Intramolecular Hydrogen-Bonding Modulates the Nucleophilic Reactivity of Ammonium-Peroxycarboxylates

Article abstract of DOI:10.1002/ejoc.201801158

The ammonium-peroxycarboxylic acid mesylates derived from γ-aminobutyric acid, β-alanine, and β-piperidinopropionic acid were synthesized and characterized by spectroscopic methods and X-ray crystallography. To study the nucleophilic reactivities of the corresponding ammonium- and amino-peroxycarboxylates, the kinetics of their reactions with a series of benzhydrylium ions (Ar₂CH⁺) were investigated in alkaline, aqueous solutions at 20 °C. Using sequential-mixing stopped-flow UV/Vis photometry, the rates of the reactions of the short-lived nucleophiles with Ar₂CH⁺ were determined and analyzed by the linear free energy relationship lg k = s_N(N + E) furnishing nucleophilicity parameters (N, s_N) of the peroxycarboxylates. Quantum chemical calculations indicate that the reactivity of the zwitterionic ammonium-peroxycarboxylates is attenuated by intramolecular N–H···O hydrogen bonding.

Full text of DOI:10.1002/ejoc.201801158

DOI: 10.1002/ejoc.201801158
Full Paper
Reaction Kinetics
Intramolecular Hydrogen-Bonding Modulates the Nucleophilic
Reactivity of Ammonium-Peroxycarboxylates
Robert J. Mayer[a] and Armin R. Ofial*[a]
Abstract: The ammonium-peroxycarboxylic acid mesylates de- flow UV/Vis photometry, the rates of the reactions of the short-
rived from γ-aminobutyric acid, ꢀ-alanine, and ꢀ-piperidino- lived nucleophiles with Ar2CH+ were determined and analyzed
propionic acid were synthesized and characterized by spectro- by the linear free energy relationship lg k = sN(N + E) furnishing
scopic methods and X-ray crystallography. To study the nucleo- nucleophilicity parameters (N, sN) of the peroxycarboxylates.
philic reactivities of the corresponding ammonium- and amino- Quantum chemical calculations indicate that the reactivity of
peroxycarboxylates, the kinetics of their reactions with a series the zwitterionic ammonium-peroxycarboxylates is attenuated
of benzhydrylium ions (Ar2CH+) were investigated in alkaline, by intramolecular N–H···O hydrogen bonding.
aqueous solutions at 20 °C. Using sequential-mixing stopped-
Introduction
ylates 2+MsO– as depicted in Scheme 1. Dissolving γ-amino-
butyric acid (1a, GABA), ꢀ-alanine (1b), and ꢀ-piperidino-
propionic acid (1c) in methanesulfonic acid and adding an ex-
cess of concentrated (> 85 %) aqueous hydrogen peroxide[3h]
generated the APOCA mesylates 2a+MsO–, 2b+MsO–, and
2c+MsO–, respectively, which precipitated when the reaction
mixture was poured on ice-cold THF.
Recent investigations into the nucleophilic reactivity of inor-
ganic bleach reagents and peroxide anions showed that per-
oxycarboxylates possess distinctly higher reactivity than the an-
ions of alkyl hydroperoxides (ROO–) and hydrogen peroxide
(HOO–).[1,2] In this context, some patents on the use of amino-
acid derived peroxycarboxylic acids intrigued us.[3a–3g] Quater-
narization of the nitrogen by alkylation or protonation along
with subsequent oxidation at the carboxyl group led to solid
–
ammonium-peroxycarboxylic acids (APOCAs, with HSO4 or
MsO– counterions).[3a] These APOCA salts had been character-
ized by elemental analysis and, in some cases, melting points.
In addition, their potential application as textile bleaches was
reported.[3a–3g] One would, furthermore, expect that APOCA
salts could serve as stable and well water-soluble precursors of
highly nucleophilic peroxycarboxylate anions.
To enhance our understanding of their reactivities, we char-
acterized a representative set of three APOCA salts by spectro-
scopic methods and X-ray analysis. We then evaluated the pH-
dependent reactivity of aqueous solutions of APOCA salts to-
ward benzhydrylium ions (as reference electrophiles) by kinetic
methods to define the potential of APOCAs as nucleophilic
oxidants and compare their reactivities with those of related
oxidants.
Scheme 1. Preparation of ammonium-peroxycarboxylic acid mesylates.
Beside their spectroscopic characterization (Supporting In-
formation), 2a+MsO–, 2b+MsO– and 2c+MsO– were investigated
by single-crystal X-ray diffraction (Figure 1).[4] The oxygen–oxy-
gen distances of 1.451 to 1.464 Å in 2a+, 2b+, and 2c+ are in
the typical range reported for peroxycarboxylic acids.[5]
In crystals of 2a+MsO– and 2b+MsO– the C(O)–O–O–H groups
adopt an almost identical conformation with C–O–O–H dihedral
Results and Discussion
Characterization of Ammonium-Peroxycarboxylates. Follow- angles of 158.2° and 154.6°. In contrast, this dihedral angle is
ing a literature procedure,[3a] we synthesized three APOCA mes- significantly lower in the piperidinium derivative 2c+MsO–
(104.7°), probably owing to different hydrogen bonding pat-
terns in the solid states of (2a–c)+MsO–. As exemplarily shown
[a] Department Chemie, Ludwig-Maximilians-Universität München,
for 2a+MsO– in Figure 2, various hydrogen bonding interactions
Butenandtstraße 5-13, 81377 München, Germany
E-mail: ofial@lmu.de
URL: http://www.cup.uni-muenchen.de/oc/ofial/
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under https://doi.org/10.1002/ejoc.201801158.
between O–H or N–H of the 2+ and the oxygen atoms of the
mesylate counterions are the prevalent interactions found for
all APOCA mesylates in crystalline state (Figure 2).
Eur. J. Org. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 2. Protonation equilibria of APOCAs.
forms of GABA (1a) and ꢀ-alanine (1b).[6] In this work, we set
out to characterize the peroxycarboxylate groups of 2 and 2–
in alkaline, aqueous solution with the aim to compare the nu-
cleophilic reactivities with those of other peroxide anions on
the basis of Mayr's linear free energy relationship (1).[7–11]
lg k2(20 °C) = sN(N + E)
(1)
In equation (1) the nucleophiles' reactivity is described by
the solvent-dependent nucleophilicity parameter N and the nu-
cleophile-specific susceptibility sN. These parameters can be de-
termined from kinetic measurements with a set of benzhydryl-
ium ions as reference electrophiles (with known electrophilicity
parameters E), employing the previously reported proce-
dure.[1,6] As shown in Scheme 3, the kinetics were studied in
alkaline, aqueous solution, in which zwitterions 2 were gener-
ated by equilibrium deprotonation of the corresponding cat-
ions 2+. The reactions of an excess of 2 with blue colored benz-
hydrylium ions Ar2CH+ 3[12] under pseudo-first-order conditions
yielded colorless reaction products, which enabled us to use
stopped-flow UV/Vis photometry to monitor the reaction
progress.
Figure 1. Solid-state structures of 2a+MsO–, 2b+MsO– and 2c+MsO– (thermal
ellipsoids are depicted at 50 % probability level at T = 100 K, counterion
MsO– not shown).[4]
Figure 2. Hydrogen bonding in the solid-state structure of 2a+MsO– (thermal
ellipsoids are depicted at 50 % probability level).
To characterize the acid/base behavior of 2+, we determined
the Brønsted acidities (pKa) of their peroxycarboxylic groups by
potentiometric titration in water. The acidities of 2+ were found
to be in the common range of pKa 7–8 for aliphatic peroxyacids
(see Table 1 below).[1] By assuming that the known pKa of
ammonium groups in protonated amino acids (pKa ≈ 10) also
apply for the APOCA cations 2+, it can be anticipated that
amino-substituted peroxycarboxylic acids exist as cations (2+),
zwitterions (2), or anions (2–) in aqueous solution (Scheme 2),
depending on the pH and in analogy to the established equilib-
ria for α-amino acids. As both, the peroxycarboxylate and the
amino group, can act as the nucleophile, the reactivity of
APOCA salt solutions is significantly affected by pH.
Scheme 3. Generation of nucleophilic zwitterions 2 in alkaline, aqueous solu-
–
tion and reactions of 2 with the benzhydrylium ions 3 (counterions: BF4 for
3, MsO– for 2+, electrophilicities E from ref[11]).
Initial sequential mixing experiments showed that conditions
successfully used to determine the nucleophilicity of various
peroxide anions[1] could not be applied without modification
to study the kinetics of the reactions of 3 with the ammonium-
peroxycarboxylates 2: When we generated 2a in a first mixing
Kinetics. The nucleophilic reactivity of the amino group in step by dissolving 2a+MsO– ([2a+]0 = 4 × 10–4
M
) in aqueous
), then al-
aminocarboxylates derived from natural amino acids or small potassium hydroxide solution ([KOH]0 = 4 × 10–4
M
peptides has already been investigated.[6] Analogous N-nucleo- lowed the solution to age for a defined time and, in a second
philicities have also been determined for the deprotonated mixing step, added solutions of 3-BF4 ([3]0 = 1.1 × 10–5
M),
–
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 3. (A) Observed decays of benzhydryl absorbances (3b) for variable aging times of the nucleophile (2a) solutions and (B) dependence of the correspond-
ing first-order rate constants kobs on the aging times. [2a+MsO–]0 = 4 × 10–4 , [OH–]0 = 4 × 10–4
M M, 20 °C. The slow reaction of 2a with 3b after 600 s aging
time did not follow a mono-exponential decay.
we observed mono-exponential decays of the benzhydrylium
absorbances under these pseudo-first-order conditions. How-
ever, the observed decay rates systematically decreased when
the time between the first and the second mixing step, that
is, the aging time of the nucleophile solution, was increased
(Figure 3A). The dependence of the observed rate contants kobs
on the aging time indicated a first-order decomposition of the
nucleophile on the seconds time scale at 20 °C (τ1/2 = 20–25 s,
Figure 3B). The analogous variation of the aging times for reac-
tions of reference electrophiles 3 with aqueous solutions of the
nucleophilic zwitterions 2b and 2c (generated from 2b+MsO–
and 2c+MsO–, respectively) gave similar results but showed that
the degradation proceeded somewhat slower (τ1/2 > 100 s, Sup-
porting Information) than for 2a.
In contrast, iodometric titration of an alkaline solution of
2a+MsO– showed that the total peroxide content of the reac-
tion mixture decreased on the minute time-scale (Figure 4),
thus, much slower than expected from the kinetic experiments
shown in Figure 3.
Scheme 4. (A) General degradation of 2 to 1 in alkaline solution. (B) Forma-
tion of the lactam 5 by ring closure of 2a+MsO– in alkaline solution and
partial trapping of thus liberated hydrogen peroxide by 3a. Yields were deter-
mined by using added dimethyl fumarate as integration standard. [a] The
hydrophilic lactam 5 was only partially extracted by dichloromethane, which
resulted in a low yield of 5 (yield refers to 2a+MsO–). [b] Yield refers to 3a-
BF4.
Figure 4. Peroxide content as determined by iodometric titration of an alkal-
The peroxy version of GABA (2a) decomposed significantly
faster than 2b or 2c owing to an additional option: As indicated
ine solution of 2a+MsO– (20 °C, [2a+] = 1.4 × 10–3 , [OH–] = 5 × 10–2
M M).
The NMR spectroscopic investigation of the alkaline, aqueous by 1H NMR spectroscopy, a mixture of the lactam 5 and the
solutions showed that degradation of the peroxycarboxylate to peroxide 4 was isolated after dichloromethane extraction of an
the carboxylate group by attack of hydroxide ions at the carb- aqueous, alkaline solution of 2a+MsO– that, after 10 min of ag-
oxyl carbon is a common decomposition pathway for all investi- ing at ambient temperature, was treated with 3a-BF4
gated zwitterions 2 (Scheme 4A). This reaction path liberates (Scheme 4B). We conclude that 2a not only degrades because
hydrogen peroxide, which decomposes slowly to molecular of the hydroxide ion attack shown in Figure 4A but also under-
oxygen and water.
goes ring closure to give γ-butyrolactam (5) under alkaline reac-
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tion conditions (pH 13). A stoichiometric amount of hydrogen carboxyl group or by amino groups that intramolecularly attack
peroxide anions is released in this self-condensation, which is at the carboxyl carbon, H2O2 is released. The iodometrically de-
interceptable by the known reaction with 3a[2] to yield the termined slow decrease of the overall peroxide content for
bis(benzhydryl)peroxide 4. Degradation by ring closure was not alkaline solutions of 2+ (Figure 4), therefore, reflects the subse-
observed for 2b, presumably because this intramolecular reac- quent, slow decomposition of the released H2O2.
tion would lead to a strained ꢀ-lactam.
Figure 3 defines the conditions for the determination of the
nucleophilic reactivities of 2: First, solutions of 2 have to be
mixed with their electrophilic reaction partners within 1 s after
generation of 2 from 2+. Second, the half-life times of the reac-
tions should be 10 s or less to avoid noticeable degradation of
2 during the reactions with 3. Determining the reaction kinetics
by using the sequential mixing setup of the stopped-flow spec-
trometer was, therefore, instrumental to access the nucleophi-
licities of solutions of 2: In a first step, aqueous solutions of the
base and 2+MsO– were mixed and allowed to age for 1 s. Then
the reaction with the reference electrophiles 3 was initiated by
mixing the solution of freshly generated 2 with that of 3.
As the degree of deprotonation of 2+ is adjustable by the
chosen hydroxide concentration, different nucleophilic species
can be addressed: Treating 2+ with an equimolar amount of
base generates the ammonium-peroxycarboxylates 2. Under
these conditions, the oxygen reactivity of zwitterions 2 can be
measured selectively because the more basic amino group is
still protected by protonation. Alternatively, a high concentra-
tion of base generates the amino-peroxycarboxylate 2– as the
major species. Because anions 2– are ambident nucleophiles
that attack via either nitrogen or oxygen at Ar2CH+, the deter-
mination of the nucleophilicities of 2a– and 2b– was not possi-
ble unequivocally. In 2c–, however, the nitrogen of the tertiary
amine is sterically hindered and does not react with highly sta-
bilized benzhydrylium ions, such as 3b–e.[13] This enabled us to
quantify the oxygen-reactivity of amino-peroxycarboxylate 2c–
without interfering reactivity of the piperidino group.
To gain evidence for the oxidation of electrophiles by
APOCAs, we studied the reaction of benzylidene malononitrile
6 with 2+ under alkaline conditions (Scheme 5). When 6 was
treated with an excess of 2a+MsO– (2.5 equiv.), the electrophile
was almost quantitatively converted into the corresponding ep-
oxide 7. Additionally, the formation of the corresponding GABA
zwitterion 1a and γ-butyrolactam 5 (both in comparable ratio)
were observed by NMR spectroscopy. The outcome of this reac-
tion corroborates our mechanistic outline: After quantitative
generation of 2a by deprotonation of 2a+ under the alkaline
conditions, one part of 2a (2.5 equiv.) reacts with 6 (1 equiv.)
in a nucleophilic epoxidation to furnish 7 (1 equiv.) and 1a. The
remaining 1.5 equivalents of the peroxycarboxylate 2a decom-
pose to hydrogen peroxide and the lactam 5.
We, therefore, conclude that the zwitterions 2 decompose
rapidly in alkaline, aqueous solution after being generated from
2+ (as seen in Figure 3). Independent of whether decomposition
of 2 is initiated by hydroxide ions that react with the peroxy-
The analysis of the observed first-order rate constants kobs
(s–1, Figure 5A) by equation (2) was performed as reported for
the anions of organic peroxides:[1,2] Brønsted acidities (pKa) of
2+ were employed to determine the concentrations of peroxy-
carboxylates and hydroxide ions, that is, [2] (or [2–]) and [OH–].
Then, [OH–] and the known second-order rate constants kOH
Scheme 5. Reaction of 2a+MsO– with malononitrile 6 as analyzed by 1H NMR
spectroscopy.
Figure 5. (A) Absorbance A (at 610 nm) vs. time for the reaction of 2b (4.00 × 10–4
M M) in aqueous solution (pH 9) at 20 °C. (B) The
) with 3b (1.14 × 10–5
slope of the linear plot of the first-order rate constant k1 (= kobs – kOH[HO–]) vs. nucleophile concentration was used to derive the second-order rate constant
k2 for the attack of the zwitterion 2b at the benzhydrylium ion 3b.
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(from ref[14]) were used for the calculation of the rates of con-
sumption of the cations 3b–e by the hydroxide ions. After sub-
traction of the contribution by hydroxide ions (kOH[OH–]) from
the observed first-order rate constants kobs, the first-order rate
constant k1 [in Equation (3)] reflect the reaction of the electro-
philes 3 with the nucleophiles with only a very minor correction
for the much slower background reaction of 3 with water (kW).
kobs = kW + kOH[HO–] + k2[R′OO–]
(2)
(3)
k1 = k2[R′OO–] + kW = kobs – kOH[HO–]
Thereby, the second-order rate constants for the reactions of
2 (or 2–) with benzhydrylium ions 3b–e can be obtained from
the slope of the linear correlation of the first-order rate constant
k1 (= kobs – kOH[HO–]) with the nucleophile concentration.
In accord with equation (3) and as exemplified in Figure 5B
for the reactions of 2b with 3b, the first-order rate constants k1
increased linearly with increasing concentration of the nucleo-
phile 2b, and the slope of these correlations corresponded to
the second-order rate constant (k2, Table 1) for the attack of 2
(or 2–) at the benzhydrylium ions 3b–e.
Figure 6. Correlation of lg k2 for the reactions of nucleophiles 2a–c (and 2c–)
with the reference electrophiles 3b–e (from Table 1) with the electrophilicity
parameters E of 3.
oxygen attack of 2b at 3a and subsequent Criegee rearrange-
ments gave rise to various oxidation products of 3a, such as 4-
(dimethylamino)benzaldehyde, and the benzhydrol and benzo-
phenone derivatives of 3a.[15] Amine 8b, which may form
The resulting second-order rate constants k2 of these reac- through nitrogen attack of 2b at 3a and subsequent degrada-
tions and the known electrophilicities E of 3b–e[11] were then tion of the peroxycarboxylic to a carboxylic acid, was not de-
1
applied in equation (1): The linear correlations of lg k2 with the tected by H NMR spectroscopy under these conditions. How-
electrophilicity E allowed us to determine the nucleophilicity ever, when 2 equiv. of KOH were used, generation of 8b be-
1
parameter N and the nucleophile-specific susceptibilities sN of came evident from the H NMR spectrum of the reaction mix-
the peroxide species 2a–2c and 2c– (Figure 6). The reactivities ture, and 8b was formed in even higher yields under more alka-
of the ammonium-peroxycarboxylates 2 vary only within less line conditions, that is, reactions of 2b with 3a in the presence
than one order of magnitude. It is noteworthy that 2 react 5 to of 3 equiv. of potassium hydroxide (Supporting Information).
10 times slower with 3b–e than structurally related peroxycarb-
oxylates (such as phenylperoxyacetate or ε-phthalimido-peroxy-
hexanoate[1]). Deprotonation of zwitterion 2c and generation of
the anionic 2c– increased the reactivity of the RCO3 group by
–
a factor of four.
The nucleophilicities of the amino groups in deprotonated
GABA 1a– and ꢀ-alanine 1b– were reported to be N = 13.3–
13.5 (sN = 0.56–0.58).[6] Such a level of reactivity can also be
anticipated for anionic 2a– or 2b– that carry free amino groups
in a similar molecular environment. We, therefore, analyzed the
Scheme 6. Chemical shifts of the N-benzhydrylated amino acids 8a and 8b
contribution of N-attack by 2– at benzhydrylium ions 3 under
(in alkaline CD3CN/D2O = 4:3).
the conditions of our kinetic experiments.
First, products of N-attack by the γ- and ꢀ-amino-carboxyl-
Owing to the fact that all rate constants for 2a–2c collected
ates 1a– and 1b– at 3a-BF4– were characterized by NMR spectro- in Table 1 were acquired by using only one equivalent of potas-
scopy (Supporting Information). The typical singlets at chemical sium hydroxide (relative to 2+MsO–), we conclude that these
shifts of 4.61–4.64 ppm for N-CHAr2 in 8a and 8b (Scheme 6) kinetics describe exclusively the oxygen attack of the peroxy-
were then used to analyse product mixtures obtained for reac- carboxylate groups of the zwitterions 2 at the cationic center
tions of equimolar amounts of 2b+MsO– and 3a-BF4– at variable of 3. Nitrogen attack becomes only relevant at higher pH (that
pH. When 1 equiv. of KOH was used to convert 2b+ to 2b, is, when > 1 equiv. of hydroxide ions relative to 2a+ or 2b+ are
Table 1. Basicities pKaH and second-order rate constants k2 for the oxygen attack of the ammoniumperoxycarboxylates 2 and the aminoperoxycarboxylate
2c– at the reference electrophiles 3b–e in alkaline, aqueous solution at 20 °C.
–1 s–1
]
N (sN)
[a]
pKaH
k2
[M
3e
3d
3c
3b
2a
2b
2c
7.77
7.35
3.10 × 102
2.06 × 102
3.08 × 102
1.26 × 103
5.20 × 102
3.32 × 102
5.55 × 102
2.10 × 103
1.39 × 103
1.07 × 103
1.66 × 103
6.07 × 103
6.33 × 103
3.93 × 103
8.45 × 103
2.89 × 104
14.33 (0.57)
14.07 (0.56)
13.94 (0.62)
15.17 (0.59)
7.48
2c–
9.92[b]
[a] Determined by potentiometric titration. [b] Deprotonation of the piperidinium group.
Eur. J. Org. Chem. 0000, 0–0 www.eurjoc.org
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
used) as the more nucleophilic 2– then starts to compete with ther nucleophiles listed in Mayr's reactivity database.[10] Intra-
2 for the electrophile 3. As nitrogen attack of 3 by 2c– is un- molecular N–H···O hydrogen bonding reduces the nucleophilic
likely[13] (see above), the kinetics of 2c– refer to oxygen attack reactivity of ꢀ- and γ-ammonium-substituted peroxycarboxyl-
at 3.
ates relative to structurally related peroxycarboxylates.
Quantum Chemical Calculations. Why are the nucleophilic
reactivities of the ammonium-peroxycarboxylates lower than
those of common peroxybenzoates or aliphatic peroxycarboxyl-
ates? The fact that the aminoperoxycarboxylate 2c– is more nu-
cleophilic than the ammoniumperoxycarboxylate 2c led us to
investigate whether intramolecular hydrogen bonding of the
peroxycarboxylate moiety with the ammonium group reduces
reactivity in zwitterions 2. To investigate the effect of intra-
molecular H-bonding in 2, quantum chemical calculations with
the Gaussian software were performed to evaluate the confor-
mational space of 2a+–2c+.[16] Conformers were optimized with
the M06-HF[17]/6-311++(d,p) method taking into account aque-
ous solvation by the SMD version of the Polarizable Continuum
Model.[18] Several methods with the 6-311++G(d,p) basis were
tested (including various DFT methods, HF, MP2 and MP3; see
Supporting Information for details), but only classical HF and
Figure 8. Comparison of nucleophilicity parameters N of 2 and 2– with those
of other anionic oxidants (from ref.[1,2,14]).
The application of APOCAs for oxygen transfer reactions in
M06-HF gave the position of the intramolecular N–H···O aqueous solution requires careful pH control, however, and is
hydrogen bond as expected from experimental basicities.
The calculations revealed that the formation of an intra-
molecular hydrogen bond is highly favored in 2a–2c (Figure 7),
which may attenuate the rates of attack of the peroxycarboxyl-
limited to very fast reactions with electrophiles because of the
competing degradation to ammoniumcarboxylates and
hydrogen peroxide under alkaline conditions.
A combination of the kinetic and thermodynamic data about
ate groups at the carbocationic centres of the electrophiles APOCAs from this study with those for further aliphatic and
3.[19]
aromatic peroxycarboxylic acids, hydroperoxides and related
bleach reagents will provide a broad fundament for under-
standing structure-activity relationships, such as, for example,
the so-called α-effect.[20]
Experimental Section
CAUTION: Pure peroxides may explode or detonate under influence
of heat, shock, spark etc. All reactions or handling of peroxides
should be carried out behind a blast shield as a precaution using
open vessels while wearing appropriate safety equipment. Rotary
evaporation of peroxides should be carried out carefully without
excessive stirring at room temperature. Generally, all handling of
peroxides should only be performed by experienced people trained
in handling explosive compounds.
Analytics: Nuclear magnetic resonance spectra were recorded on
400 or 800 MHz NMR spectrometers. The following abbreviations
and their combinations are used in the analysis of NMR spectra:
s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, br.
s = broad singlet. All 13C NMR spectra were recorded under broad-
band proton-decoupling. NMR signals were assigned based on in-
formation from additional 2D NMR experiments (COSY, gHSQC,
gHMBC, NOESY). Internal reference was set to the residual solvent
signals.[21]
Figure 7. Relative Gibbs energies (at 298 K) of the conformers of 2a–2c with
the lowest Gibbs energy as calculated at the M06-HF/6-311++G(d,p) level of
theory for aqueous solution (SMD).
Conclusions
3-Carboperoxypropan-1-aminium Methanesulfonate (2a+MsO–):
γ-Aminobutyric acid 1a (400 mg, 3.88 mmol) was dissolved in meth-
anesulfonic acid (2.03 mL, 31.3 mmol) with careful heating. The
solution was cooled to 0 °C and H2O2 (0.28 mL of a 92 % aq. solu-
tion, 3 equiv.) was added dropwise. Then the reaction mixture was
warmed up to room temperature. After 1 h the solution was poured
into ice-cold THF (30 mL) and stirred for 30 min. The formed crystal-
line precipitate was collected by filtration and then washed with
Ammoniumperoxycarboxylic acid mesylates (APOCA mesylates)
are easily preparable and stable peroxy compounds and, thus,
a convenient source of active oxygen. At pH 8–9, ammonium-
peroxycarboxylates, the corresponding zwitterions of APOCAs,
can be generated, which are highly reactive epoxidation rea-
gents. Their reactivities were characterized by Mayr's nucleo-
philicity parameters N and sN, which allows for a direct compari-
son with reactivities of other peroxy anions (Figure 8) and fur- THF (5 mL) and diethyl ether (5 mL). Drying of the residue yielded
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
2a+MsO– as a colorless, crystalline solid (704 mg, 84 %); m.p. 80 °C;
KOH solution and a solution of 2+MsO– were mixed. Kinetics aiming
to characterize the reactivity of zwitterions 2 were carried out by
mixing approx. equimolar amounts of 2+ and KOH in the first mix-
ing step. Kinetics aiming to characterize the reactivity of the anionic
active oxygen content: 100 % (iodometric titration). 1H NMR
1
(400 MHz, D2O): δ = 7.68 (br. t, JH,N ≈ 40 Hz, 0.02 H, +NH3), 3.05 (t,
J = 7.8 Hz, 2 H), 2.79 (s, 3 H), 2.56 (t, J = 7.3 Hz, 2 H), 2.00 (quint,
J = 7.4 Hz, 2 H) ppm. 13C{1H} NMR (101 MHz, D2O): δ = 173.6 (Cq),
38.43 (CH2), 38.36 (CH3), 27.3 (CH2), 21.8 (CH2) ppm. IR (ATR, neat):
2c– were performed by mixing 2+ [(2–5) × 10–4
of KOH (2.5 ×10–2
these solutions were mixed at a second mixer with an equal volume
M] with an excess
M) in the first mixing step. After 1 s (“aging time”)
ν = 3139, 2891, 1760, 1631, 1529, 1477, 1417, 1376, 1340, 1323,
˜
1298, 1257, 1159, 1119, 1066, 1044, 995, 974, 968, 928, 872, 781, of a solution of the electrophile 3-BF4. By using a high excess of the
763, 702, 660 cm–1. HRMS (FAB+): m/z calcd. for [C4H10NO3+]:
120.0665, found 120.0675. Elemental Analysis: Calcd: C, 27.90; H,
6.09; N, 6.51; S, 14.90; found: C, 27.94; H, 6.15; N, 6.49; S, 14.99.
zwitterions 2 over the electrophiles 3, the peroxide concentrations
remained almost constant during the kinetic runs, resulting in
mono-exponential decays of the electrophiles' absorptions. First-
order rate constants kobs (s–1) were obtained by least-squares fitting
the time-dependent absorbances with the single-exponential func-
tion At = A0 exp(–kobst) + C. After converting kobs to k1 by applying
equation (3), the second-order rate constants for the reactions of 2
(or 2–) with benzhydrylium ions 3b–e were obtained from the slope
of the linear correlation of the first-order rate constants k1 with the
nucleophile concentration.
2-Carboperoxyethan-1-aminium Methanesulfonate (2b+MsO–):
ꢀ-Alanine 1b (200 mg, 2.24 mmol) was dissolved in methanesul-
fonic acid (1.01 mL, 15.6 mmol) with careful heating. The solution
was cooled to 0 °C and H2O2 (0.13 mL of a 85 % aq. solution,
2 equiv.) was added dropwise. Then the reaction mixture was
warmed up to room temperature. After 1 h the solution was poured
into ice-cold THF (20 mL) and stirred for 20 min. The formed crystal-
line precipitate was collected by filtration, washed with THF (10 mL)
and diethyl ether (10 mL). Drying of the residue yielded 2b+MsO–
as colorless, crystalline solid (365 mg, 81 %). The active oxygen con-
tent of this sample was determined by iodometric titration to be
91 %. Therefore, an overall yield of 74 % of 2b+MsO– was calculated.
1H NMR (400 MHz, D2O): δ = 7.85 (br. t, 1JH,N ≈ 42 Hz, 0.01 H, +NH3),
3.38 (t, J = 6.6 Hz, 2 H), 2.91 (t, J = 6.6 Hz, 2 H), 2.83 (s, 3 H) ppm.
13C{1H} NMR (101 MHz, D2O): δ = 171.5 (Cq), 38.4 (CH3), 34.8 (CH2),
28.2 (CH2) ppm.
Details of the individual kinetic measurements are given in the Sup-
porting Information.
Computational Analysis: First, all studied species were subjected
to a conformational search with the OPLS3 force field[22] as imple-
mented in the Macromodel software package[23] applying a MCMM
search. The thus obtained set of conformers was subsequently opti-
mized with the M06-HF/6-311++G(d,p) method[17] taking aqueous
solvation into account by the SMD model.[18] Thermal corrections
were obtained at the same level of theory from vibrational frequen-
cies and are unscaled. Selection of an appropriate theoretical
method for the structural optimization in the Gaussian software
package[16] was found to be difficult as most methods are not able
to correctly represent the position of the N···H···O hydrogen bond
in a way that is in accord with experimental results (the amino
group has the higher pKaH value than the peroxycarboxylate, there-
fore the hydrogen should be located closer to nitrogen). Calcula-
tions in gas-phase as well as standard DFT (and MP) methods in
solution (water) localize the hydrogen atom at the oxygen site. Ge-
ometries were all optimized with the 6-311++G(d,p) basis both in
gas phase and aqueous solution (SMD and IEF-PCM were tested).
Of the investigated methods (HF, B3LYP, B3LYP-D3, B2PLYPD,
M06-2X, M06-HF, ωb97xd, MP2, MP3) only HF and M06-HF in combi-
nation with aqueous solvation were able to correctly locate the
hydrogen bond.
1-(3-Hydroperoxy-3-oxopropyl)piperidin-1-ium Methanesulfon-
ate (2c+MsO–): 3-(Piperidin-1-yl)propanoic acid 1c (197 mg,
1.25 mmol) was dissolved in methanesulfonic acid (0.81 mL,
12.5 mmol). The solution was cooled to 0 °C and H2O2 (0.15 mL of a
85 % aq. solution, 4 equiv.) was added dropwise. Then the reaction
mixture was warmed up to room temperature. After 1 h the solution
was poured into ice-cold THF (20 mL) and stirred for 60 min. The
formed crystalline precipitate was collected by filtration, washed
with THF (10 mL) and diethyl ether (5 mL). Drying of the residue
yielded 2c+MsO– as a colorless, crystalline solid (291 mg, 86 %); m.p.
134 °C (ref:[3d] m.p. 132 °C); active oxygen content: 94 % (iodometric
titration). Therefore, an overall yield of 81 % of 2c+MsO– was calcu-
1
lated. H NMR (400 MHz, D2O): δ = 8.72 (br. s, 0.06 H, +NH3), 3.53–
3.50 (m, 2 H, 5-H), 3.44 (t, J = 7.1 Hz, 2 H), 2.99–2.93 (m, 4 H), 2.78
(s, 3 H), 1.96–1.87 (m, 2 H), 1.83–1.63 (m, 3 H), 1.53–1.39 (m, 1 H)
ppm. 13C{1H} NMR (101 MHz, D2O): δ = 170.9 (Cq), 53.4 (CH2), 51.4
(CH2), 38.4 (CH3), 25.7 (CH2), 22.6 (CH2), 20.8 (CH2) ppm. IR (ATR,
Details of the conformational analysis and calculated geometries
are given in the Supporting Information.
neat): ν = 3021, 2961, 2775, 2721, 1773, 1480, 1451, 1349, 1208,
˜
1139, 1122, 1087, 1066, 1037, 979, 944, 884, 845, 782, 733 cm–1
.
Acknowledgments
pKa Determination: A Metrohm Titrando system (pH 0.001) was
applied for automated titrations. Solutions of the acids in water
were prepared at a constant ionic strength of I = 0.1 with a NaCl
We thank Nathalie Hampel (LMU, preparation of 2 and 7) and
Dr. Peter Mayer (LMU, single-crystal X-ray crystallography). The
authors are grateful to Professor Herbert Mayr (LMU) for contin-
uous support, Dr Jérôme Gomar (L'Oréal) and Dr Stéphane
Sabelle (L'Oréal) for helpful discussions and to Prof. Claude Y.
Legault (Sherbrooke, Canada) for advice on quantum chemical
calculations. Financial support by L'Oréal Research & Innovation
(Aulnay sous Bois, France), the Deutsche Forschungsgemein-
schaft (SFB 749, project B1), and the Fonds der Chemischen
Industrie (Kekulé fellowship to RJM) is gratefully acknowledged.
stock solution and titrated with 0.1
M KOH. The temperature during
the titration was maintained constant at (20 0.1) °C with a circulat-
ing bath thermostat. The titration curve was recorded automatically
and the equivalence points were determined by the control soft-
ware of the titration system. For every acid, the titration was re-
peated at least three times, the results were averaged and gathered
in Table 1.
Kinetic Measurements: Kinetic measurements were performed on
commercial stopped-flow UV/Vis photometry systems (Applied Pho-
tophysics SX.20). The temperature (20.0 0.2 °C) was maintained
constant by using circulating bath cryostats. To prevent alkaline de-
composition of the peroxide solutions, the sequential-mixing setup
of the instrument was employed. In a first step, equal volumes of a
Keywords: Nucleophilicity · Kinetics · Bleaching ·
Peroxides · Linear-free energy relationships
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
7
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[1] R. J. Mayer, T. Tokuyasu, P. Mayer, J. Gomar, S. Sabelle, B. Mennucci, H.
Mayr, A. R. Ofial, Angew. Chem. Int. Ed. 2017, 56, 13279–13282; Angew.
Chem. 2017, 129, 13463–13467.
nucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D.
Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Pet-
rone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G.
Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E.
N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J.
Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Och-
terski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman, D. J. Fox,
Gaussian 16, Revision A.03, Gaussian, Inc., Wallingford CT, 2016.
[2] R. J. Mayer, A. R. Ofial, Org. Lett. 2018, 20, 2816–2820.
[3] a) C. Venturello, C. Cavallotti, EP 0316809 A2, 1988; b) J. R. Darwent, K. C.
Francis, J. Oakes, D. W. Thornthwaite, US4904406A, 1988; c) C. Venturello,
C. Cavallotti, F. Achilli, US5117049A, 1991; d) C. Venturello, C. Cavallotti,
US5245075, 1993; e) C. Venturello, C. Cavallotti, US5294362A, 1994;
f) J. Oakes, D. W. Thornthwaite, WO 94/21605, 1994; g) J. Oakes, D. W.
Thornthwaite, WO 97/16515, 1997; h) N.-D. H. Gamage, B. Stiasny,
J. Stierstorfer, P. D. Martin, T. M. Klapötke, C. H. Winter, Chem. Eur. J. 2016,
22, 2582–2585.
[17]
[18]
Y. Zhao, D. G. Truhlar, J. Phys. Chem. A 2006, 110, 13126–13130.
[4] CCDC 1849148 (for 2a+MsO–, uv697), 1849149 (for 2b+MsO–, vv407),
1849150 (for 2c+MsO–, uv132) contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre..
[5] J. Hartung, I. Svoboda in The Chemistry of Peroxides Vol. 2 (Ed.: Z. Rappa-
port), Wiley, Hoboken, 2006, Chapt 6, pp. 93–144.
[6] F. Brotzel, H. Mayr, Org. Biomol. Chem. 2007, 5, 3814–3820.
[7] H. Mayr, M. Patz, Angew. Chem. Int. Ed. Engl. 1994, 33, 938–957; Angew.
Chem. 1994, 106, 990–1010.
[8] H. Mayr, A. R. Ofial, SAR QSAR Environ. Res. 2015, 26, 619–646.
[9] H. Mayr, Tetrahedron 2015, 71, 5095–5111.
A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B 2009, 113,
6378–6396.
[19]
It might be argued that also the strong electron-withdrawing effect of
-+NH3 contributes through inductive effects to the reduction of the nu-
cleophilic power of the peroxycarboxylate groups in 2. Yet, reactivity
studies on reactions of benzhydrylium ions with peroxybenzoate ions
[ArC(O)OO–] showed only a marginal effect on the nucleophilicities of
these ions when the ring substituents were varied from p-NO2 to p-OMe
(see ref[1]). We, therefore, assume that ammonium groups in ꢀ- or γ-
position along an aliphatic chain, as in 2a–c, do not significantly affect
the reactivity of the peroxycarboxylate group by inductive effects.
[10] A database of reactivity parameters E, N, and sN can be accessed at
[20]
[21]
Reviews: a) N. J. Fina, J. O. Edwards, Int. J. Chem. Kinet. 1973, 5, 1–26;
b) E. Buncel, I.-H. Um, Tetrahedron 2004, 60, 7801–7825.
http://www.cup.lmu.de/oc/mayr/DBintro.html.
[11] H. Mayr, T. Bug, M. F. Gotta, N. Hering, B. Irrgang, B. Janker, B. Kempf, R.
Loos, A. R. Ofial, G. Remennikov, H. Schimmel, J. Am. Chem. Soc. 2001,
123, 9500–9512.
[12] Detailed procedures for the synthesis of 3-BF4 are given in: R. J. Mayer,
N. Hampel, P. Mayer, A. R. Ofial, H. Mayr, Eur. J. Org. Chem. 2018, 10.1002/
ejoc.201800835.
[13] J. Ammer, M. Baidya, S. Kobayashi, H. Mayr, J. Phys. Org. Chem. 2010, 23,
1029–1035.
[14] S. Minegishi, H. Mayr, J. Am. Chem. Soc. 2003, 125, 286–295.
[15] I. A. Yaremenko, V. A. Vil', D. D. Demchuk, A. O. Terent'ev, Beilstein J. Org.
Chem. 2016, 12, 1647–1748.
G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, A. Nudelman,
B. M. Stoltz, J. E. Bercaw, K. I. Goldberg, Organometallics 2010, 29, 2176–
2179.
[22]
[23]
E. Harder, W. Damm, J. Maple, C. Wu, M. Reboul, J. Y. Xiang, L. Wang, D.
Lupyan, M. K. Dahlgren, J. L. Knight, J. W. Kaus, D. S. Cerutti, G. Krilov,
W. L. Jorgensen, R. Abel, R. A. Friesner, J. Chem. Theory Comput. 2016,
12, 281–296.
MacroModel (Schrödinger Release 2017–1), Schrödinger, LLC, New York,
NY, 2017.
[16] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li,
M. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Men-
Received: July 24, 2018
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
8
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Reaction Kinetics
R. J. Mayer, A. R. Ofial* .................... 1–9
Intramolecular Hydrogen-Bonding
Modulates the Nucleophilic Reactiv-
ity of Ammonium-Peroxycarboxyl-
ates
The kinetics of the reactions of Quantum chemical calculations indi-
ammonium-peroxycarboxylates with cate intramolecular N–H···O hydrogen
benzhydrylium ions (Ar2CH+) were in- bonding in ꢀ- and γ-ammonium-sub-
vestigated in aqueous solutions at stituted peroxycarboxylates, which ac-
20 °C (pH 8–9). Application of the counts for their attenuated nucleo-
Patz–Mayr equation (1) furnished their philicities relative to those of other
nucleophilicity parameters (N, sN). peroxycarboxylates.
DOI: 10.1002/ejoc.201801158
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
9
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim