PN Junction Surface Potential Images Measured by Kelvin Probe Force Microscopy
PO-CHIH SU, CHIH-MING HSIEH, and BING-YUE TSUI Department of Electronics Engineering & Institute of electronics National Chiao-Tung University 1001 Ta Hsueh Road, Hsinchu, 300 TAIWAN, R.O.C.
Abstract: - Kelvin probe force microscopy (KFM) was applied for two-dimensional pn junction surface potential imaging. It is demonstrated that the surface potential of semiconductor materials can be measured by controlling the voltage between the tip and the sample. Besides noise of the equipment, the sample surface treatment exhibit significant influence on the potential images. In addition, qualitative measurement of dopant concentration is also shown in this paper.
Key-Words: - Kelvin probe force microscopy, dopant concentrations, surface potential
1 Introduction and sample constant, we employ the tip oscillating at There has been an increasing demand for frequency ω1, near the resonant frequency ω0. By quantitative two-dimensional information of sensing the variation of ω1, the distance can be dopant profiling in submicron semiconductor controlled. devices. Conventional techniques, such as secondary ion mass spectroscopy (SIMS), have been used as a popular dopant profiling technique for a long time, but it lacks lateral resolution capabilities. Some attempts have been made to realize lateral dopant profiling by applying scanning probe microscopy (SPM) techniques [1]. However, resolution is limited by the required minimum capacitance. Kelvin probe force microscopy (KFM), which has potential to be a tool for two-dimensional (2D) dopant profiling with high resolutions, is another one possible Fig.1. Schematic diagram of Kelvin probe force methodology. KFM can image surface potential microscope system. by sensing the potential differences between the
probe and the sample surface. Since the work Furthermore, an ω2 frequency signal is still function of tip is known, the work function of applied between the tip and the sample for sensing surface under tip can be detected. In the case of the surface potential. Assuming that there are no
semiconductor materials, work functions depend electric charges and surface potential –V s on the on dopant type and dopant concentrations. For this sample surface, the electrostatic force is given by reason, it is possible to infer the dopant profile by measuring the surface potential [2]. 1 C 1 C F V 2 V V V sin t2 2 z 2 z dc s ac 2 2 Measurement Principle of KFM ,where Vdc (Vac) is the dc (ac) components of the Fig.1 is a schematic diagram of the KFM applied voltage between the tip and the sample, C is measurement system. The principle of KFM is to the effective capacitance, z is the distance between measure the electric interaction force by applying an them, and ω2 is the frequency of the applied ac ac voltage between a conductive tip and sample voltage. By expanding the terms, the force can be surface. In order to keep the distance between tip written by Fig.2. Energy band diagram of pn junction imaging. EF: Fermi level, Ep: vacuum level, eVp: work function in p-type region, eVn: work function in n-type region. (a) Simplified diagram of the connection. (b) Energy band diagram when no bias applied. (c) Over an n-type region. (d) Over a p-type region. [4]
1 C C V 2 A p-type silicon wafer was patterned by F (V V ) 2 ac typical photo-lithography technique, then As+ ions 2 z dc s z 4 2 were implanted at 20 KeV or 50 KeV to various C C Vac dosages. After implantation, photoresist was Vdc Vs Vac sin2t cos(22t) z z 4 removed, and a 250 nm thick SiO2 was deposited
Three components, Vdc, ω2, 2ω2, are in the force in a plasma-enhanced chemical vapor deposition system to passivate the sample surface. Samples equation. The 1ω2-component of the force depends were then annealed at 950 ℃ in N2 ambient for 30 on the dc voltage Vdc and surface potential –Vs; the min to activate the n-type dopants. The 2ω2-component depends only on the ac voltage Vac and the capacitance C between the tip and the passivation oxide was removed before KFM sample. Using a lock-in amplifier, it is easy to measurement. obtain the amplitude of the 1ω -component. During 2 3.2 Surface treatment KFM measurement, the Vdc must be modulated in order to nullify the output signal of the lock-in Since the KFM method detects the surface potential of sample, it is sensitive to surface amplifier. In other words, the relation Vdc-Vs=0 is always maintained by controlling V , and the condition. Therefore, the surface treatment before dc KFM measurement was studied at first. Initially, the surface potential –V s is just Vdc exactly [3]. Fig.2 shows a p-n junction measured by KFM sample was dipped into a 1% HF solution for about [4]. Fig.2 (a) is a simplified diagram of the probe 20 sec to remove the native oxide, and ultrasonically and the sample connection. When no voltage is oscillated in acetone for 3 min. Then, it was applied, the Fermi level of the semiconductor immersed in D.I. water. The potential image of the sample and the tip is aligned as shown in Fig.2 sample is shown in Fig.3 (a). The image measured (b). For KFM measurement, the vacuum level immediately after HF dip is shown in Fig.3 (b). difference is reduced to zero by adjusting the Compared to the previous one, the potential image is applied V . When the tip is over an n-type region, blurred which possibly result from the residual dc fluorine bond on the sample surface. the dc bias (Vm-Vn) is applied to nullify the vacuum level difference as shown in Fig.2 (c). When the tip is over a p-type region, the dc bias
(Vm-Vp) is applied as shown in Fig.2 (d). Thus, the surface potential difference between the two regions is obtained as the applied bias difference.
(a) (b) 3 Experiment and Result Fig.3. The surface potential image of samples after (a) HF 3.1 Sample preparation dip, acetone oscillation, and DI water rinse and (b) HF dip only. dose of 2E15, 2E14, 2E13 /cm2, respectively. The bright lines are un-implanted p-type regions, and the We also attempted to measure the samples with dark regions are As+ ions implanted n-type regions. oxidized surface. Fig.4 (a) shows the potential image The brighter color represents the higher potential, so of sample after rapid thermal oxidation at 900℃ for we can differentiate between the two type regions by 1 min. Fig.4 (b) shows the potential image of sample the potential height. The potential difference is 138 after immersing in H2SO4 : H2O2 3 : 1 solution at mV, 73 mV, and 30 mV with respect to the dosage 100℃ for 10 min. The qualities of potential image of 2E15, 2E14, 2E13 /cm2. in these two samples are also very poor. The clearest obtained potential image is the sample which was dipped in diluted HF, oscillated in acetone, and the rinsed in D. I. water. It is suspected that Si-F bonds on the HF dipped sample and the Si-O layer on the oxidized sample changes the surface potential so that the contrast of surface potential image degrades [5]. (a) (b)
(a) (b) (c) Fig. 6. The image of the samples with implant energy 20 Fig. 4. The surface potential image of the sample with keV and dose (a) 2E15, (b) 2E14, (c) 2E13 /cm2. (a)rapid thermal oxide and (b) chemical oxide.
107 1E21 106 -> 30Si <- 75As 1E20 105
4 3 1E19 10 m c c n /
3 t m
o 10 t
a 1E18 102 (a) (b) 1E17 101
1E16 100 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 um
107 1E21 6 -> 30Si 10 <- 75As (c) (d) 1E20 105
3 4
m 1E19 10 c c / n
t m 3
o 10 t
Fig. 5. The surface potential image of samples with As+ ion a 1E18 102 implantation at 50 KeV to a dose of (a) 2E15, (b) 2E14, (c) 1E17 101 2E13 /cm2. (d) shows the surface potential across the sample 1E16 100 shown in (a). 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 um
3.3 PN junction potential difference Fig.5 (a), (b), and (c) shows the potential image of the sample with As+ implantation at 50 KeV to a Fig.7. SIMS depth profiles of the samples with As+ ion [3] G. H. Buh, H. J. Chung, J. H. Yi, I. T. Yoon, and implantation at 20 keV to a dose of (a) 2E15 and (b) Y. Kuk, Electrical Characterization of an 2E13. Operating Si pn-junction Diode with Scanning Capacitance Microscopy and Kelvin Probe Force Fig. 6 (a), (b), and (c) show the surface potential Microscopy, J. Appl. Phys, Vol. 90, No.1, 2001, + image of samples after As ion implantation at 20 pp. 443-448. 2 KeV to a dose of 2E15, 2E14, 2E13 /cm , [4] A. Kikukawa, S. Hosala, and R. Imura, Silicon respectively. The corresponded potential difference pn Junction imaging and Characterization Using is 133 mV, 73 mV, and 34 mV. sensitive enhanced Kelvin Probe Force Fig.7 compares the SIMS depth profiles of the Microscopy, Appl. Phys. Lett., Vol. 66, No. 25, samples with As implantation at 20 KeV to a dose of 1995, pp. 3510-3512. 2E15 and 2E13. Higher implantation dose results in [5] H. Sugimura, Y. Ishia, K. Hayashi, O. Takai, higher surface concentration, which in turn results in and N. Nakairi, Potential Shielding by the higher surface potential difference detected by Surface Water Layer in Kelvin Probe Force KFM. Microscopy, Appl. Phys. Lett., Vol. 80, No. 8, 2002, pp. 1459-1461. 4 Conclusion Kelvin probe force microscopy is a tool for 2D surface potential imaging. For semiconductor material, it is able to measure dopant type and dopant concentration. In this work, we develop a reliable surface treatment technique before KFM measurement. Stable and repeatable results on As+ ion implanted n+-p junctions are demonstrated. Since the potential difference is high enough, it is possible to correlate the potential difference to the surface concentration after suitable calibration. The spatial resolution of KFM depends on the tip diameter. If single wall carbon nanotube can be mounted on the metal tip, the spatial resolution could be improved to be better than 0.5 nm. Then, KFM would be a very powerful characterization method for nano CMOSFETs.
Acknowledgement: The authors thank Prof. S. Gwo and Mr. S.-D. Tzeng in the Department of Physics, National Tsing- Hua University for the operation of KFM system and helpful discussion.
References: [1] S. Kitamura and M. Iwatsuki, High-resolution Imaging of Contact Potential Difference with Ultrahigh Vacuum Noncontact Atomic Force Microscope, Appl. Phys. Lett., Vol. 72, No. 24, 1998, pp. 3154-3156. [2] P. De Wolf, R. Stephenson, T. Trenkler, T. Clarysee, and T. Hantschel, and W. Vandervorst, Status and Review of Two-dimensional Carrier and Dopant Profiling Using Scanning Probe Microscopy, J. Vac. Sci. Technol. B, Vol. 18, No. 1, 2000, pp. 361-368.