Page 1 of 84
`
`Mylan Ex. 1018
`Mylan v. Novartis
`IPR2015-00265
`
`

`
`
`
`Kenyon8cKenyon LL?
`!.’\!'TlZl.l‘.ECTTUr\l. PKOPIERTY LA“?
`
`LIBRAR¥
`
`Advanced organic chemistry
`
`Carey & Snndberg - 3d
`ed., 1990
`A
`
`CHEM
`
`QD
`
`25 1 . 3
`. C37
`1 9 9 0
`
`$
`
`4+........»:m».w,‘¢._gg§$_Mw‘».:.m:».u».<.~...-.=-TW.~.+‘y,,,_,.qA_Wm,,,_WNW‘MAA
`
`
`
`
`a—r«o~a.««»av;««A-saw»-.«Ap:«,.»«~.......mr..w...-nu‘W.5-5.:-,\.u-rm.amm~mwwn“ww m‘x»“uxwxuM*-'r2e‘-!1u,-——»w-
`
`Page 2 of 84
`
`Page 2 of 84
`
`

`
`
`
`94:?
`
`fiat?
`
`%g $; [?‘%E_ 1
`
`‘£9 .
`
`Advanced Organic
`Chemlstry
`
`Part A Structure and Mechamsms
`
`\_\§L'1\1Y3W
`
`7
`m3;nu3t“‘°*“’°“
`mg 1‘ 3 ‘?§.'s\‘§
`HECEWEB
`
`Page 3 of 84
`
`Page 3 of 84
`
`

`
`Advanced Organic Chemistry
`
`PART A: Structure and Mechanisms
`PART B: Reactions and Synthesis
`
`Page 4 of 84
`
`Page 4 of 84
`
`

`
`Advanced Organic
`Chemistry
`EDITION
`
`THIRD
`
`Part A: Structure and Mechanisms
`
`FRANCIS A. CAREY
`and RICHARD J. SUNDBERG
`
`L'?:i‘),€K$‘l7jl of Vizginia
`Charfottesville, Virginia
`
`PLENUM PRESS 0 NEW YORK
`
`LONDON
`
`Page 5 of 84
`
`Page 5 of 84
`
`

`
`
`
`V
`
`,,
`
`Library of Congress Cataloging in Puhiication Data
`
`(Revised For 3rd ed.)
`Carcy. Francis A., 1937-
`Advanced organic chemistry.
`lnciudes bibliographical references.
`Contents: pr. A. Structure and mechanisms—pt. B. Rea:-tiens and synthesis.
`1. Chemistry, Organic. I. Sundberg, Richard J.. 3933-
`. E. '1'311:.
`QD25l.2.C36 £990
`547
`ISBN 0-306-43440-7 (Part A)
`ISBN (1-306—43447—4 (pbk: Part A)
`ISBN 0-306-434566 (Part 3)
`ISBN 0-306-43457-1 (pbk.; Part B}
`
`90-6851
`
`109876
`
`@1996. £983, 1977 Plenum Press, New York
`A Diyision of Plenum Publishing Corporation
`233 Spring Street, New York, N.Y. 300:3
`
`All rights reserved
`
`No part of this book may be reproduced, stored in a retrievai system, or transmitted
`in my form 61' by any mcans, electronic. mechanical, photocopying. microfilming.
`recording, or otherwise. withouz written pcrmissiarx from the Publisher
`Printed in Hz: United Stares of America
`
`Page 6 of 84
`
`Page 6 of 84
`
`

`
`12
`
`Free-Radical Reactions
`
`12.}. Generation and Characterization of Free Radicals
`
`12.1.1. Background
`
`A free-radical reaction is a chemical process in which molecules having unpaired
`electrons are involved. The radical species could be a starting compound or a
`product, but in organic chemistry the most common cases are reactions that involve
`radicals as intermediates. Most of the reactions discussed to this point have been
`heterolytic processes involving polar intermediates and/or transition states in which
`all electrons remained paired throughout the course of the reaction. In radical
`reactions, homolyxic bond cleavages occur. The generalized reactions shown below
`illustrate the formation of alkyl, vinyl, and aryl free radicals by hypotheticai
`homolytic processes.
`
`Y» + R,c—:‘-x —» n,c- + x—v
`
`if
`R3C=C'¥< ~—-o nzc_—.c:/
`' x
`
`R
`
`4. x.
`
`The idea that substituted carbon atoms with seven valence electrons could be
`
`involved in organic reactions took firm hold in the l930s.'1'wo experimental studies
`have special historical significance in the development of the concept of free-radical
`reactions. The work of Gomberg around 1900 provided evidence that when triphenyh
`651
`
`Page 7 of 84
`
`Page 7 of 84
`
`

`
`
`
`methyl chloride was treated with silver metal, the resulting solution contained P1-LSC.
`in equilibrium with a less reactive molecule. It was originally thought that the mate
`stable molecule was hexaplrenyietltane, but eventually this was shown not to be SQ
`The dimcric product is actually 3 cyclohexadiene derivative.‘
`
`u=nnr:)<:3___< # zphsg
`
`H I
`
`,
`
`Flt
`
`Ph
`
`The dissociation constant is small, only about 2 x 10"‘ M at room temperature. T113
`presence of the small amount of the radical at equilibrium was deduced from
`observation of reactions that could not be reasonably attributed to a normal hydro-
`carbon.
`
`The second set of experiments was carried out in 1929 by Paneth. The decompo-
`sition of tetramethyllead was carried out in such a way that the decomposition
`products were carried by a flow of inert gas over a film of lead metal. The lead was
`observed to disappear, with re-formation of tetramethyilead. The conclusion was
`reached that methyl radicals must exist long enough in the gas phase to be transported
`from the point of decomposition to the lead film.
`
`Pbicflalm 151:5 Pbm ‘I’ 4CH3':g>
`
`4CH3‘u> ‘I’ Pbm
`
`Pb(CH3)€e.;
`
`Since these early experiments, a great deal of additional information about the
`existence and properties of free-radical intermediates has been developed. In this
`chapter, we will discuss the structure of free radicals and some of the special
`properties associated with free radicals. We will also discuss some ofthe key chemical
`reactions in which free-radical intermediates are involved.
`
`12.1.2. Stable and Persistent Free Radicals
`
`Most organic free radicals have very short lifetimes, but various structural
`features enhance stability. Radicals without special stabilization rapidly dimerize
`or disproportionate. The usual dispreportionation process involves transfer of a
`hydrogen from the carbon [3 to the radical site, leading to formation of an alkane
`and an alkene.
`
`Dimerizerion
`
`Disproportiomttian
`
`I
`-—(l2-—
`
`I
`C-
`i
`
`i
`_g
`E
`
`-
`
`2
`
`-—>
`I
`I
`2 —$—(E» —-
`H
`
`l
`l
`._—?——$»-
`H H
`
`/
`\
`-1- ‘/C=C\
`
`1. H. Lankamp, W. Th. Name. and C. Mactean, Tetrahedron Lesa, 2‘-$9 E1968}; 3. M. McBride.
`Tetrahedron 36, 2069 (I974); K. J. Skinner, H. S. Hochcster, and I. M. McBride, J’. Am. Chem. Soc.
`96, 4301 (191%).
`
`Page 8 of 84
`
`Page 8 of 84
`
`

`
`"652
`
`CH APTI-SR 12
`FREE-RADlC/‘\L
`REACTIONS
`
`methyl chloride was treated with silver metal, the resulting solution contained Ph3C_
`in equilibrium with a less reactive molecule. It was originally thought that the more
`stable molecule was liexaaplienylethane, but eventually this was shown not to be so_
`The dimeréc product is actually a cyclohexadicne derivative.‘
`
`
`
`7
`
`‘V
`
`79??
`
`Ph
`
`“MC 7:: 2 Ph3C-’
`
`H
`
`«
`
`The dissociation constant is small, only about 2 X 10" M at room temperature. T113
`presence of the small amount of the radical at equilibrium was deduced from
`observation of reactions that could not be reasonably attributed to a normal hydro-
`carbon.
`
`The second set ofexperiments was carried out in 1929 by Paneth. The decompo-
`sition of tetrarneehyllead was carried out in such a way that the decomposition
`products were carried by a flow of inert gas over a film of lead metal. The lead was
`observed to disappear, with re-formation of tetrametlzyllead. The conclusion was
`reached that methyl radicals must exist long enough in the gas phase to be transported
`from the point of decomposition to the lead film.
`450°C
`Pb(CH334m """ Pbm "' 4CH3'ss)
`
`4C1-(3-.,. + ram “L”-E ps(cH,).,,,,
`
`Since these early experiments, a great deal of additional information about the
`
`existence and properties of free-radical intermediates has been developed. In this
`chapter, we will discuss the structure of free radicals and some of the special
`properties associated with free radicals. We will also discuss some of the key chemical
`reactions in which free-radical intermediates are involved.
`
`12.1.2. Stable and Persistent Free Radicals
`
`Most organic free radicals have very short lifetimes, but various structural
`features enhance stability. Radicals without special stabilization rapidly dimerize
`or disproportionate. The usual disproportionation process involves transfer of a
`hydrogen from the carbon 5 to the radical site, leading to formation of an alkane
`and an alicene.
`
`l
`i
`2 ——(."I* -+ —Cl3— -
`
`Dimerizarlbn
`
`Dispmpnrfionefion
`
`1. H. Lankamp, W. Th. Nauia, and C. Maclxan, Telrahedmn Lem, 249 (1968); J. M. Mcaridc.
`‘Tetrahedron 30, 2009 (3.914); K. J. Skinner, H. S. Hochcsler, and J. M. McBride, I Am. Chem. Soc-
`96, 4301 (I974).
`
`Page 9 of 84
`
`Page 9 of 84
`
`

`
`..my
`
`553
`SECTION 12.3,
`GENERATION AND
`CHARAC'I'ERIZATl0N
`OF FREE RADICALS
`
`
`
`
`
`uu»mmu,_;4¢;;».z,,....~..,_...»«~st;.1;,_...,
`
`
`
`E1
`
`..
`l
`~.~:
`
`. -
`
`1’
`‘J
`
`Page 10 of 84
`
`Radicals also rapidly abstract hydrogen or other atoms from many types of solvents,
`and most radicals are highly reactive toward oxygen.
`
`fiydrogen atom abstraction
`
`Addition to oxygen
`
`I
`.....C. + }—[—Y —s
`]
`
`E
`...C...[-{ + y.
`3
`
`l
`—<i:-+0.
`
`E
`—» —<I:—o-—o
`
`A few free radicals are indefinitely stable. Entries 1, 4, and 6 in Scheme 12.1
`are examples. These molecules are just as stable under ordinary conditions of
`temperature and atmosphere as typical closed—shell molecules. Entry 2 is somewhat
`less stable to oxygen, although it can exist indefinitely in the absence of oxygen.
`The strecturcs shown in entries 1, 2, and élfll permit extensive delocalization of
`the unpaired electron into aromatic rings. These highly delocalized radicals show
`no tendency toward dimcrization or disproportionation. Radicals that have long
`lifetimes and are resistant to dimerization or other routes for bimolecular self-
`annihilation are called stable free radicals. The term inert free radical has been
`
`suggested for species such as entry 4, which is unreactive under ordinary conditions
`and is thermally stable even at 300°C.’
`a significant
`Entry 3 in Scheme 12.1 has only allcyl substituents and yet
`lifetimes in the absence of oxygen. The tris{ t-butyllmethyl radical has an even longer
`lifetime, with a half-life of about 20 min at 25°C? The steric hindrance provided by
`the t—butyl substituents greatly retards the rates of dimerization and disproportion-
`ation of‘ these radicals. They remain highly reactive toward oxygen, however. The
`term persistent radicals is used to describe these species, since their extended lifetimes
`have more to do with kinetic factors than with inherent stability.“ Entry 5 is a
`sterically hindered perfluorinated radical, which is even more stable than similar
`alkyl radicals.
`There are only a few functional groups that contain an unpaired electron and
`yet are stable in a Wide variety of structural environraems. The best example is the
`nit;-oxide group, and thereare numerous specific nitroxide radicals which have been
`prepared and characterized.
`
`R\
`
`R
`
`R\
`
`R
`
`Many of these compounds are very stable under normal conditions, and hetcrolytic
`reactions can be carried out on other functional groups in the molecule without
`destroying the nitroxiée group.‘
`
`2. M. Ballestcr, Am Chem. Res. 18. 33{1(l985).
`-
`DJ
`G. D. Mendenhall, D. Griller, 5. Lindsay, T. T. Tidwell, and K. U. Ingold, J’. Am. Chem. Soc. 96,
`2441 (19?4).
`4. For a review of various types of persistent radicals, see D. Griller and K. U. lngold, Acc. Chem.
`Res. 9, 13 (l9‘76).
`S. For reviews of the‘ preparation, reactions, and uses of nitroxide radicals, see 3. F. W. Keane, Chem.
`Rev. 78, 3? (1978); L. J. Berliner (ed), Spin-Labelling, Vol. 2, Academic Press, New York, 1979.
`
`
`
` ll
`
`Page 10 of 84
`
`

`
`554
`
`CHAFTER I2
`FREE-RADICAL
`REACTIONS
`
`Structure
`‘
`
`1°
`
`
`
`Scheme 12.1. Stahility of Some Free Radicals
`
`Conditions for stability
`'
`'
`
`Indefirzitely stable as a solid, evcn in thc presence of
`air.
`
`2"
`
`Pt‘:
`
`P3:
`
`P5
`
`93"
`
`Ph
`
`3‘
`
`(CH3'}3C
`
`H
`
`\é_C(CH3)3
`
`Crystalline substance is not rapidly attacked by
`oxygen, although solutions are: air sensilive; the
`compound is stable to high temperature in the
`absence of oxygen.
`
`Stable in dilute solution {<l9"5 M) below --30"C in
`the abscace of oxygen, cm of 50 sec at 25°C.
`
`Stable in solution {or days, aver: in the prescncc of
`air. Indcfinitcly stable in solid state. Thermally
`stable up to 300°C.
`
`Stab]: to ofiygcn; themzaliy stable to ?0°C.
`
`Stable to oxygen; stable to extended storage as a
`sofid. Slowly decomposes in solutian.
`
`/
`
`Stable to oxygen ave:-1 above 1eD“C.
`
`{CHa)aC
`
`a. C. F. Koelsch. J. Am. Chem. Soc. 751, 4439:1957).
`h. K. Ziegler‘ and B. Schrsell, Juxlns licéigs Ann. Sitem. 445, 266 (l92S)_
`c. G. D. Mendcnhall, D. (killer, D. Lindsay, T. ‘F. Tidwcfl. and K. UE Ingafd, J. Am. Chem. Soc. 96, 244] (1974).
`d. M. Ballester, J. Riera. J. lI3::s:m'vcr, C. Badéa, and J. M. Mo.-zsfi, J. Am. Chem. Soc 93. 2215 (1971).
`6. IL Y. Schcrer. Jr., T. Ono. K. Yamanoucifi, R. Fernandez, and P.‘H:nderson, J. Am. Chem. Soc. 107, 718 ( 1985}.
`f. G. M» Cappingcy. J. Am. Chem Soc. 79, 513! §£957); P. D. Barzlcat andT. Funafiashi, J. Am. Chem. Soc. 84, 2596 (1962).
`g. A, K. Hofimann and A. T, Hemiersxm, I. Am. Chem. Soc. 83, 4671 (196).
`
`KI.
`‘1
`
`Page 11 of 84
`
`Page 11 of 84
`
`

`
`Although the existence of the stable and persistent free radicals we have
`discussed is of significance in establishing that free radicals can have extended
`lifetimes, most free—radical reactions involve highly reactive intermediates that have
`relatively fleeting lifetimes and can only he studied at very low concentrations. The
`techniques for study of radicals under these conditions are the subject of the next
`section.
`
`655
`
`SECTION 125;,
`GE-NERATl0N mun
`cunnacremzpmon
`OF FREE RADICALS
`
`12.1.3. Direct Detection of Radical Intermediates
`
`The distinguishing characteristic of free radicals is the presence of an unpaired
`electron. Species with an unpaired eiectron are said to be paramagnetic. The most
`useful method for detecting and characterizing uustabie radical intermediates is
`electron paramagnetic resonance (EPR) spectroscopy. Electron spin resonance (ESR}
`spectroscopy is synonymous. This method of spectroscopy detects the transition of
`an electron between the energy levels associated with the two possible orientations
`of electron spin in a magnetic field. An EPR spectrometer records the absorption
`of energy when an electron is excited from the lower to the higher state. The energy
`separation is very smail on an absolute scale and corresponds to the energy of
`microwaves. EPR spectroscopy is a highly specific tool for detecting radical species
`since only molecules with unpaired electrons give rise to EPR spectra. As with other
`spectroscopic methods, detailed analysis of the absorption spectrum can give rise
`to structural information. One: feature that is determined is the g value, which
`
`specifies the separation of the two spin states as a function of the magnetic field
`strength of the spectrometer.
`
`ho: E =g;.oEH
`
`where #3 is a constant, the Bohr magnetron (=9.274 x 18"“ erg;’(}), and H is the
`magnetic field in gauss. The measured value of g is a characteristic of the particular
`type of radical, just as the line positions in IR and NMR spectra are characteristic
`of the absorbing species.
`A second type of structural information can be deduced from the hyperfine
`Splitting in EPR spectra. The origin of this line splitting is closely related to the
`factors that cause spin—spin splitting in proton NMR spectra. Certain nuclei have
`a magnetic moment. Those which are of particular interest in organic chemistry
`include ‘H, “C, “N, “F, and "P. Interaction of the unpaired electron with one or
`more of these nuclei splits the signal arising from the electron. The number of lines
`is given by the equation
`
`number of iincs = Zn: + 1
`
`where I is the nuclear spin quantum number, and n is the number of equivalent
`interacting nuclei. For ‘H, “C, "F and “P, I =
`Thus, a single hydrogen splits a
`signal into a doublet. Interaction with three equivalent hydrogens, as in a methyl
`group, gives rise to splitting that produces four lines. This splitting is illustrated in
`Fig. 12.] . Nitrogen (“‘N), with I = 1, splits each energy level into three lines. Neither
`
`Page 12 of 84
`
`Page 12 of 84
`
`

`
` 656
`REACTIONS
`
`A.‘C‘a..j;«Lufliwlm 4$|%>MEMA~J€f‘>-ea»;.A
`
`C E-lA¥"l‘ER £2
`FREE-RA.D?.C.AL
`
`No interacting
`hydrogen nucleus-
`one absorption
`line
`
`One interacting
`hydrogen nucleus—
`two absorption
`lines
`
`
`
`Two eq uivalent
`interacting
`hydrogen nur:lei—
`three absorption
`lines
`
`(a)
`
`[Prong
`lb) st;
`
`cherf
`
`radlg
`
`5EgrlwfM£.lvL*rr1‘‘*l~“M%“ §5-9“
`Qnor—C!owl~<.»er:»~::.s~.~;:«g;/-w=.-:;~x~=:zn-tu-..~
`
`
`(M
`
`
`
`Page 13 of 84
`
`Fig. 12}. Hyperfine splitting in EPR spectra.
`
`‘EC not ‘50 has a nuclear magnetic moment, and just as they cause no splitting in
`NMR spectra, they have no effect on the multiplicity in EPR spectra.
`
`A great deal of structural information can be obtained by analysis of the
`hyperfine splitting pattern of a free radical. If we Iimit our discussion for the moment
`to radicals without heteroatoms, the number of lines indicates the number of
`
`interacting hydrogens, and the magnitude of the splitting, given by the hyperfine
`splitting constant as, is a measure of the unpaired eiectrort density in the hydrogen
`h is orbital. For planar systems in which the unpaired electron resides in a rr-orbital
`system, the relationship between electron spin density and the splitting constant is
`given by the McConnell equation‘:
`
`a=:2Q
`
`where a is the hyperfine coupling constant for a proton, Q is a proportionality
`constant (about 23 G), and p is the spin density on the carbon to which the hydrogen
`is attached. For example, taking Q = 23.0 G, the hyperfine splitting in the benzene
`radical anion may be readily calculated by taking p = é, since the one unpaired
`electron. must be distributed equally among the six carbon atoms. The calculated
`
`value of a = 3.83 is in good agreement with the observed value. The spectrum (Fig.
`12.2w) consists of seven lines separated by a coupling constant of 3.75 G.
`The EPR spectrum of the ethyl radical presented in Fig. 12.2b is readily
`interpreted, and the results are of interest with respect to the distribution of unpaired
`electron density in the molecule. The l2~line spectrum is 3. triplet of quartets resulting
`from unequal coupling of the electron spin to the ac and [3 protons. The two coupling
`constants are a,,, = 22.38 G and a5 = 26.87 G and imply extensive deloealization of
`“spin density through the 0' bonds.
`EPR. spectra have been widely used in the study of reactions to detect free-radical
`interrnediates. An interesting example involves the cyelopropylmethyl radical. Much
`
`6. H. M. McConnell, J. Chem. Phys. 24, 764 (1956).
`
`Q
`
`Page 13 of 84
`
`

`
`
`
`(all
`
`
`
`26.9Ci
`
`32.4 G
`
`657
`
`secrzow 12,1,
`GENERATION AND
`CHARACTERIZATION
`OF FREE RADICALS
`
`
`
`Fig. 122. Some EPR spectra of small organic free radicals: (3) Spectrum of the benzene radical anion.
`[From 1. R. Bolton, Mal‘. Phys. 6, 219 {I963}. Reproduced by permission of Taylor and Francis, Lid.)
`(b) Speeirum of the ethyl ‘radical. fFrom R. W‘. Fcssenden and R. H. Schulcr, J. Cissm. Phys. 33, 935
`(1960); 39, 2147 (1963); Reproduced by permission of the American Instituae of Physics.)
`
`chemical experience has indicated that this radical is unstable, givfing rise to 3—butenyl
`radical rapidly after being generated.
`
`>U~CH; ———»
`
`Cfizx
`|
`CH2
`
`cHmCH2
`
`The radical was generated by photolytic decomposition of di—t-but)-1 peroxide in
`methylcyclopropane, a process that leads to selective abstraction of a methyl hydro»
`gen from mcthyicyclopropane.
`
`(CH3}3COOC(CH3)3 J1» 2(ClI;}3C~—O-
`
`zCH31;C—<3’ + [>._cu, -—» [>~CI-E; + [CHfl_;€.‘0H
`
`Below —l‘40°C, the EPR spectrum observed was that of the cyclopropylmethyl
`radical. If the photolysis was done above 440°C, however, the spectrum of a second
`species was seen, and above —1€)0“C,
`this was the only spectrum observed. This
`spectrum could be shown to he that of the 3—butenyI radficalf This study also
`established that the 3-butenyl radical did not revert to the cyclopzopylmethyl radical
`on being cooled back to —140"C. The conclusion is that the ring opening of the
`
`‘J. J. K. Kochi, X’. J. Krustie, and D. R. Eaton. J.
`
`Cfzem. Soc 91, 187? (1969).
`
`
`
`
`
`wan»,Vjmgnm‘Tog.-A,-,r.,_.~»nu;;;,«x;x,.
`
`
`
`
`
`r-—3:.-,,W7.-.;.<¢A.,5...\.\._
`
`Page 14 of 84
`
`Page 14 of 84
`
`

`
` _
`
`658
`
`CHAPTER 12
`FREE-RADICAL.
`REACTIONS
`
`cyclopropyl radical is a very facile process, so that the lifetime of the cyelopropyl
`radical above —I€!0"C is very short. The reversal of the ring opening can be detected
`by isotopic labeling experiments, which reveal
`the occurrence of deuterium
`migration:
`
`H CH2‘
`
`CH2*“““CHCH2CD'z‘ 1 HzC— D; %'“ CH2-“~'CHC‘i}zCH2I
`
`The rates of both the ring opening (Ic : 2 X 1{}‘s"‘ at 25°C) and the ring closure
`(k = 3 x 103 s‘“‘) have been measured and show that only a very small amount of
`the cyclopropylmethyl radical is present at equilibrium, in agreement with the EPR
`results.g
`
`It is important to‘ emphasize that direct studies such as those carried out on
`the cyclopropylmethyl radical can be done with low steady-state concentrations of
`the radical. In the case of the study of the cyclopropylmethyl radical, removal of
`the source of irradiation would lead to rapid disappearance of the EPR spectrum,
`because the radicals would react rapidly and not be replaced by continuing radical
`formation. Under many conditions, the steady—state concentration of a radical
`intermediate may be too low to permit direct detection. Failure to observe an EPR
`signal therefore cannot be taken. as conclusive evidence against a radical inter-
`mediate.
`
`A technique called spin trapping can sometimes be used to study radicals in
`this circumstance. A diamagnetic molecule that has the property of reacting rapidly
`with radicals to give a stable paramagnetic species is introduced into the reaction
`system being studied. As radical intermediates are generated, they are trapped by
`the reactive molecule to give more stable, detectabie radicals. The most useful spin
`traps are nitrosc compounds. They rapidly reactpwith radicals to give stable nitroxide
`radicals.’ Analysis of the EPR spectrum of the nitroxidc radical product can often
`provide information about the structure of the original radical.
`R
`
`12- + R.'N=O ——»
`
`:;si—t§
`12*’ ”
`
`Another technique that is highly specific for radical processes is known as
`CIDNP, an abbreviation for cbemicaliy induced dynamic nucIear_poIarization.“' The
`instrumentation required for such studied is a normal NMR spectrometer. CIDNP
`is observed as a strong ‘perturbation of the intensity of NMR signals for products
`formed in certain types of free-radical reactions. The variation in intensity results
`
`8. A. Eifiio, D. Gtiller, K. U. lngold, A. I... J. Bcckwith, and A. K. Serclis, J. Am. Chem, 50:. m2_ 1734
`(1930); L. Mathew and J. Warkeruin, J. Am. Chem. Soc. 108, 7981 (1936).
`9. E. G. Janzen, /lcc Chem’. Res. 4, 31 (1937:); E. G. Janzen, in Free Radical: in Biology, W‘. A. Pryor
`(ea), Vol. 4, Academic Press, New York, 198%), pp. 115-154.
`10. H. R. Ward, Acc. Chem. Res. 5, 18 (1972); R. G. Lawier, Acc. Chem. Res, 5, 25 (I972).
`
`Page 15 of 84
`
`
`
`1'.1M‘W\L'imw§é¢§?i§Mlli4lMi@u:F.’NMxi.‘:::;$:.'(‘A.L..,..
`
` n»
`
`S:
`3;*2fewfi
`
`F 4
`
`
`
`
`
`
`
`.4.New-«.:......_:;,.:.:;.~
`
`Page 15 of 84
`
`

`
`
`
`659
`
`.SECTlON 12.1.
`GENERATION AND
`CHIuRACTERlZATl0N
`OF FREE RADICALS
`
`.4
`
`4 .LA
`
`-T
`
`
`
`“%;7s=ssJt«_.lvwyr3—4#;r.,--so-f\«-j
`
`
`‘~:~‘*3-i.“=:S$s.'s-ei;—.<7~?“
`
` .,‘.,,_.......,.i.,,,._c,.e,...,_,
`
`LLI
`
`l
`
`Page 16 of 84
`
`
`
`BPO 0.05 M in cyclohexanone
`100 MHz 110°C
`
`l=l2min
`
`:==&min
`
`t=4min
`
`t=0
`
`Fig. 12.3- NMR spectra recorded during thermal decompo-
`sition of dibenzoyl peroxide. Singlet at high field is due to
`benzene; other signals are due to dibcnzoyl peroxide. [ From
`H. E-‘lseher and J. Bargon, Ace. Chem. Res. 2, 110 (1969).
`Reproduced by permission of the American Chemical
`Society]
`
`when the normal population of nuclear spin states dictated by the Boltzmann
`distribution is disturbed by the presence of an unpaired electron. The magnetic
`moment associated with an electron causes a. redistribution of the nuclear spin states.
`Individual nuclei can become overpopulated in either the lower or upper spin state.
`If the lower state is overpopulated, an enhancecl absorption signal is observed. If
`the upper state is overpopulated, an emission signal is observed. The CIDNP method
`is not as general as EPR spectroscopy because not all free-radical reactions can be
`expected to exhibit the phenomenon."
`Figure 12.3 shows the observation of CIDNP during the decomposition of
`dibenzoyl peroxide in cyelohexanone:
`
`ii
`‘i
`PhCOQCPh -—§ 2Ph» + 2C0;
`
`?h' + S-‘H —+ C6H5 + S-'
`
`The emission signal corresponding to benzene confirms ‘that it is formed by a
`free-radical process. As in steady-state EPR experiments, the enhanced emission
`and absorption are observed only as long as the reaction is proceeding. When the
`
`11. Ear a discussion of the theory of CHJNP and the conditions under which spin polarization occurs,
`see G. L. Class, Ado. Magn. Resen. ‘I, 157 (1974); R. Kaptein, Ads. Free Radical Chem. 5. 33 (19753;
`G. L. Class, R. J. Miller, and O. D. Redwine, Acc. Chem. Res. 18, 195 (2985).
`
`Page 16 of 84
`
`

`
`Page 17 of 84
`
`reaction is complete or is stopped in some way, the signals rapidly return to their
`normal intensity, because the equilibrium population ofthc two spin states is rapidly
`reached.
`
`One aspect of both EPR and CIDNP studies that should be kept in mind is
`that either technique is capable of detecting very small amounts of radical intermedi-
`ates. This aspect makes both techniques quite sensitive but can also present a pitfall.
`The most prominant features of either EPR or CIDNP spectra may actually be due
`to radicals that account for only minor products of the {oral reaction process. An
`example ofthis was found in a CIDNP study of the decomposition of trichlorcacetyl
`peroxide in alkcnes:
`
`0 0
`ll
`,
`fl
`
`*3’ 2 Cl3C-
`
`'5' 2 CO2
`
`CI_~,C- + CH;-=C{CH132 -5 Cl_-,CCHgC(CH3)2
`
`CBC‘
`
`“F C13CCHsC(CH.-:32 “"* CECE 4' Cl3CCHg('T’=-"CH;
`CH3
`
`In addition to the emission signals of CHCI3 and Cl,CCl-l;C(CH3)=CI-"lg, which
`are the major products, a strong emission signal for Cl3CCI-ICI2 was identified.
`However, this compound is a very minor product of the reaction, and when the
`signals have returned to their normal intensity, Cl3CCHCl; is present in such a
`small amount that it cannot be detected.”
`
`12.1.4. Sources of Em Radicals
`
`There are several reactions that are quite commonly used as sources of free
`radicals, both for the study of radical structure and reactivity and also in synthetic
`processes. Some of the most general methods are outlined here. Examples of many
`of these will be encountered again when specific reactions are discussed. For the
`most part, we will defer discussion of the reactions of the radicals until then.
`
`Peroxides are a common source of radical intermediates. An advantage of the
`generation of radicals from peroxides is that reaction generally occurs at relatively
`low temperature. The oxygen-oxygen bond in peroxides is weak (~30 kcal/mol),
`and activation energies for radical formation are low. Diacyl peroxides are sources
`of alkyl radicals because the carboxyl radicals that are initially formed lose CO2
`very rapidly.” In the case of aroyl peroxides, products may be derived from the
`carboxyl radical or the radical formed by decarboxylation."
`
`12. H. Y. Loicen, R. G. Lawler, and H. R. \lls'ard, J". Org. Chem. 38, I06 (1973).
`I3. 1. C. Martin, J. W. Taylor, and E. H. Drew, J. Am Chem. Soc 89, 129 (1967); F. D. Greene, H. P.
`Stein, C,~C. Chu, and F. M. Vane, J. Am. Chem. Soc. 86, 2080 (1954).
`I4. I). F. DcTar, K A. .1. Long, .1. Rendleman, 1. Bradley, and P.Dunca;r_. J. Am. Chem. Soc. 89,4051 (1967).
`
`
`
`»,........;..u......-3-:1
`
`660
`
`CHAPTER I2
`FREE-RADICAL
`REACTIGNS
`
`
`
`‘l.
`.3.
`
`
`
`‘“.r‘-‘r$:=s’.g;:§>e«r‘«.r.m«.«.w..:.s=A:.~::;:1:/..-<:;»we-'=n%?*i~:.3
`
`
`Page 17 of 84
`
`

`
`661
`
`SECTION 12.].
`GENERATION AND
`CHARACTERIZATION
`BF FREE RADICALS
`
`o
`0
`5:
`n
`I!
`.
`cnlcooccrg —i‘i'1°°—“—‘-> 2CH3CO-i —> 2CH3- + zcoz
`
`o I
`
`ll
`.
`.1:
`rhi:oocph “i'—'E> 2PhC0* —» 29:1» + 2:20,,
`
`Alkyl hydroperoxides give alkoxy radicals and the hydroxyl radical. I-Butyl hydro-
`pcroxide is often used as a radical source. Detailed studies have been reported on
`the mechanism ofthe decomposition, which is a somewhat more complicated process
`than simple unimolecular decomposition.” Dialkyl peroxides decompose to give
`two allcoxy radicals.“
`
`gH,ooc,H5 3» 2C:I~I,0.
`
`Peroxyesters ere also sources ofradicals. The acyloxy portion normally loses carbon
`dioxide, so peroxyesters yield an alkyl (or aryl) and an alkoxy radical."
`
`0l
`
`l
`RCOOC{CH3)3 -> R: + C0; + 'OC(CH3}3
`
`The decomposition of peroxides, which occurs thermally in the examples cited
`above, can also be readily accomplished by photochemical excitation. The alkyl
`hydroperoxides are also sometimes used in conjunction with a transition metal ion.
`Under these conditions, an alkoxy radical is produced, but the hydroxyl portion
`appears as hydroxide ion as a result of one-electron reduction by the metal ion.“
`
`{cH,),cooH + M“ ~+ (CH3)3C0- + “OH + M“
`
`The thermal decompositions described above are unimolecular reactions that
`should exhibit first-order kinetics. Under many conditions, peroxides decompose
`at rates faster than expected for unimolecular thermal decomposition, and with
`more complicated kinetics. This behavior is known as induced decomposition and
`occurs when part of the peroxide decomposition is the result ofbimolecular reactions
`with radicals present in solution, as illustrated specifically for diethyl peroxide:
`
`x- + CH,CH2OOCH;CH3 -+ CH,CHOOCH2CI-I3 + I-l.--X
`CH3f;l-KOOCI-l2CH3, —» Cl-l3CH=0 + AOCHECH,
`
`The amount of induced‘ decomposition that occurs depends on the concentration
`and reactivity of the radical intermediates and the susceptibility of the substrate to
`radical attack. The racial X» may be formed from the peroxide, but it can also be
`derived from subsequent reactions with the solvent. For this reason, both the structure
`
`.. R. H
`iatt, 1‘. Mill, and F. R. Mayo, J. Org. Chem, 33, 1416 (1,968), and accompanying papers.
`)9-Al-vmm
`. W. ‘
`A. Pryor, D. M. Huston, ‘l". R. Fiskc, T. L. Pickering, and E. Ciuifarin, J. Am. Chem. Soc: 86,
`423?
`(3964).
`1?‘. P. D
`. Bartlett and R. R. I-liatt, J. Am. Chem. Soc. 80, 1398 (1958).
`18. W.
`H. Richardson, I. Am. Chem. Soc. 87, 247 (1965).
`
`
`
`.“".3‘
`
`NO
`
`
`
`at"-<"mo‘*<om
`
`20:3’
`
`Page 18 of 84
`
`Page 18 of 84
`
`

`
` I
`
`lllEE
`
`662
`
`CHAPTER 12
`FREE-RADICAL.
`REACTIONS
`
`of the peroxide and the nature of the reaction medirzm are important in determining
`the extent of induced decomposition relative to unimolecular homolysis.
`Another quite general source of free radicals is the decomposition of azo
`compounds. The products are molecular nitrogen and the radicals derived from the
`substituent groups.
`
`R—N=N——R'
`
`:19?
`
`is
`
`R’ + NEN + -R’
`
`Both symmetrical and unsymmetrical azo compounds can be made so 3. single radical
`or two different ones may be generated. The energy for the decomposition can be
`either thermal or photochemical.” In the thermal decomposition, it has been estab-
`lished that the temperature at which decomposition occurs depends on the nature
`of the substituent groups. Azomethane does not decompose to methyl radicals and
`nitrogen until temperatures above 400°C are reached. Azo compounds that generate‘
`relatively stable radicals decompose at much lower temperatures. Azo compounds
`derived from ally} groups decompose somewhat above 100°C, for example:
`
`cH3cH,cri,N=NcH,cH=ct-i, 3333* cH,cH.<:H,» + N2 + CH,=CHCH,-
`
`Roi". 20
`
`Unsymmetrical azo compounds must be used to generate phenyl radicals because
`azobenzene is very stable thermally. Phenylazotriphenylmethane decomposes readily
`because of the stability of the triphenylmethyl radical.
`
`PhN=NC(Ph)3 35» rs + ph,c- + N2
`
`Ref. 21
`
`Many azo compounds also generate radicals when photolyzed. This can occur
`by a thermal decomposition of the cis azo compounds that are formed in the
`photochemical step.” The cis isomers are thermally much more labile than the trans
`isomers.
`
`R
`
`./
`NED:
`
`-o
`
`N.-:Es'
`
`s‘'’
`
`R!’
`
`‘R
`
`—» R- N,
`
`-R
`
`N-Nitrosoanilides are an alternative source of aryl radicals. There is a close
`mechanistic relationship between this route and the decomposition of azo oom—
`pounds. The N-nitrosoanilidcs rearrange to an intermediate with a nitrogen-nitrogen
`double bond. This intermediate then decomposes to generate aryl radicals-.23
`
`s=or
`
`},‘==<o
`
`_.._g is, ‘ER ._g - + N2 + RCO}-
`
`ra. P. s. Engcl, Chem. Rev. so. 99 (1950).
`20. K. Talcagi and R. J. Crawford, J. Am. Chem. Soc. 93, 5913' (1971).
`‘21. R. F. Bridger and G. A. Russell, J. Am. Chem. Soc. 85, 3154 {I963}.
`22. M. Schmittel and C. Riichardt, J. Am. Chem. SOC. 1119., 2750 (193?).
`23. C.‘RfichardI and B. Freudenbcrg, Tetrahedron Lett, 3623 (1964); J. I. G. Cadogan, Acc. Chem. Res.
`4, 186 (1971).
`
`
`
`
`
`..............................,....-._.s-._...s....._....
`
`Page 19 of 84
`
`Page 19 of 84
`
`

`
`A technique that is a convenient source of radicals for study by EPR involves
`photolysis of a mixture of di-I-butyl peroxide, triethylsilane, and the alkyl bromide
`corresponding to the radical to be studied.“ Photolysis ofthe peroxide gives I-butoxy
`radicals, which selectively abstract hydrogen from the silane. This reactive silicon
`radical in turn abstracts bromine, generating the alkyl radical at a steady-state
`concentration suitable for EPR study.
`
`663
`
`SECTION 12.1.
`GENERATION AND
`CHARACTERIZATION
`or FREE RADICALS
`
`(cH.>3cooc<cH.;. iv 2 (camco-
`
`(CH3}3C0‘ + (CgH5};SiH -——+ (CH3)3COI-I
`
`-5- (C;H5)3Si-
`
`(CzHs}55i' + RB‘ "’ lC2Hs)3SiBT “‘ R‘
`
`The acyl deri