Conductive Atomic Force Microscopy E-Book
144,99 €
Mehr erfahren.
- Herausgeber: Wiley-VCH Verlag GmbH & Co. KGaA
- Kategorie: Wissenschaft und neue Technologien
- Sprache: Englisch
The first book to summarize the applications of CAFM as the most important method in the study of electronic properties of materials and devices at the nanoscale.
To provide a global perspective, the chapters are written by leading researchers and application scientists from all over the world and cover novel strategies, configurations and setups where new information will be obtained with the help of CAFM.
With its substantial content and logical structure, this is a valuable reference for researchers working with CAFM or planning to use it in their own fields of research.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 786
Veröffentlichungsjahr: 2017
Ähnliche
Table of Contents
Cover
Title Page
Copyright
Oxide Films and Conduction AFM
References
List of Contributors
Chapter 1: History and Status of the CAFM
1.1 The Atomic Force Microscope
1.2 The Conductive Atomic Force Microscope
1.3 History and Status of the CAFM
1.4 Editor's Choice: On the Use of CAFM to Study Nanogenerators Based on Nanowires
1.5 Conclusions
References
Chapter 2: Fabrication and Reliability of Conductive AFM Probes
2.1 Introduction
2.2 Manufacturing of Conductive AFM Probes
2.3 How to Choose Your C-AFM Tip
2.4 Tip Wear and Sample Damage: Applicable Forces and Currents in C-AFM
2.5 Conclusions
References
Chapter 3: Fundamentals of CAFM Operation Modes
3.1 Introduction
3.2 Tip-Sample Interaction: Contact Area, Effective Emission Area, and Conduction Mechanisms
3.3 Work Function Difference and Offset Voltage
3.4 Operation Modes
3.5 Case Studies
3.6 Conclusion and Future Perspectives
Acknowledgment
References
Chapter 4: Investigation of High-k Dielectric Stacks by C-AFM: Advantages, Limitations, and Possible Applications
4.1 Introduction
4.2 Comparison Between Macroscopic
I–V
Measurements and C-AFM
4.3 Influence of Displacement Currents on the Sensitivity of C-AFM Measurements
4.4 Applications of C-AFM
4.5 Conclusion
References
Chapter 5: Characterization of Grain Boundaries in Polycrystalline HfO2 Dielectrics
5.1 Introduction
5.2 Experimental Details and Sample Specifications
5.3 Formation of Grain Boundaries and Its Local Electrical Properties in HfO
2
Dielectric
5.4 RVS and CVS Stressing of HfO
2
/SiO
x
Dielectric Stack
5.5 Uniform Stressing with Successive Scanning in CAFM Mode
5.6 Conclusions
References
Chapter 6: CAFM Studies on Individual GeSi Quantum Dots and Quantum Rings
6.1 Introduction
6.2 Conductive Properties of Individual GeSi QDs and QRs
6.3 Modulating the Conductive Properties of GeSi QDs
6.4 Simultaneous Measurements of Composition and Current Distributions of GeSi QRs
6.5 Conclusions
References
Chapter 7: Conductive Atomic Force Microscopy of Two-Dimensional Electron Systems: From AlGaN/GaN Heterostructures to Graphene and MoS2
7.1 Introduction
7.2 Nanoscale Electrical Characterization of AlGaN/GaN Heterostructures
7.3 CAFM Characterization of Graphene and MoS
2
7.4 Conclusions
Acknowledgments
References
Chapter 8: Nanoscale Three-Dimensional Characterization with Scalpel SPM
8.1 Introduction
8.2 SPM Metrology with Depth Information
8.3 Scalpel SPM: A Tip-Based Slice-and-View Methodology
8.4 Applications
8.5 Conclusions and Outlook
References
Chapter 9: Conductive Atomic Force Microscopy for Nanolithography Based on Local Anodic Oxidation
9.1 Introduction to AFM Nanolithography
9.2 Local Anodic Oxidation
9.3 Kinetics of LAO
9.4 Measurement of Electrical Current During LAO
9.5 Conclusions
Acknowledgments
References
Chapter 10: Combination of Semiconductor Parameter Analyzer and Conductive Atomic Force Microscope for Advanced Nanoelectronic Characterization
10.1 Introduction
10.2 Combination of SPA and CAFM for Local Channel Hot Carrier Degradation Analysis
10.3 Combination of CAFM and SPA for Resistive Switching Analyses
10.4 Conclusions
References
Chapter 11: Design and Fabrication of a Logarithmic Amplifier for Scanning Probe Microscopes to Allow Wide-Range Current Measurements
11.1 Introduction
11.2 Fabrication of a Logarithmic Preamplifier for CAFMS
11.3 Conclusions
References
Chapter 12: Enhanced Current Dynamic Range Using ResiScope™ and Soft-ResiScope™ AFM Modes
12.1 Introduction
12.2 Conductive AFM
12.3 ResiScope™ Mode
12.4 Soft-ResiScope™ Mode
12.5 Conclusions
References
Chapter 13: Multiprobe Electrical Measurements without Optical Interference
13.1 Introduction
13.2 The Multiprobe Platform: Design and Key Features
13.3 The Present and the Future
13.4 Conclusions
References
Chapter 14: KPFM and its Use to Characterize the CPD in Different Materials
14.1 Introduction
14.2 Kelvin Probe Force Microscopy
14.3 Applications of KPFM
14.4 Conclusion and Outlook
Acknowledgment
References
Chapter 15: Hot Electron Nanoscopy and Spectroscopy (HENs)
15.1 Introduction
15.2 Coupling Schemes
15.3 Plasmonic Device and Optical Characterization
15.4 Theoretical Section
15.5 HENs Measurements: Plasmon-Assisted Current Maps and Ultimate Spatial Resolution
15.6 Kelvin Probe, HENs, and Electrical Techniques
15.7 Fast Pulses in Adiabatic Compression for Hot Electron Generation
15.8 Conclusion
Acknowledgments
References
Index
End User License Agreement
Pages
xi
xii
xiii
xiv
xv
xvi
xvii
xviii
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
Guide
Cover
Table of Contents
Begin Reading
List of Illustrations
Chapter 1: History and Status of the CAFM
Figure 1.1 Photograph of the dimension icon AFM from Bruker. This is the most representative image of an AFM, as this equipment (and previous models with very similar structure) are by far the most widespread (sold) AFM configuration.
Figure 1.2 Schematic displaying how the cantilever deflections in an AFM can be processed to obtain a topographic map. The deflection of the cantilever is detected with a laser, and the changes of the laser position in the photodiode are sent to the controller which corrects the position of the tip through the piezo tube. The data are used to build a topographic map using an image processing software compatible with AFMs.
Figure 1.3 Block diagram of a conventional conductive atomic force microscope. Compared to the AFM, the three new elements are the conductive tip, preamplifier, and sample bias.
Figure 1.4 Schematic of the effective emission area through which electrons can flow (
A
eff
) in a CAFM when the tip is placed on (a) a flat insulating sample and (b) a flat metallic electrode deposited on an insulating sample.
Figure 1.5 (a) Scanning electron microscope images of (a) metal-varnished silicon nanoprobe, (b) a sharpened solid Pt wire compatible for CAFMs, and (c) a metal-varnished silicon nanoprobe coated with a sheet of single-layer graphene. The picture in (a) intentionally shows a tip with the metallic varnish worn off, so that the core bulk of silicon can be observed.
Figure 1.6 (a, b) Photographs of two preamplifiers for CAFMs, the first one with fixed gain and the second with variable gain. (c) Photographs of different application modules for Bruker AFMs, including the CAFM module (which contains a preamplifier). (d) Simplified schematic of a conventional preamplifier used in CAFMs. The main parts are (i) an operational amplifier with high input impedance (OPAMP); (ii) a feedback resistor (
R
f
) and its parasite capacitor (
C
s
) and Johnson noise (
e
t
)-associated effects; (iii) a noise voltage source associated with the operational amplifier (
e
n
); and (iv) a capacitance associated with the input interconnections (
C
i
).
Figure 1.7 (a) First
I–V
curve measured with a CAFM. The sample was a 20 nm thick SiO
2
film grown on silicon. (b) Tunneling voltage image over an area 450 nm × 330 nm. The contrast is proportional to the threshold voltage required at each image point to generate a tunnel current of 0 pA. White to black corresponds to applied voltages of 15.0 and 40.5 V, respectively. (c) AFM topography of oxide surface recorded in the repulsive mode. Black to white correspond to a total vertical excursion of 90 nm.
Figure 1.8 (a) Topographical and (b) current images obtained at 6.5 V on the surface of a polycrystalline HfO
2
/SiO
2
/Si stack (area 1 µm × 1 µm. The CAFM revealed that the crystals are more insulating than the grain boundaries.
Figure 1.9 Statistic analysis of the (a) publication number and (b) citation about CAFM in each year
Figure 1.10 (a) Experimental setup and the whole CAFM measurement procedures in contact mode. (b) Schematic of a freestanding nanowire. (c) Physical strain distribution along the
Z
-axis of the nanowire under transverse stress. (d) Corresponding electrical filed distribution. (e) Coincidentally potential distribution along the tested nanowire. (f, g) Schematic of forward and reversed Schottky barrier respectively.
Figure 1.11 The simultaneously collected profiles of the topograsphy and CAFM current map. The corresponding images of size of 6.5 µm × 3.2 µm are displayed in the inset. A schematic in the left-bottom corner depicts the accurate tip location based on the corresponding cross-section line of topography.
Figure 1.12 Nanowire characterization. (a, b)Top and cross-sectional SEM images of the nanowire array. (c, d) Simultaneously collected topographic and current maps on the same sample using a Pt-Ir tip exerting a force of 0.1 nN and without bias. The scale bar for images (c, d) is 1 µm.
Chapter 2: Fabrication and Reliability of Conductive AFM Probes
Figure 2.1 SEM micrograph of a mold-and-release thin-film probe (NanoWorld® Pyrex Nitride Probe).
Figure 2.2 SEM micrograph of a corner tip, fabricated by a silicon-on-insulator technology (MikroMasch® Opus™ AFM probe).
Figure 2.3 SEM micrograph of an etched silicon AFM probe (NANOSENSORS™ PointProbe® Plus).
Figure 2.4 High-resolution SEM image of a platinum-coated tip (NANOSENSORS™ Electrostatic Force Microscopy probe).
Figure 2.5 SEM micrograph of a CVD diamond-coated tip (NANOSENSORS™ Conductive Diamond Tips).
Figure 2.6 AFM probe with extreme cantilever bending.
Figure 2.7 SEM micrograph of a full-diamond AFM probe.
Figure 2.8 Metal tip on a silicon cantilever [24].
Figure 2.9 Diamond tip on a silicon cantilever [26].
Figure 2.10 High-resolution SEM image of a NANOSENSORS™ platinum silicide probe.
Figure 2.11 Direct comparison of tip apices of different C-AFM probes. (a) Diamond-coated silicon, (b) platinum-coated silicon, and (c) platinum silicide. All images are captured with identical magnifications. The scale bar size is 100 nm.
Figure 2.12 Comparison of tip wear. Contact-mode scanning with a platinum tip with (a) 200 nN, (b) 500 nN, and (c) 1 µN interaction force. Overlay of SEM images before and after the scan.
Figure 2.13 Comparison of tip wear. Contact-mode scanning with a (a) platinum tip, (b) a platinum silicide tip, and (c) a conductive diamond tip with an interaction force of 500 nN. Overlay/comparison of SEM images before and after the scan.
Figure 2.14 Cross sections of current images of (a) a Pt probe, (b) a PtSi probe, and (c) a conductive diamond probe. Sample bias −0.25 V, force 1 µN, tip travel distance 13 mm. For each tip is shown a cross section out of the first and the last scan.
Figure 2.15 Platinum-coated tip apex after a series of voltage rampings with a maximum current of about 0.5 µA.
Figure 2.18 Platinum-coated tip apex after a series of voltage rampings with a maximum current of about 10 mA.
Figure 2.16 Platinum silicide-coated tip apex after a series of voltage rampings with a maximum current of about 0.5 µA.
Figure 2.17 Platinum-coated tip apex after a series of voltage rampings with a maximum current of about 1 mA.
Figure 2.19 (a) AFM tapping-mode image of a silicon surface which was previously scanned in contact mode with a diamond-coated probe with a 10 µN interaction force. (b) Cross section of scanned line.
Chapter 3: Fundamentals of CAFM Operation Modes
Figure 3.1 Measurement setup for conductive samples (a) with topography section (b). Features on position
x
2
or
x
3
result in a higher or lower current and position
x
4
showing a varying current due to particles (c). Measurement setup for measuring thin insulating layers (d) with topography section (e). Filament formation resulting in a high current at
x
1
, conductive inclusion at
x
2
, lower current due to higher insulator film thickness at
x
3
, and higher current due to lower insulator film thickness at
x
4
(f).
Figure 3.2 Topography images of identical areas of 3 × 1.5 µm
2
obtained in intermittent contact mode with high aspect ratio nonconductive tip (a) and in contact mode with low aspect ratio conductive tip (b) showing differences in spatial resolution. The comparison of the topography images (a or b) with the current mapping (c) reveals nanoscale features of the electrical conductivity that are not visible in the topography images.
Figure 3.3 Schematic of a hertzian contact between an AFM probe tip of radius and a semi-infinite sample, following Bhushan [9] and Olbrich [10].
Figure 3.4 Calculated extreme values of the contact area versus the contact force for SiO
2
substrate (a) and Si substrate (b), respectively. Tip material and tip radius are indicated. The dashed grid is for orientation. Material parameters, typical values: platinum, GPa, ; diamond, GPa, ; thermal SiO
2
, GPa, ; silicon, GPa, .
Figure 3.5 Schematic energy-band diagrams of metal-semiconductor structures, n-type (left), p-type (right): (a) thermal equilibrium. (b) Forward bias. (c) Reverse bias. (Adopted from Ref. [26].)
Figure 3.6 Water capillary bridge at the interface between a pyramid-shaped AFM probe tip, as observed in an environmental scanning electron microscope (ESEM). (Reprinted from Ref. [39].)
Figure 3.7 Scanning electron micrographs of CoCr-coated tips in the initial state (a) and after approximately 10 sequences of 25
I–V
measurements each (b). The tip radius increased owing to wear from approximately 50 to 300 nm. Severe wear may lead to fatal damage (c).
Figure 3.8 Different fundamental conduction systems. MOS (MIS)-structures: (a) direct tunneling, (b) Fowler-Nordheim tunneling, (c) thermionic emission, and (d) Frenkel-Poole emission.
Figure 3.9 Offset voltage as a function of substrate dopant density for different tip materials for p-type Si (a) (Eq. (3.10)) and n-type Si (b) (Eq. (11)) for substrate emission. In addition, the values for highly doped () poly-Si gates, n
+
and p
+
, are provided. Material parameters [10, 26, 54, 55], typical values: , , , , , , , , , , .
Figure 3.10 Topography image (tapping mode) of a 1.3 nm SiO
2
/n-Si sample subject to CAFM stress. The center area was stressed by V, 0,5 Hz scan rate. The surface protrusions in the center are a result of local
I–V
measurements ( V, 0.3 Hz ramp rate). CAFM: Cont-Pt (PtIr5) tip, . Tapping mode: (uncoated Si) tip, .
Figure 3.11 Schematic of a time-dependent force curve for one complete PF cycle from
A
to
E
(a). Simplified current reading for the contact current (b) representing the current while the tip is in physical contact with the surface between point
B
and
D
. The peak current (c) is recorded at the maximum force at point
C
.
Figure 3.12 Measurement setup for TR CAFM (a) and the front view of the cantilever and tip with indicated torsional movement (b). As shown schematically, the sample can be conductive or consist of a thin insulating surface layer.
Figure 3.13 (a) Topography and (b) simultaneously measured current image obtained at the 6H-SiC sample irradiated at the He
+
ion fluence of . The 3.0 mV tip voltage is retained for the CAFM measurement. (c) Local
I–V
characteristics of nonirradiated and He-ion irradiated () 6H-SiC samples. (d) The values for nonirradiated 6H-SiC sample as a function of applied voltage. The straight lines represent the Schottky emission fits to the experimental data. The intersection with the
y
-axis gives the value for the saturation current density . These fits lead to the Schottky barrier height () of .
Figure 3.14 Topography using the PF mode for an area of 1 × 0.5 µm
2
(a) and current mappings of sample A for different operation modes: contact mode (b), TR mode (c), and PF mode (d). In this case study, the contact mode and PF CAFM deliver the best current mappings, whereas in the TR mode, currents on the surface pits locations could not be resolved.
Figure 3.15 Local
I–V
characteristics on sample A in forward bias taken on and off surface pits. The dotted line represents an arbitrary threshold current of 100 nA leading to a turn-on voltage shift of about 400 mV between
V
th,on
and
V
th,off
.
Figure 3.16 Superposition of a 5 × 5 µm
2
IC-AFM topography (a) on sample B and indicated surface pits with a CAFM current mapping (b) at 22 V forward bias.
Figure 3.17 Topography image of area 1 × 0.5 µm
2
. (a) Blurred picture of the surface (contact mode). current mapping (b) containing information that cannot be assigned to the topographical features. The
I–V
characteristics exhibit different conduction regimes (c). Most of the characteristics show an ohmic behavior (blue and yellow arrows/characteristics), whereas on low conductive areas, a nonlinear characteristic occurs (red arrow/characteristics).
Figure 3.18 Topography of an area of 1 × 0.5 µm
2
(a). Current mappings for different contact forces reaching from 600 nN (b), 2.2 µN (c), and 5.5 µN (d). The tip voltage was 20 mV. Low forces lead to a depiction of the immediate surface structures with a very high lateral resolution. At forces above 1 µN (c and d) a germanium phase transition beneath the tip leads to a significant resistance drop and the meander structure underneath the immediate surface structures dominates the current map.
Chapter 4: Investigation of High-k Dielectric Stacks by C-AFM: Advantages, Limitations, and Possible Applications
Figure 4.1 Comparison of macroscopic and nanoscale current characteristics. (a) Macroscopic current density voltage curves for TiO
2
MIM structures using a TiCl
4
and a TTIP (Ti-tetra-isopropoxide) precursor, respectively. The symbols indicate the points of the curves where C-AFM maps with the same applied bias are shown in (b). (b) Leakage current distribution maps of samples TiO
2
(TiCl
4
) and TiO
2
(TTIP) with various applied bias voltages (in the top row, maps are shown for a voltage of 2.0 V). Note the different current scales. The scan area of all maps is 2 × 0.5 µm
2
[30].
Figure 4.2 Comparison of conventional
I–V
characteristics with
I–V
curves measured using C-AFM on samples with SiO
2
-gate dielectric on p-type silicon; all measurements were performed on MIS capacitor structures.
Figure 4.3 (a) Comparison of
J–V
characteristics as obtained from standard C-AFM
I–V
measurements (tip in direct contact with the SiO
2
film) and characteristics from conventional and C-AFM measurements on MIS test structures from Figure 4.2. (Reprinted from [32], Copyright 2009, with permission from Elsevier). (b) Comparison of
J–V
characteristics for a 4 nm SiO
2
layer on p-type silicon with MIS capacitor test structures of different gate areas. Conventional: conventional probe station and semiconductor parameter analyzer; C-AFM tip on metal: C-AFM tip on metal gate and C-AFM current amplifier; C-AFM tip on dielectric: C-AFM tip on dielectric and C-AFM current amplifier.
Figure 4.4 Comparison of conventional
J–V
characteristics with
J–V
curves measured using C-AFM on samples with a HfO
2
/SiO
2
-gate dielectric; solid lines represent measurements performed on MIS capacitor structures, dotted lines represent measurements with tip in direct contact with the HfO
2
film.
Figure 4.5 Local C-AFM
I–V
curves at the edge of a thermally oxidized silicon sample (SiO
2
thickness is about 5 nm) using (a) standard probes and (b) shielded coplanar probes with different sweep rates
sr
(as indicated in the figure); negative (positive) currents were measured in forward (backward) direction with, for example, forward direction from 0 to −6 V.
Figure 4.6
I
disp
in dependence on
sr
. Symbols: evaluated data from measurements shown in Figure 4.5; lines: corresponding regression fits with the resulting effective capacitance and offset current
I
0
, respectively.
Figure 4.7 Local C-AFM
I–V
curves at the center of the same sample as used for the measurements shown in Figure 4.5 using different sweep rates
sr
(as indicated in the figure) and both standard probes and shielded coplanar probes. Symbols are only used for indication purposes.
Figure 4.8 (a) Forward
I–V
characteristics of the same thermally oxidized silicon sample as that in Figure 4.5 for different sweep rates as indicated in the Figure (always at least eight representative characteristics are shown) using shielded probes. (b)
I–V
characteristics of a sample with a 10 nm HfO
2
ALD layer for 0.2 and 4 V/s using shielded probes.
Figure 4.9 (a) Forward
I–V
characteristics of the same thermally oxidized silicon sample as that in Figure 4.5 for different sweep rates and voltage ranges as indicated using shielded probes (the data for voltages up to −10 V is the same as shown in Figure 4.8a). Symbols are only used for indication purposes. (b) Forward
I–V
characteristics shown in (a) but corrected for both
I
disp
and
I
offset
using the proposed method described in the text.
Figure 4.10 (a) Forward
I–V
characteristics as in Figure 4.9b but presented in logarithmic current scale; (b) same data as in (a) but with reduced noise by applying a simple moving-average filter using 25 data points
Figure 4.11 (a) TUNA topography map (rms roughness of about 0.5 nm) and (b) the corresponding TUNA current map of nanocrystalline HfSi
x
O
y
samples, scan size: 2 × 1 µm
2
, diamond-coated silicon tip, applied bias voltage: −8.2 V, topography scale: 0 nm (dark) −7 nm (bright), current scale: 0 pA (bright) to −5 pA (dark).
Figure 4.12 Topography map of the ZrO
2
sample annealed at 450 °C for 30 s, AFM operated in tapping mode, scan size: 500 × 500 nm
2
; rms roughness: 0.2 nm.
Figure 4.13 TUNA current maps of ZrO
2
samples: (a) as-deposited, (b) annealed at 450 °C for 30 s, and (c) annealed at 450 °C for 1 min, scan size: 2 × 0.5 µm
2
, Pt/Ir coated silicon tip, applied bias voltage: −2.7 V, current scale: 0 pA (bright) to −5 pA (dark).
Figure 4.14 HR-TEM images of: (a) as-deposited, (b) RTA at 450 °C, 30 s, and (c) RTA at 600 °C, 1 min samples. The images clearly show an increase in SiO
2
interfacial layer thickness and crystallization of ZrO
2
as a result of RTA.
Figure 4.15 Correlation between (a) topography and (b) leakage current distribution (applied bias voltage of −0.9 V, measured with 1 nA/V sensitivity) of TiO
2
(TiCl
4
) sample. Scan area is 2 × 0.5 µm
2
. On the right, a zoom-in image of the leakage current map is presented (as indicated by the dashed frame in the original map). Note that the current scale in (b) includes positive currents to improve visibility. Actual currents do not exceed 0 A. The vertical lines point to corresponding areas in the topography and current maps [30].
Figure 4.16 (a) TEM cross-section image of sample TiO
2
(TiCl
4
) on the polycrystalline RuO
2
layer grown by MOCVD. (b) Explanatory sketch: the TiO
2
is locally thicker over the RuO
2
grain boundaries and thinner over the RuO
2
grain tips with respect to the smoothening effect [30].
Figure 4.17 (a) TUNA
I–V
curves of as-deposited and at different conditions annealed ZrO
2
samples. In the inset, the
I–V
curves of the samples are represented in Fowler–Nordheim coordinates. The Fowler–Nordheim plot reveals the existence of two conduction mechanisms, which are also schematically presented in the Figure (b) Fowler–Nordheim plot (ln(
I
/
V
s
2
) vs 1/
V
s
) of
I–V
curves in the higher voltage region (
V
g
> 4.7 V). Here, the voltage drop
V
s
across SiO
2
is used to calculate the Fowler–Nordheim coordinates, as the current in this region is dominated by tunneling through the thin interfacial SiO
2
layer.
Figure 4.18
I–V
hysteresis of as-deposited and annealed (450 °C, 1 min) samples.
Figure 4.19
I–V
hysteresis of the sample annealed at 600 °C, 1 min with different end voltages
V
end
: (a) 3.7, (b) 4.0, and (c) 4.2 V. The curves indicate different trapping phenomena resulting in different hysteresis (positive, zero, or negative). These measurements were performed with Pt/Ir tips.
Figure 4.20 Evolution of
I–V
curves of the sample annealed at 600 °C, 1 min with successive measurements at one point: (a) forward
I–V
curves, (b) backward
I–V
curves, and (c) comparison of hysteresis of the 1st and 100th
I–V
curves.
Figure 4.21 Hysteresis (voltage shift between forward and backward TUNA
I–V
curve at 5 pA) after several measurement repetitions for 8 nm samples
Z
,
ZS
1, and
ZS
2.
Figure 4.22 Comparison of conventional and C-AFM-based
I–V
measurements for Si
3
N
4
-type sample; different gate contact areas and tip directly on dielectric were used – destructive dielectric breakdown is indicated by bold circles.
Figure 4.23 Weibull plot for 100 measurements on a SiO
2
/HfO
2
sample and extraction of time-to-failure 63rd percentile and Weibull slope parameter.
Figure 4.24 (a) Time-to-failure plots (assuming Weibull distribution) of different samples for CVS measurement on gate contacts and directly on dielectric, respectively. (b) Measurement data for an 8 nm SiO
2
sample with compensation for different work-function materials between gate electrode and C-AFM tip; the respective slopes from voltage acceleration using Eq. (4.9) are listed.
Figure 4.25 Evaluation of measurement data for SiO
2
sample; voltage and area extrapolation were performed.
Figure 4.26 Evaluation of measurement data for (a) SiO
2
/Si
3
N
4
and (b) SiO
2
/HfO
2
sample; voltage and area extrapolation were performed.
Figure 4.27 (a) TDDB distributions obtained at different temperatures. A dashed line outlines the bimodal character of the distributions. (b) Cumulative distribution of the initial leakage currents,
I
(
t
= 0 s). Two types of locations with low (G) and high (GB) currents can be discerned.
Figure 4.28 Example of TDDB Weibull distributions obtained at 60 °C, at G (squares) and GB (circles) sites.
Figure 4.29 Experimental (filled symbols) and modeled (open symbols) values of β (a) and η (b). Squares represent G and circles GB sites (in (a), β was assumed to be the same at both site types). The lines in (b) show the Arrhenius dependency of the η values.
E
a
is the extracted activation energy for the G and GB data sets.
Figure 4.30 TDDB distributions at 60 °C (circles) and 210 °C (squares) fitted using the Eq. (4.11) (continuous line). Broken lines outline the bimodal character of the distributions.
Chapter 5: Characterization of Grain Boundaries in Polycrystalline HfO2 Dielectrics
Figure 5.1 Simultaneously collected topography-current images on the surface of (a) as-deposited and (b) 1000 °C annealed HfO
2
(2.5 nm)/SiO
2
(1 nm)/Si stack. The images clearly show changes in structural and electrical properties of the HfO
2
dielectric.
Figure 5.2 TEM micrographs of (a) as-deposited HfO
2
and (b) 600 °C annealed HfO
2
on Si substrate with SiO
x
as IL.
Figure 5.3 (a) CAFM topography and (b) corresponding CAFM current map (at +5 V) of the HfO
2
gate dielectric depicting the grain and GB profile. (c) Line profile of topography and current map along line ‘
X
1
–Y
2
’ shown in (a) and (b).
Figure 5.4 Evolution of current through the grain (G_IV) and GB (GB_IV) during sequential RVS. The maximum current measured by CAFM is 2 nA. The first
I–V
curves (IV1) were taken on a fresh site.
Figure 5.5 Evolution of current through the grain (G_IV) and GB (GB_IV) during sequential RVS using E-CAFM. The maximum current measured by E-CAFM is 1 mA. The first
I–V
curves (IV1) were taken on a fresh site.
Figure 5.6 CAFM leakage current evolution with stress time under constant voltage stress of +6 V at different locations on the dielectric. Note the CAFM current saturates at 1 nA in these experiments.
Figure 5.7 Weibull plot of TDDB for CAFM-induced BD locations (∼20) in the HfO
2
/SiO
x
stack shown in Figure 5.6.
Figure 5.8 CAFM topography map of a HfO
2
/SiO
x
dielectric stack after (a) first scan and (b) second scan at +6 V sample bias. (c, d) Show the corresponding current maps after the first and second scans.
Figure 5.9 Line profile of current map along the line “
X
1
–
X
2
” after the first and second scans shown in Figure 5.8c,d.
Figure 5.10 (a) CAFM topography and leakage current profile at +6 V after the (b) first scan and (c) second scan. (d) Line profile of topography and current along the line
X–Y
shown in a, b and c.
Figure 5.11 Simulated cross-sectional electric field distribution in HfO
2
/SiO
x
under different breakdown phases; (a) fresh device, (b) after SiO
x
IL SBD, and (c) subsequent SiO
x
IL and HK BD.
Chapter 6: CAFM Studies on Individual GeSi Quantum Dots and Quantum Rings
Figure 6.1 Topographic (a) and current (b) images of Ge QDs grown at 550 °C measured at a sample bias voltage of 1.0 V. (c) Respective cross-sectional profiles along the solid lines marked in (a, b).
Figure 6.2 Topographic (a) and current (b) images of Ge QDs grown at 640 °C obtained at 1.4 V. (c) Respective cross-sectional profiles along the solid lines marked in (a, b).
Figure 6.3 Topographic images of Ge QDs grown at 550 °C (a) and 640 °C (b) after etching in 30% H
2
O
2
solution for 30 min.
Figure 6.4 Topographic and current images of GeSi QDs without Si capping (a, b), after 0.32 nm Si capping (c, d), and after 2 nm Si capping (e, f).
Figure 6.5 Topographic and Ge content images of GeSi QDs before capping (a, b), after 0.32 nm Si capping (c, d), and after 2 nm capping (e, f), respectively.
Figure 6.6 Topography (a) and current (b) images of GeSi QRs after etching by HF for 30 s at a sample bias of −1 V. The topographic and current profiles of an individual QR as marked in (a) and (b) are shown in (c).
Figure 6.7 Height image measured simultaneously with the CPD image before (a) and after BPA etching (e). CPD image obtained at an AC modulation of 2 V and a lift height of 10 nm before (b) and after BPA etching (f). Current image measured at a sample bias of −1 V before (c) and after BPA etching (g). d
C
/d
V
amplitude image obtained by applying 2 V AC modulation to the sample before (d) and after BPA etching (h).
Figure 6.8 (a)
I–V
characteristics of the QD under different normal forces: experimental data (symbols) and FNT fitting results (solid lines). (b)
I–V
curves obtained as a function of exposure time after etching with two fitted conductance models. The dotted lines are experimental data and the solid lines are fitted data. S1, S2, and S3 correspond to Schottky emission fitting at 30, 45, and 60 min after etching and F1, F2, and F3 correspond to FNT fitting at 150, 240, and 360 min after etching, respectively.
Figure 6.9 Topographic (a–d), current (e–h) images of GeSi QDs, and current profiles along the marked lines (i–l) measured at different biases from −0.5 to −2 V, respectively. All image sizes are 0.5 µm × 0.5 µm.
Figure 6.10 The topographic (a–d), current (e–h) images of GeSi QDs, and current profiles along the marked lines (i–l) measured at different biases from −2.5 to −4 V. All image sizes are 0.5 µm × 0.5 µm.
Figure 6.11 Topography and current images of SL sample A (a, b) and BL sample B (d, e) with low dot densities. The height and current profiles of the marked small QDs of samples A and B are plotted in (c, f), respectively.
Figure 6.12 Topography and current images of SL sample C (a, b) and BL sample D (d, e) with high dot densities. The height and current profiles of the marked small QDs of samples C and D are plotted in (c, f), respectively.
Figure 6.13 The obtained average current as a function of bias voltage. (a) Large QDs of samples A/B, (b) small QDs of samples A/B, (c) large QDs of samples C/D, and (d) small QDs of samples C/D.
Figure 6.14 Topography images of GeSi QRs before (a) and after etching in NHH for 10 s (b), 60 s (c) and 600 s (d) and in BPA for 90 s (e). (f) The height profiles of the same QR along the same marked line before and after each etching step. The curves are vertically offset by subtracting the etched height of the WL in the corresponding etching step.
Figure 6.15 Topography (a) and surface composition distribution (b) of the same area including a single QR. The two dashed circles indicate the inner and outer rings of the QR. (c) Cross-sectional topographic change before and after each etching process. (d) A schematic diagram of the vertical composition distribution of single GeSi QRs.
Figure 6.16 The topography and current distributions of GeSi QRs before (a, f) and after etching in NHH for 10 s (b, g), 60 s (c, h) and 600 s (d, i) and in BPA for 90 s (e, j). The sample was biased at −1.0 V and the tip was grounded. (I), (II), and (III) give the height, composition, and current profiles of the same QR along the marked line, respectively. The curves in (I), (II), and (III) are offset to various values for clarity.
Chapter 7: Conductive Atomic Force Microscopy of Two-Dimensional Electron Systems: From AlGaN/GaN Heterostructures to Graphene and MoS2
Figure 7.1 (a) Schematic representation of the measurement setup for CAFM current mapping on the AlGaN surface. AFM topography (b) and the corresponding current map (d) on the LD sample. (c) Cross-section TEM image of the LD sample. AFM topography (e) and corresponding current map (g) on the HD sample. (c) Cross-section TEM image of the HD sample in which a V-shaped defect is visible. (h) Specific contact resistance
R
c
as a function of the temperature
T
for Ti/Al Ohmic contact formed on the HD sample (after annealing at 500 °C) and on the LD sample (after annealing at 800 °C). Cross-section TEM micrographs of annealed Ti/Al bilayers on (i) LD sample and (j) HD sample.
Figure 7.2 (a) Cross-section TEM analysis of the as-grown AlGaN/GaN heterostructure and high-resolution image of the AlGaN layer (insert). Surface morphology (b) and CAFM current map (c) of the as-grown sample. The white circles indicate the morphological surface defects and the corresponding current spots. (d) Cross-section TEM analysis of the oxidized sample. A high-resolution image of the AlGaN layer is reported in the insert on the top and the oxygen chemical map in the proximity of a V-defect is reported in the bottom insert. Surface morphology (e) and CAFM current map (f) of the oxidized sample.
Figure 7.3 (a) Forward
I–V
characteristics at different temperatures of the NiO/AlGaN/GaN MIS diode (see schematic in the insert). (b) Poole–Frenkel plot of the measured currents. (c) Schematic of the experimental setup for CAFM characterization of the NiO thin film. Surface morphology of the NiO film (d) and two-dimensional current map (e) collected for an electric field of 1 MV/cm applied to the dielectric.
Figure 7.4 (a) Typical AFM morphology and (b) height line profile of as-grown epitaxial Gr on nominally on-axis 4H-SiC(0001). (c) Typical micro-Raman spectrum on the same sample.
Figure 7.5 (a) Schematics of the experimental setup for CAFM measurements on epitaxial Gr on 4H-SiC (0001). (b) Representation of the different contributions to the measured resistance
R
. Surface morphology (c), current map (d), height (e), and local resistance (f) line profiles for 1L Gr over ∼1.5-nm-high step of SiC substrate. Surface morphology (g), current map (h), height (i), and local resistance (j) line profiles in a sample region with local inhomogeneity in the number of Gr layers.
Figure 7.6 (a) Schematics of the experimental setup for CAFM measurements on MoS
2
and (b) equivalent circuit. (c) Set of 25
I–V
tip
characteristics measured on a 500 nm × 500 nm array of tip positions with ∼100 nm spacing on MoS
2
. (d) Representative forward bias
I–V
tip
characteristic from this set of measurements and fit with the thermionic emission law to extract the SBH and ideality factor. (e)
H
function plot for the determination of the series resistance
R
. Histograms of the local SBH Φ
B
(f), the ideality factors
n
(g), and resistivities ρ
loc
(h) extracted from the full set of
I–V
tip
characteristics reported in panel (c). (i) Plot of ρ
loc
versus Φ
B
.
Figure 7.7
I–V
curves on the reference AlGaN/GaN samples LD (a) and HD (b) and on the Gr-coated samples LD (c) and HD (d). Representative AFM analyses of the topography of these samples are reported in the inserts.
Chapter 8: Nanoscale Three-Dimensional Characterization with Scalpel SPM
Figure 8.1 Schematic of the progressive evolution of chip architecture from 2D to 3D (based on international technology roadmap for semiconductors ITRS) [56].
Figure 8.2 SPM approaches adding the third dimension. (a) Alternating chemical etching and SPM sensing. (b) Physical removal by ion beam or microtome followed by SPM sensing. (c) Dedicated test structure creating a pseudo-3D volume (see text).
Figure 8.3 Tip-induced material removal. (a) Crater formation by material removal on a blanket oxide sample (3-nm-thick HfO
2
) using successive scans with a diamond tip. The diamond tip is scanned over 500 × 500 nm
2
at high force (≈µN). A topographic cross section (see insert) through the crater reveals a removal of ≈200 pm. (b) By repeating the number of scans progressively a higher amount of material is removed. This is seen in the cross section reported in (b); a removal rate of ≈200 pm/scan is found with a total removal of 1 nm HfO
2
after six scans.
Figure 8.4 Tip-induced material removal. (a) Removal of 30-nm-thick Cu electrode with end point control on the layer (TiN) underneath. (b) Circuit editing for failure analysis by interrupting an interconnect line. The insert shows a topographical scan across the line cut.
Figure 8.5 SPM tomography procedure. The wear-resistant diamond tip is scanned on the surface at high pressure (≈GPa) inducing a controlled material removal. The acquired 2D images are aligned and interpolated by a dedicated software to generate a 3D tomogram.
Figure 8.6 Tip comparison and practical implementation. (a) SEM and TEM images of full diamond tip and (b) commercially available diamond-coated Si tip. (c) SEM observation of a pristine versus a used tip showing residues of the removed material on the tip body. (d) Maximum probing depth as determined by the angle formed between the tip apex and the sample surface (tip scale bar 20 nm).
Figure 8.7 Structure of the memory elements and their resistive switching characteristics. (a) Schematic of the crossbar device. High-resolution TEM images for CBRAM (top) and VCM (bottom) material systems. (b) Bipolar resistive switching characteristics, one representative cycle is reported for CBRAM. (c) VCM operated at 50 µA current compliance.
Figure 8.8 (a) Planar 2D C-AFM, performed on a cross-point memory element in SET-state. The C-AFM is completely ineffective due to the presence of the TE shielding the local conductivity changes linked to the CF (scale bar 100 nm). (b) Schematic of the C-AFM tomography procedure: the scraping with the diamond tip creates multiple C-AFM maps taken at different heights of the CF after the removal of the TE. (c) 3D stacking of the collected 2D C-AFM slices. Note the average spacing between each slice is ≈0.5 nm. (d) Collection of 2D slices constituting the dataset for the 3D interpolation (scale bar 80 nm). The CF appears in the middle of the active area after top electrode removal. The highly conductive features on the top-left and bottom-right corners are the exposed parts of the TiN BE, which is progressively exposed during the removal of Al
2
O
3
. (e) Observation of the reconstructed 3D tomogram for the CF in a 5-nm-thick Al
2
O
3
(Cu electrode was on top and TiN bottom). (f) CF observation (blue shape) by volume rendering within a 5-nm-thick HfO
2
. The Hf electrode was on top and the TiN at the bottom. In both figures, the low current contribution in the tomogram is suppressed to enhance the contrast of the highly conductive features.
Figure 8.9 Illustration of the electrochemical processes during resistive switching. (a) CBRAM: (1) The Cu oxidizes and Cu
+
ions are injected in the Al
2
O
3
. The high electric field might also lead to the formation of oxygen vacancies in the dielectric layers (white balls in the cartoon). (2) The slow migration of Cu
+
ions in the switching layer implies that a reduction reaction occurs before the Cu
+
reaches the inert electrode. (3) The filament growth continues and the CF eventually shorts the two electrodes, thereby creating the low resistive state. (b) VCM: the oxygen atoms start to leave their lattice position and drift toward the anode, leaving behind a local conductive path that joins the sub-stoichiometric
V
o
⋅⋅
reservoir. On each subsequent memory cycle, the number of
V
o
⋅⋅
will be reshuffled at the constriction of the CF inducing the resistance change and flipping of the bit-state.
Figure 8.10 (a) Current maps of two representative CFs; note the different sizes/conductivity as measured with the same conditions (scale bar 20 nm). (b) Comparison between CFs' sizes observed at the top electrode interface, for cation-versus anion-based devices. In the inset we show the schematic of the final shape difference induced by the different switching mechanisms in CBRAM and VCM.
Figure 8.11 Analysis of a device with variable switching. (a) Resistance state fluctuation observed in consecutive reset traces of CBRAM. Note that the multiple LRS and reset transitions observed suggest that different CFs are alternating. (b) Cross sections extracted from 3D tomogram, (top) two cross-section showing two filaments. (middle) CF-size at the top-electrode interface (right). 3D section through CF shortening the electrodes. (c) Proposed model for the two-filament instability. During the forming process, CF
1
and CF
2
attempted to grow in the Al
2
O
3
. Finally, CF
1
connects and its morphology changes during cycles. Due to the structural rearrangements in the CF
1
driven by the subsequent reset, on the third cycle CF
2
becomes the favored CF for the next formation, leading to a resistance state variability associated with the potential fluctuation between the two CFs.
Figure 8.12 Failure analysis of a device stuck in LRS. (a) Endurance failure with resistive window collapse due to stuck in LRS. (b) 2D current maps of the device under test. Note that after the top electrode removal the CF appears as a large conductive platelet (scale bar 50 nm). (c) Surface section and cross section extracted 3D tomogram of a failed device, showing the CF degenerated into a conductive wall. The latter shorts the bottom and top electrodes and runs through the entire switching layer. (d) Schematic of the proposed model for the stuck in LRS. Three original CFs join together into one platelet during cycling.
Chapter 9: Conductive Atomic Force Microscopy for Nanolithography Based on Local Anodic Oxidation
Figure 9.1 Four possible approaches to scanning probe lithography and respective examples of patterns: (a) STM displacement of atoms on a surface, which can generate patterns such as (b) a “quantum corral” ring of 48 Fe atoms on Cu. (Reprinted with permission from Ref. [3]). (c) Dip-pen nanolithography, which can direct the deposition of SAMs such as (d) 70-nm-wide features of 16-mercaptohexadecanoic acid on Ag, to be used as an etch resist. (e) Nanoshaving removal of SAM regions, which can pattern features such as (f) a square hole within octadecanethiolate SAMs on Au. (g) Local anodic oxidation with a carbon-nanotube-modified AFM tip, which can selectively oxidize a surface, making patterns such as (h) 10-nm-wide (2-nm-tall) silicon oxide lines spaced by 100 nm [12]. Adapted from Ref. [4]. Copyright 2005 American Chemical Society.
Figure 9.2 (a) AFM images of a silicon nanowire connected to platinum electrodes; the height profile of the nanowire is also reported. Adapted from Ref. [15]. Copyright 2010 IOP Publishing, LDT. (b) The AFM height image of a single-electron transistor with oxide tunneling barriers. Between source (S) and drain (D) an island is defined by two oxide lines (light lines). Side gates (G) have been mechanically machined at the sides of the island (dark lines). Adapted from Ref. [16]. Copyright 2000 AIP Publishing.
Figure 9.3 (a) Schematics of the LAO process on Si with an AFM operating in contact mode. The feature fabricated on Si(100) is the oxide pattern, the corresponding AFM topography and height profile are also reported. (b) The AFM topography of oxide squares fabricated on silicon carbide before and after a wet etching in HF. The oxide protrusions are readily etched, as expected for SiO
2
. (c) The height profile before and after the etching; from this kind of experiment it is possible to calculate the volume mismatch between the pristine material (dotted line) and the newly formed oxide.
Figure 9.4 (a) The LAO mechanism scheme in detail. (b) 3D AFM topography of oxide dots fabricated on silicon carbide at various pulse times/bias. Note that the maximum oxide height on such substrates is much higher than the few nanometers reported for silicon. In (c) we report the fitting of LAO data (oxide height vs log of exposition time) to the “two-path” model as reported in Ref. [31]. (Panel (c) adapted from Ref. [31]. Copyright 2006 WILEY-VCH Verlag GmbH & Co.)
Figure 9.5 (a) AFM height image of oxide lines fabricated at a constant speed of 1.0 µm s
−1
and voltage
V
ox
= 7 and 8 V for the vertical and horizontal lines, respectively. (b) Detected current measured during the writing process; during LAO in contact mode at constant speed, current is stable and constant. Peaks/spikes in current are produced at the beginning of each line, while instant drop in current happens when the tip crosses an oxide line previously fabricated. (Adapted from Ref. [41] Copyright 2004 IOP Publishing, LDT.) (c) Time evolution of the total instant current during a scanning probe oxidation experiment (see Ref. [42]).
Chapter 10: Combination of Semiconductor Parameter Analyzer and Conductive Atomic Force Microscope for Advanced Nanoelectronic Characterization
Figure 10.1 Schematic of the probe station setup used for standard device level characterization.
Figure 10.2 Typical (a)
I
D
–V
D
, (b)
I
D
–V
G
, and (c)
I
G
–V
G
characteristics of a MOSFET recorded before (continuous line) and after NBTI (open symbols) and CHC (full symbols) stress. (d) Breakdown statistics from which
V
BD
(63%) was estimated.
Figure 10.3 AFM 3-D topographic image of a pMOSFET after the removal of the gate electrode, showing a very flat surface (RMS < 0.2 nm) between the two spacers. Placing the CAFM tip on the gate oxide region, the conduction through the oxide can be studied with nanometer lateral resolution. The dimensions of the area under study and the applied polarization voltage are indicated.
Figure 10.4 Typical current images obtained at a gate voltage of 3.6 V in (a) nonstressed and (b) CHC-stressed MOSFETs. (c) Average current measured with the CAFM tip along the channel for the nonstressed (circles) and CHC (squares) stressed MOSFETs.
Figure 10.5 (a) Current maps of Hg/NiO/Pt nanocapacitors that have been fixed at HRS and LRS (after the Hg electrode removal). (Reproduced with permission from [31]. Copyright 2008 AIP Publishing LLC.) (b) High BD hardness. (Reproduced with permission from [32]. Copyright IEEE 2012.) (c) Current map of a polycrystalline HfO
2
stack annealed at 100 °C showing leakage current only at the grain boundaries. (Modified and reproduced with permission from [36]. Copyright 2010 AIP Publishing LLC.) (d) Three-dimensional characterization of CFs in RRAM cells using CAFM tomography. (Reproduced with permission from [33]. Copyright IEEE 2013.) (e) Current image of Rh-tip/Ni
1+δ
O/Pt film after set and reset process at specific locations. (Reproduced with permission from [34]. Copyright 2008 AIP Publishing LLC.) (f) Statistical analysis of the current spots measured with CAFM in RRAM cells. (Reprinted from [35]. Copyright 2012 IOP Publishing.) (g) Time-of-flight secondary ion mass spectroscopy (TOF-SIMS) depth profile and maps of
18
O ions in areas in HRS and LRS. The ability of the CAFM to scan large areas through current maps allows chemical characterization with tools of lower spatial resolution. (Reproduced with permission from [34]. Copyright 2008 AIP Publishing LLC.) (h) Example of RS behavior observed only at grain boundaries of polycrystalline HfO
2
films.
Figure 10.6 (a) Schematic structure of the sample under study. MIM structures with an HfO
2
layer as dielectric. Areas ranging from 2 × 2 µm
2
to 100 × 100 µm
2
. (b) Topographic map collected with the CAFM on the capacitor displayed in (a).
Figure 10.7 Sequences of
I–V
curves (a–f) collected in a TiN/HfO
2
/Pt device. The forming process and the currents before set and reset in HRS and LRS can be clearly observed.
Figure 10.8 Current images of (a) HRS and (b) LRS (with different scales) when scanning the capacitor with the tip of the CAFM by applying 1.5 V. The red dotted line outlines the capacitor.
Figure 10.9 Resistance measured in HRS and LRS for devices with different areas.
Figure 10.10 Voltage/current ranges that can be applied/measured with three different CAFM setups: (i) a standard setup (red dashed schematic), (ii) a standard setup working in vacuum, and (iii) CAFM working in vacuum and connected to an SPA. The part of the graph that each setup can measure has been highlighted with dashed lines of different colors.
Figure 10.11 (a) Nanoscale forming processes at 12 different positions of the HfO
2
layer of sample A. For clarity in this Figure we only show 12 of the 24 locations analyzed. (b) Typical bipolar RS behavior obtained only at those locations where the
V
BD
during the forming process was remarkable lower (red circles in panel a).
Chapter 11: Design and Fabrication of a Logarithmic Amplifier for Scanning Probe Microscopes to Allow Wide-Range Current Measurements
Figure 11.1 Block diagram of the logarithmic current-to-voltage preamplifier.
Figure 11.2 Simplified circuit for describing the working principle of the logarithmic preamplifier.
Figure 11.3 Circuit diagram of the logarithmic converter block.
Figure 11.4 Circuit diagram of the adder block.
Figure 11.5 Distribution of grounding planes in the logarithmic preamplifier.
Figure 11.6 Layout of the PCF of the LogAmpDev07 displaying rounded tracks.
Figure 11.7 Configuration for the characterization in the answer
I
IN
–
V
OUT
of the devices.
Figure 11.8 LogAmpDev07 response time with C2 = 100 pF for low-level input currents.
Figure 11.9 Undesirable operation using LogAmpDev07 without C2.
Figure 11.10 Answer speed depending on the input current with C2 = 220 pF. The response of the amplifier becomes slower than needed in the current range from 10 to 20 pA. On increasing the C2 capacitor, the preamplifier becomes slower when it works in the low current range.
Figure 11.11 Answer speed depending on the input current with C2 = 33 pF.
Figure 11.12 CAFM system configuration with a log
I–V
converter.
Figure 11.13 Input current to output voltage characteristic of the logarithmic
I–V
converter.
Figure 11.14 Current image (at 5.5 V) of an oxide area where an RVS was applied until BD. The color bar is in logarithmic scale. Different BD spots are observed. Spot 1 is related to the RVS and the others to the high voltage applied during the scan. The table shows the maximum current and area of the different BD spots registered in the image. Reprinted from [22].
Figure 11.15
I–V
curves (forward green, backward red) measured on the HfO
2
sample after being stressed with consecutive RVSs. Reprinted from [22].
Chapter 12: Enhanced Current Dynamic Range Using ResiScope™ and Soft-ResiScope™ AFM Modes
Figure 12.1 Scheme of the standard conductive AFM (C-AFM) setup. A conductive probe is used to scan the topography in contact mode. A DC bias is applied between the tip and the substrate so that an electric current
i
is induced. This current can be measured with a current-to-voltage converter placed in series with the tip-substrate. The current-to-voltage converter is based on a linear operational amplifier (OPAMP) circuit with a constant gain factor. Typical preset gain factors range from 0.1 to 10 A/V. Current ranges measured in a C-AFM experiment exceed quite often the range provided with a single linear amplifier setup.
Figure 12.2 Standard scheme of a logarithmic amplifier. The p–n junction diode is connected to the negative entrance of the OPAMP. As the current characteristic of the diode is exponential, the output voltage
V
s
has a logarithmic relationship with the input voltage
V
e
. An actual logarithmic amplifier requires more complex electronics to compensate temperature dependence effects of the diode.
Figure 12.3 Bimetallic effect of a chip with two Pt/Cr-coated cantilevers with different geometries. (a, b) Three-dimensional measurement of the deflections of the cantilevers at 22 °C (room temperature) and 26.3 °C, respectively. The color scales represent the deflection with respect to the clamped end of the cantilever and have been kept equal in (a) and (b) to show the negative bending with temperature increase of the cantilevers. (c) Photograph with a side view of the same two cantilevers. The inset in (c) shows a profile along the length of the longer cantilever at RT (black line) and 4.3 °C above RT (gray line), respectively. For clarity, the approximate position of the cantilever tip has been added to each profile. (d) Free-end deflection measurement of both cantilevers (red - short cantilever; black - long cantilever) during the temperature cycle. As the temperature increases, the differences in the thermal expansion coefficient of silicon and Pt/Cr metallic coating generates an extended force that bends the cantilevers downward. As the temperature is decreased again, the deflection is restored.
Figure 12.4 Schematics of the ResiScope™ setup for conductive measurements. A variable resistance is placed in series with the sample. The ResiScope™ module adjusts in real time the amplifier gain and the variable resistance according to the actual value of the resistance of the sample. Conductive measurements performed in this manner allow to minimize the actual current that flows through the sample and have a big dynamic range (100 fA–1 mA).
Figure 12.5 Schematics of the ResiScope™ and Soft-ResiScope™ modules. The ResiScope™ module has three main internal submodules. The “smart” analog module provides the gain selection for the HPA to measure the current and control the variable resistance in series with the sample. The digital module calculates the resistance of the sample and the theoretical current, provided that
V
bias
, the total current, and the variable resistance are known. The two magnitudes
R
and
I
(theoretical) can be monitored as auxiliary channels in AFM software. The software interface controls the bias and allows to monitor
R
and
I
(theoretical).
Figure 12.6 Comparison of conductive measurements on a gold substrate with graphene oxide domains. Topography and current images obtained in standard conductive-AFM mode (a, b). Topography and current images obtained in the ResiScope™ Mode (c, d). The current image shows unstable values of the measured current due to the lack of control in the current that flows through the tip-sample ensemble. On the contrary, the current image obtained in the ResiScope™ mode is stable over the entire scanned area. Additionally, the higher dynamic range allows to measure differences in conductivity on the graphene oxide regions that can be correlated with the thickness of the layer.
Figure 12.7 Resistance signal measured with ResiScope™ with a set of several resistors with known values. (a) Logaritmic plot of current versus time for a constant bias of 1 V when the resistor was switched from 10 to 50 GΩ every 250 ms. (b) Logaritmic plot of current versus bias when the resistor was switched from 10 to 50 GΩ while varying the bias from 0 to 10 V. It can be noted that no capacitance effects are seen in the current signal in the transitions between different resistances.
Figure 12.8 Different approaches to implement conductive measurements with intermittent mechanical contact with the sample. (a) Schematics illustrating the approach based on a sinusoidal regime, typically corresponding with a natural resonance frequency of the cantilever. The inset shows both the cantilever deflection and the electrical signal on a real experiment with this approach. (b) Cantilever deflection and electrical signal on an experiment based on the force spectroscopy curves (force volume mode) where the cantilever performs a quasi-static approach-retract cycle on the surface.
Figure 12.9 Schematics of the Soft-Resiscope™ mode concept. The cantilever position is modulated at scan rates of a few kilohertz. This modulation accomplishes the requirements of fast contact time and constant applied force during the electrical measurement as illustrated in Figure 12.8.
Figure 12.10 Comparison of the Soft-ResiScope™, ResiScope™, and oscillating modes on a PMMA soft sample. The top part (green dashed rectangle) corresponds to the area imaged in oscillating mode, where the absence of frictional forces does not significantly alter the polymer. The bottom part of the image (red dashed rectangle) has been imaged using the Soft-ResiScope™ mode and provides similar results to those of the oscillating mode. The middle part (blue dashed square) of the image shows a region imaged in contact mode (i.e., the standard ResiScope™ mode) where the polymer has been deformed owing to the frictional forces generated by the cantilever sweep. Image size: 2 µm.
Figure 12.11 Quantitative similarities of the electrical signal obtained in both ResiScope™ modes. (a, b) Topopgraphy and resistance images on an SRAM memory obtained in Soft-ResiScope™ (top) and standard ResiScope™ (bottom), respectively. The centered black dashed line limits the regions of the image where each one of them was used. (c) Resistance cross section along two equivalent parts of different unit bit cells. Other than the difference in the contact time between the two modes, quantitatively, the resistance values are similar, although the Soft-ResiScope™ mode shows better stability and lateral resolution due to the absence of frictional forces.
Figure 12.12 Soft-Resiscope™ mode characterization of an organic solar cell (a) and topography and resistance images on the active surface of the solar cell (b). The resistance image shows domains of higher resistance (intense blue regions) between lower resistance values (light blue domains). The lower resistance areas are correlated to regions where the organic coating partially covers the surface as can be seen in the topography image (the yellow areas correspond to the organic layer, brown areas to partially covered regions).
Chapter 13: Multiprobe Electrical Measurements without Optical Interference
Figure 13.1 The Nanonics Imaging MultiView 4000™ MultiProbe SPM System. Two probes are seen, each on an independent SPM scanner. In the inset image (upper left corner), two probes touching in a multiprobe configuration are shown.
Figure 13.2 In this image, the scanning probe images the stationary NSOM illumination probe showing that these probes can remain in contact even while one probe images a second one.
Figure 13.3 Close-up view of the Nanonics MultiView 4000 system. Two probes in close proximity are shown under an objective highlighting the open optical access, permitting easy viewing, and rapid alignment of multiple probes. A wide variety of high numerical aperture objectives can be used with the system.
Figure 13.4 (a) A drawing of a cylindrical piezo, routinely employed in AFM systems, showing why such piezos make optical integration challenging. (b) A drawing of a Nanonics Flat Scanner™ showing the four-piezo design. (c) Picture of Nanonics Flat Scanner™ design which permits both optical access and multiprobe operation. (d) A drawing of two standing cylindrical piezos showing it is impossible to bring the centers (probes) together. Picture and drawings courtesy of Nanonics Imaging.
Figure 13.5 A MultiView 4000™ MultiProbe SPM with four SPM probes.
Figure 13.6 SEM image of a glass AFM probe from Nanonics while controlled pressure is applied, emphasizing the glass probe strength and robustness.
Figure 13.7 An illustration of a traditional laser-based AFM feedback mechanism where a laser bounces off the probe's back and movement is monitored by a position sensitive detector (PSD).
Figure 13.8 Depiction of probe mounted on a tuning fork, where the tip is oscillated perpendicular to the sample allowing for “normal force” AFM operation.
Figure 13.9 (a) Multiprobe electrical measurement configuration where the tuning fork is rotated 90° and is parallel to the surface. An ideal geometry for multiprobe electrical measurements since the oscillation of the tuning fork is parallel to the surface allowing for a strong contact with the surface without damaging it. (b) Picture showing the two probes positioned on one gold line. Both images were taken through an optical microscope with a 10× objective.
Figure 13.10 Image of two multiprobe thermal probes. In such a configuration, one nanoheater probe can heat up the sample and the second nanothermocouple probe measures the thermal transport as a function of distance.
Figure 13.11 (a) Simultaneously obtained topographic and (b) thermal conductivity images on test sample of voids in a semiconductor chip provided by Sematech. (c) The line-scan across the thermal conductivity image demonstrates a lateral resolution of 30 nm.
Figure 13.12 Probes can be easily visualized by a high numerical aperture objective directly from above and brought into close distance to one another via software.
Figure 13.13 A scanning thermocouple probe measuring heat transport of gold electrodes on silicon. (a) The topography map where the red circle shows the position of the nanoheater probe imaged by the scanning thermocouple probe. (b) The temperature map. (c) A 3D collage of the temperature map overlayed on the topography. The red arrow highlights the nanoheater probe.
Figure 13.14 A variety of multiprobe scanning probe microscope platforms from Nanonics Imaging Ltd.: (a) The MultiView 4000™: a two-probe multiprobe room temperature system. (b) Multiprobe system integrated with Raman. (c) The MultiView 4000™ four-probe configuration with full optical integration from above and below. (d) High vacuum multiprobe system with up to four probes and optical integration. (e) The CryoView MP™ multiprobe cryogenic system featuring three SPM probes. (f) The CyroView MP™ multiprobe cryogenic system with full optical integration.
Figure 13.15 Multiprobe conductivity and Raman measurements of single-layer graphene devices with graphene bridges of variable widths: (a) Raman spectrum shows that the 2D band of the graphene is shifted as a function of the width and conductivity of the graphene channel. (b) Charge-coupled device (CCD) image showing the two electrical probes located on the electrical pads connecting the graphene channels. (c) A voltage of 1 V was applied between the electrical pads using Nanonics electrical probes. The measurements were done between every pair of the electrical pads connected with the graphene channels.
