Mechanistic insights into oxidative oligomerization of p?Phenylenediamine and resorcinol

Article abstract of DOI:10.1021/acs.jpca.6b08571

An efficient synthesis of a green dye from oxidative coupling of p-phenylenediamine (PPD) and resorcinol (in a 2:1 ratio) has been developed. Reactivity studies of this dye molecule with a variety of reagents (PPD, resorcinol, the oxidized form of the gre

Full text of DOI:10.1021/acs.jpca.6b08571

Article
pubs.acs.org/JPCA
Mechanistic Insights into Oxidative Oligomerization of
p‑Phenylenediamine and Resorcinol
Aaron D. Bailey,†,‡,§ Bryan P. Murphy,*,†,⊥ and Hairong Guan*,‡
†The Procter & Gamble Company, 8700 Mason-Montgomery Road, Mason, Ohio 45040, United States
‡Department of Chemistry, University of Cincinnati, P.O. Box 210172, Cincinnati, Ohio 45221, United States
S
* Supporting Information
ABSTRACT: An efficient synthesis of a green dye from oxidative coupling of p-phenylenediamine (PPD) and resorcinol (in a
2:1 ratio) has been developed. Reactivity studies of this dye molecule with a variety of reagents (PPD, resorcinol, the oxidized
form of the green dye itself, and a dinuclear indo dye) demonstrate that it cannot be the key reactive intermediate in reported
oxidative oligomerization of PPD and resorcinol. However, the trinuclear species does form large aggregates. At least one viable
pathway of oligomerization has been demonstrated with the dinuclear indo dye.
readily. Since the dyes produced with both oxidants are
identical, this has been accepted as a valid substitution.
It has been much more challenging to generate complete
kinetic profiles for reactions that use unblocked couplers such
as resorcinol, m-aminophenol, or m-phenylenediamine. One of
the earliest clues that additional reactions occur beyond the
dinuclear indo dye was the observation of a short-lived
intermediate with a different color from what would be
suggested by the final dye chromophore.12 In these cases,
kinetic studies have been focused on determining the rates for
the disappearance of the starting materials. A full understanding
of the details of these reactions and the identity of their
products is still lacking.
In early work, the greenish pigment from the PPD−
resorcinol reaction was isolated from a complex mixture
(eight different color bands) by preparative thin-layer
chromatography and identified as a trinuclear species that
incorporates two PPD moieties and one resorcinol.13,14 It was
speculated that this trinuclear species might be formed through
a transient dinuclear intermediate, although such a compound
has never been isolated.12 Furthermore, based on derivatization
experiments it was proposed that an oligomer of at least 18
PPD−resorcinol repeating units was produced under the
INTRODUCTION
■
Permanent hair coloring relies on the reactivity of small
precursors that form dyes inside the hair to enhance its natural
color. The most frequently used materials, p-phenylenediamine
(PPD, 1) or p-aminophenol, known in the hair color industry
as primary intermediates, are oxidized by alkaline hydrogen
peroxide to yield p-quinonediimine (QDI, 2) or p-quinonemo-
noimine, respectively. These electrophiles in turn attack
couplers such as meta-disubstituted benzenes, in which the
two substituents generally are amino or hydroxyl or the
combination of thereof.1,2 As exemplified in Scheme 1 for the
oxidative coupling of PPD and resorcinol, the resulting leuco
form (3) is oxidized to the indo dye (4), which is likely to
undergo further coupling reactions.
Given the complexity of the dye forming process, historically
two strategies have been adopted to simplify mechanistic
investigation. First, some of the nucleophilic sites are blocked
so that the couplers can only react once with an oxidized
primary intermediate. Second, potassium ferricyanide is used as
a substitute oxidant for hydrogen peroxide.2−10 Although
potassium ferricyanide is not typical of an ingredient in
commercial hair products, it does simplify the kinetics when
compared to the more consumer-relevant oxidant, hydrogen
peroxide.11 Rapid oxidation of the primary intermediate by
potassium ferricyanide renders the coupling step rate-limiting.
As a result, the coupling rate constant can be determined
Received: August 24, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jpca.6b08571
J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A
Article
Scheme 1. Mechanism of Oxidative Coupling
oxidative conditions.15 This oligomerization was believed to
occur from additional oxidative coupling of monomers (PPD,
resorcinol), or dinuclear species (3 or 4) to the trinuclear dye
(5).12,15 Despite extensive efforts, no efficient, large-scale
preparation of the trinuclear species is known prior to this
work, nor are the identities of the proposed oligomeric
products and the mechanisms of their formation. Yet, these
details are crucial for the design and improvement of
commercial products.
Our goals for this project were to (a) understand in greater
detail one of these key reaction pairs, PPD and resorcinol, (b)
provide an efficient synthesis of the PPD−resorcinol trinuclear
species, and (c) determine the role and reactivity of the
trinuclear species in generation of the previously proposed
oligomers.
steady stream with stirring. After 8 h, the desired product was
collected by vacuum filtration and washed with water (500
mL). The filtrate contained unreacted starting materials;
therefore, it was collected and placed under oxygen to react
further. The collected solid was washed with EtOAc (500 mL)
and ethanol (100 mL). The 8 h coupling/filtration cycle was
repeated until the purity of the solid collected from the recycled
filtrate was below 90%. The combined solids were triturated
with ethanol and the resulting powder was dried under reduced
pressure to afford 1.33 g (45% yield) of dark powder of >95%
1
purity. H NMR (DMSO, 600 MHz) δ 10.41 (br, 2H, NH),
7.06 (d, J = 8.6 Hz, 4H, CHAr), 6.56 (d, J = 8.6 Hz, 4H, CHAr),
5.88 (s, 1 H, CH), 5.41 (s, 4H, NH2), 5.11 (s, 1H, CH).
13C{1H} NMR (DMSO, 151 MHz) δ 171.6, 152.5, 147.4,
124.5, 123.8, 112.9, 97.4, 88.6. HRMS (ESI): m/z calcd for
[C18H17N4O2]+ 321.13515; found 321.13396. λmax (EtOH/
H2O, pH 10.4), nm (ε/dm3 mol−1 cm−1): 420 (9491), 532 (sh,
3871), 600 (3085).
EXPERIMENTAL SECTION
■
Materials. 1,4-Phenylenediamine (PPD; 1) and resorcinol
were purchased from Jos. H. Lowenstein and Sons and were
sublimed prior to use. All other commercial chemicals were
used as received. Water refers to deionized water. Compressed
O2 and H2 were purchased from Wright Brothers, Inc.
Synthesis of 2,4-Bis(benzyloxy)-1-nitrobenzene (7).16
To a dry round-bottom flask with a magnetic stirbar was added
benzyl alcohol (2.719 g, 25.14 mmol) and MeCN (24 mL).
The solution was stirred at 0 °C while potassium tert-butoxide
(2.962 g, 26.40 mmol) was added in one portion. The addition
caused a rapid rise in temperature, and the reaction mixture was
allowed to cool back to 0 °C. 2,4-Difluoronitrobenzene (2.000
g, 12.57 mmol) was then added dropwise over 15 min so that
the temperature of the reaction mixture did not exceed 15 °C.
The reaction mixture was slowly warmed to room temperature
and stirred overnight. After 14 h, the resulting mixture was
filtered and the filtrate was concentrated in vacuo to afford a
dark solid. Recrystallization of the solid from dichloromethane
and diethyl ether afforded an orange solid. The mother liquors
were collected, concentrated, and the residue was also subjected
to the recrystallization conditions. After three recrystallizations
of the material isolated from the mother liquors, the final
1
Instruments. H and 13C{1H} NMR were recorded on a
600 MHz Bruker spectrometer. Spectra were referenced
internally to solvent residual peaks. Liquid chromatography/
mass spectrometry (LC/MS) analysis was carried out on an
Agilent Hewlet Packard series 1100 high-performance liquid
chromatograph (HPLC) equipped with a UV detector using a
SunFire C18 5 μm column and tandem MS analysis using
electrospray ionization (ESI) by a Micromass ZQ spectrometer.
Preparative HPLC was conducted on a Waters 2535 quaternary
gradient module with a Waters 2489 UV−vis detector. Thin-
layer chromatography (TLC) was performed using Analytech
silica gel plates with fluorescent indicator. Flash chromatog-
raphy was performed using a Teledyne Isco CombiFlash
chromatography system equipped with a UV detector.
1
combined yield for the product was 2.33 g (55% yield). H
NMR (CDCl3, 600 MHz) δ 7.99 (d, 1H, J = 9.0 Hz, CHAr),
7.47 (d, 2H, J = 4.8 Hz, CHAr), 7.42−7.32 (m, 8H, CHAr), 6.66
(d, 1H, J = 2.4 Hz, CHAr), 6.59 (dd, 1 H, J = 9.0, 2.4 Hz, CHAr),
5.19 (s, 2H, CH2), 5.10 (s, 2 H, CH2). 13C{1H} NMR (CDCl3,
151 MHz) δ 163.7, 154.6, 135.6, 135.6, 133.7, 128.9, 128.8,
128.6, 128.5, 128.3, 127.7, 127.0, 106.2, 102.1, 71.3, 70.8.
HRMS (ESI): m/z calcd for [C20H17NO4Na]+ 358.10553;
found 358.10444.
Stopped-flow experiments and absorbance spectra were
collected on an Olis-RSM 1000 with USA stopped-flow
accessory 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 of compressed air. Temperature was controlled using
a Julabo CF31 cryo-compact circulator.
Synthesis of 2,4-Bis(benzyloxy)aniline (8). To a 500 mL
round-bottom flask with a magnetic stir bar and a reflux
condenser was added zinc powder (5.69 g, 87.03 mmol),
NH4Cl (7.71 g, 144.13 mmol), acetone (208 mL), and water
(42 mL). The suspension was stirred while a 2,4-bis-
(benzyloxy)-1-nitrobenzene (2.335 g, 6.963 mmol) was added
Synthesis of Trinuclear Species 5. To a 250 mL round-
bottom flask was added PPD (1.995 g, 18.45 mmol), resorcinol
(1.016 g, 9.23 mmol), and water (60 mL). The pH was
adjusted to 10.4 by adding NH4OH (29% NH3 in water)
dropwise. A pipet attached to a compressed gas was submerged
into the flask and oxygen was bubbled through the solution at a
B
DOI: 10.1021/acs.jpca.6b08571
J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A
Article
in one portion, after which the mixture was heated to reflux and
stirred for 6 h. After cooling to room temperature, the reaction
mixture was filtered and the solid was washed with acetone.
The filtrate was concentrated in vacuo to afford a dark solid,
which was treated with CH2Cl2 and washed with a pH 10
carbonate buffer solution. The precipitated zinc salts were
filtered off and the filtrate was washed with brine. The resulting
organic layer was dried over anhydrous magnesium sulfate,
filtered, and concentrated to afford a dark oil. The crude
product was purified by flash chromatography using hexane/
EtOAc as the mobile phase on a 120 g SiO21column. The pure
product was isolated in 66% yield (1.4 g). H NMR (CDCl3,
600 MHz) δ 7.46−7.33 (m, 10H, CHAr), 6.69 (d, 1H, J = 8.4
Hz, CHAr), 6.65 (d, 1H, J = 2.5 Hz, CHAr), 6.49 (dd, 1H, J =
8.4, 2.5 Hz, CHAr), 5.06 (s, 2H, CH2), 5.00 (s, 2H, CH2), 3.60
(br, 2H, NH2). 13C{1H} NMR (CDCl3, 151 MHz) δ 152.2,
147.4, 137.5, 137.0, 130.5, 128.7, 128.6, 128.1, 127.9, 127.7,
127.6 115.4, 106.4, 101.9, 70.9, 70.6. HRMS (ESI): m/z calcd
for [C20H20NO2]+ 306.14940; found 306.14831.
Synthesis of 2,4-Bis(benzyloxy)-N-(4-nitrophenyl)-
aniline (9). To an oven-dried round-bottom flask equipped
with a magnetic stir bar, a reflux condenser, and a N2 inlet was
charged Pd(OAc)2 (140 mg, 0.624 mmol, 5 mol %), Ph3P (491
mg, 1.87 mmol, 15 mol %), Cs2CO3 (12.21 g, 37.47 mmol),
and 1-bromo-4-nitrobenzene (2.523 g, 12.49 mmol). The flask
was evacuated and refilled with N2 three times prior to the
addition of 8 (4.00 g, 13.1 mmol) in dry toluene (125 mL).
The mixture was heated to reflux using an oil bath and stirred at
reflux. After 24 h, the reaction mixture was cooled to room
temperature and filtered through a pad of SiO2. The solid was
washed with EtOAc, and the combined filtrate was concen-
trated in vacuo to yield an oil. The crude product was
crystallized by dissolving in minimal CH2Cl2 followed by the
addition of hexane. The desired product was isolated as an
orange solid (4.2 g, 79% yield). 1H NMR (CDCl3, 600 MHz) δ
8.07 (d, 2H, J = 9.2 Hz, CHAr), 7.42−7.24 (m, 11H, CHAr),
6.81 (d, 2H, J = 9.2 Hz, CHAr), 6.70 (d, 1H, J = 2.7 Hz, CHAr),
6.59 (dd, 1H, J = 8.7, 2.7 Hz, CHAr), 6.18 (br, 1H, NH), 5.04
(s, 2H, CH2), 5.03 (s, 2H, CH2). 13C{1H} NMR (CDCl3, 151
MHz) δ 156.9, 152.4, 151.4, 139.3, 136.8, 136.2, 128.8, 128.4,
128.3, 127.7, 127.6, 126.3, 123.8, 122.5, 113.3, 106.0, 102.2,
70.9, 70.6. HRMS (ESI): calcd for [C26H22N2O4Na]+
449.14773; found 449.14677.
Synthesis of 4-((4-Aminophenyl)amino)benzene-1,3-
diol (3). To a Fisher-Porter bottle with a magnetic stir bar was
added 9 (500 mg, 1.17 mmol), methanol (10 mL), and 10% wt
Pd on activated carbon (70 mg). The vessel was sealed and
purged seven times with H2 before the H2 pressure was set to
75 psig. The reaction mixture was stirred at room temperature
for 12 h, after which H2 was vented and the bottle was refilled
with N2. The vessel was taken into a drybox, and the contents
were filtered through a pad of dried Celite, which was rinsed
with ethanol. The combined filtrate was concentrated in vacuo
to afford 240 mg (95% yield) of dark solid. The solid was
transferred to a vial and stored under vacuum. 1H NMR
(DMSO, 600 MHz) δ 9.14 (br, 1H, OH), 8.75 (br, 1H, OH),
6.71 (d, J = 8.5 Hz, 1H, CHAr), 6.66 (d, J = 8.8 Hz, 2H, CHAr),
6.43 (d, J = 8.8 Hz, 2H, CHAr), 6.31 (d, J = 2.7 Hz, 1H, CHAr),
6.09 (dd, J = 8.5, 2.7 Hz, 1H, CHAr), 6.01 (br, 1H, NH) 4.39
(br, 2H, NH2). 13C{1H} NMR (DMSO, 151 MHz) δ 151.8,
148.9, 141.6, 135.7, 124.8, 119.3, 118.4, 113.8, 105.5, 103.1. MS
(ES+, m/z): 217.21 ([M + 1]+, 100%).
Oxidative Oligomerization Following Hydrogenolysis
of 9. Compound 9 (60.8 mg, 143 μmol), 10% wt Pd on
activated carbon (70 mg), and ethanol (10 mL) were added to
a Fisher-Porter bottle. The vessel was sealed and flushed with
H2 seven times before the H2 pressure was set to 70 psig. The
mixture was stirred under H2 for 12 h, after which H2 was
vented and the vessel was backfilled with N2. The mixture was
filtered (using a 0.45 μm glass syringe filter) into a flask
containing 5 mL of ethanol. O2 was bubbled through the
solution until the solvent had evaporated, resulting in the
isolation of 50 mg of dark solid. MS (ESI, m/z): 321.20
(C18H17N4O2, [M + H]+ of 5), 429.19 (C24H21N4O4), 643.31
(C36H31N6O6), 855.27 (C48H39N8O8).
Resorufamine Generated from 9. Compound 9 (61.6
mg, 145 μmol), 10% wt Pd on activated carbon (70 mg), and
ethanol (10 mL) were added to a Fisher-Porter bottle. The
vessel was sealed and flushed with H2 seven times before the H2
pressure was set to 70 psig. After stirring at this pressure for 10
h, the excess hydrogen was vented, and the colorless solution
was filtered into a pH 10.4 aq solution containing K3Fe(CN)6
(119 mg, 361 μmol). The resulting mixture was stirred for 30
min before filtering through a 0.45 μm glass filter. Removal of
the solvent from the filtrate afforded a dark solid, which was
treated with EtOH and filtered again. The filtrate was
concentrated under vacuum, and the resulting pink solid was
dissolved in 1 M HCl and purified by preparative HPLC
(reversed-phase C18-capped silica, MeCN−water as the mobile
1
phase) to afford 15 mg (49% yield) of dark solid. H NMR
(DMSO, 600 MHz) δ 7.50 (d, J = 8.8 Hz, 1H, CH), 7.43 (d, J
= 9.6 Hz, 1H, CHAr), 7.02 (s, 2H, NH2), 6.71 (dd, J = 8.8, 2.4
Hz, 1H, CH), 6.62 (dd, J = 9.7, 2.1 Hz, 1H, CHAr), 6.49 (d, J =
2.4 Hz, 1H, CH), 6.16 (d, J = 2.1 Hz, 1H, CHAr).
Initial Rate Study of the Oxidation of 5 by K3Fe(CN)6
To Generate 10. To a 25 mL volumetric flask was added 5
(9.2 mg, 28.7 μmol), which was diluted to the line with pH 10.4
buffer and sonicated in a 30 °C bath for 5 min. An aliquot (435
μL) was taken from stock solution and diluted with phosphate
buffer to 10 mL in a volumetric flask so that the final
concentration was 50 μM. The final solution was loaded into a
10 mL disposable syringe and placed at stopped-flow injection
slot A. The buffer solution was loaded into another disposable
10 mL syringe and placed at injection slot B. A background
scan was collected by injecting 0.25 mL of each syringe driven
by a pneumatic piston. The final solution of 5 in the
observation cell was 25 μM. This spectrum of 5 was used as
the absorbance baseline so that kinetic run would record the
change in absorbance relative to 5. To a 5 mL volumetric flask
was added K3Fe(CN)6 (21.9 mg, 66.5 μmol), which was diluted
to the line with pH 10.4 buffer and sonicated in a 30 °C bath
for 5 min. An aliquot (75.1 μL) was taken from the stock
solution and diluted with phosphate buffer to 10 mL in a
volumetric flask so that the final concentration was 100 μM.
The solution was loaded into a syringe and replaced the buffer
syringe located in injection slot B. The Julabo water circulator
was set to 25 °C, and the solutions were kept in the loading
syringes for 5 min to allow for temperature equilibration. The
kinetic runs occurred by injecting 0.25 mL of each syringe
through the observation cell bringing the final concentrations of
the reagents as follows: [5] = 25 μM and [K3Fe(CN)6] = 50
μM. The rate was determined by observing the change in
absorbance at 532 nm (used due to greatest observable change
in absorbance from 5 to 10 without interference of the
ferricyanide absorbance). Product concentration was calculated
C
DOI: 10.1021/acs.jpca.6b08571
J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A
Article
Scheme 2. Synthesis of the Leuco Dye 3
by using ε = 1249 dm3 mol−1 cm−1 at 532 nm and Beer’s Law.
The initial rate was calculated by taking the slope of the line of
[10] versus time for the first 10% conversion. Characterization
data of 10: HRMS (positive ion, m/z) [M + H]+ calcd for
C18H15N4O2 319.119501; found 319.12000. λmax (EtOH/
H2O), nm (ε/dm3 mol−1 cm−1): 450 (6295), 532 (sh, 5167).
General Procedure for Stopped-Flow Reaction of 10
with PPD. Similar stock solutions of 5 and K3Fe(CN)6 were
prepared as above and mixed in a 10 mL volumetric flask so
that the final concentrations of the reagents were 50 μM for 5
and 100 μM for K3Fe(CN)6. After 2 min, the reaction to
generate 10 was complete and the resulting solution was
transferred to a 10 mL disposable syringe and placed at
stopped-flow injector slot A. A background scan was taken for
the oxidized trinuclear dye 10 using the buffer as a baseline. A
50 μM solution of PPD in pH 10.4 buffer was added to a 10 mL
syringe and replaced the buffer solution at injector slot B. The
rate at 25 °C was determined by observing the change in
absorbance at 532 nm. For the starting material 10 and the
product 5, concentrations were calculated by using ε5 = 3871
dm3 mol−1 cm−1 and ε10 = 5167 dm3 mol−1 cm−1 at 532 nm
and Beer’s Law. The initial rate was calculated by taking the
slope of the line of [10] versus time for the first 10%
conversion.
convenience, the material generated in situ was used for the
mechanistic studies.
Attempted Isolation of Indo Dye 4. It was found that the
best way to generate 4 for reactivity studies was under stopped-
flow conditions. All other attempts resulted in additional
products. Oxidation of 3 in alkaline solution by atmospheric O2
gave trinuclear species 5 along with a small amount of
oligomers (Scheme 3). Presumably, base-promoted hydrolysis
Scheme 3. Attempted Synthesis of Indo Dye 4 from 9
RESULTS AND DISCUSSION
■
To understand potential oligomerization pathways and
mechanistic details, syntheses of pure 3, 4, and 5 needed to
be developed. The following approaches provide either isolated
or in situ generated material of sufficient purity for the
mechanistic investigations.
of 4 generated PPD/QDI, which subsequently reacted with the
remaining dinuclear species to form 5. Using a one-electron
oxidant also was ineffective. Concentrating the reaction of
potassium ferricyanide with 3 yielded a cyclization product
resorufamine (Scheme 3).2,6,18−22
Preparation of Trinuclear Species 5. Trinuclear species 5
is the final compound needed to study the kinetics and
mechanisms of oligomerization. This compound was previously
isolated from a small-scale synthesis that used repeated
preparative TLC to purify the final product.13,23 Purification
of the compound was complicated by the appearance of several
byproducts potentially arising from oligomerization of the
desired product.13
Several other oxidation methods were tried but most had side
reactions that led to low purity of the desired dye. For example,
although K3Fe(CN)6 gave the desired product, as verified by
mass spectrometry, removal of the Fe2+ salts proved difficult
(Table 1, entry 1). Alternatively, using an aqueous H2O2
solution resulted in slow conversion of PPD and resorcinol
to the desired compound with very low purity (Table 1, entry
2). Although adding catalytic amounts of redox metals drives
the reaction,24 it was not possible to isolate the product in
sufficient purity (Table 1, entries 3−5).
Preparation of Leuco Dye 3. Leuco dye 3 has been
proposed as the first coupling product en route to trinuclear
species 5 and oligomers (Scheme 1).12 Attempts to isolate pure
3 directly from oxidative coupling of PPD and resorcinol were
unsuccessful. Thus, an independent synthesis of 3 via coupling
of two aromatic moieties followed by reduction/deprotection
was pursued (Scheme 2).
Preparation of the cross-coupling precursor 8 was accom-
plished in two steps: reaction of 2,4-difluoronitrobenzene with
benzyl alcohol followed by zinc-promoted reduction of the
nitro group.16,17 Subsequent palladium-catalyzed C−N cou-
pling of 8 with 1-bromo-4-nitrobenzene afforded 9 in good
yield. Compound 9 was converted readily to leuco dye 3 via
palladium-catalyzed reduction of the nitro group and hydro-
genolysis of the benzyl ethers in one pot. However, the
obtained material or its protonated form (following acidic
workup) is oxidatively unstable, thus preventing the isolation of
a pure sample. Nevertheless, NMR and mass spectral data of
the crude product strongly support the proposed structure. For
D
DOI: 10.1021/acs.jpca.6b08571
J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A
Article
monomer addition to 5. However, the addition of the dinuclear
species 4 to 5 presents an alternative oligomerization
mechanism. Compound 4 was generated in situ under dilute
conditions in a stopped-flow apparatus because of its high
reactivity and inability to be isolated. Mixing PPD and
resorcinol (100 μM each) with 400 μM of K3Fe(CN)6 in pH
10.4 water−ethanol (1:1) generated a new species with an
absorbance maximum at 492 nm, which closely matches the
absorbance of similar dinuclear dyes.3 Mass spectrometry
suggested that this species has a mass number of 215.1, further
confirming the formation of 4 ([M + H]+ calcd for C12H11N2O2
m/z 215.08). Adding 100 μM of 5 to the aforementioned
mixture gave an absorbance spectrum that is simply a
combination of the spectra of 4 and 5 (Figure 2), suggesting
that no reaction occurred between these two compounds.
Oxidized Trinuclear Dye 10 as a Potential Intermedi-
ate for Oligomerization. The inertness of 5 toward other
species present in the reaction mixture indicates that it is not
directly involved in oligomer formation. However, since 5 is a
major product in the oxidative coupling of PPD and resorcinol,
it is reasonable to assume that it is either a precursor to the key
reactive intermediate or it is a saddle point, as described by
More O’Ferrall and Jencks, off the reaction coordinate.28,29
Alternatively, oligomerization could merely be a side reaction
and the oligomers could be impurities in the major dye (5)
forming reaction. To differentiate these possibilities, oxidation
of 5 to 10 was explored using potassium ferricyanide. The
reaction went to completion within 30 s, and the absorbance
spectrum contained two isosbestic points (Figure 3). High-
resolution mass spectrometry revealed a new product with a
mass corresponding to an oxidized species. However, the
oxidized product (10) could not be isolated, and all attempts
led to the recovery of 5. Hence the reaction of 10 with 5 would
not be expected to be important in oligomer formation.
Interpretation of initial rates showed oxidation of 5 to be of
first-order dependence on both 5 and potassium ferricyanide,
with an overall second-order rate constant of (1.08 0.10) ×
104 M−1 s−1 at pH 10.40 0.05 (see Supporting Information
for details).
Table 1. Synthetic Attempts to Access the Trinuclear Dye 5
a
b
entry
oxidant (equiv)
additive (equiv)
none
time
purity (yield)
c
1
K3Fe(CN)6 (4)
H2O2 (40)
H2O2 (40)
H2O2 (40)
H2O2 (40)
oxone (0.5)
oxone (1.0)
oxone (1.5)
oxone (2.0)
O2
2.5 h
8 h
<80%
<10%
∼10%
30%
2
none
3
K3Fe(CN)6 (0.05)
FeCl3 (0.05)
CuCl2 (0.05)
none
3 h
4
1 h
5
1 h
25%
6
6 h
40%
7
none
3 h
30%
8
none
2 h
20%
9
none
2 h
20%
10
none
2.5 d
>98% (45%)
a
Based on disappearance of the starting materials from TLC analysis
b
c
1
(eluent: ethyl acetate). Based on H NMR. Product could not be
isolated from the Fe(II) salts.
Likewise, the water-soluble oxidant oxone gave a material
with low purity (Table 1, entries 6−9). As with all other
attempts that gave an impure material, the desired product and
the byproducts have limited solubility in water and organic
solvents even polar ones such as DMSO, DMF, and NMP,
making purification difficult. However, by bubbling O2 through
a buffered (pH 10) solution of PPD and resorcinol, the desired
product was obtained in good yield and high purity, albeit
slowly (Table 1, entry 10).
Compound 5 was characterized by a variety of analytical
methods. High-resolution mass spectrometry (HRMS) gave the
expected mass for 5. Both 1H and 13C{1H} NMR spectra of the
product (in DMSO-d6) support a C2-symmetric molecule as
illustrated in Figure 1. This type of symmetry has been
observed in related systems and the core structure has been
proposed to be zwitterionic based on density functional theory
(DFT) calculations.25−27
Addition of PPD or resorcinol to the oxidized trinuclear
species 10 was examined next. Oligomerization potentially
could occur through the nucleophilic attack of PPD on 10;
similar reactivity has been observed in the formation of the
Bandrowski’s base.2 However, when PPD was added to an in
situ generated solution of 10, no coupled product was formed;
instead, the resulting absorbance spectrum matched that of the
1
reduced trinuclear dye 5, and H NMR analysis showed the
presence of QDI 2. Additionally, mass spectrometry confirmed
that no coupled products were present. These results support a
redox process occurring between 10 and PPD (Scheme 4).
Evidently, the order of the reaction depends on the initial
concentration of 10. It showed zero-order dependence on PPD
with 50 μM of [10]0, while first-order dependence on PPD was
observed with 10 μM of [10]0. In the latter case, the reaction
was determined to be second order overall (first order with
respect to both 10 and PPD) and the calculated rate constant
was 31.7 2.8 M−1 s−1.
For the 50 μM reaction, a radical chain mechanism could be
used to rationalize the zero-order dependence on PPD,
assuming a steady-state concentration of a radical intermediate
derived from 10 is maintained in solution. However, the
reaction of 10 and PPD in the presence of 5,5-dimethyl-1-
pyrroline-N-oxide (DMPO) failed to yield a spin-trapped
Figure 1. Zwitterionic form of 5.
Reactivity of Dye Products. With pure 3, 4, and 5 in hand
or generated in situ, their roles in oligomerization through the
reactions with the monomeric reactants (PPD and resorcinol)
were probed. Pathways to larger oligomers begin to bifurcate as
the oligomer grows, and multiple possibilities can be proposed.
Our interest was to examine the initial steps and determine the
key intermediates for oligomer growth.
Trinuclear Dye 5 as a Potential Intermediate for
Oligomerization. In the absence of an exogenous oxidant, no
reaction was observed between either PPD or resorcinol and 5.
It is thus clear that oligomerization does not occur through
E
DOI: 10.1021/acs.jpca.6b08571
J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A
Article
Figure 2. Comparison of calculated additive spectrum of 4 + 5 and observed spectrum of 4 + 5.
dissociation step were rate-limiting, PPD would not be
expressed in the rate law. This can be explained mathematically
using eqs 1−9.
K1
10 + 10 ⇌ 102 where 102 = K1[10]2
(1)
K2
102 + 10 ⇌ 103 where 103 = K1K2[10]3
(2)
If the aggregation follows isodesmic growth,13,30 then K1 =
K2 = K3 = ... = Ki−1 and for the reaction with PPD:
k1
10 ⇀ 10 + 10 where Ki−1 = k−1/k1
i
i−1
↽⎯⎯⎯
(3)
(4)
k
−1
k2
10 + PPD → 5 + QDI
d[5]
dt
= k2[10][PPD]
(5)
assuming steady state for [10]
d[10]
= k1[10] − k−1[10i−1][10] − k2[PPD][10] ≈ 0
i
dt
(6)
Figure 3. Oxidation of 5 by K3Fe(CN)6.
k1[10]
k−1[10i−1] + k2[PPD]
i
[10] =
Scheme 4. Redox Reaction of 10 with PPD
(7)
then
k1k2[10][PPD]
k−1[10i−1] + k2[PPD]
d[5]
dt
i
=
(8)
(9)
assuming k2 ≫ k−1, the observed rate becomes
d[5]
= k1[10]
i
dt
product, suggesting that the radical mechanism is unlikely to be
operative.
Following the isodesmic growth model, PPD would not appear
in the rate law (eq 9) unless [10]0 were low enough to prevent
significant aggregation from occurring. Future work will be
carried out to further elucidate the aggregate growth
mechanism.31
In the final attempt to form oligomers, 10 was reacted with
resorcinol. Under stopped-flow conditions with diluted
solutions, no reaction occurred using a 1:1 stoichiometric
ratio. Even with a large excess of resorcinol (100 equiv), only
minor amounts (<5%) of unidentifiable products were
Previous reports have documented hydrogen-bonding and π-
stacking behavior in molecules with similar core structures.26,27
Using dynamic light scattering (DLS) method, particles sizes of
334 and 529 nm were observed for 50 μM solutions of 5 and
10, respectively. Formation of aggregates under these
concentrations may limit the amount of free 10 available to
react in solution. The redox reaction with PPD can only occur
after sufficient dissociation of the aggregates occurs. If the
F
DOI: 10.1021/acs.jpca.6b08571
J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A
Article
a
Scheme 5. Proposed Mechanisms for the Oligomerization from 3/4
a
Hartree-Fock (6-31G* in EtOH) calculations (refs 32 and 33) show that 4 is 9.8 kcal/mol more stable than its tautomer 4T.
detected. Scaling up the reaction and performing it in the
presence of hydrogen peroxide resulted in only minor amounts
of products (<5%) with molecular weight of <350 Da. Isolation
of these minor products was unsuccessful, but the absence of
oligomeric species, as confirmed by matrix-assisted laser
desorption ionization (MALDI) or ESI, eliminated resorcinol
addition to 10 as a viable oligomerization route.
CONCLUSIONS
■
The success in developing a large-scale synthesis of trinuclear
dye 5 has allowed us to investigate its reactivity (or lack
Scheme 6. Current Understanding of PPD−Resorcinol Dye
Reactivity
Dinuclear Dyes 3 and 4 as Potential Intermediates For
Oligomerization. At this point, neither 5 nor 10 has been
demonstrated as the key intermediate in oxidative oligomeriza-
tion of PPD−resorcinol, as suggested by their limited or no
reactivity under a variety of conditions that are representative of
hair color products. Our attention was then focused on
reactions of the two dinuclear species, 3 and 4.
Bubbling O2 through an ethanol solution of 3 resulted in a
highly insoluble dark powder. Mass spectrum (MS-ESI) of the
product indicated the presence of an octanuclear species (m/z
855.27 consistent with a formula of C48H39N8O8). The
observation of larger molecules arising from oxidation of the
dinuclear species has not been shown previously. Our result
here provides strong evidence that 3 and 4 are key reactive
intermediates for the oligomerization process. Two oligomeri-
zation pathways can be proposed: (a) aza-Michael addition of 3
to 4, or (b) attack on tautomer 4T by the electron-rich
resorcinol moiety of 3 (Scheme 5). Although other
oligomerization pathways are possible (e.g., reaction with 10
with 3), mechanisms depicted in Scheme 5 represent two most
likely paths to the proposed oligomers. We propose that the
rate of coupling is slow; therefore, slow oxidative conditions are
needed so sufficient quantities of 3 are available to react with 4
for oligomerization. Reactions that have a fast rate of oxidation
of 3 (potassium ferricyanide or alkaline O2) will not generate a
significant amount of oligomers due to the relatively low
concentrations of the redox pairs. Additional work is required
to deepen our understanding of oxidative oligomerization of
PPD and resorcinol. Among our future directions are (a)
investigation of the pH profile for the rates of the reactions
listed in Scheme 5, (b) understanding of reactivity of 10 with 3,
and (c) determination of rate constants for the individual steps
once the main operative pathway is identified.
thereof) toward oligomerization. Overall the reactivity of 5 is
limited. It does not react with PPD, resorcinol, or the dinuclear
dye 4. Upon oxidation by potassium ferricyanide, 5 is converted
to 10 following a second-order reaction (first order in both
potassium ferricyanide and 5). The oxidized trinuclear dye 10 is
also unreactive toward resorcinol or 5, ruling out these
reactions as possibilities for oligomerization. Compound 10,
however, reacts with PPD to give 5 and QDI via a redox
G
DOI: 10.1021/acs.jpca.6b08571
J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A
Article
(6) Corbett, J. F. Benzoquinone Imines. Part X. The Mechanism and
Kinetics of the Reactions of p-Benzoquinone Di-imine and p-
Benzoquinone Monoimine with C-Methoxy-m-Diamines and p-
Methoxy- and p-Chloro-phenols. J. Chem. Soc., Perkin Trans. 2 1972,
No. 8, 999−1005.
(7) Corbett, J. F.; Gamson, E. P. Benzoquinone Imines. Part XI.
Mechanism and Kinetics of the Reaction of p-Benzoquinone Di-imines
with Aniline and its Derivatives. J. Chem. Soc., Perkin Trans. 2 1972,
No. 11, 1531−1537.
(8) Corbett, J. F. Benzoquinone Imines. Part VIII. Mechanism and
Kinetics of the Reaction of p-Benzoquinone Monoimines with
Monohydric Phenols. J. Chem. Soc. B 1970, 1502−1509.
(9) Cox, H. E. Hair Dyes II. The Functions and Reactions of Phenols.
Analyst 1940, 65 (772), 393−398.
process. Both kinetics and DLS experiments suggest that 5 and
10 readily undergo significant aggregation or agglomeration.
In situ generation of intermediate 4 using two independent
routes (via PPD−resorcinol oxidative coupling, or via synthesis
of the precursor leuco dye 3) has allowed us to observe
previously unknown reaction pathways from this unstable
species. Scheme 6 outlines our current understanding of the
PPD−resorcinol dye reactivity, which has been expanded
significantly from the original mechanistic profile. Our results
suggest that oligomerization through this intermediate is a
viable route, and further work will be carried out to understand
the mechanistic details and to determine whether other routes
are possible.
(10) Brown, K. C.; Corbett, J. F.; Labinson, R. Benzoquinone Imines.
Part 14. The Kinetics and Mechanism of the Coupling of p-
Benzoquinone Monoimines with m-Aminophenols. J. Chem. Soc.,
Perkin Trans. 2 1978, No. 12, 1292−1296.
(11) Feng, Y.; Chan, A. Kinetics and Mechanism of Indo-Aniline Dye
Formation in Aqueous Peroxide Solution. J. Soc. Cosmet. Chem. 1994,
45, 299−308.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jpca.6b08571.
■
S
Details on coupling optimization to generate 9, hydro-
genolysis optimization to generate 3, kinetics of
oxidation of 5, kinetics of reduction of 10, dynamic
light scattering table, and NMR data (PDF)
(12) Corbett, J. F. The Role of Meta Difunctional Benzene
Derivatives. J. Soc. Cosmet. Chem. 1973, 24, 103−134.
(13) Shah, M. J.; Tolgyesi, W. S.; Britt, A. D. Cooxidation of p-
Phenylenediamine and Resorcinol in Hair Dyes. J. Soc. Cosmet. Chem.
1972, 23, 853−861.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: Bryan_Murphy@cotyinc.com. Phone: 513-936-4362.
*E-mail: hairong.guan@uc.edu. Phone: 513-556-6377.
́
(14) Bonnet, A.; Barre, G.; Gilard, P. Trinoyaux. Determination des
■
1
structures par RMN H, 13C et 15N. J. Chim. Phys. Phys.-Chim. Biol.
1995, 92 (10), 1823−1828.
(15) Brody, F.; Burns, M. S. Studies Concerning the Reactions of
Oxidation Dye Intermediates. J. Soc. Cosmet. Chem. 1968, 19, 361−
379.
Present Addresses
§Coty, Inc., 2180 E. Galbraith Road, Building D, Cincinnati,
(16) Sythana, S. K.; Naramreddy, S. R.; Kavitake, S.; Kumar, CH V.;
Bhagat, P. R. Nonpolar Solvent a Key for Highly Regioselective SNAr
Reaction in the Case of 2,4-Difluoronitrobenzene. Org. Process Res.
Dev. 2014, 18 (7), 912−918.
OH 45237.
⊥Coty, Inc., 10123 Alliance Road, Suite 310, Blue Ash, Ohio
45241.
(17) Boiteau, J.-G.; Bouquet, K.; Talano, S.; Barbuis, C. M. 4-
(Azacycloalkyl)benzene-1,3-diol Compounds as Tyrosinase Inhibitors,
Process for the Preparation Thereof and Use Thereof in Human Medicine
and in Cosmetics. U.S. Patent 8,299,259 B2, Oct 30, 2012.
(18) Chen, B.; Lv, C.; Tang, X. Chemoselective Reduction-Based
Fluorescence Probe for Detection of Hydrogen Sulfide in Living Cells.
Anal. Bioanal. Chem. 2012, 404 (6−7), 1919−1923.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
(19) Kende, A. S.; Koch, K. Intramolecular Radical Cyclization of
Phenolic Nitronates: Facile Synthesis of Annelated Tropone and
Tropolone Derivatives. Tetrahedron Lett. 1986, 27 (50), 6051−6054.
(20) Leboff, A.; Carbonnelle, A. C.; Alazard, J. P.; Thal, C.; Kende, A.
S. Intramolecular Radical Coupling of a Phenolic Enolate: Oxidative
Fragmentation of the Spirodiketone Intermediate. Tetrahedron Lett.
1987, 28 (36), 4163−4164.
(21) Kende, A. S.; Koch, K.; Smith, C. A. Intramolecular Radical
Cyclization of Phenolic Enolates. J. Am. Chem. Soc. 1988, 110 (7),
2210−2218.
(22) McDonald, P. D.; Hamilton, G. A. On the Mechanism of
Phenolic Oxidative Coupling Reactions. Ferricyanide Oxidation of
2,3′,4-Trihydroxybenzophenone, an Example of a Radical Aromatic
Substitution Mechanism. J. Am. Chem. Soc. 1973, 95 (23), 7752−7758.
(23) Sabelle, S.; Leduc, M. Tri-Aromatic Azomethine Direct Dyes
Comprising at Least One Unit Derived from Resorcinol, Dyeing
Composition, Method and Use. U.S. Patent 2015/0027483 A1, Jan 29,
2015.
ACKNOWLEDGMENTS
■
The authors acknowledge Dr. Bob Strife for acquiring the
HRMS data, Dr. Mike Weaver for assistance in obtaining the
DLS data, and Dr. Casey Kelly for assistance with the
computational methods, Dr. Keith Brown for review of the
manuscript and helpful comments, and Dr. Guiru Zhang and
Dr. John Gardlik for mechanistic discussion.
REFERENCES
■
(1) Morel, O. J. X.; Christie, R. M. Current Trends in the Chemistry
of Permanent Hair Dyeing. Chem. Rev. 2011, 111 (4), 2537−2561.
(2) Corbett, J. F. Benzoquinone Imines. Part IV. Mechanism and
Kinetics of the Formation of Bandrowski’s Base. J. Chem. Soc. B 1969,
818−822.
(3) Corbett, J. F. Benzoquinone Imines. Part V. Mechanism and
Kinetics of the Reaction of p-Benzoquinone Monoimines with m-
Phenylenediamines. J. Chem. Soc. B 1969, 823−826.
(4) Corbett, J. F. Benzoquinone Imines. Part VI. Mechanism and
Kinetics of the Reaction of p-Benzoquinone Di-imines with m-
Phenylenediamines. J. Chem. Soc. B 1969, 827−835.
(5) Corbett, J. F. Benzoquinone Imines. Part IX. Mechanism and
Kinetics of the Reaction of p-Benzoquinone Di-imines with m-
Aminophenols. J. Chem. Soc., Perkin Trans. 2 1972, No. 5, 539−548.
(24) Sawyer, D. T.; Sobkowiak, A.; Matsushita, T. Metal [MLx; M =
Fe, Cu, Co, Mn]/Hydroperoxide-Induced Activation of Dioxygen for
the Oxygenation of Hydrocarbons: Oxygenated Fenton Chemistry.
Acc. Chem. Res. 1996, 29 (9), 409−416.
́
(25) Holtzl, T.; Veszpremi, T.; Nguyen, M. T. Chemical Bonding in
̈
Zwitterionic Diamino-meta-quinonoids and their Isomers. J. Phys. Org.
Chem. 2005, 18 (11), 1123−1131.
H
DOI: 10.1021/acs.jpca.6b08571
J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A
Article
(26) Siri, O.; Braunstein, P. Unprecedented Zwitterion in Quinonoid
Chemistry. Chem. Commun. 2002, 3, 208−209.
(27) Tamboura, F. B.; Cazin, C. S. J.; Pattacini, R.; Braunstein, P.
Reactions of Amines with Zwitterionic Quinoneimines: Synthesis of
New Anionic and Zwitterionic Quinonoids. Eur. J. Org. Chem. 2009,
2009 (20), 3340−3350.
(28) Jencks, W. P. General Acid-Base Catalysis of Complex Reactions
in Water. Chem. Rev. 1972, 72 (6), 705−718.
(29) More O’Ferrall, R. A. Relationships between E2 and E1cB
mechanisms of β-elimination. J. Chem. Soc. B 1970, 274−277.
(30) De Greef, T. F. A.; Smulders, M. M. J.; Wolffs, M.; Schenning,
A. P. H. J.; Sijbesma, R. P.; Meijer, E. W. Supramolecular
Polymerization. Chem. Rev. 2009, 109 (11), 5687−5754.
(31) We cannot rule out the possibility of cooperative aggregate
growth prior to a nucleation point.
(32) Shao, Y.; Molnar, L. F.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.;
Brown, S. T.; Gilbert, A. T. B.; Slipchenko, L. V.; Levchenko, S. V.;
O’Neill, D. P.; DiStasio, R. A., Jr.; Lochan, R. C.; Wang, T.; Beran, G. J.
O.; Besley, N. A.; Herbert, J. M.; Yeh Lin, C.; Van Voorhis, T.; Hung
Chien, S.; Sodt, A.; Steele, R. P.; Rassolov, V. A.; Maslen, P. E.;
Korambath, P. P.; Adamson, R. D.; Austin, B.; Baker, J.; Byrd, E. F. C.;
Dachsel, H.; Doerksen, R. J.; Dreuw, A.; Dunietz, B. D.; Dutoi, A. D.;
Furlani, T. R.; Gwaltney, S. R.; Heyden, A.; Hirata, S.; Hsu, C.-P.;
Kedziora, G.; Khalliulin, R. Z.; Klunzinger, P.; Lee, A. M.; Lee, M. S.;
Liang, W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E. I.; Pieniazek, P.
A.; Min Rhee, Y.; Ritchie, J.; Rosta, E.; David Sherrill, C.; Simmonett,
A. C.; Subotnik, J. E.; Lee Woodcock Iii, H.; Zhang, W.; Bell, A. T.;
Chakraborty, A. K.; Chipman, D. M.; Keil, F. J.; Warshel, A.; Hehre,
W. J.; Schaefer Iii, H. F.; Kong, J.; Krylov, A. I.; Gill, P. M. W.; Head-
Gordon, M. Advances in Methods and Algorithms in a Modern
Quantum Chemistry Program Package. Phys. Chem. Chem. Phys. 2006,
8 (27), 3172−3191.
(33) Spartan’10; Wavefunction, Inc.: Irvine, CA, 2010.
I
DOI: 10.1021/acs.jpca.6b08571
J. Phys. Chem. A XXXX, XXX, XXX−XXX