Hydroxyl Radicals via Collision-Induced Dissociation of Trimethylammonium Benzyl Alcohols

Article abstract of DOI:10.1071/CH16602

The hydroxyl radical is a well known reactive oxygen species important for interstellar, atmospheric, and combustion chemistry in addition to multiple biochemical processes. Although there are many methods to generate the hydroxyl radical, most of these are inorganic based, with only a few originating from organic precursor molecules. Reported herein is the observation that trimethylammonium benzyl alcohols and their corresponding deuterated isotopologues act as a good source of hydroxyl and deuteroxyl radicals in the gas-phase under collision-induced dissociation (CID) conditions. Attempts to replicate this chemistry in the condensed phase are described.

Full text of DOI:10.1071/CH16602

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem.
Full Paper
http://dx.doi.org/10.1071/CH16602
Hydroxyl Radicals via Collision-Induced Dissociation
of Trimethylammonium Benzyl Alcohols
A
A
B
Peter W. Moore, Jordan P. Hooker, Athanasios Zavras,
B C
,
A
A
George N. Khairallah,
Gina Quach, Evan G. Moore, Richard A. J. O’Hair,
and Craig M. Williams
Elizabeth H. Krenske, Paul V. Bernhardt,
A
A
B D
,
A D
,
A
School of Chemistry and Molecular Biosciences, University of Queensland, Brisbane,
Qld 4072, Australia.
B
School of Chemistry, Bio21 Institute of Molecular Science and Biotechnology,
University of Melbourne, Melbourne, Vic. 3010, Australia.
C
Accurate Mass Scientific P/L, Keilor, Vic. 3036, Australia.
D
Corresponding authors. Email: rohair@unimelb.edu.au; c.williams3@uq.edu.au
The hydroxyl radical is a well known reactive oxygen species important for interstellar, atmospheric, and combustion
chemistry in addition to multiple biochemical processes. Although there are many methods to generate the hydroxyl
radical, most of these are inorganic based, with only a few originating from organic precursor molecules. Reported herein
is the observation that trimethylammonium benzyl alcohols and their corresponding deuterated isotopologues act as a good
source of hydroxyl and deuteroxyl radicals in the gas-phase under collision-induced dissociation (CID) conditions.
Attempts to replicate this chemistry in the condensed phase are described.
Manuscript received: 25 October 2016.
Manuscript accepted: 26 November 2016.
Published online: 9 January 2017.
Introduction
highest homolytic bond dissociation energies (DH298(HO–H) ¼
118.8 kcal mol
ꢁ1 [9] means that alternative methods to H-atom
The hydroxyl radical (HOꢀ) is arguably the most famous of all
small molecule reactive intermediates due to its wide ranging
importance encompassing the galaxy, earth, and society.[1] As
early as 1963, the hydroxyl radical was observed in interstellar
space[2,3] and later as masers that are associated with very young
stars.[4] Beyond these stellar observations, hydroxyl radicals
also play a vital role in the Earth’s atmosphere, essentially
scrubbing the atmosphere of hydrocarbons, halogenated com-
pounds, carbon monoxide, and sulfur dioxide.[5] From a ter-
restrial perspective, the high oxidation potential of the hydroxyl
radical[6] has been exploited in physiological processes,[7] and
made impacts on both human health and disease.[8]
)
abstraction reactions are required. Thus the hydroxyl radical has
been generated from a range of inorganic precusors (e.g. Fenton
chemistry).[1] However, apart from homolysis of alkyl hydro-
peroxides such as tert-butylhydroperoxide,[10] generation of the
hydroxyl radical from organic precursor molecules remains
rare. In the gas phase, treatment of various alkenes with ozone
has been found to deliver hydroxyl radicals on fragmentation of
the primary ozonide 1,[11] whereas exposure of substituted
phenyl radicals (e.g. 2) to oxygen proceeded similarly
(Scheme 1).[12] Interestingly, however, only a single case exists
in the solution phase. Namely, the photolysis of 1-hydroxy-2
(1H)-pyridinethione (3) (Scheme 1),[13] commonly associated
with the Barton decarboxylation reaction,[14] smoothly and
Not surprisingly, methods to detect and measure hydroxyl
radicals have been prolific.[1] The fact that water has one of the
O
O
O
H
HO
O
O
O
O
O
O
O2
ꢀ
R1
H
S
R2
R2
R1
R2
2
1
O
R2
H
H
OH
O
hu
HO
N
Hydroxyl radical
OH
O
3
N
S
Scheme 1.
Journal compilation ꢀ CSIRO 2017
www.publish.csiro.au/journals/ajc

B
P. W. Moore et al.
mildly produces hydroxyl radicals such that 3 has found appli-
cations within biological settings.[15]
Results and Discussion
A survey of the literature revealed that phenyltrimethyl
ammonium salts (e.g. 4) have been reported to undergo N-alkyl
fragmentation to give a PhNR2 radical cation 5 on mass spec-
trometry promoted collision-induced dissociation (CID)
(Scheme 2).[16,17] Competing scission of the N-phenyl bond can
also occur to give a NR3 radical cation 6, but this pathway is a
higher energy process.[16–18] However, closer inspection of
radical cation 5, especially when viewed as the para-resonance
contributor 7, suggested that if para-hydroxymethyl substitution
of the aromatic ring could be accommodated (e.g. 8), a hydroxyl
radical precusor molecule could be engineered (Scheme 2).
Related sequential radical losses that give rise to quinones under
CID conditions have been described.[19]
For proof of principle, trimethylammonium benzyl alcohol
(9) was first investigated. Synthesis of 9 was pedestrian in that
reduction of the commercially available aldehyde 10 gave 11 on
treatment with sodium borohydride, and exposure to methyl
iodide gave 9. The iodide anion could be easily substituted for
either a tetraphenyl borate, giving 12, or a tetrafluoroborate to
give 13 (Scheme 3), all of which provided sufficient quality
crystals for X-ray crystallographic analysis (Fig. 1).
With the view to devising new organic-based hydroxyl
radical precusor molecules for the solution phase, we disclose
herein the results of a gas phase survey of the sequential
fragmentation reactions of a series of para-substituted pyridi-
nium ions; and based on discovering a promising precursor, also
report the results of attempts to translate this chemistry to the
solution phase.
H
ꢁ R
ꢀ
N
ꢀ
ꢀ
R
R
R
R
N
N
5
4
7
R
R
R
ꢁ Ph
OH
ꢀ
R
R
ꢀ
HO
R
N
N
6
R
8
R
R ꢂ Alkyl
Scheme 2.
Iꢁ
OH
OH
O
NaBH4
88 %
MeI
ꢀ
75 %
N
N
N
10
11
9
NaBPh4
NaBF4
77 %
89 %
ꢁBPh4
OH
ꢁBF4
OH
ꢀ
ꢀ
N
N
13
12
Scheme 3.
I1
B2
F4
C10
B1
F1
F3
O1a
C3
O1
C56
C55
C4
C2
F2
C67a
C54
C3 C10
C4
C57
C2
C5
O1
C1
N1
C49
C6
C7
C52
N1
C1
N1
C8
C53
C9
C5
C50
C6
C8
C7
C9
C51
Fig. 1. ORTEP diagrams showing the X-ray crystal structure of ammonium salts 9, 12, and 13 shown with 30 % probability ellipsoids. Structure 12 comprises
two independent formula units in the asymmetric unit. Disorder in the hydroxymethyl groups in both cations was observed, most likely due to the absence of H-
bonding interactions with the tetraphenylborate counter ions. The hydroxymethyl group in 13 formed a bifurcated H-bond to a pair of fluorine atoms from the
tetrafluoroborate anion. The tetrafluoroborate anions and –NMeþ3 groups from an adjacent cation form a double layer. Structure 9 comprised pairs of cations
and anions linked by H-bonds between the hydroxyl group and the iodide ion, which is qualitatively similar to 13 with double layers of iodide ions and –NMe3þ
groups a feature.

Hydroxyl Radicals
C
Electrospray ionisation mass spectrometry (ESI-MS) of 9 in
methanol provided a peak at m/z 166 in the positive ion mode,
corresponding to the benzyl alcohol ammonium ion 14 (Sup-
plementary Material Fig. S1). CID of cation 14 showed loss
of 15 Da (Fig. 2a), relating to a loss of a methyl radical, resulting
in the expected radical cation 15 at m/z 151. Gratifyingly, further
isolation and CID of 15 lead to loss of 17 Da relating to the loss
of a hydroxyl radical (Fig. 2), giving a strong peak at m/z 134
attributed to the aminium ion 16 (Scheme 4). Note: no loss of the
substituted phenyl radical was detected.
The deuterium isotopologue 17 was then investigated to
determine any kinetic isotope effects associated with the N–C
bond cleavage reaction. Synthesis of 17 from 11 was easily
achieved using d3-methyl iodide. Alcohol 17 displayed a molec-
ular ion peak at m/z 169, which upon CID (Supplementary
Material Fig. S2a) gave two major peaks at m/z 151 (ꢁ18 Da),
corresponding to loss of the CD3 radical (i.e. 15), and a peak at
m/z 154 consistent with loss of a CH3 radical to give the radical
cation 18. Subsequent CID of the m/z 154 peak (Supplementary
Material Fig. S2b) resulted in the loss of 17 mass units, again
consistent with the hydroxyl radical and formation of 19 (m/z
137), along with the observation of a hydroxymethylphenyl
cation (20) peak at m/z 107 (Scheme 5). By integrating the area
under the peaks at m/z 154 and 151, and taking into account that
there are two CH3 groups but only one CD3 group in 17, a normal
secondary kinetic isotope effect of kH/kD ¼ 1.19 was deter-
mined.[20] Finally, based on the mass shifts when comparing
the CID spectra of 15 and 18, the ions at m/z 122 and 123 in
Fig. 2b are tentatively assigned as arising via the losses of
HOCH2ꢀ and OCH2.
(a)
151
100
50
166
∗
136
134
0
(b)
100
To understand the chemoselectivity of C–N bond homolysis,
the thermodynamics of various possible C–N dissociation reac-
tions of the phenyltrimethylammonium ion were computed with
CBS-QB3.[21,22] Loss of a methyl radical was found to be
thermodynamically much more favourable than loss of a phenyl
radical, the relative free energies of reactions (DGrel) being 0 and
32.0 kcal molꢁ1, respectively. Loss of a d3-methyl radical was
50
0
122 123
107
100
151
∗
50
75
125
175
200
150
slightly less favourable than loss of a methyl radical (DGrel
¼
þ0.8 kcal molꢁ1). These calculated thermodynamic values
agree with the experimental observations (i.e. ease of dissocia-
tion is Me . CD3 . Ph). In the case of methyl versus phenyl
m/z
Fig. 2. CID mass spectra of (a) 14; (b) 15.
CID
CID
OH
OH
ꢀ
ꢀ
ꢀ
N
N
N
OH•
CH3
16
C9H12Nꢀ
15
C9H13NO•ꢀ
14
C10H16NOꢀ
m/z 166
m/z 151
m/z 134
Scheme 4.
ꢁCD3
ꢁ OH
OH
ꢀ
N
ꢀ
N
16
15
C9H13NO•ꢀ
C9H12Nꢀ, m/z 134
m/z 151
Iꢁ
OH
ꢁ OH
ꢀ
ꢀ
N
N
D3C
17
CD3
19
m/z 169, C10H13D3NOꢀ
ꢁCH3
OH
m/z 137, C9H9D3Nꢀ
ꢀ
N
18
CD3
m/z 154, C9H10D3NO•ꢀ
OH
ꢀ
20
m/z 107, C7H7Oꢀ
Scheme 5.

D
P. W. Moore et al.
O
O
ꢀ
N
N
N
N
ꢁ
I
21
R
R ꢂ CH3 (22)
R ꢂ CD3 (23)
OH
OH
OH
CID
ꢀ
N
ꢀ
N
ꢀ
N
ꢀ
N
N
N
R
ꢁ
I
ꢁ
ꢁ
X
R
Rꢃ
I
R
R ꢂ CH3 (30)
R ꢂ CD3 (31)
R ꢂ Rꢃ ꢂ CH3 (26)
R ꢂ Rꢃ ꢂ CD3 (27)
R ꢂ CH3, X ꢂ I (24)
R ꢂ CD3, X ꢂ I (25)
R ꢂ CH3, Rꢃ ꢂ CD3 (28)
R ꢂ CD3, X ꢂ BPh4 (29)
CID
CID
OH
OH
OH
ꢀ
ꢀ
N
ꢀ
ꢀ
N
ꢀ
N
CID
N
R
N
N
R
Rꢃ
Rꢃ
R
R ꢂ CH3 (32)
R ꢂ CD3 (33)
R ꢂ Rꢃ ꢂ CH3 (37)
R ꢂ Rꢃ ꢂ CD3 (38)
R ꢂ Rꢃ ꢂ CH3 (34)
R ꢂ Rꢃ ꢂ CD3 (35)
R ꢂ CH3, Rꢃ ꢂ CD3 (36)
R ꢂ CH3, Rꢃ ꢂ CD3 (39)
Scheme 6.
removed. In solution, the iodide counter ion could exchange
with other anions, for example, tetraphenylborate (e.g. 29)
(Scheme 6).
O1
C14
C15
C9
C17
C16
C13
C4
Subjecting cations 24 and 25 to CID (Supplementary Mate-
rial Figs S3a and S4a) resulted in loss of either a methyl or
d3-methyl radical to give radical cations 30 and 31, which upon
mass selection and CID underwent loss of hydroxyl radicals to
produce Michler dyes 32 and 33 (Supplementary Material Figs
S3b and S4b). A similar observation was made for CID of the
bis-ammonium cations 26–28 (Supplementary Material Figs
S5–S7) in that radical cations 34–36 were formed on loss of
either methyl or d3-methyl radicals, which was followed by loss
of a hydroxyl radical to give Michler derivatives 37–39
(Scheme 6). In terms of kinetic isotope effects (KIEs), the kH/
kD for 23 was ,1.19, whereas for 25 a value of ,1.22 was
obtained in addition to ,1.20 for 27, and ,1.72 for 28.
Although it is unclear why the latter value is substantially larger
than the others (all KIEs have been statistically corrected for the
number of Me versus CD3 groups), these normal secondary
isotope effects values are consistent with the DFT calculated
ease of dissociation of Me . CD3.
A variation on this theme was explored using cinammyl
alcohol 43, and although loss of a methyl radical was observed
upon CID (i.e. 41, Supplementary Material Fig. S8a), further
fragmentation to give the hydroxyl radical and ammonium ion
42 was not found (Supplementary Material Fig. S8b). Confir-
mation of this lack of reactivity was undertaken by examination
of the deuterated derivative 43 (Supplementary Material Fig.
S9), which was found not to give a molecular ion arising from
loss of a deuteroxyl radical (Scheme 7).
C3
N2
C18
C10
C11
C5
C2
C12
N1
C6
C1
C8
C7
Fig. 3. ORTEP diagram of the X-ray crystal structure for ketone 22, shown
with 30 % probability ellipsoids. In structure 22 the cation bears no H-bond
donors so the iodide ions form an H-bonded chain with a water molecule that
runs parallel to the a-axis (not shown).
dissociation, Wesdemiotis has previously reported that the latter
only occurs above a certain energy threshold.[16]
The scope of hydroxyl radical generation via this system was
further explored through the use of N-alkylated derivatives of
Michler’s ketone 21 (Scheme 6). Initially, it was discovered that
the reduced form of Michler’s ketone (carbinol) was difficult to
handle, as it would undergo reoxidation in air. Therefore, the
mono trimethylammonium ketone 22 (Fig. 3), and correspond-
ing mono d3-methyl derivative (23), were synthesised facilitat-
ing reduction with sodium borohydride to give the
corresponding carbinols 24 and 25. Further N-alkylation with
either methyl or d3-methyl iodide gave the bis-ammonium salts
26–28, although traces of starting material could not be

Hydroxyl Radicals
E
OR
CID
OR
CID
ꢀ
ꢀ
ꢀ
N
N
N
Iꢁ
CH3
42
R ꢂ H (40)
R ꢂ D (43)
41
OH
Scheme 7.
CID
CID
OD
OD
ꢀ
ꢀ
ꢀ
N
N
N
OD
CH3
45
16
C9H12Nꢀ
44
C9H12DNO•ꢀ
C
10H15DNOꢀ
m/z 152
m/z 167
m/z 134
CD3OD
OH
Iꢁ
ꢀ
N
9
Scheme 8.
OH
hu
Iꢁ
OH
O
ꢀ
ꢀ
N
9
46
47
OH
hu
O
ꢀ
O
I
I
ꢀ
N
ꢁ
O
49
48
CF3
H
Scheme 9.
Since the trimethylammonium benzyl alcohol system
provided a convenient source of hydroxyl radicals (i.e. 17,
24, and 26–28) analogous formation of the deuteroxyl radical
was pursued. The simplest approach found was to treat 9 with
d4-methanol for 5 min to facilitate H/D exchange. Analysis of
this mixture via mass spectrometry gave rise to a molecular
ion corresponding to 44, which on CID (Supplementary
Material Fig. S10a) gave radical cation 45 (m/z 152). Subse-
quent fragmentation (Supplementary Material Fig. S10b)
pleasingly led to loss of the deuteroxyl radical, and a peak
at m/z 134 for the previously observed aminium ion 16
(Scheme 8).
Having established a method for the generation of both
hydroxyl and deuteroxyl radicals in the gas phase under CID
conditions, it was important to determine whether the observed
gas phase results could be translated directly into the solution
phase. Gas phase mass spectroscopic observations being suc-
cessfully developed into synthetically useful reactions are not
unprecedented. Prager noted the characteristic loss of carbon
dioxide (CO2) in the mass spectra for a series of isoxazo-
lones,[23] that subsequently gave rise to synthetically viable
heterocycles, including imidazoles, pyrimidines, oxazoles, and
thiazoles, using both photolysis and pyrolysis.[24] Such inspira-
tion provided the motivation to subject 9 to photochemical
studies. Irradiation of 9 at 300 nm in acetonitrile (Rayonet
photochemical reactor) over an extended period led only to
minor amounts of benzyl alcohol (46) and benzaldehyde (47)
being detected (Scheme 9), as determined by gas chromatogra-
phy–mass spectrometry (GCMS) analysis, which increased in
conversion on irradiation at 254 nm. These results indicated that
the preferred reaction pathway involved radical cleavage of the
CAr–N bond, which has been previously reported for phenyl-
trimethylammonium salts containing an iodide counter
ion.[25,26] Irradiation of 12 at 254 nm also led to decomposition,
giving biphenyl derived from the tetraphenylborate counter ion,
whereas 13 (tetrafluoroborate counter ion) was stable under
photolytic conditions. These results supported the notion that the
iodide mediated direct charge transfer excitation pathway fol-
lowed by homolytic cleavage was preferred.[26] In an attempt to
detect hydroxyl radical formation even at trace levels (i.e. as a
minor pathway), nanosecond laser flash photolysis using 9 was
explored, with the rationale that even if only minute amounts of

F
P. W. Moore et al.
hydroxyl radical were to form, detection of the iminium by-
product 16 should be possible using this technique, since it has a
characteristic UV-visible absorption band at 415 nm.[27] How-
ever, our efforts to observe this species as a transient intermedi-
ate were unsuccessful, with obtained spectra instead revealing
only the prompt excitation and then decay of the initially excited
singlet state, with a lifetime of ,9.8 ꢂ 1 ns (Supplementary
Material Fig. S11) suggesting that the photolysis solely pro-
motes CAr–N bond cleavage. (Note: UV-vis action spectroscopy
of the isolated pyridinium ions in the gas-phase[28] would be a
worthwhile pursuit to examine the fundamental photodissocia-
tion chemistry in the absence of counter ions and solvent
effects.)
Exposing 9 to thermal conditions (heating or microwave
irradiation) led to rapid dequaternisation as previously observed
with similar quaternary ammonium salts,[29] whereas 12 led
predominantly to the formation of biphenyl due to the known
sensitivity of the tetraphenylborate anion.[30] The tetrafluoro-
borate 13, however, was perfectly stable after heating at 1008C
for 18 h. Not to be perturbed by these disappointing results,
photolysis was further explored in pursuit of solution phase
hydroxyl radical generation. This motivation was driven by the
calculations that indicate the N-methyl bond is the preferred
bond breakage pathway, and that nickel catalysed cross coupling
chemistry involving phenyltrimethylammonium salts has
shown that either phenyl[31] or methyl[32] bond breakage can
occur depending on the selected conditions. Cyclic voltammetry
of 13 showed no electrochemical reactions, while cyclic vol-
tammetry of 9 showed two reversible processes associated with
the iodide anion not the ammonium cation.
In the view that quinoline has previously been reported to act
as a suitable methyl radical trap,[33] and that the trifluoroacetate
salt of quinoline is known to improve selectivity in photochemi-
cally promoted radical reactions[34] these conditions were
investigated in the context of our system. On irradiation of 9
in the presence of one equivalent of quinoline and trifluoroacetic
acid, GCMS analysis indicated the formation of p-iodobenzyl
alcohol (48) and p-iodobenzaldehyde (49) (Scheme 9). p-Iodo-
benzaldehyde (49) was also detected by GCMS when exploring
laser flash photolysis of 9 (i.e. irradiation for 48 h using an
amplified laser system delivering 250 nm (,1 mW)). Although,
this type of aryl iodination has not been observed previously,
and will be further explored in due course as a synthetic method,
it demonstrated that CAr–N bond cleavage could not be avoided.
CaH2 still. Anhydrous CH3CN was distilled over CaH2.
Reagents were purchased from Sigma Aldrich and were used
directly and without further purification. NMR spectra were
recorded using either a Bruker AV300 (300 MHz, 75 MHz),
AV400 (400 MHz, 375 MHz, 100 MHz, 60 MHz), or AV500
(500 MHz, 125 MHz) instrument, with all data being processed
with MestreNova ver. 9.1.0. Chemical shifts are given in parts
per million (ppm) and referenced according to solvent; CDCl3
(dH 7.26, dC 77.0), D2O (dH 4.80) with 1,4-dioxane reference (dC
66.7) and d6-DMSO (dH 2.50; dC 39.5). Coupling constants (J)
are given in Hz. Infrared spectra were measured using a Perkin
Elmer FT-IR spectrometer (Spectrum 2000). Microanalyses
were performed by the University of Queensland Microanalyt-
ical Service. Melting points were recorded using a Digimelt
MPA161 SRS apparatus.
Mass Spectrometry (MS)
For characterisation, high resolution electrospray ionisation
(HRESIMS) accurate mass measurements were recorded in
positive or negative mode on a Bruker MicroOTOF-Q (quad-
rupole-time of flight) instrument equipped with a Bruker ESI
source using sodium formate as a reference calibrant (University
of Queensland). GCMS was recorded using a Shimadzu GCMS-
QP5000 machine with auto sampler and analysed using
GCMSsolution v1.20 software at the University of Queensland.
Tandem-mass and high resolution mass spectrometry experi-
ments were conducted at the University of Melbourne (Bio21
Institute) on a Thermo Scientific LTQ FT hybrid mass spec-
trometer equipped with a Finnigan ESI source. The instrument
consists of a linear ion trap (LTQ) coupled to a Fourier-
transform ion cyclotron resonance (FT-ICR) mass spectrometer.
Ions generated through ESI were mass selected with a window
width of 1 m/z and subjected to low-energy CID with the helium
bath gas. A normalised collision energy of between 15 and 30 %,
a Q value of 0.25, and a reaction time of 30 ms were used in these
CID experiments. Accurate mass measurements of precursor
and/or product ions were performed by transferring the ions
from the linear ion trap to the FT-ICR cell for high resolution
mass spectrometry. The instrument was calibrated immediately
before commencing all experiments.
Experimental Procedures
[4-(N,N-Dimethylamino)phenyl]methanol (11)
4-(N,N-Dimethylamino)benzaldehyde (10, 504 mg, 3.38
mmol) was dissolved in methanol (10.0 mL) and cooled to 08C
in an ice bath. Sodium borohydride (194 mg, 5.12 mmol) was
added, and the reaction stirred at 08C before being allowed to
warm to room temperature over 30 min. The reaction mixture
was then quenched with water (10.0 mL) and stirred for 10 min
before being concentrated to half volume. The reaction mixture
was extracted with diethyl ether (4 ꢃ 20 mL) and the organic
fractions were combined, washed with brine (30 mL), dried,
filtered, and concentrated, yielding 11 (451 mg, 88 %) as a
viscous green oil. The data matched that reported by Jiang
et al.[36] dH (300 MHz, CDCl3) 1.58 (br s, 1H), 2.95 (s, 6H), 4.57
(s, 2H), 6.74 (d, J 9.0, 2H), 7.25 (d, J 9.0, 2H). dC (100 MHz,
CDCl3) 41.2, 65.9, 113.2, 129.2, 129.5, 150.9. m/z (GCMS) 151
(27 %, Mþ), 134 (47, [M – HO]þ), 44 (100, [M – C7H7O]þ).
Conclusion
Trimethylammonium benzyl alcohols have been identified as
suitable precursor molecules for the generation of both hydroxyl
and deuteroxyl radicals under collision-induced dissociation
mass spectrometry conditions in the gas phase. The generation
of hydroxyl radicals from organic based precursor molecules is
rare, making this contribution one of very few methods available
to those studying this reactive intermediate. Unfortunately, all
attempts at translating our gas phase results to the solution phase
have been unsuccessful, and thus further efforts are required to
develop viable precursor molecules for solution phase genera-
tion of hydroxyl radicals.
Experimental
4-(Hydroxymethyl)-N,N,N-trimethylbenzenaminium
Iodide (9)
General Experimental
All reagents and solvents were purified before use according to
literature methods.[35] Anhydrous CH2Cl2 was distilled from a
Alcohol 11 (204 mg, 1.35 mmol) was dissolved in methanol
(1.0 mL) followed by dropwise addition of methyl iodide

Hydroxyl Radicals
G
(160 mL, 2.57 mmol). The reaction mixture was stirred at room
temperature for 18 h. The mixture was filtered, washed with
diethyl ether, and dried under vacuum, yielding 9 (298 mg,
1.02 mmol, 75 %) as a white solid. Mp 218.08C (dec.). dH
(300 MHz, d6-DMSO) 3.59 (s, 9H), 4.57 (d, J 6.0, 2H), 5.39
(t, J 6.0, 1H), 7.54 (d, J 9.0, 2H), 7.91 (d, J 9.0, 2H). dC
(100 MHz, d6-DMSO) 56.4, 61.7, 120.1, 127.4, 144.6, 145.7.
HRMS m/z 166.1224, calcd for C10H16NOþ [M – I]þ 166.1226.
nmax /cmꢁ1 712, 813, 846, 931, 950, 1014, 1060, 1122, 1248,
1334, 1396, 1412, 1466, 1483, 1512, 2877, 3012, 3338. Anal.
Calc. for C10H16INO: C 40.97, H 5.50, N 4.78. Found: C 40.97,
H 5.55, N 4.67 %.
HRMS m/z 169.1417, calcd for C10H13D3NOþ [M – I]þ
169.1415. nmax /cmꢁ1 575, 677, 764, 821, 844, 931, 1124,
1193, 1367, 1454, 1527, 1547, 1600, 1629, 2261, 2823, 2902,
3013.
4-[4-(N,N-Dimethylamino)-benzoyl]-N,N,N-
trimethylbenzenaminium Iodide (22)
Michler’s ketone 21 (1.00 g, 3.74 mmol) and methyl iodide
(4.0 mL, 64.3 mmol) were added to a pressure tube (with a
Q-Tube pressure release valve set at 120 psi) and the reaction
heated in an oil bath at 808C for 16 h (Caution: use blast shield).
On cooling to room temperature the solid was partitioned
between dichloromethane (100 mL) and water (100 mL). The
aqueous layer was removed and the organic layer washed with
water (3 ꢃ 100 mL). The aqueous layers were combined and
evaporated under vacuum, providing 22 (1.49 g, 3.64 mmol,
97 %) as a yellow-green solid. Mp 168.3–169.08C. dH
(300 MHz, d6-DMSO) 3.05 (s, 6H), 3.68 (s, 9H), 6.79 (d, J
9.0, 2H), 7.64 (d, J 9.0, 2H), 7.83 (d, J 9.0, 2H), 8.12 (d, J 9.0,
2H). dC (100 MHz, d6-DMSO) 36.6, 56.4, 110.8, 120.7, 122.8,
130.1, 132.2, 140.5, 148.6, 153.6, 191.9. HRMS m/z 283.1801,
calcd for C18H23N2O [M – Iꢁ]þ 283.1805. nmax /cmꢁ1 579, 678,
765, 821, 842, 930, 1065, 1116, 1154, 1192, 1231, 1294, 1323,
1366, 1454, 1466, 1482, 1526, 1547, 1600, 1629, 2901, 3013.
Anal. Calc. for C18H23IN2O: C 52.69, H 5.65, N 6.83. Found: C
52.44, H 5.67, N 6.70 %.
4-(Hydroxymethyl)-N,N,N-trimethylbenzenaminium
Tetraphenylborate (12)
Alcohol 9 (249 mg, 0.85 mmol) was dissolved in water
(4.0 mL). Sodium tetraphenylborate (346 mg, 1.01 mmol) was
separately dissolved in water (4.0 mL), and the two solutions
combined. The precipitate was then filtered, washed with water,
and dried under vacuum providing 12 (367 mg, 89 %) as a white
powder. The powder was crystallised from acetone/water
(50 : 50). Mp 202.28C (dec.). dH (300 MHz, d6-DMSO) 3.57
(s, 9H), 4.57 (d, J 6.0, 2H), 5.40 (t, J 6.0, 1H), 6.79 (t, J 6.0, 4H,),
6.93 (t, J 7.5, 8H), 7.15–7.21 (m, 8H), 7.54 (d, J 9.0, 2H), 7.90 (d,
J 9.0, 2H). dC (75 MHz, d6-DMSO) 56.8, 62.2, 120.5, 121.9,
125.7 (q, JC-B 2.8), 127.9, 136.0 (d, JC-B 1.2), 145.1, 146.1, 163.8
(q, JC-B 49.0). HRMS m/z 166.1232, calcd for C10H16NOþ
[M – C24H20B]þ 166.1226; 319.1662, calcd for C24H20Bꢁ [M –
C10H16NOþ]ꢁ 319.1664. nmax /cmꢁ1 584, 741, 754, 810, 893,
908, 1050, 1133, 1152, 1276, 1441, 1491, 1501, 1565, 2979,
3008, 3055, 3525. Anal. Calc. for C20H24INO4: C 84.12, H 7.47,
N 2.89. Found C 83.18, H 7.48, N 2.96 %.
4-[4-(N,N-Dimethylamino)benzoyl]-N,N-dimethyl-N-
(methyl-d3)benzenaminium Iodide (23)
Michler’s ketone 21 (503 mg, 1.88 mmol) and d3-iodo-
methane (2.0 mL, 32 mmol) were added to a pressure tube (with
a Q-Tube pressure release valve set at 120 psi), and the reaction
mixture heated to 808C for 16 h (Caution: use blast shield). On
cooling to room temperature the solid was partitioned between
dichloromethane (150 mL) and water (150 mL). The aqueous
layer was removed, and the organic layer washed with water
(3 ꢃ 50 mL). The aqueous layers were combined and evaporated
under vacuum, yielding 23 (690 mg, 89 %) as a yellow-green
solid. Mp 172.2 – 173.08C. dH (400 MHz, d6-DMSO) 3.05
(s, 6H), 3.69 (s, 6H), 6.79 (d, J 9.0, 2H), 7.64 (d, J 9.2, 2H),
7.82 (d, J 9.2, 2H), 8.13 (d, J 9.0, 2H). dD (60 MHz, H2O) 3.64
(s, CD3), 4.81 (HOD). dC (100 MHz, d6-DMSO) 39.6, 56.3,
110.8, 120.7, 122.7, 130.0, 132.2, 140.4, 148.6, 153.5, 191.9.
HRMS m/z 286.1991, calcd for C18H20N2Oþ [M – I]þ 286.1993.
nmax /cmꢁ1 575, 677, 764, 822, 844, 931, 1124, 1154, 1193,
1294, 1324, 1527, 1547, 1600, 1629, 2261, 2826, 2903, 3013.
4-(Hydroxymethyl)-N,N,N-trimethylbenzenaminium
Tetrafluoroborate (13)
Alcohol 9 (947 mg, 3.23 mmol) was dissolved in water
(20.0 mL). Silver tetrafluoroborate (666 mg, 3.42 mmol) was
separately dissolved in water (10.0 mL) and then added to the
above solution. The reaction mixture was then filtered, and the
filtrate concentrated. The resulting solid was washed with ethyl
acetate (15.0 mL) and dried under vacuum, yielding 13 (629 mg,
77 %) as a fine, pale blue solid. Mp 177.58C (dec.). dH
(400 MHz, d6-DMSO) 3.58 (s, 9H), 4.57 (d, J 5.6, 2H), 5.39
(t, J 5.6, 1H), 7.53 (d, J 9.0), 7.90 (d, J 9.0). dC (100 MHz,
d6-DMSO) 56.4, 61.7, 120.0, 127.5, 144.7, 145.7. dF (375 MHz,
d6-DMSO) 148.83 (d, J 3.75), 148.88 (d, J 3.75). HRMS m/z
166.1229, calcd for C10H16NOþ [M – BF4]þ 166.1226. nmax
/cmꢁ1 705, 827, 848, 939, 956, 1030, 1218, 1415, 1475, 1494,
1602, 1701, 3017, 3051, 3557. Anal. Calc. for C20H24INO4: C
47.47, H 6.37, N 5.54. Found C 47.08, H 6.38, N 5.43 %.
4,4’-(Hydroxymethylene)bis(N,N,N-
trimethylbenzenaminium) Iodide (26)
Ketone 22 (202 mg, 0.49 mmol) was dissolved in methanol
(50 mL) and cooled to 08C before addition of sodium borohy-
dride (70 mg, 1.84 mmol). The reaction was stirred at 08C for
5 min, before being allowed to warm to room temperature and
heated to reflux for 30 min. The reaction mixture was cooled to
room temperature, and then to 08C, at which point water
(3.0 mL) was added. The mixture was concentrated to dryness,
providing a crude mixture of 24 that was dissolved in methanol
(5.0 mL) and treated dropwise with iodomethane (0.24 mL,
3.86 mmol). The reaction was stirred at room temperature for
18 h, before being filtered, washed with diethyl ether (10 mL),
and dried under vacuum to yield 26 (221 mg, 82 % over two
steps) as a white solid containing ,1 % impurity of 22. dH
4-(Hydroxymethyl)-N,N-dimethyl-N-(methyl-d3)
benzenaminium Iodide (17)
Alcohol 11 (428 mg, 2.83 mmol) was dissolved in methanol
(2.0 mL). d3-iodomethane (0.35 mL, 5.6 mmol) was added drop-
wise to the mixture and then stirred at room temperature for 20 h.
The resulting precipitate was filtered, washed with diethyl ether
(10.0 mL), and dried under vacuum, yielding 17 (715 mg, 85 %)
as a white powder. Mp 219.78C (dec.). dH (400 MHz, d6-DMSO)
3.60 (s, 6H), 4.57 (d, J 6.0, 2H), 5.38 (t, J 5.6, 1H), 7.54 (d, J 9.0,
2H), 7.92 (d, J 9.0, 2H). dD (60 MHz, H2O) 3.42 (s). dC
(100 MHz, d6-DMSO) 56.4, 61.7, 120.1, 127.5, 144.7, 145.7.

H
P. W. Moore et al.
(300 MHz, d6-DMSO) 3.58 (s, 18H), 5.94 (s, 1H), 6.34 (s, 1H),
7.66 (d, J 9.0, 4H), 7.93 (d, J 9.0, 4H). dC (75 MHz, d6-DMSO)
56.4, 72.1, 120.5, 127.3, 145.9, 146.9. HRMS m/z 150.1099,
calcd for C19H28N2O2þ [M – 2Iꢁ]2þ 150.1095. nmax /cmꢁ1 592,
693, 813, 845, 933, 951, 1014, 1056, 1120, 1230, 1396, 1465,
1486, 1509, 1616, 1738, 3017, 3416(br).
dissolved in water (4.0 mL). Sodium tetraphenylborate
(86.3 mg, 0.25 mmol) was separately dissolved in water
(4.0 mL) and the solution added dropwise to the reaction
mixtiure. After addition the mixture was filtered, washed with
water (10.0 mL), and dried under vacuum to yield 29 (95 mg,
63 %) as a white solid. Mp 80.58C (dec.). dH (300 MHz,
d6-DMSO) 2.84 (s, 6H), 3.54 (s, 6H), 5.68 (s, 1H), 5.86 (s,
1H), 6.66 (d, J 8.9, 2H), 6.79 (t, J 7.1, 4H), 6.92 (t, J 7.5, 8H),
7.15–7.21 (m, 10H), 7.56 (d, J 8.8, 2H), 7.85 (d, J 9.1, 2H). dC
(75 MHz, d6-DMSO) 40.2, 56.3, 73.1, 120.0, 121.5, 125.3 (q, JC-
4,4’-(Hydroxymethylene)bis(N,N-dimethyl-N-(methyl-
d3)benzenaminium) Iodide (27)
Ketone 23 (204 mg, 0.49 mmol) was dissolved in methanol
(50 mL) and cooled to 08C before addition of sodium borohy-
dride (62 mg, 1.6 mmol). The reaction was stirred at 08C for
5 min before being allowed to warm to room temperature and
heated to reflux for 30 min. The reaction mixture was cooled to
room temperature, and then to 08C, at which point water
(3.0 mL) was added. The mixture was then concentrated to
dryness, providing a crude mixture of 25, that was dissolved
in methanol (3.0 mL) and treated dropwise with d3-iodomethane
(0.24 mL, 3.9 mmol). The reaction was stirred at room tempera-
ture for 18 h, before being filtered, washed with diethyl ether
(10.0 mL), and dried under vacuum to yield 26 (116 mg, 42 %
over two steps) as a white solid containing ,1 % impurity of 25.
dH (300 MHz, d6-DMSO) 3.57 (s, 12H), 5.93 (s, 1H), 7.66 (d, J
9.0, 4H), 7.91 (d, J 9.1, 4H). dD (60 MHz, H2O) 3.57 (s). dC
(75 MHz, d6-DMSO) 56.3, 72.1, 120.4, 127.3, 145.9, 146.9.
HRMS m/z 153.12840, calcd for C19H22D6N2O2þ [M – 2I]2þ
153.12836. nmax /cmꢁ1 588, 827, 984, 1014, 1055, 1124, 1180,
1231, 1396, 1508, 1615, 3015, 3057, 3415.
2.8), 127.0, 127.1, 132.7, 135.5 (d, JC-B 1.2), 145.3, 148.4,
B
149.6, 163.3 (q, JC-B 49.0). HRMS m/z 286.21483, calcd for
C19H22D3N2Oþ [M – C24H20B]þ 286.21497. nmax /cmꢁ1 576,
612, 677, 706, 734, 822, 845, 932, 1066, 1125, 1155, 1193, 1295,
1325, 1479, 1524, 1602, 2264, 2908, 3010, 3060, 3521(br).
(E)-4-(3-Hydroxyprop-1-en-1-yl)-N,N,N-
trimethylbenzenaminium Iodide (40)
4-(N,N-Dimethylamino)benzaldehyde (503 mg, 3.37 mmol)
was dissolved in dichloromethane (20 mL) at room temperature
followed by the addition of (carbethoxymethylene)triphenyl-
phosphorane (199, 1.410 g, 4.05 mmol). The reaction was heat-
ed to reflux for 24 h before the reaction was allowed to cool to
room temperature, and concentrated under reduced pressure to
provide the desired product as a crude mixture with triphenyl-
phosphine oxide. This mixture was purified by column chroma-
tography (1 : 2 ethyl acetate/petroleum spirt) affording ethyl
(E)-3-(4-(N,N-dimethylamino)phenyl)acrylate (517 mg, 70 %)
as a 50 : 1 mixture of E : Z isomers. Data matched that from
Diaz[37] dH (300 MHz, CDCl3) 1.33 (t, J 7.1, 3H), 3.02 (s, 6H),
4.24 (q, J 7.1, 2H), 6.22 (d, J 15.8, 1H), 6.71–6.62 (m, 2H), 7.47–
7.37 (m, 2H), 7.62 (d, J 15.9, 1H). m/z (GCMS) 219 (100 %,
Mþ), 190 (25, [M–C2H5]þ), 174 (62, [M–C2H5O]þ), 146 (43,
[M–C3H5O2]þ).
4-{4-[N,N-Dimethyl(N-methyl-d3)ammonio]phenyl}
(hydroxymethyl)-N,N,N-trimethylbenzenaminium
Iodide (28)
Ketone 23 (200 mg, 0.48 mmol) was dissolved in methanol
(50 mL) and cooled to 08C before addition of sodium borohy-
dride (63 mg, 1.7 mmol). The reaction was stirred at 08C for
5 min before being allowed to warm to room temperature and
heated to reflux for 30 min. The reaction mixture was cooled to
room temperature, and then to 08C at which point water (3.0 mL)
was added. The mixture was then concentrated to dryness
providing a crude mixture of 25 that was dissolved in methanol
(3.0 mL) and treated dropwise with iodomethane (0.24 mL,
3.9 mmol). The reaction was stirred at room temperature for
18 h, before being filtered, washed with diethyl ether (10.0 mL),
and dried under vacuum to yield 28 (95 mg, 36 % over two steps)
as a white solid containing ,1 % impurity of 25. dH (400 MHz,
d6-DMSO) 3.567 (s, 6H), 3.573 (s, 9H), 5.93 (s, 1H), 6.33 (s,
1H), 7.66 (d, J 9.0, 4H), 7.91 (d, J 9.0, 4H). dD (60 MHz, H2O)
3.57 (s). dC (100 MHz, d6-DMSO) 56.3, 56.4, 72.1, 120.4, 127.3,
145.8, 146.9. HRMS m/z 151.61899, calcd for C19H25D3N2O2þ
[M – 2I]2þ 151.61895. nmax /cmꢁ1 588, 688, 846, 932, 1014,
1055, 1122, 1396, 1508, 1618, 3016, 3415.
Following the method of Brunhofer et al.[38] ethyl (E)-3-(4-
(N,N-dimethylamino)phenyl)acrylate (305 mg, 1.39 mmol) was
disolved in anhydrous toluene (10.0 mL) under an argon atmo-
sphere and cooled to ꢁ788C. DIBAL-H (3.50 mL, 3.50 mmol)
was added dropwise and the reaction stirred at ꢁ788C for 3 h
before being warmed to 08C and then quenched with sodium
potassium tartrate (5.0 mL). The reaction mixture was extracted
with dichloromethane (3 ꢃ 10 mL), passed through a silica plug,
and then concentrated to afford (E)-3-(4-(N,N-dimethylamino)
phenyl)prop-2-en-1-ol (230 mg, 93 %). Data matched that as
previously reported.[38] dH (300 MHz, CDCl3) 1.32 (t, J 5.9, 1H),
2.96 (s, 6H), 4.28 (td, J 6.0, 1.3, 2H), 6.18 (td, J 15.8, 6.2, 1H),
6.52(d, J 15.8, 1H), 6.73–6.64 (m, 2H), 7.31–7.26 (m, 2H). m/z
(GCMS) 177 (50 %, Mþ), 160 (31, [M – OH]þ), 146 (23, [M –
CH3O]þ), 134 (62, [M – C2H5N]þ).
(E)-3-(4-(N,N-Dimethylamino)phenyl)prop-2-en-1-ol (102 mg,
0.58 mmol) was dissolved in methanol (1.0 mL) followed by
dropwise addition of methyl iodide (140 mL, 2.30 mmol). The
reaction was stirred at room temperature for 18 h after which
time the solution was filtered, washed with ether, and dried
under high vacuum to afford the titled compound (40) as a white
solid (180 mg, 98 %). Mp 178.58C (dec.). dH (500 MHz, d6-
DMSO) 3.59 (s, 9H), 4.20–4.09 (m, 2H), 4.97 (t, J 5.4, 1H), 6.57
(dt, J 16.0, 4.4, 1H), 6.64 (d, J 16.1, 1H), 7.67 (d, J 9.0, 2H), 7.89
(d, J 9.0, 2H). dC (125 MHz, d6-DMSO) 56.4, 61.2, 120.7, 126.0,
127.2, 134.0, 138.6, 145.7. nmax /cmꢁ1 720, 783, 836, 853, 868,
932, 951, 969, 1014, 1098, 1113, 1232, 1299, 1318, 1396, 1415,
1460, 1485, 1511, 2849, 3014, 3053, 3371. HRMS m/z
192.1377, calcd for C7H10NOþ [M – I]þ 192.1383. Anal. Calc.
4-{4-[N,N-Dimethylamino)phenyl]}(hydroxymethyl)-N,
N,N-trimethylbenzenaminium tetraphenylborate (29)
Ketone 23 (103 mg, 0.25 mmol) was dissolved in methanol
(25 mL) and cooled to 08C before addition of sodium borohy-
dride (31 mg, 0.81 mmol). The reaction was stirred at 08C for
5 min before being allowed to warm to room temperature and
then heated to reflux for 1 h. The mixture was cooled to room
temperature and cooled to 08C at which point water (2.0 mL)
was added. The reaction mixture was then concentrated and

Hydroxyl Radicals
I
for C18H23IN2O: C 45.16, H 5.68, N 4.39. Found: C 45.23,
H 5.69, N 4.38 %.
and diluted to have an absorbance at 250 nm of ,0.8 over the
2 mm pathlength cell used. Samples were continuously stirred
mechanically, and were changed every 5–10 min to avoid sig-
nificant decomposition of the parent compound. Resulting
traces of the time-resolved DOD data were analysed globally
using commercially available software (Igor, version 6.1.2.1,
Wavemetrics).
Crystallography
Single crystal intensity data were collected on an Oxford Dif-
fraction Gemini S Ultra CCD diffractometer using Mo-Ka or
Cu-Ka radiation operating in the v-scan mode. Data reduction
and empirical absorption corrections were performed with the
CrysAlis program. All other computations were performed
within the WinGX suite of programs.[39] Structures were solved
by direct methods with SHELXS and refined by full matrix least-
squares analysis with SHELXL97.[40] All non-H atoms were
refined with anisotropic thermal parameters, and H-atoms were
constrained at estimated positions using a riding model.
Disorder in the hydroxymethyl groups in the structure of 12
was modelled by resolution of alternate positions for the
hydroxyl groups and also the methylene protons, which were
refined with complementary occupancies. The atomic nomen-
clature is defined in Figs 1 and 2, drawn with ORTEP3.[41]
Crystallographic data in CIF format are available from the
Cambridge Crystallographic Data Centre with deposition num-
bers 1508906–1508909.
Supplementary Material
Copies of mass spectra, CBS-QB3 calculation data, copies of 1H
and 13C NMR spectra, and flash laser photolysis spectra are
available on the Journal’s website.
Acknowledgements
The authors thank Mr Timothy J. Zerk for assistance with cyclic voltam-
metry. Financial support by the University of Queensland, the University of
Melbourne, and the Australian Research Council (ARC) is gratefully
acknowledged. E.H.K. (FT120100632), C.M.W. (FT110100851) and
E.G.M. (FT100100795) gratefully acknowledge the ARC for Future
Fellowship awards. P.W.M. is grateful for an Australian Postgraduate
Award. Nanosecond Flash Photolysis measurements were undertaken at the
Photochemistry and Ultrafast Laser Spectroscopy (PULSE) facility, School
of Chemistry and Molecular Biosciences, established with financial support
by the University of Queensland (MEI-2013000106). We thank the National
Computational Infrastructure National Facility for computer time.
Compound 9: C10H16INO: M 293.14, temperature 190 K,
monoclinic, space group P21/c, a 12.5141(5), b 8.1503(4),
3
c 11.4731(6) A, b 95.449(5)8, V 1164.9(1) A , Z 4, F(000)
˚
˚
ꢁ1
576, Dc 1.671 g cmꢁ3, l (Mo-Ka) 0.71073 A, m 2.715 mm
,
˚
2049 unique data (Rint 0.0423, 2ymax 508), R1 0.0330 (for 1609
observed reflections I . 2s(I)), wR2 0.0896 (all data), CCDC
1508909.
References
[1] S. Gligorovski, R. Strekowski, S. Barbati, D. Vione, Chem. Rev. 2015,
115, 13051. doi:10.1021/CR500310B
Compound 12: C34H36BNO: M 485.45, temperature 190 K,
ꢀ
triclinic, space group P1, a 10.7741(4), b 10.8770(5), c 27.031
[2] N. H. Dieter, H. I. Ewen, Nature 1964, 201, 279. doi:10.1038/
201279B0
[3] J. W. V. Storey, D. M. Watson, C. H. Townes, Astrophys. J. 1981, 244,
L27. doi:10.1086/183472
[4] A. L. Argon, M. J. Reid, K. M. Menten, Astrophys. J. 2003, 593, 925.
doi:10.1086/376592
[5] C. N. Hewitt, R. M. Harrison, Atmos. Environ. 1985, 19, 545.
doi:10.1016/0004-6981(85)90033-2
3
˚
˚
(1) A, a 80.401(4)8, b 88.944(3)8, g 60.949(5)8, V 2723.3(2) A ,
Z 4, F(000) 1040, Dc 1.184 g cmꢁ3, l (Mo-Ka) 0.71073 A, m
˚
0.529 mmꢁ1, 8619 unique data (Rint 0.0364, 2ymax 508), R1
0.0601 (for 6872 observed reflections I . 2s(I)), wR2 0.1848
(all data), CCDC 1508906.
Compound 13: C10H16BF4NO: M 253.05, temperature
190 K, monoclinic, space group P21/c, a 12.8146(9), b 8.5381
[6] A. A. Fokin, P. R. Schreiner, Chem. Rev. 2002, 102, 1551. doi:10.1021/
CR000453M
[7] See, for example, S.-x. Chen, P. Schopfer, Eur. J. Biochem. 1999, 260,
3
(4), c 11.1871(8) A, b 105.971(8)8, V 1176.8(1) A , Z 4, F(000)
˚
˚
ꢁ1
528, Dc 1.428 g cmꢁ3, l (Mo-Ka) 0.71073 A, m 0.132 mm
,
˚
726. doi:10.1046/J.1432-1327.1999.00199.X
2069 unique data (Rint 0.0285, 2ymax 508), R1 0.0609 (for 1545
observed reflections I . 2s(I)), wR2 0.1752 (all data), CCDC
1508907.
[8] B. Lipinski, Oxid. Med. Cell. Longev. 2011, 2011, 809696.
[9] S. J. Blanksby, G. B. Ellison, Acc. Chem. Res. 2003, 36, 255.
doi:10.1021/AR020230D
[10] (a) J. Sanchez, T. N. Myers, Peroxides and Peroxide Compounds,
Organic Peroxides, in Kirk-Othmer Encyclopedia of Chemical Tech-
nology 2000 (John Wiley: Hoboken, NJ).
Compound 22: C18H25IN2O2: M 428.3, temperature 190 K,
ꢀ
triclinic, space group P1, a 6.3038(3), b 9.1178(5), c 17.4003(9)
3
˚
˚
˚
A, a 102.932(5)8, b 92.472(4)8, g 103.450(4)8, V 943.22(8) A , Z
(b) S. Hsieh, R. Vushe, Y. T. Tun, J. L. Vallejo, Chem. Phys. Lett. 2014,
591, 99. doi:10.1016/J.CPLETT.2013.11.008
2, F(000) 432, Dc 1.508 g cmꢁ3, l (Cu-Ka) 1.54184 A, m
13.42 mmꢁ1, 2958 unique data (Rint 0.0486, 2ymax 1258), R1
0.0433 (for 2848 observed reflections I . 2s(I)), wR2 0.1167
(all data), CCDC 1508908.
[11] (a) R. Gutbrod, R. N. Schindler, E. Kraka, D. Cremer, Chem. Phys.
Lett. 1996, 252, 221. doi:10.1016/0009-2614(96)00126-1
(b) S. M. Aschmann, J. Arey, R. Atkinson, Atmos. Environ. 2002, 36,
4347 and references therein. doi:10.1016/S1352-2310(02)00355-2
(c) R. Atkinson, Atmos. Chem. Phys. 2003, 3, 2233. doi:10.5194/ACP-
3-2233-2003
[12] (a) M. B. Prendergast, P. A. Cooper, B. B. Kirk, G. da Silva, S. J.
Blanksby, A. J. Trevitt, Phys. Chem. Chem. Phys. 2013, 15, 20577.
doi:10.1039/C3CP53690D
Nanosecond Flash Photolysis
The excitation source utilised for nanosecond flash photolysis
was an amplified laser system (Spitfire ACE, Spectra Physics)
coupled to an OPA system (Topas Prime, Light Conversion)
delivering tunable excitation pulses at 250 nm, and a 1 kHz
repetition rate. Ground and excited state difference spectra
(DOD) at various delay times were measured using a sub-
nanosecond spectrometer (EOS, Ultrafast Systems) incorpo-
rating two 512 pixel CCD sensors as the sample and reference
channel. The instrument response function (IRF) of this
experimental setup has a full width half maximum (FWHM) of
,100 ps. A sample of compound 9 was dissolved in CH3CN,
(b) For a related process see: M. B. Prendergast, B. B. Kirk, J. D. Savee,
D. L. Osborn, C. A. Taatjes, K.-S. Masters, S. J. Blanksby, G. da Silva,
A. J. Trevitt, Phys. Chem. Chem. Phys. 2016, 18, 4320. doi:10.1039/
C5CP02953H
[13] J. Boivin, E. Cre´pon, S. Z. Zard, Tetrahedron Lett. 1990, 31, 6869.
doi:10.1016/S0040-4039(00)97193-4
[14] (a) D. H. R. Barton, D. Crich, W. B. Motherwell, J. Chem. Soc., Chem.
Commun. 1983, 939. doi:10.1039/C39830000939

J
P. W. Moore et al.
(b) D. H. R. Barton, D. Crich, W. B. Motherwell, Tetrahedron 1985,
41, 3901. doi:10.1016/S0040-4020(01)97173-X
[24] (a) R. H. Prager, Y. Singh, B. Weber, Aust. J. Chem. 1994, 47, 1249.
doi:10.1071/CH9941249
(c) D. H. R. Barton, M. Samadi, Tetrahedron 1992, 48, 7083.
doi:10.1016/S0040-4020(01)88249-1
(d) E. J. Ko, G. P. Savage, C. M. Williams, J. Tsanaktsidis, Org. Lett.
(b) K. H. Ang, R. H. Prager, C. M. Williams, Aust. J. Chem. 1995, 48,
567. doi:10.1071/CH9950567
(c) K. H. Ang, R. H. Prager, J. A. Smith, B. Weber, C. M. Williams,
Tetrahedron Lett. 1996, 37, 675. doi:10.1016/0040-4039(95)02240-6
(d) R. H. Prager, C. M. Williams, Aust. J. Chem. 1996, 49, 1315.
doi:10.1071/CH9961315
(e) R. H. Prager, J. A. Smith, B. Weber, C. M. Williams, J. Chem. Soc.,
Perkins Trans. 1 1997, 2665. doi:10.1039/A700134G
(f) R. H. Prager, M. R. Taylor, C. M. Williams, J. Chem. Soc., Perkins
Trans. 1 1997, 2673. doi:10.1039/A700646B
(g) J. Khalafy, C. E. Svensson, R. H. Prager, C. M. Williams, Tetra-
hedron Lett. 1998, 39, 5405. doi:10.1016/S0040-4039(98)01014-4
(h) J. Khalafy, R. H. Prager, C. M. Williams, Aust. J. Chem. 1999, 52,
31. doi:10.1071/C98093
2011, 13, 1944. doi:10.1021/OL200290M
(e) J. Ho, J. Zheng, R. Meana-Pan˜eda, D. G. Truhlar, E. J. Ko, G. P.
Savage, C. M. Williams, M. L. Coote, J. Tsanaktsidis, J. Org. Chem.
2013, 78, 6677. doi:10.1021/JO400927Y
(f) E. J. Ko, C. M. Williams, G. P. Savage, J. Tsanaktsidis, Org. Synth.
2012, 89, 471. doi:10.15227/ORGSYN.089.0471
[15] (a) K. M. Hess, T. A. Dix, Anal. Biochem. 1992, 206, 309. doi:10.1016/
0003-2697(92)90371-D
(b) B. Epe, D. Ballmaier, W. Adam, G. N. Grimm, C. R. Saha-Mo¨ller,
Nucleic Acids Res. 1996, 24, 1625. doi:10.1093/NAR/24.9.1625
(c) W. Adam, G. N. Grimm, S. Marquardt, C. R. Saha-Mo¨ller, J. Am.
Chem. Soc. 1999, 121, 1179. doi:10.1021/JA9811849
[25] M. Stark, D. R. Arnold, J. Chem. Soc., Chem. Commun. 1982, 434.
doi:10.1039/C39820000434
[26] T. D. Walsh, R. C. Long, J. Am. Chem. Soc. 1967, 89, 3943.
doi:10.1021/JA00991A073
[27] T. Latowski, B. Zelent, Rocz. Chem. 1977, 51, 1405.
[28] C. S. Hansen, S. J. Blanksby, N. Chalyavi, E. J. Bieske, J. R. Reimers,
A. J. Trevitt, J. Chem. Phys. 2015, 142, 014301. doi:10.1063/
1.4904267
[29] (a) T. L. Ho, Synth. Commun. 1973, 3, 99. doi:10.1080/
00397917308062015
(b) G. R. Newkome, V. K. Majestic, J. D. Sauer, Org. Prep. Proced.
Int. 1980, 12, 345. doi:10.1080/00304948009356488
[30] M. J. Barnes, Waste Management 91 CONF-910270-66, 1991.
[31] F. Zhu, J.-L. Tao, Z.-X. Wang, Org. Lett. 2015, 17, 4926. doi:10.1021/
ACS.ORGLETT.5B02458
[16] S. Beranova´, C. Wesdemiotis, Int. J. Mass Spectrom. Ion Process.
1994, 134, 83. doi:10.1016/0168-1176(94)03972-0
[17] A. G. Harrison, J. Mass Spectrom. 1999, 34, 1253. doi:10.1002/(SICI)
1096-9888(199912)34:12,1253::AID-JMS898.3.0.CO;2-W
[18] If a more labile radical fragmenation pathway is available, the NR3
moiety remains intact as a phenyl substituent. See for example: B. B.
Kirk, A. J. Trevitt, S. J. Blanksby, J. Am. Soc. Mass Spectrom. 2013,
24, 481. doi:10.1007/S13361-012-0549-X
[19] D. Kuck, S. Heitkamp, J. Spross, M. C. Letzel, I. Ahmed, K. Krohn,
R. G. Parker, Y. L. Wang, V. J. Robbins, W. M. Ames, P. D. Schettler,
R. R. Hark, Int. J. Mass Spectrom. 2015, 377, 23. doi:10.1016/J.IJMS.
2014.06.013
[20] P. J. Derrick, Mass Spectrom. Rev. 1983, 2, 285. doi:10.1002/MAS.
1280020204
[21] (a) J. A. Montgomery, Jr, M. J. Frisch, J. W. Ochterski, G. A. Petersson,
[32] T. Uemura, M. Yamaguchi, N. Chatani, Angew. Chem. Int. Ed. 2016,
J. Chem. Phys. 1999, 110, 2822. doi:10.1063/1.477924
55, 3162. doi:10.1002/ANIE.201511197
(b) J. A. Montgomery, Jr, M. J. Frisch, J. W. Ochterski, G. A.
Petersson, J. Chem. Phys. 2000, 112, 6532. doi:10.1063/1.481224
[22] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A.
Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F.
Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr, J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N.
Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega,
J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma,
V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz,
J. Cioslowski, D. J. Fox, Gaussian 09, Revision E.01 2013 (Gaussian,
Inc.: Wallingford, CT).
[33] F. Minisci, F. R. Bernardi, F. Bertini, R. Galli, M. Perchinummo,
Tetrahedron 1971, 27, 3575. doi:10.1016/S0040-4020(01)97768-3
[34] D. A. DiRocco, K. Dykstra, S. Krska, P. Vachal, D. V. Conway,
M. Tudge, Angew. Chem. Int. Ed. 2014, 53, 4802. doi:10.1002/ANIE.
201402023
[35] D. D. Perin, W. L. F. Armarego, Purification of Laboratory Chemicals
(3rd edn) 1988 (Permagon Press: Oxford).
[36] F. Jiang, D. Bezier, J. Sortais, C. Darcel, Adv. Synth. Catal. 2011, 353,
239. doi:10.1002/ADSC.201000781
[37] J. L. D´ıaz, B. Villacampa, F. Lo´pez-Calahorra, D. Velasco, F. Lo,
Chem. Mater. 2002, 14, 2240. doi:10.1021/CM011293D
[38] T. Erker, G. Brunhofer, U. Ja¨ger, K. Vanura, V. Dirsch, E. Heiß,
WO 2012/013725 A1.
[39] L. J. Farrugia, J. Appl. Cryst. 1999, 32, 837. doi:10.1107/
S0021889899006020
[40] G. M. Sheldrick, Acta Crystallogr. Sect. A: Found. Adv. 2008, A64,
112. doi:10.1107/S0108767307043930
[41] L. J. Farrugia, J. Appl. Cryst. 1997, 30, 565. doi:10.1107/
S0021889897003117
[23] Y. Singh, R. H. Prager, Aust. J. Chem. 1992, 45, 1811.