Atomic force microscopy is a powerful technique for measurement and mapping of nanoscale topography and electrical and mechanical sample properties. The Nanosurf DriveAFM is a new generation instrument that combines ease of use and high performance through full motorization, CleanDrive photothermal excitation, and a mechanical and electrical design that allows for both high-resolution and large-range imaging.

Since its invention in 1986,1 atomic force microscopy (AFM) has evolved from a technology for surface topography imaging with sub-nanometer resolution into a multifunctional toolbox, which is also capable of measuring electrical and mechanical sample properties with nanometer spatial resolution. Recent technological developments provide new AFM systems with unprecedented ease of use, flexibility, and performance. In particular, the Nanosurf DriveAFM, Fig. 1(a), represents a new generation of the atomic force microscope, combining a highly versatile tip-scanning design with a compact scanner, full motorization, and CleanDrive photothermal excitation. In this work, we cover key aspects of the DriveAFM design and show how these features combine and unite both ease of use and high performance to make the DriveAFM a powerful and multifunctional AFM.

FIG. 1.

(a) DriveAFM scanhead that can be used as a benchtop system or on top of an inverted optical microscope. (b) Schematic illustration of the optical beam tracking concept used in the DriveAFM. The scanner (dashed box), equipped with a mirror and a focus lens, is moved along the axis indicated by the arrow. The collimated light beam emitted by the light source is redirected toward and focused onto the cantilever, reflected from the cantilever, and in the return path recollimated and redirected toward the photodetector. The physical arrangement of the light source and the photodetector on opposite sides of the scanning motion, along with the layout of the mirrors, maintains the end position of the outgoing light beam on the photodetector regardless of the position of the scanner (two scanner positions are shown, red and yellow light paths). Only when the cantilever is deflected, a signal is measured at the photodiode (blue light path). This principle, illustrated in one axis, is extended in two axes to provide XY scan motion optical tracking. (c) Rendering of the scanner and motorized optomechanical adjustment modules of the DriveAFM.

FIG. 1.

(a) DriveAFM scanhead that can be used as a benchtop system or on top of an inverted optical microscope. (b) Schematic illustration of the optical beam tracking concept used in the DriveAFM. The scanner (dashed box), equipped with a mirror and a focus lens, is moved along the axis indicated by the arrow. The collimated light beam emitted by the light source is redirected toward and focused onto the cantilever, reflected from the cantilever, and in the return path recollimated and redirected toward the photodetector. The physical arrangement of the light source and the photodetector on opposite sides of the scanning motion, along with the layout of the mirrors, maintains the end position of the outgoing light beam on the photodetector regardless of the position of the scanner (two scanner positions are shown, red and yellow light paths). Only when the cantilever is deflected, a signal is measured at the photodiode (blue light path). This principle, illustrated in one axis, is extended in two axes to provide XY scan motion optical tracking. (c) Rendering of the scanner and motorized optomechanical adjustment modules of the DriveAFM.

Close modal

Tip-scanning AFMs excel in several ways over sample-scanning AFMs. With all the essential components of an AFM system incorporated into a single unit, the DriveAFM, Fig. 1(a), is a truly portable and configurable multifunctional device. It can be operated as a benchtop system with a small footprint for opaque samples or easily transferred to an inverted optical microscope within minutes to allow for simultaneous AFM and unperturbed optical observation, thanks to a stationary sample. Furthermore, a tip-scanning AFM such as the DriveAFM can readily investigate large and heavy samples without compromising the AFM scanner’s performance. Nevertheless, tip-scanning AFMs are intrinsically more difficult to design compared to sample-scanning AFMs.2 The DriveAFM introduces a new concept to track the cantilever’s scanning motion with the laser beam of the deflection detection unit. This new design avoids mounting a heavy deflection detection unit onto the scanner that negatively impacts its performance, instead introducing only low-weight passive scanner-mounted elements, Fig. 1(b). This new patented beam tracking design3 allows for placing the photodetector and light source off of the scanning unit and even enables adding an additional light source for photothermal excitation and robust motorization of all adjustments. Thanks to the full motorization, Fig. 1(c), the DriveAFM can be operated fully remotely, allowing the user to adjust and optimize the system without disturbing the controlled measurement environment.

Many popular AFM modes require oscillating the cantilever near, or at, its resonance frequency. The oscillation is usually indirectly driven by using a shaker piezo placed near the cantilever. This drive mechanism typically leads to many spurious signals.4 The CleanDrive option of the DriveAFM represents an alternate approach, also known as photothermal excitation, that uses an additional intensity-modulated laser that can be independently positioned, Fig. 2(a), to directly excite cantilever oscillations, typically through a local bimorph effect.5 

FIG. 2.

(a) CleanDrive photothermal excitation schematic: standard readout beam at the free end of the cantilever and the second independently positioned intensity-modulated laser beam near the base of the cantilever to excite cantilever oscillations. (b) Excitation of an AC40 cantilever in buffer solution using either standard piezo-acoustic excitation (black line) or CleanDrive photothermal excitation (red line). Only CleanDrive photothermal excitation shows the proper resonance peak of the cantilever. With piezo-acoustic excitation, a “forest of peaks” masks the response of the cantilever. (c) At higher frequencies with piezo-acoustic excitation, the cantilever response can be difficult to separate from spurious structural resonances of similar magnitude. Exciting cantilevers at higher frequencies in air is simplified with CleanDrive photothermal excitation, and the resonance peaks remain clearly distinguishable up to several MHz excitation. (d) A cantilever excited using CleanDrive maintains a stable excitation amplitude even in changing environmental conditions, such as a slowly evaporating water droplet surrounding the cantilever.

FIG. 2.

(a) CleanDrive photothermal excitation schematic: standard readout beam at the free end of the cantilever and the second independently positioned intensity-modulated laser beam near the base of the cantilever to excite cantilever oscillations. (b) Excitation of an AC40 cantilever in buffer solution using either standard piezo-acoustic excitation (black line) or CleanDrive photothermal excitation (red line). Only CleanDrive photothermal excitation shows the proper resonance peak of the cantilever. With piezo-acoustic excitation, a “forest of peaks” masks the response of the cantilever. (c) At higher frequencies with piezo-acoustic excitation, the cantilever response can be difficult to separate from spurious structural resonances of similar magnitude. Exciting cantilevers at higher frequencies in air is simplified with CleanDrive photothermal excitation, and the resonance peaks remain clearly distinguishable up to several MHz excitation. (d) A cantilever excited using CleanDrive maintains a stable excitation amplitude even in changing environmental conditions, such as a slowly evaporating water droplet surrounding the cantilever.

Close modal

CleanDrive photothermal excitation offers significant advantages over the conventional piezo-based excitation. Particularly in liquids, CleanDrive excels with a reliable textbook-like amplitude and phase response, Fig. 2(b). Piezo-based excitation typically produces a so-called forest of peaks as not only the cantilever but also the cantilever holder, and its surroundings are excited and eventually coupled back into its motion, Fig. 2(b).4,6 The peak distribution in such a forest of peaks depends on many factors (e.g., tip–surface distance) and may vary with time. CleanDrive also allows for exciting the cantilever at frequencies of up to several MHz, where piezo-based excitation suffers from significant loss in efficiency, Fig. 2(c). Without the limitations of the piezo-based excitation, the oscillation amplitude of the cantilever can be maintained over hours even in liquid environments with drastically changing volume, Fig. 2(d).

Rapid and accurate scanning, a critical AFM functionality, requires a combination of high resonance frequency scanner mechanics and precise low-noise electronics. Most large-range AFMs mechanically amplify a small piezo motion to achieve a larger scan range. This amplification, however, causes parasitic force losses in the amplifying structure and requires softening of the scanner structure. In contrast, the DriveAFM uses a direct drive scanner architecture with large-range piezo actuators. This 1:1 piezo motion to scanner motion coupling, Fig. 3(a), provides more driving force and allows for maximizing the scanner stiffness. Combining this approach with the new beam tracking concept results in a very compact scanner design with minimized moving mass and high stiffness that yields a large 100 × 100 × 20 μm3 scanner with high resonance frequencies. The primary system resonance at 3.5 kHz, Fig. 3(b), represents an excellent figure for such a large-range scanner.

FIG. 3.

(a) DriveAFM XY scanner schematic. The 1:1 direct drive scanner approach results in a very stiff and high resonance frequency scanner, important for stable and fast AFM imaging. (b) Frequency response of the DriveAFM scanner motion when actuated in Z. Despite its large scan range of 100 µm in XY and 20 µm in Z, the complete XYZ tip scanner shows a primary resonance at 3.5 kHz. (c) Scan ramp signal (72 μV) as a function of time showing improved accuracy with the full 28-bit output resolution of the CX controller (red line) vs a simulated 20-bit output resolution (black line). (d) Power spectral density of a high-resolution output of the CX controller at 0 V output (black line) and 9 V output (red line), demonstrating low-noise output capability even at non-zero voltages.

FIG. 3.

(a) DriveAFM XY scanner schematic. The 1:1 direct drive scanner approach results in a very stiff and high resonance frequency scanner, important for stable and fast AFM imaging. (b) Frequency response of the DriveAFM scanner motion when actuated in Z. Despite its large scan range of 100 µm in XY and 20 µm in Z, the complete XYZ tip scanner shows a primary resonance at 3.5 kHz. (c) Scan ramp signal (72 μV) as a function of time showing improved accuracy with the full 28-bit output resolution of the CX controller (red line) vs a simulated 20-bit output resolution (black line). (d) Power spectral density of a high-resolution output of the CX controller at 0 V output (black line) and 9 V output (red line), demonstrating low-noise output capability even at non-zero voltages.

Close modal

The CX controller provides high 28-bit resolution output signals to the DriveAFM scanner, allowing very small yet smooth and accurate scan signals in comparison to the 20-bit resolution of most competitor systems, Fig. 3(c). Furthermore, the electronic design of the CX controller ensures outstanding output noise performance over the whole +/−10 V output range, not only at 0 V, Fig. 3(d). Above 10 Hz, the noise floor of <20 nV/√Hz is equal at both output values, with only a slight increase in the 1/f noise floor of the output at 9 V. Such noise performance allows the user to exploit the full performance of the DriveAFM system over the entire scan range.

CleanDrive excitation provides reliable cantilever excitation, even at small oscillation amplitudes of a few nm and in liquid environments. In this example, these capabilities were used to image double-stranded DNA (dsDNA) at high resolution in buffer solution using a 20 μm long and 10 µm wide cantilever (USC-F0.3-k0.3, Nanoworld). The detection optics of the DriveAFM permits using such small, high resonance frequency cantilevers for faster imaging than with traditional AFM systems. The DNA molecules shown in Fig. 4(a) clearly show a banding pattern all over the molecules that originates from the major and minor groves of the DNA double helix. Thanks to the stable CleanDrive excitation, the banding pattern can be stably imaged without the need for imaging parameter adjustments. The inset of Fig. 4(a) shows a surface profile of a stretch of dsDNA double helix. The profile is characterized by a wavy pattern with a periodicity of 3.42 nm for a motif of two valleys.

FIG. 4.

(a) High-resolution AFM topography image of dsDNA in buffer solution revealing secondary structure information. Inset: height profile along the line indicated. Image size: 82 nm; color scale: 1.6 nm. Inset scale bar: 5 nm/0.2 nm (x,z). (b) AFM topography overview images of HSV-1 capsids imaged before and after indentation. A high-resolution 3D height image after indentation is also shown. The force curve shows the indentation curve recorded on the virus capsid. Markings of the labels are as follows: (1) the contact point of the AFM tip with the capsid, (2) capsid perforation force, (3) compression and perforation of the second capsid layer, and (4) pushing the tip onto the substrate. The orange line indicates the linear response regime. Image sizes: overview images: 1220 × 600 nm2, color scale: 70 nm; 3D render: 700 × 400 nm2, color scale: 70 nm; force curve scale bars: 100 nm tip sample separation and 2.5 nN. (c) Moiré superlattice of twisted graphene, image size: 190 nm. (d) Superlattice also showing atomic lattice resolution; image size: 68 nm. (e) Digital zoom of (d); image size: 20 nm. (f) Fourier transformation of (d) showing the superlattice. Sample courtesy: HSV-1 capsids: Alex Evilevitch, University Lund, Sweden; double-layered graphene: Nanoelectronics group TIFR, India.

FIG. 4.

(a) High-resolution AFM topography image of dsDNA in buffer solution revealing secondary structure information. Inset: height profile along the line indicated. Image size: 82 nm; color scale: 1.6 nm. Inset scale bar: 5 nm/0.2 nm (x,z). (b) AFM topography overview images of HSV-1 capsids imaged before and after indentation. A high-resolution 3D height image after indentation is also shown. The force curve shows the indentation curve recorded on the virus capsid. Markings of the labels are as follows: (1) the contact point of the AFM tip with the capsid, (2) capsid perforation force, (3) compression and perforation of the second capsid layer, and (4) pushing the tip onto the substrate. The orange line indicates the linear response regime. Image sizes: overview images: 1220 × 600 nm2, color scale: 70 nm; 3D render: 700 × 400 nm2, color scale: 70 nm; force curve scale bars: 100 nm tip sample separation and 2.5 nN. (c) Moiré superlattice of twisted graphene, image size: 190 nm. (d) Superlattice also showing atomic lattice resolution; image size: 68 nm. (e) Digital zoom of (d); image size: 20 nm. (f) Fourier transformation of (d) showing the superlattice. Sample courtesy: HSV-1 capsids: Alex Evilevitch, University Lund, Sweden; double-layered graphene: Nanoelectronics group TIFR, India.

Close modal

With its low noise and high force resolution, the DriveAFM can be used as a tool for investigation of the mechanical properties of a variety of different samples. Figure 4(b) shows an example of imaging and mechanically probing virus capsids, which are the protein shells of viruses. The stable mechanical loop of the DriveAFM allows for imaging the virus capsids in buffer solution at high resolution, revealing the icosahedral structure and even the protein complex arrays that form the capsid’s protein cage. In this experiment that was performed on an inverted optical microscope, the virus capsids were first imaged [Fig. 4(b) (before)], mechanically indented with the AFM tip [Fig. 4(b) (force curve)], and then reimaged [Fig. 4(b) (after)] to observe the structural changes induced by the indentation. The force–distance curve reveals that before perforation, the capsid shows a linear force response with a spring constant of ∼0.3 N/m [Fig. 4(b), force curve, orange line]. Perforation of the protein layer on the opposite side of the capsid is characterized by a steep force increase followed by a small drop in force. Upon reimaging the capsids, a hole is visible in the capsid, caused by the indenting tip.

In this application, CleanDrive photothermal excitation was used to excite the cantilever on the contact resonance frequency peak during contact mode imaging (force modulation) of graphene layers twisted by 0.5°. The Moiré superlattice, Fig. 4(c), in the phase image results from variations of the contact resonance frequency when scanning the sample. The photothermal excitation and the stiff scanner design allow for zooming in to reveal both Moiré and the atomic lattice in a single image, Figs. 4(d)4(f).

The DriveAFM is a new generation AFM with a unique combination of features that are enabled by a novel patented design. Full motorization provides ease of use and the ability to adjust the system without disturbing a controlled environment. The stable mechanical design, direct drive scanner, and low-noise electronics of the CX controller yield high-performance imaging from the atomic lattice resolution up to 100 × 100 µm2 scan sizes. CleanDrive photothermal excitation and readout optics for cantilevers down to 10 μm in width provide straightforward and stable cantilever excitation, even in liquid environments, and imaging at faster rates than traditionally achievable in AFM. These features, together with the broad range of available AFM modes, accessories, and options, make the DriveAFM a powerful and truly multifunctional AFM.

The authors have no conflict to disclose.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
G.
Binnig
,
C. F.
Quate
, and
Ch.
Gerber
,
Phys. Rev. Lett.
56
,
930
(
1986
).
2.
C.
Yang
,
J.
Yan
,
M.
Dukic
,
N.
Hosseini
,
J.
Zhao
, and
G. E.
Fantner
,
Scanning
38
,
889
(
2016
).
3.
J. D.
Adams
, “
Atomic force microscope with optical guiding mechanism,
” US Patent 10,564,181 (August 27,
2018
).
4.
T. E.
Schäffer
,
J. P.
Cleveland
,
F.
Ohnesorge
,
D. A.
Walters
, and
P. K.
Hansma
,
J. Appl. Phys.
80
,
3622
(
1996
).
5.
G. C.
Ratcliff
,
D. A.
Erie
, and
R.
Superfine
,
Appl. Phys. Lett.
72
,
1911
(
1998
).
6.
J.
Kokavecz
and
A.
Mechler
,
Appl. Phys. Lett.
91
,
023113
(
2007
).