Download Visualization of Charge-Carrier Propagation in Water

11890
Langmuir 2007, 23, 11890-11895
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
volume.
Introduction
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
to hundreds of micrometers from the respective surfaces.1-2 The
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
charged electrodes. In this case, the charge would be continuously
supplied electrolytically, rather than being statically lodged within
the surface, and could conceivably be more ample.
We found indeed, that next to each electrode, the injection of
charge produced unexpectedly vast zones with extreme pH values.
Methods
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
concentrations of 0.2-2 mM, to produce optimum color, depending
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
* To whom correspondence should be addressed. E-mail: ghp@
u.washington.edu.
† 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.
(2) Zheng, J.-M.; Pollack, G. H. In Water and the Cell; Pollack, G. H., Cameron,
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.
Figure 1. Experimental apparatus. See text for detailed explanation.
small chamber with a transparent bottom. The chamber was situated
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
supply, which could deliver the desired voltage and current. The pH
indicator dye color, as well as the motion of negatively charged
microspheres, were captured by a color video camera (Logitech
QuickCam), equipped with a lens whose working distance was 4-8
cm. The camera was connected through USB port to a computer.
The chamber, whose floor was made of glass, had 7-mm high
Plexiglas walls covered by polymerized silicone made from Silicone
Elastomer Kit Sylgard-184 (Figure 1A,B). The platinum elctrodes
were mounted in the Plexiglas and silicone walls. The chamber had
two compartments. In the A compartment the electrodes had a length
of 26 mm and were 13 mm horizontally apart from one another.
Their distances from the floor and the top of the solution were 2 mm,
and from the nearest wall 2-3 mm. The electrodes were passed
horizontally through compartment A, turned vertically upward, turned
horizontally, and then turned vertically downward to compartment
B separated by 10 mm from one from another. One of the
compartments, either A or B, was filled with liquid for experiments.
Electrical current drawn from a regulated power supply was divided
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
10-6 A. After the ammeter, the current was sent to switch SW1, 2,
which could change the polarity of current, and then to the electrodes
10.1021/la701742v CCC: $37.00 © 2007 American Chemical Society
Published on Web 10/16/2007
Charge-Carrier Propagation in Water
Langmuir, Vol. 23, No. 23, 2007 11891
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
pH indicator.
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).
For video recording, the program Microsoft VidCap.exe was used
in the Capture/Frames mode. When the cursor was positioned in the
VidCap window on the command button “OK,” the left-hand mouse
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
captured frames or the number of frames to be captured. For storage,
the files were saved in the *.avi format. For review of videos on a
Mac OS, files were converted to the format *.mov using QuickTime.
Results
Figure 3 shows the results of applying a potential difference
between electrodes for some time. Five volts were applied. Initial
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
electrode and red color near the positive electrode. With continued
flow of current these zones grew progressively larger, and within
3 min the zones with changed color (red, pH ≈ 4 and gray-green,
pH ≈ 7-8) became substantial; between them, the zone with the
initial color (pH ≈ 5) remained. The borders between these three
zones were rarely straight; generally they appeared curved or
convoluted.
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
electrodessat least up to 3 h of observation in some experiments.
The curvature changed with time, the dynamic somewhat
resembling that of a forest fire.
In some experiments we added microspheres to examine
opacity. Results are illustrated in Figure 4. Initial color corresponds
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
charged microspheres moved from the negative electrode toward
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 earlier.1,3 By 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,
dye molecules get progressively excluded from this region. Earlier,
we found that several different dyes were excluded from the
zone near various surfaces such as Nafion.3 This result looks
much the same.
The right panel of Figure 4 shows the effect of polarity reversal.
Just after capturing the 21-min video on the left, electrode polarity
was reversed, and the results are shown in the right panel of
Figure 4. Immediately, the microspheres began moving in the
opposite direction. The frame taken 7 min after the polarity change
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
zone near the positive. That is, dye molecules were progressively
excluded from that zone.
In the next series of experiments we explored the pH behavior
using vertically oriented electrodes (Figure 1B). Representative
results are shown in Figure 5. Initially, low voltages were studied.
For small incremental increases of voltage with time, 220-1300
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
11892 Langmuir, Vol. 23, No. 23, 2007
KlimoV and Pollack
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
was reversed.
was no measurable voltage or current flow through the voltmeter resistance. After 9 min without external power, the color
everywhere in the chamber solution was the same, indicating
pH 7.
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
violet (pH 10). At the positive electrode (right), the color changed
from yellow-green (pH 7) to orange (pH 4) and red (pH < 4).
These zones widened progressively until they met (∼300 s);
from then on they remained relatively stable, with only a narrow
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
time from 11 µA up to 14 µA, presumably because of the increase
of number of charge carriers. This increase was seen consistently
(n ) 9).
The last two frames in Figure 5b show what happens during
the several minutes following cessation of current flow. When
the power supply was switched OFF and no external voltage was
applied to the electrodes, the color pattern remained for some
time, and disappeared only rather slowly.
Lingering pH difference implies long-lasting charge separation.
In some experiments we measured the ability of separated charge
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
current flow through this resistance. With time, voltage and current
decreased respectively from 4 to 6 mV and 4-6 µA initially, to
near zero in 5-10 min (n ) 5 experiments).
Charge-Carrier Propagation in Water
Langmuir, Vol. 23, No. 23, 2007 11893
Figure 5. (a) Vertically oriented electrodes, running perpendicular to the plane of each image. Voltage stepped up progressively, as indicated.
(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.
The effect of voltage reversal was explored in other experiments. 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.
During the first few moments the shapes were typically cylindrical,
but very soon they took on various odd shapes and evolved
unpredictably. The only consistent feature was that when two
zones with different colors collided, they produced narrow curved
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
larger positively charged clusters of “pure water”; the other with
an excess of OH- or larger negatively charged clusters. With pH
< 4 and pH ≈ 10 in the respective zones, the ratio of charge
carriers is more than one million. Hence, different charge carriers
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 >
10. The relative rapidity of the change presumably arose because
of the relatively higher conductivity of the salt solution, compared
to distilled water. The pH pattern was more uniform close to the
positive electrode than to the negative electrode (see panels 40110 s), where current passed through a relatively narrow channel
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.
11894 Langmuir, Vol. 23, No. 23, 2007
KlimoV and Pollack
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.
Discussion
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
negative electrode. The result was seen using pH-sensitive dyes,
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
grow away from the respective electrodes. Distinct borders could
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.
This is not the first report in which solutions of vastly different
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.4-5 In these studies, the
border between zones of different pH appears straight, not jagged
as in our experiments.
In another study6 proton migratory fronts were visualized using
the acid-base indicator bromocresol green, pH range 3.8-5.4.
(4) Macounova, K.; Cabrera, C. R.; Holl, M. R.; Yager, P. Anal. Chem. 2000,
72, 3745-3751.
(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.
The dye was added to an electrolyte solution containing 100 mM
ZnSO4 + 1 mM H2SO4. With constant electrical current through
the zinc wire electrodes, it was shown that an increase in viscosity
through glycerol additions yielded a more uniform deposit of
zinc with smaller separation between branches of metal aggregates, 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
the behavior of microspheres. When current was applied, charged
microspheres could move. Negatively charged microspheres
translated away from the negative electrode, leaving a clear zone
through which writing on a film beneath the chamber could be
clearly discerned (Figure 4). The microspheres gathered in the
positively charged zone, creating an opacity through which writing
could not be seen. The boundary between opaque and transparent
zones was jagged, much like the boundary between high-pH and
low-pH solutions.
This experiment also provided a clue for linking another
interesting phenomenon. Negatively charged microspheres are
consistently expelled from a broad zone next to many hydrophilic
surfaces.3 We found recently that pH-sensitive dyes are also
excluded.7 Results of the current experiments are similar: Both
the microspheres, and also the pH-sensitive dye, were progres(7) Yoo, H.; Pollack, G. H. Unpublished results.
Charge-Carrier Propagation in Water
sively excluded from the high-pH zone (Figure 4, lower panels).
Hence the high-pH zone, which is negatively charged, may be
similar in character to the negatively charged solute-exclusion
zonesthe latter being physically different from ordinary bulk
water.3
When the current was applied through vertically rather than
horizontally oriented electrodes, the behavior was largely similar
in that the established zones did not mix. They remained
separated by a small layer of neutral-pH water (Figure 5). Here,
the dye formed jagged boundaries with triangular emanations
from the start, moving progressively closer toward one another
with time.
When physiological solutions containing various salts were
substituted for distilled water, application current produced
qualitatively similar results, but with some notable differences
(Figure 6). Whereas the low-pH zone that formed around the
positive electrode was similar to those formed in the absence of
salts, the events around the negative electrode were unexpected.
The high-pH region first grew more or less parallel to the electrode,
but it soon gave rise to a finger-like protuberance extending
toward the positive electrode, a structure that implied a more
conductive path. Such protuberances were stable in their position
but grew in girth. Their presence indicated that grossly nonlinear
Langmuir, Vol. 23, No. 23, 2007 11895
features might be characteristic, particularly in physiological
solutions. Although such patterns were noted, they were not
studied in detail as they were not the main focus of the study.
The experimental results also showed that excess positive
charge in the low-pH zone, and excess negative charge in the
high pH zone, could drive current through a resistive load. Current
flow could persist for minutes. In the absence of the resistive
load, however, the zones were maintained next to one another
for an extended period, in spite of the huge pH and charge
gradients. Almost no mixing occurs. The thin, neutral zone
between compartments remains, and the zones themselves remain
of uniform color for an extended period.
The key question emerging from these results is why these
regions with different pH do not immediately neutralize one
another. Either the charges must somehow be neutralized by
counterionsswhich begs the question of how current could then
be driven through the resistor, or the charges must somehow be
stabilized within a matrix, as they are for example, in p- and
n-type semiconductor materials. Future experiments will need
to address these important electrochemical questions.
LA701742V