Visualization of Charge-Carrier Propagation in Water
Andrey Klimov†and Gerald H. Pollack*
Department of Bioengineering, Box 355061, UniVersity of Washington, Seattle, Washington 98195
ReceiVed June 12, 2007. In Final Form: August 7, 2007
The electrical properties of water in the region between parallel electrodes were investigated using pH indicator
dyes. Different pH values corresponded to different colors, which could be registered by a video camera. Imposition
of electrical current was able to produce zones of constant pH around, and well beyond each electrode: extremely
low pH around the positive electrode and extremely high pH around the negative electrode. The border between
alkaline and acid zones was jagged and separated by only a narrow layer of water with neutral pH. When the water
was replaced by various salt solutions, similar zones were observed. Again, passage of current produced large zones
of extreme pH values near and beyond each electrode. Alkaline zones appeared to propagate from the negative to
the positive electrode in narrow channels through the neutral solution. When the power supply was disconnected from
the electrodes and replaced by a resistive load, a potential difference was registered, and current flowed through the
resistor for some period of time. Hence, the acid and alkaline zones appear to carry opposite charges throughout their
Previous work from this laboratory revealed an unexpected
observation: solutes were profoundly excluded from aqueous
zones in the vicinity of various charged surfaces. These surfaces
included hydrogels, ion-exchange resins, and polymers as well
as biological entities, and the exclusion zones could extend up
water in this zone was physically different from bulk water, and
appeared to be charged.3
The question arose whether similarly charged aqueous zones
might be found when nucleating surfaces were replaced by
the surface, and could conceivably be more ample.
We found indeed, that next to each electrode, the injection of
The following components were used to study electrolytic
generation of charge carriers:
(1) A universal pH Indicator (Sigma no. 36803), made from a
mixture of different dyes whose color depends on pH, added in
on thickness of the water layer.
(2) Carboxylated polysterene microspheres, 1 µm diameter, the
surfaces of which are covered with hydroxyl groups which are able
to assume negative charge in aqueous solutions at pH > 3.
(3) In one experiment a physiological salt solution was used,
containing universal pH indicator, 0.5 mM EGTA, 100 mM KCl,
20 mM potassium phosphate, 0.2 mM MgCl2, 2 mM DTT. Initial
pH ) 7.0.
In general, two electrodes, made from platinum-wire welding
rods, diameter 0.32 mm, were situated parallel to one another in a
on a table with transparent glass surface, illuminated from below
with light from tungsten bulb, scattered by white paper to produce
uniform illumination. The electrodes were connected to a power
indicator dye color, as well as the motion of negatively charged
microspheres, were captured by a color video camera (Logitech
cm. The camera was connected through USB port to a computer.
The chamber, whose floor was made of glass, had 7-mm high
Elastomer Kit Sylgard-184 (Figure 1A,B). The platinum elctrodes
were mounted in the Plexiglas and silicone walls. The chamber had
of 26 mm and were 13 mm horizontally apart from one another.
and from the nearest wall 2-3 mm. The electrodes were passed
horizontally, and then turned vertically downward to compartment
B separated by 10 mm from one from another. One of the
10 times by external resistors. In this way, external voltage could
be regulated with 10 times better accuracy than with the power
supply regulator. Current from the divider was sent to a Fluke 8020
multimeter, which was able to measure current with a precision of
10-6A. After the ammeter, the current was sent to switch SW1, 2,
* To whom correspondence should be addressed. E-mail:
†Current address: Institute of Theoretical and Experimental Biophysics,
Puschino, Russia 142290.
(1) Zheng, J. M.; Pollack, G. H. Phys. ReV. E: Stat., Nonlinear, Soft Matter
Phys. 2003, 68, 031408.
I. L., Wheatley, D. N., Eds.; Springer: New York, 2006; pp 165-174.
(3) Zheng, J.-M.; Chin, W.-C; Khijniak, E.; Khijniak, E., Jr.; Pollack, G. H.
AdV. Colloid Interface Sci. 2006, 127, 19-27.
Langmuir 2007, 23, 11890-11895
10.1021/la701742v CCC: $37.00© 2007 American Chemical Society
Published on Web 10/16/2007
in the chamber. The voltage applied to the electrodes was measured
with the help of a second high impedance multimeter. When switch
SW 3 was OFF, the power supply was disconnected, and the same
multimeter could measure the voltage produced in the chamber.
Before each experiment electrodes were cleaned mechanically and
by passing electrical current in water containing 1 M KOH with
direct and reversed polarity of current.
When microspheres were used, a transparent plastic film with
text was placed under the transparent glass floor of the chamber.
This allowed us to check solution opacity.
The video camera was positioned so that a sharp image of the
electrodes was visible along their entire length. pH was determined
by comparing the color on the video image with the color on the
color scale that came with the pH indicator (cf. Figure 2).
button could be actuated not only manually but also on command
from a program, written for selecting arbitrary intervals between
frames at any moment of time during the experiment. When the
program activated the command button “OK,” electrical impulses
were supplied through the serial port of the computer, resistor, and
diode on the base of the n-p-n transistor 2N2206, hooked up in
parallel to the left mouse button. When the cursor was placed in one
of the program’s text boxes, we were able to change time between
the files were saved in the *.avi format. For review of videos on a
Figure 3 shows the results of applying a potential difference
current was ∼20 µA. The first frame was captured 10 s after
current was applied to water whose initial pH was ∼5. One may
see a very narrow strip of gray-green coloring near the negative
zones were rarely straight; generally they appeared curved or
After some time, the acid (red, pH < 4) and alkaline (violet,
pH ≈ 10) zones met one another, and only a very narrow layer
with the initial color remained between them (right panel). This
layer persisted as long as current continued to pass through the
The curvature changed with time, the dynamic somewhat
resembling that of a forest fire.
In some experiments we added microspheres to examine
to pH 6. The presence of microspheres increases opacity, which
blurs the visibility of characters on a film placed beneath the
chamber. When 5 V were applied, the pH around the electrodes
began to change as described above. Meanwhile, the negatively
the positive. After 21 min, text could be seen under the bottom
of the chamber, as microspheres in solution migrated from the
negative electrode, leaving the wide zone transparent. This zone
resembled the microsphere-free “exlusion zone” reported ear-
lier.1,3By contrast, the concentration of microspheres in the red
zone (pH < 4) close to the positive electrode grew, as did the
opacity. The letters on a paper are not at all visible.
A notable feature is that the negative electrode zone looks
“cleaner” than the positive electrode zone; i.e., the green-violet
color (pH > 8) gets progressively fainter with time. Apparently,
we found that several different dyes were excluded from the
zone near various surfaces such as Nafion.3This result looks
much the same.
was reversed, and the results are shown in the right panel of
Figure 4. Immediately, the microspheres began moving in the
is shown at the top. The yellow color in the middle zone
corresponds to pH ≈ 6. By 27 min, most of the microspheres
were concentrated close to the new positive electrode, and the
zone near the new negative electrode became cleaner than the
excluded from that zone.
using vertically oriented electrodes (Figure 1B). Representative
mV (n ) 9 experiments), current remained less than 1 µA, and
only small changes in solution features were seen (Figure 5a).
When the highest-level voltage (1.3 V) was switched off, there
Figure 2. pH color scale for universal pH indicator (Sigma Aldrich).
Figure 3. Time course of color change in images captured in chamber “A” (Figure 1), with horizontal electrodes and water containing a
Charge-Carrier Propagation in WaterLangmuir, Vol. 23, No. 23, 2007 11891
was no measurable voltage or current flow through the volt-
meter resistance. After 9 min without external power, the color
everywhere in the chamber solution was the same, indicating
At higher voltages, progressive color changes could be
observed, and they were largely similar to those observed with
the horizontal electrodes. Figure 5b shows the time course of
color change at an applied voltage of 4.3 V. At the negative
electrode (left), the color changed from yellow-green (pH 7) to
from yellow-green (pH 7) to orange (pH 4) and red (pH < 4).
These zones widened progressively until they met (∼300 s);
zone of neutral pH situated in between.
Meanwhile, the current did not remain constant. As can be
seen from the panels, the current increased progressively over
(n ) 9).
The last two frames in Figure 5b show what happens during
the several minutes following cessation of current flow. When
applied to the electrodes, the color pattern remained for some
time, and disappeared only rather slowly.
to drive current. When the external power supply is in the OFF
position, the output resistance of the power supply (112 Ω)
remains connected to electrodes, and it is possible to measure
decreased respectively from 4 to 6 mV and 4-6 µA initially, to
near zero in 5-10 min (n ) 5 experiments).
Figure 4. Effect of current flow on solution opacity. On the left panels, the positive electrode is at the bottom. On the right side, the polarity
11892 Langmuir, Vol. 23, No. 23, 2007KlimoV and Pollack
The effect of voltage reversal was explored in other experi-
ments. In Figure 5c, voltage with reversed polarity was applied
following the 1100 s sequence shown in Figure 5b.
In the figure it is possible to see new zones growing around
each electrode. Their colors are complementary to the originals.
but very soon they took on various odd shapes and evolved
unpredictably. The only consistent feature was that when two
zones with neutral pH (gray/green), where negative OH-and
positive H+or H3O+ions apparently annihilated one another
and produced neutral water. Ultimately, the new pattern was the
inverse of the one obtained prior to reversal, indicating that the
process is reversible, and dye composition was not appreciably
changed by prolonged flow of currents.
From Figure 5, perhaps the most interesting finding is that in
solutions of pure water with dye, two distinct zones exist with
different charge carriers: one with an excess of H+, H3O+, or
< 4 and pH ≈ 10 in the respective zones, the ratio of charge
carry current in the respective zones.
The effect of replacing distilled water with physiological salt
solution is shown in Figure 6. The salt solution contained the
Universal pH indicator, 0.5 mM EGTA, 100 mM KCl, 20 mM
potassium phosphate, 0.2 mM MgCl2, 2 mM DTT. Initial pH )
7.0. When a potential difference of 4.3 V was applied between
the electrodes, the pH around the positive electrode changed
very quickly to <3, and around the negative electrode to pH >
to distilled water. The pH pattern was more uniform close to the
resembling a finger. With time, channel width increased, and
finally the alkaline solution occupied 70-80% of the space
between electrodes, the remainder filled with acid solution.
(b) Time course of pH change near vertically oriented electrodes at 4300 mV. Last panel shows color change following turnoff of input power
at 900 s. (c) Continuation of previous experiment (Figure 5b) with V ) 4.3 V and electrical polarity reversed.
Charge-Carrier Propagation in WaterLangmuir, Vol. 23, No. 23, 2007 11893
Electrolytic imposition of current into a chamber with water
is able to produce zones with low and high pHsextremely low
around the positive electrode and extremely high around the
both in distilled water and in salt solution.
The surprising aspect was the extent of these two zones. Soon
after onset of current flow, the acid and alkaline zones began to
be seen between each growing zone and the neutral zone in
between. Boundaries between zones were initially straight and
ran more or less parallel to the horizontally oriented electrodes
(Figure 3: 10 s, 180 s). The zones continued to grow and finally
met one anothersseparated only by a narrow zone of neutral
water (Figure 3, 240 s). At this stage the border between zones
became more jagged, constantly moving about.
pH could coexist beside one another. For example, pH gradients
were electrochemically formed in microfluidic channels and
optically quantified using acid-base indicators, for isoelectric
focusing of sample biological analytes.4-5In these studies, the
as in our experiments.
the acid-base indicator bromocresol green, pH range 3.8-5.4.
through glycerol additions yielded a more uniform deposit of
zinc with smaller separation between branches of metal ag-
gregates, i.e., a change in morphology from more separated
compact trees having branches of metal aggregates, to a more
dense, fractal-like structure. The border between acid and more
neutral zones generally appeared straight.
Thus, others have reported extensive zones of differing pH
adjacent to one another, although the emphasis of these studies
was considerably different from the emphasis here.
In addition to noting extensive pH separation, we examined
microspheres could move. Negatively charged microspheres
through which writing on a film beneath the chamber could be
clearly discerned (Figure 4). The microspheres gathered in the
This experiment also provided a clue for linking another
interesting phenomenon. Negatively charged microspheres are
surfaces.3We found recently that pH-sensitive dyes are also
excluded.7Results of the current experiments are similar: Both
the microspheres, and also the pH-sensitive dye, were progres-
(4) Macounova, K.; Cabrera, C. R.; Holl, M. R.; Yager, P. Anal. Chem. 2000,
(5) Cabrera, C. R.; Finlayson, B.; Yager, P. Anal. Chem. 2001, 73, 658-666.
(6) Gonzalez, G.; Marshall, G.; Molina, F. V.; Dengra, S.; Rossoc, M. J.
Electrochem. Soc. 2001, 148, 479-487.
(7) Yoo, H.; Pollack, G. H. Unpublished results.
Figure 6. Images of pH changes seen in ordinary physiological buffer during the imposition of current. Electrodes run horizontally. V )
4.3 V. (Top) Positive. (Bottom) Negative.
11894 Langmuir, Vol. 23, No. 23, 2007 KlimoV and Pollack