Biomedical Engineering Reference
In-Depth Information
respectively. In other words, the measured photoemf by means of the null-current method
as described by Eq. (15.9) is the apparent photoemf, the effectiveness of which is reduced
by shunting, whereas the measured conductance during illumination actually contains a
contribution from the dark ionic conductance. The photoconductance as determined by
Eq. (15.10) is more appropriately referred to as the
combined conductance during illumination
(
G
p
G
m
), or simply
apparent photoconductance
for brevity. Since
G
m
can be independently
measured, the true photoemf
E
p
and the true photoconductance
G
p
can be calculated by
virtue of Eqs. (11) and (12). The rigorous mathematical derivation of Eqs. (11) and (12) has
been published elsewhere [69].
The result of a typical null-current measurement is shown in Figure 15.13. We confirmed
a widely reported observation [70]: the photocurrent and the apparent photoemf are
linearly dependent on the applied voltage, and the photocurrent reverses its polarity at a
certain applied voltage (
70 mV in Figure 15.13A). In contrast, the true photoemf and the
photoconductance are voltage independent (Figure 15.13B and Figure 15.13C). Our inter-
pretation of the linear voltage dependence of the photocurrent and its polarity reversal is
therefore quite different from what were offered by other investigators (discussed in [69]).
Our experimental result indicates that the fraction of photocurrent driven by light is volt-
age independent (Figure 15.13B). The apparent linear dependence of the photocurrent is the
consequence of the current driven by the applied voltage through the
same
proton conduc-
tance channel activated by light. Therefore, the apparent linear dependence reflects the lin-
ear dependence of the voltage-driven current on the applied voltage by virtue of Ohm's
law. Furthermore, the polarity reversal is a consequence of applying an increasingly oppos-
ing voltage, which eventually overcomes the true photoemf. Thus, the simple result here
actually justifies the use of a potentiometric method—the null-current method. A simple
explanation of the linear voltage-dependence of the photocurrent was thus made possible.
So far, we have ignored the “spike” transients that appear at the onset and at the cessa-
tion of illumination (Figure 15.12B). Elementary analysis indicates that a linear high-pass
RC
filter exhibits a similar waveform (cf. Figure 15.4). Thus, the transient spikes are a man-
ifestation of chemical capacitance. However, the nature of chemical capacitance (a built-in
high-pass filter) alone does not explain the asymmetry of the two peaks. In Figure 15.12B,
the positive spike is more prominent and decays faster than the negative spike. This asym-
metry can be explained on the basis of “step-function” photoswitching:
R
s
(or 1/
G
p
)
assumes two different values, being much greater (virtually infinity) in the dark than in
the light (a finite value). Therefore, the symmetry of the two spikes predicted by the equiv-
alent circuit analysis on the assumption that
R
s
is a constant (or infinity, as assumed in the
computation of the time course of the AC photoelectric signal) and does not hold any
longer. The decay of the transient will be faster in the light than in the dark because of the
effect of light upon
R
s
(cf. Eq. (15.8)). Again the condition of the zero time-integral requires
that the area enclosed by the two transient decay curves be equal. This is possible only if
the “on” peak is more prominent than the “off” peak.
Our experimental result indicates that the proton conduction pathway is
not rectified
: the
current driven by the applied voltage encounters the same resistance in either direction
during illumination, that is, bR is not a photodiode in the strict sense (Figure 15.14). The
notion that the photoemf and the applied transmembrane voltage are interchangeable for
driving a transmembrane proton current is however consistent with the basic principle of
bioenergetics. According to the chemiosmotic theory of Mitchell [71] (see also a similar
theory by Williams [72]), the converted energy by bR photoreaction is stored as a trans-
membrane
electrochemical
gradient of protons. Furthermore, both the electrical component
(
pH) of this gradient are equivalent and are both avail-
able for ATP synthesis. The latter condition holds because of the common pathway for
light-driven and voltage-driven proton currents.
V
) and the chemical component (
