Better Scanners for AFM-IR

Background

Motivation

AFM-IR (sometimes called nano-IR) is an exciting new analytical technology that provides researchers with nanoscale chemical analysis. The current leading technique, PiFM, can provide chemical absorption maps and FTIR-like spectra at a resolution that is 1 billion times better than even ATR-FTIR [Figure 1]. Couple that with the simple power of AFM topography, and the result is a technique that has incredible applications ranging from studying organic-inorganic biominerals in seashells, to nanoscale analysis of cancerous tissues, to defect analysis in the semiconductor industry. With such a wide range of uses, PiFM is the natural extension of both atomic force microscopy and infrared spectroscopy.

AFM-IR

The term AFM-IR refers to a handful of techniques which combine infrared (IR) spectroscopy with an atomic force microscope (AFM). AFMs are a type of scanning probe instrument where a sharp probe is raster-scanned over the surface to record topographic features. To add infrared capabilities, an IR laser must be precisely focused onto the end of a metal coated AFM tip. The metal-coated tip acts like an antenna to provide a highly localized field enhancement. The enhanced electric field will excite the sample locally, and that excitation can be measured using either the AFM tip or via detecting tip-scattered photons.

AFM-IR optical response detection

There are a few ways to detect the optical excitation of the sample. The most obvious method might be to do an optical detection as is done in nano-FTIR. However, collecting scattered photons off the tip of the AFM and excluding the far-field background signal has huge disadvantages including fussy optical alignment, poor SNR, long data collection times, and poor laser power control [nano-FTIR vs PiF-IR comparison].

Another detection technique is to measure the thermal expansion of the sample mechanically via the AFM. This is what is done in techniques like photothermal induced resonance (PTIR) which is often called tapping AFM-IR by manufactures. This technique relies on the coefficient of thermal expansion of the sample material(s). Therefore, it is most effective on organic samples since many oxides and other inorganics have small thermal expansion coefficients. This is one of the most significant limitations of tapping AFM-IR, but there are other consequences including the possibility for sample damage, tip contamination problems, resonant frequency instability, and material dependent signal convolution. All these issues stem from the fact that tapping mode AFM-IR acts in the repulsive regime of the van der Waals (vdW) force curve where the tip must hit the sample hard to induce the frequency mixing necessary for heterodyne detection. For more details, see our explanation of tapping AFM-IR.

The last way to detect the optical response of the sample in AFM-IR is to use photo-induced force microscopy (PiFM). PiFM uses mechanical detection techniques like tapping AFM-IR; however, it doesn’t make the other compromises listed before. This is because PiFM is the only true non-contact AFM-IR technique available. By operating in the attractive regime of the vdW force curve, PiFM can not only detect vdW-mediated thermal expansion forces, but it can also detect other forces including opto-mechanical damping forces and, most notably, dipole-dipole attractive forces created by local polarization of the sample from the tip-enhanced field. This is only possible if the AFM is operated in non-contact mode, and the result is a technique that works well on both organic and inorganic materials, making it far more versatile and sensitive than either nano-FTIR or tapping AFM-IR [1].

Detection considerations for PiFM

Because PiFM is based off non-contact AFM (NC-AFM), the detection scheme is important for the best IR performance. The most obvious method is to detect the forces directly. However, better spatial resolution and sensitivity is possible by detecting the force gradient instead.

The reason force gradient detection provides better results is because it increases the sensitivity of the system to the tip-sample gap. If the force interaction depends on the tip-sample spacing (r) by a factor of r^−x, then the force gradient measurement will depend on r^−(x+1). This will increase the spatial resolution of the measurement because the atoms at the apex of the AFM tip will contribute to the IR signal much more than the adjacent atoms that are a little further away. Therefore, making the IR signal more dependent on the z distance between the tip and the sample increases the resolution, and by extension the sensitivity, of the PiFM measurements. In PiFM this is done via a sideband measurement technique.

Why feedback matters

Based off the arguments above, it is clear there are two features required to make the best AFM-IR instrument possible: a non-contact AFM-based IR detection scheme to measure all relevant optical forces, and a sideband measurement technique to increase resolution and sensitivity via force gradient detection. PiFM is the only AFM-IR technique to offer both features, and this introduces some important instrument design requirements.

It may not be immediately obvious, but there is one critical instrument design consideration that is important for both a non-contact AFM and force gradient detection: precise control of the gap spacing between the tip and the sample. A non-contact AFM must have an extremely robust feedback loop to maintain the tip-sample gap without crashing the tip. Additionally, force gradient detection must maintain a constant gap spacing to avoid signal strength artifacts from a changing topography. Thus, AFM feedback is an extremely important part of any PiFM instrument.

Traditional AFM feedback

To maintain a constant tip-sample space, an AFM must have an extremely robust feedback system to track surface topography by moving the tip up and down. To put it into perspective, the job of the feedback loop is equivalent to trying to fly a 747 airliner only a few inches above the ground without crashing. The feedback loop must be able to respond fast enough to changes in the surface height so that the tip does not crash, while also not being unstable. This is like a race car driver that can respond quickly to avoid an obstacle while not over-steering and causing her own crash by losing control of the car.

In a traditional AFM the tip-sample gap spacing is controlled by moving the tip up and down to track topography. The AFM may also scan the tip in x and y, though some AFMs will instead scan the sample in x and y. Regardless, the z-scanner is usually on the tip itself. This has some significant advantages. The very small mass of the tip is easy to move. Therefore, the feedback loop can be tuned to be very sensitive while still providing many microns of z-range.

­

AFM feedback for IR-PiFM

Unfortunately, a tip-scanning system isn’t practical for PiFM. The focused IR light will have a spot size that is approximately the wavelength. For the mid-IR region, that means that the spot will be about 5 µm in diameter at the smallest. Therefore, if the AFM tip were moved up and down for scanning, the tip would be moving in and out of the IR beam. This is unacceptable as it would make IR measurements impossible. In fact, because of the field enhancement, it is best if the tip doesn’t move more than 0.5 µm. Because the beam and tip must be kept in perfect alignment with each other, AFM-IR instruments generally have the z-scanner on the sample rather than on the tip.

Moving the sample rather than the tip is not without its challenges. An AFM cantilever is minuscule compared to the mass of a sample stage plus the sample itself. Therefore, the z-scanner piezos on a sample scanning system must be much larger and stronger than they would be on a tip scanning system. Finding piezos that can move a much larger mass and still have a large range is not hard; however, these piezos cannot be as stiff and so they cannot move quickly. With such piezos, maintaining a constant gap over varying topography necessitates unacceptably slow scanning speed. One may design a sample scanner with only a small range to improve the tracking speed, but that limits the variety of samples that can be imaged. In some cases, this value could be as small as 2.5 µm which effectively limits the AFM to 2D materials since some extra range is needed for the sample slope and thermal drift between the tip-sample distance.

Dual z piezos

With IR PiFM, there is a unique feedback system to address the need for a practical z-range and constant sample-tip gap spacing: a dual z piezo system.

A dual z piezo system makes no compromises by combining the best traits of both a sample-scanning AFM and a tip-scanning AFM. In a dual z AFM, there are two z-piezos responsible for the tip-sample spacing. First there are large piezos in the sample scanner the provide a very large z-range. They can move large distances to allow the system to scan very rough samples. However, they cannot maintain constant gap spacing for fast scanning speeds. Therefore, a second piezo is added to move just the tip. This small and fast piezo can move the tip very quickly to account for errors in sample scanner feedback when tracking a complex surface topography. However, the range of the tip z-piezo is limited to only ~± 0.5 µm which will not affect the optical alignment for PiFM.

This elegant solution solves a host of problems very effectively. It allows rapid scanning of the surface in non-contact mode, ensures great topography tracking for PiFM, and maintains critical optical alignments.

Current state of AFM-IR feedback

Based on all this information, one may look at the various AFM-IR instruments available to see how they are solving this feedback problem. If having a dual z piezo system is so critical for the best AFM-IR performance, then naturally every instrument would have it. However, that is not true. In fact, as of summer 2022 Molecular Vista is the only AFM-IR company to offer dual z scanning systems. Why is that?

Fundamentally, the reason other companies do not make instruments with such sophisticated AFM feedback is that they don’t have to. Tapping AFM-IR relies on repulsive force interactions rather than attractive ones. By operating in tapping mode rather than non-contact mode one can sidestep these critical feedback issues for IR measurements. However, all the other compromises inherent in a tapping mode system are still present. The story for nano-FTIR based systems is similar. The optical scattering detection method requires a larger cantilever oscillation amplitude to effectively reject the background signal. Therefore, precise gap control is not as important. Yet, making a robust AFM is something that should be done for any quality instrument, and so there is no reason to inhibit IR performance when there is a solution that makes no compromises.

Summary

AFM-IR is the next generation for both infrared spectroscopy and atomic force microscopy. PiFM is the leading AFM-IR technique because it is based on non-contact AFM and uses force gradient detection.

Using a non-contact mode AFM is critical to get good data on the widest variety of samples and surfaces. Non-contact mode allows the system to be sensitive to a wider variety of force interactions which avoids problems such as mechanical signal convolution, dependence on thermal expansion, and poor SNR that is inherent in other AFM-IR detection schemes.

Another important consideration is the spatial resolution of the IR signal detection. To increase the resolution, one can use a PiFM sideband type detection scheme to measure the force gradient rather than the force itself. This makes the signal more dependent on the tip-sample gap spacing, which in turn increases both the resolution and sensitivity.

Because both these details are critical for getting the best PiFM data, and because both are reliant on having a perfect tip-sample gap spacing, the z-scanner is a critical component of the instrument. Many AFMs scan the tip up and down to track topography. However, the precise optical alignments required for PiFM make this option impossible, so the sample is moved instead. Because large sample z-scanners are slow, a fast z-scanner on the tip is used simultaneously to improve the tracking. This solution of using a dual z system for PiFM makes no compromises. Therefore, most people choose PiFM and a thoughtfully engineered AFM to get the best AFM-IR capabilities possible. With such care in the design, PiFM has a stunning array of applications and it offers capabilities no other analytical technique can.

References

1. J. Jahng, E. O. Potma, and E. S. Lee, “Nanoscale spectroscopic origins of photoinduced tip–sample force in the midinfrared,” Proc. Natl. Acad. Sci. USA. 116, 26359-26366 (2019)