and two-dimensional dopant/carrier profiling for ULSI

One- and two-dimensional dopant/carrier profiling for ULSI W. Vandervorst, T. Clarysse, P. De Wolf, T. Trenkler, T. Hantschel, R. Stephenson, and T. Janssens Citation: AIP Conference Proceedings 449, 617 (1998); doi: 10.1063/1.56849 View online: http://dx.doi.org/10.1063/1.56849 View Table of Contents: http://scitation.aip.org/content/aip/proceeding/aipcp/449?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Two-dimensional effects on ultralow energy B implants in Si J. Vac. Sci. Technol. B 20, 414 (2002); 10.1116/1.1424277 Status and review of two-dimensional carrier and dopant profiling using scanning probe microscopy J. Vac. Sci. Technol. B 18, 361 (2000); 10.1116/1.591198 Dopant characterization round-robin study performed on two-dimensional test structures fabricated at Texas Instruments AIP Conf. Proc. 449, 741 (1998); 10.1063/1.56919 Fast low energy SIMS depth profiling for ULSI applications AIP Conf. Proc. 449, 782 (1998); 10.1063/1.56865 Metrology aspects of SIMS depth profiling for advanced ULSI processes AIP Conf. Proc. 449, 169 (1998); 10.1063/1.56792

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 54.210.20.124 On: Sat, 07 Nov 2015 20:16:27

One- and two-dimensional dopant/carrier profiling for ULSI W.Vandervorst°, T.Clarysse, P.De Wolf, T.Trenkler, T.Hantschel, R.Stephenson and T.Janssens Imec, Kapeldreef 75 B-3001 Belgium "~":KULeuven, INSYS, Kard. Mercierlaan 92, B-3001 Belgium Dopant/carrier profiles constitute the basis of the operation of a semiconductordevice and thus play a decisive role in the performance of a transistor and are subjected to the same scaling laws as the other constituents of a modern semiconductor device and continuously evolve towards shallower and more complex configurations. This evolution has increased the demands on the profiling techniques in particular in terms of resolution and quantification such that a constant reevaluation and improvement of the tools is required. As no single technique provides all the necessary information (dopant distribution, electrical activation,..) with the requested spatial and depth resolution, the present paper attempts to provide an assessment of those tools which can be considered as the main metrology technologies for ULSI-applications. For 1D-dopant profiling secondary ion mass spectrometry (SIMS) has progressed towards a generally accepted tool meeting the requirements. For 1D-carrier profiling spreading resistance profiling and microwave surface impedance profiling are envisaged as the best choices but extra developments are required to promote them to routinely applicable methods. As no main metmlogy tool exist for 2D-dopant profiling, main emphasis is on 2D-carrier profiling tools based on scanning probe microscopy. Scanning spreading resistance (SSRM) and scanning capacitance microscopy (SCM) are the preferred methods although neither of them already meets all the requirements. Complementary information can be extracted from Nanopotentiometry which samples the device operation in more detail. Concurrent use of carrier profiling tools, Nanopotentiometry, analysis of device characteristics and simulations is required to provide a complete characterization of deep submicron devices.

INTRODUCTION Dopant/carrier profiles constitute the basis of the operation of a semiconductor device and thus play a decisive role in the performance of a transistor. These dopant distributions are subjected to the same scaling laws as the other constituents of a modem semiconductor device and continuously evolve towards shallower and more complex configurations. Facing the increasing costs of the actual processing, new designs of process technology (and thus doping profiles) start off with an exploration of the various fabrication possibilities using process and device simulations (TCAD). If these simulations can be fine tuned to promising results, the experiment can be completed in silicon. Upon fabrication electrical charateristics can be measured and, in the ideal case, should match the expected performance. Provided our understanding/modeling of doping processes and process control would be advanced to such a level that these simulations can be viewed as the perfect mimic of the actual processes, metrology (and more specifically dopant/carrier profiling) should be totally obsolete in a m o d e m process

environment. Unfortunately the latter is not yet achieved and process control as well as T C A D calibration hinges very strongly on feedback from dopant/carrier profiling tools. Moreover, since too many variables in the simulators are not exactly known, process splits during process development are a common approach during the first iteration as well. Again, assessment of the impact of these splits can only be performed when appropriate measurement tools are available. Finally, in order to prevent the recurrence of deficient processing and non-operational devices, failure analysis is required based on tools which provide detailed information on the structure under investigation. As indicated in Fig. 1 a strong interplay exists between the process flow, process engineers, T C A D and experimental verification necessitating that metrology tools stay in line with the demands from processing technology. With the increasing dominance of two- (three-dimensional) effects in devices, it is obvious that the requirements placed on metrology have increased from simple 1D-dopant profiling for ancient technologies towards full 3Dchemical and electrical characterization. As simple 1D-models and experiments are no longer sufficient

CP449, Characterization and Metrologyfor ULSI Technology: 1998 International Conference edited by D. G. Seiler, A. C. Diebold, W. M. Bullis, T. J. Shaffner, R. McDonald, and E. J. Walters © 1998 The American Institute of Physics 1-56396-753-7/98/$15.00 617 This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 54.210.20.124 On: Sat, 07 Nov 2015 20:16:27

Facing the limitations of the techniques (dopant or carrier sensitive) and the various applications (homogeneous samples or device like structures) alternative methods can be used. Consequenty this paper is divided into one-dimensional and twodimensional profiling, distinguishing as well between dopant and carrier profiles.

to explain the observed processing effects and device characteristics, availability of adequate metrology for ULSI technology is a very important issue and efforts should be directed towards making these available for routine applications. EXl~inn~l ~nilkafion

R a ~ ard

The outline of the paper is therefore based on the classification included in table 1. T A B L E 1. Classification of characterization methods available for dopant/carrier profiling.

FIGURE 1. Interplay between experimental verification, process engineering and TCAD. Indicated are the main technologies for the experimental verification,

The changes in technologies have also increased the demands on the profiling techniques in particular in terms of resolution and quantification such that a constant reevaluation and improvement of the tools is required. Ideally one single technique should provide all the necessary information (dopant distribution, electrical activation,..) with the requested spatial and depth resolution. Since at present no single technique is available satisfying all the industrial needs, the present paper attempts to provide an assessment of those tools which can be considered as (candidates for) the main metrology technologies for ULSIapplications. The latter implies that these tools do (or hold the prospect to) achieve all the requirements in terms of : 1. independent quantification, 2. quantification accuracy, 3. sensitivity and dynamic range 4. spatial resolution 5. depth resolution 6. direct applicability to any arbitrary device (i.e. no need for special test structures or extra processing)

Supporting metrology technologies can be viewed as those methods which fail on one of these criteria but may, within their limits, still provide substantial support to the main technologies. Where possible reference to these methods will be made but it is not the objective of this paper to provide a detailed overview of them also.

Main technolo[~ies

Supportin~ technologies

1D-profiles

SIMS

Nuclear methods

2D-profiles device analysis

non-existent

2D-SIMS Lateral SIMS Tomography SIMS

ID-profiles

SRP

C-V FPP (MicroSRP)' (Microwave Surface Impedance Profiler)'

2D-profiles device analysis

SSRM SCM

Selective etching + TEM Selective etching + AFM Secondary electron emission Other scanning probe methods (SKM, EFM, STM,..)

Dopant profiling

Carrier profiling,

Nanopotentiometry

'These could move to the main technologies if a commercial instrument were to be developed.

MEASUREMENT REQUIREMENTS When relating the required measurement capabilities with the needs of the semiconductor industry as expressed in the SIA roadmap [1], one faces the requirement for highly repeatable and quantitative tools (< 5 % accuracy) for ever decreasing profile depths (< 50 nm) with a dynamic concentration range spanning 10'LI02° a t / c m 3 . Moreover in view of the increasing importance of the 2 (and 3)D-interactions

618 This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 54.210.20.124 On: Sat, 07 Nov 2015 20:16:27

between dopants and doping processes (among others leading to the Reverse Short Channel Effect [2,3]), a

strong interest has surfaced in tools suited for 2Dcharacterization of doping profiles. As the intent is to

TABLE 2. Metrology requirements in relation to processing technology evolution according to NTRS for dopant profiling. Year

1997

1999

2001

2003

2006

2009

2012

100-200

70-140

60-120

50-100

40-80

15-30

10-20

50-100

36-72

30-60

26-52

20-40

15-30

10-20

4-6 10 ~7

6-1010 '7

7-13 10 t7

1-210 '~

2-3 10TM

> 3.5 10t~

> 7 10TM

5

3

3

2

1.5

1

0.8-0.6

13 n m

11 nm

10 n m

9 nm

7.5 nm