branson Instruments Co.) at a depth that produced the maximum agitation. After 20 minutes of sonication, the CF 212 solution was added dropwise over 6 minutes and the heterogeneous reaction mixture was sonicated for an additional 3 hours while the bath temperature climbed slowly to 580C. All volatile material was vacuum transferred from the mixture and the clear, colorless solution was examined by 19F NMR. Yield of 55 was 38%.

Free-Radical Additions of
CF212 to Aikenes

Addition of CF212 to 1-Hexene

Into a 10-mL round-bottomed flask were weighed 0.325g of CF212 (1.07 mmol), 1.108g of 1-hexene (13.2 mmol), and 10mg of benzoyl peroxide. After refluxing the magnetically stirred solution for 10.5h, excess olefin was removed by rotary evaporation, and the pale pink residue was purified


by flash chromatography,76eluting with hexane, to yield

0.304g of l,l-difluoro-l,3-diiodoheptane 61; 73%. H NMR: 6 0.95 (t, J=6 Hz, 3H)

6 1.2 3.1 (br m, 6H)

6 2.8 3.6 (m, 2H)

6 4.32 (quint, J=7 H,lH)
19F ~MR:4 36.18 (midpoint, AB, JAB= 172 Hz, A=1.75)

downfield F (dd, J=18 Hz, J=10 Hz)

upfield F (dd, J=17 Hz, J=14 Hz)

Also visible was 43% of the opposite regioadduct


S38.1 (midpoint AB of d, JAB=172, A=1.75, JHF10 Hz) 13
13C NMR:6 99.6 (t, J=316 Hz, C1)

6 58.2 (t, J=40 Hz, C2)

6 39.2 (s) 6 31.5 (s)

6 25.3 (br s, C3)

6 21.7 (s)

6 13.9 (s, C7)

IR: 895(s), 1060, 1165, 2870, 2880, 2930(s), 2970(s)cm-1 MS: 388 (1.0) M

261 (4.8) M+-I 41 (100) C3H5

and many smaller fragments


Addition of CF2 I2 to Methyl Propenoate

A mixture of 1.02g of methyl propenoate (11.8 mmol, MCB, freshly distilled from phenothiazene onto molecular sieves), 2.15gc of CF2 12 (7.07 mmol), about 4 mL of benzene and 0.1g of benzoyl peroxide (0.1 mmol) was treated as described above for 19h at 90-1060C to yield 36% of unreacted CF2 12 and 1.29g of methyl 4,4-difluoro-2,4-diiodobutanoate 63 47%. H NMR: 6 4.55 (dd, J=10 Hz, J=3.8 Hz, 1H)

6 2.84 3.95 (m, 2H)

6 3.77 (s, 3H)
19F NMR: 40.1 (midpoint, AB, JAB=176 Hz, A$=0.3, JHF=15 Hz) 1C NMR:6 170.3 (s, C=O)

6 97.0 (t, J=315 Hz, CF2I)

6 54.5 (t, J=21 Hz, CH2)

6 53.3 (s, CH3)

6 7.8 (t, J=2.4 Hz, CHI) MS: 359 (1.3) M-OCH
331 (0.3) M-CO2CH3

263 (20.8) M-I

254 (0.8) 12
136 (100)

127 (24.5) I

77 (27.7) C3H3F2 64 (3.0) C2H2 2 59 (17.3) CO2CH3


31 (13.9) OCH3

and other fragments No M was observed.

Addition of CF2 I2 to trans-4-Octene

A mixture of 0.70g of CF2 12 (2.3 mmol), 0.81g of

trans-4-octene (7.2 mmol) and 0.10gc of benzoyl peroxide was allowed to react as described above at 1100 for 24h to yield 84% of a 1:1 mixture of the two possible diastereoisomeric 4-(difluoroiodomethyl)-5-iodooctanes 65. The 1H-NMR of the reaction mixture shows the hydrogens on C5 as multiplets 64.45.
19F NMR: 36.6 (midpoint, AB of d, A$=3.02, JAB=187 Hz)

downfield F (d, JHF=14 Hz) upfield F (d, J HF=17.5 Hz)

37.5 (midpoint, AB of d, A6=3.17, JAB=188 Hz

downfield F (d, JHF=13.5 Hz)

upfield F (d, JHF=15 Hz)

Radical Addition of CF2 I2 to Cyclohexene

A mixture of 1.00g of CF2 12 (3.3 mmol), 1.20g of cyclohexene (14.6 mmol), and 0.03g of benzoyl peroxide (0.1 mmol) was allowed to react at 90-970C for 20h to yield 61% of a mixture consisting of 65% trans-1-(difluoroiodomethyl)-2-iodocyclohexane 67, 28% cis-1-(difluoroiodomethyl)-2-iodocyclohexane 68, 4% unreacted CF212, and 3% of an unidentified product (642, midpoint, AB .7=10 Hz,


A =l ppm). The major diastereomeric products were isolated by preparative GLC. The column was 20% OV-210 on Chromasorb WHP 60/80, 10' x 0.25", at 1500C. Some slight decomposition was noted under these GC conditions. The retention times were trans = 27 min, and cis = 33 min. 67
19F NMR:4 34.5 (d, J=10.3 Hz) H NMR: 6 4.27 (ddd, J=8.5 Hz, J=7.0 Hz, J=4.5 Hz, 1H, H-CI)

6 1.1 2.5 (m, 9H) 13
C NMR:6 108 (t, J=319 Hz, CF2I)

6 57.3 (t, J=16.5 Hz, C1)

6 38.3 (s) 6 27.9 (s)

6 27.5 (dd, J=3.6 Hz, J=2.5 Hz)

6 26.2 (s) 6 22.7 (s)

19F NMR: 42.4 (midpoint, AB, J AB=172 Hz, AQ=5.12)

downfield F (d, JHF=13.6 Hz)

upfield F (d, JHF=12.8 Hz) IH NMR: 6 4.75 (br s, 1H, H-CI)

6 1.1 2.3 (m, 9H)
13C NMR:6 105.7 (dd, J=321 Hz, J=320 Hz, CF2I)

6 57.5 (t, J=9.8 Hz, C1)

6 37.2 (s)

6 32.1 (dd, J=3.7 Hz, J=2.4 Hz, C6)


6 24.8 (s)

6 23.6 (t, J=2.4 Hz, C2)

6 22.2 (s)

Addition of CF2 12 to 2,3-Dimethyl-2-butene 54

Refluxing 0.56g of CF2 12 (1.8 mmol), 0.86g of 54

(10 mmol), and 0.02g of benzoyl peroxide (0.08 mmol) for 24h gave 56 as the only F-containing product. More than half of the CF212 was recovered unchanged. The yield was 95% based on converted CF2 12 and 19F NMR integration.

Zinc Reductions of 1,1-Difluoro1,3-diiodoheptane 61

Into a 10-mL flask were placed 0.180g of 61 (0.464 mnmol),

1 mL of THF (freshly distilled from LiAlH4 and storred over CaH2 until use), 0.083g of Zn dust (1.28 rmmol, acid washed) and a small crystal of ZnCl2. The magnetically stirred mixture was refluxed for 8.5 h. All volatile material was then vacuum transferred and analyzed by 19F NMR. The relative yields of F-containing products were 52% 59, 25% 60 and 23% of other mostly unidentified products. When a similar reaction was carried out in ethanol, the relative yields were 36, 23 and 41% respectively. The 19F NMR spectrum of 60 was identical to that reported above for 60 prepared from photolysis of CF2 I2 with 1-hexene. 59
196.1 (dt, J=57 Hz, J=17 Hz)
F NMR:4 116.1 (dt, J=57 Hz, J=17 Hz)


The thermal rearrangement of 1,1-difluorospiropentane to 2,2-difluoromethylenecyclobutane was shown to proceed via two competing mechanisms. Initial C1-C2 bond cleavage is three times faster than initial C4-C5 bond cleavage.

The kinetics for the thermal isomerization of

cis,cis-l-fluoro-2-methyl-3-vinylcyclopropane indicate that a cis-monofluoro substituent in the 1 position lowers the activation energy for the 1,5-homodienyl hydrogen shift by 3.3 kcal/mol relative to the hydrocarbon. This value is less than half of the Ea lowering observed for the rearrangement of the gem-difluoro species.

Difluorodiiodomethane is now available in practical quantities through the treatment of CI4 with HgF2. This versatile reagent has been used in photochemical and SimmonsSmith cyclopropanations and undergoes facile free-radical additions to alkenes.



Table A-1. Kinetic Data for Thermolysis of 12 at 339.6"C.

min % 12 % 13 % std

1.9 38.14 61.50
1.9 38.36 61.64
1.9 39.19 61.37

26.2 37.13 1.54 61.02
26.2 37.20 1.50 60.80
26.2 37.28 1.35 60.81

52.6 34.28 2.93 62.32
52.6 34.14 2.90 62.22
52.6 34.75 3.03 61.78
52.6 34.48 3.01 61.71
52.6 34.40 2.98 62.13

81.7 32.48 4.68 62.26
81.7 32.09 4.62 61.63
81.7 32.25 4.47 62.27
81.7 32.21 4.51 62.45

163.7 27.31 8.22 62.86
163.7 27.52 8.33 63.21
163.7 27.36 8.33 63.10
163.7 27.50 8.28 63.20

201.7 24.98 9.73 63.41
201.7 25.20 9.76 63.38
201.7 25.09 9.89 63.41
201.7 25.26 9.90 63.52

244.5 23.60 11.47 63.75
244.5 23.29 11.29 64.20
244.5 22.53 11.32 64.06
244.5 23.37 11.34 64.40



Table A-2. Kinetic Data for Thermolysis of 19 at 339.60 C.

min % 19% % 23+24+27 % std

1.5 32.15 -67.20
1.5 32.92 -67.08
1.5 32.42 -67.58
1.5 32.68 -66.62

20.6 31.70 0.76 67.54
20.6 31.37 0.68 67.95
20.6 31.56 0.77 67.50

62.3 28.73 2.55 67.96
62.3 28.37 2.49 68.66
62.3 29.04 2.41 68.06

92.4 26.63 3.78 68.24
92.4 27.21 3.78 67.94
92.4 27.09 3.73 68.62

128.1 25.99 4.71 68.38
128.1 26.09 4.67 68.11
128.1 25.70 4.54 67.95

168.0 23.57 6.37 69.43
168.0 23.58 6.40 68.58
168.0 23.91 6.40 68.74
168.0 23.18 6.38 69.22

263.4 19.72 8.67 70.67
263.4 19.62 8.88 70.51
263.4 20.04 8.95 70.09


Table A-3. KinetiC Data for Thermolysis of 31 at 152.80C.

min % 31 % Pi % P2 % P3 % P4 % std

0 92.912 0.631 0.649 0.025 0.022 3.450
0 92.886 0.623 0.632 0.018 0.008 3.450
0 92.840 0.609 0.584 3.462

15 84.579 7.758 0.852 1.231 0.098 3.381
15 84.558 7.684 0.858 1.211 0.101 3.372
15 84.125 7.563 0.944 1.243 0.140 3.401

30 75.877 14.719 1.086 2.362 0.204 3.465
30 76.203 14.722 1.173 2.348 0.195 3.392
30 75.604 14.475 1.290 2.260 0.206 3.436

45 72.896 15.576 2.564 2.359 0.260 3.660
45 72.973 15.350 2.728 2.246 0.225 3.685
45 73.002 15.082 2.875 2.190 0.168 3.755

60 62.364 25.782 1.791 3.867 0.381 3.463
60 62.489 25.574 1.808 3.859 0.324 3.507
60 62.523 25.521 1.867 3.921 0.411 3.499

75 59-008 26.651 3.097 4.262 0.503 3.707
75 58.964 26.294 3.166 4.193 0.489 3.702
75 58.865 26.008 3.331 3.970 0.576 3.726

105 45.212 39.386 2.596 6.109 0.601 3.483
105 45.191 39.231 2.522 6.190 0.587 3.470
105 45.024 38.977 2.593 5.864 0.300 3.471


Table A-4. Kinetic Data for Thermolysis of 31 at 157.1'C.

min % 31 % Pi % P2 % P3 % P4 % std

12 82.833 7.533 1.007 1.260 0.182 4.091
12 81.955 7.828 1.195 1.407 0.249 4.120
12 80.591 7.820 1.497 1.459 4.267

24 73.921 13.644 1.425 2.503 4.137
24 73.571 14.576 1.462 2.411 0.155 4.060
24 73.360 14.448 1.437 2.448 0.250 4.096

36 65.692 20.788 1.757 3.772 0.441 4.184
36 65.491 20.649 1.692 3.698 0.342 4.192
36 65.397 20.580 1.901 2.867 4.326

48 59.569 26.269 1.945 4.462 0.327 4.206
48 59.183 25.929 1.967 4.506 0.364 4.285
48 59.538 26.124 1.841 4.416 0.296 4.258

60 58.565 23.186 5.230 3.602 0.629 4.826
60 57.992 23.120 5.124 3.463 0.460 4.925
60 59.150 23.133 5.527 3.557 0.488 4.944

72 51.575 27.697 6.454 4.088 0.498 4.857
72 51.966 27.355 6.691 3.878 4.881
72 50.668 26.622 6.655 3.407 4.875

84 50.384 25.713 9.131 2.825 5.401
84 50.499 25.678 9.340 2.858 0.559 5.297
84 50.753 25.520 9.393 2.737 0.444 5.275