Abstract
Electrical, optical, and mechanical properties of thin films significantly differ from those of bulk materials. Therefore, characterization
methods for evaluation of thin film properties became highly important. A novel approach to the well known ‘‘Hot-Probe’’ method is
proposed and applied in our work. The conventional Hot Probe characterization method enables only the definition of a semiconductor
type, P or N, by identifying the majority charged carriers. According to the new Hot Probe technique, one can measure and calculate the
majority charged carriers concentration and its dynamic parameters. Feasibility proof of the upgraded Hot Probe method was done in Si
and Ge bulk, and in thin film semiconductor samples.
r 2006 Elsevier Ltd. All rights reserved.
Keywords: Hot-Probe method; Charged carriers concentration; Type of charged carriers; Semiconductors film properties
0026-2692/$ - see front matter r 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.mejo.2006.01.014
ARTICLE IN PRESS
G. Golan et al. / Microelectronics Journal 37 (2006) 910–915 911
2. Experimental details and all readouts were stored and analyzed in the
computerized Multi Log 710 multimeter of ‘‘ExTech’’.
The experimental laboratory setup is shown in Fig. 1. A current within the closed loop circuit of: multimeter,
Two spring-loaded Au-coated electrodes placed at a hot probe, semiconductor film, cold probe and multimeter,
distance of L ¼ 10 mm from each other create the closed is created due to a concentration drop of free charged
loop circuit with the digital multimeter. A thermo-couple carriers which generate diffusion within the sample due to
equipped heater is attached to one of the electrodes to the hot probe. There are two optional cases for such closed
warm it up to 150 1C. A ‘‘WS 81’’ soldering iron of Weller loops: (a) Ohmic contact between the probe electrodes and
was used also as heater and created a stable controllable the sample surface, (b) blocking contact. In the first case,
temperature up to 450 1C. Voltage and current measure- current is measured and follows the polarities of the
ments were carried out using a Keithley automatic test measured voltage. In the second case, closed circuit appears
station. The sheet resistance of all samples was measured as a two back to back diodes pair. Thus, the current is
using a conventional four-point probe station of Cascade defined by backward saturation current only [5]. The value
Microtech. Silicon and germanium crystalline samples and of this reversed current is very small and cannot be
germanium thin semiconductor films doped with aluminum measured by conventional techniques.
were studied in the experiment. Germanium thin films were
deposited by thermal evaporation. Aluminum powder was
attached to the germanium to provide deposition of doped 3. Results
semiconductor coatings. In order to obtain a good Ohmic
contact between the electrodes and the sample surface, we As the distance of 10 mm between the hot and cold
used an In–Ga eutectic grout. probes is not sufficiently high, the hot electrode conducts
The measurement technique was as follows: heat toward the cold probe on top of the surface plane. In
A hot electrode was connected to the positive input of a order to check this temperature influence on the second
digital voltmeter. Then, the hot electrode was heated up to electrode we measured simultaneously the temperatures in
the required temperature and attached to the sample both positions of the hot electrode and cold electrode by
surface. Measurement was done for 40–60 s approximately independent thermo-couple temperature sensors. Fig. 2
presents the obtained temperature characteristics of the
cold electrode in various processing times. It can be seen
that maximum temperature at the second electrode (cold
probe) after 40 s has not exceeded more than 30 1C, for
the highest possible temperature of the hot electrode
(450 1C). In lower temperatures, this difference was lower
and for 150 1C on the hot electrode, the cold probe
temperature was no more of 10 1C. Generally, the maximal
variation of the cold point temperature was not more than
6.7% from other hot-probe temperature. Therefore, we
may assume that the second electrode temperature devia-
tion can be neglected in the initial stage of measurements
and it must be considered farther.
Fig. 3 presents a hot probe characteristics family of
curves, measured in a germanium sample for various
Measurement 150°C
60
Cold electrode temperature,°C
200°C
55
250°C
50
300°C
45
350°C
40 400°C
35 450°C
30
25
Fig. 1. Experimental laboratory setup for a hot-probe measurement. Fig. 2. Cold electrode temperature while measurement.
ARTICLE IN PRESS
G. Golan et al. / Microelectronics Journal 37 (2006) 910–915 913
40 150°C 14
Measured voltage, mV
35
200°C Temperature 150° C
30 12
25 250°C
10
20 300°C
8
15 400°C
10 6
5 4 (1.2, 0.01) Initial state
0
0 10 20 30 40 50 60 70 80 90 100 110 120 2
Processing time, sec 0
0 2 4 6 8 10
Fig. 3. Hot-probe characteristics for n-type germanium in various Time,sec
temperatures.
Fig. 4. Hot-probe characteristic of germanium, measured at 150 1C.
Measured current, µA
-5
Fig. 4 presents an initial part of the hot-probe -10
characteristic for a germanium chip measured at 150 1C. -10
This figure demonstrates the relationship between the -20
-15 voltage
initial state and the steady state. Averaged hot-probe
characteristics for a p-type mono-crystal silicon sample and current
-20 -30
for a p-type polycrystalline silicon sample are presented in
Figs. 5 and 6. These characteristics are very similar for -25 Temperature 100 °C
various crystal forms of silicon. Fig. 6 shows the voltage -40
and current characteristics measured in the same sample. 0 10 20 30 40 50 60 70 80
Processing time, sec
As shown, the voltage and current curves are almost of the
same nature and the conductance of these samples is Fig. 6. Averaged hot-probe characteristics for p-type polycrystalline
defined by the conventional Ohm’s law. silicon sample.
Hot-probe characteristics for germanium thin films on
glass substrates are presented in Fig. 7. These character-
istics show measured voltages as a function of heating time material changes the sign of the majority charge carriers
for two different temperatures, and shows the same in the grown germanium film. Comparing hot-probe
behavior of a semiconductor regardless its type. Obtained characteristics measured for bulk, and thin film semicon-
films are of p-type and n-type as shown from the negative ductor materials (see Figs. 3, 5–7) demonstrates that these
(a) and positive (b) values of the measured voltages. materials behave the same way irrespective of their crystal
Increasing of aluminum in the precursor evaporating structure and/or thickness. Thus, the same measurement
ARTICLE IN PRESS
914 G. Golan et al. / Microelectronics Journal 37 (2006) 910–915
0.0 P-type germanium film. Here D is the diffusion coefficient of the charged carriers.
Now, using the relation Q ¼ qNðx; tÞ the Poisson’s
-0.1 100° C equation (1) takes the following form:
Measured voltage, mV
140° C
-0.2
qE q Q x2
¼ pffiffiffiffiffiffiffiffi exp . (13)
qx 2pe0 er pDt 4Dt
-0.3 If we designate
-0.4 q Q x2 pffiffiffiffiffiffi
B¼ pffiffiffiffiffiffiffiffi ; y2 ¼ and dx ¼ 2 Dt dy,
2pe0 er pDt 4Dt
-0.5
0 10 20 30 40 50 Eq. (13) becomes:
(a) Processing time, sec
qE 2
¼ B ey (14)
0.8 N-type germanium film. qx
0.7 which is a classic equation that cannot be solved using
elementary functions [6]. However, the right part of
Measured voltage, mV
0.6
Eq. (14) may be represented in a series form for yo1:
0.5
0.4 2 y2 y4 y2n
ey ¼ 1 þ þ ð1Þ2 . (15)
0.3
1! 2! n!
60° C
0.2 In our case, for x ¼ L ¼ 1 cm and a time of 5–10 s at
100° C temperatures of 300–423 K, yo1. Therefore, Eq. (14) may
0.1 be integrated partially
0.0
0 10 20 30 40 50 60 Z y
E 2 y3 1 y5
(b) Processing time, sec pffiffiffiffiffiffi ¼ B ey dy ¼ By B þB . (16)
2 Dt 0 3 25
Fig. 7. Hot-probe characteristics for germanium thin films doped with The voltage V measured between the hot probe and the
aluminum during deposition in glass substrates: (a) p-type film, (b) n-type
cold probe may express from E by integration of the
film.
equation