摘要:

The occurrence in SCM images of contrast reversal (CR) phenomena (i.e. a non-monotonic relationship between the amplitude of the SCM signal and the local carrier concentration) has been investigated theoretically by a three dimensional (3D) simulation model and experimentally by measurements of calibrated staircase structures. The CR effect is commonly ascribed to an improper DC bias voltage, far away from the flat band voltage, and this assumption has been supported in the past by two-dimensional (2D) simulations. Simulations with a 3D model show that CR phenomena are less likely to occur than foreseen by the 2D model. The experimental dC/dV curves are in better agreement with the 3D simulations. However, CR phenomena are observed sometimes in a sporadic way even when the DC bias voltage is correctly set at the experimental flat band voltage. A sudden degradation of the probe-tip is probably the cause of the sporadic occurrence of CR phenomena. ©2001 American Institute of Physics.

ISSN:0094-243X    

DOI:10.1063/1.1354471

出版商:AIP    

年代:1901

数据来源: AIP

摘要:

Scanning Capacitance Microscopy (SCM) can be used to measure doping density in silicon with a spatial resolution of about 10 nm. In order to make such measurements on fabricated devices, the device must be cross-sectioned, and an oxide sufficient to form an MOS capacitor with the SCM probe tip must be grown on the cut face without affecting the device structure. Thus thermal oxidation under conditions usually employed to grow the gate oxide (≈900&hthinsp;°C) cannot be used. We have systematically investigated oxide growth on silicon under mild conditions. An oxide sufficient for SCM can be grown thermally on a clean, polished silicon surface at temperatures of 300&hthinsp;°C and above. In addition, such an oxide can be made near room temperature in the presence of ozone generated by intense ultraviolet light. SCM measurements using such an oxide show stable flatband voltages and breakdown voltages above 3 V. The oxide thickness, measured ellipsometrically, is(2.0±0.2)nm.©2001 American Institute of Physics.

ISSN:0094-243X    

DOI:10.1063/1.1354472

出版商:AIP    

年代:1901

数据来源: AIP

摘要:

Secondary Ion Mass Spectrometry (SIMS) has proven to be an essential tool for ultra-shallow junction metrology. Accurate and precise measurement of ultra-low energy implants can be accomplished if SIMS artifacts are understood and controlled. These artifacts include excessive atomic mixing, ion yield and sputter rate variations at the onset of sputtering, and the formation of sputter-induced surface topography. Control of these artifacts can be accomplished through the use of sub-keV primary ion beams incident at select angles. From results in the literature and from International SEMATECH’s Materials Analysis Lab, it appears likely that SIMS can provide the necessary accuracy and precision to support ultra-shallow junction metrology to at least the 100 nm node in 2005. ©2001 American Institute of Physics.

ISSN:0094-243X    

DOI:10.1063/1.1354473

出版商:AIP    

年代:1901

数据来源: AIP

摘要:

Low energy Secondary Ion Mass Spectrometry (SIMS) was employed to study graded-baseSi1−xGex:Cheterostructure bipolar transistors (HBTs) structural properties. Using anO2+beam at 500 eV with normal incident angle, Ge profiles can be quantified by minimizing the influence of matrix effect. This is achieved with a correction based on the linearity of the ratio(IGe/ISi)with respect to the ratio(x/1−x).The SIMS results showed an excellent correlation with the graded Auger Ge profile in 5–20&percent; range. Moreover, SIMS analysis revealed an enhanced Ge diffusion with the carbon incorporation, which was used to suppress base boron diffusion during the rapid thermal annealing (RTA) process. Based on the SIMS Ge profile, device simulation can be used to design Ge profile shape in order to optimize process throughput without impact on the device performance. The high depth resolution SIMS data is essential forSi1−xGex:CHBTs structural characterization and process optimization. ©2001 American Institute of Physics.

ISSN:0094-243X    

DOI:10.1063/1.1354474

出版商:AIP    

年代:1901

数据来源: AIP

摘要:

A feasibility study was undertaken to determine if radiochemical neutron activation analysis (RNAA) can be used to certify the retained dose of phosphorus implanted in silicon, with the goal of producing a phosphorus SRM. Six pieces of silicon, implanted with a nominal phosphorus dose of8.5×1014&hthinsp;atoms⋅cm−2were irradiated at a neutron flux of1.05×1014&hthinsp;cm−2⋅s−1.The samples were mixed with carrier, dissolved in acid, the phosphorus isolated by chemical separation, and32Pmeasured using a beta proportional counter. A mean phosphorus concentration of(8.35±0.20)×1014&hthinsp;atoms⋅cm−2(uncertainty=1standard deviation) was determined for the six samples, in agreement with the nominal implanted dose. ©2001 American Institute of Physics.

ISSN:0094-243X    

DOI:10.1063/1.1354475

出版商:AIP    

年代:1901

数据来源: AIP

摘要:

The Metrology section of the 1999 International Technology Roadmap for Semiconductors (ITRS) specifies in-line dopant profile concentration precision requirements ranging from a value of 5&percent; in 1999 to a value of 2&percent; in 2008. These values are to be accomplished with “low systematic error.” Secondary ion mass spectrometry (SIMS) has a demonstrated capability to meet these requirements for B, As, and P. However, the detailed analytical protocols required to achieve these goals have not been completely specified. This paper reports the parameters that must be controlled to make highly repeatable dose measurements of arsenic implants in silicon with magnetic sector SIMS instruments. Instrument conditions that were investigated include arsenic analytical species, matrix ion species, energy bandpass, and sample holder design. With optimized settings, we demonstrate the ability to distinguish arsenic implant doses differing by 5&percent;. Low systematic error is achieved by referencing the measurements to NIST SRM 2134, which has a certified arsenic dose value of7.33×1014&hthinsp;cm−2and an expanded dose uncertainty of only 0.38&percent;. ©2001 American Institute of Physics.

ISSN:0094-243X    

DOI:10.1063/1.1354476

出版商:AIP    

年代:1901

数据来源: AIP

摘要:

We are evaluating the use of polyatomic and cluster primary ion beams for characterization of semiconductor materials by secondary ion mass spectrometry using both magnetic sector and time-of-flight SIMS instruments. Primary ion beams ofSF5+,C8−andCsC6−have been used to analyze low energy arsenic implants in silicon, boron delta-doped structures, thin gate oxides, metal multilayers, organic surface contamination and photoresist thin films. Compared to monoatomic bombardment under the same conditions, cluster ion beams offer improved depth resolution for silicon depth profiling and a reduction in sputter-induced topography for metals. For organic materials, the use of a cluster ion beam can give large improvements in yield for characteristic secondary ions and can minimize beam-induced degradation in some materials. ©2001 American Institute of Physics.

ISSN:0094-243X    

DOI:10.1063/1.1354477

出版商:AIP    

年代:1901

数据来源: AIP

摘要:

With the lowering of the implantation energies it becomes more and more a problem to perform a correct quantification of the implanted species, especially if a major part of the implant is deposited in a screening or native oxide. Usually the Secondary Ion Mass Spectrometry (SIMS) is applied for the analysis of those implants. The problem using SIMS is the strong sensitivity for changes in the composition of the matrix. This matrix effect causes a dramatic change of the Relative Sensitivity Factor (RSF) dependent on the oxidation state of the sample. In this paper a correction procedure for the changing oxidation state is proposed. ©2001 American Institute of Physics.

ISSN:0094-243X    

DOI:10.1063/1.1354478

出版商:AIP    

年代:1901

数据来源: AIP