University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange
Doctoral Dissertations Graduate School
12-2002
The synthesis and reactions of substituted perfluoroadamantanes, fluorofullerenes and norbornane carbocations
Kenneth Smith University of Tennessee
Follow this and additional works at: https://trace.tennessee.edu/utk_graddiss
Recommended Citation Smith, Kenneth, "The synthesis and reactions of substituted perfluoroadamantanes, fluorofullerenes and norbornane carbocations. " PhD diss., University of Tennessee, 2002. https://trace.tennessee.edu/utk_graddiss/6311
This Dissertation is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council:
I am submitting herewith a dissertation written by Kenneth Smith entitled "The synthesis and reactions of substituted perfluoroadamantanes, fluorofullerenes and norbornane carbocations." I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the equirr ements for the degree of Doctor of Philosophy, with a major in Chemistry.
Jamie L. Adcock, Major Professor
We have read this dissertation and recommend its acceptance:
Bob Compton, Lee Magrid, David Brian
Accepted for the Council:
Carolyn R. Hodges
Vice Provost and Dean of the Graduate School
(Original signatures are on file with official studentecor r ds.) To the Graduate Council:
I am submitting herewith a dissertation written by Kenneth M. Smith entitled "The Synthesis and Reactions of Substituted Pertluoroadamantanes, Fluorofullerenes and Norbomane Carbocations." I have examined the final paper copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Chemistry.
l·
We have read this dissertation and recommend its acceptance:
Vice Provost and Dea Graduate Studies
The Synthesis and Reactions of Substituted Perfluoroadamantanes, Fluorofullerenes and Norbornane Carbocations
A Dissertation
Presented for the
Doctor of Philosophy
Degree
The University of Tennessee, Knoxville
Kenneth Smith
December, 2002 Dedication
To Mom and Dad
Without your support and encouragement this would not have been possible
11 Acknowledgments
I would like to thank my research advisor, Dr. Jamie Adcock for her advice and help with this work. I would also like to thank Dr. Richard Pagni for his direction and guidance with the norbomane project.
The author also thanks Rose Boll and Jim Green for their help and encouragement.
You two are the role models that have made me a good teacher.
The author also thanks Dr. Albert Tuinman and Dr. Pan for their help and guidance with the mass spectroscopy and NMR.
The author also thanks the other members of his dissertation committee (Dr. Bob
Compton, Dr. Lee Magid, and Dr. David Brian) for their aid and cooperation.
The author also thanks those who have made life enjoyable the last 8 years. The original crew, Ben, John, Paul Weaver, and especially Chris and Heather, I miss you all.
Paul and Darin, I'm sure I will miss you both soon.
Finally I would like to thank two of my favorite people in the entire world, Becca and Amanda. Becca, your unwaivering love and friendship have shaped much of what I am today. And Amanda, there is nothing I will miss more than your smile. Your company is something I hope I can always look forward too.
ill Abstract
Fluoro and trifluoromethyl groups are becoming increasingly popular in pharmaceuticals, as their inductive effects can be used to tailor a drug's specific interactions.
Fluorous syntheses are also now becoming commonplace due to the efficient partitioning possible between fluorous and organic phases. Several new F-adamantyl derivatives were synthesized from F-Adamantyl lithium as targets for future incorporation into pharmaceuticals or for use as fluorous tags. They are: Ethyl-F-Adamantane from ethyl iodide, methyl F adamantyl ketone and enolate from acetyl chloride, and TIPS protected F-Adamantyl ethanol from TIPS protected 2-bromoethanol. The synthesis of F-adamantyl sulfonyl halides was attempted, however, the F-adamantyl anion attacks the chlorine of sulfonyl chloride fluoride rather than the sulfur in an SNX reaction. Also, F-adamantyl sulfonyl bromide formed by the reaction ofF-Adamantyl Lithium with sulfur dioxide and bromine desulfonylated to form F adamantyl bromide.
No fluorofullerene other than CJ48 has been produced as a pure compound. Also, no derivatized fluorofullerenes have been produced. Our goal here was to either produce a new fluorofullerene as a pure compound, or produce significant amounts of a functionalized fluorofullerene. Fullerene was reacted with fluorine at various temperatures ( 150 °, 100 °, and
50°C) and the product distribution was observed using negative ion chemical ionization mass
spectrometry. These distributions ranged from CJ48 to Cwf3 6 with lower temperatures favoring lower distributions. No products with less than 36 fluorines were observed, although
unreacted fullerene was recovered at 50°C. The cyanofullerenes C6JICN and C60(CN)2 were also fluorinated at various temperatures. (250°, 150°, 100°, and 50°C) Although substituted
IV cyanofullerenes such as CJ47 CN were observed, it was only in very small quantities. The majority of the products were similar to fluorinating fullerene at the same temperatures. A
unique exception to this was the presence of the series CJ18 to C60F24 as well, in these two systems.
The reactions producing 1-chloronorbomane are widely used, but the mechanistic aspects have not been thoroughly explored. We have observed an ion, which is an intermediate in these reactions and have characterized it by high resolution 2D NMR.
Although the 2-norbomyl carbocation has been widely studied in the past, we have reexplored this ion using high resolution 2D NMR techniques which were not possible when this work was first completed. COSY and HMQC agreed with previous assignments. An EXSY kinetics study of the system yielded rate constants for the proton exchange in this system at
193K, 208K, and 223K. The 2-chloro-2-norbomyl carbocation was observed by NMR for the first time. This new cation was characterized using NOESY, COSY, HMQC, as well as
1D spectra.
V Table of Contents
1. INTRODUCTION ...... 1
2. EXPERIMENTAL PROCEDURE ...... 8
A. The Aerosol Fluorinator...... 8
B. Vacuum Line ...... 10
C. Instrumentation ...... 10
1. Gas Chromatography ...... 10
2. Nuclear Magnetic Resonance ...... 10
3. Mass Spectra ...... 10
D. Commercially Available Materials ...... 11
E. Preparation of Starting Materials ...... 13
1. Synthesis of 1-Adamantane Carboxylic Acid Methyl Ester...... 13
2. Synthesis of TIPS protected 2-Bromoethanol...... 13
3. Synthesis ofCJICN, C60(CN)z, and C(i()(CI-Ig)CN ...... 14
4. Synthesis of2,2-dichloronorbomane ...... l 5
F. F-Adamantanes ...... 16
1. Aerosol Direct Fluorination ...... 16
a. Aerosol Fluorination of 1-Chloroadamantane...... 16
b. Aerosol Fluorination of 1-Adamantane Carboxylic Acid Methyl
Ester ...... 16
2. Formation of 1-Hydryl-F-Adamantane...... 18
a. Reaction ofF-1-Chloroadamantane with LiAIH4 ..••....••.••...... •...... 18 b. Hydrolysis of F-1-Adamantane Carboxylic Acid Methyl Ester ...... 19
c. Methanolysis ofF-1-Adamantane Carboxylic Acid Methyl Ester
and subsequent reduction ...... 19
3. Synthesis of F-Adamantane derivatives ...... 20
a. Reaction ofF-Lithioadamantane and TIPS protected
2-Bromoethanol...... 20
b. Reaction ofF-Lithioadamantane and acetyl chloride ...... 21
c. Reaction ofF-Lithioadamantane and sulfuryl chloride fluoride ...... 21
d. Reaction ofF-Lithioadamantane and sulfur dioxide ...... 22
e. Reaction ofF-Lithioadamantane and ethyl bromide ...... 23
G. Fluorofullerenes ...... 23
1. Synthesis ofCJ48 ...... 23
2. Reaction ofFullerene with fluorine at various temperatures ...... 25
3. Reaction of C6JfCN, C60(CN)z, and C60(CH3)CN with fluorine at various
temperatures ...... 25
4. Reaction of CJ48 with reducing agents ...... 26
a. Heat...... 26
b. Thiophenol, borane/dimethyl thioether, Mg/Fe/Zn metals, Fe(CO)526
c. Hydrogen ...... 26
H. Norbomyl Cations ...... 27
1. Formation ofCation...... 27
3. RESULTS AND DISCUSSION ...... 28
Vll A Methods for the formation of 1-Hydryl-F-Adamantane ...... 28
B. Synthesis of new F-Adamantane derivatives ...... 32
C. The electronegativity of the F-Adamantyl group by NMR...... 34
D. The large scale synthesis of CJ48 and fluorination of fullerene at lower
temperatures ...... 38
E. The fluorination of cyanofullerenes ...... 40
F. The reduction of CJ48 ...... 43
G. Low temperature NMR studies ofnorbomyl carbocations ...... 45
1. Low temperature COSY and HMQC of the non-classical 2-norbomyl
carbocation ...... 4 5
2. Low temperature kinetic study of the 2-norbornyl carbocation using
EXSY ...... 46
3. Characterization of a new ion, the 2-chloro-2-norbornyl carbocation .. .4 7
4. CONCLUSIONS ...... 53
REFERENCES ...... 56
APPENDICES ...... 59
APPENDIX A - Substituted F-Adamantane Spectra...... 60
APPENDIX B - Fluorofullerene Mass Spectra ...... 66
APPENDIX C - Norbomyl Cation Spectra...... 86
VITA ...... 99
Vlll List of Tables
1. Aerosol Fluorination of 1-Chloroadamantane ...... 17
2. Aerosol Fluorination of 1-Adamantane Carboxylic Acid Methyl Ester...... 18
3. Electronegativity vs. Chemical Shift data for Ethyl halides ...... 36
4. Electronegativity of the Trifluoromethyl group ...... 38
5. The Fluorination of C60 at various temperatures ...... 40
6. The Fluorination of C6JICN at various temperatures ...... 42
7. The Fluorination of C60(CN)2 at various temperatures ...... 43
8. Summary ofHM:QC results ...... 51
9. Summary of COSY and NOESY results ...... 51
lX List of Schemes
1. Formation of Cyanofullerenes ...... 14
2. Methods for the formation ofF-adamantyl hydride ...... 28
3. Reactions of Perfluorinated Adamantyl Esters ...... 31
4. SNX Attack of Sulfuryl Chloride Fluoride ...... 34
5. Reactions ofF-Adamantyl Lithium ...... 35
6. Reactions of Chloronorbomanes ...... 49
X List of Figures
1. Fluorinated Pharmaceuticals ...... 2
2. Fluorous Separation ...... 3
3. The Aerosol Fluorinator...... 9
4. System for Fluorinating Fullerenes ...... 24
5. Carbon spectrum ofTIPSO protected F-adamantylethanol...... 33
6. Graphs of Chemical Shift vs. Electronegativity ...... 37
7. Low temp NMR of the Norbomyl cation ...... 45
8. EXSY Kinetics Data ...... 48
9. The 2-Chloro-2-Norbomyl Carbocation ...... 50
A-1. 1-Chloro-F-Adamantane (19F) ...... 61
A-2. 1-Hydryl-F-Adamantane (19F and 1H) ...... 62
A-3. F-Adamantyl Methyl Ketone and Enolate (1H) ...... 63
A-4. TIPSO protected F-Adamantylethanol (1H) ...... 64
A-5. TIPSO protected F-Adamantylethanol (13C) ...... 65
B-1. Pure C~48 ...... •...... •...... 67
B-2. C60 + F2 at 150°C ...... 68
B-3. C60 + F2 at 100°C ...... 69
B-4. C60 + F2 at 50°C ...... 70
B-5. C~CN + F2 at 250°C ...... 71
B-6. C~CN + F2 at 150°C ...... 72
B-7. C~CN + F2 at l00°C...... 73
XI B-8. CwlfCN + F2 at S0°C ...... 74
B-9. CwlfCN + F2 at 50°C (blow up) ...... 75
B-10. C60(CN)2 + F2 at 250°C ...... 76
B-11. C60(CN)2 + F2 at 150°C ...... 77
B-12. C60(CN)2 + F2 at I00°C ...... 78
B-13. C60(CN)2 + F2 at S0°C ...... 79
B-14. C60(CH3)CN + F2 at 250°C ...... 80
B-15. CJ48 with N2 and H2 at 150°C ...... 8l
B-16. C60F48 with Mg, Fe, and Zn metals ...... 82
B-17. CJ48 + Thiophenol...... 83
B-18. C6of48 + Borane ...... 84
B-19. CJ48 + Fe(CO)5 ...... 85
C-1. COSY of the 2-Norbomyl Cation at -80°C...... 87
C-2. HMQC of the 2-Norbomyl Cation at -80°C ...... 88
C-3. 2D EXSY of the 2-Norbomyl Cation at 193K...... 89
C-4. 2D EXSY of the 2-Norbomyl Cation at 208K...... 90
C-5. 2D EXSY of the 2-Norbomyl Cation at 223K...... 91
C-6. 1H of the 2-Chloro-2-Cation at -90°C ...... 92
C-7. 13C of the 2-Chloro-2-Cation at -90°C ...... 93
C-8. HMQC of the 2-Chloro-2-Cation at -90°C ...... 94
C-9. COSY of the 2-Chloro-2-Cation at -90°C...... 95
C-10. NOESY of the 2-Chloro-2-Cation at -90°C ...... 96
XII C-11. 1H of the 2-Chloro-2-Cation warmed to -20°C...... 97
C-12. 13C ofthe 2-Chloro-2-Cation warmed to -20°C ...... 98
Xlll
Chapter 1 - Introduction
Since the isolation of elemental fluorine by Moisson in the l 890's, the goal of using the most reactive of the elements had gradually grown into a substantial field, with arms into nearly every area of chemistry. Fluorine chemistry, however, had a rather slow start - as it was not until the Manhattan project that the element was first widely explored. Due to the necessity of using fluorine to separate uranium isotopes, large amounts of valuable information were gathered about the element, and more importantly all aspects of handling the dangerous element were tackled.
Today, fluorinated compounds are widely used due to their inertness to chemical attack, and often non-toxicity. Of course, Dupont's Teflon is the most widely used fluorochemical, but both 1st generation and today's 'ozone safe' refrigerants are fluorocarbons, and the list goes on.
It is not well appreciated, however, how often fluorine appears in some of today's most prescribed pharmaceuticals. The fluoro and trifluoromethyl groups have unique properties based on their high electronegativity and electron withdrawing effects on neighboring groups. Today, when derivatives of a new drug are being explored, fluoro and trifluoromethyl groups are often included. This is because their unique electronegativities can tailor the required interactions, as well as the problem that the use of chloro groups is now frowned upon in pharmaceuticals. Finally, fluorinated groups aid in transport of a pharmaceutical through the blood/brain barrier if this is required. Trifluoromethyl groups are found in diuretics (hydroflumethiazide ), anti-psychotics (penfluridal and fluphenazine ), neuroleptics ( oxaflumazine), and even experimental analgesics ( antraferic). Fluoro groups
1 are more widely used, being popular in inhaled anesthetics, beta-lactamase resistant penicillins, and most importantly in several widely prescribed steroids, which have many uses from dermatosis treatments ( clobetasone), to bronchial anti-inflamatory agents for asthma
(betamethasone), which alleviate the symptoms of millions of asthma sufferers. See Figure
1 for structures of several of these compounds.
Another rather new development in organo-fluorine chemistry exploits the often immiscibility of fluorocarbons in other common organic solvents. This leads to what are now called fluorous separations. Large perfluorinated groups are tethered to synthons, to allow
Fluphi:nazinc Pcnlluridol
{'11,0ll I - c=c, Oil I,, ( IL
16 alpha: Dcxamclhasonc 16 beta: Bctamcthasonc
Figure 1 - Fluorinated Pharmaceuticals
2 the product to be partitioned in the fluorous phase, while unreacted material remains in the organic phase. The opposite is also possible, where a reactant and an unwanted byproduct cany the fluorous group, removing them from the desired product in the organic layer. These methods often remove the need for chromatography between reactions, and could feasibly be used in combinatorial synthesis. Fluorous phase syntheses in this manner, could revolutionize the synthesis of derivatives in pharmaceutical chemistry. See Figure 2. 1
These perfluorinated groups may be attached to the target reagent/precursor in several ways. Most recently, a perfluorinated group has been attached to a silicon, which is then used to protect an alcohol. Alcohols may also be protected by forming fluorine containing ethers2,
R-CN ( ... · lmr:urities"J 9
1) (C 6F uCH;.CH;:)::,SnN3 (8) BTF. 80 C 2) Benzene/ FC-72
Organic , F!uorous
R-·CN' N----N"Sn(CH2CH 2C6F 13 ):; I I 9 R.....-"-.. N', N (-.. "lrnpur,tics') 10
3) HCliether
: FluorOcJS Organic i ' t
12 Figure 2 - Fluorous Separation
3 or a reagent such as tributyl tin hydride may be substituted with fluorinated groups, allowing easy removal of the unreacted hydride, as well as the reacted tin containing byproduct. 3
Over the years, our lab has perfected the technique of aerosol direct fluorination
(AF).4'5 There are two problems with directly reacting fluorine and an organic molecule. The first is the high exothermicity of the attachment of the first fluorines. This heat often causes, among other things, fragmentation of carbon skeletons. Efficient removal of this heat, however - as with low temperature fluorination techniques, often causes much of the organic substrate to fail to be perfluorinated. This is due to the second factor, that with increasing fluorination, the last few hydrogens are very difficult to replace. On the other hand, not removing the excess heat would cause run-away exothermic reactions.
The aerosol direct fluorination technique stabilizes both of these effects. AF is a flow technique, meaning that the organic is introduced in the reactor in a stream of inert gas.
Sodium fluoride is heated to a temperature high enough for it to melt and then vaporize into a flow of an inert gas. This flow is rapidly cooled to liquid nitrogen temperature which creates an aerosol of sodium fluoride. This aerosol is then combined with the fluorine and organic in a reactor tube, in which the amounts of fluorine can be controlled in four separate phases, as can the temperatures. Fluorination occurs on the surface of the aerosol particles, which act as a heat sink to dissipate the heat, most of which is lost to the inert gas. Cooling the outside of the reactor tube to zero to -10 ° C also helps ensure that all heat produced in the system is efficiently removed ..
These four phases are carefully controlled. The first phase has low temperature and low fluorine concentration. This minimized the amount of fragmentation and crosslinking as
4 only the most reactive hydrogens are replaced with fluorines. The second and third stages have higher temperatures and fluorine concentrations helping to increase the efficiency of collisions with fluorine radicals. In the fourth stage, UV irradiation is applied. This increases the concentration and energy of the fluorine radicals to insure that the last, most unreactive hydrogens in the now mostly fluorinated molecule are efficiently replaced with fluorine.
This method, then, controls the reaction exothermicity, fragmentation, and crosslinking. It results in high yields of nearly I 00% of perfluorinated product, and makes efficient use of the rather expensive fluorine.
This lab has used the aerosol fluorination method to perfluorinate a wide variety of organic functionalities. These include alkanes, which are effectivly perfluorinated as long as no ring strain exists6, ethers7' 8, esters8, ketones9' 10, and alkyl halides. Alkyl chlorides give good yields of perfluorinated products11' 12, occasionally with rearrangement. Alkyl bromides can be fluorinated, but lose the bromine as it is attacked during the reaction and is likely lost
as BrF3. This results in perfluorinated product without functionality.
The aerosol fluorinator has shown itself to be especially useful in fluorinating diamondoid alkanes13' 14. Although adamantane and its deriviatives are often solids, they have high volatility, and no ring strain, which allows us to fluorinate them in high yield and purity.
More importantly, their derivatives may also be fluorinated. Good results have been shown with l-chloroadamantane15, 2-chloroadamantane, 1,3-dichloroadamantane, 1,3,5- trichloroadamantane, 1,3 ,5, 7-tetrachloroadamantane16, 2-adamantanone17, and 1-adamantane carboxylic acid methyl ester18 .
This lab has also shown that F-chloroadamantanes can be reduced to the
5 corresponding hydryl adamantanes by a variety of methods 15• F-2-adamantanone can be reduced to 2-hydryl-F-adamantan-1-0117, and although the perfluorinated 1-adamantane carboxylic acid methyl ester is unstable to hydrolysis, the trifluoromethyl group may easily be replaced by a methyl in a transesterification reaction, resulting in that part of the ester not being fluorinated which greatly increases its resistance to hydrolysis18 . These esters can also be reduced to hydryl compounds with lithium aluminum hydride ..
These hydryl F-adamantanes are the primary focus of this work. They very readily react with methyl lithium to form methane and a very stable, highly ionic F lithioadamantane19. The lithio compound leads to a wide variety of reactions, which may be useful in introducing the F-adamantyl group into organic molecules, potentially yielding pharmaceuticals with unique properties, or by using this group as a tag for fluorous phase separations. The ways we have explored doing this include using the F-lithioadamantane to react with alkyl halides, alkyl halides containing a protected functionality, and acyl halides.
Our lab is also the first in the world to produce significant amounts of fluorofullerene compounds. Fullerene is fluorinated in a sodium fluoride matrix at high temperature, which
20 produces high purity C6of48 . Another focus of this work is to explore the possibility of derivitizing CJ48 to give it greater synthetic usability, as well as reduce it using a variety of reagents to possibly form other pure fluorofullerenes. No functionalized fluorofullerenes have yet been produced. These compounds could show interesting oxidizing and fluorinating properties on their own. There is also the possibility of further modifying a functionalized fluorofullene, again with the goal of attaching these to other molecules.
The final focus will be to revisit the low temperature formation of the norbomyl
6 carbocation in superacid media. The goal of this work is to use new high resolution NMR techniques to probe the structure and kinetics of this unique cation, as well as explore the reactivity of chlorinated norbornanes in super-acid media. Also, I will report the first NMR characterization of a new ion formed from 2,2-dichloronorbornane and antimony pentafluoride. This ion is the 2-chloro-2-norbornyl carbocation. It is extremely stable at -
80°C, and was characterized at -100°C. Upon warming this ion isomerizes to a different ion, which we believe is the l-chloro-2-norbornyl carbocation. This work is of great interest to us because it potentially gives new mechanistic information on the well known reaction which converts 2,2-dichloronorbornane to 1-chloronorbornane in the presence of a strong Lewis acid.
7 Chapter 2 - Experimental Procedure
A. The Aerosol Fluorinator
The principle of the aerosol direct fluorinator was discussed earlier and it has been described in detail elsewhere5• To summarize, nucleating aerosol particles of sodium fluoride are generated in a furnace which is heated to above 1050° C. These particles are then carried in a flow of helium through a trap at -196 ° C and then into the main reactor section. The hydrocarbon (in a solvent which is co-fluorinated) is introduced into the reactor by liquid solution injection. Both the hydrocarbon and the solvent are evaporated and carried by a flow of helium into the reactor as well. Finally fluorine, also diluted with helium in introduced into the reactor through four modules establishing a fluorine gradient along the reactor tube, which is comprised of a porous monel tube. The aerosol, hydrocarbon, and fluorine are carried through the reactor modules, the last of which is irradiated by an ultraviolet lamp and finally into a stainless steel product trap which is cooled by liquid nitrogen. The trap also contains sodium fluoride pellets to absorb much of the HF which is produced. After the fluorination is complete, the product trap is disconnected and connected to a vacuum line for further separation. See Figure 3 for a schematic of the fluorinator.
The ratio of hydrocarbon to co-fluorinating solvent is typically 1:3.5(g:g). The solution is injected via a syringe pump at the rate of 1.5 ml/hr. Trap to trap fractionation follows, typically including traps at -22°, -76°, and -196°C. The fluorinated adamantanes collect in the -22 ° trap, fluorinated solvent in the -76 ° trap, and other byproducts in the -196 ° trap.
8 HYDROCARBON EVAPORATOR
SKoodary ---- Hydre>c•rbon
REACTOR MODULE 1
Porous Monc,I lnaert 2 moor,
Coolant l~Hlt on Remove~ S.ctlcn Fluorine 1,-;IIJt 1-}
Fluo,1,- Inlet 2-1
REACTOR Coo.,,nt Ou11et MODULE 2 Porous Mone! risen ;., mlcr::,n
PHOTOCHEMICAL STAGE Coor.ng Water Outlet
T•!lon FEP Medium Pressure Mercury Ale Un.d Oua,rtz Reflector Not Shown Ructor Tub41
Ouertz l~imer&Jcn Well
Coonn,1 Waler Inlet
AEROSOL PRODUCT TRAP
Product Trap Valvee --- TO ALUM~'lA TRAPS I -r
Sodium nuorlde Pellet• -4 o Mier on n11er Ditk
Liquid Nitrogen - Dewer Fleak
Figure 3 - The Aerosol Fluorinator
9 B. Vacuum line
An all pyrex vacuum line was used for transfer and fractionation of perfluorinated compounds. All stopcocks were Kontes Hi Vacuum Teflon and a Welch Duo-seal (1402B) pump supplied the vacuum.
C. Instrumentation
1. Gas Chromatography
The fluorinated products were separated and purified by gas chromatography using a Bendix 2300 prep scale instrument. The column used was a 7 meter by 3/8 inch column filled with 13% fluorosilicone QF-1 stationary phase on 60-80 mesh, acid washed chromosorb
P conditioned at 225°C for 12 hours. The carrier gas was helium.
2. Nuclear Magnetic Resonance
All 19F, 1H and 13C spectra are obtained on a Varian Mercury 300MHz 4 nucleus spectrometer. Typically deuterochloroform containing 1% tetramethylsilane was used as the
1 solvent and roughly 100 ul ofCFC13 was added as the standard for 9F. Low temperature and
2D spectra were obtained on a Varian INOV A 600MHz or on a Bruker AVANCE 400MHz spectrometer.
3. Mass spectra
Mass spectra were recorded on a VG ZAB-EQ mass spectrometer operating in negative chemical ionization mode. D. Commercially Available Materials
All materials used were reagent grade and were used as received unless noted otherwise.
Acetyl Chloride: Aldrich Chemical (98%)
1-Adamantane Carboxylic Acid: Eastman
Aluminum Chloride, anhydrous: Aldrich Chemical (99%)
Antimony Pentafluoride: Cationics Inc. (technical)
Bromine: Acros
2-Bromoethanol: Acros (97%)
1-Chloroadamantane: Aldrich Chemical (98%) or Acros (98%)
Chloroform-d: Aldrich Chemical (99.8%D, 1% v/v TMS)
exo-2-Chloronorbornane: Aldrich Chemical (98%)
Cyclohexane: Acros (99+%)
1,2-Dichlorobenzene: Acros (99%)
N,N-Dimethylformamide: Acros (99%)
Ethanol Absolute: Aaper
Ethyl Acetate: Fisher
Ethyl Ether (anhydrous): Fisher
Ethyl Iodide: Fisher
Fluorine: Air Products, 97% technical grade
Fullerene: Southern Chemical group (99.5%)
Hydrochloric Acid (3 7% ): Fisher
11 Iron pentacarbonyl: Aldrich
Lithium Aluminum Hydride (powder): Aldrich Chemical (95+%)
Methanol: Fisher
Methyl Lithium: Aldrich Chemical (1.4M in ether) or Acros (1.6M in ether)
Methyl Trifluoromethansulfonate: Aldrich (99+%)
N orbomane-2-carboxylic Acid: Alfa (98%) (bicyclo[2.2. l ]heptane-2-carboxylic acid)
Norcamphor
Pentane
Phosphorus trichloride: Aldrich (98%)
Phosphorus pentachloride: Aldrich (95%)
Silicagel: Acros (0.060-0.200mm) 6nm pore diameter
Sodium Bifluoride (pellets): Harshaw Catalysts
Sodium Cyanide: Acros
Sodium Fluoride: Aldrich Chemical (99.8%)
Sodium Hydroxide: Mallinckrodt Chemical
Sodium Sulfate (anhydrous): Fisher
Sulfuric Acid: Fisher
Sulfuryl Chloride Fluoride: Lancaster (99%)
Sulfuryl Fluoride: Lancaster (99%)
Toluene: Fisher
p-Toluenesulfonyl cyanide: Aldrich (95% tech)
12 Tri-n-butyl tin hydride: Aldrich (97%)
1, 1,2-Trichloroethane: Fluka (>98%)
Trichlorofluoromethane (R-11 ): Pennwalt Corporation
Trifluoroacetic Acid: Fisher
Triisopropylsilyl chloride: Aldrich (98%)
E. Preparation of Starting Materials
1. Synthesis of 1-Adamantane carboxylic acid methyl ester
1-Adamantane carboxylic acid methyl ester was made by a simple acid catalyzed esterification reaction due to its high stability to acid hydrolysis. 1-adamantane carboxylic
acid (6.0 g, .033 mol) was dissolved in 50 ml methanol. 18M H2SO4 was then added slowly and the mixture was refluxed for 4 hours. After cooling, the methanol solution was poured into water and the water was extracted with ether. The ether layers were dried and the ether was evaporated to yield an oil which solidified when placed in the freezer. It showed the expected five peaks in the carbon spectrum: 184.62, 40.44, 38.49, 36.362, and 27.77.
2. Synthesis of TIPS protected 2-Bromoethanol
2-Bromoethanol was protected with a triisopropylsilyl group (TIPS) usmg a generalized literature procedure21 for TIPS protecting alcohols. 2-bromoethanol (10 g, 0.080 mol) was added to 20 ml ofDMF in which imidazole (13.6 g, 0.200 mol) had been dissolved.
TIPS chloride (18.5 g 0.96 mol) was then slowly added. The mixture was stirred overnight at room temperature. It was then poured into water and extracted with methylene chloride
13 which is then evaporated to yield a clear liquid (7.36 g, 65.4%) 1H peaks at l .04(s), 1.05(s),
3.37(t), 3.94(t). 13C peaks at 12.0, 17. 7, 33.0, 63.8.
Synthesis of these fullerene derivatives followed the literature procedures22. To a
solution of C60 in o-dichlorobenzene (250 mg in 20 ml) was added a solution ofNaCN (3.5 ml of 0. lM solution in DMF) dropwise over a 5 minute period. The color immediately changed from purple to blue and gradually to very deep green. After stirring for 15 min,
various electrophiles are added to the C60(CNr anion which has been formed.
To prepare C~CN, 0.25 ml of tritluoroacetic acid were added to the above solution.
After 30 minutes, the mixture was loaded onto a silica column. Unreacted C60 eluted with cyclohexane/toluene 8:2, and then the product with cyclohexane/toluene 1: 1. The solvents were removed under reduced pressure and the product was washed with cyclohexane,
Scheme 1 - Formation of cyanofullerenes
14 methanol and then with ether.
To prepare C60(CH3)CN, the mixture was treated with methyl trifluoromethanesulfonate (0.2 ml). After stirring overnight, the brown mixture was
precipitated with ether and the solid was isolated by decantation. It was then dissolved in CS2
and loaded on a silica column. Unreacted C60 was eluted with cyclohexane/CS2 9: 1, and the product with cyclohexane/toluene 7:3. Solvents were removed under reduced pressure and the residue was first washed with methanol and then with ether.
To prepare C6(J(CN)z, the mixture was treated with 130 mg of p-Toluenesulfonyl cyanide in o-dichlorobenzene (5 ml). After 2 hours the brown mixture was loaded onto a
silica column. The unreacted C60 was eluted with cyclohexane/toluene 8:2, and the desired product with cyclohexane/toluene 2:8. The solvents were removed under pressure and the residue was first washed with methanol and then with ether.
4. Synthesis of2,2-dichloronorbomane
2,2-Dichloronorbomane was prepared by literature procedures23 . Phosphorus
trichloride (10 ml) was charged into a 250-ml three-neck flask equipped with a CaC12 drying tube. The flask was cooled to -5 ° C in an ice-salt bath and norcamphor ( 15 g) was added in portions. After the norcamphor had dissolved, phosphorus pentachloride (31. 7 g) was added
over 20 minutes with efficient stirring and cooling. The mixture was allowed to warm to
room temperature and stand over night with stirring. It was then poured onto 200 g of ice
in a 500-ml flask and the hydrolysis was allowed to proceed while the flask was kept in an ice
salt bath. After 30 minutes, the mixture was allowed to warm to room temperature and was
15 extracted with pentane ( 4x80 ml). The combined pentane extracts were washed with water
(2x 100 ml) and dried over anyhydrous Naz SO 4. The pentane was evaporated to give 19 g of a yellowish oil. Distillation of this yellowish oil under vacuum gave 14. 5 g of a colorless liquid which solidified to a low-melting solid. The 13C spectrum of the 2,2- dichloronorbomane showed the expected seven peaks at 94.10, 55.97, 55.37, 37.69, 27.58, and 25.38.
F. F-Adamantanes
1. Aerosol Direct Fluorination
a. Aerosol Fluorination of 1-Chloroadamantane
Details of the aerosol fluorination parameters for 1-chloroadamantane are listed in
Table 1. 1-Chloroadamantane (2.50 g) was fluorinated as a solution in 1, 1,2-trichloroethane
(5.50 g) over a period of 4.5 hours. (Total volume of solution-? ml, injected at 1.5 ml/hour.)
The products were transferred to the vacuum line and fractionated. The F-1- chloroadamantane was trapped at -22°C. After warming to room temperature, the product is dissolved in a 4/1 CFCI/chloroform mixture (the chloroform raised the boiling point of the
solution to make injecting the solution easier) and purified by prep scale GC, as described above. The 19F NMR spectrum showed three peaks at -114, -120, and -219ppm integrating to 6:6:3.
b. Aerosol Fluorination of 1-Adamantane Carboxylic Acid Methyl Ester
16 Details of the aerosol fluorination parameters for 1-Adamantane Carboxylic Acid
Methyl Ester are listed in Table 2. 1-Adamantane Carboxylic Acid Methyl Ester (2.50 g) was fluorinated as a solution in 1, 1,2-trichloroethane (5.50 g) over a period of 4.5 hours. (Total volume of solution ~ 7 ml, injected at 1. 5 ml/hour.) The products were transferred to the vacuum line and fractionated. The F-1-Adamantane Carboxylic Acid Methyl Ester was trapped at -22°C. The product was then immediately hydrolyzed (see below).
Table 1
Aerosol Fluorination of 1-Chloroadamantane
Parameters FLUORINE FLOWS (ml/min)
Module 1-1 8 Module 1-2 70 Module 1-3 63 Module 1-4 6 REACTION TEMPERATURES (°C)
Module 1 -10 Module 2 0 Evaporator 110 Injector 115
AMOUNT OF ST ARTING MATERIAL 2.6g AMOUNT OF SOL VENT 7.3g STARTING MATERIAL/SOLVENT RATIO 1/2.8 PRODUCT YIELD (grams) 3.70g PRODUCT YIELD (GC pure, grams) 1.70g FINAL PRODUCT YIELD (GC pure,%) 25.3
17 Table 2
Aerosol Fluorination of 1-Adamantane Carboxylic Acid Methyl Ester
Parameters
FLUORINE FLOWS (ml/min)
Module 1-1 8 Module 1-2 70 Module 1-3 63 Module 1-4 6
REACTION TEMPERATURES (°C)
Module 1 -10 Module 2 0 Evaporator 120 Injector 115
AMOUNT OF ST AR TING MATERIAL 2.2g AMOUNT OF SOL VENT 6.0g STARTING MATERIAL/SOLVENT RATIO 1/2.7
2. Formation of 1-Hydryl-F-adamantane
a. Reaction of F-1-Chloroadamantane with LiAIH4
1.33 g oflithium aluminum hydride were transferred to a flask in a dry box and 50 ml of ethyl ether were added. This solution was cooled in an ice bath and stirred for 10 minutes.
2.5 g of F-Adamantyl Chloride were dissolved in 10 ml of ethyl ether. This solution was added dropwise to the LAH solution with stirring. The solution was then allowed to warm to room temperature over a period of 2-3 hours and then was stirred at room temperature overnight. The solution is again cooled to near zero degrees. 16 ml of cold absolute ethanol are slowly added, ensuring that the ether solution never boiled. Finally 120 g of 50% sulfuric
18 acid was added, slowly at first. Enough ether was then added and the ether layer separated.
The aqueous layer was extracted with ether a total of three times, the ether was then washed once with water and dried over sodium sulfate. The ether was then evaporated to yield 0.570 g F-adamantyl hydride. (25.0% yield) 19F peaks at -108.7(s), -120.9(m), -219.3(m). 1H peak at 3. 90. An alternate workup was to pump on the mixture after adding the ethanol. This eliminates the need for the acid hydrolysis and liquid work-ups. The vessel is pumped on at
RT and all volatiles are collected at -196°C. The volatile products are then fractionated with pure F-adamantyl hydride remaining in the -22°C trap. (Does not improve yield)
b. Hydrolysis ofF-1-Adamantane Carboxylic Acid Methyl Ester
After being fractionated from the reactor trap, the F-ester was held at -22°C. A solution of3 g ofNaOH in 10 ml of water was then added, and the solution was allowed to warm to room temperature for an hour. To ensure complete decarboxylation, the solution was then acidified with 3M sulfuric acid, and again allowed to stand for an hour. Finally the solution was extracted with ethyl ether 3 times, washed with water, dried with sodium sulfate, and evaporated to yield 510 mg of F-adamantyl hydride. (Yield from unfluorinated starting material - 11. 0%)
c. Methanolysis ofF-1-Adamantane Carboxylic Acid Methyl Ester and subsequent reduction
As above, after being fractionated from the reactor trap, the F-ester was held at -
22 ° C. A large excess of methanol in ethyl ether was added ( 10 ml of each) and the solution
19 was wanned to room temperature. The organic was washed twice with water, and dried with sodium sulfate. The ether was then carefully evaporated to yield 630 mg of a colorless liquid, shown to be NMR. pure, methyl F-1-adamantanoate. (Yield from unfluorinated starting material - 12.0%) 19F peaks at -108.2, -120,1, and -219.3
The ester was then reduced with lithium aluminum hyride as above. 600 mg of ester were reacted with 400 mg of LAH in a total of 20 ml of ethyl ether. After hydrolysis, extraction, and evaporating the solvent, 200 mg ofF-adamantyl hydride were produced (38% yield).
3. Synthesis ofF-adamantane derivatives
a. Reaction ofF-lithioadamantane and TIPS protected 2-Bromoethanol
1-Hydryl-F-adamantane (0.090 g) dissolved in ethyl ether (1.0 ml) was placed in an ampule (-5 ml) capped with a septum. A diethyl ether solution of methyl lithium (1.6M, 0.2 ml) was injected dropwise through the septum at -80°C. The mixture was allowed to sit for
2 hours. After 2 hours, an excess of TIPS protected 2-bromoethanol (0.080 ml) in 0.100 ml of ethyl ether was added to the above mixture, still at -80 °. The solution was then allowed to wann to room temperature for 2 hours. The contents of the tube are pumped off any solid formed (LiBr) to a trap held at-196°C. The trap is then warmed to -22°C, allowed the ether and unreacted protected alcohol to be pumped off 110 mg of a colorless solid results, shown by NMR to be TIPS protected F-adamantyl ethanol (82.1% yield). 19F peaks at -l 13.2(s), -
121.5(s), -219.3(s). 1H peaks at l.05(s), l.07(s), 2.5l(t), 4.03(t). 13C peaks at 12.0, 17.8,
23.6, 54.3(m), 57.8, 86.3(m), 90. l(m), 106.7(m), 109.7(m), l l 1.3(m), 114. l(m), l 15.8(m),
20 l 16.5(m). The full carbon spectrum with assignments can be found in Appendix A
b. Reaction of F-lithioadamantane and acetyl chloride
1-Hydryl-F-adamantane (0.450 g) dissolved in ethyl ether (2.5 ml) was placed in an ampule (-10 ml) capped with a septum. A diethyl ether solution of methyl lithium ( 1. 6M, 0. 9 ml) was injected dropwise through the septum at -80°C. The mixture was allowed to sit for
2 hours. 0.200 ml of acetyl chloride in 1 ml of ethyl ether was then added. An immediate
reaction is observed with LiCl being formed. The solution was allowed to warm to room temperature for an hour. The contents of the tube are pumped off any solid formed (LiCl) to a trap held at -196°C. The trap is then warmed to -22°C, allowed the ether, unreacted
acetyl chloride, and any acetone by product to be pumped off The resulting solid was
analyzed by NMR to be a mixture of methyl F-adamantyl ketone and its enol. The mixture can
then be converted completely to the keto form by adding a methanol/HCl mixture and
extracting with ether. '9F peaks at: ketone- -l 10.8(s), -121.l(s), -218.9(s), enolate- -1 l 1.5(s),
-121.2(s), -219.3(s). 'H peaks at: ketone- 2.2(s), enolate- 5.6(d), 5.9(d). 13C peaks at:
ketone- 20.7, enol- 118.5, 119.6.
c. Reaction ofF-lithioadamantane and sulfuryl chloride fluoride
1-Hydryl-F-adamantane (0.260 g) dissolved in ethyl ether (2.0 ml) was placed in an
ampule (-5ml) capped with a septum. A diethyl ether solution of methyl lithium (1.6M, 0.6
ml) was injected dropwise through the septum at -80°C. The mixture was allowed to sit for
2 hours. After 2 hours, the solution was frozen with liquid nitrogen and a large excess of
21 sulfuryl chloride fluoride (0.900 g) was condensed onto the above mixture. The mixture was allowed to warm first to -80 ° C, and then over a period of 2 hours to room temperature. A small amount of white precipitate was formed. The ampule was then reattached to the vacuum line and while holding the tube at -22°C, the volatiles were pumped out. The tube was then returned to atmospheric pressure with dry nitrogen and warmed to room
1 temperature. The product was extracted with CFC13 analyzed by 9F NMR. The sample was found to be F-adamantyl chloride.
d. Reaction ofF-lithioadamantane and sulfur dioxide
1-Hydryl-F-adamantane (0.260 g) dissolved in ethyl ether (2.0 ml) was placed in an ampule (-5ml) capped with a septum. A diethyl ether solution of methyl lithium (1.6M, 0.6 ml) was injected dropwise through the septum at -80 ° C. The mixture was allowed to sit for
2 hours. After 2 hours, the solution was frozen with liquid nitrogen and a large excess of sulfur dioxide was condensed onto the above mixture (760 mg). The solution was warmed to -30°C, at which time a precipitate formed and settled to the bottom. After an hour the solution, while still held at -30°C, was pumped on, removing the ether solvent and any
remaining sulfur dioxide. 1 ml ofCFCl3 was added, followed by 200 mg of bromine dissolved
in 1 ml ofCFC13. This mixture was allowed to sit at room temperature for 30 minutes, after which time the solvent and excess bromine were evaporated, the solids were again dissolved
in CFC13 and the solution was filtered and the solvent evaporated again. Finally the remaining solid was dissolved in deuterochloroform and its fluorine NMR spectrum was obtained. The
NMR showed the product to be a mixture ofF-adamantyl bromide and F-adamantyl hydride.
22 e. Reaction ofF-lithioadamantane and ethyl bromide
1-Hydryl-F-adamantane (0.090 g) dissolved in ethyl ether (1.0 ml) was placed in an ampule (~5 ml) capped with a septum. A diethyl ether solution of methyl lithium (1.6M, 0.2 ml) was injected dropwise through the septum at -80°C. The mixture was allowed to sit for
2 hours. After 2 hours, an excess of ethyl bromide (0.2 ml) was added to the above mixture, still at -80°C. After 3 min the solution is allowed to warm to RT over 30 minutes after which the ampule is attached to the vacuum line. Following trap to trap fractionation, 0.045 mg of
Ethyl-F-adamantane was isolated (47% yield). 19F peaks at -l 13.7(s), -121.2(s), -219.2(s).
1H peaks at l.34(t), 2.39(q).
G. Fluorofullerenes
1. Synthesis of CJ48
100 mg offullerene was ground in an agate mortar until very fine. 2.75 g ofNaF was slowly added to the ground mixture with further grinding until the NaF was a uniform brown/grey color. The reactor consisted of a 3/8" monel tube 8 inches long, affixed with
Monel swagelock fittings. The tube was then packed with the NaF/C60 mixture. This was done by inserting a roughly 3/8" solid glass rod which has teflon tape wrapped around one end 1/4 of the way into the monel tube. The NaF/C(i() mixture is then added using a funnel and is very slightly tamped down. The plug needs to stay in its spot in the tube without cascading down or moving when gas pressure is applied, on the other hand, if the tube is packed too
tightly the F2 /N2 mixture is not able to pass through the tube. The tube is then carefully connected to the fluorine system, which consisted of two fluorine-proof gas transducers, one
23 of which was connected to 100%, F2 , the other to N2 . Using a valve system, any percentage
ofF2 /N2 mixture or either alone is possible. The other end of the monel tube is attached to a teflon tube which was buried in a flask of alumina to trap excess fluorine. The first half of the monel tube was wrapped with heating tape, and the very end of the monel tube was wrapped with a heat sink made from aluminum foil. (See figure 4 for an illustration of the apparatus.) The front of the tube is heated to 250°C under a flow of nitrogen to ensure everything is dry for 30 minutes, and then the fluorine percentage is gradually increased to pure fluorine. The reaction is allowed to proceed for 20 hours. The tube is cooled to RT and then purged with nitrogen for 30 minutes.
The tube is then washed through with CFC13 (washing out the plug), the solid is
extracted a few times with CFC13 and the solvent is then dried and evaporated. An otf white/tan solid results. This solid is again triturated with 3. 3 g of sodium fluoride, and a new
JieltmlTlpe HaitSink NaF/PullcnnePlug / ,I
~ 3/8" Mooel Tube
Pluorinelnmclmr Alumim taUblMr
F1uomu, IOlllce Figure 4 - System for Fluorinating Fullerenes 24 plug is formed in the monel tube. This plug is fluorinated at 275°C for 16-18 hours. The extraction process is repeated. The solid resulting from the extraction now should be snow
1 white, indicative of very pure CJ48 . Purity was verified using 9F NMR and/or neg CI MS.
Typical amount produced was 160 mg, 70% yield.
In later runs, the middle extraction step was omitted, simply allowing the fluorine to flow through the plug for roughly 40 hours at 250°C. This method worked nearly as well with only a slight reduction in yield.
2. Reaction of fullerene with fluorine at various temperatures
The apparatus and methods used were the same as in the synthesis of C6cf 48, however lower tempuratures were used for 20 hours, the cream to brownish· solids resulting from the extractions were then analyzed using neg CI MS. The product consisted offluorofullerenes
from Cc,of 48 to C&if36. (m/z 1632 to 1404 with the predominant peaks following the formula
1632- (n X 38)).
3. Reaction of C~CN, C60(CH3)CN, and C60(CN)2 with fluorine at various temperatures
The apparatus and methods used were the same as in the synthesis of C&iF 48, however lower tempuratures were used for 20 hours, the cream to brownish solids resulting from the extractions were then analyzed using neg CI MS. The product consisted offluorofullerenes
from C&iF 48 to C&iF 18, as well as small amounts of substituted fluorofullerenes.
25 4. Reaction of CJ48 with reducing agents
a. Heat
A small sample (50 mg) ofCJ48 was placed in a 1/4" diameter quartz tube, roughly
4 inches long. The tube was attached to a nitrogen source using swagelock with teflon ferrules. A slow stream of nitrogen was passed over the fluorofullerene for 90 minutes while heating the sample to 150°C. After this time period, the tube was allowed to cool, the sample was removed and analyzed using negative CI/MS. The product consisted offluorofullerenes from CJ48 to CJ36.
b. Thiophenol, Borane/Dimethyl thioether, Mg/Fe/Zn metals, Fe(CO)5
In all cases, a small sample (50 mg) ofCJ48 was placed in a 2 dram vial. The solid
was then dissolved in 1 ml of CFCl3. The reducing agents were then added in large excess,
100 µI of the liquids/solutions or 100 mg of the metals. The vials were then tightly capped and placed in a cool dark spot for 16 hours. After this time the solution was decanted from
any solids, solids were washed with CFCl3, and the solvent was then evaporated and the resulting yellow/tan material was analyzed using negative CI/MS. The product consisted of fluorofullerenes from CJ48 to CJ36 .
c. Hydrogen
A small sample (50 mg) of CJ48 was placed in a 1/4" diameter quartz tube, roughly
4 inches long. The tube was attached to hydrogen and nitrogen sources using swagelock with teflon ferrules. A slow stream of hydrogen was passed over the fluorofullerene for 90
26 minutes while heating the sample to 150°C. After this time period, the tube was allowed to cool, the sample was removed and analyzed using negative CI/MS. The product consisted
offluorofullerenes from CJ48 to CJ36_
H. Norbomyl cation
1. Formation of cation
In a typical experiment 0.25 g of2-chloronorbomane or 2,2-dichloronorbomane was
placed into a valved ampule and attached to the vacuum line. 0.500 g of SO2CIF was then condensed onto the chloride and the solution was brought back to atmospheric pressure with nitrogen and allowed to warm to room temperature. In similar fashion, 1.25 g of SbF 5 and
2.0 g of SO2CIF were mixed, also forming a solution at room temperature. In a glove bag equal amounts of the two solutions were mixed, with care being exercised to keep both solutions below -50° Celsius. After being mixed together, the solution containing the carbocation is added to an NMR tube fitted with a J Young valve and after being sealed it is kept at -80°C until the NMR is ready to receive the sample.
27 Chapter 3 - Results and Discussion
A. Methods for the formation of 1-hydryl-F-adamantane
Our lab has now successfully produced 1-hydryl-F-adamantane by a variety of methods. In short these consist of either reducing 1-chloro-F-adamantane or reducing or hydrolyzing an F-adamantane ester. See Scheme 2.
The first method for the reduction of F-adamantyl chloride is using tributyltin hydride15 . Tributyltin hydride is ideal for converting a Cl to an H, and using AIBN as an initiator proceeds via a free radical mechanism. This reaction has a relatively short reaction time (5 hrs), and has an easy workup, simply pumping on the mixture held at -22°C pulls everything off of the hydryl. The yield for the reaction, however, is still typically less than
50%. While this isn't currently understood, it is a running theme in these reduction reactions.
Also, with possible intent to scale up these reaction mixtures, tributyltin hydride is expensive and not trivial to handle. This reaction would also be useful in systems with other
H
('H ~OH.,('F( t3:lw "
KOH- J 8-c1,)Yrn-6
Scheme 2 - Methods for the Formation of F-adamantyl Hydride
28 functionalities, as long as they are not sensitive to free radical hydride donors.
F-adamantyl chloride can also be reduced using zinc metal in dioxane. This method requires a long reflux and standing time (24 hrs)24 . Luo also reports that some starting material precipitates. This is due to the poor solubility ofF-adamantyl chloride in ethers. As above, the reported yield was only 42%. This reaction would be poorly suited to scale up due to the limited amount of active zinc surface at any particular time.
The most promising reaction with respect to yield is the photochemical reduction of
24 F-adamantyl chloride . In this reaction, chloride is dissolved in a methanol/CFC13 mixture and irradiated using a 550W medium-pressure mercury lamp. Again the reaction most likely proceeds with a free radical mechanism. Luo reports yields as high as 86%, but this was on a small scale, still using long irradiation times. To achieve those yields on a larger scale irradiation times as long as 70 hours were necessary. This reaction is ideal on a small scale for high yields, or for systems where several chlorides are being replaced. In these systems, the hydride donor is in large excess and the chloride solution is rather dilute which lessens the
amount of side reactions. This is probably why this reaction is highly efficient with respect to yield. This reaction may not be as useful with more volatile halides because it is difficult to control the temperature near the mercury lamp.
Zhang reported17, in an attempt to make F-adamantyl alcohol from F-adamantyl
chloride and sodium hydroxide, that hydryl was formed instead. The reaction conditions are
quite harsh, 120 ° C for 48 hours in the presence of 18-crown-6 using benzene/water as the
solvent in a sealed stainless steel canister. The 18-crown-6 could be omitted if tetraethylene
glycol dimethyl ether was used as the solvent. In similar reactions, poly F-adamantyl
29 chlorides could be reduced to the corresponding hydrides. The yield for this reaction is quite good, reported as 73%.
Zhang also has reported using lithium aluminum hydride17 to reduce F-adamantyl chloride. Although the yield again is always less than 50%, it is the method prefered by the author, and it has been scaled up to reduce several grams of F-adamantyl chloride at a time.
Although LAH is difficult to handle, after unreacted LAH is hydrolyzed, the work-up is simple. It is possible to proceed with a conventional work-up (solvent extraction), or can attach the vessel to the vacuum line and pump it. Using the conventional work-up, some product is lost due to heating of the ether solution during hydrolysis which is very difficult to control. Using the vacuum work-up, all product is collected, although some may remain either physically or chemically bound to the aluminum byproducts because they are never fully solubilized. Either way, yields are similar and high quality F-adamantyl hydride is formed and is the only compound (including starting material) which is isolated by either work-up. This seems to indicate that byproducts are either being formed which are volatile (broken cages), involatile (polymer), or ionic compounds which are only slowly hydrolyzed by acid. This reaction has little utility with other functionalities because of the high reactivity of LAH, but
for this particular compound, this reaction is ideal.
The last two methods involve F-1-Adamantane Carboxylic Acid Methyl Ester18 • This
ester is rather sensitive to hydrolysis which makes it ideal for modification. In the first
method, this ester can be transesterified to a non-perfluorinated methyl ester. This ester is
much more stable to water and can be isolated and purified easily. It has been shown before
that perfluorinated esters are reduced to hydryl compounds, which is also true in this case.
30 Again the yields, however, are less than 50%.
An even simpler way to produce the hydryl compound from this perfluoroester is to hydrolyze it essentially in situ with aqueous base. Perfluorocarboxylic acids readily decarboxylate, therefore by saponifying the ester, you immediately form a pathway to the hydryl compound. The F-carboxylate is somewhat more stable than the ester however, so I have also then included an acid step in the work-up to ensure that everything is protonated
and hence, decarboxylates. Although this synthesis is simple, just adding a work-up step to
isolating the product from the reactor, it was very difficult to obtain clean product. Due to these impurities, even if the actual yields were good, the need for GC purification will
decrease the overall yield by over 50%. See Scheme 3.
o-ci-·_,
LiAlt-1.fl:QO
Scheme 3 - Reactions ofF-Adamantyl Esters
31 B. Synthesis of new F-adamantane derivatives
The facile formation of F-lithioadamantane19 allows a huge variety of synthetic derivatives containing the F-adamantyl group. F-lithioadamantane undergoes reactions characteristic of an alkyl lithium and is even more effective than non-perfluorinated alkyl lithiums at some reactions, such as a simple SN2 on a primary alkyl halide. At low temperatures, methyl lithium is ineffective in this transformation due to the large covalent character of the Li-C bond. F-lithioadamantane, however, is highly ionic, and quickly undergoes the SN2 reaction with ethyl iodide to form ethyl-F-adamantane.
With a potential future application of this group being included in drug design, or as fluorous tags, I have adapted the ethyl iodide reaction to include a protected functionality for future modification. F-lithioadamantane also reacts with TIPS protected bromoethanol to form TIPS protected F-adamantylethanol. Due to the high solubility of this compound in deuterochloroform, we were able to see, for the first time, the 13C peaks from the F-adamantyl moiety itself. See Figure 5.
Another reaction of organolithium reagents that we have shown to also take place with F-adamantyl lithium, is nucleophilic attack of carbonyl groups. F-adamantyl lithium attacks acetyl chloride very quickly, even at -80°C. This reaction forms an interesting system.
Due to the basic conditions of the reaction and stabilization due to the electron withdrawing
F-adamantyl group, the major product is the lithium enolate. Upon addition of acid, the enolate tautamerizes to the ketone, in this case F-adamantyl methyl ketone. Again, should the goal be adding functionality, the acyl halide would contain whatever linker or synthon was desired.
32 Ul:l'IH)lll'S
h
d
cord a b A
11 0 100 90 80 70 60 50 PPl1 Figure 5 - Carbon spectrum of the F-adamantyl group
Another interesting possibility is the formation of F-adamantyl sulfonic acid. This acid is potentially a rather strong acid which may be soluble both in aqueous and fluorous media. If we could prepare this acid, it would then be trivial to form di and tri acids in this system and perhaps even F-1,3,5,7-adamantyl tetrasulfonic acid, a very interesting aqueous species.
In our first attempt to form the F-adamantyl sulfonic acid, we reacted F-adamantyl lithium with sulfuryl chloride fluoride, expecting it to attack the sulfonyl group just as it attacked the carbonyl group, losing chloride and forming F-adamantyl sulfuryl fluoride. This unfortunately was not the case. Instead of the anion attacking the sulfonyl, it underwent an
SNX reaction, preferentially attacking the chloride of the sulfuryl chloride fluoride. This
formed F-adamantyl chloride and a lithium salt which probably lost SO2, leaving LiF. See
33 scheme 4.
To circumvent this problem, we decided to use S02 to react with the anion instead.
25 It is well known that Grignard reagents react with both CO2 and S02. The lithium salt thus formed would then be oxidized with bromine, hopefully forming the F-adamantyl sulfonyl bromide which could then be easily hydrolyzed to the sulfonic acid. Unfortunately again, this wasn't the case. The product formed instead was F-adamantyl bromide. It appeared as though the reaction with sulfur dioxide had occurred with the formation of the lithium salt which precipitated. Upon oxidation with bromine, however, the compound apparently desulfonylated which left F-adamantyl bromide.
See Scheme 5 for a summary of these new compounds.
C. The electronegativity of the F-adamantyl group by N1MR.
The electronegativity of a new group can easily be determined by NMR.25. The chemical shift of protons neighboring a group is roughly proportional to the electronegativity of the group. We graphed the chemical shift of ethyl halides, both the methylene and methyl protons versus the electronegativity of the halide. This method only works, of course, if there
lJ
+ SO: + 1.tF
Scheme 4 - SNX Attack of Sulfuryl Chloride Fluoride
34 Li
t I
( II;,( ll( I
j
Ur
Scheme 5 - Reactions of F-Adamantyl Lithium
35 is zero or constant magnetic anisotropy in the groups. Although there are fluorines in close proximity to the ethyl group, the F- adamantyl group is highly symmetric and should not show anisotropy. Table 3 presents the data and Figure 6 shows the two plots used to determine the results discussed below. Note that the chemical shift for the methyl protons in the ethyl group are inversely related to electronegativity.
The proton chemical shifts for ethyl-F-adamantane are 2.39 ppm for the methylene and
1.34 ppm for the methyl. Using the data in Table 3, the methylene protons give an abnormally low chemical shift, this is most likely due to excess shielding from the six fluorines which are very close in proximity to the methylene protons. The methyl protons give a more reliable chemical shift resulting in an electronegativity near that of chlorine (3.0).
The electronegativity of the trifluoromethyl group has been calculated by a variety of methods.26 Table 4 shows a summary of these results. All of the various methods in this paper agree that the electronegativity of the trifluoromethyl group is somewhat greater than that of chlorine. Our result, then, for the F-adamantyl group having an electronegativity very
Table 3 - Chemical Shifts of Ethyl Halides
Electronegativity Methylene Chemical Shift Methyl Chemical
(Pauling) (ppm) Shift (ppm)
EtF 4.0 4.36 1.24
EtCl 3.0 3.47 1.33
EtBr 2.8 3.37 1.66
Etl 2.5 3.16 1.88
36 Chemical Shift vs. Electronegativity of the Methylene Protons of Ethyl Halides
,-.. 4 Cl C :::, Ill Q;, ]!;- .... Ill Cl ·~Q) 3 eC • i3 Q) [i
2 -,------,------~ 3 4 Chemical Shift (ppm)
Chemical Shift vs. Electronegativity for the Methyl Protons of Ethyl Halides
5~------~
•
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 Chemical Shift (ppm) Figure 6 - Graphs of Chemical Shift vs. Electronegativity
37 T a bl e 4 - Th e El ectronegattv1tv o f t eh n T "fl uoromet hiv l Group Wells Huheey Sanderson Bratch Mullay
CF3 3.35 3.46 3.45 3.49 3.10 close to that of chlorine seems to make sense.
D. The large scale synthesis of C6of 48 and fluorination of fullerene at lower temperatures
In this lab, Gakh et al, were the first to prepare CJ48 , and properly report its composition20 . It is a unique molecule. Space filling models show that the 48 fluorines essentially fill up the space surrounding the sphere. Fitting any more fluorines on the molecule would require breaking open the sphere. This is why we can use rather intense
(250°C for 2 days) conditions to ensure that all of the fullerene is fluorinated, and also, is fluorinated faster than it is decomposed at the same temperature. Also aiding us in this process, is that the higher fluorofullerenes are more volatile than fullerene itself, so they tend to leave the plug and migrate to the cooler areas of the reaction tube. Gakh did this reaction on a several milligram scale using a small diameter quartz tube. Since the price of fullerene had dropped considerably over the years, I attempted to scale up his procedure to make large quantities of fluorofullerene at once. The apparatus and procedure were described above.
This technique is also useful because the temperature can be carefully controlled, so
I have also run the experiment at other temperatures and shorter times to see what the distribution of products would look like, hoping to possibly hit on a 'magic' temperature, where enough of another fluorofullerene compound would predominate that it could be
purified in some manner and studied just as we have with CJ48 .
38 Before discussing the results, it is important to consider the processes occurring in the mass spectrometer. Chemical ionization is a technique which minimizes the amount of fragmentation occurring. This is achieved by using a reagent gas to thermalize the electrons.
The electrons are first absorbed by this gas (usually ammonia or methane). When observing negative ions, we are interested in the following processes: the ionized gas then participates in electron transfer from itself to the parent, or abstracts a proton from the parent. In our case, the fluorofullerenes form stable negative ions, so these are the ions which we observe.
The fragmentation which occurs is from loss of fluoride from the parent. Thus, the pattern seen is in multiples of 19, with the major peaks at multiples of 3 8. Using the spectrum of pure
CJ48 as a benchmark for the amount of fragmentation occurring, we can hypothesize that peaks in higher abundance are actually due to compounds in the sample. The only exception to this would be compounds with widely different electron affinities, but due to the extraction work-up, all products injected are fluorofullerenes, and therefore have similar electron affinities to each other.
The results are as expected and are summarized in Table 5. In all cases except as noted above at 250°C, a distribution of products was formed (see Appendix B for spectra).
At 150°C, a sizeable amount ofCJ44 was formed, as well as C6(if36, with all those in between
also being present. At l00°C, CJ42 and CJ44 were formed and at 50°C, CJ36 was the major product with very little density of peaks with less than 36 fluorines. This seems to indicate that similar to 48 fluorines, 36 fluorines may be a sort of 'magic' number due to its high symmetry. This also suggests that 18 fluorines would also be a more stable compound,
although at 50°C, CJ36 predominates and the spectrum is then rather clean at lower masses,
39 Table 5 - Fluorination of C60 at various temperatures
I Temperature (for 20 hours) II Major Product I Other Products I 250°C (40 hours) CJ48 (m/z=l6J2)
1so 0 c CJ44 (m/z=l556) C6of3 6 (m/z=l404) 100°c C6of42(mlz=l518) C6of 44 (m/z=l556)
50°C Ca.F~,; (m/z=l404) C"nF~o unreacted C,;"
but unreacted C60 could be recovered from the reaction mixture. The hypothesis is then that CJ18 is not as thermodynamically stable, but perhaps could be formed with kinetic controls.
E. The fluorination of cyanofullerenes
Furthur expanding on this project, I attempted to directly fluorinate the cyanofullerenes, CJICN, C60(CN)2, and C60CN(CH3). The goal here was either to make substituted fluorofullerenes, hopefully as a large percentage of the reaction mixture if the CN group is retained. Conversely if theCN group falls off easily, the group may activate some or all of the sphere to attack by fluorine at lower temperatures, forming perhaps Fl8 or F36 cleanly.
Unfortunately, neither was the case. The results are very similar to those when fluorinating fullerene itself, with only a few anomalies. At 250 ° C, CJ48 and CJ46 are by far the major products. Careful inspection of the spectum however, shows two peaks which do not follow the 48F-xF pattern. These are at m/z=l639 and m/z=l601. These peaks correspond to CJ47CN, and probably from loss of fluorine in the mass spectrometer
40 C60F 45CN. These compounds are present in very tiny amounts. It is well known that perfluorinated nitriles are highly stable, especially to oxidizers, including fluorine. Also, knowing that the starting material is stable at this temperature to loss of HCN, we can speculate that in the presence of fluorine, the formation and loss of HF is favored at this temperature. This loss of HF is then subsequently catalyzing the loss of FCN, or furthur
fluorinated products such as CF4 and NF3 as the fullerene then regains the pi bond. In summary, it appears as though the HCN substitution may be activating itself, or the surrounding area to attack by fluorine, with the functionality being lost very early in the fluorination. As we explore the fluorination of other similar compounds, we'll see if this holds true.
At 150°C, the results are nearly identical to the tluorination ofC60 at that temperature.
The only difference is again the added presence ofCJ47CN, still in very small amounts. At
I 00 ° C, however, we see an interesting change. The distribution of products is slightly lower than that of fullerene itself Looking at 50°C, we now see a significant change. Instead of
F36 being the dominant species, we are now seeing CJ18 and CJ18 0 as the major species, as well as unreacted starting material. The oxide is due to the fact that water plays a role in this tluorination because of the temperature of the reaction being only 50°C. This seems to verify that attack on the cyanofullerene is easier. Similar to fluorinating fullerene itself at lower temperatures where products with 38-40 fluorines are ending up as thermodynamically
more stable CJ36; products with 20-24 fluorines are ending up as CJ18 . See Table 6.
The fluorination of C60(CN)2, shows similar trends. At 250°C, the distribution is similar, although now we have caught a new product, also in very small amounts - this
41 Table 6 - Fluorination ofC6JICN at various temperatures
Temperature (for 20 hours) II Major Product I Other Products I 2so 0 c C6of 48 (m/z=l632) CJ47CN (m/z=l639)
1so 0 c CJ44 (m/z=l556) C6of 47CN (m/z=l639)
100°c C6of40 (m/z=l480) C6of4 2 (m/z=l518) 50°C CnF1RO (m/z=l078) C£nF1R(m/z=l062)
appears to be C60F47(CF=NF). This would be due to partial fluorination of the CN group.
At 150°C, C6of47CN is also observed in this system. Again at I00°C, the distribution is lower than with fullerene and at 50°C, Fl 8 again is prominant, although F36 is also present.
A new feature in this spectrum is the presence of the entire series from F18 to F36. Again, in the case of fullerene itself, no species below F36 are observed. Looking back to the case ofC~CN, since this compounds shows similar results, it may not be the loss of HF which is catalyzing the loss of these groups. It is more likely that for some reason, perhaps loss of symmetry, that cyano substituted tluorofullerenes are simply thermodynamically unstable. See
Table 7 for a summary of these results.
Finally, one experiment was done on C60CN(CH3) at 250°C. In a very surprising result, no methyl or trifluoromethyl groups were seen on the fullerene skeleton, although again CJ47 CN is observed. It is obvious that this fluorination process, or the fluorofullerenes themselves, are not friendly to substituents, but it is difficult to hypothesize why or at what stage these groups are being lost.
We can make some guesses about the CJ:w'C60F18 abundances. With fullerene itself,
it appears that at 50°C, products higher than CJ36 are formed, but then lose fluorine to form
42 Table 7 - Fluorination of C60( CN)2 at various temperatures
I Temeerature ~for 20 hours} II Major Product II Other Products I
250°C C6of 48 (m/z=l632) C6of 4lCFNF) (m/z=l677)
150°C C6of 42 (m/z=l518) CJ47CN (m/z=l639)
100°c CJ40 (m/z=l480) CJ42 (m/z=l518)
50°C Ccnf 111 (m/z=l062) C,,nF,gO C,,nfl."
the more thermodynamically stable CJ36. In the cyanofullerenes, however, the attack of fluorine on the sphere occurs at an overall lower energy. Thus the amount of overall fluorination is lower, and some molecules remain at C60F 18 which is perhaps marginally stable
at these lower temperatures. In the case offullerene itself at 50°C, attack is more difficult,
so those molecules with enough energy to initially react also have enough energy to take them
above F36.
F. The reduction of CJ48
Galch et.al. also showed that C6of 48 was a strong oxidizer7, reacting with iodide (4.3
moles of iodine released per mole of CJ48), bromide (only 0.2 moles of bromine per mole
of CJ48 ) and also reacting with 2-propanol at elevated temperatures to form acetone. Seeing
that these fluorofullerenes barely react with bromide tells us that their oxidizing power is
somewhere near that of bromine. More importantly when analyzing the fluorofullerene left
over after the reaction with iodine, it contained CJ36 - CJ40 . Is it possible then, that there
would be a reducing agent that would give a good yield of one fluorofullerene isomer,
43 perhaps CJ36?
We know that the formation ofCJ48 as the sole product is difficult, requiring long fluorination times at high temperatures. At 250°C, an equilibrium is established between the formation of CJ48 and its decomposition. In an atmosphere of fluorine, and allowing the more volatile CJ48 to sublime out of the hottest portions of the tube, we can optimize the formation of C60F 48_ We wanted to see what product would be formed by heating CJ48 in an inert atmosphere, in this case nitrogen. Heating it for 2 hours at 150 ° C, gave a distribution
of products, with the largest being C60F44, followed by CJ46 and C6of 42. Heating with hydrogen instead of nitrogen did not change this distribution, and essentially the hydrogen had no effect on the reduction.
We also tried using metals as reducing agents, allowing CJ48 to react with the metal
powder for at least 12 hours in CFC13. These reactions were very poor, magnesium, iron, and zinc - all giving similar results. A wide range of products including starting material to CJ36 •
Also, two organic reducing agents gave similar results, thiophenol and borane/dimethylthioether. Both of these reagents gave distributions of products, with the
largest being C6of 44, followed by CJ46 and CJ42 . The only reducer which gave a notable
result was Fe(CO)5. It gave a primary product of C60F40, followed by C60F42, and then C60F38 and CJ36 . This reagent gave a much stronger reaction resulting in a lower distribution of products. This may be due to the fact that the higher tluorofullerenes were acting both as oxidizers and fluorinators, as it is well known that xenon difluoride fluorinates and oxidizes zero valent metal carbonyl complexes28 .
44 G. Low temperature NMR studies of norbornyl carbocations (see Appendix C for spectra)
1. Low temperature COSY and HMQC of the non-classical norbornyl carbocation
Olah et.al. were the first to observe the interesting character of the norbornyl
29 carbocation by treating 2-chloronorbomane with SbF5 in SO2CIF at -80°C . This ion is considered a "non-classical" carbocation, with the positive charge delocalized on the face of the norbornane (Cl, 2 and 6). The proton and carbon spectra of the ion at this temperature consist only of three peaks. This is due to rapid exchange of 2 and 6 protons ( 6, 2 hydride shift) on the NMR timescale, see Figure 7.
These original results were obtained using low-field instruments, and although the assignments seemed sound, we decided to revisit this extremely interesting ion with modem
2D techniques. Both the COSY and HMQC of this ion do agree with Olah's original
2
:g "' ...... 9 o .. 0 ... .. "' "'.. "'
~...,.--.,~,...... ,.....,..--~~-,--,---r-·"T'~~~·~-.---r---.---,---,- -~--, .., 7 , , ~ g a, C, .."' Figure 7 - Low Temp NMR of the Norbornyl Cation
45 assignments. As expected on the COSY spectrum, there is no crosspeak between the peak corresponding to H4 and the peak for H1, H2, and~, but H3, Rs and H7 show crosspeaks to both other peaks. This is because H4 and H., H2, and ~ show no scalar coupling. The
HMQC shows that the most deshielded peaks in both the carbon and proton spectra correspond to Hand C 1, 2, and 6, the smaller middle peak to Hand C 4, and the most shielded peak to Hand C 3, 5, and 7.
2. Low temperature kinetic study of the norbomyl carbocation using EXSY
The kinetics of this ion are also of interest. Although Olah has estimated the rate constants using line shape analysis,26 this method relies strongly on theoretical models. We have determined the kinetic data for this ion directly, using EXSY.
2D NOESY-EXSY is a technique which can be used to probe exchange at low temperatures. Qualitatively, a phase-sensitive proton 2D EXSY spectrum shows only
NOESY crosspeaks at a temperature where the exchange is frozen out, but new, low intensity crosspeaks begin to appear at higher temperatures where exchange is occurring. The 2D
EXSY data can also be treated qualitatively using methods summarized by Perrin and
Dwyer.30
In order to quantitate this experiment, the exchange rate constants must be estimated, then the relaxation times of the exchanging protons are determined using the standard inversion-recovery technique to find T 1. Olah had already reported these rate constants using line shape analysis, '19 these values, as well as the TI are then plugged into Equation 1. These
46 1 optimal mixing time~ 1/(T,· + kAB + k8 ,J (I) values are summarized on top of Figure 8. The 2D EXSY spectra were then obtained and the data was processed using equations 2 and 3, where tm is the mixing time, X is the mole
fraction, k' is the sum of the foiward and reverse exchange rate constants, IAA and l88 are the
diagonal crosspeak intensities and IAB and 18 A are the crosspeak intensities. Considering that
k'=(l/1m) x ln[(r+l)/(r-1)] (2)
r={ 4XAXiIAA +IBB)/ (IAB +IB,J} - (XJXB)2 (3) the foiward and reverse rates are equal, the rate constant for the conversion of A to B is given by equation 4. These rate constants are also listed in Figure 8.
kAB=k'/(l+XA/XB) (4)
Activation parameters for the exchange process were then obtained by an Eyring analysis of the k values. They agree well with the values obtained by the line shape analysis by Olah.
3. Characterization of a new ion, the 2-chloro-2-norbornyl carbocation
The reactions for the formation of 1-chloronorbornane from norcamphor are very interesting from a couple of standpoints. One of these is the now well-studied rearrangement that occurs in the second reaction. We believe we have caught this Wagner-Meeiwein 1, 2 alkyl shift31 by NMR. First I will describe the steps in the reaction, see Scheme 6. In the first chlorination step, norcamphor 1s treated with PC15 in PC13, forming
47 Summary of EXSY and Tl experiments of 2-norbomyl cation
Peaks ' 193K 208K 223K (ppm) i Tl(s) Mixing time (s) Tl (s) ! Mixing time (s) Tl (s) Mixing time (s) i . 4.93 I 0.95 1.47 I 2.11 2.82 1.06 0.3 1.52 ! 0.06 2.10 0.01 : 1.85 1.60 1.73 I 2.06
Exchange rate constant from EXSY experiment
1 Exchange sites Rate Constant k (s"') Rate constant k (s" ) Rate constant k (s" 1) I at 193k at 208K at 223K --·- I I 1 - 4 -- 0.52 5.42 45.99 2,6-3,5 0.45 6.47 50.38 Average k 0.485 5.945 48.185
Eyring plot of exchange rate constant I 193K 208K 223K I ln(k1T) ! -5.986 -3.555 -1.532 I l,'T i 5.181 * 10·J 4.808 * 10•j 4.484 * 10·J ! ··-
Eyring plot for the e,changa rate constants
.:;: .
. , . - ,:,i. _.,, -
C
-5 •
I ·•. ! I
-7 -- I .. •• ,a 50 52 s., 1/T x ~000 Linear fitting result: Slop = -6393.428, Constant= 27.153
From the equation: ln(k!T) = -.0.H/(RT) + (t.S/R) + ln(k0/h)
R = 8.314 J / (mol*k), ka = 1.38 * 10·23 J/k, h = 6.626 * 10·34 J / Hz
t.H = 6393.428 * R = 6393.428 * 8.314 = 53.15 * 103 J /mol.
.0.S = R*{27.153 -ln(k8/h)}=8.314 * { 27.153-23.760} = 8.314 * 3.393) = 28.2J / (mo! * k) = 6.74 eu
Figure 8 - EXSY Kinetics Data
48 o__ ccr l'ct_. PC~ c6 • .! hydrile a bl.lr.V:-lion !
Cl Y.unuin,gt,J. . -20 d.;-£JCCS cb Scheme 6 - Reactions of Chloronorbornanes
2,2-dichloronorbomane. The second step is also Lewis acid mediated, this time using A1Cl3.
A carbocation/tetrachloroaluminate complex is formed which then undergoes the Wagner
Meerwein rearrangement. Finally a hydride is abstracted forming 1-chloronorbomane. Our goal was to reproduce the Lewis acid conditions by treating 2,2-dichloronorbomane with
SbF5 in SO2CIF. The superacid media allowed us to study the carbocation formed at low temperature. We believe this new ion to be the 2-chloro-2-norbomyl carbocation, formed by the loss of chloride. We obtained proton and carbon spectra of this ion, as well as COSY,
HMQC, DEPT, and NOESY spectra. See appendix C for all spectra.
The first thing we found interesting about this cation was that unlike the norbomyl cation formed from either 1 or 2-chloronorbomane, this cation was classical. It showed seven clear resonances in the 13C spectrum, only one of which showed an extreme amount of deshielding. Were the carbocation non-classical, this deshielding due to the positive charge would have been spread over 3 carbons. The most deshielded peak was the carbocation at
284.5 ppm. The DEPT spectrum showed that this peak indeed did not have a proton attached to it. The DEPT also showed the presence of two C-H carbons and four H-C-H carbons.
49 Already knowing that 1-chloronorbomane does not form a bridgehead cation, and that these are highly unstable, we were not surprised by the presence of the two bridgehead C-H's.
Essentially this was enough information to deduce the structure shown below (Figure 9). At first glance, a halogen directly attached to a carbocation would be thought to be unstable, however, these are well known32•33 . The chlorine (and fluorine as well) tends to stabilize the carbocation by back donation from the filled halogen p orbital into the empty carbocation p orbital.
We assigned the proton and carbon spectra using the HMQC, as well as previous assignments made by Olah for the 2-methyl-2-norbomyl cation. Table 8 is a summary of those results. Finally, we checked our results by verifying crosspeaks on the COSY and
NOESY spectra. Those results are summarized in Table 9.
There are a few very interesting properties to this ion. The proton spectrum of this
Cl
Figure 9-The 2-Chloro-2-Norbornyl Cation
50 Table 8 - Summary of HMQC Results
13C peak Correlated Proton Peaks Assignment
284.5 a
76.9 4.93 b
61.0 ~3.6*(exo), 3.52/3.48(endo) d
42.5 3.40 C
40.5 ~3.6*(exo), 2.32(endo) f
36.0 2.47 e 22.2 2.07(exo), 1.66(endo) g
T abl e 9 - Summary o f COSY an d NOESY resu ts Proton peak(s) COSY Crosspeaks NOESY Crosspeaks
b fl weak) e(weak), flweak), g(weak)
C d(weak) d(weak)
d d(strong), c(weak) d(strong), c(weak)
e d or flweak)* b(weak), flweak), g(weak)
f g(strong), b(weak) g(strong), b(weak), e(weak)
g g(strong), f(strong) f(strong), g(strong), b(weak), e(weak) *NOTE - the chemical shift of one F and D proton lie on top of each other
51 ion matches closely with the spectrum of the 2-phenyl-2-norbomyl cation, much more so than with the 2-methyl and 2-ethyl cations. Again this implies that the chlorine bond does indeed have some double bond character, just as the phenyl cation would due to resonance with the aromatic system. In our system, as well as the other three, the f protons have widely different chemical shifts, over 1.2 ppm in our system. Olah attributes this, as well as the very deshielded b methine proton (4.9 ppm in our system) to the fact that there is indeed some nonclassical nature to these carbocations, with significant amounts of deshielding being experience by Hb and Hi(exo). Also, we were able to resolve the d pair, and g pair. Thee pair, however, in our system is not diastereotopic.
We also predicted that if we warmed the sample to -20°C we may see the ion beginning the Wagner-Meerwein rearrangement. Unfortunately the ions are highly unstable and quickly polymerize at that temperature, but we were able to catch a quick carbon and proton spectrum. Both are very interesting, the carbon spectrum coalesced to only 5 peaks, as well as two very broad peaks. These peaks indicate exchange, most likely a hydride shift between the two secondary centers. The carbocation peak moved very little (294.4), but the key in this spectrum is a new peak at 91.1 which is possibly the carbon attached to the now bridgehead chlorine. In the proton spectrum, the peak for He has moved from 3.4 ppm to 3.6 ppm (now closer to the carbocation), but more interestingly the peak for I\, moved from 4. 9 ppm to 5.6 ppm, possible indicative of this old bridgehead proton now becoming the proton on the carbocation. So we can confidently hypothesize that the 2-chloro-2-carbocation is indeed becoming the 1-chloro-2 or 3-carbocation at this higher temperature.
52 Chapter 4 - Conclusions
This project has explored many new aspects of the chemistry of the F-adamantyl group. The F-adamantyl anion is remarkable, showing incredible stability as a lithium salt, being an excellent nucleophile, both in classical SN2 reactions, as well as attack on carbonyl and sulfonyl groups. Unfortunately, these properties also make it ideal for undergoing decarboxylation and as we have now shown, and more surprising - desulfonylation as well.
Also, because of its bulk, it undergoes an SNX reaction with sulfuryl chloride fluoride, rather than attack the sulfonyl itself. Even though, in the case of the sulfonic acids, we couldn't form the target molecule, the ability of this ion to undergo these new types of reactions reveals that there is much very interesting chemistry still hidden here.
More importantly, we have shown that it is possible to use this group as a substituent or building block for larger molecules, or as a fluorous tag. It is well known that the trifluoromethyl group has an electronegativity (hence inductive effect) between that of fluorine and chlorine, and thus can be used to fine tune the specific activity of a pharmaceutical. The F-adamantyl group is very close to chlorine in electronegativity, but would have unique properties due to its large size and fluorophilicity. We have also shown the ability to substitute the F-adamantyl group so it can be used as a fluorous tag. We suspect that this group would have unique solubility properties which would need to be explored. For
example F-adamantyl chloride is soluble in CFC13, but only slightly in chloroform. F
adamantyl hydride is soluble in CFC13, more soluble in chloroform than the chloride and rather soluble in ether, probably with hydrogen bonding. Ethyl-F-adamantane is not particularly
soluble in either CFC13 or ethanol, but is very soluble in a 50/50 mixture. Finally, the TIPSO
53 protected F-adamantylethanol is highly soluble in chloroform and ether. Obviously, the F adamantyl group would show novel partitioning useful in many new areas. Not only could this group be used as a fluorous tag, but simply by adding a substituent, the partitioning of the compound could be fine tuned to whatever solvent or polarity was needed. This could even extend as far as making these compounds soluble in water, fluorous phases, and organic phases, and altering this tuning in the middle of a separation.
Although making controlled substitutions and otherwise functionalizing fullerene is now a field of its own, no one had investigated fluorinating these new compounds, or
derivitizing CJ48 itself Even though we were not successful in producing either a substituted fluorofullerene, or a single fluorofullerene apart from other compounds as is
possible with CJ48 . We did learn a great deal about these fascinating compounds. The various cyanofullerenes fluorinated very similarly to fullerene itself and did not retain the cyano, or other functionality that may be present in any useful amount. We did see though,
that through these compounds, we could access compounds between CJ18 and CJ36, which were not seen by simply fluorinating fullerene itself
Gakh et.al. showed that CJ48 is an oxidizer, but is it possible to use this property to form a single fluorofullerene isomer? Most reducing agents showed the same trend of
removing 2 to 6 fluorines, similar to letting CJ48 decompose at a slightly elevated
temperature with no reducing agent. Only Fe(CO)5 produced a different result, showing more
reduction. This is most likely due to the fact that Fe(CO)5 could be fluorinated as well as
oxidized, proving that similar to other highly fluorinated compounds, CJ48 is an oxidative fluorinator.
54 Although the rearrangements of norbomanes have been known now for over a century, and the formation of 1-chloronorbomane from norcamphor via 2,2- dichloronorbomane is a well known reaction. We believe we are the first to catch this rearrangement by NMR. We have also characterized a new ion using these same techniques.
The 2-chloro-2-norbomyl ion is very interesting in itself, being rather stable and having chemical shifts similar to the 2-phenyl-2-norbomyl ion, showing the impressive degree of double bonding in the carbocation-chlorine bond. As the wider picture, in catching this ion and watching it rearrange, we believe we can now explore the next step in the reaction. This step is the hydride abstraction which forms the final 1-chloronorbomane product. It is believed that the hydride is abstracted from another chloronorbomane resulting in a polymer which lowers the yield of the reaction below 50%. We can now show that it may be possible to form the 2-chloro-2-norbomyl carbocation at low temperature, warm it to allow it to rearrange and again freeze out the cations at low temperature, and then finally quench with a weak hydride donor before warming the solution. Carefully done, this could give a nearly quantitative yield of product.
Although the three focuses ofthis thesis are not apparently related, the goals were still the same. That is to explore these three very unique chemical systems to learn more about how they undergo chemistry, but more importantly to then apply them to areas where they may be useful, now or in the future.
55 References
56 I .Studer A; Hadida S; Ferritto R; Kim S; Jeger P; Wipf P; Curran D. Science, 1997, 275, 823.
2.Rover S; Wipf P. Tetra. Lett., 1999, 40, 5667.
3.Curran D; Hadida S; Kim S; Luo Z. J Am. Chem. Soc., 1999, 6607.
4.Adcock, J. L.; Horita, K.; Renk, E.B .. J. Am. Chem. Soc. 1981, 103, 6937
5.Adcock, J. L.; Cherry, M. L. Ind Eng. Chem. Res., 1987, 26, 208.
6.Adcock, J. L.; Horita, K.; Renk, E.B.J. Am. Chem. Soc. 1981, 103, 6937.
?.Adcock, J. L.; Robin, M. L. .I. Org. Chem. 1984, 49, 191.
8.Cherry, M. L. Ph.D. Dissertation, The University of Tennessee at Knoxville, March 1986.
9.Adcock, J. L.; Robin, M. L. .I. Org. Chem. 1983, 48, 2437.
IO.Adcock, J. L.; Robin, M. L. .I. Org. Chem. 1984, 49, 1442.
I I.Adcock, J. L.; Evans, W. D.; Heller-Grossman, L. .I. Org. Chem.1983, 48, 4953.
12.Adcock, J. L.; Evans, W. D . .I. Org. Chem. 1984, 49, 2719.
13.Adcock, J. L.; Robin, M. L. .I. Org. Chem. 1983, 48, 3128.
14.Adcock, J. L.; Luo, H.J. Org. Chem. 1992, 57, 2162.
15.Adcock, J. L.; Luo, H.; Zuberi, S. S. J. Org Chem. 1992, 57, 4749.
16.Adcock, J. L.; Luo, H. J Org. Chem. 1993, 58, 1704.
17 .Zhang, H. Ph.D. Dissertation, The University of Tennessee at Knoxville, May 1995.
18.Zuberi, S. S. Ph.D. Dissertation, The University of Tennessee at Knoxville,
19.Adcock, J. L.; Luo, H.J. Org. Chem. 1993, 58, 1999.
20.Gakh, A.; Tuinman, A.; Adcock, J.L.; Sachleben, R.; Compton R . .I. Am. Chem. Soc., 1994, 116, 819.
21.Corey, E.J.; Venkateswarlu, A. .I. Am. Chem. Soc., 1972, 94, 6190.
57 22.Keshavarz-K, M.; Knight, B.; Srdanov, G.; Wud~ F . .I. Am. Chem. Soc. 1995, 117, 11371.
23.Ward, G.A.; Bower, B. K.; Findlay, M.; Chien, J.C. W. lnorg. Chem. 1974, 13,614.
24.Luo, H. Ph.D. Dissertation, The University of Tennessee at Knoxville, December 1992.
25.March, J. "Advanced Organic Chemistry", Forth ed., John Wiley & Sons, New York, 1991.
26.Mullay, J. .I. Am. Chem. Soc. 1985, 107, 7271
27.Gakh, A.; Tuinman, A.; Adcock, J.L.; Compton, R. Tetra. Lett., 1993, 34, 7167.
28.Holloway, J.H.; Hope, E.G. .I. Fluorine., 1996, 76, 209.
29.0lah, G.A.; White, AM.; DeMember, J.R.; Commeyras, A.; Lui, C.Y. .I. Am. Chem. Soc., 1970, 92, 4627.
30.Perrin, C.L.; Dwyer, T.J. .I. Chem. Rev., 1990, 90, 935.
31.De Mayo, P. "Molecular Rearrangemenf', John Wiley & Sons, 1963.
32.Laube, T.; Bannwart, E.; Hollenstein, S. .I. Am. Chem. Soc., 1993, 115, 1731.
33.Christe, K.; Zhang, X.; Bau, R.; Hegge, J.; Olah, G.; Prakash, G.; Sheehy, J. .I. Am. Chem. Soc., 2000, 122,481.
58 Appendices
59 Appendix A Substituted F-Adamantane Spectra
60 1...
N • >o tH"ln---- -N I _;
..-N I
-N -I
-"' ..I
Cl-.. I 0..... I
-..ID I ..1 : ~--=----=--- ~ - .,...... ------, "'-.. ~-=1:· I '· i ~ -..... i[~ I
111·111----._ __,______=,; :::: } ICl"Hl• .. I -.. I .. -..I .. I
Figure A-1 1-Chloro-F-Adamantane {1 9F)
61 ----108.741
..I
---:_-:--..=s _ ... -~:-__ __- --•::.' ..I ==·~- ==~ -•-• ...... -~-- ·==--==--~ ---
..I ..0>
..I .."'
I ..N
I ..N
----111.457 ,. •
Figure A-2 1-Hydryl-F-Adamantane {1 9F and 1H)
62 90?"1 ---··
-- .f'"; )
==-,,~a· "" ' ,..,
oos·r 915 '£ """
.r, .,
"" "" J
Figure A-3 F-Adamantyl Methyl Ketone and Enolate (1H)
63 ,..
N ... 0 ... OD ... a, ......
... 0
N
OD
N a,
N...
N <:>' - -i "' - a, -... - N - 0
"' Figure A-4 TIPSO protected F-Adamantylethanol (1H)
64 0 N
0 n
.,0
.,0
0..
..0
0..
Figure A-5 TIPSO protected F-Adamantylethanol (oC)
65 Appendix B Fluorofullerene Mass Spectra
66 File,D454B Ident,6_26 SHO(l,7) Acq, l-DEC-1997 16113:34 +2158 Cal,D454B ZAB-SEQ4P CI- Magnet BpI,5922560 TIC,578289856 Pile TextsRua C60F48 (old sln) S250 RO AFSe-6 1009< 163,1. 9 20V
19
18
17
16 ~ 15 -· 70 14 ~ 65 13 (1) ec 60 I 11 ~ 55 O'I so rlO -..J 45 ~-9 "'C t E; 40 8 (1) 7 35 719.9 n 30 6 O'I 0 ~ 25 5 ~ 00 20 4
15 3 1517. j 10.. 2
5 1
0 ~4,1.V._.,q,/\ 1,1 ,110 , ,· , , , ,,,µ-t-"44-JW\J\.i\",J\!. , , , , . ·, , to 1100 1200 1300 1500 1600 1700 1800 1900 2300 2400 m/Z P~l d)394A Ident,14 ___17 Win SOOPPM Acq: 1-JUL-1997 16:24:38 ~1,40 c .. l,D394A Z -SEQ4P CI- Magnet BpH,1556 Bpl:6774388 TIC:99066472 Flags:NORH P le Text,K.Slllitb. (Compton) "C60,150" S240, r0-65005, AFSe-6 10~ 1555.9 2557mV
95 2430
2302 90 1517.9 85 I 2174 80 , 2046 u 0 75 1 1!118 ""1 0 10) 1790 Ir) ,-....4 1662 65 · U03.9 ~ 60 1534 ~N 55 1441 .1407
50 1279 + 0 00 1151 \0 4 s..:; u \0 40_! 1023
35 '. .895 I N I 30 · 720.0 767 ~ 25 639 s~s. 0 20 511 I 1 15.: 384 -~i 10 '. 256 ~
~- 759. D il 128 (, ' I ' lJ 2.9 o ; ..• ____.,.J_ 1.l.~.L.L_ .. 1 0 ··r·. ....'.'.'L:~ 1 I .1_ ··-. -----· 600 700 800 900 1000 1100 1200 1300 1-l CO lSOO J60G POO 1 i100 Vi0c 1 20 C•O 2100 2200 2:l°oo m:'z
m/z. m/z.
5 5
6 6
4 4
8 8
g g
10V 10V
7 7
~3 ~3
t t t:
Eo Eo
2000 2000 2100
1900 1900
1800 1800
1700 1700
1600 1600
1556 1556
1518 1518
1s·oo 1s·oo
1ioo 1ioo
Cal,D424A Cal,D424A
00 00
1
12
+l:52 +l:52
11100136 11100136
e220,r0,APSe-7 e220,r0,APSe-7
TIC,939262272 TIC,939262272
1000 1000 1100
0 0
9 9
"C60,100" "C60,100"
BpI,7185408 BpI,7185408
Acq116-SKP-l997 Acq116-SKP-l997
Magnet Magnet
720 720
Idant13_S Idant13_S
CI-
7 7 0 8 0
Text,Smith(Adcock) Text,Smith(Adcock)
10 10
ZAB-SKQ4P ZAB-SKQ4P Pile Pile
Pile1D424A Pile1D424A
0 0
I I
+ +
~ ~ """""" """"""
~ ~
to to
(I) (I)
w w
~ ~ ~ ~ n n
n n
~ ~
0 0
0 0
(IQ (IQ
0 0
O'I O'I
N N
O'I O'I \0 \0 ile1D439A ldent,10_14 Acq,24-0CT-1997 15155145 +4r25 CalrD439A [AJI-SB(l4F CI- Magllet BpI,36094976 TIC,3706864640 ile Text:Slllith(Adc:oc:k) "C60:50" S200 r0-650~2, AFSe-6 0, 720.0 43V
95~ _u
38 901' 85 _36
80 ~-3' 1 u 75~ 32 0 0 ,oj f30 tr) 28 ~ ::1 26 ~N 55~ 23 j ! 50 21 + 0 0 45 l19 uv:> r-...
40 17
35 _15 I 30 __ .13 (::Q'""'" 25 11 (l) 1404. 0 20 9 s 15 _6 -~ ~ 10 j L• 5
•.. ·1 . . ! . .. ..1,.... i· ...... ,~~uLL~J,1~. . m/z 7 O 8 0 850 9 0 9 0 14'00 1~·00 16·00 ~:11'00 760 1000 1050 1100 1150 1200 1250 1300 1350 1450 1550 1650 File,0322A Ident:10_19 Acq:18-DEC-1996 13,54:11 +4152 Cal:D322A ZAB-SBQ4F CI- Magnet BpI:31569152 TIC,5611371520 File Text,Smith (Adcock) Spl 2 S200, R0-1.3$1, AFS •-6 lOO!
JA/Z
Figure B-5 C60HCN + F2 at 250°C
71
Pile1D395A Pile1D395A
File File
ZAB-SE04P ZAB-SE04P
1ocr.. 1ocr..
25: 25: 15 15
20 20 10 10
70' 70'
45 45
75.' 75.'
60 60
35: 35:
55. 55.
651 651
4 4 80; 80;
85 85 so, so,
10 10
go_j go_j
95 95
0 0
5 5
0 0
600 600
.. ..
.: .:
: :
:: ::
: :
: :
: :
i i i i
j j
Text:K,Smith Text:K,Smith
700 700
720,0 720,0
CI-
Idant:l0_l4 Idant:l0_l4
777.9 777.9
Hagnac Hagnac
800 800
(Compton) (Compton)
BpM:1556 BpM:1556
Win Win
900 900
SOOPPM SOOPPM
"C60HCN,150" "C60HCN,150"
---~ ---~
lOOC lOOC
BpI:9725645 BpI:9725645
Acg: Acg:
l l
l l
DO DO
l-JIJL-1997 l-JIJL-1997
S240, S240,
TTC,356413216 TTC,356413216
1~'00 1~'00
r0-650~5, r0-650~5,
16:46:30 16:46:30
~lOC ~lOC
t+J t+J
111:trn 111:trn
Fla~e,NORH Fla~e,NORH
14·)]. 14·)].
AFS.,-6 AFS.,-6
1-ioc, 1-ioc,
,1:21 ,1:21
9 9
l l
1517.9 1517.9
Cal:D.H4A Cal:D.H4A
SO SO
1ss
0 0
I I
; ;
1 1
~ ~
1
s.9 s.9
,,, ,,,
l l
. .
17 17
00 00
1800 1800
h h
P
18
SP SP
·°-1.899. ·°-1.899.
1900 1900
. .
l l
~iJQ(, ~iJQ(,
~100 ~100
' '
~200 ~200
·--·• ·--·•
23'00 23'00
I I
l l
: :
:_s51 :_s51
l l f f
' '
~ ~
t t
' '
f f :. :. ; ;
-.1652 -.1652
. .
c c
.. ..
.. ..
3304 3304
29 29 7H 7H
3121 3121
27 27
3488 3488
0 0
3672mV 3672mV
2387 2387
1101 1101
918 918
2203 2203
2570 2570
2019 2019
1285 1285
1469 1469
184 184
367 367
l836 l836
37 37
54 54
m/Z m/Z
-~ -~
~ ~
o:l o:l
'° '°
u u
tt:: tt::
u u
z z
oJ) oJ)
~ ~
0 0
~ ~
u u
0 0
,-...4 ,-...4
V) V) + +
~ ~
I I
\0 \0
0 0
N N
r--- N N ~ -· File,04268 Ident,1+2+4+6 SMO(l,7) Acq,16-SEP-1997 13:26,01 +1,56 Cal:0424A ZAB-SBO'F CI- Magnet BpI,359672 TIC:160567504 1 File Text,Smith(Adcock) "C60HCN" &220,rO,AFSe-7 r392mV t:c 1oi6r I 90J I ~344 I -....J ' 80 1305 70 ;:_267 E p 60 1480 t 229 0 I 749 ' 50. 915 f.191 ~ ~ I151e w n ,o_ ~153 z 30. 1556 115 20_ 955 10,41 I; . 1181 1248 76 + 19 12 10 999 I , ' 11280 !il I~:~ 1823 , ~W ,·;[I,,!~ ' ' 1763 11862 ' 2147 38 N~ 0 w! :,, ·. ' '~~~i~~". r/i~~. , • l. , . ~..' ' i.Ll...i.L1995 20,96' I. _o ' ' + ., ' ' ,'""'14U,J ' ' ~+.1.,-.J, I.,,'---' 0 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 m/z ~ ...... 0 0 n0 File
File1D440A
ZAB-SBQ4F
1oorir.
"j
70 1:j
15.
30. 65
60~
25.
20
351 40~
55j ,s~ 80i
85
95
S0-1 90
0.
1
1
1 ~
1
i
1 ; ' j
j
Wr1
L
Text1Slllitb(Adcook)
634
:1 : : 1 : : T
L
650
CI-
I4ent,l.2_16
700
Magnet
1~0
~ ~
it
7~0
BpI,9461248
c
. 1
Acq,2,-0CT-1997
•c60BCN1SO"
791
860
J..
.
,
: 1
. r
850
TIC,633436160
,
t
S200
•
..
900
e
1
16130120
r
r0-650~2,
I
950
r
+4158
00 1050 1000
. . , . . . ,
+
APSa-6
r
104~145
•
Cal,D439A
,.._
,
1078
10 1150 1100
,
-.
1
,
12°00
,.....,,_
__
20 1300 1250
.,.,..,...,..,...,...,.-,-.,...,...---,.-,-,--~----,.----'-
1350
14°00
1450
1500
1550
16'00
[_
.3
_,
11V
10
10
2
9
7
7
8 0 8 1 3 4
6
6
5
1
ra/
z
.
~
P=l 00
u
~
u z
0
~
'r) 0
u
+
s
oJ)
~
(l) ~
I
0
\0
N
t--
-.:I" \
Pile1D440A Ident,12_16 Aeq,24-0CT-1997 16130120 +4158 Cal1D439A ZAB-SSQ&P CI- Magnet BpI,9461248 TIC1633436160 Pile Text,Sllith(Adcock) •c60HCN150" S200 r0-650i2, AJl'Se-6 ~~ ~8 t335mV ~ f ~- 318 302 1059 1 .285 to 80_ 268 I '° 75 .. 251 70 235
.218 0\n 65 0 201 ~ 60. n 55 184 50 1043 168 -...J z VI + 45 151 1094 ~ 40 134 N 1128 1145 35 117 ~ 1110 Vl 30 101 0 25 lOU 94 0 1003 r 67 n 20 f 15 r50 • to 10 t-3'
0""""'" 5 ;_17
0 r·1•, , .. , i_O l 10·,o 1080 11'00 1140 ' ' 1ioo m/z File,D372A Ident,23_35 Win l000PPH Acq, 9-JUN-1997 14:43:0: +3:U6 Cal:D.l'/lA ZAB-SEQ4P CI- Magnet BpK,1518 BpI:18204870 TIC:337072512 P:ags:NORH File Text:ICSJDith (Adcock) "CN2" SlSO, RC-650Q2, AFS e-5 100'!,, ,. ... ------•------···x25,00 l48JSU ,- 7 577mV 1 1 f 95.j 1556 t.7199 i 90_] [_6820 I j t u es.i 1632 r 6441 0 0 BO '· 6062 ' tr) 1594 75 [_5683 N .,. 70. (_5304 ~ N 65 r.4925 t i ~ 60., 4546 759 + 55 "4168 N 1442 ,.-._ 50 797 f-3789 z t IO 45.1 740 t. 3410 u r--- ,. ~ 0 40, :.3031 \0
I 816 3 5 , : 2652 u i 30: : 2273 0 25.: 1894 ,...... 721 1404 I 20,J 1515 i O'.:l 15 : 1137 (1) i 49<1-531,57461 46 10 :. 758 702 s oJ) 5 :,_37 9 -~ ! 17 l S ~ o l ··---~ .,.., •• ,17 6 C>t800184tl . u 700 800 900 1000 1100 1~ 00 l.~)(J 14 i)(J 1 ~-.I) u it: 00 17 0 C 1 a'oo H·o0 111/Z file1D396A Ident:10.15 Win 500PPM Acq: l-,JUL-1997 17:0~:48 +l:25 Cal,D394A ZAB-SEQ4P CI- Magnet BpM:1518 BpI:5700950 TlC:161558464 Pl~gs:NORM Pil" Text,K.Smith (Compton) "C60CN2:150" S240, r0-650@5, APSe-6 100"-,, 1517. 9 2152rnV ! 95j 1555.8 2045 1 90; 1937 ...... ~ j ~ 85 i 1829 1 E; 80 j 1722 ~ -~ 75~ :.1614 t:c I 1441. I 10] .1507 lo-- lo-- 1399
60. 1291 "1 1403.9 ss 1 1184 I10 ,..-... 50 .: 1076 " Q 453 :. 968 40. 861 ~ ' f 753 + 35 :! t 30; ~646 N~ I 25_ !_538 594.8 a 20. :.. 430 lo-- j' Vl 15~ ;_323 J 0 10' 720.0 i 215 0 n f"108 :~---~,~1-7~~;~:··~·9··------·--·•-....,_ I 1e _J J:Mlt'IC■ l o 600 160 a6o 900 1000 11.00 1200 1300 1400 16·00 1700 I 8.0 0 1900 2000 2100 22no 23.00 ml% u 0 0 Pile1D425A Ident,1+2+3+6 SMO(l,7) Acq,16-SBP-1997 11:32105 +1151 Cal1D424A 0 ZAII-SKQ4P CI• Magnet BpI,593312 TIC,405550112 ,--4 Pile Text,Slllith(Adcock) "C60(CN)2,l00• a220,r0,APSe-7 10~6~4 ~ 'j ;' i:::•V 90J i 1480 ~N [· 530 720 ' fc464 + 1556 JI.., f.397 ~ 50 .. ' z 00 40 ....._,u r--- 1 1-::: , j · 0 JO· 1081 j i • 1608 f.H9 3 915 : • • '. • ll ( 20 cJ 818 I I: 1,~ . I. ' :: .·.I ! !_132 1247 ,6, I I ' l~ .• t ' I ! ! 10_05 136 8 ' . . •· : ; I 17 94 ; 10. ,,1,11.. ' ti, 113211_91 . 'I' 'I; . ,ij ', I.. ' 18_45 . 2US (·66 o..: .I N . ' ~f...JJ.,,,;..'J.i41l~~_i,i ~¥:~l~,,~1~~.::.r,•c~,~ . , Ii• 700 800 900 1000 1100 1200 1)00 1400 1500 1600 1700 1800 1900 2000 2100 m/z ,--4 I ~ s(1) -~on ~ FiletD441A Ident,ll __l5 Acq12&-0CT-U97 l6t48t48 +4142 cal,D439A ZAB-9RQ4F CI- Magnet BpI,17215488 TIC,5589796864 Fil• Text,Slnith(Ad.cock) "C60CN2t50" 9200 r0-650<12, AFSe-6 lOO'll 720 20V ( 95 19
go 18 ~ t CJQ. 85 l11 f f 16 ~ 80. (1) 75 ilS
a:, _14 I 70 ~ w 65:! _13 12
11 £1 ::1 0 50 ~10 -....) ,.-._ \0 t n 45 [-g ,o ~8 > 3 f N 35 p t + 30 t.6 t N~ 25. ls r" r ~ 20 1062 ~' 1404 t 15 Vl 3 0 1366 l f 10 0 I 1y2 1214 1252 1290 1328 I - 420 f·2 n 5 102 15.32 r1 0 ·L1~'"~~~u~~~UJlk~(:~ ,- rr- • r ..-r · ,' r r , , 1-i r-r--r-r r,--. ·-r .. r ,. , • , ~------...... -.,.-rr,,. ,- r, ""fr, 1 ,-- • r __,.., ; I ; ,...... , .o 1050 1100 1150 1200 1250 1300 1350 1400 1450 1500 1550 1600 1650 m/:z: File,D371B Ident:22_37 Win lOOOPPM Acq, 9-JUN-1997 14:'l0:49 +3:13 Cal:D371A ZAB-SEQ4F CI- Magnet BpM,1518 BpI,14544080 TIC,234646224 Flags:NORM File Text,K.Smith (Adcock) ""CNCH3" 5150, R0-650@2, AFS e-5 100'.. 4------x25. 00 15;19 6054mV
95..; 1556 5751 i 5448 u 90~ 1480 0 ;' 0 1594 5146 lr) "j80. 484 3 N 75 4540 ~ ~ 4238 N 70) 1532 ~ 651 3935 + 60.; ' 3632 ,..-.. M 1 3330 55.j ::r:: 50 1 3027 u i 1 '-' 0 45 j 2724 00 759 1442 z 40 ; 2421 u ha 0 I.O 35 : · 2119 HO u 30 '. ' : 1816
25 ..'. 1513 11 IH? 1211 ~ 20: H04 """'I j 721 15_; 908 ~ IH49 53 5? I II la~6 I (1) 10 I ; 605 l~~HJ. •: .[ ti?I u; 77 2 1 5 ,03 s 9 b.1} 1 1 15 .~ o_ ',jJ] LluLl~11lli11l~lJJ::it·\ l ?6Qiaoo 0 700 800 900 10·00 llOQ 1200 1100 i-l 00 1 !:.•JO lfi,00 17\,0 1aoo l 90(J mlz ~ File:D421A Ident:6 18 Acq: 4-SEP-1997 11:02:25 +5:35 Cal:D422A ZAB-SEQ4F CI- Magnet BpI:13512448 TIC,1184583040 File Text,Srnith(Adcock) "F48H2" s200, R0-650@2, afse-6 100:~ 1556 ,.20V l19 90 t 18 r17 r-16
14 13 12 rrll t:o 1-8' 35 t7 3') . t:-6 148 ls 1 20 r-4 15 : t:.3 1 10 720 I t 2 1401 I r ~l 7 1 0 :-:'---~i-+~-'-'i~1--.------..;.l.;.,1 !J..-0 t~ )J;[~-~-~--~--~- e6o 1000 1200 14°00 1600 1eoo 2000 m/Z File:D422A Ident:6_18 Acq1 4-SEP-1997 11:27:43 +5:36 Cal:D422A ZA~-SEQ4F CI- Magnet Bpl:9799424 TIC:851583808 Fil,, Text,Srnith(Adcock) "F48N2" s200, R0-650@2, afse-6 100,, 14V 1y6 :_14 :.13 I :_12 80_ I :.11 :_11 ;o· , 10 6S, '- 9 60_, :..9 55 : ,_e r 50' 1-7 4S_: r.6 40_; ~r 6 35 ' 30_: 148 25_'. 20= ~: f:-3 15: 1404 f2 ': "',,. Ii Ip"' 1 o_----·-r----4------,-~--,....,.._,...... _...... llu!~_i.--.--,--.-. --.,--,----.---~~--.--~: 800 1000 1200 1400 1600 1800 2000 rn/ z Figure B-15 C60F48 with N 2 and H2 at 150°C
81 File1D409A
File1D408A File File,D407A File ZAB-SEQ4F ZAB-SEQ4P Fila ZA8-SBQ4F
10~
100'!1 10,
20J
40 20.J
80
40~
20
80
60. 60 60
401
0oj
o1,
0
oJ
:
-'---~~,
j 1 ➔ j
Text1KSmith(Compton)"Fa" Text,KSmith(Compton)"Zn"
Text1KSmith(ComptoD)"Mg"
• ' • •
6~~•g
CI-
CI-
CI-
Ident,21 Ident1l9
Ident,22
1
,
1cfo_.,--,-.,....,.ii6
100
700
716.9
I
Magnet
Magnet
Magnet
740.0
I
1
75~;0
I
• •
777.9
I
Acq,21-JUL-1997 Acq,21-JUL-1997
Acq12l-JUL-1997
797
I,
8pM11480
0
BpM,1632
8pM11632
sbo
eoo
l
.0
o.
•
-,----~~-r--r--r--·~
BpI138912000 BpI,12156928
s200,
s200,
a200, BpI:24387584
960
900 900
-
•.
···-----·~--,
18,07,20
18118137 18127134 r0-650@2, r0-650@2, r0-650@2,
~00
1000
10'00
TIC,364675360 TIC:846653312 TIC:159.02448
-,---·
+2,19 +2101
+2:15
AFSe-7 AFSe-7 AFSe-7
1070.0
..
·-··-
11'00
1i'oo ll"oo
Cal,D408A Cal:D408A
Cal:D408A
-
~-...,...~~-----~--~---
:··--.
......
~•;~.:
Flags:NORM
Flags,NORM Flags:NORM
"!""
t··-·--
1z'oo 1z'oo
1300
u·ao
uoo
...
,-.,-~
-.'..'~,JJ
'~'.uuJJ1 1365.
.Ll
.
1403.9,
1400
uoo 14·00
9
1441.9
1441.9'
14
3.0
Jilltu,,.,
1479.9
147J.9
1500 1500 1500
1517.9
15~8.9:
15
li
I
8.9
15
15r.9LJ
.
·
1
Ll
_
6.9
lS9i3.9
1593.9
J1
16'00 1600
:
I i
16
1631. 1631.
l
.
3.1.
l
---.,..~,.,I,'
~,--T,
..
9
9
9
~,•-
1700
70 1800 1700
..
..
,
···-•,-
·,--
1800
··r:
·r
r
[o
. f:
~1128
t:
'."
2255
3383 4511
o
5638mV
18V 14
11
m/z
.,.
m/Z
•
~ '° u
,....-1 i3
•1"""'4
~
~
~ '"O
N ~
~
,+-,I
01)
l"""'4 ::s
(1.)
M
~ e
Cd (1.) 01) r=:
en (1.)
§
I
I.O
0
00
-.::t"
"'
"'
00
N
m/z m/z
58V 58V
6 6
9 9
1S 1S
52 52
17 17 32 32
35 35
55 55
20 20
23 23
49 49
29 29
12 12
38 38 41 41
44 44
_46 _46
~ ~
~ ~
~ ~
......
~ ~
~3 ~3
t.26 t.26
t t
f-o f-o
2200 2200
00 00
1
21
Cal1D434A Cal1D434A
+5153 +5153
2000 2000
;T, ;T,
13104102 13104102
tr tr
-, -,
1700 1700 1800 1900
Acq,10-0CT-1997 Acq,10-0CT-1997
632.0 632.0
1600 1600
593.9 593.9
155,5.9 155,5.9
. .
1517. 1517.
JWll.Lt~-,-,-r,· JWll.Lt~-,-,-r,·
1479. 1479.
j j
9 9
! !
1403. 1403.
SPEC(Areao,Cantroid) SPEC(Areao,Cantroid)
1366. 1366.
a-7 a-7
•··•1-,--•).J~ •··•1-,--•).J~
1NORM 1NORM
■
APS APS
r·,·-. r·,·-.
Plag
1200 1200 1300 1400 1500
·' ·'
·r ·r
,---
f f
[email protected], [email protected],
-r-
rT" rT"
SlS0, SlS0,
TIC,104026520 TIC,104026520
1000 1000 1100
-1-:-
PKD(S,3,5,0.01%,0.0,0.00%,P,P) PKD(S,3,5,0.01%,0.0,0.00%,P,P)
T-i T-i
"SB" "SB"
900 900
BpI11191700 BpI11191700
SMO(l,5) SMO(l,5)
(Adcock) (Adcock)
Ma1111et Ma1111et
S111ith S111ith
Ident,19 Ident,19
720.0 720.0
CI-
111:. 111:.
Text Text
6 6
5 5
0 0
JS JS
10 10
15. 15.
50 50
55~ 55~
60~ 60~
65_ 65_
30 30
40 40
45 45
85 85
80 80
20 20
25 25
70 70
75 75
90 90
95 95
100'16 100'16
ZAB-SEQ4P ZAB-SEQ4P
Pile1D43,A Pile1D43,A
Pile Pile
I I
+ +
~ ~
""'"'. ""'"'.
~ ~
to to
~ ~
(I) (I)
~ ~
.....J .....J
""'"'. ""'"'.
~ ~ =--
(J (J (I) (I)
0 0
0 0
= =
; ;
(Jq (Jq
0 0
°' °'
00 00
~ ~
.j::,. .j::,.
00 00 w w File File10428A ZAB-SEQ4P
10,
70
eo 90j
20_, 60. 10_~
50 40] 30
0
~
1
j
'
Text1Smith(Adcook)
700
CI-
Ident13_6
720
! I I
I I
Magnet
.
800
Acq116-SEP-1997
BpI,9297920
"BB3"
900
s220,r0,APSe-7
1000
TIC,747984640
' i
•
12157143
11
1
00
'
+1154
1200
'i.
Cal10424A
1300
,
1
1~,.e~~....
40 1500 1400
,.,,J
1480
~i,UlJ,!.~C.
I
15.56
I
1594
1600
!
'
70 1800 1700
,
1766,
,
,
.1~0s
1900
20·00
2100
7
12 ll 8
3 ' UV
5 1 0
m/z
-~
00 ,-.-4 u ~ ~
~ ~
+
s
(l.) o.O (l.)
0 ~ ~
I
0
00 IO -.:t"
00
-.:t" Pile,D436B Ident,8_30 SKO(l,5) PKD(5,3,5,0.01%,0.0,0.00%,P,P) SPBC(Areas,Centroid) Acq:10-0CT-1997 15:59127 +9,19 Cal,D434A ZAB-SEQ4P CI- Magnet BpI,31959296 TIC:7530713600 Plags,NORM Pile Text,K.Smith (Adcock) "Pe(CO)5• S200, r0-650~2, APS e-7 10 148:°"0 ..619V ! 95 598
90 557
518.0 526 ~ 851 ·495 OQ""'"'. 80 E; 75 464 (T) 433
a, 402 I ~ .371 \0 1442. 340
309 00 VI n O'I 279 0 .i:,.~ 40 248 j I 00 1404. ,_217 i I I + JOj I 196 ~ I (T) 25' ' ,,-.._ 155 n 20 124 15 93 '-._/0 VI 10
31
10·00 1100 1200 llOO 16'00 17'00 1e·oo 19·00 2000 2100 m, Appendix C Norbomyl Cation Spectra
86 H3, "5• t17
Hl, Hz, H6 _A______.HA..____jL_ a a . I :::i:: 0 L-: ::,: N. ::,: en
0
... "'
... 0 .,.:c
·o ~
a ~ . a ...... N 9 0 .. "' .. "'
n ~ -<
Figure C-1 COSY of the 2-Norbomyl Cation at -80°C
87 l
:,: N
---
-1 j i , I j < I i ------· ------J
r~..,.....-·.,. -,--~--·.,.·---,-~---.-,--.---,---T·-.---,--,···•--'"--,--~~-,-, ·-r·-, ..., 7 ~ ~ g ~ g ~ ~ ~ ~ :,: 3 ,CJ n
__,
Figure C-2 HMQC of the 2-Norbomyl Cation at -80°C
88 ------····--·------··------,
Iii
rtl"l"lTflTTTTTTTlTTll tTnTrtn rt 1n1 nn 11 ,11111,,, t rnrnrnrn--rTTJTTn-rnnrnn nTnp11 rnn1Tn n nTH 1n nn n,,, 1 u I rn' I' ppm 10 8 6 4 2
Figure C-3 2D EXSY of the 2-Norbomyl Cation at 193K
89 t,J 67
4
G
8
10
12 ppm
I 11111111, rnrnTTnTTTTTip"lTTTTTT1TTT1'TTTrTTJT• I I I' I I I ii I I I I I I I.,,,, I I I I I I ,TTTTITT1TJTTTTTTTTTTTTTT1TTTTflTTI11TI"lTTJT111 '' 'I 1 ppm 10 8 6 4 2
Figure C-4 2D EXSY of the 2-Norbomyl Cation at 208K
90 0
0
r-nT11rr·TTTITTT1i,,,, • r' •'I• I ITmTTTTnnnu11 ,n1T11 I' ,1 r1 * '''' J•Tm·11, t' ''•I' • •np·m--rrnTTTTT11, 111 ~• 1111' ,·n11n11111 'l' ppm 10 8 6 4
Figure C-5 2D EXSY of the 2-Norbomyl Cation at 223K
91 UI . ,.H ■ 0 C -----· --- , .• ,1 .... UI .... 0
.w UI
w 0
.w UI
w 0
,.. . -· UI
,.. . - 0
Figure C-6 1H of the 2-Chloro-2-Cation at -90°C
92 ..., 0 0
284.52'
N 0 0
U1'"" 0
'""0 0
er
76.929
Q. 61.015
U1 0 " - ,2.Ul
- --- ..----·- 2Ll88
Figure C-7 13 C of the 2-Chloro-2-Cation at -90°C
93 ex, ...:i 0\ U1 .... w 0 0 0 0 0 0 . ' . ~-·-'·---'- _:._. _. .!..._..:. .. ; _ _i_J __ ~ ____ 1 _; .:_.l .. .!. ••• ~ •. ~ _ _J_ - ..•.• .! ... 1 ...•.. -.-~.:.. -· : ..•.. 1 _,.._; · ..... _, ___ i ,y. ~., . ....:i r::r ! .U1 0 . U1
.... 0 ...,
w - 0
U1
w 0 tlj tJ ...... 0 'O tJ 'O \1 ...... a U1 ,~, CJQ
t,.)
0
.I-' U1
.I-' 0
.0 U1
Figure C-8 HMQC of the 2-Chloro-2-Cation at -90°C
94 F2 - g (ppm); !
8 2.0 f G
2.5
3.0
C
,, .. (.; 3 .5 ·, ~G,; ~ dand f
4.0
4. 5 · b p 5.0
5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5
Fl {ppm)
Figure C-9 COSY of the 2-Chloro-2-Cation at -90°C
95 Fl (ppm) g
2.0
2.5
3.0
3.5
,.o
4.5
5.0
5.0 4.5 ,.o 3.5 3.0 2.5 2.0 1.5 F2 (ppm)
Figure C-10 NOESY of the 2-Chloro-2-Cation at -90°C
96 --5.515
u, -
--2.708
--2.UI
97 -- 0 IO
ttt"Ht ------· ------a
Figure C-11 13C of the 2-Chloro-2-Cation warmed to -20°C
98 VITA
Kenneth Smith was born in Hanover, Pennsylvania on October 41\ 1972. He graduated from Delone Catholic High School in 1990. He received a B. S. in chemistry from
Shippensburg University in 1994. He entered the graduate school at the University of
Tennessee in 1994.
His interests in the field of chemistry include noble gas compounds, the chemistry of strong oxidizers, and stable carbocations, as well as pharmaceutical chemistry and biochemistry, specifically the production and uses of plant secondary products.
His interests outside of chemistry are widely varied including astronomy, computers, tropical fish. His hobbies include computer gaming, as well as drawing and writing.
99 em 5979J1 31r.x1 1113 lil3/B5/03 f