Theory of the single contact electron beam induced current effect

Abstract

All publications on the single contact electron beam induced
current (SC-EBIC) technique so far have been concerned with the
application of the technique. This paper seeks to examine the theory
behind the technique and supports it with experimental observation. It
will be shown that the technique can be used, not only on electron and
ion beam machines, but also on any scanning equipment that is capable of
generating electron-hole pairs within a semiconductor device, e.g., with
the use of a fine laser beam

Full text

Page 1

This document is downloaded from DR-NTU, Nanyang Technological This document is downloaded from DR-NTU, Nanyang Technological
University Library, Singapore.University Library, Singapore.
Title
Theory of the single contact electron beam induced
current effect.( Published version )
Author(s)Ong, Vincent K. S.; Lau, K. T.; Ma, Jianguo.
Citation
Ong, V. K. S., Lau, K. T., & Ma, J. G. (2000). Theory of
the single contact electron beam induced current effect.
IEEE Transactions nn Electron Devices, 47(4), 897-899.
Date2000
URLhttp://hdl.handle.net/10220/4714
Rights
© 2000 IEEE. Personal use of this material is permitted.
However, permission to reprint/republish this material for
advertising or promotional purposes or for creating new
collective works for resale or redistribution to servers or
lists, or to reuse any copyrighted component of this work
in other works must be obtained from the IEEE. This
material is presented to ensure timely dissemination of
scholarly and technical work. Copyright and all rights
therein are retained by authors or by other copyright
holders. All persons copying this information are
expected to adhere to the terms and constraints invoked
by each author's copyright. In most cases, these works
may not be reposted without the explicit permission of the
copyright holder.

Page 2

IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 47, NO. 4, APRIL 2000897
peak current density of 8.9 kA/cm?was obtained for an unstrained
AlInAsSb/InGaAs double-barrier resonant tunneling diode.
REFERENCES
[1] L.L.Chang,L.Esaki,andR.Tsu,“Resonanttunnelinginsemiconductor
double barriers,” Appl. Phys. Lett., vol. 24, pp. 593–595, 1974.
[2] T.J.Shewchuk
etal., “Resonant
GaAs-Al Ga
As heterostructure at room temperature,” Appl.
Phys. Lett., vol. 46, pp. 508–510, 1985.
[3] C. I. Huang et al., “AlGaAs/GaAs double barrier diodes with high
peak-to-valley current ratio,” Appl. Phys. Lett., vol. 51, pp. 121–123,
1987.
[4] M. Tsuchiya, H. Sakaki, and J. Yoshino, “Room temperature observa-
tion of differential negative resistance in an AlAs/GaAs/AlAs resonant
tunneling diode,” Jpn J. Appl. Phys., vol. 24, pp. L446–L468, 1985.
[5] M. Tsuchiya and H. Sakaki, “Dependence of resonant tunneling cur-
rentonwellwidthsinAlAs/GaAs/AlAsdoublebarrierdiodestructures,”
Appl. Phys. Lett., vol. 49, pp. 88–90, 1986.
[6] T. Inata et al., “Excellent negative differential resistance of InAlAs/In-
GaAs resonant tunneling barrier structures grown by MBE,” Jpn. J.
Appl. Phys., vol. 25, pp. L983–L985, 1986.
[7] Y.Sugiyama
etal.,“Current-voltage
In
GaAs/In AlAs resonant tunneling barrier structures
grown by molecular beam epitaxy,” Appl. Phys. Lett., vol. 52, pp.
314–316, 1988.
[8] J. H. Smet, T. P. E. Broekaert, and C. G. Fonstad, “Peak-to-valley
current ratios as high as 50:1 at room temperature in pseudomorphic
In
GaAs/AlAs/InAs resonant tunneling diodes,” J. Appl. Phys.,
vol. 71, pp. 2475–2477, 1992.
[9] J. S. Su, W. C. Hsu, W. Lin, and S. Y. Jain, “High-breakdown char-
acteristics of the InP-based heterostructure field-effect transistor with
Al
InAsSb Schottky layer,” IEEE Electron Device Lett.,
vol. 19, pp. 195–197, 1998.
[10] G. Bastard, “Theoretical investigations of superlattice band structure
in the envelope function approximation,” Phys. Rev. B, vol. 25, pp.
7584–7597, 1982.
[11] L. Hrivnák, “Exciton binding energy as a function of the well width,” J.
Appl. Phys., vol. 72, pp. 3218–3219, 1992.
[12] R. People, K. W. Wecht, K. Alavi, and A. Y. Cho, “Measurement of
the conduction-band discontinuity of molecular beam epitaxial grown
In
AlAs/In GaAs N-n heterojunction by C/V profiling,”
Appl. Phys. Lett., vol. 43, pp. 118–120, 1983.
[13] Y. Sugiyama et al., “Conduction band edge discontinuity of
In
Ga As/In(GaAl )
erostructures,” Jpn. J. Appl. Phys., vol. 25, pp. L648–L650, 1986.
[14] L. Bornstein, Numerical Data and Functional Relationships in Science
and Technology, O. Madelung, M. Schulz, and H. Weiss, Eds.
Germany: Springer-Verlag, 1982, vol. 17a and 17b.
[15] S. Adachi, “Band gaps and refractive indices of AlGaAsSb, GaInAsSb,
and InPAsSb: Key properties for a variety of the 2–4-?m optoelectronic
device applications,” J. Appl. Phys., vol. 61, pp. 4869–4876, 1987.
[16] R. People and S. A. Jackson, Semiconductors and Semimetals, T. P.
Pearsall, Ed. New York: Academic, 1990, vol. 32, p. 154.
[17] J. I. Pankove, Optical Process in Semiconductors.
NJ: Prentice-Hall, 1971, p. 412.
[18] A. D. Andreev and G. G. Zegrya, “Theoretical study of thresholdless
Auger recombination in compressively strained InAlAsSb/GaSb
quantum wells,” Appl. Phys. Lett., vol. 70, pp. 601–603, 1997.
[19] P. W. Yu et al., “High valence-band offset of GaSbAs-InAlAs quantum
wells grown by molecular beam epitaxy,” Appl. Phys. Lett., vol. 61, pp.
2317–2319, 1992.
tunnelingoscillations in
characteristicsof
As ??
??
?? het-
Berlin,
Englewood Cliffs,
Theory of the Single Contact Electron Beam Induced
Current Effect
V. K. S. Ong, K. T. Lau, and J. G. Ma
Abstract—All publications on the single contact electron beam induced
current (SC-EBIC) technique so far have been concerned with the appli-
cation of the technique. This paper seeks to examine the theory behind the
technique and supports it with experimental observation. It will be shown
that the technique can be used, not only on electron and ion beam ma-
chines, but also on any scanning equipment that is capable of generating
electron-hole pairs within a semiconductor device, e.g., with the use of a
fine laser beam.
Index Terms—Displacement current, EBIC, electric displacement , elec-
tromagnetics, SC-EBIC, SEM.
I. INTRODUCTION
Ithasbeenshownrecentlythatitispossibletocaptureelectronbeam
induced current (EBIC) images of integrated circuit (IC’s) using only
a single contact, as opposed to the multiple contacts required by the
traditional EBIC technique [1], [2]. This new technique is called the
single contact electron beam induced current technique or SC-EBIC.
An example of a micrograph captured by this technique is shown in
Fig. 1. It was also shown in [1] and [2] that EBIC images of entire IC’s
can be captured using this method and that it was possible to use it to
detect device failures. This same technique was also used successfully
on ion beam machines [3], [4].
Traditional EBIC requires that a complete electrical circuit be
formed across a junction that is to be imaged. For every extra junction
that needs to be imaged, an additional connection needs to be made
on the semiconductor. For a magnified image of a small area of an
IC, which may have hundreds of junctions, more than that number of
connections need to be made on the IC. This is not only unfeasible in
practice, but the connections, which is normally made with probe pins
[5]–[7], will block the electron beam and prevent it from reaching the
IC, causing massive shadowing.
The idea behind the technique described in [1]–[4] is to completely
remove all but one of the probes. This is possible in the transient mode
since the circuit, which is traditionally completed by the second probe,
is now replaced by the displacement current created by the transient.
II. SC-EBIC EXPERIMENT FOR A SINGLE JUNCTION
A simple SC-EBIC experiment is set up by wiring up a piece of
semiconductor sample with a single metallurgical junction fabricated
on it as shown in Fig. 2. At time ???, say, an electron beam is turned on
to impinge directly upon the junction. A few tens of milliseconds later,
the beam is turned off. The current measured by the current meter is
then plotted against time. The result is shown in Fig. 3.
III. SC-EBIC THEORY FOR A SINGLE JUNCTION
Consider a single semiconductor junction connected as shown in
Fig. 2. The device is initially in thermal equilibrium. Under this con-
dition the current meter registers a zero current. At time ???, a beam
Manuscript received October 6, 1998; revised February 12, 1999. The review
of this paper was arranged by Editor J. N. Hollenhorst.
The authors are with the School of Electrical and Electronic Engineering,
Nanyang Technological University, Nanyang Avenue, Singapore 639798, Re-
public of Singapore (e-mail: eongks@ntu.edu.sg).
Publisher Item Identifier S 0018-9383(00)00172-6.
0018-9383/00$10.00 © 2000 IEEE

Page 3

898IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 47, NO. 4, APRIL 2000
Fig. 1.Example of an SC-EBIC image.
Fig. 2. Single function SC-EBIC configuration.
of electrons begins to impinge upon the semiconductor junction. Elec-
trons entering into the semiconductor will ionize the material and gen-
erate electron-hole pairs (EHP’s). The electric field at the junction is
suchthattheholesandelectronsdrifttowardsthep-andn-typeregions,
respectively.
The p-type region has a short-circuit to ground through the current
meter, and is therefore always at ground potential. The n-type region,
however, is floating. The beam generated electrons that are swept by
the electric field into the n-type region will therefore be trapped and
accumulates in the n-type region. These negatively charged electrons
accumulating in the n-type region will cause two groups of electric
displacements to terminate into it. This is illustrated in Fig. 4, where
the space charge region (SCR) is removed. The first group of electric
displacements, denoted as? ? ?? in Fig. 4, are those which originate from
the accumulation of positively charged holes in the p-type region. The
second group, denoted as ? ? ???in Fig. 4, are the electric displacements
whichoriginatefromnearbygroundplanes,e.g.,theobjectivelenspole
piece and housing of the SEM, and the grounded specimen holder.
The rate of change of this second group of electric displacements
give rise to displacement currents flowing from the nearby ground
planes into the n-type region [8]. This displacement current is balanced
by an equal amount of electric current flowing from the p-type region
to ground through the current meter. This explains the initial flow of
current at ??? ? ? ms in Fig. 3.
As the electrons accumulate in the n-type region, the strength of the
electricdisplacementsalsoincreases.Thisincreaseinelectricdisplace-
ment will cause the potential of the n-type region to fall. This in turn
causes the junction electric field to weaken, and the junction will begin
to conduct in the forward direction.
Asthejunctionforwardbiases,electronsgetinjectedfromthen-type
region into the p-type region, hence the rate of electrons accumulating
Fig. 3.Graph of a single junction SC-EBIC experiment.
Fig. 4.Displacement vectors.
in the n-type region decreases. This reduces the rate of change of the
second group of electric displacements, ? ? ???in Fig. 4, which directly
decreasesthedisplacementcurrentsflowingintothen-typeregionfrom
the nearby ground planes. Consequently, this reduction in the displace-
ment current is balanced by a reduced flow of electric current out of
the p-type region to ground through the current meter. This explains
the reduced flow of SC-EBIC current in Fig. 3 between ???and ? ? ??
ms.
An equilibrium takes place between 20–30 ms in the graph of Fig. 3.
ThisiswhenthegenerationcurrentduetoEHPcreationbytheelectron
beam and separation by the junction electric field balances the forward
biascurrentinthe junction.As aresult ofthis balance,the accumulated
electrons in the n-type region neither increase nor decrease. The elec-
tricdisplacementsfromnearbygroundplanes terminatinginthen-type
region,? ? ???, therefore become constant, and the displacement current,
which is the partial time derivative of this electric displacement, van-
ishes. The zero displacement current is consequently balanced by zero
SC-EBIC currentflowing through thecurrentmeter. However, the con-
stant electron beam current itself continues to flow through the current
meter. This current is about 0.1 nA, and therefore cannot be readily
seen in Fig. 3.
Whentheelectronbeamisremovedfromthejunctionattime???,the
generation current suddenly disappears. This causes a sudden change
fromanequilibriumstatewheretheaccumulatedelectronsinthen-type
region neither increase nor decrease, to a state in which the depletion
rate of the accumulated electrons in the n-type region equals the junc-
tion forward bias current. This change consequently gives rise to dis-
placementcurrentsflowingfromthen-typeregiontothenearbyground
planes (opposite in direction to ? ? ???), which is again balanced by the
SC-EBIC current flowing from the ground to the p-type region through
the current meter. This can be seen in Fig. 3 where ??? ? ?? ms.
Aselectronsfromthen-typeregioncontinuestoinjectintothep-type
region, the amount of accumulated electrons in the n-type region de-
creases. This causes the electric displacements to also decrease, re-
sulting in a rise in the potential of the n-type region. This in turn causes
the junction to be less and less strongly forward biased. The rate of de-

Page 4

IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 47, NO. 4, APRIL 2000 899
crease of the accumulated electrons decreases, resulting in a decrease
in the displacement current flowing from the n-type region to ground.
TheSC-EBICcurrentwhichbalancesthedisplacementcurrentalsode-
creases, as can be seen in Fig. 3 from ??? ? ?? ms onwards.
Itistobenotedthatsincethejunctionbecomeslessandlessstrongly
forwardbiased aselectronsflow out ofthe n-typeregion,therate ofthe
flow of electrons also decreases in a very gradual manner. Although it
is not obvious in Fig. 3, the SC-EBIC current in this figure does not
actually become zero at the 40 ns mark. It continues on asymptotically.
If one were to look closely at Fig. 3, it can be seen that the wiggly
line lies mainly above the zero current line. This effect can also be seen
inFig.1,wherethewhitestreakscontinueonstronglyrightthroughthe
right hand edge of the micrograph. It is important to note here that the
areas underthe curveofFig. 3 beforeand after???mustbe equal ifone
were to plot it for a few hundred ms, since each of these areas represent
the total amount of excess charge stored in the n-type region.
It is also to be noted that the entire process described above sits on
top of the generation and recombination processes that are going on in
the semiconductor. The process described above is mainly electromag-
netic in nature, and is therefore a function of the excess charges which
accumulate in the n-type region. These excess charges are not affected
by the generation and recombination processes since the n-type region
is floating in the electrical sense.
The time constant of the entire measurement circuit has been mea-
sured to be of the order of about half a microsecond. This is about four
ordersofmagnitudebetterthanthetimeconstantsinthegraphofFig.3.
In fact the superior time constant of the measurement circuit is self evi-
dent from the fact that the electrical noise is reproducedquite faithfully
by the measurement circuit in Fig. 3. If the time constant of the mea-
surement circuit had been significantly poorer, then the electrical noise
would have been smoothed out.
TheexplanationfortheSC-EBICphenomenongivenaboveassumes
the use of energetic electrons to generate the EHP’s. It can be seen that
the explanation given above is similarly valid, regardless of the means
of generating the EHP’s. So far, the technique has be used successfully
with electron and ion beams. Successful preliminary experiments have
been conducted with the use of laser beam.
IV. CONCLUSION
The explanation behind the SC-EBIC effect was described in this
article using well known electromagnetics theory. It was shown that
the explanation agrees qualitatively with experiment. It is hoped that
this will fill up the void in the literature on the SC-EBIC effect.
ACKNOWLEDGMENT
ApartofthisworkwasdoneattheNationalUniversityofSingapore.
V. K. S. Ong would like to acknowledge interesting discussions with
D. S. H. Chan and J. C. H. Phang on this work.
REFERENCES
[1] V. K. S. Ong, J. C. H. Phang, and D. S. H. Chan, “Novel EBIC observa-
tion of unconnected junctions of large area VLSI circuits,” in Proc. 20th
Int. Symp. Testing and Failure Analysis, Los Angeles, CA, Nov. 1994,
pp. 49–56.
[2] V.K.S.Ong,P.C.Liu,andK.T.Lau,“Largeareaelectronbeaminduced
currentimagingwithasinglecontact,”Solid-StateElectron.,vol.43,pp.
17–25, Jan. 1999.
[3] S. Kolachina et al., “Ion beam induced charge (IBIC) imaging for the
failure analysis of multi-level metal VLSI circuits,” in Proc. 21st Int.
Symp. Testing and Failure Analysis, Santa Clara, CA, Nov. 1995, pp.
19–24.
[4] S. Kolachina et al., “Unconnected junction contrast in ion beam induced
charge microscopy,” Appl. Phys. Lett., vol. 68, pp. 532–534, 1996.
[5] C. Coakley and A. Marquez, “Microprobing and EBIC for VLSI tech-
nology,” in Proc. 18th Int. Symp. Testing and Failure Analysis, Los An-
geles, CA, 1992, pp. 43–48.
[6] E. Hutchinson, “CMOS characterization using EBIC: Applications and
performance,” in Proc. 17th Int. Symp. Testing and Failure Analysis,
Santa Clara, CA, 1991, pp. 199–204.
[7] G. V. Lukianoff, “Beam-induced current testing in the line support
environment,” in Electron Beam Testing Technology, J. T. L. Thong,
Ed.New York: Plenum, 1993, pp. 433–444.
[8] D. K. Cheng, Field and Wave Electromagnetics.
dison-Wesley, 1989.
Boston, MA: Ad-