Anthraquinone catalysis in the glucose-driven reduction of indigo to leuco-indigo

Article abstract of DOI:10.1039/b814149e

Anthraquinone immobilised onto the surface of indigo microcrystals enhances the reductive dissolution of indigo to leuco-indigo. Indigo reduction is driven by glucose in aqueous NaOH and a vibrating gold disc electrode is employed to monitor the increasing leuco-indigo concentration with time. Anthraquinone introduces a strong catalytic effect which is explained by invoking a molecular "wedge effect" during co-intercalation of Na⁺ and anthraquinone into the layered indigo crystal structure. The glucose-driven indigo reduction, which is ineffective in 0.1 M NaOH at 65°C, becomes facile and goes to completion in the presence of anthraquinone catalyst. Electron microscopy of indigo crystals before and after reductive dissolution confirms a delamination mechanism initiated at the edges of the plate-like indigo crystals. Catalysis occurs when the anthraquinone-indigo mixture reaches a molar ratio of 1:400 (at 65°C; corresponding to 3 μM anthraquinone) with excess of anthraquinone having virtually no effect. A strong temperature effect (with a composite E_A≈ 120 kJ mol^-1) is observed for the reductive dissolution in the presence of anthraquinone. The molar ratio and temperature effects are both consistent with the heterogeneous nature of the anthraquinone catalysis in the aqueous reaction mixture.

Full text of DOI:10.1039/b814149e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Anthraquinone catalysis in the glucose-driven reduction of indigo
to leuco-indigo
a
b
a
c
Anne Vuorema, Philip John, Marjo Keskitalo, Mary F. Mahon,
d
c
M. Anbu Kulandainathan and Frank Marken*
Received 14th August 2008, Accepted 1st December 2008
First published as an Advance Article on the web 14th January 2009
DOI: 10.1039/b814149e
Anthraquinone immobilised onto the surface of indigo microcrystals enhances the reductive
dissolution of indigo to leuco-indigo. Indigo reduction is driven by glucose in aqueous NaOH and a
vibrating gold disc electrode is employed to monitor the increasing leuco-indigo concentration with
time. Anthraquinone introduces a strong catalytic effect which is explained by invoking a molecular
+
‘
‘wedge effect’’ during co-intercalation of Na and anthraquinone into the layered indigo crystal
structure. The glucose-driven indigo reduction, which is ineffective in 0.1 M NaOH at 65 1C,
becomes facile and goes to completion in the presence of anthraquinone catalyst. Electron
microscopy of indigo crystals before and after reductive dissolution confirms a delamination
mechanism initiated at the edges of the plate-like indigo crystals. Catalysis occurs when the
anthraquinone–indigo mixture reaches a molar ratio of 1 : 400 (at 65 1C; corresponding to 3 mM
anthraquinone) with excess of anthraquinone having virtually no effect. A strong temperature effect
À1
(
with a composite E
A
E 120 kJ mol ) is observed for the reductive dissolution in the presence of
anthraquinone. The molar ratio and temperature effects are both consistent with the heterogeneous
nature of the anthraquinone catalysis in the aqueous reaction mixture.
1
5
16
1
. Introduction
beds, or at nickel contacts represent alternative mediator-
free approaches.
1
For centuries indigo has been a popular dye, the current
Glucose has been investigated for sulfur dye reduction
17
processes and it is considered to be a possible benign
18
alternative reducing agent for indigo. Glucose is degraded
annual world consumption of indigo and other vat dyes being
2
about 33 million kg. Indigo as a vat dye requires that the
water-insoluble oxidised form is reduced to its water-soluble
in alkaline solutions in a complex sequence of reaction
19–21
leuco-form prior to the dyeing process. Traditionally, a
3
fermenting woad bath was used to reduce indigo. Anthra-
steps
and the reducing effect of glucose is associated with
22
a not-fully-characterised electron-rich intermediate.
We
quinone-containing madder extracts were invariably added to
this bath, and it has been shown with pure cultures of the
indigo-reducing bacterium Clostridium isatidis, that soluble
have recently demonstrated that the glucose-driven reductive
dissolution of indigo can also be catalysed in the presence
of anthraquinone derivatives such as the water-soluble
18
4
,5
anthraquinones stimulate bacterial indigo reduction. With
the replacement of natural indigo by the synthetic product at
the end of the 19th century, the bacterial fermentation bath
was replaced by sodium dithionite, which for a long time was
the sole industrial reducing agent. But recently, a plethora of
less environmentally harmful reducing agents and processes
1
,8-dihydroxyanthraquinone. The use of this catalyst allows
for lower levels of added alkali and for lower operating
temperatures during reductive dissolution. Anthraquinones
can be electrochemically reduced in aqueous alkaline solution
at a potential similar to that for the reduction of indigo,
forming the two-electron two-proton anthraquinol product,
which then transfers electrons to the suspended indigo.
Usually, a soluble anthraquinone derivative is used as catalyst.
However, the parent anthraquinone compound is not
water-soluble and the alkali metal salts of the parent anthra-
6
–10
have been developed
among which direct or indirect
(
mediated) electrochemical methods are perhaps the most
significant. Mediator-based electrochemical reactions with
1
1
12
the Fe(III) triethanolamine (TEA) complex, Fe(III) Hedta,
3+
2+
3+
with Fe -D-gluconate or Ca –Fe -D-gluconate complexes,
13
23
quinol exhibit only limited water-solubility. Two major
1
4
and with anthraquinone derivatives have been studied.
Direct reduction processes of indigo in fixed and fluidised
benefits of the parent anthraquinone compared to other more
soluble anthraquinone derivatives are slightly lower toxicity
and lower cost. It is shown here that the parent anthraquinone
is a highly effective catalyst for indigo reduction and that the
low solubility of anthraquinone during the indigo reduction
process actually is a beneficial feature.
a
MTT Agrifood Research Finland, Plant Production Research,
Crop Science and Technology, FIN-31600 Jokioinen, Finland
School of Biological Sciences, The University of Reading,
PO Box 221, Reading, UK RG6 6AS
Department of Chemistry, University of Bath, Bath, UK BA2 7AY.
E-mail: f.marken@bath.ac.uk
Electro-organic Division, Central Electrochemical Research Institute,
Karaikudi, 630 006, India
b
In this study, the catalytic effect of the water-insoluble
anthraquinone on the reductive dissolution of indigo is studied
quantitatively. Electrochemical experiments are reported that
monitor the formation of leuco-indigo. The anthraquinone
c
d
1
816 | Phys. Chem. Chem. Phys., 2009, 11, 1816–1824
This journal is ꢀc the Owner Societies 2009

View Article Online
Fig. 2 Schematic drawing of the 250 Hz vibrating electrode system
used to measure leuco-indigo content during reduction in aqueous
media with glucose as the reducing reagent.
Fig. 1 Schematic representation for (A) solution phase catalysis and
B) surface catalysis mechanisms for the glucose-driven reduction of
at the basal plane pyrolytic graphite electrode by rubbing on a
25,26
filter paper.
(
The electrode was then immersed into solution
indigo to leuco-indigo.
and cyclic voltammograms recorded for a range of different
scan rates. Each measurement consisted of four consecutive
potential cycles to explore the time-dependent reductive
dissolution of leuco- product away from the electrode surface.
catalyst is immobilised onto the indigo micro-crystal surface
prior to the glucose-driven indigo reduction process. The
schematic drawings in Fig. 1 compare the case of (A) a
solution soluble catalyst affecting the indigo reduction and
(
B) a surface immobilised catalyst. The latter case allows
2
.4 Procedure for the voltammetric determination
catalyst to be effective only at the point where the reaction
occurs.
of leuco-indigo
Kinetic data for the indigo reduction are obtained with a
vibrating electrode system which allows the leuco-indigo
content to be monitored. The effect of the reductive indigo
dissolution on the crystal morphology is investigated by
monitoring indigo crystals (grown by sublimation) with
scanning electron microscopy.
Anthraquinone catalyst was immobilised onto indigo particles
by adding an acetone solution (concentration 0.3 mM) to solid
indigo powder. The suspension was homogenised with 24 kHz
ultrasound (Hielscher UP200G) and the acetone solvent was
evaporated to leave anthraquinone coated onto the indigo
powder.
3
An aqueous NaOH solution (100 cm ) was thermostated to
5
1
5, 65 or 75 1C and 30 mg indigo added (corresponding to
.1 mM solution) after pre-dispersion in a small volume of
2
. Experimental
solution by treatment with 24 kHz ultrasound. Next, 400 mg
3
glucose (corresponding to 22 mM solution in 100 cm of
2
.1 Reagents
Chemical reagents including indigo (Fluka), NaOH (Aldrich),
anthraquinone (Sigma), acetone (Sigma), and D-(+)-glucose
NaOH) was added under an atmosphere of argon and
voltammograms (limiting currents for the oxidation of
24
(
Sigma), were obtained commercially and used without further
leuco-indigo) at the vibrating gold disc electrode were
purification. Demineralised water was taken from an Elga
recorded at regular time intervals. Polishing the gold electrode
surface before each measurement was required.
purification system with at least 18 MO cm resistivity. Argon
(
BOC) gas was employed for solution de-aeration.
2
.2 Instrumentation
2
.5 Procedure for indigo sublimation and SEM
For voltammetric studies a micro-Autolab II potentiostat
system (EcoChemie, Netherlands) was employed with a Pt wire
counter electrode and a saturated Calomel (SCE) reference
electrode (Radiometer, Copenhagen). Solid-state electro-
chemical experiments were carried out with a 4.9 mm diameter
basal plane pyrolytic graphite electrode (Le Carbon, UK). A
Indigo crystals were formed following sublimation in a short
path vacuum sublimation system (made from glass) with
internal water-cooled deposition finger. The system was heated
in silicone oil to a temperature of B240 1C. Sublimation under
À3
oil pump vacuum (ca. 10 Torr) was continued for 3 days
(
nights excluded). Macroscopic but very thin plate-like indigo
5
00 mm diameter vibrating gold disc electrode acted as a
crystals were obtained (see Fig. 3). Indigo crystals were
collected and attached to the double-sided conducting carbon
sticky pads for SEM, which were mounted on glass slides. The
glass slide with indigo crystal samples were immersed in the
reduction solution (33 mM glucose in 0.2 M NaOH at 65 1C).
Samples of indigo, exposed to reducing conditions for different
lengths of time, were rinsed in water, dried, and used directly
for electron microscopy (after gold sputter coating).
working electrode for the quantitative determination of
leuco-indigo. The electrode was embedded in a home-build
PEEK probe connected to a sonic probe (Braun, Oral-B
Sonic Complete, 250 Hz) to produce a sonic vibration with
2
4
approximately Æca. 200 mm amplitude (see Fig. 2).
Experiments were conducted in a thermostated electro-
chemical cell (with a Haake B3 circulator) under constant
de-aeration with high purity argon.
In order to apply anthraquinone catalyst, a 0.3 mM anthra-
3
quinone solution in acetone (ca. 0.1 cm ) was added to indigo
2
.3 Procedure for solid-state electrochemical measurements
crystal samples (ca. 10 mg) and allowed to evaporate. The
catalyst-impregnated indigo crystals were then mounted,
reacted, and imaged as before.
Aqueous 0.1 M NaOH was thermostated to 20, 55, 65 or 75 1C
and microcrystals of indigo or anthraquinone were immobilised
This journal is ꢀc the Owner Societies 2009
Phys. Chem. Chem. Phys., 2009, 11, 1816–1824 | 1817

View Article Online
Fig. 3 Typical scanning electron microscopy (SEM) image of the
plate-like indigo micro-crystals grown by sublimation and immobi-
lised onto double sided carbon tape.
3
. Results and discussion
3
.1 Solid-state voltammetry study of the temperature-
dependent reduction of indigo and anthraquinone
Indigo is a highly water-insoluble material due to strong
hydrogen-bonding interactions in the layered solid state
2
7
structure. The industrial reduction of indigo (and other
vat-dyes) is performed in alkaline aqueous environments in
order to dissolve the 2-electron reduction product leuco-indigo
2
8
29
(
pKa1 = 8.0, pKa2 = 12.7 ) for the vat-dyeing process. The
solid state electrochemistry of indigo micro-particles immobilised
at graphite electrodes has been previously reported in the
3
0
31
literature for aqueous media and for non-aqueous media.
The aqueous process has been shown to be highly pH-dependent.
At a pH o 13 a reversible 2-electron 2-proton process occurs
without significant dissolution of indigo. Only at significantly
À1
Fig. 4 Solid-state voltammetry data (scan rate 0.2 V s ) (A) for
the reduction of indigo microcrystals and (C) for the reduction of
anthraquinone microcrystals immobilised at a basal plane pyrolytic
graphite electrode and immersed in aqueous 0.1 M NaOH. Data
were obtained at temperatures (i) 20 1C, (ii) 55 1C, (iii) 65 1C, and
+
more negative potentials does a further Na -cation-dependent
intercalation process occur, which ultimately causes the break-up
3
0
of the indigo solid during reductive dissolution. At pH 13 these
(
iv) 75 1C. (B) Arrhenius plot of the indigo reduction peak current
+
+
À1
two processes, based on H and Na , merge into a single
reductive dissolution response, but even then the dissolution of
indigo remains rate-limited by the relatively slow dissolution
versus T
.
process may be limited by the rate of sodium cations diffusing
into the reaction zone, e.g. the edges of the indigo lattice
(vide infra). Cyclic voltammograms obtained over a range of
30
process.
Fig. 4A shows typical cyclic voltammograms obtained for
solid indigo particles immobilised at a basal plane pyrolytic
graphite electrode surface immersed in aqueous 0.1 M NaOH.
The reversible reduction response at À0.75 V vs. SCE is
consistent with the 2-electron reduction of indigo to
leuco-indigo. The shape of the voltammetric signal is complex
with a peak superimposed on a slower constant process. A
minor oxidation response at À0.1 V vs. SCE has been
À1
scan rates (0.05 to 2.0 V s , not shown) show an increase in the
peak current (but not in the underlying step) at higher scan rate.
In contrast to the temperature-dependent characteristics
observed for the reduction of solid immobilised indigo, micro-
particles of anthraquinone immobilised at a basal plane
pyrolytic graphite electrode show a reversible reduction
response at ca. À0.85 V vs. SCE without a significant
temperature effect. The process is consistent with the 2-electron
3
0
attributed to the 2-electron oxidation of indigo.
3
2
The reduction of solid indigo immobilised at a basal plane
pyrolytic graphite electrode is strongly temperature-dependent.
Fig. 4A shows voltammograms obtained at 20 1C, 55 1C, 65 1C,
and 75 1C. The cathodic current response increases with
temperature and this suggests that the rate of the dissolution
process is increased. An Arrhenius-type plot (see inset, Fig. 4B)
reduction of anthraquinone. The anthraquinol product
appears to be sufficiently water insoluble to remain at the
electrode surface. Cyclic voltammetry experiments conducted
À1
over a range of scan rates (0.05 to 2.0 V s , not shown) reveal
increased irreversibility at 65 1C and 75 1C (where loss due to
diffusion away from the electrode becomes noticeable) at scan
À1
À1
reveals an approximate activation energy of 14 kJ mol . The
rates of less than 0.1 V s . The relatively insignificant
1
818 | Phys. Chem. Chem. Phys., 2009, 11, 1816–1824
This journal is ꢀc the Owner Societies 2009

View Article Online
temperature effect on the voltammetric response suggests a
facile solid-state conversion which is approaching completion.
The reversible reduction potential for solid anthraquinone
appears to be ca. 100 mV negative of that for solid indigo
occurs at À0.65 V vs. SCE consistent with the oxidation of
leuco-indigo to indigo. A plateau current is observed and the
limiting current is determined at À0.6 V vs. SCE (see P1).
Scanning the potential further into the positive potential range
allows a further oxidation response to be observed (see P2)
which is associated with the presence of glucose (as reducing
agent) in the aqueous solution. The characteristic shape of this
glucose oxidation process is caused by the electrocatalytic
nature of gold towards glucose oxidation and this feature is
(
in aqueous 0.1 M NaOH) and therefore anthraquinone
should be an effective redox mediator and catalyst for indigo
reduction.
3
.2 Monitoring the glucose-driven reduction of indigo
22
consistent with literature reports. Upon reversal of the scan
to leuco-indigo at vibrating gold electrodes
direction, a further prominent peak response is observed at
À0.8 V vs. SCE (see P3). This response is associated with the
reductive dissolution of indigo from the electrode surface back
into the solution phase. Closer inspection shows that there are
two peak features associated with process P3 and this is in
good agreement with data in Fig. 4A. The first reduction peak
is linked to the fast surface reduction of indigo (fast surface
process) and the second broader peak is associated with the
follow-up dissolution kinetics (slow bulk process). The
The progress of the reduction of solid indigo can be monitored
conveniently by measuring the solution concentration of
leuco-indigo. A reliable methodology based on a vibrating
2
4
gold electrode has been developed recently. In this method,
fast mass transport at the surface of a vibrating (250 Hz) gold
disc electrode (500 mm diameter) is exploited for the steady-
state mass-transport-limited determination of the leuco-indigo
concentration in solution.
Fig. 5A shows an overlay of cyclic voltammograms (as a
function of time) recorded during the reduction of indigo by
glucose. Upon scanning the potential of the vibrating electrode
from À1.0 V vs. SCE positive, a sigmoidal oxidation response
Fig. 6 (A) Plots of the limiting currents of the leuco-indigo oxidation
versus time for the reduction of 30 mg of indigo (corresponding to 1.1 mM)
3
with 22 mM glucose in 100 cm aqueous 0.1 M NaOH at 65 1C.
À1
Fig. 5 (A) Cyclic voltammograms (scan rate 0.2 V s ) monitoring the
The concentration of anthraquinone (present as solid) was (i) 30 mM,
(ii) 15 mM, (iii) 3.0 mM, (iv) 1.5 mM, and (v) 0.8 mM. (B) Plots of the
limiting currents of the leuco-indigo oxidation versus time for the reduction
of 30 mg of indigo (corresponding to 1.1 mM) with 22 mM glucose in
glucose-driven reduction of 30 mg indigo (corresponding to 1.1 mM in
3
00 cm aqueous 0.2 M NaOH) in the presence of 22 mM glucose at 65 1C
data taken every 3 min: (i) back scan, (ii) 4 min, (iii) 7 min, (iv) 10 min).
B) Plot of the limiting currents of the leuco-indigo oxidation versus time at
i) 0.1 M and (ii) 0.2 M NaOH.
1
(
3
(
100 cm aqueous NaOH at 65 1C in the presence of 15 mM anthraquinone
with (i) 0.1 M, (ii) 0.08 M, and (iii) 0.05 M NaOH.
(
This journal is ꢀc the Owner Societies 2009
Phys. Chem. Chem. Phys., 2009, 11, 1816–1824 | 1819

View Article Online
37
Fig. 8 Crystal structure of indigo showing (A) from the top onto an
indigo sheet, (B) from the side on two indigo sheets, and (C) from the
side into a single indigo sheet. Within sheets, oxygen (red) and
nitrogen (blue) atoms are inter-linked via hydrogen bonds (dashed
lines).
in Fig. 5B. It can be seen that at 65 1C the leuco-indigo
concentration increases steadily over a 10–15 min period after
which it reaches a plateau. At this point the glucose-driven
reduction is complete and the colour has changed from blue to
orange. The slope of the initial time-course (see dotted line)
1
8
represents the initial rate of leuco-indigo formation. If the
concentration of NaOH is lower (see Fig. 5Bii) the rate of
leuco-indigo formation is severely reduced and the endpoint
apparently different. The colour of the suspension in 0.1 M
NaOH remains dark, consistent with incomplete conversion of
indigo. Therefore, the concentration of NaOH is a crucial
parameter in the effectiveness of the overall reduction process.
It is possible that the ability of glucose to generate a powerful
reducing agent depends on the level of hydroxide present in
solution. Furthermore, the lack of progress in the reductive
Fig. 7 (A) Plots of the limiting currents of the leuco-indigo oxidation
versus time for the reduction of 30 mg of indigo (corresponding to
3
.1 mM) with 22 mM glucose in 100 cm aqueous 0.1 M NaOH at
1
(
i) 55 1C, (ii) 65 1C, and (iii) 75 1C. (B) Plots of the estimated reaction
À1
rate for leuco-indigo formation (see dotted lines in A) versus T
.
(C) Plot of the limiting currents of the leuco-indigo oxidation versus
time for the reduction of 30 mg of indigo (corresponding to 1.1 mM)
with (i) 2.8 mM, (ii) 5.5 mM, (iii) 11 mM and (iv) 22 mM glucose in
3
100 cm aqueous 0.1 M NaOH in the presence of 15mM anthraquinone
and at 65 1C.
+
dissolution is also linked to the activity of Na which is
experiment has to be conducted with a sufficiently fast scan
rate in order to avoid blocking of the electrode due to
competing with protons for the anionic binding sites on the
surface of the reacting indigo crystal (vide infra). In order to
improve the rate and efficiency of the indigo reduction (while
limiting the need for alkali during the process), redox catalysts
can be introduced. One such redox catalyst as demonstrated in
this study is anthraquinone.
2
4
formation of the indigo deposits.
The limiting current plateau determined at À0.6 V vs. SCE is
a direct measure of the leuco-indigo concentration. The
formation of leuco-indigo in 0.2 M NaOH with time is shown
1
820 | Phys. Chem. Chem. Phys., 2009, 11, 1816–1824
This journal is ꢀc the Owner Societies 2009

View Article Online
Fig. 9 SEM images (low magnification) of indigo crystals (A) before reduction, (B) after 3 min reduction, (C) after 14 min reduction,
(D) anthraquinone-dosed after 3 min reduction, (E) anthraquinone-dosed after 15 min reduction, and (F) anthraquinone-dosed after 25 min
reduction. The reduction was achieved by immersion into aqueous 0.2 M NaOH with 22 mM glucose at 65 1C which was followed by rinsing with
water and drying.
3
.3 Anthraquinone catalysis of the glucose-driven reduction
within 10–15 min (compare Fig. 5B). Low levels of anthra-
quinone effectively increase the rate and conversion level of the
reduction process. At a level of 0.3 mol% (corresponding to
of indigo to leuco-indigo
Anthraquinone derivatives are well-known redox mediator sys-
tems and they have been employed previously for the mediated
3
mM) the catalyst is still fully active and only at 0.1 mol% is
the effect lost. In comparison, a solution of 3 mM water-soluble
,8-dihydroxyanthraquinone catalyst was reported to have
33
reduction of indigo suspensions. Perhaps surprisingly, anthra-
quinone derivatives also play an important catalytic role in the
1
1
0
4,5
virtually no effect. With insufficient amounts of anthra-
quinone (less than 3 mM) the indigo reduction process fails
to go to completion and the suspension remains dark.
Increased amounts of catalyst above the critical level do not
significantly speed up the reduction process either (see
Fig. 6A). Excess of anthraquinone catalyst is likely to remain
insoluble and inactive under the reaction conditions employed.
These findings suggested that it would be profitable to
determine if the NaOH concentration could be lowered even
further in the presence of anthraquinone catalyst, but Fig. 6B
demonstrates that lowering the concentration of NaOH
significantly below 0.1 M is not compatible with the full
reduction of indigo (at 65 1C). It was noted that the suspension
remained dark and therefore the reduction incomplete even
after the plateau level of leuco-indigo had been reached.
The rate of leuco-indigo formation was followed by voltam-
metry at temperatures of 55 1C, 65 1C, and 75 1C. The plots in
bacterial indigo reduction process. Many anthraquinone deri-
vatives are toxic but the parent anthraquinone is relatively safe to
34
use and readily available. The insignificant solubility of anthra-
quinone in aqueous media can be sidestepped in the present work
by direct impregnation of the redox catalyst onto the indigo
substrate prior to the reduction experiments. The anthraquinone
catalyst was dissolved in acetone, indigo particles suspended into
this solution by sonication, and after complete evaporation of the
solvent, we obtained indigo on which anthraquinone had been
uniformly immobilised. The amount of anthraquinone on the
indigo surface was altered to determine the lowest effective level
of catalyst.
Fig. 6A shows plots for the formation of leuco-indigo from
anthraquinone-dosed indigo. In contrast to what happens in
the absence of anthraquinone, the reduction process is highly
effective in the presence of the lower concentration (0.1 M) of
NaOH, and indigo is completely converted to leuco-indigo
This journal is ꢀc the Owner Societies 2009
Phys. Chem. Chem. Phys., 2009, 11, 1816–1824 | 1821

View Article Online
Fig. 10 SEM images (high magnification) of indigo crystals (A) before reduction, (B) after 3 min reduction, (C) after 14 min reduction,
(D) anthraquinone-dosed before reduction, (E) anthraquinone-dosed after 15 min reduction, and (F) anthraquinone-dosed after 25 min reduction.
The reduction was achieved by immersion into aqueous 0.2 M NaOH with 22 mM glucose at 65 1C which was followed by rinsing with water and
drying.
Fig. 7A demonstrate the dramatic effect of temperature on the
indigo reduction process. Increasing the temperatures signi-
ficantly accelerates the reduction of indigo and from the
approximate rate of leuco-indigo formation the Arrhenius
activation energy can be estimated (see Fig. 7B).
3.4 Electron microscopy study of the glucose-driven reduction
of indigo to leuco-indigo
In order to complement the kinetic data obtained in indigo
suspension experiments, it was desirable to observe directly
indigo crystals during the reduction. Indigo crystals of almost
macroscopic size can be grown in short-path sublimation
experiments (see experimental) to obtain plate-like crystals
that owe their appearance to the layered packing in the indigo
crystal structure. Indigo crystallises monoclinic and in space
À1
The magnitude of the activation energy, 120 kJ mol , is
relatively high, and even after correction due a component
introduced by the diffusion process of leuco-indigo, the
activation energy is still indicative of a high activation barrier,
almost twice as high as that observed for the same process in
3
5
2
4
group P2 /c. Fig. 8 shows the indigo structure from three
1
the absence of the anthraquinone catalyst. A substantial
part of the observed composite activation energy could be
associated with the increase in solubility of the anthraquinone
catalyst system.
directions (red = O, blue = N, green = C) and the sheet-like
structures held by the strong inter-molecular hydrogen bonds
are clearly visible. The sheet structure at molecular level is
reflected in the formation of thin crystal plates at macroscopic
level. During the indigo reduction process, both the inter-
actions between sheets and the stronger interactions within
sheets need to be overcome.
Fig. 7C demonstrates the effect of the glucose concentration
on the indigo reduction process. Glucose is present typically
in 20-fold excess and halving the concentration of glucose
appears to not affect significantly the progress of the reaction.
Halving the concentration of glucose again, however,
does affect the process. An excess of at least 10-fold of
glucose versus indigo appears to be necessary for the reaction
to go to completion (in the presence of 15 mM anthraquinone
and at 65 1C).
Indigo crystals grown by sublimation were immobilised
onto sticky carbon tape (for mounting SEM samples) and
imaged by electron microscopy (after gold sputter coating).
Fig. 9A shows a typical assembly of plate-like crystals with
smaller crystal debris visible initially. The characteristic shape
1
822 | Phys. Chem. Chem. Phys., 2009, 11, 1816–1824
This journal is ꢀc the Owner Societies 2009

View Article Online
Fig. 11 shows a schematic representation of three types of
reductive dissolution. In the presence of protons the reduction
is limited to the edges of indigo sheets and proton intercalation
followed by delamination and further dissolution is unlikely to
occur. Protons are likely to be bound in the form of hydroxyl
groups and therefore less likely to intercalate. A similar
conclusion for related dye systems has been reached by
3
6
Dome
chronoamperometry techniques.
‘block’’ further progress of the reductive dissolution into the
´
nech et al. who employed in situ AFM
and
3
7
Protons are believed
‘
solid. In contrast, sodium cations can intercalate, which, under
sufficiently alkaline conditions, will lead to slow delamination
and disintegration of the indigo sheets. Finally, in the presence
of anthraquinone, strong adsorption of the planar anthra-
quinone molecule into the gap between indigo sheets occurs
and the catalyst is moving into the crystal like a ‘‘wedge’’
causing more rapid delamination and faster reductive
dissolution.
Further and more quantitative kinetic studies are required
to resolve the complex catalytic and dye dissolution processes
in more detail. In future, advanced imaging techniques such
3
8
in situ AFM monitoring are most likely to provide further
essential kinetic information.
4. Conclusions
It has been demonstrated that anthraquinone, although
insoluble in water, is a highly effective catalyst in the glucose-driven
reduction of indigo to leuco-indigo in aqueous 0.1 M NaOH.
Anthraquinone is adsorbed/immobilised onto the indigo
powder prior to the reduction. For indigo crystals of micron
dimension, a molar ratio of 1 : 400 for anthraquinone : indigo
is effective for complete conversion with 10-fold excess of
glucose at 65 1C (corresponding to 3 mM anthraquinone). The
temperature has a considerable effect on the anthraquinone-
Fig. 11 Schematic representation of the indigo reduction process
at the edge of a layered indigo crystal with (A) proton intercalation,
+
(
B) Na intercalation, and (C) simultaneous anthraquinone and
+
Na intercalation.
of the crystals originates from the preferred growth direction
planar to the sheet.
À1
catalysed reductive dissolution (EA E 120 kJ mol ). A
In order to monitor the indigo reduction, samples of immo-
bilised indigo crystals were immersed in glucose-containing
reduction solution and withdrawn at various times. Fig. 9B
and C show samples after 3 min and after 14 min reduction. The
smaller indigo crystal debris was immediately removed. Larger
crystals remained and were still observed after 14 min of
reduction.
molecular mechanism based on adsorption of anthra-
quinone between indigo sheets has been proposed to explain
the remarkably enhanced reductive dissolution. Anthra-
+
quinone and Na ions are believed to penetrate between the
indigo sheets and thereby help break up the solid crystal. This
conclusion reached from studies in which glucose was the
origin of the reducing potential is consistent with studies of
4
,5
When the experiment was repeated with anthraquinone-
dosed indigo (see Fig. 9D–F) similar images to those obtained
with untreated indigo were obtained. In all cases the plate-shape
of indigo crystals remains and no holes are formed. It is
therefore most likely that the reductive dissolution process
occurs from the edges and that this process is relatively slow
perpendicular to the indigo sheets. Closer inspection of Fig. 9C
and F shows etch patterns and terrace-like features which are
typical for reductive dissolution occurring from the edges.
Fig. 10 shows a set of higher magnification images. In the
presence of anthraquinone there appear to be more irregular
features possibly associated with local anthraquinone concen-
tration variations. Although quantitative information is not
obtained from these images, qualitatively they support the idea
of a reductive dissolution process occurring at the edges of
indigo crystals.
bacterial reduction
where it was concluded that the
stimulatory effect of soluble anthraquinone derivatives on
bacterial indigo reduction was best explained by an alteration
of the surface properties of the bacteria or indigo particles.
Acknowledgements
A. V. thanks the Academy of Finland and Finnish Cultural
Foundation for financial support.
References
1 K. G. Gilbert and D. T. Cooke, Plant Growth Regul., 2001, 34, 57.
2
3
A. Roessler and X. Jin, Dyes & Pigments, 2003, 59, 223.
A. N. Padden, V. M. Dillon, P. John, J. Edmonds, M. D. Collins
and N. Alvarez, Nature, 1998, 396, 225.
4 S. K. Nicholson and P. John, Biocatal Biotransform., 2004, 22, 397.
This journal is ꢀc the Owner Societies 2009
Phys. Chem. Chem. Phys., 2009, 11, 1816–1824 | 1823

View Article Online
5
6
7
S. K. Nicholson and P. John, Appl. Microbiol. Biotechnol., 2005,
8, 117.
M. A. Kulandainathan, K. Patil, A. Muthukumaran and
R. B. Chavan, Color. Technol., 2007, 123, 143.
¨
T. Bechtold, E. Burtscher, G. Kuhnel and O. Bobleter, J. Soc.
Dyers Colour., 1997, 113, 135.
23 G. Matricali, M. M. Dieng, J. F. Dufeu and M. Guillou,
Electrochim. Acta, 1976, 21, 943.
24 A. Vuorema, P. John, M. Keskitalo and F. Marken, J. Appl.
Electrochem., 2008, 38, 1683.
25 T. Grygar, F. Marken, U. Schroder and F. Scholz, Collect. Czech.
¨
6
Chem. Commun., 2002, 67, 163.
8
9
A. Roessler and D. Crettenand, Dyes & Pigments, 2004, 63, 29.
M. A. Kulandainathan, A. Muthukumaran, K. Patil and
R. B. Chavan, Dyes & Pigments, 2007, 73, 47.
26 A. M. Bond, F. Marken, C. T. Williams, D. A. Beattie,
T. E. Keyes, R. J. Forster and J. G. Vos, J. Phys. Chem. B,
2000, 104, 1977.
10 A. Vuorema, P. John, A. T. A. Jenkins and F. Marken, J. Solid
State Electrochem., 2006, 10, 865.
27 S. J. Holt and P. W. Sadler, Proc. Royal Soc. London, Ser. B, 1958,
148, 495.
1
1
1 T. Bechtold and A. Turcanu, J. Electrochem. Soc., 2002, 149, D7.
2 T. Bechtold, E. Burtscher, A. Amann and O. Bobleter, J. Chem.
Soc., Faraday Trans., 1993, 89, 2451.
28 J. N. Etters, J. Soc. Dyers Colour., 1993, 109, 251.
29 T. Vickerstaff, The physical chemistry of dyeing, Oliver and Boyd,
London, 1954, ch. IX.
1
1
3 T. Bechtold and A. Turcanu, J. Appl. Electrochem., 2004, 34, 1221.
4 T. Bechtold, E. Burtscher and A. Turcanu, J. Electroanal. Chem.,
30 A. M. Bond, F. Marken, E. Hill, R. G. Compton and H. Hu
J. Chem. Soc. Perkin Trans. II, 1997, 9, 1735.
¨
gel,
1
999, 465, 80.
31 A. Domenech, M. T. Domenech-Carbo and M. L. V. de Agredos
Pascual, J. Phys. Chem. C, 2007, 111, 4585.
32 T. Grygar, S. Kuckova, D. Hradil and J. Hradilova, J. Solid State
Electrochem., 2003, 7, 706.
33 T. Bechtold, E. Burtscher and A. Turcanu, J. Electroanal. Chem.,
1999, 465, 80.
1
1
5 A. Roessler, D. Crettenand, O. Dossenbach and P. Rys, J. Appl.
Electrochem., 2003, 33, 901.
6 A. Roessler, O. Dossenbach, W. Marte and P. Rys, J. Appl.
Electrochem., 2002, 32, 647.
7 R. S. Blackburn and A. Harvey, Environ. Sci. Technol., 2004, 38, 4034.
8 A. Vuorema, P. John, M. Keskitalo, M. A. Kulandainathan and
F. Marken, Dyes & Pigments, 2008, 76, 542.
1
1
34 The Merck Index, ed. S. Budavari, Merck & Co., New Jersey,
12th edn, 1996.
1
2
2
2
9 R. J. Ferrier and P. M. Collins, Monosaccharide chemistry,
Penguin Books, Harmondsworth, 1972, ch. 3, p. 89.
0 J. M. De Bruijn, A. P. G. Kieboom and H. van Bekkum,
Starch/Sta¨rke, 1987, 39, 23.
1 J. M. De Bruijn, A. P. G. Kieboom and H. van Bekkum, Sugar
Technol. Rev., 1986, 13, 21.
35 P. Suesse, M. Steins and V. Kupcik, Z. Kristallographie, 1988, 184,
269.
36 A. Dome
2008, 10, 1238.
37 A. Domenech and M. T. Dome
Electrochem., 2006, 10, 949.
nech and M. T. Domenech-Carbo, Electrochem. Commun.,
´ ´
´
´
nech-Carbo, J. Solid State
2 M. A. Ghanem, R. G. Compton, B. A. Coles, A. Canals,
A. Vuorema, P. John and F. Marken, Phys. Chem. Chem. Phys.,
38 S. R. Higgins, L. H. Boram, C. N. Eggleston, B. A. Coles,
R. G. Compton and K. G. Knauss, J. Phys. Chem. B, 2002, 106,
6696.
2005, 7, 3552.
1
824 | Phys. Chem. Chem. Phys., 2009, 11, 1816–1824
This journal is ꢀc the Owner Societies 2009