Etching of graphite surface with STM needle
At present, there is no complete understanding of the mechanism of etching the surface of graphite using STM. Thus, in  it was assumed that the surface is destroyed by electrons emitted from the needle. Most researchers suggest a mechanism for electrochemical surface oxidation. In [5, 6], a significant reduction in the tunnel voltage required for etching the surface in the presence of water vapor is indicated. It is assumed that carbon atoms react with water molecules and form gaseous carbon oxides:
CO(gas) + H2(gas),
С(graphite) + 2H2О (l.) →
CO2(gas) + 2H2(gas).
Paper  reported a decrease in the tunnel voltage for etching graphite during an experiment in an atmosphere of methanol. The authors suggest the following chemical reactions:
CH3OH (l.) + C (graphite) → CH4 (gas) + CO (gas),
2CH3OH (l.) + C (graphite) → 2HCHO (gas) + CH4 (gas).
The advantage of using methanol is associated with its low surface tension, which should lead to a reduction in the size of the meniscus and an improvement in the resolution of lithography.
Most often in the STM the gap between the probe and the sample is maintained by a feedback system provided that the tunneling current is constant. However, in the process of lithography, both the magnitude of the flowing electric current and the surface relief can vary substantially, which requires special attention to the tuning of the feedback links. The present work is devoted to the investigation of the etching of the graphite surface by the STM probe and the effects associated with the peculiarities of the feedback of the microscope.
METHODS AND MATERIALS
The experiments were carried out using the FemtoScan scanning probe microscope (Advanced Technologies Center), equipped with an STM head . The measurements were carried out in air with controlled humidity at room temperature. As probes, a mechanically sharpened wire was used from the alloy of platinum and iridium. The studies were carried out using highly oriented pyrolytic graphite with a mosaic structure of 0.4° (Atomgraph-Crystal LLC), the surface of which was chipped before the experiment. The scanning was performed with a tunnel current of 300 pA and a voltage of –50 mV. A surface area of 200 nm in size, free of defects, was selected. Lithography was performed at a scanning frequency of 7.3 Hz with a tunnel current of 300 pA. The frame size was 512 Ч 512 pixels. Voltage varied in the range from –4 to –8.5 V.
A feedback system with a proportional (Kp) and integral (Ki) links is implemented in the microscope, whose operation is described by the formula (1):
where U(t) is the signal on the z-manipulator, and E(t) is the error signal, which is proportional to the deviation of the tunnel current from the value chosen in the experiment. During the lithography, various feedback links were used.
Data processing and imaging were carried out in the FemtoScan Online software .
RESULTS AND DISCUSSION
Experiments have shown that the typical values of the feedback links used to obtain STM images are large for lithography. Even with the minimum value of the proportional link and the zero value of the integral link, needle jumps are observed with respect to the surface with an amplitude of up to 10–20 nm, and accordingly the tunnel current increases abruptly. At the zero value of the proportional link and the minimum value of the integral link, the needle loses the surface. Such needle jumps during lithography can lead to additional mechanical fracture of the surface and unstable etching results. Therefore, the data presented in Fig.1 were obtained with the minimum possible values of feedback links.
In the case of low voltage modulus, etching of the surface was not observed. When the voltage modulus was increased, at first, individual points appeared in the lithography region, which, in our assumption, are vacancies in the upper carbon layer of graphite. With further increase in the voltage modulus, the formation of cavities was observed. The increase in air humidity led to an increase in the etching rate of the surface. Thus, at –6 V and 25% relative humidity, the cavity depth was 7 nm, and at 55% – 55 nm. It should be noted that the bottom of the formed cavities is not atomically smooth, as the initial surface of graphite. It can be assumed that fragments of the destroyed graphite lattice remain at the bottom, since when scanning the cavity with the help of STM, multiple failures appear on the images.
Taking into account the dependence of cavity depth on air humidity and the character of the bottom relief, we can assume a mixed etching mechanism of graphite: emitted electrons from the needle create defects in the graphite lattice, which are followed by electrochemical oxidation.
Particular attention should be paid to the cavity with a depth of more than 10 nm (Fig.2). Their shape resembles the imprint of an STM needle made in the surface of graphite. An important role is played by feedback, since the height of the tip of the needle above the surface must constantly decrease so that it can follow the changing surface relief during etching.
When carrying out experiments with the same oxidation parameters, but with a zero proportional link, when the position of the needle does not vary in height, the depth of the cavities is much smaller (Fig.3). In particular, at a voltage of –8.5 V and a relative humidity of 15%, the surface is etched in such a way that the distance between the tip of the needle and the graph becomes equal to 10 ± 3 nm. In Fig.3, the graphite surface has a slope, so the depth of the cavity varies from 4 nm to 12 nm from left to right.
During etching, there is no direct current between the probe and the sample, although the electrochemical reaction must be accompanied by its appearance. The current value, assuming no side electrochemical reactions, can be estimated as the ratio of the amount of leakage charge (ΔQ) to the time of etching of the cavity (Δt):
I = ΔQ/Δt, (2)
ΔQ = –4en(Vox/Vgraph), (3)
where e is the electron charge, n is the degree of carbon oxidation, Vox is the volume of the cavity, Vgraph is the volume of the unit cell of graphite. Parameters of the hexagonal cell of graphite: a = 0.246 nm, c = 0.67 nm . The cell contains four atoms. Assuming an etching depth of 10 nm and oxidation of graphite to CO (n = 2), the current value is about 0.2 pA. Taking into account the microscope noise (about 10 pA), this current can not be registered.
When etching without feedback, needle jumps near the surface are absent. Despite this, the bottom of the cavities is uneven, which additionally confirms our hypothesis about the nature of the etching mechanism.
On the basis of the experimental data, it was suggested that the etching of the graphite surface by STM under the conditions of the presence of water vapor in the atmosphere has a mixed mechanism combining the destruction of the surface by electrons emitted from the needle and its local anodic oxidation.
The selection of the feedback parameters of the microscope is important, especially when etching deep structures. To improve reproducibility of results at small etching depths (up to 10 nm), it is expedient to carry out a process without feedback at a constant height of the tip of the needle along the Z coordinate above the surface. ■
The study was carried out with the financial support of the Russian Foundation for Basic Research in the framework of the scientific project № 17-52-560001.