An analysis of the surface potential of monocrystalline silicon solar cell using kelvin probe force microscopy

This paper investigates the effect of defects on the electrical properties of monocrystalline silicon solar cells (mono-Si/m-Si). It characterises work function in terms of surface potential (SP) using scanning probe microscopy (SPM), and Kelvin probe force microscopy (KPFM) in force modulation mode, as opposed to the usual amplitude modulation mode.

International Journal of Mechanical Engineering and Technology (IJMET)

Volume 10, Issue 12, December 2019, pp 578-588, Article ID: IJMET_10_12_053

Available online at http://www.iaeme.com/ijmet/issues.asp?JType=IJMET&VType=10&IType=12 ISSN Print: 0976-6340 and ISSN Online: 0976-6359

© IAEME Publication

AN ANALYSIS OF THE SURFACE POTENTIAL

OF MONOCRYSTALLINE SILICON SOLAR CELL USING KELVIN PROBE FORCE

MICROSCOPY

G Osayemwenre

Department of Physics, University of Fort Hare, Discipline of Mechanical of Engineering, University of KwaZulu-Natal, South Africa

F Inambao*

Discipline of Mechanical of Engineering, University of KwaZulu-Natal, South Africa https://orcid.org/0000-0001-9922-5434

*Corresponding Author Email: inambaof@ukzn.ac.za

ABSTRACT

This paper investigates the effect of defects on the electrical properties of monocrystalline silicon solar cells (mono-Si/m-Si) It characterises work function in terms of surface potential (SP) using scanning probe microscopy (SPM), and Kelvin probe force microscopy (KPFM) in force modulation mode, as opposed to the usual amplitude modulation mode A decrease in SP poses a huge challenge for the reliability of photovoltaic (PV) solar cells because of the effect of long-term degradation on the cells which reduces the efficiency of PV devices In this study, SP was obtained by measuring localised work function with KFPM and then defect localisation and characterisation were carried out with SPM to shed light on the origin of the defects Furthermore, an in-depth comparison of the SP measurements from identified defective and non-defective regions of the solar cell was carried out to quantify the effect of defects on the SP of the cell

Keywords: Scanning probe microscopy (SPM), Kelvin probe force microscope

(KPFM), Monocrystalline silicon solar cell (mono-Si), Interfaces, p-n junction,

Surface potential (SP)

Cite this Article: G Osayemwenre and F Inambao, An Analysis of the Surface

Potential of Monocrystalline Silicon Solar Cell using Kelvin Probe Force Microscopy

International Journal of Mechanical Engineering and Technology 10(12), 2019,

pp 578-588

http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=10&IType=12

1 INTRODUCTION

Over the last two decades from when Kelvin probe force microscopy (KPFM) was invented, it gained increasing relevance [1] Today, its applications relate to semiconductors, nano-devices, various photovoltaic (PV) technologies and electronic devices KFPM is used for measuring work function at a nanoscale level and charge distribution at an atomic scale level [2-5] Its technique is a combination of atomic force microscopy (AFM) and Kelvin probe Except when it is otherwise stated, the images presented in this study as scanning probe microscopy (SPM), are AFM images KPFM has different scanning modes such as contact and non-contact In this study, all measurements were taken in the non-contact mode and in most cases, in the tapping mode In the contact mode, the cantilever makes contact with a sample during scanning but this is not the case with the non-contact mode In the tapping mode there is minimal contact which does not cause damage to the interface of a sample The tapping mode is generally preferred because of the fragile nature of the interfaces of solar cell samples and it is commonly used for structural characterisation Therefore, all samples were measured in the tapping mode with high-resolution imaging without damaging the tip of the cantilever During scanning, care was taken to select a special probe in which oscillation was above the surface of the sample at the right resonant frequency The SPM technique depends

on a micro sharp tip that deflects an incoming laser as a result of its interaction with a sample With the aid of a photodiode sensor, the reflected laser is detected and analysed The analysed signal includes both the normal and lateral deflections [6] The results are presented as a three-dimensional map in nano-scale resolution The conductive nature of the sharp tip of the cantilever helps to focus the incident laser beam for better spatial resolution [7–9] Due to its numerous advantages as a non-destructive tool, KFPM is used to measure material work function or its evolution known as surface potential (SP) It can also be used to study surface charge distribution [10, 13-14] The basic theory of the KFPM is its ability to evaluate material work function from contact potential difference (CPD) of sample surfaces and cantilever tips The CPD between the conductive cantilever and the surface of a sample is the change in their electronic parameters The electrical properties of PV cells at nano scale are better characterised with the KFPM, which also provides the morphological information about analysed materials A KFPM uses two vibrating electrodes that are joined to the cantilever and an analysed sample to form a capacitor when the applied current is a direct current (DC) This causes the reference electrode to vibrate and a proportionate current is generated according to Eq 1

t cos C V

) t (  CPD  

(1) where is the frequency of vibration, the generated current is ( t ), Cis the variation

in the capacitance, and VCPD is the contact potential existing between plates

The micrograms presented in the results section include the images obtained from SPM and KFPM when voltage was applied to the grounded electrode of the sample with the cantilever fixed at a constant current and the AFM operating in frequency modulation mode [15-16] It is worth noting that at a constant height, the image corresponding to the tip height must operate in a predefined current [16] In most measurements, an oscillation amplitude of 0.5 Armstrong at zero voltage is encouraged For the accuracy of results, slow scans are encouraged throughout operation to minimise noise level during the scanning processes KPFM images are mapped through recorded spectra on the lateral grid of every single point in

a parallel plane of the surfaces of samples [17] Unlike the conventional surface probing of materials to determine work function, this study investigated the electrical properties of the samples via their cross-sectional areas To get the actual electrical behaviour of different

G Osayemwenre and F Inambao

regions of the monocrystalline silicon solar cell (mono-Si) samples, it is preferable to analyse their cross sectional areas The justification for the cross sectional probing is the limitation experienced by SPM when acquiring deep information SPM is limited to the few nanometres (about 4 nm to 10 nm) that are below the surfaces of samples, especially when performed

under an ultra-high vacuum (UHV) environment This is evident from Rosenwaks et al.’s

work which revealed that KFPM in ambient air can only measure defects of less than 2 nm below the surface of a sample [18] In the case of optical and electron microscopy, when the device is equipped with the right parameter it can detect depth signals which can provide information about deep or bulky defects Nevertheless, in the case of SPM, deep profiles can only be accessed when cross-sectional measurements are performed These types of electrical measurements done across the layers of samples help to distinguish bulk effects from surface contributions [19-20]

2 METHODS AND MATERIALS

Five mono-Si solar cells were assessed by visual inspection for defects with one being found

to show defects This cell was further subjected to infrared (IR) inspection to see if there were hotspots in the defective region of the cell (the IR inspection is reported in detail in a previous study [21]) From this mono-Si cell, two samples were prepared (one from a defective region and one from a non-defective region) for further investigation at nanoscale level using SPM and KFPM The samples were cleaved, prepared and published via their cross sections so as

to have full access to their interfaces The reason for the above procedure was to correlate visual defects with intrinsic defects (also known as bulk defects) to see if any relationship existed between them This study found that most surface defects did not have bulky defect origins However, in a few cases, surface phenomena can be due to deep defects and it is believed that such a defect can negatively affect the electrical behaviour of solar materials or devices To ensure that a close to actual experimental and industrial environment was achieved, all measurements were taken at room temperature without vacuum using SPM and KFPM in non-contact/tapping modes with a frequency of 257 kHz and spring contact of 25 N/m Figure 1 presents the optical image of the mono-Si samples The final samples were prepared from similar regions and were analysed to see if such defects are detrimental to the

elecrical behavoiur of mono-Si

KPFM allows topography and surface potential measurements to be conducted simultaneously on the sample spot For the SPM and KFPM measurements, a single electrical bias was performed in the range of -2 V to 1.5 V and applied between the p+ and the n-side contacts of the sample A set up voltage of 0.35 V was used for all measurements while the p-side was grounded The KPFM measures the change between the work function of a sample and the tip work function through the amplitude of the cantilever when a laser is beamed on the edge of the cantilever [13]

Thus, the work function of the samples was measured with KPFM and this provided the surface potential of each sample The measurements were used to assess the electrical properties across the cross-sectional areas of the samples Furthermore, two regions of the solar cells were cleaved The cleaved samples were labelled ‘defective’ and ‘non-defective’ samples, and their cross-sectional areas were mechanically polished using diamond grinding disks to achieve a smoothness of less than 40 nm, with a smoothness of 15 nm being achieved This reduced roughness helped to facilitate better electrical measurements since the effect of topography image imprint was reduced because of the smooth surface topography Furthermore, the two samples were placed on a sample holder of a good electrical contact after the right-hand side of the sample was ground with a silver (Ag) paste Throughout the scanning process, the tip work function was unchanged and the change observed between the

tip and the sample was the change in the work function of the sample; this is known as surface potential For precision, a second scan was conducted after each scan with the sample in the same condition to ensure that the tip had not degraded following the series of scans performed

Figure 1 Optical microscopy images of a mono-Si solar cell (a) shows the landing position; the red

boxes indicate possible bulk defect regions and (b) shows Ag finger and landing position; the black

box represents the region without defect

3 RESULTS AND DISCUSSIONS

The topography images were taken at different magnifications to demonstrate structural damage The structural imperfection is clearly seen in Figure 2 (d and e) below, while the 3D image of Figure 2 (b-e) is shown in Figure 2 (f-i) The morphological images in 2D (Figure 2 a-e) became clearer as the scanned areas decreased, hence the topography images were better examined in a smaller area

(a)

(b)

G Osayemwenre and F Inambao

Figure 2 Topography images of part of the mono crystalline silicon solar cell with different

magnifications under standard room temperature (b), (c) (d) and (e) are topographies at 10  m, 5  m,

1  m and 496 nm magnification respectively; (f), (g) (h) and (i) are the 3D topography images of ‘b’

‘c’ ‘d’ and ‘e’ respectively

Figure 2 shows the topographies of the defective and non-defective sample of the mono-Si solar cell The scan provided spatial dimension images at different magnifications to detect possible structural defects Figure 2 (d-e) indicates the topography also known as the height sensor which shows some granular variations in a few areas of the image Such features are usually difficult to see clearly when a large area is scanned For the mechanical measurement (topography), scans were conducted in the tapping mode This mode is very soft, hence it has less effect on the samples and as such the surface roughness is as a result of structural deformation which may be due to degradation It is observed that the interface region has different topography roughness due to inhomogeneous degradation experienced by the

mono-Si cell The cross-sectional area of the samples used for SP measurements are presented in Figure 3 below To maintain the electronic property of the layers during the scanning process, the non-contact mode was preferred to avoid the degradation of the interfaces of the samples

Figure 3 3D topography images of (a) non-defective sample and (b) defective sample, used for the

electrical characterisation

The micrograph presented in Figure 3 includes the images of the topographies of the defective and non-defective samples They are similar except for their roughness The maximum height of the non-defective sample was 12 nm and that of the defective was 57.6

nm The minimum height roughness of the non-defective and defective samples used for the electrical characterisation were 5 nm and 16.2 nm respectively It was observed that the brighter side of Figure 3a did not to correspond with height because of charge accumulation, which had no effect on the electrical behaviour of the sample

3.1 Surface Potential Measurement

The most viable parameter for defect analysis is electrical, hence the electrical measurement

of the samples are presented in this section In section 3 the measurements presented are all mechanical, but here only electrical analysis from KFPM is presented

Figure 4 KPFM images of mono-Si (a) non-defective sample, (b) defective sample in 2D Figure 4 presents the 2D image of the samples in Figure 3 obtained from the surface potential measurements Two distinct regions are seen with the p-Si corresponding with the brighter side The dark side corresponds with the n-Si at a lower potential, and the depletion region is not clearly seen The results in 2D show a clear demarcation between the pn junction of the mono-Si solar cell for the non-defective and defective samples without illumination The 3D version of these images is presented below

Figure 5 KPFM images of mono-Si (a) non-defective sample, (b) defective sample in 3D

Unlike the 2D images presented in Figure 4, the depletion region in the 3D images presented in Figure 5 is clearly seen The charging effect on the n-Si is observed in Figure 5b, the charged region corresponding with the effect indicated by the black rectangular shape in Figure 3a The identification of the area of interest was confirmed by the topography profiles obtained from the mechanical characterisation of the samples which was done before the commencement of the electrical measurement This allowed the positioning of the edge of the sample In each measurement, an average of 35 consecutive horizontal scan lines was used to

reduce noise effect from the surface potential measurement According to Narchi et al [22],

multiple AFM images have the ability to drift As such, the line profiles presented in Figure 6 (a and b) below are from the region of the samples with the most identified areas

(a)

(b)

G Osayemwenre and F Inambao

0.0 0.5 1.0 1.5 2.0 2.5 3.0 -550

-500 -450 -400 -350 -300 -250 -200 -150 -100 -50 0 50 100 150 200 250 300

Distance (  m)

Single line profile of the

no defective sample

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

-120 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 200

Distance (  m)

Defective sample i-Si region

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

O r i g i n P r o 8 E v a l u a t i o n O r i g i n P r o 8 E v a l u a t i o n

Figure 6 Surface potential line profiles across the cross sections of mono-Si solar cell under room

condition for: (a) non-defective sample, (b) defective sample; (c) inside the depletion region of the sample presented in Figure 6 (a) and (d) inside the depletion region of the sample presented in Figure

6 (b)

Each profile showed good coherent features as expected from a photo-voltage measurement In Figures 6 (a and b) the profiles showed higher surface potentials in the p-Si layer with band bendings in the intrinsic layers, while the n-Si layer showed the lowest surface potential In the defective sample, the surface potential decreased in both the p-Si and n-Si layers because of the effect of degradation Furthermore, two different regions in the samples (inside the intrinsic layers) were scanned and the results are presented in Figures 6 (c and d) In Figure 6d, the surface potential showed a high noise level related to thermal fluctuations According to Narchi, this level of noise even after averaging is unexpected in a perfect intrinsic layer like the one in Figure 6c [22-24]

The results in Figures 6b and Figure 6d show a huge discrepancy in the surface potential

of the sample from the non-defective region; this is not expected [22, 25] Theoretically the observed discrepancy may be due to the measuring device degradation phenomenon, like tip oxidation or abrasion However, the precautions taken and the use of frequency-modulated KPFM (FM-KPFM) in the non-contact mode in this work eliminated the above assumption Therefore, an abnormal phenomena capable of affecting the opto-electrical property of the sample might have occurred for the reasons stated below Firstly, layer and interface delamination; this allows particle migration to form an unintended doping effect Thus, it

could form an n-dope or p-dope and create an n-n, a p-p or an n-p-n junction instead of a p-n

junction Secondly, it could be because of a high concentration of the interface defect layer

0.0 0.5 1.0 1.5 2.0 2.5 3.0

-300

-200

-100

0

100

200

300

Non-defective sample

Distance across the interface (µm)

-20

0

20

40

60

80

Distance (µm)

(d) (c)

between the p-Si and n-Si sides If the latter had happened in this case it would have led to electrical fluctuations across the interface regions and produced inhomogeneous potential behavours, which when controlled could produce good fluctuations Hence, KFPM can also

be used to study impurity concentration or impurity photovoltaic effect in PVmaterials

As described above, the variations of surface potentials across i-Si regions can be measured with KPFM These types of measurements at nanoscale together with an electrical nanodevice can be used to study defects in such regions and the application can be used in a wide range of PV cells The net surface potential presented above provides the overall performance of the sample However, scanning only the intrinsic layer provides a better view

of its spatial resolution and therefore is more advantageous for analysing the interfacial degradation of the samples Some areas in the 3D image of the SP image obtained from the

i-Si region showed lower values than those obtained from the non-defective sample This is because the region had a weak bonding strength as revealed by previous studies Hence, the AFM laser beaming from the cantilever led to the formation of imprints in the sample Such features of laser on surface potential measurements can be linked to previous identifications

and discussions by Osayemwenre et al and Feijfar et al [26-27] Another reason is due to the

closeness of the p-Si and i-Si junctions to the edge of the region since a-Si is a pin device Literature revealed that the carrier transport at this region via the interface provides an Ohmic conduction path The Voc in this affected region dropped as seen from the surface potential values

The two samples (defective and non-defective) in this work were examined to show the existence of photo-signal degradation In each case, a slow scan was used to obtain reliable results that are reproducible As earlier stated, tip oxidation, sample modification during the measurements and laser-induced effects were not found on the surface of the samples Tip oxidation and sample modification could not have been possible because different tips of the same specifications and calibrations were used for each sample For the surface induced illumination, all conditions for each scan and the laser intensities were kept constant throughout measurements The only possible explanation for the surface potential decrease is

that it could be due to defect or trap effects previously studied by Zhang et al [10, 28] The

hypothesis of defects is supported by the nonexistence of the interface junction between the

i-Si and n-i-Si layers in the defective sample This makes the band bending less pronounced or absent as obtained in Figure 6d; this is similar to Kikukawa’s report on the effect of surface states on band bending [6, 21, 29]

3.2 The Electrical behaviour Path across the P-N Junction

The results from the 3D images in Figure 5b showed a drift in the surface potential when the non-defective sample was compared with the defective sample The range of drifts varies from one region to another; in this case, the variation was less than 100 mV 1000 nm area in the interface region of each of the sample was scanned to see if there were changes in the band bending Hence the interested is on the curves smoothness and not on the value of the

SP, hence the scanned region in the depletion region are not on same scale Figure 6b shows a different shape that was contrary to the expected result as an undulated shape was observed The surface potential line seemed like a contamination of interface; this is not because of interface modification but because of changes in an inbuilt electrical field Such an effect is not attributable to change in the work function that results from inhomogeneous surface, but

to the available electrostatic field due to charging [8, 16, 17, 29] Built in electrical charge normally influences KFPM at the interface region material [17, 29 The tip and sample work functions, when rightly calibrated, are represented in Eq 2 below The work function

G Osayemwenre and F Inambao

measured by KFPM is converted to surface potential By taking a spatial local derivative of

Eq 2, Eq 3 is obtained [24, 30]

Note that Φs and Φt stand for samples and tip work function respectively, and Vsp is the surface potential Figure 6 presents significant information regarding the effect of degradation inside the p-n junction Such information can be used to investigate interface defects, especially in shallow levels, provided the samples are mechanically polished to expose the various layers

4 CONCLUSION

In this study, the structural defects and electrical behaviours of a mono-Si cell were examined via the cell’s cross-sections SPM and KPFM were used to identify the defective regions of the samples by measuring the surface potential of samples from a defective region and a non-defective region of a mono-Si cell As expected, the 2D and 3D images of the surface potential showed two- and three-stepwise profiles from the p-Si to the n-Si layers respectively The 3D image clearly showed the interface between the p-Si and the n-Si layers With KPFM the electrical behaviour of the two cleaved samples from a single mono-Si solar cell was analysed to show the effect of defects on PV cells The KPFM result of the sample from the defective region decreased compared to the surface potential of the sample from the non-defective region To further explore the effect of degradation on the sample, the depletion regions were analysed by scanning the layer or region between the p-Si and the n-Si layers In addition, a further analysis of the depletion regions showed an abnormal band bending pattern contrary to what was found in the non-defective region of the cell Furthermore, the possible challenges of interpreting KPFM images and profile results are itemised from a range of physical features Based on theoretical explanations, most of these features were eliminated This novel technique proffers methods that can be used to investigate surface photovoltage changes in various materials at nanoscale of different crystalline silicon solar and inorganic hybrid perovskite PV cells In conclusion, the presence of morphological defect does not always correspond to bulk defect in a mono-Si solar cell However, the roughness of the sample increases in the defective region of a mono-Si cell

FUNDING

This research received no external funding

ACKNOWLEDGMENTS

The authors would like to express their gratitude to the following organization:

University of Fort Hare, University of South Africa, and University of KwaZulu-Natal

DECLARATION OF CONFLICTING INTERESTS

The authors declare no potential conflicts of interest in respect of this research, authorship and/or publication of this article

DATA AVAILABILITY

The data used to support the findings of this study are included within the article However, in support of open access research, all underlying data can be accessed upon request via email to

the corresponding author

REFERENCES

[1] Nonnenmacher, M., O’Boyle, M P and Wickramasinghe, H K Kelvin Probe Force

Microscopy Applied Physics Letters, 58(25), 1991, pp 2921–2923 https://doi.org/10.1063/1.105227

[2] Melitz, W., Shen, J., Kummel, A C and Lee, S Kelvin Probe Force Microscopy and its

Application Surface Science Reports, 66(1), 2011, pp 1–27 https://doi.org/10.1016/j.surfrep.2010.10.001

[3] Doukkali, A., Ledain, S., Guasch, C and Bonnet, J Surface Potential Mapping of Biased

pn Junction with Kelvin Probe Force Microscopy: Application to Cross-Section Devices

Applied Surface Science, 235(4), 2004, pp 507–512

[4] Moores, B., Hane, F., Eng, L and Leonenko, Z Kelvin Probe Force Microscopy in

Application to Biomolecular Films: Frequency Modulation, Amplitude Modulation, and Lift Mode Ultramicroscopy, 110(6), 2010, pp 708–711 https://doi.org/10.1016/j.ultramic.2010.02.036

[5] Okamoto, K., Sugawara, Y and Morita, S The Imaging Mechanism of Atomic-Scale

Kelvin Probe Force Microscopy and its Application to Atomic-Scale Force Mapping

https://doi.org/10.1143/JJAP.42.7163

[6] Kikukawa, A., Hosaka, H., and Imura, R Silicon pn Junction Imaging and

Characterization using Sensitivity Enhanced Kelvin Probe Force Microscopy Applied Physics Letters, 66, 1995, pp 3510 http://dx.doi.org/10.1063/1.113780

[7] Barth, C., Foster, A S., Henry, C R and Shluger, A L Recent Trends in Surface

Characterization and Chemistry with High Resolution Scanning Force Methods Advanced Materials, 23(4), 2011, pp 477–501 https://doi.org/10.1002/adma.201002270

[8] Sommerhalter, C., Glatzel, T., Matthes, T W., Jager-Waldau, A and Lux-Steiner, M C

Kelvin Probe Force Microscopy in Ultra High Vacuum Using Amplitude Modulation

Detection of the Electrostatic Forces Applied Surface Science, 157(4), 2000 pp 263–268

https://doi.org/10.1016/S0169-4332(99)00537-1

[9] Douheret, O., Swinnen, A., Berthot S et al High-Resolution Morphological and Electrical

Characterisation of Organic Bulk Heterojunction Solar Cells by Scanning Probe

Microscopy Progress in Photovoltaics: Research and Applications, 15(8), 2007, pp 713–

726 https://doi.org/10.1002/pip.795

[10] Tal, O., Gao, W., Chan, C K., Kahn, A and Rosenwaks, Y Measurement of Interface

Potential Change and Space Charge Region across Metal/Organic/Metal Structures using

Kelvin Probe Force Microscopy Applied Physics Letters, 85(18), 2004, pp 4148–4150

[11] Zerweck, U., Loppacher, C., Otto, T., Grafstrom, S and Eng, L M Kelvin Probe Force

Microscopy of C60 on Metal Substrates: Towards Molecular Resolution Nanotechnology,

18(8), 2007 https://doi.org/10.1088/0957-4484/18/8/084006

[12] Lee, S., Shinde, A and Ragan, R Morphological Work Function Dependence of

Rare-Earth Disilicide Metal Nanostructures Nanotechnology, 20(3), 2009 https://doi.org/10.1088/0957-4484/20/3/035701

[13] Sadewasser, S., Jelinek, P., Fang C -K, et al New Insights on Atomic-Resolution

Frequency-Modulation Kelvin-Probe Force Microscopy Imaging of Semiconductors Physical Review Letters, 103(26), 2009 https://doi.org/10.1103/PhysRevLett.103.266103

[14] Sadewasser, S., Glatzel, T., Rusu, M., Jager-Waldau, A and Lux-Steiner, M C

High-Resolution Work Function Imaging of Single Grains of Semiconductor surfaces Applied Physics Letters, 80(16), 2002, pp 2979–2981 https://doi.org/10.1063/1.1471375

[15] Ternes, M., Lutz, C P., Hirjibehedin, C F., Giessibl, F J., and Heinrich, A J., The Force

Needed to Move an Atom on a Surface Science, 319(5866), 2008, pp 1066–1069

https://doi.org/10.1126/science.1150288

[16] Albrecht T R., Grutter P., Horne D., and Rugar D Frequency-Modulation Detection using

High-Q Cantilever for enhanced Force Microscope Sensitivity, Journal of Applied Physics, 69(2), 1991, pp 668–673