Department of Electrical & Electronic Engineering ORT Braude College Advanced Laboratory for Characterization of Semiconductor Devices - 31820 Semiconductor diode March 24, 2016 Dr. Radu Florescu Dr. Vladislav Shteeman Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 The goal. The goal of this experiment is to measure the Current-Voltage (I-V) and Capacitance-Voltage (C-V) characteristics of semiconductor diode (at room temperature and under heating conditions) and extract from the measurements basic physical parameters / characteristics of the device. You will use Keithley SCS 4200 measurement system and Agilent 4284A C-R-L analyzer. The following parameters will be acquired from the I-V measurements: Saturation current I sat Parasitic series resistance Rseries Parasitic shunt resistance Rshunt The following parameters will be acquired from the C-V measurements: Built-in voltage of pn-junction, V0 Doping densities N A and N D Depletion layer width W and depletion layer widths x p , x n on the p- and n-sides of the junction unit unit Electric charges (per unit area) on the p- and n-sides of the junction, Q area and Q area . Dr. Radu Florescu Dr. Vladislav Shteeman 2 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Short theoretical background. Diode[1] is a two-terminal electronic component that conducts electric current in one direction (from positive (p) to negative (n) terminal) and blocks it the opposite direction. Semiconductor diodes (represented physically by a p-n junction) are most common type of contemporary diodes. They are widely used as a rectifiers, stabilitrons, voltage-dependent capacitors etc. Special modifications of semiconductor diodes are used as solar cells, photodiodes, light emitting diodes (LEDs) and laser diodes. p-n junction consists of two regions (in a single semiconductor crystal) with opposite doping type (Figure 1). The region on the left is p-type with an acceptor density N A , while the region on the right is n-type with a donor density N D . The electrons (holes) density in the n-type (p-type) region is approximately equal to the donor (acceptor) density (i.e. p0 N A , n0 N D ). Figure 1. Sketch of p-n junction (after [12]). I. Electrostatics of pn – junction. 1. pn – junction at zero-bias (applied voltage VD = 0) (after [5]) To reach the thermal equilibrium, electrons/holes close to the pn-junction diffuse across the junction into the p-type/n-type region where hardly any electrons/holes are present. This process leaves the ionized donors (acceptors) behind, creating a region around the junction, which is depleted of mobile carriers. This region is called depletion region, extending from x x p 0 to x xn 0 . The charge due to the ionized donors and acceptors creates built–in difference in potentials between the two sides of the pn-junction, qV0 (where V0 is a built-in voltage). This built-in potential qV0 is also expressed by the existence of the built-in electric field, which in turn causes a drift of carriers in the opposite directions. The diffusion of carriers continues until the drift current balances the diffusion current, thereby reaching thermal equilibrium (zero total current) as indicated by a constant Fermi energy. This situation is shown on Figure 2 and Figure 3: Dr. Radu Florescu Dr. Vladislav Shteeman 3 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Figure 2. Concentration of carriers, electric charge density, electric field and electric potential distribution in pn-junction at zero bias (after [1]). Figure 3. Energy band diagram of a p-n junction at zero bias (after [1]). Dr. Radu Florescu Dr. Vladislav Shteeman 4 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 The built-in voltage V0 can be expressed in terms of concentrations of donors N D , acceptors N A and intrinsic concentration ni : V0 kT N A N D ln q ni2 (where 0.0256 ln at room temperature T 300K V N AND ni2 (1) k 8.61733 10 5 eV K is Boltzmann constant, T K is a temperature and q 1.6 10 19 C is an electron charge) Typical value of V0 for a standard Si diode (for N D 1015 cm 3 and N A 1018 cm 3 ): V0 0.7 V . Capacitance of a pn–junction at zero bias, C j 0 , originates mainly from the stationary ions in the depletion region (i.e. from so-called junction capacitance C j at VD 0 ). It is expressed via the total width of the depletion layer, W0 , and the relative dielectric constant of the semiconductor, r , as follows: C j0 0 r W0 A (2) (where A is a cross-section area of the pn-junction and 0 is the vacuum permittivity) 2. Biased pn-junction (applied voltage VD ≠ 0). Depending on the applied external voltage, VD , there are two modes of diode biasing: forward biased conducting mode and reverse biased non-conducting mode. Band diagrams of reverse- and forward- biased pn-junction are shown on Figure 4 and Figure 5. Dr. Radu Florescu Dr. Vladislav Shteeman 5 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Figure 4. Band diagram of a p-n junction under reverse bias (after [4]). Figure 5. Band diagram of a p-n junction under reverse bias (after [4]). Dr. Radu Florescu Dr. Vladislav Shteeman 6 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Depletion layer width W. The total width of the depletion layer W x p xn for biased pnjunction is voltage-dependent. Figure 6. Depletion layer in biased pn-junction. 2 V V 1 1 r 0 0 D q ND N A W 2 V V 1 1 r 0 0 D q N N A D VD 0 (3) VD 0 The width of the depletion layers on the p- and n- sides of the junction, x p and x n : xn W , ND 1 NA xp W N 1 A ND (4) Capacitance of ideal biased pn – junction. A total capacitance of biased pn-junction is a sum of 2 different capacitances, connected in parallel, namely: Dr. Radu Florescu Dr. Vladislav Shteeman 7 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 1. C j (junction capacitance) - capacitance of stationary ions in the depletion layer (negatively charged acceptor ions at the p-side vs positively charged donor ions at the n-side). This capacitance is dominant at the reverse bias VD 0 . 2. CS (stored or diffusion capacitance) capacitance of mobile carriers (electrons at the p-side vs holes at the n-side), stored (as the result of diffusion) outside the depletion layer. This capacitance is dominant at the forward bias VD 0 and not exists at the reverse bias VD 0 (since for VD 0 , there is no excess carriers outside the depletion layer). Thus a total capacitance of diode is C j for reverse bias VD 0 C j C S for forward bias VD 0 For reverse bias, C j (and therefore a total capacitance of a diode) depends on VD as follows: Cj A 2 2q r 0 N A N D A r 0 W V0 VD N A N D (5) (where A is a junction’ cross-section area) At the forward bias, a total capacitance of diode is C j C S , not C j , and therefore (with increasing VD ) it rises faster than measured 1 C 2 VD Dr. Radu Florescu 1 V0 VD , as expected from Eq. (5). For this reason, characteristics of a real diode experience drastic decrease for VD 0 (see ). Dr. Vladislav Shteeman 8 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Figure 7. Measured total capacitance C VD and C 2 VD of 1N4007 diode. Calculation of built-in voltage, V0 from C-V measurements. From Eq. (5), 1 C 2j is a linear function of VD : 1 2 N A ND V0 VD 2 2 C j q 0 semiconductor A N A N D Dr. Radu Florescu Dr. Vladislav Shteeman (6) 9 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Figure 8. Sketch of a total capacitance C VD and C 2 VD for an ideal pn – junction (after [4]). From the latter equation, for VD V0 1 0 (i.e. C j ). This allows one to find the builtC 2j in voltage V0 from the C-V measurements, as shown on Figure 8. V0 is obtained at the intersection of the 1 C 2j curve and VD axis. Calculation of dopant concentrations N A and N D from C-V measurements. N A and N D can be obtained from the slope of the graph 1 C 2j vs VD (see e.g. Figure 8) and Eq. (1). The slope d C j 2 VD is given by: dVD d C j 2 VD dVD NA ND q 0 r A2 N A N D 2 (7) Therefore, q 0 r A2 2 d C 2 j VD NA ND dVD N AN D (8) From Eq. (1), at the room temperature (i.e. kT q 0.0256 V ), for the V0 and intrinsic concentration ni known, we get another expression for N A and N D : Dr. Radu Florescu Dr. Vladislav Shteeman 10 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 V0 N A N D ni2 exp 0.0256 V (9) From Eq. (8) and Eq. (9) one can find the 2 unknowns, N A and N D . Electric charge over the junction for reverse bias. The total electric charge over the junction (for reverse biase) is given by: Figure 9. Sketch of a biased pn-junction and a static charge distribution over it. Q qAxn N D Q qAx p N A n side charge p side charge (10) where A is a junction’ cross-section area. Electric charges per unit area on both sides of the junction are: unit area unit area n side charge p side charge Q qxn N D Q qx p N A Dr. Radu Florescu Dr. Vladislav Shteeman (11) 11 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 II. I-V characteristics of diode. 1. I-V characteristics of an ideal pn – junction. Current via an ideal pn-junction obeys Shockley equation: kTVD q I D I sat e 1 (12) where I sat is a saturation current, q 1.6 10 19 C is an electron charge, k is Boltzmann constant and T is temperature. Figure 10 shows a sketch of I-V characteristics of an ideal diode. Figure 10: I-V characteristics of an ideal diode (after [6]). Asymptotic behavior of Shockley equation. Note that e 1 e for 5 and e 0 for 5 . In our case, the condition 5 corresponds (at the room temperature, for kT 0.026 V ) to the external voltageVD 0.13 V , q while 5 corresponds to the VD 0.13 V . Therefore, for the forward biased junction with VD 0.13 V , Dr. Radu Florescu Dr. Vladislav Shteeman 12 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 I D I sate VD kT q VD 0.13 V (13) while for the reverse biased junction with VD 0.13 V VD 0.13 V I D I sat (14) In other words, for reverse biased ideal pn-junction, current I D quickly reaches its saturation value I sat and remains constant for all the range of voltages VD 0.13 V . As opposite to the former, the forward bias, current increases exponentially all along; there is no current saturation. Typical values of the saturation current are: I sat 109 A for silicon diodes I sat 106 A for germanium diodes. 2. Deviations of real diodes from Shockley model. Deviations under forward bias. Parasitic series resistance Rseries . When (at forward bias) VD reaches V0 (i.e. VD V0 ), the potential barrier between p- and n- sides of diode disappears: qV0 VD 0 . Thus, I D cannot continue rise exponentially, as before (as in Eq. (13)). In order to describe a slower rising of I D for VD V0 , a model of parasitic series resistance Rseries is introduced. (This is so-called “second-order” I-V analysis of diode, extending ideal Shockley model to the case, where it is actually inapplicable (i.e. to VD V0 ).) According to the model, for VD V0 a diode should suffer from a parasitic series resistance Rseries (see Figure 11). This resistance reduces the voltage drop over the junction from VD to VD I D Rseries . Thus, Eq. (13) should be re-written as: I D I sate V D I D R series kT q Dr. Radu Florescu (15) Dr. Vladislav Shteeman 13 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Figure 11. Sketch of I-V characteristics of "ideal" and "real" pn – junction (after [7]). Transformation of this equation gives: ln ID V I R D D series I sat kT q kT q (16) This allows us to express Rseries as follows: Rseries 1 ID kT I D ln VD I sat q (17) Thus, Rseries (according to the model) should be a linear function of VD . Typical values of Rseries for Si diode are: Rseries 5 50 . Deviations under reverse bias. Parasitic shunt resistance Rshunt . Experimentally, at reverse bias, for VD 0.13 V , I D does not saturate to I sat but rather continues to increase (in abs value, see Figure 12). Dr. Radu Florescu Dr. Vladislav Shteeman 14 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Figure 12. I-V characteristics of reverse-biased 1N4148 diode. This increase is due to the generation current (drift current) I gen qAWni g , originating from thermal generation of electron-hole pairs in the volume of depletion layer (where A and W are cross-section of width of depletion layer of pn-junction, ni is intrinsic concentration and g is a lifetime of electron-hole pair from thermal generation to recombination). In a good approximation, I D increases linearly with VD (see e.g. I D VD graph on Figure 12 for 1 VD 0.3 ). One can modify Shockley model to account for I gen by introducing a parasitic shunt resistance Rshunt , connected in parallel to the ideal diode and assuming I gen VD Rshunt (see Figure 13). For VD 0.13 V , Shockley equation should be replaced with: I D I sat I gen I sat Dr. Radu Florescu qAWni g Dr. Vladislav Shteeman I sat VD Rshunt (18) 15 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Figure 13. Explanation to the model of parasitic shunt resistance Rshunt and saturation current I sat for reverse biased diode. Note that Rshunt is a single value, independent on VD . 2 3 Typical values of Rshunt for a Si diode are: Rshunt 10 10 M . One can compute both the Rshunt and I sat of a diode using trendline of the I D VD curve in the “deep” reverse bias region (say, in the range 1 VD 0.3 ). Rshunt 1 of this trendline, slope while I sat free term (see Figure 13). 3. Diode response vs. temperature. Diode performance is strongly affected by the temperature factor. Consider again the Shockley model, I sat in Eq. (12) has a form: I sat (n) De( p ) 2 Dhole ni qA ( p ) (n) Le N A Lhole N D (19) (n ) where: q is an electron charge, De( p ) and Dhole are the diffusion coefficients of the electrons on the p-side and the holes on the n-side, L(ep ) and L(nhole) are the diffusion lengths of the electrons on the Dr. Radu Florescu Dr. Vladislav Shteeman 16 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 p-side and the holes on the n-side, A is the pn-junction cross-section area, N A , N D are the donors and acceptors concentrations and ni is the intrinsic concentration of Si. (n ) For simplicity, assume, that De( p ) , Dhole , L(ep ) and L(nhole) are temperature independent. Nevertheless, the current I sat in Eq. (19) does depend on the temperature, since n i is strongly dependent on the temperature. It is possible to show, that I sat increases approximately for 7% for each temperature degree. Thus, for two different temperatures T1 and T2 holds: I sat T2 I sat T1 1.07T2 T1 (20) 10 Since 1.07 2 , there is a mnemonic rule, which says, that I sat doubles itself each 10 degrees: I sat T2 I sat T1 2 T2 T1 10 (21) The higher the temperature, the faster grows the forward branch of the I-V characteristics and the larger the absolute value of the breakdown voltage at the reverse bias. Figure 14 illustrates the aforesaid. Figure 14. Diode response vs. temperature. Dr. Radu Florescu Dr. Vladislav Shteeman 17 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Assignments and analysis. Part I. Room temperature measurements. 1. Acquire I-V measurements of diode using Keithley measurement system. (See Appendix 2 for details about pin connections and Keithley program parameters.) 2. Acquire C-V measurements of diode using Agilent R-C-L meter and Keithley measurement system. (See Appendix 3 for details about pin connections and Keithley program parameters.) Note: after executing the measurements and before processing the acquired data, save this Excel template on your computer (double click on the Excel icon File Save as … ). Diode measurements empty.xlsx Then close the Excel template and open the Excel file, saved recently. Copy the results of the measurements (located in the measurements folder of Keithley in the subdirectory “tests/data”) to the Excel template, saved on your computer. Evaluation of physical parameters of diode from I-V measurements. 1. Shunt resistance Rshunt : I-V characteristics of diode in the “deep” reverse bias (e.g. 0.5 V VD 2 V ) in a good approximation is a straight line. 1 1 of the trendline of section is Rshunt : Rshunt slope slope 2. Saturation current I sat : I D I sat I gen I sat VD Rshunt Free term of the trendline of the section above is I sat : I sat free term Dr. Radu Florescu Dr. Vladislav Shteeman 18 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 3. Series resistance Rseries : for forward bias region VD 0.2 V , from the vs V and Eq. (17), it is possible to find the range of graph ln I D D I sat VD , corresponding to the ideal diode (red line) and another range, corresponding to the diode with series resistance (blue line). It is also possible to find the Rseries itself. To do this: in Excel, fill in the column ln I D I sat vs V ; on the graph, find (visually) the sections, corresponding to the make a graph ln I D D I sat red and to the blue lines. vs V for the “blue” range only. add to the graph plot ln I D D I sat add to the graph trend line for the “blue” range points (with “show equation” and “show R square” boxes marked) to get the linear equation coefficients. (Note that in our case ln I D is the “y” axis and VD is the “x” axis). I sat fill in a column, “ Rseries ”, in the datasheet using Eq. (17) make 2 additional plots: Rseries vs VD and Rseries vs I D . Evaluation of physical parameters of the diode from C-V measurements. 4. Built-in junction potential, V0 . This potential should be obtained at the intersection of the C j 2 curve and the horizontal axis ( VD voltage). In order to get V0 : in Excel, build the graph C j 2 vs VD . add to the graph trend line with the option “show equation” marked. from the trend line equation, find V0 (note that in our case 1 C 2j is the “y” axis and VD is the “x” axis). Dr. Radu Florescu Dr. Vladislav Shteeman 19 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 5. Dopants’ concentration N A and N D . If the slope d C j 2 dVD , the built-in voltage V0 , the intrinsic carriers’ concentration ni and the cross section area of the pn- junction A are known ( see List of symbols in Appendix 1 and Appendix 3 for details)) – one can find N A and N D by solving Eq. (8) and Eq. (9): q A2 d C j 2 N N D A 0 r 2 N AN D dVD 2 V0 n exp i 0.0256 V N A N D the built-in voltage V0 and the slope d C j 2 dVD were found on the previous step. r and ni for the specific diode could be found elsewhere (see e.g. Appendix 1). 2 2 pn-junction’ cross-section area, A : for the 1N4148 Si diode, assume A 1mm 0.01cm . 6. Depletion layer width W vs VD for the reverse bias VD 0 . in Excel, compute (fill in the corresponding columns) depletion layer width W and depletion layers width at the p- and n- sides ( x p and x n ) for the reverse bias VD 0 (see Eq. (3) and Eq. (4) ). in Excel, make on a single graph 3 plots: W vs VD , x p vs VD and xn vs VD . 7. Electric charge (per unit area) over the junction for the reverse bias VD 0 . in Excel, compute (fill in the corresponding columns), using Eq. (11), electric charges per unit unit unit area on both sides of the junction Q area and Q area for reverse bias VD 0 . unit unit in Excel, make on a single graph 2 plots: Qarea vs VD and Qarea vs VD . Dr. Radu Florescu Dr. Vladislav Shteeman 20 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Part II. Diode measurements under heating conditions. For a room temperature and 4-5 other temperatures in a range of 40 C 70 C , acquire the I-V measurements of the diode using the Keithley measurement system. (See Appendix 2 for details about the pin connections and the Keithley program parameters.) Important: execute the 1st measurement (under room temperature) using the green button , and all the rest of the measurements (under heating conditions, for 4-6 different temperatures) using the yellow-greed button (“append”). DO NOT use the green button for the measurements under heating conditions: it will override all your previous measurements.) After each measurement save the data in the Keithley program. After finishing the measurements, you can automatically process the data using Matlab program. To do this: 1. Double-click on the zip-file “Diode heating processing.zip”. In the newly opened Diode heating - processing.zip window, folder “Diode heating processing” will appear. 2. Drag this folder to the Desktop of your computer. 3. In the Keithley measurements connect_pin_agilent#1@1.xls folder ( C VD & C 2 (subfolder VD tests/data) find the files data), IU_forward#1@1.xls ( I D VD measurements – forward bias data) & IU_backward#1@1.xls ( I D VD measurements – reverse bias data) Dr. Radu Florescu Dr. Vladislav Shteeman 21 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 4. Copy the files connect_pin_agilent#1@1.xls, IU_forward#1@1.xls & IU_backward#1@1.xls to the folder “Diode heating processing.zip”: 5. Double-click on the file processed_data_DIODE_sf.m in the directory “Diode heating processing”. This will start Matlab. Wait for 1-2 minutes to allow Matlab start. Dr. Radu Florescu Dr. Vladislav Shteeman 22 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 6. Go to the Matlab Editor window and run the file processed_data_DIODE_sf.m (press F5 or Debug Run on the Editor menu bar) 7. The program will ask you to input the temperatures, at which you measured the diode. Input the temperatures in the square parentheses with the spaces between the different values, e.g. [25 35 45 55 65]. Press Enter to continue. 8. Wait for approximately 1 minute, until the program will finish the processing of the measured data. 9. The results of the computations (Excel file processed_data.xls, Matlab files and figures) are located in the subfolder Results. Dr. Radu Florescu Dr. Vladislav Shteeman 23 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Dr. Radu Florescu Dr. Vladislav Shteeman 24 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Final Report content. Final Report must include the following diode parameters and graphs with explanations: Part I – room temperature measurements: [1] I-V characteristics of diode (2 graphs: I D VD for forward and for reverse bias) [2] C-V characteristics of diode (a single graph, including C VD & C 2 VD curves) [3] ln I D I sat vs VD graph [4] Shunt resistance Rshunt (a single value) [5] Saturation current I sat (a single value) [6] Series resistance Rseries (2 graphs: Rseries vs VD and Rseries vs I D ) [7] Built-in voltage V0 (a single value) [8] Dopant concentrations N A and N D (two single values) [9] Depletion layer width W vs VD and depletion layers widths at the n- and p-sides x p vs VD , x n vs VD (a single graph) unit [10]Electric charge per unit area Q area vs VD (a single value) Part II – measurements under heating conditions: [1] The graph of forward biased I V measurements for all the temperatures. (A single figure, 5 different curves for 5 different temperatures.) [2] The graph of backward biased I V measurements for all the temperatures. (A single figure, 5 different curves for 5 different temperatures.) Dr. Radu Florescu Dr. Vladislav Shteeman 25 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 [3] The graph of the saturation current as a function of temperature, I sat T . (A single graph, 5 points, corresponding to 5 different temperatures.) [4] The graph of the relative variation of the anode current versus anode voltage (forward bias) I 100% I T I Troom for different temperatures. I Troom [5] The graph of the relative variation of the anode current versus anode voltage (backward bias) I 100% I T I Troom for different temperatures. I Troom [6] The graph of the total width of the depletion layer W0 at zero bias as a function of temperature. (The total width of the depletion layer W0 can be evaluated from Eq. (2).) Dr. Radu Florescu Dr. Vladislav Shteeman 26 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Experimental set-up and samples to be studied The experimental setup includes Keithley matrix and Agilent L-C-R analyzer (Figure 17). (a) For the room temperature measurements, you will use Test fixture probe station (Figure 15) (with dual in-line package (18 pins) for diodes on Teaching chip No 3, or two-pins connection table for standard stand-alone diodes), connected by the triax cables No 9,10,11,12 to the Keithley switching matrix. (b) For the measurements under the heating conditions, you will use the temperature controlling oven (Figure 16) connected by the triax cables No 5,6,7,8 to the Keithley switching matrix. Monitor Keithley 708A Switching Matrix Figure 15. Test fixture probe station. Agilent 4284A LCR meter Keithley SCS 4200 I-V AND Parameter analyzer Figure 16. Temperature controlling oven. Figure 17. Keithley and Agilent L-C-R measurement setup. Table 1. Diode samples for available for study. Test chip No 3 Appendix 3 datasheet Dr. Radu Florescu 1N4148 diode 1N4001 diode 1N4148_1N4448 datasheet.pdf 1n4001.pdf Dr. Vladislav Shteeman 27 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Acknowledgement Electrical Engineering Department of Braude College would thank to Alexander Goldenberg, Vadim Goyhman, Adi Atias and Moran Efrony for their extensive help in preparation of this laboratory work. Several parts of this guide were adapted from the pn-junction manual of the Advanced Semiconductor Devices Lab (83-435) of School of Engineering of Bar-Ilan University. We would like to thank Dr. Abraham Chelly for the granted manual. Dr. Radu Florescu Dr. Vladislav Shteeman 28 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Appendix 1 : List of symbols and definitions. List of symbols. A - pn-junction’ cross-section area cm2 (See Appendix 3 for the details about the crosssection area of different diodes on the Teaching chip No 3). For the 1N4148 Si diode assume A 1mm2 0.01cm 2 . V0 - built-in voltage V . ni - intrinsic carriers concentration cm 3 . The following table presents ni for the basic semiconductors at the room temperature (T = 300 K): Si Ge GaAs 10 13 3 1.5 10 2.4 10 1.79 106 ni cm p0 , n0 - holes (electrons) concentration in the quasi-neutral regions of p- and n- sides of pn- junction cm 3 . N A , N D - acceptors (donors) concentration at the p- and n- sides of pn-junction cm 3 . Rseries - parasitic serial resistance [ ]; a voltage VD - controlled variable; usually ~ 5 50 . Rshunt - parasitic serial resistance [ ]; a voltage VD -independent constant; usually ~ 1 M . r - relative dielectric constant of a semiconductor [dimensionless]. The following table presents relative dielectric constants of the basic semiconductors: Si Ge GaAs 11.9 16 13.1 semiconductor 0 - permittivity of vacuum. 0 8.85 10-14 F cm 8.85 10-12 F m C j - pn-junction capacitance, originating from the stationary ion charges in the depletion region C j 0 - C j at zero bias VD 0 F F (n ) De( p ) and Dhole - diffusion coefficients of the electrons on the p-side and the holes on the n-side of the pn-junction. I D - diode current A. I gen - generation current A. Originates from thermal generation of electron-hole pairs in the volume of depletion layer: I gen qAWni g I sat - saturation current A. L(ep ) and L(nhole) - diffusion lengths of the electrons on the p-side and the holes on the n-side of the pn-junction cm . VD - voltage applied to pn-junction V . W - total depletion layer width cm . Dr. Radu Florescu Dr. Vladislav Shteeman 29 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 W0 - total depletion layer width at zero bias VD 0 cm . x p , x n - depletion layers width in the p- and n- regions cm . W x p x n . 8.617 10 5 eV - Boltzmann constant k 1.38 10 23 Joule deg .K deg .K T - temperature [deg. K] q 1.6 1019 C - electron charge. E F , Ei - Fermi and intrinsic Fermi level in semiconductor . E F , p , E F ,n - Fermi level on the p- (n-) side of the pn-junction. EC , EV - energy of the bottom of the conduction band and at the top of the valence band . Q , Q - electric charge (originating from the stationary ions in the depletion region) on the pand n- sides of the pn-junction C ; voltage VD -dependent. Qarea , Qarea - total electric charge per unit area and electric charges per unit area on the p- and unit unit n- sides of the pn- junction C cm 2 g - generation time. An average time between thermal generation and recombination of an electron-hole pair in the depletion layer. List of definitions n - high doping density of n-type ( N D 1019 cm 3 ). p - high doping density of p-type ( N A 1019 cm 3 ). Inversion - change of carrier type in a semiconductor obtained by applying an external voltage. Inversion layer - the layer of free carriers of opposite type at the semiconductor interface (layer of electrons in p-type semiconductor and layer of holes in n-type semiconductor). Dr. Radu Florescu Dr. Vladislav Shteeman 30 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Appendix 2 : Kite settings for I-V and C-V measurements. 1. pin connection scheme: SMU 1 (cable 9) LoPin (cable 12) SMU 2 (cable 10) HiPin (cable 11) I-V measurement s C-V measurement s 2. I-V Keithley settings Connect pins Dr. Radu Florescu Dr. Vladislav Shteeman 31 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 forward bias settings Expected results – forward bias Dr. Radu Florescu Dr. Vladislav Shteeman 32 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 backward bias settings Expected results – backward bias Dr. Radu Florescu Dr. Vladislav Shteeman 33 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 3. C-V Keithley settings frequency 1kHz Expected results – C-V measurements (10 kHz) Dr. Radu Florescu Dr. Vladislav Shteeman 34 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Appendix 3 : Teaching chip No 3 diodes specifications Diodes’ details and pins Dr. Radu Florescu Dr. Vladislav Shteeman Chip appearance 35 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Bibliography and internet links 1 Diode at wikipedia: http://en.wikipedia.org/wiki/Diode. 2 B. Streetman, S. Banerjee, “Solid state electronic devices” (6th edition), Prentice Hall, 2005. 3 К.В. Шалимова, «Физика полупроводников» (3е издание), Энергоатомиздат, 1985. 4 B. Van Zeghbroeck, “Principles of semiconductor devices”, Lectures – Colorado University, 2004. 5 A. Chelly, “pn-junction”, Lab manual - Advanced Semiconductor Devices Lab (83-435), School of Engineering of Bar-Ilan University. 6 J. Singh, “Semiconductor devices: basic principles”, Whiley, 2001. 7 A. del Alamo. “pn diode characterization” – project in the framework of course “Microelectronic Devices and Circuits” (6.012), MIT, 2003. 8 D. Neamen, “Semiconductor Physics and Devices: Basic Principles” (3rd edition), McGraw Hill, 2003. 9 S. Kasap, “pn-junction: the Shockley model”. An e-booklet (2001). 10 pn-junction Simulation using Java Applet: http://jas.eng.buffalo.edu/education/pn/iv/index.html 11 pn-junction properties calculator: http://www.ee.byu.edu/cleanroom/pn_junction.phtml 12 http://en.wikipedia.org/wiki/P%E2%80%93n_junction Silicon chip with PN junction 2 mm Dr. Radu Florescu Dr. Vladislav Shteeman 36 Department of Electrical and Electronic Engineering ORT Braude College of Engineering Advanced Laboratory for Characterization of Devices – 31820 Preparation Questions 1. 2. 3. 4. Explain (in short) the principle of diode operation Plot the qualitative graph of diode I-V characteristics Plot the qualitative graph of diode C-V characteristics How can you find from the I-V characteristics: a. saturation current I sat b. series resistance Rseries 5. How can you find from the C-V characteristics: built-in voltage of the pn-junction V0 doping densities N A and N D total depletion layer width W and depletion layer width x p , x n on each side of the junction Dr. Radu Florescu Dr. Vladislav Shteeman 37