Extremely Fast Hydrogen Atom Transfer between Nitroxides and HOO · Radicals and Implication for Catalytic Coantioxidant Systems

Article abstract of DOI:10.1021/jacs.8b06336

We report a novel coantioxidant system based on TEMPO (2,2,6,6-tetramethylpiperidine-1-oxyl radical) that, in biologically relevant model systems, rapidly converts chain-carrying alkylperoxyl radicals to HOO·. Extremely efficient quenching of HOO· by TEMPO blocks the oxidative chain. Rate constants in chlorobenzene were measured to be 1.1 × 10⁹ M^-1 s^-1 for the reductive reaction TEMPO + HOO· → TEMPOH + O₂ and 5.0 × 10⁷ M^-1 s^-1 for the oxidative reaction TEMPOH + HOO· → TEMPO + H₂O₂. These rate constants are significantly higher than that associated with the reaction of HOO· with α-tocopherol, Nature's best lipid soluble antioxidant (k = 1.6 × 10⁶ M^-1 s^-1). These data show that in the presence of ROO·-to-HOO· chain-transfer agents, which are common in lipophilic environments, the TEMPO/TEMPOH couple protects organic molecules from oxidation by establishing an efficient reductive catalytic cycle. This catalytic cycle provides a new understanding of the efficacy of the antioxidant capability of TEMPO in nonaqueous systems and its potential to act as a chemoprotective against radical damage.

Full text of DOI:10.1021/jacs.8b06336

Subscriber access provided by READING UNIV
Article
Extremely fast hydrogen atom transfer between nitroxides and
HOO• radicals and implication in catalytic co-antioxidant systems
Andrea Baschieri, Luca Valgimigli, Simone Gabbanini, Gino
DiLabio, Eduardo Romero-Montalvo, and Riccardo Amorati
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b06336 • Publication Date (Web): 19 Jul 2018
Downloaded from http://pubs.acs.org on July 19, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
Extremely fast hydrogen atom transfer between nitroxides and HOO• radicals and
implication in catalytic co-antioxidant systems
7
8
9
Andrea Baschieri,a Luca Valgimigli,*a Simone Gabbanini,b Gino DiLabio,c,d Eduardo Romeroꢀ
Montalvo,c Riccardo Amorati*a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a Department of Chemistry “G. Ciamician”, University of Bologna, Via S. Giacomo 11, 40126
Bologna, Italy. Eꢀmail: riiccardo.amorati@unibo.it; luca.valgimigli@unibo.it
b R&D division, BeC s.r.l. Via C. Monteverdi 49, 47122 Forlì, Italy
c Department of Chemistry, University of British Columbia, 3247 University Way, Kelowna, British
Columbia, Canada
d Faculty of Management, University of British Columbia, 1137 Alumni Ave, Kelowna, British
Columbia, Canada
Abstract: We report a novel coꢀantioxidant system based on TEMPO (2,2,6,6ꢀtetramethylpiperidineꢀ1ꢀ
oxyl) that, in biologicallyꢀrelevant model systems, rapidly converts chainꢀcarrying alkylperoxyl
radicals to HOO•. Extremely efficient quenching of HOO• by TEMPO blocks the oxidative chain. Rate
constants in chlorobenzene were measured to be 1.1×109 M‒1s‒1 for the reductive reaction TEMPO +
HOO• → TEMPOH + O2 and 5.0×107 M‒1s‒1 for the oxidative reaction TEMPOH + HOO• → TEMPO
+ H2O2. These rate constants are significantly higher than that associated with the reaction of HOO•
with αꢀtocopherol, Nature’s best lipid soluble antioxidant (k = 1.6×106 M‒1s‒1). These data show that in
the presence of ROO•ꢀtoꢀHOO• chainꢀtransfer agents, which are common in lipophilic environments,
TEMPO/TEMPOH couple protects organic molecules from oxidation by establishing an efficient
reductive catalytic cycle. This catalytic cycle provides a new understanding of the efficacy of the
antioxidant capability of TEMPO in nonꢀaqueous systems and its potential to act as a chemoprotective
against radical damage.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 25
1
2
3
4
Introduction
5
6
7
8
Nitroxides are stable and persistent free radicals that find wide use in (bio)chemical applications, such
as polymerization catalysts, spin labels, organic battery electrodes, and antioxidants.1ꢀ5 In biomedical
applications, recent work explored the use of nitroxides as antioxidants in the inhibition of
ferroptosis,6,7 protection from retinopathy,8 and from ischemiaꢀreperfusion.9 Nitroxides are also crucial
intermediates formed during the antioxidant action of aromatic amines and of hindered amine light
stabilizers (HALS), which are widely used as polymer and oil stabilizers,10 and as active moieties of
nanoantioxidants.11 In these examples, nitroxides protect organic molecules from autoxidation.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Lipid peroxidation is sustained by alkylperoxyl radicals (ROO•), which are formed from the reaction of
carbonꢀcentered radicals (R•) with O2 that is ubiquitous in biologically relevant environments.12 In
water, nitroxides react very quickly with alkylperoxyl radicals (k = 2.8–10 × 106 M−1s−1) leading to the
oxidation of the nitroxide to form the oxoammonium cation (Reaction 1), via protonꢀcoupled electron
transfer mechanisms.3,5 The formation of oxoammonium ion is also the key step in the reaction with
•‒
superoxide (HOO• / O2 ) in water, where nitroxides behave in a catalytic fashion by cycling between
the oxoammonium and the nitroxide redox states (Reactions 2 and 3).4
These reactions explain the antioxidant activity of nitroxides in aqueous systems.13 However, this
mechanism cannot be invoked in the many cases in which nitroxides reduce the extent of autoxidation
in lipophilic environments such as the interior of membranes8,14 and in oils,11 particularly near room
temperature. In fact, dialkyl nitroxides are only weak retardants of the autoxidation of hydrocarbons
and of polyunsaturated lipids in model apolar solvents.15 The limited activity of nitroxides in this
ACS Paragon Plus Environment

Page 3 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
connection is due to the quenching of alkyl radicals (Reaction 4)15 or to the (inefficient) Hꢀatom
transfer reaction from a CꢀH bond of the alkyl peroxyl radical to the nitroxide to afford the
corresponding hydroxylamine (Reaction 5).16 Some of us reported the first case of efficient quenching
of ROO• by nitroxides in nonꢀaqueous solvents wherein the reaction of dialkyl nitroxides with ROO• in
MeCN can be substantially accelerated by the addition of a weak acid.17 The proposed mechanism is a
protonꢀcoupled electron transfer from the nitroxide and the acid to ROO•, producing the corresponding
hydroperoxide and the oxoammonium salt (Reactions 1 and 6, where H+ is provided by a carboxylic
acid). 10,17
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Interestingly, nitroxides have been reported to act as catalytic antioxidants in organic solvents when the
chainꢀcarrying radicals are hydroperoxyl (HOO•), such as during the autoxidation of alcohols and
amines at temperatures of 323ꢀ348K. Inhibition was suggested to occur via reactions 7 and 8, where 7
is the rate determining step.18b
O
N
OH
N
(7)
(8)
+ O2
+ HOO
+ HOO
OH
N
O
N
+ H2O2
Pliss et al. recently reported that 2,2,6,6ꢀtetramethylpiperidineꢀ1ꢀoxyl (TEMPO) catalytically inhibited
the autoxidations of substituted ethylenes and 1,3ꢀbutadienes at 323 K, in which reactions HOO• serves
as the chain carrying radical.19
Based on these observations, we envisaged that the catalytic quenching of HOO• could be the key to a
general strategy to achieve efficient inhibition of autoxidation in nonꢀaqueous systems. Although
reaction 7 could explain the observed antioxidant behavior, it has not been demonstrated directly.20
With the aim of rationally exploiting the catalytic antioxidant activity of nitroxides, we report herein
the first experimental measurement of the rate constants and the kinetic solvent effect of reactions 7
and 8, in the case of prototypical nitroxide TEMPO and of its hydroxylamine TEMPOH. We also
describe how coꢀantioxidants present in common lipophilic biomaterials interact with TEMPO to
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 25
1
2
3
4
5
6
remove chainꢀcarrying alkylperoxyl radicals through a highly efficient catalytic cycle. We further
discuss how this catalytic system can be used to underpin the design novel coꢀantioxidant systems.
7
8
9
Results and discussion
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1) Experimental determination of the values of k7 and k8
Hydroperoxyl (HOO•) is involved in the initiation of lipid peroxidation of membranes when exposed to
a source of superoxide,21 as well in the propagation of the peroxidative chain of alkylamines,22
alcohols,18b and cyclohexadiene derivatives23 in nonꢀaqueous media. To study this reaction, we
measured the rate of the autoxidation of 1,4ꢀcyclohexadiene (CHD), initiated at constant rate by
azobisisobutyronitrile (AIBN) at 30 °C, by measuring the O2 uptake as previously reported (see
Scheme 1).23,24
Scheme 1. Autoxidation mechanism of 1,4ꢀcyclohexadiene (CHD).
The rate of the CHD autoxidation was monitored by O2 consumption; Figure 1A shows that this
autoxidation was strongly inhibited by the addition of TEMPO. Preliminary experiments showed that
the direct reaction between TEMPO and CHD (i.e. without the radical initiator AIBN) does not occur
on the timeꢀscale of the autoxidation experiments, see Figure 1S.
ACS Paragon Plus Environment

Page 5 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. A) Inhibition of the O2 consumption during the autoxidation of CHD in PhCl initiated by
AIBN at 30° in the presence of either 5 ꢀM of TEMPO (red line) or 5 ꢀM of TEMPOH (blue line) and
simulation with COPASI (solid lines). B) Concentrations of TEMPO and TEMPOH under the same
conditions as panel A. C) EPR spectroscopic (dots) determination and simulation with COPASI (solid
lines) of the concentration of TEMPO during CHD autoxidation in the presence of (a) TEMPO (inset:
EPR spectra) or (b) TEMPOH, both at 5 ꢀM. D) Time evolution of the concentration of TEMPO and
TEMPO monitored by GCꢀMS on exposing a solution of TEMPO to atmospheric oxygen at 303 K.
On the other hand, during the autoxidation there is a rapid quantitative reduction of TEMPO to
TEMPOH as demonstrated through GCꢀMS (Figure 1B, Figure 2S and Figure 3S), UPLCꢀMS (Figure
4S) analysis of the reaction mixture. These data demonstrate the reduction of TEMPO to TEMPOH
whilst showing no formation of the (most expected) oxidized oxoammonium species. Parallel
autoxidation experiments monitored by EPR spectroscopy (Figure 1C and Figure 5S) showed a timeꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 25
1
2
3
4
evolution of TEMPO that was superimposable on the CGꢀMS data, and these were used for quantitative
analysis of reaction kinetics (vide infra).
5
6
7
8
9
The conversion of TEMPO to TEMPOH is incomplete. After the first few minutes of autoxidation, a
steadyꢀstate is reached wherein the [TEMPO] and [TEMPOH] are in a ratio of ~96:4, and this ratio was
maintained throughout the experiment (Figure 1B). This suggests that TEMPOH is oxidized back to
TEMPO through the reaction with chainꢀcarrying HOO• radicals (Reaction 8, as suggested by
Denisov18 and Pliss19) during the inhibition process. Indeed, when tested on the autoxidation of CHD,
TEMPOH had an identical antioxidant effect to that of TEMPO (Figure 1A).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We investigated other potentially competing reactions that may lead to the oxidation of the
hydroxylamine. TEMPOH is efficiently oxidized to TEMPO by atmospheric oxygen: This is a wellꢀ
known reaction that challenges the purification of the hydroxylamine following its synthesis (Reaction
ꢀ7).
Since this reaction would deplete the hydroxylamine and yield chainꢀcarrying hydroperoxyl radicals, it
would have proꢀoxidant effect, rather than an antioxidant effect as shown in Figure 1A. The kinetics of
ꢀ7 was investigated using GCꢀMS to follow the disappearance of TEMPOH in the presence of O2 in
different organic solvents. The reaction was slow, with a t1/2 of about 3 hours, and quantitatively
yielded TEMPO (see Figure 1D, Scheme 1S, Figures 6S and 7S). Second order rate constants, kꢀ7, were
measured to be 0.026 Mꢀ1sꢀ1 and 0.016 Mꢀ1sꢀ1 in PhCl and EtOAc, respectively, at 303 K (vide infra):
The values that are too low to justify the steady states of TEMPO and TEMPOH, which require rapid
interconversion.
In the polar aprotic solvents MeCN and AcOEt, the antioxidant abilities of TEMPO and TEMPOH
were smaller than those measured in PhCl (O2 consumption plots are shown Figure 8S and Figure 9S),
suggesting that reactions 7 and 8 are negatively affected by hydrogen bonding with the solvent. This is
expected for the transfer of a hydrogen atom linked to an electronegative atom.25
Autoxidation of CHD inhibited by TEMPO or TEMPOH in acetonitrile containing D2O showed a
dramatically smaller antioxidant effect with respect to the corresponding experiment in the presence of
ACS Paragon Plus Environment

Page 7 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
H2O (Figure 2 and Figure 10S). This deuterium kinetic solvent effect confirms that the key mechanism
is the formal transfer of an H atom.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Autoxidation of CHD in MeCN without inhibitors (a) or in the presence of 1.3% v/v H2O (b),
1.3% v/v D2O (c), TEMPO (d), 1.3% v/v H2O + TEMPO (e), 1.3% v/v D2O + TEMPO (f).
The foregoing findings indicate that the reaction between TEMPO and HOO• (Reaction 7) can be
explained as an efficient Hꢀatom transfer to afford TEMPOH and O2. A second important aspect of the
results shown in Figure 1A is that the number of radicals trapped by each TEMPO molecule
significantly exceeds the value of 1 that would be expected from a stoichiometric reaction between
TEMPO and HOO•. As anticipated, this is explained by the fact that TEMPOH is also able to
efficiently react with HOO•, with the formation of H2O2 and TEMPO (Reaction 8). Therefore,
reactions 7 and 8 couple to produce a reductive catalytic cycle for TEMPO + HOO•. Notably, this cycle
operates in a fashion that is alternative to the oxidative cycle previously established for TEMPO in
aqueous solutions.
To obtain the values of rate constants k7, kꢀ7 and k8, we quantitated by EPR spectroscopy the timeꢀ
evolution of TEMPO during autoxidation experiments inhibited either by TEMPO or TEMPOH. The
findings from these measurements matched our chromatographic results. Therefore,
TEMPOH/TEMPO relative concentrations determined by EPR and CGꢀMS served as constraints for
the analysis of the O2 consumption traces of CHD autoxidations performed using the kinetic simulation
software “COPASI”.26 The numerical analysis of O2 consumption plots required literature values of the
rate constants for CHD autoxidation,23 and the rate of initiation that was measured in our preliminary
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 25
1
2
3
4
5
6
7
8
9
experiments using the inhibition period method (see Scheme 2S, Table 1S and Figure 11S).27 The
procedure resulted in excellent fits of the experimental traces and yielded values of k7 and k8 in PhCl
and AcOEt, which are reported in Table 1. A similar analysis of the kinetics of TEMPOH
disappearance under O2 (Figure 1D) provided the values of kꢀ7 (Table 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Values of k7, kꢀ7 and k8 (M‒1s‒1). The ratio of the rate constants reflects the effect of solvent on
the reactions.
rate constant
PhCl[a]
AcOEt[a]
solvent effect
k7
kꢀ7
k8
(1.1±0.1) × 109
(2.6±0.2) × 10‒2
(5.0±0.4) × 107
(1.9±0.3) × 107
(1.6±0.1) × 10‒2
(1.7±0.2) × 106
58
1.6
29
[a] mean of three measures, ±Standard Deviation.
The values reported in Table 1 indicate that the reaction between TEMPO and HOO• in PhCl occurs
with nearly diffusionꢀcontrolled rate constant, in alignment with the large exothermicity of the reaction,
(ꢁH ≈ ‒9 kcal/mol in DMSO).28 The results also indicate that reaction 8 is one order of magnitude
slower than reaction 7, making it the rateꢀdetermining step of the catalytic antioxidant cycle.
In a polar solvent, k7 decreases by 58ꢀfold. The kinetic solvent effect (KSE) in k7 can be attributed to
the hydrogen bonding between HOO• and the solvent (i.e. S…HOO•), which renders it less reactive as
an Hꢀatom donor (see Scheme 2A). Unexpectedly, k8 in PhCl is significantly larger than the rate
constant for the reaction between TEMPOH and ROO• (k = 1.6 x 106 M‒1s‒1 at 303 K in the case of the
peroxyl radicals derived from styrene).17 This result can be explained by considering the smaller steric
requirement of HOO• relative to ROO•. The value of k8 decreases in AcOEt relative to the rate constant
in PhCl by an amount that is about half of the decrease observed for k7. The decrease in k8 likely result
from two parallel effects: The formation of a weak TEMPOH…S hydrogen bond complex,17 and the
“remote” Hꢀbond effect in the abstracting radical, wherein S…HOO• has a somewhat lower reactivity
than HOO• toward Hꢀatom abstraction (Scheme 2B).24a The negligible KSE found for kꢀ7 can be
attributed to a weak Hꢀbond of TEMPOH with solvents, inꢀline with previous findings. For
ACS Paragon Plus Environment

Page 9 of 25
Journal of the American Chemical Society
1
2
3
4
5
MeCN
comparison, the KSE for the reaction of TEMPOH with alkylperoxyl radicals was kinhPhCl / kinh
1.3 at 303 K.17
=
6
7
8
From the measured values of k7 and k-7, the free energy change of the TEMPO + HOO• reaction is
9
determined to be ꢀ14.5 kcal/mol and ꢀ12.4 kcal/mol in PhCl and AcOEt, respectively.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Kinetic solvent effect on the reactions of HOO• with TEMPO (A) and TEMPOH (B).
The possibility that singlet O2 is formed in Reaction 7 was considered, but it was discounted on the
1
basis of results from autoxidation experiments in the presence of the O2 trap 9,10ꢀdiphenylanthracene,
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 25
1
2
3
4
5
as the typical UVꢀvis spectrum of the trap was unchanged throughout the reaction (see Figure 12S).29
The results of theoretical calculations, (see below) also discount the formation of 1O2 in reaction 7.
6
7
8
9
2) Theoretical calculation of the barriers of reactions 7 and 8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We began our modeling efforts with an exploration of the spin state associated with reaction 7.
We considered the possibility that the reaction might occur on the singlet spin surface, thereby
1
generating O2. Transition state (TS) structures for the hydrogen atom transfer reaction involving the
triplet and openꢀshell singlet systems were examined, and both were found to be genuine TSs with
single imaginary vibrations along the hydrogen atom transfer coordinate. At the level of theory
employed in the examination of reactions 7 and 8 more generally (see below), i.e. CAMꢀB3LYPꢀ
D3/augꢀccꢀpVTZ//B3LYPꢀD3/6ꢀ31+G(d,p),30 the triplet TS was found to be 2.4 kcal/mol lower in
electronic energy than the singlet. Additional calculations were performed in order to assess the
sensitivity of the tripletꢀsinglet TS energy difference to the level of theory. Singleꢀpoint energy
calculations using the B2PLYPD331 and DSDPBEP8632 doubleꢀhybrids and the wavefunctionꢀbased
composite CBSꢀQB333 method verified that the triplet TS was lower than the singlet by 3.2, 2.6 and 2.3
kcal/mol, respectively. Thermal and entropic corrections to fee energy further increase the tripletꢀ
singlet separation by about 0.8 kcal/mol. On the basis of these results, we conclude that reaction 7
occurs on the triplet surface with a rate constant that is at least two orders of magnitude higher than that
1
associated with the singlet surface. The results of our calculations and our O2 trap experiments support
the conclusion that the reaction occurs on the triplet spin surface.
The calculated relative free energies associated with reaction 7 on the triplet surface are
summarized in Figure 3a. In PhCl, the reactants form a hydrogenꢀbonded preꢀreaction complex (ꢁG298
= ꢀ1.0 kcal/mol) prior to surmounting a relative free energy barrier of 5.7 kcal/mol. The computed
transition state (TS) structure is shown in Figure 4a. The reaction is calculated as overall exergonic by ꢀ
10.4 kcal/mol and leads to the formation of TEMPOH and 3O2. Calculations using AcOEt as the
solvent give a barrier of 6.3 kcal/mol, which is in qualitative agreement with our experimental findings.
These results are consistent with the stronger hydrogenꢀbond accepting ability of AcOEt compared to
PhCl, and the consequences this has on the reaction rate constants.34
ACS Paragon Plus Environment

Page 11 of 25
Journal of the American Chemical Society
1
2
3
4
5
A
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Calculated free energy reaction coordinate associated with reactions A) 7 and B) 8, relative
to reactants. Key: PreꢀRC = preꢀreaction complex, TS = transition state complex, PostꢀRC = postꢀ
reaction complex.
Figure 3b shows the calculated results for reaction 8. The results demonstrate that a very strong
hydrogenꢀbonded, preꢀreaction complex forms between TEMPOH and HOO•, viz. ꢁG = ꢀ4.2 kcal/mol
in PhCl, see Fig. 4b. This double hydrogen bond results from a strong orbital interaction between the
TEMPOH nitrogen loneꢀpair and the OꢀH σ* orbital of HOO• and a weaker interaction between a
HOO• O loneꢀpair and the OꢀH σ* orbital of TEMPOH. The zeroꢀpoint corrected electronic energy
calculated for this hydrogen bond is 16.4 kcal/mol in the gasꢀphase. For comparison, the highꢀlevel ab
initio value for the strong hydrogen bonding in the uracilꢀuracil dimer is 17.4 kcal/mol.35 Relative to
reactants, the free energy barrier to hydrogen atom transfer is computed to be 9.1 kcal/mol (TS shown
in Figure 4c) and is strongly downhill to products. In AcOEt, calculations predict weaker hydrogen
bonding and a higher reaction barrier, consistent with the experimental results.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 25
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Computed structure of the a) TS associated with reaction 7, and the b) preꢀreaction and c) TS
associated with reaction 8. N = blue; O = red; C = grey; H = white. Distances are in Angstroms.
The computed TS structures associated with reactions 7 and 8 are shown in Figure 4 (a,c), along with
the strongly hydrogenꢀbonded preꢀreaction for reaction 8 (Fig. 4b). We note that the TS show in Fig. 4a
has a cisoid structure. This is expected on the basis of previous work that showed that hydrogen atom
transfer reactions in which lone pairꢀlone pair overlap can occur in the TS develop partial bonding
character that stabilize the structures relative to transoid TSs wherein such overlap cannot occur (see
Figure 13S).36
3) Catalytic co-antioxidant systems
Our results show that TEMPO is a powerful catalytic quencher of HOO• in nonꢀaqueous solvents.
However, it is not obvious how this property is connected to antioxidant activity toward substrates that
autoxidize through ROO• formation, such as the unsaturated lipids in bioꢀmembranes. We
ACS Paragon Plus Environment

Page 13 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
hypothesized that TEMPO unleashes its antioxidant activity in the presence of molecules that act as coꢀ
antioxidants able to transfer the chainꢀreaction from ROO• to HOO•. Indeed, a number of structures
commonly found in natural oxidizable materials or in biological systems could serve as chainꢀtransfer
coꢀantioxidants. For instance, CHD is the model for widespread terpenoid components of vegetable oils
(e.g. αꢀterpinene, γꢀterpinene, αꢀphellandrene, etc.) known to oxidize via HOO•.37 1,4ꢀHydroquinones
are ubiquitous in biological systems (e.g. ubiquinol) and are known to form HOO• upon reaction with
ROO•.38 Additionally, biologically ubiquitous aliphatic amines22 and alcohols18 are known to form
HOO• during their autoxidation. Since k7 and k8 far exceed the rate constants of reactions of ROO• with
common chainꢀbreaking antioxidants, it is reasonable to expect that the TEMPO/TEMPOH system
would be efficient even in the presence of relatively small amounts of coꢀantioxidants.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We demonstrate the efficacy of catalytic coꢀantioxidant system by studying the autoxidation of styrene,
which is a typical substrate that forms ROO•. Inhibition of this autoxidation by TEMPO is sought in
the presence of CHD and three other model compounds that yield HOO• upon reaction with free
radicals, namely 2,5ꢀdiꢀtertꢀbutylhydroquinone (QH2),38 trimethylamine22 and benzylic alcohol39 – as
summarized in Scheme 3.
O
O2
N
H2O2 or ROOH
HOO
A B C, D
,
,
Initiation
R
ROO
OH
N
O2
HOO or ROO
RH
ROOH
III) catalytic HOO
decomposition
II) trasformation of ROO
into HOO
I) autoxidation of the
substrate RH
Scheme 3. Mechanism explaining the coꢀantioxidant effect of TEMPO in the presence of (A) 1,4ꢀ
cyclohexadienes, (B) triethylamine, (C) 1,4ꢀhydroquinones or (D) benzyl alcohol.
The results presented in Figure 5 provide support for our coꢀantioxidant hypothesis. Styrene
autoxidation is inhibited nearly to the point of elimination by the CHDꢀTEMPO, (panel A), QH2ꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 25
1
2
3
4
5
6
TEMPO (panel B), Et3NꢀTEMPO (panel C) coꢀantioxidant couples. Significant, albeit incomplete,
inhibition is obtained with the PhCH2OHꢀTEMPO coꢀantioxidant couple (panel D).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. O2 consumption during the autoxidation of styrene (4.3 M) in chlorobenzene initiated by
AIBN at 30° without antioxidants (a in panels A, B, C and D), or in the presence of: TEMPO 5 ꢀM (b
in panels A, B, C and D), CHD 0.27 M or CHD 0.27 M and TEMPO 5 ꢀM (c and d in panel A), Et3N
0.018 M or Et3N 0.018 M and TEMPO 5 ꢀM (c and d in panel B), QH2 5 ꢀM or QH2 5 ꢀM and
ACS Paragon Plus Environment

Page 15 of 25
Journal of the American Chemical Society
1
2
3
TEMPO 5 ꢀM (c and d in panel C), PhCH2OH 0.24 M or PhCH2OH 0.24 M and TEMPO 5 ꢀM (c and
4
5
6
d in panel D).
7
8
9
Inspection of the kinetic traces in Fig. 5 reveals that while some chainꢀtransfer coꢀantioxidants like QH2
and Et3NH are able to retard autoxidation on their own due to quenching of ROO• radicals and
replacing with faster terminating HOO•, addition of micromolar amounts of TEMPO boosts the
antioxidant protection by largely suppressing the radical chain. This coꢀantioxidant system outꢀ
performs Nature’s best antioxidant, αꢀtocopherol (see Figure 14S). More importantly, and unlike
typical antioxidants that quench peroxyl radicals on stoichiometric basis by blocking one to two
radicals per antioxidant molecule, TEMPO acts in a catalytic fashion by inhibiting autoxidation over a
longer duration, the extent of which depends on the availability of the chainꢀtransfer agent. The
relatively lower performance of PhCH2OHꢀTEMPO coꢀantioxidant couple can be explained by the
lower efficiency of the alcohol in affording ROO• to HOO• chainꢀtransfer. Despite the lower
efficiency, we were able to obtain relevant coꢀoxidant activity with TEMPO even in the case of
unactivated primary alcohol nꢀoctanol (Figure 15S). In general, the very large rate constants for HOO•
quenching by both TEMPO and TEMPOH (reactions 7 and 8) suggest that, under most common
experimental conditions, the limiting step governing the efficacy of the coꢀantioxidant system lies in
the chainꢀtransfer process.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To provide proofꢀofꢀconcept of the validity of our “coꢀantioxidant hypothesis” in a biological
context, we investigated the oxidation of turpentine oil. Turpentine oil, obtained from pine wood, is a
commodity produced on a scale of ca. 250,000 tons annually, and has a broad range of uses: from the
manufacturing of paints and coatings, to that of plastics, food, cosmetics and pharmaceuticals.40 It is
based on αꢀpinene, that undergoes facile autoxidation mediated by alkylperoxyl radicals,41 a process
that may be involved in its toxicity.42 This natural material also contains variable minor amounts of
other components such as αꢀ and γꢀterpinene (e.g. 0.95% and 1.65%, respectively, in the specimen we
investigated, see Figure 16S and Table 2S). Figure 6 shows that, at micromolar levels, TEMPO
completely blocks turpentine autoxidation by exploiting the natural content of CHDꢀlike terpinenes that
act as coꢀantioxidant chainꢀtransfer agents.37b
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 25
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Oxygen consumption during the autoxidation of vegetable turpentine oil (Pinus spp) 75%
V/V initiated by AIBN (0.05 M) in PhCl at 30°C in the absence of inhibitors (a) or in the presence of
TEMPO 0.5 µM (b) or 5 µM (c).
Conclusions
The prototypical nitroxide TEMPO reacts very efficiently with HOO•, forming the corresponding
hydroxylamine that is oxidized back to TEMPO by a second HOO•. The rate constants for these two
processes are ca. 2.8 and 1.5 orders of magnitude faster, respectively, in comparison to Nature’s most
effective lipidꢀsoluble, phenolic antioxidant αꢀtocopherol (kHOO = 1.6×106 M‒1s‒1 in PhCl at 30 °C).24a
The efficiency of reactions 7 and 8 gives rise to a reductive catalytic cycle alternative to the established
oxidative cycle observed in aqueous solution (Scheme 4). Our findings help resolve the longꢀstanding
mysterious antioxidant behavior of nitroxides in nonꢀaqueous solvents, and open new insights in the
role of nitroxides as HOO• sensors in the biological milieu.43
ACS Paragon Plus Environment

Page 17 of 25
Journal of the American Chemical Society
1
2
3
4
H2O2
HOO +H+
H2O2
HOO
5
6
7
8
9
Reductive cycle
Oxidative cycle
O
N
O
N
OH
N
Water and
organic solvent with
H+ donors
Mainly organic
solvents
10
11
12
13
14
15
16
17
18
O2 + H+
HOO
O2
HOO
Scheme 4. Catalytic quenching of HOO• by TEMPO.
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The catalytic antioxidant activity of TEMPO can be fully exploited in the presence of molecules able to
transfer the chainꢀcarrying radicals from ROO• to HOO•. Using cyclohexadienes, hydroquinones,
alcohols and amines as model compounds in combination with TEMPO, we demonstrated the efficacy
of these coꢀantioxidant systems in protecting naturallyꢀoccurring organic molecules from oxidation.
The models are representatives of ubiquitous natural compounds in animal tissue, for example
ubiquinol present in lipoproteins and in cellular membranes, and in plants (e.g. γꢀterpinene37).
Therefore, TEMPO can be used to provide extraordinary protection of biomaterials from the damage
due to exposure to reactive oxygen species.
More generally, the effectiveness of the nitroxideꢀbased catalytic cycle depends on the ability of the
oxidizable substrate to form HOO• radicals during autoxidation,44 and on the availability of suitable
chainꢀtransfer agents. We expect this chemistry to be very useful in the rational design of versatile,
highly effective coꢀantioxidant systems.
Experimental section
Materials
All solvents (PhCl, MeCN, ethyl acetate) were of the highest grade commercially available (≥99,9%
HPLC grade), and used as received. 1,4ꢀcyclohexadiene (CHD, 97%) was percolated on alumina and
silica before each experiment in order to remove traces of stabilizer. The initiator 2,2′ꢀazobis(2ꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 25
1
2
3
4
5
6
7
8
9
methylpropionitrile) (AIBN), was recrystallized from methanol before use. 9,10ꢀDiphenylanthracene
(97 %), 2,2,6,6ꢀtetramethylpiperidinꢀ1ꢀyl)oxyl, TEMPO (99 %), 2,2,5,7,8ꢀpentamethylꢀ6ꢀchromanol
(97 %), 2,5ꢀdiꢀtertꢀbutylhydroquinone (99%) and triethylamine (≥99%) were from Sigma Aldrich and
were used as received. 1ꢀHydroxyꢀ2,2,6,6ꢀtetramethylpiperidine, TEMPOH, was synthesized by
reducing TEMPO with sodium ascorbate as previously reported, and its characterization was consistent
with literature values.45 Turpentine was purchased from Muller & Koster and used as received. Its
composition was determined by GCꢀMS, as reported in the Supporting Information.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Measurements of autoxidation rates
Autoxidation experiments were performed by measuring the oxygen consumption in a twoꢀchannel gas
uptake apparatus, immersed in a thermostatted bath, based on Validyne DP15 pressure
transducer.24a,27,46 The rate of initiation (Ri) was calculated from preliminary set of experiments from
the length of the inhibition period, tinh, using 2,2,5,7,8ꢀpentamethylꢀ6ꢀchromanol (TOH) as a reference
antioxidant during autoxidation of styrene.47 The values of k7 and k8 (see manuscript) were obtained by
numerical fitting of the O2 consumption traces measured during the CHD autoxidation, by using the
kinetic simulation software COPASI, freely available on the internet.26 This method was previously
used by our group and gave excellent results in the case of the CHD autoxidation inhibited by
2,2,5,7,8ꢀpentamethylꢀ6ꢀchromanol24a and in the case of nonꢀclassical antioxidants.47 The reaction
scheme used to simulate the experimental data is reported in Scheme 1S.
Determination of TEMPO and TEMPOH by GC-MS
The air equilibrated autoxidizing reaction mixture containing 10 % CHD in MeCN or PhCl or EtOAc
with 0.05 M AIBN and TEMPO (5 ꢀM) was sampled every 8 minutes and subjected to GCꢀMS
analysis in a Trace 1310 – ISQꢀQD GCꢀMS equipment (Thermo Fisher Scientific) equipped with a
Combipal thermostatted autosampler (CTC) using a Zebron ZBꢀ5 column 30m x 0.25 mm x 0.25 µm
(Phenomenex) ramped from 50°C to 150°C at 20°C/min and eluted with He at 1.2 mL/min. Mass
spectra were acquired in the range 40ꢀ650 amu and in multiple SIM mode at m/z 141, 142, 156, 157.
Several calibration curves for TEMPO and TEMPOH were built with genuine standard solutions using
either the full scan signal or the SIM at m/z 141 and 156 for TEMPO and at m/z 142 and 157 for the
hydroxylamine (R2 > 0.99) to confirm quantitative analysis under different acquisition modes.
Determination of TEMPO and TEMPOH by UPLC-MS
ACS Paragon Plus Environment

Page 19 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
UPLCꢀMS analysis of the autoxidizing mixture in MeCN was performed with an Accela LCQ Fleet
equipment (Thermo Fisher Scientific) on Hypersil Gold (Thermo) 100 mm x 2.1mm x 1.9 µm C18
column, thermostatted at 35°C, eluted with 1% aqueous formic acid (A) and acetonitrile (B) with
gradient from 80%Aꢀ20%B to 30%Aꢀ70%B in 8 min, with flow of 250 µL/min. Injection volume was
2 µL. ESI+ spectra were recorded in the range m/z 50ꢀ500 with Ion spray voltage 5 KV, capillary
voltage 15V and tube lens 110V. Since the signalꢀtoꢀnoise ratio for TEMPO was too low for reliable
quantitative analysis, no calibration was performed, and analysis was used only to confirm GCꢀMS on
qualitative grounds. Identification of TEMPO and TEMPOH and absence of TEMPO+ was obtained by
injecting genuine standards.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Determination of TEMPO by EPR spectroscopy
EPR experiments were performed with a Bruker Elexsys 500 spectrometer equipped with a Super Xꢀ
Band ER049 microwave bridge and a quartz Dewar. Temperature was maintained at the desired value
by a Bruker BꢀVT100 variable temperature unit and monitored before and after each experiment with a
Delta OHM HD9218 type K thermocouple and was stable within ± 0.1°C.
An air saturated solution containing 10 % CHD in either PhCl or AcOEt with 0.05 M AIBN and
TEMPO or TEMPOH (5 ꢀM) was put in the cavity of the EPR spectrometer at 303K in an open quartz
tube (4 mm ID). The time evolution of the concentration of TEMPO was monitored at regular intervals,
both from the intensity of the first spectral line (aN= 15.6 G, g = 2.0062) and from the double integral
of the EPR spectrum.
Computational Details
The geometries of reactants, products, preꢀ and postꢀreaction complexes and the transition state
structures were optimized using B348LYP49ꢀD350/6ꢀ31+G(d,p) with the SMD51 implicit solvent model.
Frequency calculations at this same level of theory and basis set confirmed that the optimizations lead
to local maxima for the transition state structures and local minima for all other species and provided
thermal corrections to free energies. Singleꢀpoint energies were calculated on all optimized structures
using CAMꢀB3LYP30ꢀD3/augꢀccꢀpVTZ basis sets with SMD solvent. All calculations were performed
with the Gaussianꢀ09 program.52 Optimized geometries and calculated energies are reported in the ESI
(Tables S2ꢀS7).
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 25
1
2
3
4
Supporting Information
5
6
7
8
Additional experiments, experimental details, composition of turpentine, calculations and simulations
(44 pages).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acknowledgements
RA and LV thank the University of Bologna for funding. GAD thanks the National Sciences and
Engineering Research Council, Canada Foundation for Innovation, and the BC Knowledge
Development Fund for funding. ERM is grateful for the support provided by the Government of
Mexico through CONACyT (Ph.D. Scholar/Scholarship 308773/472432).
References
1
a) Tesch, M.; Kudruk, S.; Letzel, M.; Studer, A. Chem. Eur. J. 2017, 23 (25), 5915ꢀ5919; b)
Hansen, K.ꢀA.; Nerkar, J.; Thomas, K.; Bottle, S. E.; O’Mullane, A. P.; Talbot, P. C.; Blinco, J. P. ACS
Appl. Mater. Interfaces 2018, 10, 7982−7988; c) Karthikeyan, G.; Bonucci, A.; Casano, G.; Gerbaud,
G.; Abel, S.; Thomé, V.; Kodjabachian, L.; Magalon, A.; Guigliarelli, B.; Belle, V.; Ouari, O. Mileo, E.
Angew.Chem. Int. Ed. 2018, 57,1366 –1370; d) Zhang, L.; Laborda, E.; Darwish, N.; Noble, B.B.;
Tyrell, J. H.; Pluczyk, S.; Le Brun, A. P.; Wallace, G. G.; Gonzalez, J.; Coote, M. L.; Ciampi, S. J. Am.
Chem. Soc. 2018, 140, 766−774; e) Prescott, C.; Bottle, S. E. Cell Biochem. Biophys. 2017, 75, 227–
240.
2
a) Soule, B. P.; Hyodo, F.; Matsumoto, K.ꢀI.; Simone, N. L.; Cook, J. A.; Krishna, M. C.;
Mitchell, J. B. Free Radic. Biol. Med. 2007, 42 (11), 1632–1650; b) Lewandowski, M.; Gwozdzinski,
K. Int. J. Mol. Sci. 2017, 18, 2490.
3
4
5
Goldstein, S.; Samuni, A. J. Phys. Chem. A 2007, 111 (6), 1066–1072.
Goldstein, S.; Merenyi, G.; Russo, A.; Samuni, A. J. Am. Chem. Soc. 2003, 125 (3), 789–795.
Griesser, M.; Shah, R.; Van Kessel, A. T.; Zilka, O.; Haidasz, E. A.; Pratt, D. A. J. Am. Chem.
Soc. 2018, 140 (10), 3798–3808.
ACS Paragon Plus Environment

Page 21 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
6
Krainz, T.; Gaschler, M. M.; Lim, C.; Sacher, J. R.; Stockwell, B. R.; Wipf, P. ACS Cent. Sci.
Zilka, O.; Shah, R.; Li, B.; Friedmann Angeli, J. P.; Griesser, M.; Conrad, M.; Pratt, D. A. ACS
2016, 2 (9), 653−659.
7
9
Cent. Sci. 2017, 3 (3), 232−243.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
Zareba, M.; Widomska, J.; Burke, J. M.; Subczynski, W. K. Free Radic. Biol. Med. 2016, 101,
446–454.
9
Lahiani, A.; Hidmi, A.; Katzhendler, J.; Yavin, E.; Lazarovici, P. ACS Chem. Neurosci. 2016, 7,
1452−1462.
10
Haidasz, E. A.; Meng, D.; Amorati, R.; Baschieri, A.; Ingold, K. U.; Valgimigli, L.; Pratt, D. A.
J. Am. Chem. Soc. 2016, 138 (16), 5290−5298.
11
a) Yoshitomi, T.; Hirayama, A.; Nagasaki, Y.; Biomaterials 2011, 32, 8021ꢀ8028; b)
Valgimigli, L.; Baschieri, A.; Amorati, R. J. Mater. Chem. B 2018, 6, 2036ꢀ2051.
12
13
Zielinski, Z. A. M.; Pratt, D. A. J. Org. Chem. 2017, 82, 2817−2825.
An exception to the oxidative cycle for the reaction with superoxide in water was reported by
Samuni et al. in the case of the 2ꢀethylꢀ2,5,5ꢀtrimethylꢀ3ꢀoxazolidinoxyl radical, which catalytically
reacts with superoxide through the nitroxideꢀhydroxylamine couple. This reductive pathway, however,
is less efficient than the oxidative one displayed by TEMPO. Samuni, A.; Krishna, C. M.; Riesz, P.;
Finkelstein, E.; Russo, A. J. Biol. Chem. 1988, 263, 17921ꢀ17924.
14
Mobbili, G.; Crucianelli, E.; Barbon, A.; Marcaccio, M.; Pisani, M.; Dalzini, A.; Ussano, E.;
Bortolus, M.; Stipa, P.; Astolfi, P. RSC Adv. 2015, 5, 98955–98966.
15
Pedulli, G. F.; Lucarini, M.; Pedrielli, P.; Sagrini, M.; Cipollone, M. Res. Chem. Intermed.
1996, 22, 1–14.
16
Tikhonov, I. V.; Pliss, E. M.; Bogoyavlenskii, D. A.; Berezin, M. P.; Sen, V. D. Russ. Chem.
Bull., Int. Ed. 2015, 64, 2433—2437.
17
Amorati, R.; Pedulli, G. F.; Pratt, D. A.; Valgimigli, L. Chem. Commun. 2010, 46, 5139–5141.
18
a) Denisov, E. T. Kinet. Catal. 2015, 56, 56ꢀ70; b) Denisov, E. T. Russ. Chem. Rev. 1996, 65,
505ꢀ520.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 25
1
2
3
4
19
Pliss, E. M.; Machtin, V. A.; Grobov, A. M.; Pliss, R. E.; Sirick, A. V. Int. J. Chem. Kin. 2017,
49, 173ꢀ181.
5
6
20
The rate constant for reaction 7 was estimated by Denisov as ≈108 Mꢀ1sꢀ1 (at 350 K) on the basis
7
8
9
of thermodynamic considerations (ref. 18a) and was suggested to be rateꢀlimiting compared to reaction
8 (ref. 18b).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21
Bedard, L.; Young, M. J.; Hall, D.; Paul, T.; Ingold, K. U. J. Am. Chem. Soc. 2001, 123, 12439–
12448.
22
23
24
Howard, J. A.; Yamada, T. J. Am. Chem. Soc. 1981, 103, 7102−7106.
Howard, J. A.; Ingold, K. U. Can. J. Chem. 1965, 43, 2729–2736.
a) Cedrowski, J.; Litwinienko, G.; Baschieri, A.; Amorati, R. Chem. Eur. J. 2016, 22, 16441 –
16445. b) Foti, M. C.; Sortino, S.; Ingold, K. U. Chem. Eur. J. 2005, 11, 1942 – 1948.
25
26
Litwinienko, G.; Ingold, K. U. Acc. Chem. Res. 2007, 40, 222ꢀ230.
Hoops, S.; Sahle, S.; Gauges, R.; Lee, C.; Pahle, J.; Simus, N.; Singhal, M.; Xu, L.; Mendes, P.;
Kummer, U. Bioinformatics 2006, 22, 3067ꢀ3074.
27
Amorati, R.; Baschieri, A.; Morroni, G.; Gambino, R.; Valgimigli, L. Chem. Eur. J. 2016, 22,
7924 – 7934.
28
The bond dissociation enthalpies of TEMPOH is 72 kcal/mol, and that of HꢀOO• is 69 kcal/mol,
in DMSO. Warren, J. J.; Tronic, T. A.; Mayer, J. M. Chem. Rev. 2010, 110, 6961–7001.
29
Tanaka, K.; Miura, T.; Umezawa, N.; Urano, Y.; Kikuchi, K.; Higuchi, T.; Nagano, T. J. Am.
Chem. Soc. 2001, 123, 2530–2536.
30
31
32
33
Yanai, T.; Tew, D.; Handy, N. Chem. Phys. Lett. 2004, 393, 51ꢀ57.
Grimme, S. J. Chem. Phys. 2006, 124, 034108.
Kozuch, S.; Martin, J. M. L. Phys. Chem. Chem. Phys. 2011, 13, 20104–20107.
Montgomery Jr., J. A.; Frisch, M. J.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 2000,
112, 6532ꢀ6542.
ACS Paragon Plus Environment

Page 23 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
34
Our modeling work does not reproduce the experimental relative free energies of reaction 7, or
the correct ordering of the reaction exoergicities in the two solvents. This is a known shortcoming of
implicit solvation methods when reproducing solvent effects due to specific Hꢀbonding to the solvent.
See: Guerra, M.; Amorati, R.; Pedulli, G. F. J. Org. Chem. 2004, 69, 5460ꢀ5467
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
35
DiLabio, G. A.; Johnson, E. R.; OteroꢀdeꢀlaꢀRoza, A. Phys. Chem. Chem. Phys. 2013, 15,
12821ꢀ12828.
36
a) DiLabio, G. A.; Johnson, E. R. J. Am. Chem. Soc. 2007, 129, 6199ꢀ6203; b) DiLabio, G. A.;
Ingold, K. U. J. Am. Chem. Soc. 2005, 127, 6693ꢀ6699; c) Zielinski, Z.; Presseau, N.; Amorati, R.;
Valgimigli, L.; Pratt, D. A. J. Am. Chem. Soc. 2014, 136, 1570−1578.
37
a) Foti, M. C.; Ingold, K. U. J. Agric. Food Chem. 2003, 51, 2758ꢀ2765; b) Amorati, R.; Foti,
M. C.; Valgimigli, L. J. Agric. Food Chem. 2013, 61, 10835−10847.
38
Valgimigli, L.; Amorati, R.; Fumo, M. G.; DiLabio, G. A.; Pedulli, G. F.; Ingold, K. U.; Pratt,
D. A. J. Org. Chem. 2008, 73, 1830ꢀ1841
39
40
Howard, J. A.; Korcek, S. Can. J. Chem. 1970, 48, 2165ꢀ2172.
https://ntp.niehs.nih.gov/ntp/htdocs/chem_background/exsumpdf/turpentine_508.pdf accessed
on July 2018.
41
42
Neuenschwander, U.; Guignard, F.; Hermans, I. ChemSusChem 2010, 3, 75–84.
Amorati, R.; Foti, M.C. Oxidative stability and antioxidant properties of essential oils. In:
Valgimigli, L. (Ed.), Essential oils as natural food additives, Nova Science Publishers, New York,
2012. 75ꢀ95.
43
Chong, K. L.; Chalmers, B. A.; Cullen, J. K.; Kaur, A.; Kolanowski, J. L.; Morrow, B. J.;
FairfullꢀSmith, K. E.; Lavin, M. J.; Barnett, N. L.; New, E. J.; Murphy, M. P.; Bottle, S. E. Free Radic.
Biol. Med. 2018, https://doi.org/10.1016/j.freeradbiomed.2018.03.008
44
A very recent example appeared in the literature while this manuscript was in preparation is the
production of HOO• in the autoxidation of alkenes at 70°C: Harrison, K. A.; Haidasz, E. A.; Griesser,
M.; Pratt, D. A. Chem. Sci. 2018, DOI: 10.1039/C8SC01575A.
45
HenryꢀRiyad, H.; Tidwell, T. T. J. Phys. Org. Chem. 2003, 16, 559ꢀ563.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 25
1
2
3
4
5
6
7
8
9
46
47
Amorati, R.; Baschieri, A.; Valgimigli, L. J. Chem. 2017, 6369358.
Amorati, R.; Lynett, P. T.; Valgimigli, L.; Pratt, D. A. Chem. Eur. J. 2012, 18, 6370 – 6379.
Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
48
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
49.
50.
Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
a) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys. 2010, 132, 154104. b) Grimme,
S.; Ehrlich, S.; Goerigk, L. J. Comput. Chem. 2011, 32, 1456–1465.
51.
52.
Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2009, 113, 6378ꢀ6396.
Gaussian 09, Revision D.01, Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji,
H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.;
Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;
Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.;
Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari,
K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.;
Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;
Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas,
Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian, Inc., Wallingford CT, 2009.
ACS Paragon Plus Environment

Page 25 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
TOC
47x26mm (300 x 300 DPI)
ACS Paragon Plus Environment