INVESTIGATIONOFBANDGAPOFA SEMICONDUCTORCRYSTAL

K.SURESH SENANAYAKE PS/2004/225 PHYS 43115

INVESTIGATION OF BAND GAP OF A SEMICONDUCTOR CRYSTAL

Investigation of Band Gap of a Ge Crystal by using Four Probe Method

Four Probe Method

PS/2004/225

INDEX
INDEX ......................................................................................................................................1 INTRODUCTION..................................................................................................................2 APPARATUS ..........................................................................................................................4 THEORY ..................................................................................................................................5 EXPERIMENTAL PROCEDURE ........................................................................................7 OBSERVATIONS ..................................................................................................................8 RESULTS ...............................................................................................................................10 ERROR ANALYSIS .............................................................................................................15 CONCLUSION.....................................................................................................................16 DISCUSSION .......................................................................................................................17 REFERENCES .......................................................................................................................18

Department of Physics

University of Kelaniya

Four Probe Method

PS/2004/225

INTRODUCTION
This introduction will be broken into two parts; a description of the Ge crystal structure, and a detailed description of the four probe Resistivity experimental technique that was used. Crystal Structure In solids, atoms are arranged in an ordered, dense packing, forming a crystal. Silicon, germanium and carbon crystals have identical structure, called diamond structure. A single atom has a wave function associated with it and discrete potential energy levels, but in a solid with a many densely packed atoms, the wave functions overlap and interact and the rather than a single energy level it is split and spread. A solid of N atoms has N times as many levels as each atom. Because of the enormous number of atoms in a crystal (1023 cm-3) this results in bands of energy levels. The band gap Eg separates the valence band and the conduction band. A semiconductor has a band gap in the order of 1 eV. At absolute zero temperature the valence band is completely filled and the conduction band is empty, thus no conduction takes place. As the temperature increases some electrons gain energy greater than the band gap and move to the conduction band, leaving a hole in the valence band. The electron in the conduction band and the hole in the valence band hence contribute to conduction in an applied electric field. This is called intrinsic conduction. The net electrical conductivity is thus the sum of both the negative (electrons) and the positive (holes) charge carriers. To enhance the conduction, and other properties, a pure material may be doped with n-type or p-type impurity elements (resulting in extrinsic and well as intrinsic conduction). At room temperature, conduction is mostly by the extrinsic dopant, but as the temperature is increased, the intrinsic conduction mechanism dominates. The effect of the dopant is to greatly reduce the band gap to 0.05 eV. Four Probe Resistivity Technique A four probe experimental technique was used to measure the band gap. This involves four small sharp probes in a straight line placed on a flat surface of the sample to be measured, a constant current is passed between the outer contacts this flows in a certain pattern in the body of the material, and the resulting floating potential between the inner contacts is measured. The resistivity is calculated from Ohms Law and the geometry of the sample. If the temperature is increased, more electrons will be excited into the conduction band so the conduction will increase, therefore the resistivity will decrease. Since the resistivity is highly dependant on the temperature, by varying the temperature while taking measurements, and hence are able to find the resistivity as a function of temperature. The motivation for using the four probe technique is that many conventional methods for measuring resistivity such as soldering contacts to the sample are unsatisfactory for semiconductors. Soldering directly onto the sample can disrupt the properties of the sample through heating and carrier injection. Metal-semiconductor contacts are usually rectifying in nature and there tends to be minority carrier injection by one of the current carrying contacts. An excess concentration of minority carriers will affect the potential of other contacts and modulate the resistance of the material, hence giving an unsatisfactory result. The four probe Department of Physics 2 University of Kelaniya

Four Probe Method PS/2004/225 technique overcomes these faults by using pressure contacts but a draw back is they tend to be noisy. The four probe method also allows measurements to be made on a smaller area; hence there will be less variation in the properties of the crystal over the small area. Another benefit over a soldered contact based technique is the precision at which the probe spacing can be measured. In order to use the four probe technique certain assumptions need to be made: 1. The resistivity of the sample is uniform in the area of measurement. 2. Any minority carriers injected into the sample from the current-carrying electrodes recombine near the electrodes so that their effect of the conductivity is negligible. 3. The surface on which the probes rest is flat. 4. The four probes lie in a straight line. 5. The diameter of the contact between the probes and the sample is small compared to the distance between the probes. 6. The boundary between the current carrying probes and the sample is hemispherical or small in diameter. It is also assumed that the current through the sample is small enough not to cause heating, and that the voltage drop at the contacts is low so that injection is minimized (this is somewhat recovered in assumption 2 above).

Department of Physics

University of Kelaniya

Four Probe Method

PS/2004/225

APPARATUS
Hot plate Thermometer Four Probe Apparatus (Constant Current generator and Digital Voltmeter) Ge sample Connecting wires

Department of Physics

University of Kelaniya

Four Probe Method

PS/2004/225

THEORY
For the current density J in a body under the influence of an electric field E Ohms law states; J = .E The proportionality factor is called electric conductivity. Since this quantity strongly depends on the material, it is common to classify materials with regard to their conductivity. Semiconductors, for example, are solids that do not conduct electric currents at low temperatures, but show a measurable conductivity at higher temperatures. The reason for this temperature dependence is the specific band structure of the electronic energy levels of a semiconductor. The valence band, i.e. the highest band that is completely or partially populated in the ground state, and the conduction band, i.e. the next unpopulated band, are separated by a band gap Eg (Ge: Eg 0.7 eV). The region between the two bands is not populated by electrons in an undoped, pure semiconductor and is called the forbidden zone. At higher temperatures, more and more electrons are thermally activated from the valence band into the conduction band. They leave holes in the valence band which move like positive charged particles, thus contributing to the current density J as do the electrons.

The conduction which is made possible by the excitation of the electrons from the valence band into the conduction band is called intrinsic conduction. Since under conditions of thermal equilibrium the number of holes in the valence band and of electrons in the conduction band are equal, the current density in the case of intrinsic conduction can be written in the form; J = ( e).n.v n + e.n.v p e: elementary charge, n: concentration of electrons or holes respectively. vp: average drift velocity of the electrons. vn: average drift velocity of the holes. The average drift velocities vn and vp of the electrons and the holes are proportional to the field strength E. Then; v n = n .E and v p = p .E Where the mobility n and p are chosen to be positive quantities,
J = e.n.( n + p ).E

By comparing this equation with the equation of Ohms law, The conductivity, = e.n.( n + p ) Department of Physics 5 University of Kelaniya

Four Probe Method PS/2004/225 Apart from the elementary charge e, all quantities in the above equation depend on the temperature T. The concentration of intrinsic conduction n is given by;

n = ( N .P ) k: Boltzmann constant, Eg: band gap of the semiconductor. And N and P are the effective state densities in the conduction band and in the valence band.
2 .m p .kT 2 2 .mn .kT 2 And P = 2 . N = 2. h2 h2 h: Planck constant, mn: effective electron mass, mp: effective hole mass. The mobility n and p also depends on the temperature. At low temperatures, the proportionality T 2 holds roughly, as does the proportionality T 2 at high temperatures. Because of the predominance of the exponential function, conductivity is well approximated and represented by; Eg Eg = 0 e 2.kT ln = ln 0 2.kT The resistivity can be written as inverse of the conductivity. Then, 1 The resistivity, = By substituting this in the above equation, Eg ln = + ln 0 2.kT With the object of confirming above equation and determining the band gap Eg, the resistivity of undoped germanium is determined as a function of the temperature T in the experiment. At a constant current; I = J .b.c Where b: breadth of the crystal, c: thickness of the crystal. V = E.a And voltage drop; a: length of the crystal. By combining above equations the conductivity can be written as; a I b.c V = . = . b.c V a I Then it can rearrange the equation as the graph of ln vs 1\T as follows; Eg 1 ln = 2.k . T + ln 0
3 3
3

Eg 1 2.kT 2 .e

Y = m. X + C Department of Physics

University of Kelaniya

Four Probe Method

PS/2004/225

EXPERIMENTAL PROCEDURE
Digital Voltmeter & Constant current generator

Probe equipment with the Ge sample Oven

1. The probe equipment with Ge sample was placed on the hot plate and then the thermometer was kept into its holder. (In this set up, the hot plate was used instead of the oven because it was not worked as well.) 2. The electric connections were made, and a current of 3.0 mA was set. 3. The hot plate was turned on and readings of the voltage were taken until the temperature reached 140 0C, the hot plate was then turned off and more readings were taken as the sample cooled down to approximately 70 0C. (At the low temperatures the semiconductor resistivity was rapidly decreased. Therefore to get the linear variation of ln vs 1/T, the slightly high temperatures were selected to plot the graphs). 4. The same procedure was repeated for the currents 5.0, 7.0, mA. 5. The graphs of ln vs 1/T were plotted to calculate the band gap.

Department of Physics

University of Kelaniya

Four Probe Method

PS/2004/225

OBSERVATIONS
I = 310-3 A Vi Vd (Increasing)/V (Decreasing)/V 0.232 0.231 0.218 0.213 0.200 0.193 0.181 0.170 0.160 0.149 0.142 0.129 0.123 0.112 0.106 0.094 0.094 0.082 0.082 0.070 0.071 0.060 0.054 0.051 0.048 0.044 0.042 0.039 0.037 0.033 0.033 0.029 0.029 0.026 0.026 0.023 0.024 0.020 0.021 0.018 0.018 0.017 0.016 0.016 I = 510-3 A Vi Vd (Increasing)/V (Decreasing)/V 0.382 0.382 0.350 0.350 0.313 0.308 0.277 0.269 0.243 0.231 0.211 0.198 0.178 0.171 0.150 0.148 0.129 0.129 0.110 0.109 0.094 0.092 0.080 0.079 0.068 0.066 0.058 0.057 0.050 0.048 0.043 0.041 0.038 0.036 0.033 0.032 0.029 0.028 0.025 0.025 0.021 0.020 0.018 0.018

T/K 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140

T/K 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140

Department of Physics

University of Kelaniya

Four Probe Method

PS/2004/225

Crystal Dimensions Length = 1010-3 m Width = 1010-3 m Thickness = 0.510-3 m The space between two probes = 210-3 m Room temperature: 30 0C Uncertainties Voltage: 0.310-3 V Temperature: 0.5 0C

Department of Physics

University of Kelaniya

Four Probe Method

PS/2004/225

RESULTS
T (0C) 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 T (K) 343 348 353 358 363 368 373 378 383 388 393 398 403 408 413 I = 310-3A 1/T (K-1) 0.002915 0.002874 0.002833 0.002793 0.002755 0.002717 0.002681 0.002646 0.002611 0.002577 0.002545 0.002513 0.002481 0.002451 0.002421 Vi (V) 0.106 0.094 0.082 0.071 0.054 0.048 0.042 0.037 0.033 0.029 0.026 0.024 0.021 0.018 0.016 Vd (V) 0.094 0.082 0.070 0.060 0.051 0.044 0.039 0.033 0.029 0.026 0.023 0.020 0.018 0.017 0.016 V (V) 0.1000 0.0880 0.0760 0.0655 0.0525 0.0460 0.0405 0.0350 0.0310 0.0275 0.0245 0.0220 0.0195 0.0175 0.0160 0.01667 0.01467 0.01267 0.01092 0.00875 0.00767 0.00675 0.00583 0.00517 0.00458 0.00408 0.00367 0.00325 0.00292 0.00267 ln -4.09434 -4.22218 -4.36878 -4.51746 -4.73870 -4.87087 -4.99821 -5.14417 -5.26553 -5.38533 -5.50084 -5.60847 -5.72910 -5.83731 -5.92693

Eg 1 ln = 2.k . T + ln 0

Department of Physics

10

University of Kelaniya

Four Probe Method According to the graph, (By using the Origin 6.0 software) The gradient of the graph, m1, = 3786.7795 K Eg According to the equation, m1 = 2k Eg Therefore, m1 = = 3786.7795 2k Where, k = Boltzmann constant =1.3810-23JK-1 Therefore, E g = 3786.7795 2 1.38 10 23 J E g = 1.0452 10 19 J 1.0452 10 19 eV 1.602 10 19 E g = 0.65eV Eg =

PS/2004/225

T (0C) 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140

T (K) 343 348 353 358 363 368 373 378 383 388 393 398 403 408 413

I = 510-3A 1/T (K-1) 0.002915 0.002874 0.002833 0.002793 0.002755 0.002717 0.002681 0.002646 0.002611 0.002577 0.002545 0.002513 0.002481 0.002451 0.002421 Vi (V) 0.150 0.129 0.110 0.094 0.080 0.068 0.058 0.050 0.043 0.038 0.033 0.029 0.025 0.021 0.018 Vd (V) 0.148 0.129 0.109 0.092 0.079 0.066 0.057 0.048 0.041 0.036 0.032 0.028 0.025 0.020 0.018 V (V) 0.149 0.129 0.1095 0.093 0.0795 0.067 0.0575 0.049 0.042 0.037 0.0325 0.0285 0.025 0.0205 0.018 0.02483 0.02150 0.01825 0.01550 0.01325 0.01117 0.00958 0.00817 0.00700 0.00617 0.00542 0.00475 0.00417 0.00342 0.00300 ln -3.69557 -3.83970 -4.00359 -4.16692 -4.32376 -4.49482 -4.64773 -4.80769 -4.96185 -5.08860 -5.21827 -5.34961 -5.48064 -5.67909 -5.80914

Department of Physics

11

University of Kelaniya

Four Probe Method

PS/2004/225

Eg 1 ln = 2.k . T + ln 0

According to the graph, (By using the Origin 6.0 software) The gradient of the graph, m2, = 4269.6883 K Eg According to the equation, m 2 = 2k Eg Therefore, m2 = = 4269.6883 2k Where, k = Boltzmann constant =1.3810-23JK-1 Therefore, E g = 4269.6883 2 1.38 10 23 J E g = 1.1784 10 19 J 1.1784 10 19 eV 1.602 10 19 E g = 0.74eV Eg =

Department of Physics

12

University of Kelaniya

Four Probe Method

PS/2004/225

T (0C) 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140

T (K) 343 348 353 358 363 368 373 378 383 388 393 398 403 408 413

I = 710-3A 1/T (K-1) 0.002915 0.002874 0.002833 0.002793 0.002755 0.002717 0.002681 0.002646 0.002611 0.002577 0.002545 0.002513 0.002481 0.002451 0.002421 Vi (V) 0.198 0.167 0.140 0.120 0.103 0.089 0.077 0.066 0.057 0.050 0.043 0.038 0.033 0.029 0.026 Vd (V) 0.178 0.153 0.130 0.111 0.095 0.082 0.071 0.062 0.053 0.047 0.040 0.036 0.032 0.028 0.025 V (V) 0.188 0.16 0.135 0.1155 0.099 0.0855 0.074 0.064 0.055 0.0485 0.0415 0.037 0.0325 0.0285 0.0255 0.03133 0.02667 0.02250 0.01925 0.01650 0.01425 0.01233 0.01067 0.00917 0.00808 0.00692 0.00617 0.00542 0.00475 0.00425 ln -3.46307 -3.62434 -3.79424 -3.95024 -4.10439 -4.25100 -4.39545 -4.54063 -4.69218 -4.81795 -4.97382 -5.08860 -5.21827 -5.34961 -5.46084

Eg 1 ln = . T + ln 0 2.k

Department of Physics

13

University of Kelaniya

Four Probe Method According to the graph, (By using the Origin 6.0 software) The gradient of the graph, m3, = 4062.1123 K Eg According to the equation, m3 = 2k Eg Therefore, m3 = = 4062.1123 2k Where, k = Boltzmann constant =1.3810-23JK-1 Therefore, E g = 4062.1123 2 1.38 10 23 J E g = 1.1211 10 19 J 1.1211 10 19 eV 1.602 10 19 E g = 0.70eV Eg = The average value of the band gap = The band gap of the Ge is 0.697 eV.
0.65 + 0.74 + 0.70 eV = 0.697eV 3

PS/2004/225

Department of Physics

14

University of Kelaniya

Four Probe Method

PS/2004/225

ERROR ANALYSIS
The band gap, E g = 2k .m
ln E g = ln (2k ) + ln m

By taking partial differentiation,

E g
Eg

m
m

m
m

Eg

For I = 310-3A m E g = 1 E g m1 By the statistical method, (by using Origin 6.0 software), m1 = 45.2263 K 45.2263 E g = 0.65eV 3786.7795
E g = 0.008eV

For I = 510-3A m2 E g = Eg m2 By the statistical method, (by using Origin 6.0 software), m2 = 41.5528 K 41.5528 E g = 0.74eV 4269.6883
E g = 0.006eV

For I = 710-3A m E g = 3 E g m3 By the statistical method, (by using Origin 6.0 software), m3 = 12.7484 K 12.7484 E g = 0.70eV 4062.1123
E g = 0.002eV

The average value of the error of the band gap =

0.008 + 0.006 + 0.002 = 0.006eV 0.01eV 3

Department of Physics

15

University of Kelaniya

Four Probe Method

PS/2004/225

CONCLUSION
Experimental value The band gap of the Ge is 0.697 0.006 eV. Standard value The band gap of the Ge is 0.67 eV.

Department of Physics

16

University of Kelaniya

Four Probe Method

PS/2004/225

DISCUSSION
The band gap of Ge was measured as 0.700.01 eV using a resistivity method. A four probe experimental technique was used to avoid the use of soldering contacts to the sample. This method has its benefits but also makes assumptions about the characteristics of the sample. The accepted band gap for Ge is 0.66 eV, the discrepancy between the measured value and the accepted value could be explained by the assumptions of the four probe method, the difference in voltage readings for cooling and heating stages of the experiment, and/or the doping effect of the sample. The impurities in the Ge sample would produce an extrinsic component of the conductivity, and although the band gap was calculated from measurements when the temperature is high and the intrinsic component dominates; the extrinsic component would still have a residual effect. During the cooling phase of the experiment the measured values of resistivity were slightly lower than in the heating phase, this is due to the thermometer and/or sample cooling at difference rates. The observed difference can be explained if the thermometer is cooling quicker than the sample. A digital thermometer would have produced more accurate results. At absolute zero temperature there would be no thermal excitation of the electrons from the valence band to the conduction band; the valence band would be completely full and the conduction band completely empty, hence there would be no conduction of electrons and the semiconductor would behave as an insulator. In this experiment it would be important for the millivoltmeter to have high impedance. There is a contact resistance in the semiconducting crustal. This is a serious matter for electrical semiconductors and to avoid these errors they used the extra probes between the current contacts. It should not be exceed the temperature over the 1800C then the device can be destroyed. Also for the accurate value the probe spacing should be greater than the thickness of the semiconductor device. In this experiment, the thermo plate was used to heat the semiconductor instead of Oven of four probe apparatus. (It was destroyed by high current). It was not so accurate for the experiment because we cannot keep at constant temperature in longer time like in the oven. This experimental technique has important limitations and assumptions; we assume that the resistivity of the material is uniform in the area of measurement, there is minimum carrier injection from the probes into the material, and a good contact is made with the surface.

Department of Physics

17

University of Kelaniya

Four Probe Method

PS/2004/225

REFERENCES
C. Kittel, Introduction to Solid State Physics, (Willey, New York). www.sestechno.com/pro1/2c.pdf www.cce.ufes.br/jair/web/gap-Ge.pdf LEYBOLD Physics Leaflets P7.2.1.5

Department of Physics

18

University of Kelaniya