Oxygen acidity of 1-arylalkanol radical cations. 4-Methoxycumyloxyl radical as -C(Me)2-O--to-nucleus electron-transfer intermediate in the reaction of 4-methoxycumyl alcohol radical cation with OH- [4]

Full text of DOI:10.1021/ja980645n

11516
J. Am. Chem. Soc. 1998, 120, 11516-11517
Scheme 1
Oxygen Acidity of 1-Arylalkanol Radical Cations.
4-Methoxycumyloxyl Radical as
-C(Me)2-O--to-Nucleus Electron-Transfer
Intermediate in the Reaction of 4-Methoxycumyl
Alcohol Radical Cation with OH-
Enrico Baciocchi,*,1a Massimo Bietti,1b
Osvaldo Lanzalunga,1c and Steen Steenken*,1d
Dipartimento di Chimica and
Centro CNR di Studio sui Meccanismi di Reazione
UniVersita´ “La Sapienza”
P.le A. Moro, 5 I-00185 Rome, Italy
Dipartimento di Scienze e Tecnologie Chimiche
UniVersita´ “Tor Vergata”
Via della Ricerca Scientifica, I-00133 Rome, Italy
Max-Planck-Institut fu¨r Strahlenchemie
D-45413 Mu¨lheim, Germany
However, when the reaction was performed at pH ) 10, a fast
decay of the radical cation was observed, resulting in the formation
of 4-methoxyacetophenone (λmax ) 280 nm).10 The rate constant
5.1 × 109 M-1 s-1 was measured for the OH--induced reaction,
a value almost identical to those previously obtained for the
reaction of other 4-MeOPhCH(OH)R radical cations under similar
conditions.2 It is therefore reasonable to assume that MCA•+ also2
is deprotonated by OH- at the alcoholic OH group with or
followed by electron transfer to yield the intermediate 4-meth-
oxycumyloxyl radical which subsequently forms 4-methoxy-
acetophenone by â-cleavage11 (Scheme 1).
Full support of this hypothesis was obtained when the reaction
was carried out at pH ) 11 where the rate of decay of the radical
cation is 10 times larger than that at pH ) 10.
Under these conditions, the spectra presented in Figure 1 were
obtained which clearly show that the product of the radical cation
decay at 440 nm (see also inset a) is a species characterized by
a broad absorption band centered at ∼660 nm which subsequently
decays (inset b, the “spike” after the pulse is due to the removal
of e-aq by S2O82-) giving rise, with the same rate, to 4-methoxy-
acetophenone (inset c).12
ReceiVed February 26, 1998
We have recently shown that 1-arylalkanol radical cations,2
e.g., [4-MeOPhCH(OH)R]•+, in aqueous solution can exhibit
oxygen acidity in addition to the expected and well-known
carbon acidity, depending on pH. Thus, whereas at pH e 5
4-MeOPhCH(OH)R]•+ radical cations undergo R-C-H deproto-
nation (R ) H, k ) 1.5 × 104; R ) Me ) 7.0 × 103 s-1
,
determined by conductance),3 at pH ) 10, a very fast OH--
induced reaction (k ≈ 5 × 109 M-1 s-1) takes place involving
deprotonation at the OH group. It was suggested that a
benzyloxyl radical forms (either directly (concerted with OH-
attack) or via an intermediate radical zwitterion) which then
undergoes a formal 1,2-hydrogen atom shift4 (R ) H, Me) con-
verting the oxyl radical into a carbon-centered radical or a
â-fragmentation reaction (R ) tBu) leading to 4-methoxyben-
zaldehyde and the radical R• (Scheme 1). However, the oxyl
radical postulated in Scheme 1 has so far not been seen.
Since decisive support for the suggested OH deprotonation
consists of the direct observation of the benzyloxyl radical in the
reaction with OH- of a suitable 1-arylalkanol radical cation pre-
cursor, we have now studied the reaction of 4-methoxycumyl
alcohol (MCA) radical cation in acidic and basic solutions. If
this species is indeed deprotonated at the OH group by OH-, the
cumyloxyl radical should form, and there should be a good chance
to detect it since the only reaction of this radical is â-fragmentation
leading to •CH3 (the least stable alkyl radical), a process expected
to be relatively slow.
On the basis of these observations, the 660-nm absorption band
is assigned to the 4-methoxycumyloxyl radical, for which a λmax
of 590 nm was reported for acetonitrile as solvent.13,14 In these
studies, it was found that the visible absorption band of cumyloxyl
radicals is solvent-insensitive when measured in a large variety
of nonaqueous solvents. We felt, however, that a red-shift of
(5) For details on this technique, see ref 6 or ref 2 or 7.
(6) O’Neill, P.; Steenken, S.; Schulte-Frohlinde, D. J. Phys. Chem. 1975,
79, 2773.
(7) Baciocchi, E.; Bietti, M.; Putignani, L.; Steenken, S. J. Am. Chem. Soc.
1996, 118, 5952.
Along these lines, the radical cation of MCA was generated
by pulse radiolysis in water using SO4•- or Tl2+ as the oxidant.5
Under acidic conditions (pH ) 4.1), at 4 µs after the pulse, the
complete formation of MCA•+ was observed which exhibits the
characteristic6,7 anisole-type absorption bands centered at ∼290
and 440 nm.
(8) Some steady state γ-radiolysis experiments9 were carried out at pH )
•-
4.0, using SO4 to generate the radical cation. Under these conditions the
only observed product was 4-methoxyacetophenone, but its yield (based on
the initial concentration of radical cation, calculated from the radiation dose)
was only 3.5%. A reasonable explanation is that the long-lived radi-
•
cal cation reacts with the radical CH2C(CH3)2OH (formed through H-atom
abstraction from 2-methyl-2-propanol by •OH), thus leading to the regeneration
of MCA. In agreement with this hypothesis is the observation of a 4-fold
acceleration in the rate of decay of MCA•+ when 2-methyl-2-propanol was
added in a concentration sufficient to scavenge 50% of the initially produced
The radical cation, as generated with Tl2+, decayed with the
(low) rate constant 2.9 × 102 s-1 by formation of H+, as measured
by time-resolved AC-conductance. On this time scale, products
of this decay were not visible with optical detection.8
•
OH• radicals, which leads to the production of CH2C(CH3)2OH instead of
Tl2+. The details of the mechanism of the decay of MCA•+ in acid medium
are currently under investigation.
(1) (a) Dipartimento di Chimica, Universita´ “La Sapienza”; (b) University
“Tor Vergata”; (c) Dipartimento di Chimica and Centro CNR di Studio sui
Meccanismi di Reazione, Universita´ “La Sapienza”; (d) Max-Planck-Institut.
(2) Baciocchi, E.; Bietti, M.; Steenken, S. J. Am. Chem. Soc. 1997, 119,
4078.
(3) As expected, this reaction shows a significant kinetic isotope effect,
i.e., for [4-MeOPhCH2OH]•+/[MeOPhCD2OH]•+, kH/kD ) 4.5, and, for the
corresponding methyl ethers, kH/kD ) 5.7.
(4) Gilbert, B. C.; Laue, H. A. H.; Norman, R. O. C.; Sealy, R. C. J. Chem.
Soc., Perkin Trans. 2 1976, 1040. Dobbs, A. J.; Gilbert, B. C.; Laue, H. A.
H.; Norman, R. O. C. J. Chem. Soc., Perkin Trans. 2 1976, 1044. Gilbert, B.
C.; Holmes, R. G. G.; Laue, H. A. H.; Norman, R. O. C. J. Chem. Soc., Perkin
Trans. 2 1976, 1047.
(9) Irradiations were carried out on Ar-saturated aqueous solutions contain-
ing 1 mM MCA, 0.5 mM K2S2O8, and 0.2 M 2-methyl-2-propanol, at room
temperature, using a 60Co γ-source at dose rates of 0.5 Gy s-1, for the time
necessary to obtain a 40% conversion with respect to peroxydisulfate. The
pH of the solution was adjusted to 4 or 10 with HClO4 or NaOH, respectively;
in the latter case, 1 mM Na2B4O7 × 10H2O was added to avoid undesired pH
changes upon irradiation. Products were identified and quantitatively deter-
mined by HPLC (comparison with authentic samples).
(10) Steady-state experiments confirmed that under these conditions,
4-methoxyacetophenone is produced from MCA•+ in quantitative yield.
(11) The methyl radical was in fact detected via EPR by trapping with
-
CH2dNO2 (We thank Dr. K. Hildenbrandt for performing this experiment
at pH ) 11).
10.1021/ja980645n CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/21/1998

Communications to the Editor
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11517
•-
Figure 2. Absorption spectra of transients obtained after 248-nm laser
flash photolysis of an Ar-saturated CH3CN solution containing 5 mM
4-methoxycumyl-tbutyl peroxide, recorded 0.1, 0.35, 2, and 7 µs after
the laser flash. Insets: (a) buildup of 4-methoxyacetophenone monitored
at 280 nm; (b) decay of the 4-methoxycumyloxyl radical monitored at
580 nm;. (c) comparison between the transient absorption spectra of
MeOCumO• generated by 248-nm LFP in CH3CN (b) and in CH3CN/
H2O 1:1 (]).
Figure 1. Time-resolved absorption spectra observed on reaction of SO4
with MCA (1 mM) recorded on pulse radiolysis of an Ar-saturated
aqueous solution (pH ) 11.0), containing 0.1 M 2-methyl-2-propanol, 1
mM Na2B4O7 and 10 mM K2S2O8, at 112 (b), 222 (]), 324 (9) and 624
ns (4) after the 20 ns, 10-MeV electron pulse. Insets: (a) decay of MCA•+
monitored at 440 nm; (b) buildup and subsequent decay, monitored at
660 nm assigned to the 4-methoxycumyloxyl radical; visible is also the
fast decay of the electron, due to eq 3; (c) buildup at 290 nm assigned to
4-methoxyacetophenone.
were obtained with the unsubstituted cumyloxyl radical where a
red shift of 30 nm was observed for the same solvent change,
thus showing the generality of the phenomenon.17
λmax could not be excluded in water in view of the special
characteristics of this solvent, which should stabilize the excited
state of the 4-methoxycumyloxyl radical, suggested to be char-
acterized by a quite large charge separation.14
In conclusion, we have obtained for the first time direct
evidence that 1-arylalkanol radical cations can be deprotonated
at the OH group to finally form a highly reactive alkoxyl radical.
Apparently, the oxygen acidity of these radical cations is displayed
only in fairly basic media, and this can be rationalized in terms
of the oxygen-bonded hydrogen being a hard acid center, requiring
a hard base (OH-) to react. With the neutral base water (at pH
e 5), the normal (weak) C-H acidity prevails, as concluded from
systems containing an R-hydrogen.2 There is an interesting
theoretical aspect in this oxygen acidity as it finally leads to an
intramolecular electron transfer from the side-chain (-O- or
-OH, depending on whether the zwitterion (see Scheme 1) is an
intermediate rather than a transition state) to the aromatic π-system
of the radical cation, involving orbitals which, different from those
of the CR-H bond, do not directly overlap. Finally, another novel
observation reported in this paper is that the visible absorption
band of cumyloxyl radicals undergoes a significant red-shift on
going from nonaqueous to aqueous solvents. This effect of
wateris in contrast to the solvent insensitivity13,14 with respect to
nonaqueous solvents.18
To test this hypothesis, we synthesized 4-methoxycumyl-t-butyl
peroxide (4-MeOPhC(CH3)2OOC(CH3)3)15 and produced the
4-methoxycumyloxyl radical, 4-MeOPhC(CH3)2O• (MeOCumO•),
by 248-nm photolysis of the peroxide in CH3CN and in CH3CN/
H2O 1:1 (v/v).16 Figure 2 displays the spectrum of MeOCumO•
recorded in CH3CN at 100 ns after the laser flash. The spec-
trum, with a λmax of 580 nm for the visible absorption band,
is in excellent agreement with that13,14 previously reported.14a
MeOCumO•, the species responsible, undergoes a first-order
change, such that at 7 µs after the flash is present, the spectrum
of 4-methoxyacetophenone whose buildup at 280 nm (inset a)
occurs with the same rate (k ) 1.0 × 106 s-1) as the decay of
MeOCumO• at 580 nm (inset b).
In inset c, the visible absorption band of the 4-methoxycumyl-
oxyl radical in CH3CN (b) is compared with that in CH3CN/
H2O 1:1 (open diamonds). It is evident that the presence of water
results in a red-shift of ∼45 nm. This observation supports the
attribution of the 660 nm absorption in water to the 4-methoxy-
cumyloxyl radical since an even larger red-shift should occur on
going from CH3CN/H2O 1:1 to pure water. Analogous results
Acknowledgment. E.B., M.B., and O. L. thank the Ministry of
University and Technological Research (MURST) and the National
Research Council (CNR) for financial support. Some of the experiments
were performed at the Paterson Institute for Cancer Research Free Radical
Research Facility, Manchester, U.K. with the support of the European
Commission through the Access to Large-Scale Facilities activity of the
TMR Program.
(12) On the basis of the values of the extinction coefficients of MCA•+
(ꢀ440 ≈ 3600 M-1 cm-1) and of MeOCumO• (ꢀ590 ≈ 1550 M-1 cm-1, as
measured in MeCN14) and taking account of the buildup and decay kinetics,
the spectral data (Figure 1) are interpreted in terms of g90% of MCA•+ being
converted into MeOCumO•. At pH 11, MCA•+ decays and MeOCumO• is
produced with k ≈ 7 × 106 s-1 (at pH 11.5, k g 1.2 × 107 s-1), and from
inset b it is evident that the rate of buildup of MeOCumO• is larger than that
of its decay. Furthermore, the decay kinetics of the 660 nm species matches
the buildup of the 290 nm one (k ≈ 3 × 106 s-1 at pH 11). Thus, we are
dealing with the 440 f 660 f 290 nm sequence.
JA980645N
(13) Avila, D. V.; Lusztyk, J.; Ingold, K. U. J. Am. Chem. Soc. 1992, 114,
6576.
(17) The cumyloxyl radical, PhC(CH3)2O•, was produced by 248-nm
photolysis of (commercially available) dicumylperoxide in CH3CN and in
CH3CN/H2O 1:1 (v/v).16 The spectrum recorded in CH3CN is in perfect
agreement with that previously reported, with a λmax of 485 nm for the visible
absorption band.13,14 This band shifts to 515 nm in CH3CN/H2O 1:1 (v/v).
(18) Since AcOH and EtOH behave the same way as aprotic solvents,13,14
the remarkable effect of water on the visible band of benzyloxyl radicals is
probably not (exclusively) related to hydrogen-bond formation: Thus, other
solvent properties (e.g., the dielectric constant) must play a role.
(14) Avila, D. V.; Ingold, K. U.; Di Nardo, A. A.; Zerbetto, F.; Zgierski,
M. Z.; Lusztyk, J. J. Am. Chem. Soc. 1995, 117, 2711. (a) The band at ∼400
nm is mainly due to the 4-HOC(Me)2C6H4O• radical which is a side-product
(by O-Me bond rupture) of the photolysis of the peroxide precursor.
(15) Hendrickson, W. H.; Nguyen, C. C.; Nguyen, J. T.; Simons, K. T.
Tetrahedron Lett. 1995, 36, 7217.
(16) In pure water, the peroxide is not sufficiently soluble.