Near Field Microscopy

Instrumental developments associated with scanning probe microscopies Champ proche

"Scanning Probe Microscopies” are a family of surface characterization techniques emerged in the early 1980s in the wake of tunneling microscopes (STM) and atomic force microscope (AFM). These techniques were introduced in the Laboratory in 1992 and have aroused since 1994 specific instrumental developments for performing local electrical measurements by conductive tip AFM, especially of resistance with the so-called "Resiscope" device. The advantage of such measurements potentially concerns all materials and devices studied in the PHEMADIC pole, and gives rise to varied external collaborations. An important AFM platform was formed over the years, now consisting of five AFM equipments - all coupled to a Resiscope - with additional features in order to address all sample types. Besides, one of these equipments is also dedicated to teaching activities (practical trainings as part of the M2 Nanosciences, Paris-Saclay University). In many cases the information from the AFM-Resiscope association may be profitably collated with spectroscopic analyzes from the XPS-AES-UPS platform. Furthermore the Resiscope facility has been recently implemented on the integrated AFM-Raman spectroscopy platform.

The Resiscope

Historically the Resiscope is the first instrument for local electrical AFM measurements developed in the laboratory. The principle is simple: a DC bias is applied (from ±10mV to ±10V) between the conductive tip and the sample, and the resulting current is measured. Successive developments of the Resiscope (which still continue) aim at achieving local resistance maps over the widest possible range of values, with the less aggressive possible contact conditions between tip and sample surface. To meet these requirements, a specific technique of real-time dynamic compression has been developed, together with a precise calibration protocol [patent EP 2567245A1 (2013)]. This solution makes it stand out compared to the state of the art world, including to obtain, compared to basic logarithmic amplifier assemblies used elsewhere, increased dynamics of several orders of magnitude (11 decades of resistance available on the latest prototype), with better precision on the measurement of very small currents.

Schematic view of the AFM-Resiscope coupling. The left side of the figure recalls the principle of standard AFM, which provides a map of the sample topography. When using a conductive tip and applying a DC bias between the tip and the sample, measurement of the resulting current (right side) allows simultaneously to get a map of local resistance.
Schematic view of the AFM-Resiscope coupling. The left side of the figure recalls the principle of standard AFM, which provides a map of the sample topography. When using a conductive tip and applying a DC bias between the tip and the sample, measurement of the resulting current (right side) allows simultaneously to get a map of local resistance.

Considering the Interest aroused by the unique performance of Resiscope device, a valorization process was engaged, in a first stage through a restricted distribution of home-made clones in 7 French laboratories, then from 2007 in the framework of a transfer to the SME Concept Scientific Instruments (CSI). The resulting commercial version is distributed by ScienTec company. To date, more than thirty units were sold in different countries. The partnership with CSI was established in a long-lasting view : the company has thus been associated with significant recent developments of Resiscope, particularly in the context of two successive projects funded by the National Research Agency (ANR) : « ALICANTE » (2007-2010 : increasing the range of measurement down to 0.1pA and even less) and « MELAMIN » (2011-2015 : measurements in intermittent contact mode, see below), the latter being coupled with a PhD thesis under industrial convention CIFRE with CSI. This innovation and transfer dynamics earned P. Chrétien, F. Houzé and O. Schneegans to receive in 2013 the 2nd Prize of Applied Research from Federation of Electrical, Electronical and Communication Industries (FIEEC), and in 2014 the Yves Rocard Prize from the French Physical Society (SFP).

Latest major evolution of Resiscope device : adaptation to an intermittent tip/sample contact mode

Developed in contact mode in which the tip continuously applies a force on the sample, the initial Resiscope technique reaches its limits on fragile samples, since tip friction may damage their surface. That is the reason why a study was begun in 2011 to modify the Resiscope device in order to operate in an intermittent contact mode, much less aggressive, in which the tip briefly hits the surface at regular intervals (ANR Project « MELAMIN », A. Vecchiola’s PhD thesis). This adaptation of the device to radically different and much more complex contact conditions aimed to answer to an increased demand for local electrical characterization on soft materials or weakly anchored nano-objects. As a result (end of 2015), a new Resiscope electronics suitable for intermittent frequencies up to 1kHz is operational, as well using classical sinusoidal intermittency actuation (the so-called Pulsed Force Mode from Witec of Peak Force Mode from Bruker), as with the specific, more elaborated actuation designed by CSI (Soft-Resiscope Mode). This new Resiscope version allows optimized resistance measurements over a range of 10 decades, from 1 kΩ to 10 TΩ [Appl. Phys. Lett. 108, 243101 (2016)].

Schematic view of the set-up used for Resiscope measurements in intermittent contact mode. Here the intermittency is obtained by introducing a sinusoidal modulation on the AFM z piezo at a frequency between 100Hz and 2kHz, thanks to a pulsed-force commercial module from Witec, with the maximum repulsive force used as feedback parameter. The graph on left side of the figure shows the successive phases of a deflection vs time curve, with jump-to-contact, force increasing up to setpoint value, then decreasing with adhesion peak at withdrawal, and lastly jump-off and free cantilever oscillation.

Schematic view of the set-up used for Resiscope measurements in intermittent contact mode. Here the intermittency is obtained by introducing a sinusoidal modulation on the AFM z piezo at a frequency between 100Hz and 2kHz, thanks to a pulsed-force commercial module from Witec, with the maximum repulsive force used as feedback parameter. The graph on left side of the figure shows the successive phases of a deflection vs time curve, with jump-to-contact, force increasing up to setpoint value, then decreasing with adhesion peak at withdrawal, and lastly jump-off and free cantilever oscillation.

This electronics has been validated on samples of known structure : self-assembled monolayers (SAMs) of alkanethiols with various chain lengths deposited on gold-coated substrates. Extensively considered as model systems to investigate electron transport mechanisms by various characterization methods (in particular STM), SAMs of thiols are indeed a simple mean to get reference objects with high molecular organisation and well-defined thickness, ideally adapted to test the new Resiscope capabilities. For such a situation of a molecular layer with a wide gap between its highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) sandwitched between two metal contacts (the AFM tip and the gold substrate), the conduction mechanism is expected to be direct tunnelling, with resistance increasing exponentially with molecular length l following the law

where ß and R0 denote the so-called attenuation factor and effective contact resistance parameters, respectively. On the figure below are shown typical resistance maps obtained for the 5 chain lengths studied (6, 8, 10, 12 and 14 carbon atoms). From a qualitative viewpoint, one can actually verify that the measured resistance globally increases with chain length, since prevailing colour evolves from orange to purple. No degradation was observed on associated topography maps.

Typical resistance maps (1µm  0.5µm) obtained by pulsed-force-Resiscope with a Cr/Pt-coated tip on the series of alkanethiol SAMs with 5 different chain lengths.
Typical resistance maps (1µm x 0.5µm) obtained by pulsed-force-Resiscope with a Cr/Pt-coated tip on the series of alkanethiol SAMs with 5 different chain lengths.

For each image an histogram of resistance values is calculated from which the mean values of Log(R) and the associated standard deviations can be derived. The next figure sums up the results; for each chain length, data correspond to different zones of the same sample (and have been slightly shifted for clarity). As expected, this graph clearly evidences a resistance increase of nearly one decade every two additional C atoms. From a linear fit the attenuation factor is found to be 1.04 per C atom, that is to say 0.83 Å-1, which is in accordance with results reported in the literature for similar configurations. These results demonstrate the ability of Resiscope in intermittent contact mode to get semi-quantitative results on ultrathin fragile materials.
Summing-up of mean Log(R) values (square symbols, slightly shifted along each C number for clarity, 1 sample per C number, from 3 to 5 zones per sample) and associated standard deviation (error bars) calculated from the resistance maps obtained by PFM-Resiscope on the SAMs layers, and linear fit of the data (dotted line). The deduced attenuation factor is =1.04 / C atom (0.83 Å-1), in agreement with values reported in the literature.
Summing-up of mean Log(R) values (square symbols, slightly shifted along each C number for clarity, 1 sample per C number, from 3 to 5 zones per sample) and associated standard deviation (error bars) calculated from the resistance maps obtained by PFM-Resiscope on the SAMs layers, and linear fit of the data (dotted line). The deduced attenuation factor is Β=1.04 / C atom (0.83 Å-1), in agreement with values reported in the literature.

Beyond molecular samples like SAMs, Resiscope in intermittent mode may open up interesting prospect for carbonaceous objects and related materials. As an illustration, the above figure shows it is possible to get resistance cartography on individual vertical carbon nanotubes, 50-70nm in diameter and about 430nm in height, which are extremely brittle nano-objects.


Application of the Resiscope in intermittent mode (max force 5-10 nN) for imaging individual vertical carbon nanotubes. The top left scanning electron microscopy image, shows one of these nano-objects. The top right image corresponds to a typical resistance map, which proved to be reproducible over several successive scans (7 in the present case). The resistance profile (bottom curve) reveals variations over 5 decades. The height profile (upper line) allows to verify that the object is not altered by the scan. Conversely, when operating in contact mode at the same force, the nanotube is broken from the first scan. Sample courtesy of NanoCarb Team, Thales Research & Technology.
Application of the Resiscope in intermittent mode (max force 5-10 nN) for imaging individual vertical carbon nanotubes. The top left scanning electron microscopy image, shows one of these nano-objects. The top right image corresponds to a typical resistance map, which proved to be reproducible over several successive scans (7 in the present case). The resistance profile (bottom curve) reveals variations over 5 decades. The height profile (upper line) allows to verify that the object is not altered by the scan. Conversely, when operating in contact mode at the same force, the nanotube is broken from the first scan. Sample courtesy of NanoCarb Team, Thales Research & Technology.

 

Examples of recent or current studies using Resiscope

  • - Surfaces engineering : advanced coatings involving graphene flakes for applications to electrical contacts and connecting devices. K. Dalla-Francesca’s PhD, 2012-2016. [Proc. 60th IEEE Holm Conf. on Electrical Contacts, pp. 116-123 (2014)]
  • - Origin of the leakage current in diamond-based Schottky barrier devices, coll. with National Institute for Materials Science (NIMS), Japan. [J. Phys. D: Appl. Phys. 47, 355102 (2014)]
  • Local modification of materials and their interfaces : resistive transition in thin films of lithium cobalt oxides. Coll. LPS, IEF and ICMMO (Paris-Saclay labs), LITEN (CEA-Grenoble) and Cyprus Univ. V.-H. Mai PhD (2011-2014), V.-S. Nguyen PhD (2014-2017). [Adv. Mater. 23 (36) 4141 (2011)]
  • - Electrical characterizations at nanoscale in the framework of IPVF
  • - Characterization of GaN nanowires for piezo-generator application, estimation of piezo conversion balance and maximum power density delivered. Coll. With LPN and IEF (Paris-Saclay labs). N. Jamond PhD (LPN, 2013-2016).

  • 3D output voltage cartography collected by a modified version of Resiscope on partially encapsulated GaN nanowires. The study has demonstrated a maximum output voltage of 443mV, which is the highest reported value for GaN-based NWs [Phys. Status Solidi RRL 8 (5) pp.414-419 (2014), presented image was selected for backcover]. Using the AFM associated with the modified Resiscope in force mode, it is also possible to investigate the evolution of piezo signal as a function of compressive load.
  • 3D output voltage cartography collected by a modified version of Resiscope on partially encapsulated GaN nanowires. The study has demonstrated a maximum output voltage of 443mV, which is the highest reported value for GaN-based NWs [Phys. Status Solidi RRL 8 (5) pp.414-419 (2014), presented image was selected for backcover]. Using the AFM associated with the modified Resiscope in force mode, it is also possible to investigate the evolution of piezo signal as a function of compressive load.
  • - Electrical properties of fractal metallic nanostructures (clusters of gold nanoparticles at the vicinity of the percolation transition). Coll. with GEMaC (Paris-Saclay lab)).
  • - Origin of contact failure on switching devices. Industrial contract.

The « Capascope »

Started in 2001, this module may be considered as a broadened version of Resiscope : a small AC component is superimposed to the applied DC bias, and the resulting quadrature component of the current is extracted using a lock-in device to get a measurement of local capacitance. It must be emphasized that the question is here to measure absolute capacitance and not capacitance variations as with the so-called Scanning Capacitance Microscopy (SCM) technique. Such a measurement is a challenging task mainly because the local interesting capacitive signal at the tip apex must be extracted from a non-local stray capacitance, 4-5 orders of magnitude higher. The original method developed in our lab combines a two-pass image acquisition (first scan in contact, second scan at a constant, close distance to the surface) with a judicious exploitation of capacitance-distance curves recorded at regular intervals. Using this principle it is possible to quantify and then to subtract parasitic capacitances, which significantly improves measurement resolution (I. Estevez’ PhD thesis) [Appl. Phys. Lett. 104 (8), 083108 (2014)]. As an illustration of the resolution capability of this technique, the figure below presents the outstanding result obtained on a test structure consisting of a matrix of square-shaped capacitors, close enough to have mutual influences : the capacitance cartography clearly evidences a specific contrast depending on each capacitor vicinity. Operated as described, the Capascope module offers a measurement range of 7 decades (from 10-10 to 10-17 F), with the limitation of scan rates slower than with Resiscope ( < 1 line/s )


Top : Schematic view of a test structure composed of a 3×3 matrix of gold square electrodes deposited on a Macor® substrate with a continuous gold under electrode.
Bottom : partial Capascope image of the structure, revealing a specific capacitance contrast depending on each square plate position (corner, middle edge, centre), attesting the high sensitivity of the device (variations of only a few tenths of femtoFarad).

The relevance of the technique from a quantitative viewpoint has been investigated using a series of calibration samples consisting of individual square metal-insulator-metal capacitors (golden plates on Macor) with various dimensions between 8µm and 7mm, for which expected theoretical capacitance values can be calculated by Finite Element Method. Numerical modeling is essential to take into account edge effects (which cannot be neglected for small size electrodes), as well as the electrostatic influence of AFM probe, chip and holder. Once the problem limits and mesh are carefully defined, experimental data prove to be in nice agreement with calculated values.
Accord entre mesures et valeurs théoriques calculées, pour une série de plots dorés carrés sur Macor de différentes tailles. On note l’écart avec la loi analytique classique C =  S/e, d’autant plus important que le plot est petit, en raison des effets de bords significatifs à ces dimensions.
Agreement between experimental capacitance measurements on a series of calibration square-shaped capacitors and corresponding values obtained from numerical simulations. The green line corresponds to the ideal parallel plate formula which becomes more and more inoperative as the electrode size decreases due to the prevalence of edge effects.