Mechanistic Insights into Substituent Effects on Reactivity of 2-(Methoxymethyl)benzene-1,4-diamine

Article abstract of DOI:10.1002/kin.21110

In the series of benzene-1,4-diamines (p-phenylenediamines) investigated, 2-(methoxymethyl)benzene-1,4-diamine (2-methoxymethyl-p-phenylenediamine) is the most slowly oxidized, with removal of the first electron being rate determining. This electron-withd

Full text of DOI:10.1002/kin.21110

Mechanistic Insights into
Substituent Effects on
Reactivity of
2-(Methoxymethyl)benzene-
1,4-diamine
AARON D. BAILEY,1 GUIRU ZHANG,1 BRYAN P. MURPHY2
1Coty Inc, Cincinnati, OH, 45237
2Coty Inc, Blue Ash, OH, 45241
Received 7 April 2017; revised 2 July 2017; accepted 2 July 2017
DOI 10.1002/kin.21110
Published online in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: In the series of benzene-1,4-diamines (p-phenylenediamines) investigated, 2-
(methoxymethyl)benzene-1,4-diamine (2-methoxymethyl-p-phenylenediamine) is the most
slowly oxidized, with removal of the first electron being rate determining. This electron-
withdrawing methoxymethyl group also drove a faster coupling step with 3-aminophenol
(m-aminophenol) than for the analogs with electron-donating groups. However, the series
parent, benzene-1,4-diamine, exhibited the fastest coupling step. Since both N(1) and N(4)
of the primary intermediates react with m-aminophenol, it seems that steric hindrance by the
2-substituents slows the overall rate. When 3-amino-2,6-dimethylphenol is the coupler, the
kinetics are biphasic. Nonproductive adducts are formed reversibly with all the primary inter-
mediates by ipso attack at C(6). For 2-(methoxymethyl)benzene-1,4-diamine, formation of this
nonproductive adduct is favored more than for the other compounds in the series, in what
seems to be a kinetically rather than thermodynamically controlled process. This slows the
ꢀC
overall rate of color formation.
2017 Wiley Periodicals, Inc. Int J Chem Kinet 1–14, 2017
INTRODUCTION
Correspodence to: Bryan P. Murphy; e-mail: bpm8@aol.com.
Present Addresses: Aaron D. Bailey, ANGUS Chemical Co.,
Buffalo Grove, IL 60089.
Oxidative hair colorants have been popular for
well over a hundred years. This technology re-
quires that small, colorless molecules combine,
generally mediated by H2O2, to produce color.
Supporting Information is available in the online issue at
www.wileyonlinelibrary.com.
ꢀC
2017 Wiley Periodicals, Inc.

2
BAILEY, ZHANG, AND MURPHY
Scheme 1 Representative reaction of a 1 and 4, in which only the dimeric indo dye product is possible [1,2]. Nitrogen N(1) is
ortho to the carbon to which R1 is bonded, and N(4) is para to N(1).
In this process, primary intermediates such as
benzene-1,4-diamines (p-phenylenediamines) and 4-
aminophenols (p-aminophenols) are oxidized to p-
benzoquinonedi- or monoimines, respectively. Sub-
sequent electrophilic attack on couplers such
as 3-aminophenols (m-aminophenols), benzene-1,3-
diamines (m-phenylenediamines), and resorcinols
yield diphenylamines (leuco dyes), which are oxidized
rapidly to form the final indo dye (azomethine). For
example, Scheme 1 shows simplified coupling chem-
istry between benzene-1,4-diamine (1) and the blocked
coupler, 5-amino-2-methylphenol (4).
In addition to improving the toxicological proper-
ties of benzene-1,4-diamines, substituents on primary
intermediates, and also couplers, are used to fine-tune
properties such as color, reactivity, solubility, and in-
tensity of the resultant indo dyes. For example, ad-
dition of electron-donating or electron-withdrawing
groups to the primary intermediate or the coupler can
be used to manipulate the rates and extents of reaction.
For the 1,3-disubstituted benzenes that are common
couplers, blocking potential reaction sites also affects
rates and limits the possibilities of the dyes that can be
formed [1,2,5].
Determination of rates of individual steps to form an
indo dye is difficult when unblocked couplers are used.
For example, the complexity of the coupling of 1 with
resorcinol has most recently been demonstrated by
Bailey et al. [6]. Therefore, when studying the reactions
of primary intermediates and couplers, it is common
practice to block sites in the coupler so study of the
first coupling reaction, i.e. when no oligomerization
occurs, is more straightforward. Additionally, when
Recently, advances to the chemistry and perfor-
mance of benzene-1,4-diamine analogs have included
addition of substituents to the aromatic ring of benzene-
1,4-diamines and have made the dye-forming reac-
tion pathway potentially more complex than what is
pictured in Scheme 1. These advances have occurred
because some of the more popular primary interme-
diates such as 1 and 2-methylbenzene-1,4-diamine
(p-toluenediamine; 7) have come under increased reg-
ulatory scrutiny over the past few decades. However,
the haircolor industry has been successful in replac-
ing materials of concern over the past 40 years in its
drive for improved performance and safety in hair dye
chemistry. For example, we have recently developed
and commercialized 2-(methoxymethyl)benzene-1,4-
diamine (8) as a primary intermediate that replaces 1
and 7 and that offers reduced allergenic potential [3,4].
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

SUBSTITUENT EFFECTS ON REACTIVITY of 2-(METHOXYMETHYL)BENZENE-1,4-DIAMINE
3
studying mechanisms of oxidation and dye formation,
use of potassium ferricyanide is a well-accepted ap-
proach [7,8]. This simplifies the kinetics by making
the rate of oxidation much more rapid than the rate
of coupling, thus allowing a more informative analysis
of the coupling kinetics. Since our goal was to under-
stand the effect of the methoxymethyl group on both
rates of oxidation and coupling, we adopted these ap-
proaches, with one exception: We did also study the
parent coupler (4) for comparison.
The above-mentioned understanding of dye prod-
ucts, rates of reaction, mechanisms of formation, and
competitive kinetics with other oxidation dye inter-
mediate pairs informs both the shade development
work required to commercialize new haircolor prod-
ucts that use these materials, and also the design of
next-generation dye intermediates. However, as new
dye oxidation dye precursors have been introduced
recently [9–11], the kinetics and mechanisms of dye
formation from these materials have not been stud-
ied as extensively as for previous materials, as in the
work of Corbett, Brown, Tong, and others, particu-
larly for the primary intermediates. Recently, though,
Mohr-Hautavoine et al. have published important in-
formation on the use and the comparative color deliv-
ery properties of 8 relative to 7, another commercial
analog [12].
Buffer refers to the corresponding phosphate buffer
adjusted to pH 7.9, 9.25, 9.45, or 10.3, and I = 0.2 M
with KCl. The buffer preparation details can be found
in the Supporting Information.
Instruments
Stopped-flow experiments and absorbance spectra
were collected on an Olis-RSM 1000 from Olis Inc,
Bogard, GA, USA, with U.S.A. stopped-flow ac-
cessory using 120 µM slit widths and gratings of
400 lines/mm. The sample cell has a path length of 2 cm
and a volume of 35 µL. All reactions were performed
by flowing 0.25 mL of each reagent through the cell
using a pneumatic piston pressurized by 80 psig com-
pressed air. Temperature was controlled using a Julabo
CF31 cryo-compact circulator from Julabo USA Inc.,
Allentown, PA, USA. Cyclic voltammetry experiments
were conducted using a BASi Epsilon potentiostat and
electrochemical cell from Bioanalytical Systems Inc.
West Lafayette, IN, USA. Three electrodes were used:
Pt working electrode, Pt wire auxiliary electrode, and
a Ag/AgCl reference electrode, and KCl was used as
the electrolyte.
General Procedure for Stopped-Flow
Reaction of a Primary Intermediate with
3-Aminophenol (4) Using K3Fe(CN)6 as
Oxidant
In general, the color from the indo dyes formed
from 8 was slightly redder than the color from those
formed from 7. Our interest was to understand the
mechanistic drivers of their results in hair dyeing and to
determine whether the same properties that shifted the
color wouldalsoresult inachangeinrateor mechanism
of color formation from 8.
To a 5-mL volumetric flask was added a primary in-
termediate and phosphate buffer (I = 0.2 M) at the
appropriate pH (9.25 or 10.3) until the final volume
was 5 mL. The resulting mixture was sonicated in a
30°C bath until dissolved. An aliquot was added to a
10-mL volumetric flask, so that the final concentration
of the primary intermediate was 100 µM. To a 5-mL
volumetric flask was added 4 (19.6 mg, 180 µMol) and
the appropriate pH phosphate buffer until the final vol-
ume was 5 mL. The resulting mixture was sonicated
in a 30°C bath until dissolved. An aliquot (2.784 mL)
was taken from the stock solution and mixed with the
volumetric flask containing the primary intermediate
aliquot. The solution was diluted to 10 mL with phos-
phate buffer (I = 0.2 M) at the appropriate pH bringing
the concentration of 4 to 10 mM. The final solution was
loaded into a 10-mL disposable syringe and placed at
stopped-flow injection slot A. To a 5-mL volumetric
flask was added K3Fe(CN)6 (45.9 mg, 139 µMol) and
phosphate buffer (I = 0.2 M) at the appropriate pH to
the to bring the volume to 5 mL. The resulting mixture
was sonicated in a 30°C bath for 5 min. An aliquot
(143 µL) was taken from the stock solution and di-
luted with the phosphate buffer (I = 0.2 M) at the
EXPERIMENTAL
Materials
Benzene-1,4-diamine (p-phenylenediamine; 1) and 3-
aminophenol (m-aminophenol; 11) were purchased
from Jos. H. Lowenstein and Sons, NY, NY, USA,
and were sublimed prior to use. All other commer-
cial chemicals were acquired from Sigma-Aldrich,
St. Louis, MO, USA, and were used as received.
Water refers to deionized water. 2-Methylbenzene-
1,4-diamine sulfate salt, 2-(methoxymethyl)benzene-
1,4-diamine, 3-amino-2,6-dimethylphenol, and 2-
propylbenzene-1,4-diamine dihydrochloride salt and
were generously supplied by Dr. John Gardlik of Proc-
ter & Gamble, Cincinnati, OH, USA. Compressed
gases were provided by Wright Brothers, Inc, Mont-
gomery, OH, USA.
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

4
BAILEY, ZHANG, AND MURPHY
appropriate pH to 10 mL in a volumetric flask so that
the final concentration was 400 µM. The solution was
loaded into a syringe and replaced the buffer syringe
located in injection slot B. The kinetic runs were car-
ried out by injecting 0.25 mL of each syringe through
the observation cell, bringing the reagents to the fol-
lowing final concentrations: [primary intermediate] =
50 µM, [4] = 5 mM, and [K3Fe(CN)6] = 200 µM.
The rate was determined by observing the change in
absorbance at the λmax. The concentration of the dye
formed was calculated using Beer’s law, and the ex-
tinction coefficient of the dye was determined by the
absorbance at tꢀ. The first-order rate constant obtained
under pseudo–first-order conditions was calculated by
assuming a 1:1 mol conversion of primary to dye and
plotting the ln[primary intermediate] versus time.
through the observation cell, bringing the reagents
to the following final concentrations: [primary inter-
mediate] = 50 µM, [3-amino-2,6-dimethylphenol] =
5 mM, and [K3Fe(CN)6] = 200 µM. The rate was de-
termined by observing the change in absorbance at the
λmax. The concentration of the dye formed was calcu-
lated using Beer’s law, and the extinction coefficient of
the dye was determined by the absorbance at tꢀ. There
are two phases for the reaction to form the dye species:
(1) a rapid initial conversion directly to the dye and
(2) a slower conversion from a nonproductive product
formed by a side equilibrium. The first-order rate con-
stant obtained under pseudo–first-order conditions for
each step was calculated by assuming a 1:1 mol con-
version of primary to dye and plotting the ln[primary]
versus time.
General Procedure for Stopped-Flow
Reaction of a Primary Intermediate with
3-Amino-2,6-dimethylphenol Using
K3Fe(CN)6 as Oxidant
General Procedure for Stopped-Flow
Oxidation of a Primary Intermediate by
K3Fe(CN)6
To a 5-mL volumetric flask was added a primary in-
termediate and phosphate buffer (I = 0.2 M) at the
appropriate pH (7.9, 9.45, or 10.3) until the final vol-
ume was 5 mL. The resulting mixture was sonicated in
a 30°C bath until dissolved. An aliquot was taken from
the stock solution and diluted with phosphate buffer
(I = 0.2 M) at the appropriate pH to 10 mL in a volu-
metric flask so that the final concentration was 100 µM.
The final solution was loaded into a 10-mL disposable
syringe and placed at stopped-flow injection slot A. To
a 5-mL volumetric flask was added K3Fe(CN)6 (32.6
mg, 99.0 µMol) and phosphate buffer (I = 0.2 M) at
the appropriate pH to bring the final volume to 5 mL.
The resulting mixture was sonicated in a 30°C bath for
5 min. An aliquot (101 µL) was taken from the stock
solution and diluted with phosphate buffer (I = 0.2 M)
at the appropriate pH to 10 mL in a volumetric flask so
that the final concentration was 200 µM. The solution
was loaded into a syringe and placed at stopped-flow
injection slot B. The Julabo water circulator was set to
25°C. The solutions were allowed to sit in the loading
syringes for 5 min for temperature equilibration. The
kinetic runs were carried out by injecting 0.25 mL of
each syringe through the observation cell, bringing the
reagents to the following final concentrations: [primary
intermediate] = 50 µM and [K3Fe(CN)6] = 100 µM.
The rate was determined by observing the change in
absorbance at 420 nm. K3Fe(CN)6 concentration was
calculated by using ε = 1000 dm3 mol−1 cm−1 at 420
nm and using the Beer’s law. The initial rate was cal-
culated by taking the slope of the line of [K3Fe(CN)6]
versus time over the first 10% of conversion.
To a 5-mL volumetric flask was added a primary in-
termediate and phosphate buffer (I = 0.2 M) at the
appropriate pH until the final volume was 5 mL. The
resulting mixture was sonicated in a 30°C bath until
dissolved. An aliquot was added to a 10-mL volumet-
ric flask and diluted with phosphate buffer (I = 0.2 M)
at the appropriate pH, so that the final concentration of
the primary intermediate was 100 µM. To a 5-mL vol-
umetric flask was added 3-amino-2,6-dimethylphenol
(25.4 mg, 184 µMol) and phosphate buffer (I = 0.2 M)
at the appropriate pH until the final volume was 5
mL. The resulting mixture was sonicated in a 30°C
bath until dissolved. An aliquot (2.700 mL) was taken
from the stock solution and mixed with the volumetric
flask containing the primary intermediate aliquot. The
solution was diluted to 10 mL with phosphate buffer
(I = 0.2 M) at the appropriate pH bringing the con-
centration of 3-amino-2,6-dimethylphenol to 10 mM.
The final solution was loaded into a 10 mL dispos-
able syringe and placed at stopped-flow injection slot
A. To a 5-mL volumetric flask was added K3Fe(CN)6
(34.7 mg, 105 µMol) and phosphate buffer (I =
0.2 M) at the appropriate pH to bring the final volume
to 5 mL. The resulting mixture was sonicated in a 30°C
bath for 5 min. An aliquot (190 µL) was taken from
the stock solution and diluted with phosphate buffer
(I = 0.2 M) at the appropriate pH to 10 mL in a volu-
metric flask so that the final concentration was 400 µM.
The solution was loaded into a syringe and replaced the
buffer syringe located in injection slot B. The kinetic
runs occurred by injecting 0.25 mL of each syringe
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

SUBSTITUENT EFFECTS ON REACTIVITY of 2-(METHOXYMETHYL)BENZENE-1,4-DIAMINE
5
electron-donating groups to the primary intermediate,
e.g., as a result of bis(2-hydroxyethylation) of the N(4)
nitrogen, is well known [2,5]. In general, relative to the
parent compound (1), electron-donating groups in the
primary intermediate increase the rate of oxidation but
decrease the rate of electrophilic coupling. Electron-
withdrawing groups in the primary intermediate de-
crease the rate of oxidation, but increase the rate of
coupling. Substituent effects are the reverse for cou-
plers: Electron-withdrawing groups decrease the rate
of coupling, and electron-donating groups increase the
rate of coupling [2,5]. The kinetics of color forma-
tion from 2-substituted-benzene-1,4-diamines is not as
well-studied as benzene-1,4-diamine, perhaps due to
the potential complications that result from formation
of product mixtures arising from coupling through the
nonequivalent N(1) and N(4) nitrogens. This is true of
8; the methoxymethyl group has not been used in oxi-
dation hair dye chemistry until recently, and its effects
on rates of oxidation of the primary intermediate and
the subsequent reaction of its oxidation product (the
benzoquinonediiminium ion) with a coupler have not
been explored. This is important as we understand the
effect of the methoxymethyl substituent on the overall
process.
Cyclic Voltammetry of a Primary
Intermediate
A 0.1 M solution of KCl in water was used as the
electrolyte. It was purged with N2 for 15 min prior
to collecting a background scan without the primary
intermediate. Once the background scan had been col-
lected, the primary intermediate (40 µMol) was added
to electrolyte solution (5 mL) and pH adjusted to
10.0
0.02 using 0.1 M KOH. The solution was
brought to 10 mL using the electrolyte solution and
purged with N2 for 2 min. Cyclic voltammograms were
acquired from 0.6 to 1.2 V at 100 mv/s scan rate. After
each run, the solution was allowed to stir for 1 min to
ensure no solids were deposited on the probe.
RESULTS AND DISCUSSION
It has been observed that at alkaline pH color forma-
tion from 8 and some couplers is qualitatively different
(and perhaps slower) than from 7. To help understand
what might be driving this, we looked at two of the po-
tential causes: rate differences for oxidation of 8 and
reaction of its diimine with blocked and unblocked
couplers.
Compound 8 also has the potential to exist in an
H-bonded form (8a), as can its diimine (9) (Fig. 1).
Owing to its electronegativity and its ability to form
an H-bond with an adjacent amino or imino hydrogen,
the oxygen atom of the methoxymethyl side chain of
8 has the potential to affect the electron density at the
nitrogens attached to carbons 1 (N(1)) and 4 (N(4))
of 2-(methoxymethyl)benzene-1,4-diamine (8), its ox-
idized forms, and any of the intermediates or transition
states in the dye-forming reaction (Fig. 1). Determining
the impact on rate of coupling will help us understand
this effect. Additionally, in either the non-H-bonded
or H-bonded forms, there is the potential for steric
interaction of the methoxymethyl side chain with the
coupler to affect the rates.
Final indo dye formation from a primary intermedi-
ate and a blocked coupler such as 4 involves a two-step
oxidation of the primary intermediate, electrophilic at-
tack of the diimine on the coupler to form the leuco
dye, and oxidation of the leuco dye to form the indo
dye (Scheme 1). When potassium ferricyanide is used
to mediate the oxidations, these steps drop from the
rate expression because they are extremely rapid rel-
ative to the coupling step [1]. This allows straightfor-
ward derivation of a rate expression and determination
of the rate constant for the coupling step. For our initial
investigation of the coupling kinetics of 8, we followed
this convention.
The effect on final dye color and the rates of the
oxidation and coupling steps caused by addition of
Figure 1 Compound 8 and its diimine (9) showing the H-bonded and non-H-bonded forms.
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

6
BAILEY, ZHANG, AND MURPHY
Figure 2 Series of benzene-1,4-diamines and 3-aminophenols used as probes to elucidate contributions of steric and electronic
effects on oxidation and coupling reactions in indo dye formation. For ring-substituted benzene-1,4-diamines, nitrogen N(1) is
ortho to the carbon bearing the alkyl substituent, and N(4) is para to N(1).
Given that the electronic or steric differences from
1 and 7 could affect the overall rate, our goal was
to understand at which step (if any) the effect was
manifested, and the specific cause for that effect.
cause conversion of the leuco dye to the indo dye (the
final step of Scheme 2) is significantly faster than the
attack of the p-benzoquinonediimine on the coupler
under these conditions.
In the case of 11, Fig. 3 shows that the relative
order of the coupling rates is 1 > 8 > 7ࣃ10. Recall-
ing the general rule that any substituent that decreases
the electron density at the nitrogen that attacks the
electrophilic coupler, two things are apparent: (1) The
methoxymethyl group is acting to withdraw electrons
by H-bonding or inductive effects and is increasing the
electrophilicity of the diimine of 8 relative to the other
2-substituted-benzene-1,4-diamine analogs (7 and 10),
and (2) 1 seems to be out of order based on electron
density at the reactive nitrogen. Additionally, the nearly
identical rates of coupling for 7 and 10 support the im-
portance of the role of electron donation to the ring
by two alkyl substituents with similar Hammett σ val-
ues (Me, σm = −0.07; σp = −0.17; Pr, σm = −0.06;
σp = −0.13) [13].
Effect of the Methoxymethyl Side Chain on
Rates of Coupling
We used a series of benzene-1,4-diamines that poten-
tially could help elucidate the roles of steric, electronic,
and H-bonding effects on the rate of coupling (Fig. 2).
Relative to 1, 7 has the added electron-donating
ability of the methyl group, without a large potential
steric interaction. 2-Propylbenzene-1,4-diamine (10)
is sterically similar to 8, but without the electroneg-
ative oxygen with its ability to H-bond. For this
comparison among these primary intermediates, we
used a sterically unhindered coupler, 3-aminophenol
(11) and a sterically hindered coupler, 3-amino-2,6-
dimethylphenol (12).
The oxidation of benzene-1,4-diamine to its di-
imine, as ferricyanide is converted to ferrocyanide, is
an equilibrium process. Corbett has demonstrated that
reaction must be above approximately pH 8–8.5 for
quantitative generation of the p-benzoquinonediimine
which is necessary to isolate and understand the cou-
pling step [7]. The final two steps in indo dye formation
(formation and oxidation of the leuco dye) also can be
viewed in aggregate as the rate of color formation, be-
The Hammett σ values do not explain why cou-
pling of 11 with 1 is faster than for 8 (CH2OCH3,
σm = −0.10; σp = 0.03) [14]. If the reaction rate
were controlled purely by the electronics, the series
should be 8 > 1 > 7 ࣃ 10. For coupling of any of
these primary intermediates through the unhindered
N(4) with 11, we would not expect any steric in-
teractions that would decrease the rate. So with the
exception of 1, we could ascribe the differences to
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

SUBSTITUENT EFFECTS ON REACTIVITY of 2-(METHOXYMETHYL)BENZENE-1,4-DIAMINE
7
Scheme 2 General reaction between 2-substituted-benzene-1,4-diamines and 3-aminophenols in which one or both of the 2-
and 6-positions may be substituted (blocked). Positional isomers for the cases in which R1 is not –H are shown. Nitrogen N(1)
is ortho to the carbon to which R1 is bonded, and N(4) is para to N(1).
electronic factors. However, what must be considered
in analysis of the rate data is that for 2-substituted-
benzene-1,4-diamines, coupling at both nitrogens is
possible (Scheme 2), and N(1) (the nitrogen on the
carbon ortho to the alkyl substituent) is more steri-
cally hindered. In fact, it has been shown that although
reaction at N(4) (the nitrogen on the carbon meta to
the alkyl substituent) predominates, there is a signif-
icant amount of N(1) coupling product (Scheme 2)
[15,16]. Slower reaction of the N(1) nitrogen could
explain why the overall rates for this series seems
to be out of order based on electronics with respect
to 1, in which the two nitrogens are equivalent, and
there is no possibility of a rate disparity between the
nitrogens.
To provide further insight into whether steric inter-
actions between the 2-substituent of the benzene-1,4-
diamine derivative affected coupling rate, we used 12
as the coupler. If electronic effects dominate, electron
donation by the two methyl groups and a concomitant
increase in nucleophilicity at the reaction site should
result in a rate increase relative to 11. If steric inter-
actions significantly slow the rate of coupling with the
N(1) nitrogen, the situation should be exacerbated by
use of 12. In that case, in some collisions between the a
coupler–diimine pair either the amino or methyl group
could interact with the 2-substituent of the diimine, and
the collisions would not result in product formation.
As can be seen in Fig. 4 for reaction of 12 and
8, conformation (b) is unfavorable and would limit
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

8
BAILEY, ZHANG, AND MURPHY
Figure 3 Observed rate constants under pseudo–first-order conditions as a function of pH for coupling of subject benzene-
1,4-diamines with 3-aminophenol with K3Fe(CN)6 as oxidant in phosphate buffer (I = 0.2 M). [Color figure can be viewed at
wileyonlinelibrary.com]
Figure 4 Two of the possible scenarios of coupler approaching benzoquinonediimine least hindered (a) and most hindered
(b) configurations for coupling of 8 to 12. The atoms and lone pairs are represented as gray for carbon, blue for nitrogen, red for
oxygen, white for hydrogen, and pink for an electron lone pair. [Color figure can be viewed at wileyonlinelibrary.com]
electrophilic attack on 12 by 8. Of course, rotation on
the C(3)–C(6) axis alleviates this poor orientation, but
the possibility of unfavorable intermolecular interac-
tion, or the requirement for specific geometric interac-
tion of 12 and 8, has the potential effect of decreasing
overall rate because not every intermolecular interac-
tion (collision) can lead productively to dye formation.
We would expect similar steric interactions for 8 and
10, perhaps less for 7 and none for 1. This could cause
an even greater disparity in coupling rates between 1
and the other primary intermediates.
dicating that the process for 12 was different than for
11. For reactions with 12 the overall rate of color for-
mation (kobs) is approximately 103 times slower than
for 11 (Fig. 5).
Although the overall rate of coupling of the
benzene-1,4-diamines with 12 was significantly slower
than with 11, we observed that initial rates of color for-
mation are up to approximately 10 times faster than for
the analogous reactions with 11, as might be expected
for a nucleophile that is more electron rich than the
parent compound (Fig. 6). The initial rates of color for-
mation were measured under pseudo–first-order con-
ditions over the first few seconds of the reaction and
were analyzed using the integrated rate law.
In fact, the kinetics were not straightforward; we ob-
served a biphasic kinetic plot, with a faster initial phase
and a slower, eventually complete color formation, in-
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

SUBSTITUENT EFFECTS ON REACTIVITY of 2-(METHOXYMETHYL)BENZENE-1,4-DIAMINE
9
Figure 5 Observed rate constants for color formation under pseudo–first-order conditions at 25°C in phosphate buffer (I =
0.2 M) for 12 with subject benzene-1,4-diamines as a function of pH with K3Fe(CN)6 as oxidant. [Color figure can be viewed
at wileyonlinelibrary.com]
Figure 6 Rates of coupling (color formation) at 25°C in phosphate buffer (I = 0.2 M) between 12 and diimine 13 (Scheme 3)
for the fast initial reaction. [Color figure can be viewed at wileyonlinelibrary.com]
At both pH 9.25 and 10.3, contrary to the case
with 11, with 12, the primary intermediates that are
more electron rich than 8 have a higher rate of initial
color formation. This is the reverse of what would
be expected if there were no competing reactions.
As mentioned earlier, electron-donating groups on
the diimine should slow the reaction and electron-
withdrawing groups should increase the rate of cou-
pling. Regardless, at pH 9.25 there is an indica-
tion that although the Hammett σ values for methyl
(σm = −0.07; σp = −0.17) and propyl (σm = −0.06;
σp = −0.13) are similar, there may be an effect exerted
by increased steric interaction of the propyl group on
rate.
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

10
BAILEY, ZHANG, AND MURPHY
Figure 7 Percent of total dye formation during the fast initial productive coupling step at 25°C between 12 and 13 at pH 9.25
and 10.3 in phosphate buffer (I = 0.2 M). [Color figure can be viewed at wileyonlinelibrary.com]
Given the biphasic nature of the reaction plots, the
above indicates that an equilibrium process that has a
slow reverse step is competing with the color-forming
reaction, and that the diimines of 1 and 8 may react
in a nonproductive manner more so than those of 7
and 10. Analysis showed approximately 15–50% initial
conversion to the indo dye, depending on pH and which
primary intermediate was used, followed by a slower
second phase in which complete conversion occurred
(Fig. 7).
Corbett has noted a similar situation for 1 and 4 in
which there was a nonproductive coupling (i.e., a dye is
not formed) of 1 via electrophilic attack at the carbon
ipso to the methyl group [1]. Apparently the methyl
group is not completely efficient as a blocking group,
but not unexpectedly, neither is it a good nucleofuge,
such as a methoxy group [1,17–19]. So, although at-
tack can occur to generate structure 14, the coupling is
nonproductive and reversible. In Scheme 3, we show
the process that begins with coupling of 12 and the
p-benzoquinonediimine derivatives.
the same behavior for 12. Above pH 8, the equilib-
rium between 1 and 13 lies far to the right, and the
conversion to the diimine is “virtually instantaneous”
[7,20]. Corbett also has reported [20] when a 1:4:1
ratio of a benzene-1,4-diamine:ferricyanide:coupler is
used, both the benzene-1,4-diamine and the leuco dye
are oxidized by ferricyanide, and the oxidation of the
leuco dye occurs after the rate-limiting step [20]. In
our case, i.e. Scheme 3, this means that coupling of 13
and 12 is the rate-determining step. Therefore, since
we used a 4:1 molar ratio of ferricyanide:benzene-1,4-
diamine above pH 8, the conversions of the benzene-
1,4-diamine derivative to leuco dye 13, and conversion
of 15 to 16 are rapid; the kinetic expression is simpli-
fied (vide infra). Assuming that the oxidations of the
are fast, the equations for Scheme 3 simplify as they
did in Corbett’s investigation, and we derive below the
kinetic expression for oxidative coupling of 12 and
13:
ꢀ
ꢁ
d [16]
dt
= k3 [15] FeIII
(1)
This nonproductive, reversible coupling of 12 with
this series of benzene-1,4-diamines made the overall
rates of dye formation much more similar across the
series, and slower overall, than for coupling with 11
(Fig. 5).
Assuming that 15 is at steady state, solving for [15]
gives
It appears that the nonproductive pathway plays an
important role in controlling the overall kinetics. Cor-
bett has observed this for the case of benzene-1,4-
diamine and 5-amino-2-methylphenol [1], and we see
ꢀ
ꢁ
d [15]
dt
= k1[12][13] − k3[15] FeIII − k−1[15] ꢁ 0
(2)
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

SUBSTITUENT EFFECTS ON REACTIVITY of 2-(METHOXYMETHYL)BENZENE-1,4-DIAMINE
11
Scheme 3 Reaction scheme depicting nonproductive equilibrium and productive color formation from diimine 13 and 12. For
13 nitrogen N(1) is ortho to the carbon to which R is bonded, and N(4) is para to N(1).
ꢂ
ꢃ
k1[12][13]
ꢀ
k
−1[15] + k−2[14]
k1[12] + k2[12]
ꢁ
[13] =
(8)
[15] =
(3)
(4)
k
−1 + k3 FeIII
Substituting for [13] gives
Substituting for [15] gives
ꢂ
ꢃ
ꢀ
ꢁ
d[16]
dt
k
−1[15] + k−2[14]
[12] (k1 + k2)
k1k3[12][13] FeIII
= k1[12]
(9)
d[16]
dt
ꢀ
ꢁ
=
k
−1 + k3 FeIII
ꢂ
ꢃ
d[16]
k
−1[15] + k−2[14]
k1 + k2
If k3[FeIII] ꢁ k−1, then
= k1
(10)
dt
d[16]
Recalling that k3[FeIII] ꢁ k−1, and substituting for
[15] in Eq. (9) gives
= k1[12][13]
(5)
(6)
dt
ꢄ
ꢅ
ꢀ
ꢁ
For the nonproductive, reversible pathway
k
−1k1[12][13]/k3 FeIII + k−2[14]
k1 + k2
d[16]
dt
= k1
d[14]
= k2[12][13] − k−2[14]
dt
(11)
If k3[FeIII] ꢁ k1k−1[12][13], then
once equilibrium has been established, it can be as-
sumed that 13 is in a steady state giving
d[16]
dt
k1k−2[14]
k1 + k2
=
(12)
d[13]
= k−1[15] + k−2[14]
dt
For simplicity, we will define k1k−2/(k1+k2) as kobs
,
− k1[12][13] − k2[12][13] ꢁ 0
(7)
the overall rate of color formation, so we can represent
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

12
BAILEY, ZHANG, AND MURPHY
Scheme 4 Oxidation of 2-substituted-benzene-1,4-diamines proceeds through two, one-electron steps, with aromaticity lost
in the second step. Nitrogen N(1) is ortho to the carbon to which R is bonded, and N(4) is para to N(1).
the equation as
mediate are important, and that as the Highest Occu-
pied Molecular Orbital-Lowest Unoccupied Molecular
Orbital (HOMO-LUMO) interactions change, nonpro-
ductive coupling can increase in importance in the pro-
cess.
d[16]
dt
= kobs[14]
(13)
Although this has been informative and helps guide
future work to fully understand the role that increased
steric interaction might play in controlling the kinetics
of the process, more conclusive work on the proposed
steric interactions is necessary. By comparison of the
results with 7 and 10, this work does show that the
methoxymethyl group of 8 is acting to increase the
electrophilicity of the diiminium nitrogens. However,
it does not explain adequately the observed differences
indicated by Mohr-Hautavoine et al. [12]. Understand-
ing the rates of oxidation of this series of benzene-
1,4-diamines can help us answer our original question,
though.
The trend toward equalization of overall rates at
a given pH is not adequately explained by just non-
productive equilibria, without considering the elec-
tronic properties that drive different rates for k2 and
k−2 among the primary intermediates. Because the ac-
tive species in this type of coupling chemistry are the
cationic diiminium ion and the phenolate [1], as pH is
increased there are competing effects in the observed
rates: an increase due to the increasing concentration of
phenolate and a decrease due to the decreasing concen-
tration of the diiminium ion. Because the pKa should
vary by substitution in each of these diiminium ions,
the effect on the rate will be different for each benzene-
1,4-diamine in our series, and we do not expect a clean
trend in color formation rates across the series with
12 as pH is varied. Although we can make compar-
isons between individual pairs of molecules at a given
pH value, we cannot make broad generalizations across
the series without invoking the further conditions men-
tioned above. Since the Hammett σ values are similar
for 7 (Me, σm = −0.07; σp = −0.17) and 10 (Pr,
σm = −0.06; σp = −0.13), we would expect the rates
to be similar based on electronic factors and degree
of protonation only. This is not the case at pH 9.25,
and the rate disparity between them gives some hint
to the importance of steric interaction as a driver of
rates.
Effect of the Methoxymethyl Side Chain on
Rate of Oxidation
It is known for benzene-1,4-diamines that oxidation to
the corresponding p-benzoquinonediimine happens as
two single electron oxidations (Scheme 4) [7,8,17].
We used 1, 7, and 8 to understand whether the
methoxymethyl group was acting to decrease electron
density, affecting the rate of either of these oxidation
steps.
In the first step, aromaticity is maintained, but when
the second electron is removed, there is loss of aro-
maticity. It is known that the substituents on the pri-
mary intermediates affect rates of oxidation, but it is
unclear in this case whether that effect is through re-
moval of the first or second electron. We used cyclic
voltammetry to first validate that for 1, 7, and 8 each
showed two distinct electron transfer steps. Table I
shows that the oxidation potential for removal of the
first electron (later shown to be rate determining) fol-
lows the series 7 < 1 < 8.
It is obvious that unlike for 11, when 12 is the cou-
pler the initial rate of productive coupling is slower
with 8 than with 7 or 10. This is not explained by
a difference in the percent conversion; all reactions
eventually go to completion. However, the rate of non-
productive coupling (k2 vs. k1) is fastest for 8. Since
all reactions eventually go to completion and form
the expected indo dyes, it appears to be a difference
in kinetic control being operative to a greater extent
for 8. This lends credence to the hypothesis that elec-
tronic effects of the substituents on the primary inter-
Having validated that these benzene-1,4-diamine
derivatives were oxidized in two one-electron steps,
we determined rates of oxidation by potassium
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

SUBSTITUENT EFFECTS ON REACTIVITY of 2-(METHOXYMETHYL)BENZENE-1,4-DIAMINE
13
Table I Average Values for Removal of First and
Second Electrons [E / V vs. Ag/AgCl (3 M KCl), pH 10.0]
for 1, 7, and 8
CONCLUSIONS
In simple reactions such as with 11, clear guid-
ance is available to understand how to ex-
ploit the color-forming reactions of this se-
ries of benzene-1,4-diamines, and in particular 2-
(methoxymethyl)benzene-1,4-diamine. Although the
overall observed rate was somewhat slower for 8 than
for 1, the rate of coupling of 8 with 11 actually is faster
than for the other 2-substituted-benzene-1,4-diamines
(7 and 10).
Compound
E1 (V)
E2 (V)
1
7
8
0.242
0.212
0.275
0.444
0.37
0.423
When more substituted primary intermediates (with
multiple coupling sites) and couplers with additional
substituents are employed, kinetics becomes more
complex. For primary intermediates, it is possible to
add substituents to the molecule to slow or acceler-
ate the rate of oxidation to make it more controllable
while also impacting the HOMO–LUMO interactions
with the coupler. Knowledge of these effects makes it
fairly easy to inform the design of new dye materials
and to direct the chemistry. In addition to these elec-
tronic effects, steric interactions also play an important
role in overall color formation rates, as is demonstrated
for 1 versus substituted primary intermediates.
Removal of the first electron from the primary in-
termediates is rate determining, and the rate constants
for that step follow the expected order, 8 < 1 < 7. As
Feng and Chan showed, this is especially important
in indo dye formation chemistry when milder (slower)
oxidants are used, such as in haircolor products [21],
and the oxidation of benzene-1,4-diamines by H2O2 is
the rate-determining step in color formation.
ferricyanide by monitoring the conversion of ferri-
cyanide to ferrocyanide at 420 nm in a stopped-flow
apparatus. Examining the initial rates and attempting
to fit the data to both first- and second-order equa-
tions verified that the reaction was second order overall
with first-order dependence on both the benzene-1,4-
diamine derivative and potassium ferricyanide. Figure
8 shows the second-order rate constants for complete
oxidation at three pH values.
The series follows exactly the order that would
be expected based on the Hammett σ values of the
substituents: 7 (Me, σm = −0.07; σp = −0.17)
with its electron-donating substituent, is faster than
the benchmark (1) (H, σm = 0.00; σp = 0.00) and
the methoxymethyl group (CH2OCH3, σm = −0.10;
σp = 0.03) of 8 is acting to withdraw electrons from
the ring and to decrease the rate of oxidation relative to
1. In fact, the rate of the first oxidation of 7 is approxi-
mately double that of 8 across the pH range, explaining
the previous observations of Mohr-Hautavoine et al. on
color formation [12].
Figure 8 Rate constants for oxidation of 1, 7, and 8 by K3Fe(CN)6 at 25°C in phosphate buffer (I = 0.2 M) at pH 7.9, 9.45,
and 10.3. [Color figure can be viewed at wileyonlinelibrary.com]
International Journal of Chemical Kinetics DOI 10.1002/kin.21110

14
BAILEY, ZHANG, AND MURPHY
We have not distinguished between the pure induc-
4. Goebel, C.; Troutman, J.; Hennen, J.; Rothe, H.;
Schlatter, H.; Gerberick, G. F.; Blo¨meke, B. Toxicol
Appl Pharmacol 2014, 274, 480–487.
5. Murphy, B. P. Hair Colorants, In: Poucher’s Perfumes,
Cosmetics and Soaps, 10th Ed., Butler, H. Ed.; Kluwer
Academic Publishers: London; 2000: Chapter 10.
6. Bailey, A. D.; Murphy, B. P.; Guan, H. J Phys Chem A
2016, 120, 8512–8520.
7. Corbett, J. F. J Chem Soc (B) 1969, 207–212.
8. Tong, L. K. J. J Phys Chem 1954, 58, 1090–
1097.
9. Henkel AG & Co. KGaA. German Patent Application
DE 10 2008 061 864 A1, 2010.
tive effect of the methoxymethyl group and the con-
tribution of H-bonding. It is clear from the work of
Corbett and others that the transition state in the rate-
determining step should involve a charge, and there-
fore future work investigating kinetic isotope and salt
effects should yield answers on this point. Addition-
ally, we plan to investigate more thoroughly the role of
H-bonding through NMR and crystallography.
Finally, we have not examined the mechanism of the
reverse step of nonproductive coupling. Further work
is needed to determine how the seemingly kinetically
preferred product (14) is being converted to indo dye
16.
10. Henkel AG & Co. KGaA. European Patent EP 1 765
267 B1, 2010.
11. L’Oreal. World Patent Application WO 2010133639 A1,
2010.
12. Mohr-Hautavoine, C. V.; Herrlein, M.; Guthrie, J. T.
Dyes Pigments 2015, 117, 157–162.
AUTHOR CONTRIBUTIONS
13. Hansch, C.; Leo, A.; Taft, R. W. Chem Rev 1991, 91,
165–195.
The manuscript was written through contributions of
all authors. All authors have given approval to the final
version of the manuscript.
14. Perrin, D. D.; Dempsey, B.; Serjeant, E. P.
pKaPrediction for Organic Acids and Bases; Springer
Publishing Company: New York, NY, 1981.
15. Scientific Committee on Consumer Products, Opin-
ion on Intermediates and reaction products of oxida-
tive hair dye ingredients formed during hair dyeing,
SCCP/1198/08, Adopted at the 19th plenary meeting
of 21 January 2009.
NOTES
The authors declare no competing financial interest.
16. Scientific Committee on Consumer Products, Opinion
on Reaction Products of Oxidative Hair Dye Ingredients
Formed during Hair Dyeing Processes, SCCS/1311/10,
Adopted at the 8th plenary meeting of 21 September
2010.
The authors acknowledge Dr. Bob Strife for acquiring the
HRMS data, Dr. Keith Brown for ongoing discussions and
insights on oxidation dye chemistry, Dr. John Gardlik for his
helpful comments on mechanistic details, and Dr. Bill Laidig
for his molecular modeling and calculations.
17. Corbett, J. F.; Brown, K. C. J Chem Soc, Perkin Trans 2
1972, 1125–1131.
18. Corbett, J. F. J Chem Soc, Perkin Trans 2 1972, 999–
1005.
BIBLIOGRAPHY
19. Brown, E. R. In: The Chemistry of Quinoid Compounds,
Vol. II; Patai, S.; Rappoport, Z. Eds.; Wiley : Hoboken,
NJ, 1988; pp. 1231–1292.
20. Corbett, J. F. J Soc Cosmet Chem 1973, 24, 103–134
21. Feng, Y.; Chan, A. J Soc Cosmet Chem 1994, 45, 299–
308.
1. Corbett, J. F. J Chem Soc, Perkin Trans 2 1972, 539–
548.
2. Morel, O. J. X.; Christie, R. M. Chem Rev 2011, 111,
2537–2561.
3. Wella AG, US Patent 6, 2003 558, 433 B2, 2003.
International Journal of Chemical Kinetics DOI 10.1002/kin.21110