Photo-electrical measurements

All the measurement systems were self-built. The specifications for the setup are described in the following paragraphs.

2.3.1 Current-voltage (I-V) measurement

Figure 2.3.1 shows the apparatus for the basic I-V measurement, which included two parts—the light source and the electrical measuring system. Newport Oriel 96000 150W Xe lamp and AM1.5G filter 81094 were used as the solar simulator; the optical power of incident light was measured by the thermopile detector 818P-010-12 and calibrated to 100mW/cm² before measuring. For characterizing the I-V properties, the electrochemical potentiostat / galvanostat EG&G model 273A was connected to the device in two probing modes; conventional solar cells were tested to identify the validity of the measuring system and the results were consistent with I-V characteristics from other equipments.

For the measurement of solar cell I-V characteristic, specific mask with defined transparent area was used so that the effective area among each measurement could be kept being constant. In this experiment, since the active area was defined during processing, no additional mask was necessary in the photo-electrical measurements.

The I-V characteristics with/without mask applied are shown in Fig.2.3.2 and the difference in these two curves was in the I_sc region, which is similar to the result in Ref.1. We, thus, thought the lowering of photocurrent is mainly due to the edge shading effect caused by the presence of mask. It seems that the I-V characteristics were not changed no matter the mask was employed or not. As a result, in the experiment all the measurements have been done without mask.

11

Experimental Methods It should be mentioned that all the photoelectrical properties presented in the experiment were characterized under stable equilibrium situations and the devices were all encapsulated. For evaluating the fabricated solar cell, the stability is indispensible. Fig. 2.3.3 shows the current-voltage characteristic for repeated measurement.

Fig.2.3.1 setup of I-V and EIS systems

12

Experimental Methods

0 400 800

-4000 -3000 -2000 -1000 0

1000 LIGHT1

DARK LIGHT2

current(uA/cm2 )

voltage(mV)

0 400 800

-6000 -4000 -2000 0 2000 4000

light dark masked light

current(uA/cm2 )

voltage(mV) edge shading caused current lowering

Fig.2.3.3 current lowering cue to use of mask in measurement Fig.2.3.2 consistency in photocurrent in repeated measurements

13

Experimental Methods 2.3.2 Electrical impedance spectroscopy (EIS) [2, 3, 4]

The setup of EIS measurement was basically the same as the I-V measurement be derived. Electrical impedance spectroscopy measures the variation of impedance at different frequencies, so it is a trace of the variation of the impedance of measured sample.

Normally, the measurement for DSSC was done under the standard illumination, which was the condition for the I-V test, with DC bias applied. The value of the DC bias was selected to be the same as the open-circuit of the measured device so that the device was under the static condition, i.e. there was almost no DC current when the measurement was proceeding. If not specially claimed, the conditions for EIS measurement in the experiment were always that the AC amplitude was of 10mV descending from 500 kHz to 0.01 Hz and DC bias equaled to open-circuit voltage.

2.3.3 Photovoltage transient

The photovoltage transient measures the relaxation behavior of the open-circuit voltage whereas the photocurrent transient measures the decay of the short-circuit current; the working point at each measurement is shown in Fig.2.3.9. Due to the unique property of DSSC, the relaxation behavior of photovoltage corresponds to the

14

Experimental Methods movement of electron quasi Fermi level of titanium dioxide. Based on the diffusion model and the proposed random walk model for electron transporting in nano-crystal [5, 6, 7], the recombination rate could be derived through the relation [8]

1 rec

kT dV q dt



 

  

 

The setup of the system is depicted in Fig.2.3.4; similar to previously described measurement, the system includes light source and electrical measuring circuit.

Usually expanded laser or LED [9, 10] was chosen to be the light source so that the energy of incident photon was the same and thus led to the excited electrons having the similar energies. To make a pulse of incident light, a solid state switch circuit controlled by a function generator was used; the produced pulse width was from few seconds to 100 nano seconds as shown in Fig.2.3.5. The photovoltage was recorded by a digital oscilloscope (Tektronix TDS series).

Fig.2.3.4 photovoltage/photocurrent transient system

15

Experimental Methods

-1.0x10^-6-8.0x10^-7-6.0x10^-7-4.0x10^-7-2.0x10^-7 0.0 2.0x10^-74.0x10^-76.0x10^-78.0x10^-71.0x10^-6 -2

-1 0 1 2 3 4 5

voltage(V)

time (sec)

2.3.4 Photocurrent transient [11, 12]

The light source for photocurrent transient was the same as photovoltage transient, but an amplifier was necessary to transform short circuit current to voltage such that the signal could be measured by digital oscilloscope. The transient behavior of the amplifier was simulated and the result is shown in Fig.2.3.6. According to the result, the limit of the amplifier can be determined and was found to be adequate in the measurement since the limit of response time is shorter than the relaxation time.

Time

0s 10us 20us 30us 40us 50us 60us

V(U1:OUT) I(I1)*2200 0

1.0m 2.0m 3.0m

Fig.2.3.5 pulse generation limit of the solid switch circuit

Fig.2.3.6 consistency in 10us transient response of the amplifier circuit

16

Experimental Methods 2.3.5 DOS measurement [13, 14]

Fig.2.3.7 shows the apparatus of DOS measurement; briefly speaking, it is a combination of photovoltage transient and a background light. During the measurement, the intensity of incident pulse remains the same while the background light intensity is modified by setting different ND filters. Since the intensity of the pulse is constant, the excess electrons injected from dye to titanium dioxide are remained constant among each measurement. This small perturbation will induce a small shift of electron quasi Fermi level, which could be different depending on the level of background intensity (Fig.2.3.8). The density of state is proportional to the inverse of perturbation induced voltage difference, so the distribution could be revealed by plotting background induced voltage level versusV^1.

17

Experimental Methods

Fig.2.3.8 the basic concept of DOS measurement Fig.2.3.7 combined DOS system

18

Experimental Methods References

[1] Naoki Koide; Liyuan Han, Review of Scientific Instruments Volume 75 Number 9;

September 2004.

[2] R. Kern; R. Sastrawan; J. Ferber; R. Stangl; J. Luther, Electrochimica Acta. 2002, 47, 4213.

[3] Juan Bisquert, J. Phys. Chem. B 2002, 106, 325.

[4] T. Hoshikawa; M. Yamada; R. Kikuchi; K. Eguchi, J. Electrochem. Soc. 2005, 152, E68.

[5] Jenny Nelson. Phys. Rev. B 1999, 59, 23.

[6] J. van de Lagemaat; A. J. Frank, J. Phys. Chem. B 2001, 105, 11194.

[7] K. D. Benkstein; N. Kopidakis; J. van de Lagemaat; A. J. Frank, J. Phys. Chem. B 2003, 107, 7759.

[8] J. Bisquert; A. Zaban; M. Greenshtein; I. Mor-Sero, J. Am. Chem. Soc. 2004, 126, 13550;

F. Fabregat-Santiago; J. Garcia-Canadas; E.Palomares; J. N. Clifford; S. A. Haque; J. R.

Durrant; G. Garcia-Belmonte; J. Bisquert, J. Appl. Phys, 2004, 96, 6903.

[9] Shogo Nakade; Taisuke Kanzaki; Yuji Wada; Shozo Yanagida, Langmuir 2005, 21, 10803.

[10] Shogo Nakade, Electron Transport in Nano-porous TiO2 Films and its Effect on Dye-Sensitized Solar Cells.

[11] Jun-Ho Yum; Shogo Nakade; Dong-Yu Kim; Shozo Yanagida, J. Phys. Chem. B 2006, 110, 3215.

[12] Shogo Nakade; Taisuke Kanzaki; Wataru Kubo; Takayuki Kitamura; Yuji Wada; Shozo Yanagida, J. Phys. Chem. B 2005, 109, 3480.

[13] B. C. O’Regan; S. Scully; A. C. Mayer, J. Phys. Chem. B 2005, 109, 4616.

[14] H. J. Snaith; A. J. Moule; C. Klein; K. Meerholz; R. H. Friend; M. Gratzel, Nano Lett. (Letter) 2007, 7(11), 3372.

Dye Sensitized Solar Cell

19