*Article* **Specific Features of Reactive Pulsed Laser Deposition of Solid Lubricating Nanocomposite Mo–S–C–H Thin-Film Coatings**

**Vyacheslav Fominski 1,\* , Dmitry Fominski <sup>1</sup> , Roman Romanov <sup>1</sup> , Mariya Gritskevich <sup>1</sup> , Maxim Demin <sup>2</sup> , Petr Shvets <sup>2</sup> , Ksenia Maksimova <sup>2</sup> and Alexander Goikhman <sup>2</sup>**


Received: 20 November 2020; Accepted: 6 December 2020; Published: 8 December 2020

**Abstract:** This work investigates the structure and chemical states of thin-film coatings obtained by pulsed laser codeposition of Mo and C in a reactive gas (H2S). The coatings were analysed for their prospective use as solid lubricating coatings for friction units operating in extreme conditions. Pulsed laser ablation of molybdenum and graphite targets was accompanied by the effective interaction of the deposited Mo and C layers with the reactive gas and the chemical states of Mo- and C-containing nanophases were interdependent. This had a negative effect on the tribological properties of Mo–S–C–H nanocomposite coatings obtained at H2S pressures of 9 and 18 Pa, which were optimal for obtaining MoS<sup>2</sup> and MoS<sup>3</sup> coatings, respectively. The best tribological properties were found for the Mo–S–C–H\_5.5 coating formed at an H2S pressure of 5.5 Pa. At this pressure, the *x* = S/Mo ratio in the MoS*<sup>x</sup>* nanophase was slightly less than 2, and the a-C(S,H) nanophase contained ~8 at.% S and ~16 at.% H. The a-C(S,H) nanophase with this composition provided a low coefficient of friction (~0.03) at low ambient humidity and 22 ◦C. The nanophase composition in Mo–S–C–H\_5.5 coating demonstrated fairly good antifriction properties and increased wear resistance even at −100 ◦C. For wet friction conditions, Mo–S–C–H nanocomposite coatings did not have significant advantages in reducing friction compared to the MoS<sup>2</sup> and MoS<sup>3</sup> coatings formed by reactive pulsed laser deposition.

**Keywords:** reactive pulsed laser deposition; solid lubricants; nanocomposite; molybdenum sulfides; coefficient of friction; wear; diamond-like carbon

### **1. Introduction**

Researchers and practitioners turned their attention to solid lubricating coatings based on transitional metal dichalcogenides (TMDs), such as MoS2, WS2, MoSe2, and WSe2, in the 1980s [1,2]. This was due to the need to improve qualitatively the tribological properties (low coefficient of friction, durability, wear resistance) of friction units operating in a vacuum or the inert environment of a spacecraft. By that time, sufficiently effective technologies for the deposition of such coatings (mainly ion sputtering) had already been developed. This made it possible to regulate flexibly the modes of deposition and the composition of the coatings [3,4]. At the initial stage of research into these coatings, the emphasis was on finding the optimal conditions for the deposition of monophase (pure) TMD coatings and for obtaining their required composition and structure. Soon the main problem of such coatings became apparent: it was low wear resistance, especially at high contact loads. Consequently, the focus of research shifted towards the search for nanocomposite and multilayer (nanolayer) coatings. The wear resistance of nanocomposite coatings can be significantly increased by combining the plastic TMD phase with a harder/stronger or corrosion-resistant component (metals, hard carbon, metal carbides or nitrides) [5–8].

Despite the breakthroughs in the formation of nanocomposite TMD-based coatings with improved tribological characteristics, the problem of obtaining solid lubricating coatings to reduce friction under various operating conditions remains topical. This is due to both the growing demand for such coatings in traditional and new hi-tech industries (space technology, vacuum/cryogenic technology, micromechanics, etc.), and the interest in new processes that can bring about a radical change in the patterns of friction and wear. Modern studies show that it is necessary to regulate the architecture of the coating, i.e., morphology and phase composition, at the nanoscale to improve the tribological properties of nanocomposite TMD-based coatings [9–14]. In this case, triboinduced processes in the contact layer can change significantly and new nanophases can form in the tribofilm.

Nanocomposite coatings containing the TMD nanophase and hard carbon/diamond-like carbon/graphene have always stirred the intense interest of researchers since the C-based nanophase causes both the strengthening of the coating and contributes to the manifestation of the "chameleon" effect upon changing friction conditions [5,9,15]. In high air humidity, nanocomposite coatings can be used for a wider array of applications. During triboactivated interaction of nanodiamonds with the ultrafine (2D) MoS<sup>2</sup> phase, onion-like inclusions can form in the tribolayer. This leads to a decrease in the friction coefficient in vacuum to very low values (~0.005) [16]. Certain composition of coatings containing nanosized MoS<sup>2</sup> phases and diamond-like carbon demonstrate superlubricity: reduced wear can be achieved in air due to the formation of a tribofilm containing nanoscrolls of graphene-like material [17]. Many researchers note a significant effect of such graphene-like carbon nanoscrolls on friction and wear [18,19].

The majority of works on the production and study of nanocomposite TMD-based films analyse the method of deposition by ion (magnetron) sputtering of multisector targets. The method of ion sputtering has been used in many experiments and is still being perfected [20,21]. Pulsed laser deposition (PLD) is also used to form sufficiently high-quality solid lubricating TMD-based coatings [22–25]. This method differs from the more traditional magnetron sputtering since PLD gives precise control of the growth rate of the coating (up to one monolayer of the material) [26]. It allows researchers to obtain numerous combinations of different materials under controlled vacuum conditions; it is also possible to supplement deposition by the implantation of high-energy ions for ion mixing [27,28]. Yet, the PLD method has a deficiency due to the specificity of pulsed laser ablation of TMD targets. It is difficult to achieve the deposition of pure vapour (plasma) during the ablation of MoS<sup>2</sup> targets prepared by pressing MoS<sup>2</sup> powder. The ablation of the MoS<sup>2</sup> target results as a rule in the formation of microparticles/microdroplets, the deposition of which contributes to the formation of the porous structure of the coating [29,30]. During pulsed laser ablation of MoSe<sup>2</sup> and WSe<sup>2</sup> targets, explosive boiling of the material and the formation of liquid metal droplets can occur. However, for these materials, most of the particles on the surface of the coating are round in shape and ~10–100 nm in size [31–33]. Such nanoparticles do not have any noticeable negative effect on the formation of high-quality monophase and nanocomposite coatings with solid lubricating components MoSe*<sup>x</sup>* and WSe*<sup>x</sup>* [24,27,33]. Still, these nanoparticles may complicate the formation of multilayer coatings with layer thickness controlled at the nanoscale [34].

The aim of this work is to study the composition, structure, and tribological properties of Mo–S–C–H thin-film coatings formed by reactive PLD (RPLD) on steel substrates at room temperature. Reactive PLD makes it possible to form smoother and more uniform layers of a molybdenum sulphide MoS*<sup>x</sup>* and alter the ratio of elements *x* = Mo/S over a wide range (1 ≤ ≤ 4) [35]. Varying the composition of the coatings has proven to be an important factor for the development of their specific applications. For instance, the MoS<sup>3</sup> coatings of clustered type demonstrated improved antifriction properties when tested at low temperatures (−100 ◦C) and low humidity. The deficiency of these coatings is

their reduced wear resistance compared to MoS<sup>2</sup> coatings. As shown above, the wear resistance of such coatings can be improved through the formation of a nanocomposite material containing both the MoS*<sup>x</sup>* nanophase and the nanophase of hard (diamond-like) carbon. During the RPLD of a Mo–S–C–H nanocomposite coating from a target containing sectors of pure molybdenum and graphite, the deposition of a pulsed laser plume occurs in a reactive medium—hydrogen sulphide (H2S). Fominski et al. [35] revealed the dependence of the composition of solid lubricate MoS*<sup>x</sup>* thin-film coatings on the pressure of H2S. The influence of H2S on the composition and properties of carbon thin films obtained by RPLD remains unexplored, which creates difficulties in choosing optimal conditions for obtaining high-quality nanocomposite coatings Mo–S–C–H.

It was found in the work that during the ablation of graphite target in H2S, the deposited a-C(S,H) films effectively captured S and H atoms. The tribological properties of nanocomposite Mo–S–C–H coatings under various sliding conditions were largely determined by the properties of amorphous a-C(S,H) nanophases, which strongly degraded with an increase in S content. The Mo–S–C–H coating formed at a relatively low H2S pressure (~5.5 Pa) turned out to be the most promising. Overall, these coatings outperformed MoS<sup>2</sup> and MoS<sup>3</sup> monophase coatings in their tribological characteristics when tested in dry friction conditions at room and low (−100 ◦C) temperatures. In humid air at room temperature, the Mo–S–C–H coatings did not show a noticeable improvement of low friction properties in comparison with the MoS<sup>2</sup> and MoS<sup>3</sup> monophase coatings previously obtained by RPLD and studied in [35].

At first glance, the RPLD technique is not quite suitable for large area deposition, especially onto shaped work pieces used in practice. Moreover, the use of H2S gas requires special safety measures since it is explosive and toxic. However, several important issues should be highlighted in the research of this technique. Within the framework of a fundamental problem, if new nanomaterials with unique/interesting properties would be formed by RPLD, these results could initiate improving more conventional deposition techniques (e.g., ion-sputter deposition). From a practical point of view, the solvent of environmental safety problem arising due to H2S is not very complicated. In the case of the unique results of the RPLD application, the area of the covered surface can be increased by moving/rotating the processed parts under a laser-induced plume. Also, the use of pulsed electric fields applied to a shaped work piece would improve the uniformity of RPLD treatment. Obviously, the RPLD technique can be applied (and even difficult to replace) when processing small-sized parts or the inner surface of pipes/rings.

#### **2. Materials and Methods**

Reactive pulsed laser deposition technique for MoS*<sup>x</sup>* thin-film coating formation was analysed in detail in [35]. The Mo target was ablated by nanosecond laser pulses with a radiation wavelength of 1064 nm. The pulse energy did not exceed 85 mJ at a pulse repetition rate of 25 Hz. A laser fluence of ~20 J/cm<sup>2</sup> ensured efficient evaporation of the Mo target without any noticeable formation of a droplet fraction. When obtaining Mo–S–C–H films, a graphite target was placed next to the Mo target. The ablation time of the Mo target was 8 s, and the C target—4 s. Before the ablation, the film deposition chamber was evacuated with a vacuum pump to a pressure no higher than 10−<sup>3</sup> Pa. Then, hydrogen sulphide was introduced into the chamber to a predetermined pressure. The pressures were chosen taking into account the results of [35]. Fominski et al. [35] established that for obtaining films MoS1.5, MoS<sup>2</sup> and MoS3, the pressure of hydrogen sulphide had to be maintained at 5.5, 9, and 18 Pa, respectively. The prepared nanocomposite coatings were designated taking into account the H2S pressure used: Mo–S–C–H\_5.5, Mo–S–C–H\_9 and Mo–S–C–H\_18.

To reveal the effect of H2S on the carbon nanophase, additional experiments on the deposition of carbon films from a graphite target at the same H2S pressures were carried out. To better understand the effect of hydrogen sulphide on the deposition rate of Mo and C, Mo-C films were obtained under vacuum conditions (at a residual gas pressure of 10−<sup>3</sup> Pa).

We used polished discs made of 95Cr18 stainless steel (C content of 0.95% and Cr content of 18%) and polished silicon wafers as substrates for the deposition of thin-film coatings. The total deposition time of the Mo–S–C–H coatings on the steel substrates was 40 min and on silicon substrates, 20 min. The total thickness of Mo–S–C–H thin-film coatings on steel substrates was ~300-400 nm.

The substrates were kept at room temperature during the formation of the coatings. Before the deposition of the Mo–C–S–H coatings on the steel substrates, a SiC sublayer was deposited on the surface of the substrates. The sublayer was obtained by PLD from the SiC target under vacuum conditions. It was considered that the silicon carbide sublayer can increase the adhesion of C-based coatings to the steel substrate [36,37].

To determine the atomic composition of the coatings, Rutherford back-scattering spectroscopy (RBS) and elastic recoil detection analysis (ERDA) techniques were used. The energy of helium ions in the analysing beam was 1.5 MeV, and the detector resolution was 20 keV. The RBS spectra were recorded in the configuration α = 0 ◦ , β = 20◦ , θ = 160◦ . The ERDA spectra were recorded in the configuration α = 80◦ , β = 80◦ , θ = 20◦ . The measured spectra data were processed using the Simnra software (Max-Planck-Institut für Plasmaphysok, Garching bei München, Germany). The surface morphologies of the Mo–S–C–H coatings were studied using scanning electron microscopy (SEM, Tescan LYRA 3, Brno, Czech Republic) before and after friction testing. The crystal structure of the coatings was examined by grazing incidence X-ray diffraction (XRD) using an angle of 5◦ and Cu Kα radiation in an Ultima IV (Rigaku, Tokyo, Japan) diffractometer. The chemical states of the coatings were studied by X-ray photoelectron spectroscopy (XPS). The XPS spectra were obtained by a Theta Probe Thermo Fisher Scientific spectrometer (Madison, WI 53711, USA) with a monochromatic Al Kα X-ray source (1486.7 eV) and an X-ray spot size of 400 µm. The photoelectron take-off angle was 50◦ with respect to the surface plane. The spectrometer energy scale was calibrated using Au4f7/2 core level lines located at E = 84.0 eV.

The structure of the coatings before and after the friction tests was studied by micro-Raman spectroscopy (MRS). Raman spectra of the samples were collected using a Horiba Jobin Yvon micro-Raman spectrometer LabRam HR800 (Horiba, Kyoto, Japan) with a 100× magnification lens. Measurements were conducted at room temperature in air. A He-Ne laser with a 632.8 nm wavelength was used to excite Raman scattering. The irradiation power density on the sample was chosen to avoid any structural changes or phase degradation in the films. The typical measurement conditions involved a laser power of ~1 mW and a laser spot with a diameter of ~30 µm.

To study the structure of the nanocomposite coatings at the nanoscale level, thin Mo–S–C–H films were deposited on NaCl substrates. The conditions for obtaining thin Mo–S–C–H films reproduced the conditions for obtaining coatings on steel discs. Thin films were studied using transmission electron microscopy (TEM, including high-resolution HRTEM) and selected area diffraction (SAED) in a JEM-2100 microscope (JEOL Ltd., Tokyo, Japan). The films deposited on NaCl crystals were first planted in water using a metal mesh and then transferred to the microscope to obtain a planar image.

The friction testing of thin-film coatings was carried out with the help of an Anton Paar TRB3 tribometer (Anton Paar GmbH, Graz, Austria) in the reciprocating motion mode, using a steel ball (100Cr6) with a diameter of 6 mm as a counterbody. The load on the ball was 1 N, and the Hertzian contact stress was ~660 MPa. The average speed of the ball over a substrate with a Mo–S–C–H coating was 1cm/s. The length of the wear track was 5 mm. A detailed description of the technique and setup for friction testing can be found in [35]. Three conditions were selected for testing, differing in ambient humidity and substrate temperature. The first tests were carried out at 22 ◦C in air at a related humidity (RH) ~58% (wet friction conditions). The second tests were carried out at a reduced atmospheric humidity (RH ~8%, dry friction condition), which was achieved by pumping argon through the testing chamber. The sample temperature was 22 ◦C. The third series of tests was carried out at low humidity (RH ~8%) and the sample was cooled to −100 ◦C (low temperature/dry friction conditions). The wear tracks were studied by MRS, SEM, optical microscopy, and optical profilometry.

#### **3. Results**

#### *3.1. Composition of Mo–S–C–H Films Obtained by RPLD*

Figure 1 shows the experimental and simulated RBS and ERDA spectra for Mo–C and Mo–S–C–H films obtained by pulsed laser codeposition of molybdenum and carbon under vacuum conditions and in H2S gas with different pressures. For the film obtained under vacuum, mathematical processing of the spectra showed that the composition of this film was described by the formula C0.81Mo0.16H0.03. There was almost no sulphur in the bulk of o–C–H film. A small amount of sulphur was found at the boundary of the Mo–C film with the Si substrate and on the surface of the Mo–C–H film. This was possibly due to the fact that the walls of the deposition chamber were not subjected to any special treatment prior to the formation of the Mo–C–H films. The walls of the chamber were covered with a thin S-containing film, formed during previous experiments on RPLD of MoS*<sup>x</sup>* films. Sulphur atoms desorbing from the walls of the chamber could have deposited on the surface of the Si plate during vacuum pumping of the chamber and on the surface of the Mo–C–H film—after its deposition and storage (for some time) in the chamber. The presence of a small amount of hydrogen in the Mo–C–H films was possibly due to the interaction of the growing film with residual water vapour in the deposition chamber.

**Figure 1.** Experimental and simulated RBS (**left**) and ESDA (**right**) spectra of the films prepared on Si substrates by pulsed laser co-deposition of carbon and molybdenum under vacuum conditions (residual gas pressure was ~10−<sup>3</sup> Pa) and in H2S gas with different pressures. The RBS spectrum of the film deposited in vacuum contains a peak at the channel number 230. This peak is due to scattering of ions by sulfur atoms that have been adsorbed on the surface of the Si substrate before the film deposition.

According to the RBS data, the thickness of the Mo–C–H film obtained during 20 min of deposition was 8.2 <sup>×</sup> <sup>10</sup><sup>17</sup> atom/cm<sup>2</sup> . In reactive H2S gas, S atoms penetrated the deposited Mo–C–S–H film. Despite this fact, the overall deposition rate of atoms decreased. For hydrogen sulphide pressures of 5.5, 9, and 18 Pa, the composition of the films was described by the formulas C0.49Mo0.13S0.28H0.1, C0.405Mo0.13S0.37H0.095, and C0.24Mo0.115S0.55H0.095 respectively. The thickness of these films was 6.2 × <sup>10</sup>17, 5.8 <sup>×</sup> <sup>10</sup>17, and 5.2 <sup>×</sup> <sup>10</sup><sup>17</sup> atom/cm<sup>2</sup> . For ablation of graphite and molybdenum targets, 1.5 <sup>×</sup> <sup>10</sup><sup>4</sup> laser pulses were used, which were divided into series of 100 pulses for the C target and 200 pulses for the Mo target. The ablation of the C target resulted in the deposition of ~2 <sup>×</sup> <sup>10</sup><sup>15</sup> atom/cm<sup>2</sup> in hydrogen sulphide. This was sufficient for the formation of approximately one monolayer of amorphous carbon, given the atomic density in amorphous carbon of ~(1 <sup>÷</sup> 1.7) <sup>×</sup> <sup>10</sup><sup>23</sup> atom/cm<sup>3</sup> . The same estimates for Mo showed that a 0.6 ÷ 1 MoS*<sup>x</sup>* monolayer could be formed after 200-pulse ablation of the Mo target in hydrogen sulphide.

Figure 2 shows a change in the rate of deposition of various elements in the Mo–C–H and Mo–S–C–H films following an increase in the hydrogen sulphide pressure. As it is shown, H2S gas did not only ensure the saturation of the films with S atoms, but also had a significant effect on the rate of carbon deposition. At a pressure of 18 Pa, the rate of carbon deposition dropped almost fivefold compared to the rate of deposition in a vacuum. Under the same conditions, the rate of deposition of Mo atoms decreased only twofold. This could be explained by the difference of collisions when light (C) and heavy (Mo) atoms moved through H2S gas. The relatively heavy Mo atoms changed their trajectory only slightly and slowly lost their kinetic energy in collisions with the relatively light H2S molecules. When colliding with the H2S molecules, the lighter C atoms are scattered at large angles and leave the area where the deposition on the substrate took place. The possibility of reactive collisions of carbon ions with H2S molecules cannot be ruled out. These collisions may have resulted in the formation of volatile hydrocarbon molecules. This is a possible explanation as to why an increase in the H2S pressure did not result in a discernible change in the rate of the saturation of the films with hydrogen.

**Figure 2.** Influence of hydrogen sulfide pressure on the composition of Mo-S-C-H films which were obtained by pulsed laser codeposition of carbon and molybdenum in 20 min.

The analysis of the RBS and ERDA data for a-C(S,H) films formed during pulsed laser ablation of graphite in H2S confirmed the assumption about the effective interaction of the ablated flux of carbon atoms and H2S molecules (see Figure S1, Supplementary Materials). An increase in the pressure of hydrogen sulphide resulted in a noticeable increase in the concentration of sulphur in the a-C (S, H) films. The compositions of the films at hydrogen sulphide pressures of 5.5, 9, and 18 Pa were described by the formulas C0.71S0.15H0.14, C0.61S0.26H0.13, and C0.42S0.4H0.18. The thicknesses of the

films were 2.2 <sup>×</sup> <sup>10</sup>18, 1.7 <sup>×</sup> <sup>10</sup><sup>18</sup> and 1.0 <sup>×</sup> <sup>10</sup><sup>18</sup> atom/cm<sup>2</sup> respectively. During the PLD of carbon films in vacuum (residual gas pressure ~10−<sup>3</sup> Pa), their thickness was 2.88 <sup>×</sup> <sup>10</sup><sup>18</sup> atom/cm<sup>2</sup> , and the H atoms concentration did not exceed 1 at.%.

#### *3.2. Morphology, Structure, and Chemical State of Mo–S–C–H Films Obtained by RPLD*

The use of RPLD for obtaining Mo–S–C–H thin films made it possible to produce sufficiently smooth and dense coatings, the morphology of which, according to the results of SEM studies (Figure 3), did not depend on the H2S pressure. There were individual round-shaped particles of a submicron size on the surface of the Mo–S–C–H coatings. These particles could have formed because of the deposition of droplets formed during the ablation of the Mo and graphite targets. Such particles were found in a-C(S,H) films produced by RPLD from a graphite target in H2S gas (Figure S2, Supplementary Materials).

**Figure 3.** SEM images of the surface of Mo–S–C–H thin-film coatings obtained by reactive pulsed laser deposition (PLD) on steel substrates at the following H2S pressures: (**a**) 5.5; (**b**) 9; (**c**) 18 Pa.

The results of the XRD of Mo–S–C–H thin-film coatings (Figure 4) show the deposition of Mo particles upon ablation of the Mo target. The X-ray diffraction pattern of the Mo–S–C–H\_5.5 coating had a weak intensity peak, which corresponded to the (110) reflection for a body-centred cubic Mo lattice. The peak was practically invisible in the X-ray diffraction patterns of the Mo–S–C–H\_9 and Mo–S–C–H\_18 coatings. This could be attributed to the fact that with an increase in the pressure of H2S gas, the surface of the Mo target interacted with the reactive gas. Following this interaction, molybdenum sulphides could form on the target surface; this changed to a certain extent the mechanism of pulsed laser ablation of the Mo target.

XRD studies showed that at all selected pressures of H2S, the Mo–S–C–H thin-film coatings had an amorphous structure with a broadened diffraction peak in the angle range from 35◦ to 50◦ . With greater hydrogen sulphide pressure, the intensity of this peak noticeably weakened. This indicated an increased disordering of the structure following a rise in the sulphur concentration. This type of XRD pattern has been extensively described in the literature; in most experiments, the Mo–S–C coatings have been obtained by ion sputtering/codeposition from MoS<sup>2</sup> and graphite targets (for example, [15,38–40]). For TMD coatings having an amorphous structure, this broad peak is usually explained by the formation of nanosize inclusions with a hexagonal lattice of the 2H-MoS<sup>2</sup> type [41]. In the cases when there was no peak at angles 2θ~13◦ , but there was a peak in the 2θ range from 35◦ to 50◦ , the turbostratic stacking of (10*L*) planes into Type I texture was supposed. With this texture, the basal planes (002) are oriented perpendicular to the surface of the substrate [42]. The absence of a peak at 2θ~13◦ shows that the reactive PLD of Mo–S–C–H\_5.5 films in H2S may not have caused the formation of a self-assembled multilayer structure MoS*x*/a-C (doped with Mo/S/H) with a periodicity in the nanometer scale, as it was the case during the magnetron sputtering of graphite and MoS<sup>2</sup> targets in Ar/N<sup>2</sup> gases [14,39]. In XRD patterns for the Mo–S–C–H\_18 coatings, a weak-intensity and a very broad band appears at 2θ~15◦ . This shows that a MoS*<sup>x</sup>* nanophase with a high sulphur concentration (

≥ 3) formed in the structure of these coatings. Such coatings are characterized by an XRD pattern with two strongly broadened bands at 2θ~15◦ and 2θ~40◦ [35,43].

**Figure 4.** In plane grazing incidence X-ray diffraction patterns of Mo-S-C-H thin-film coatings obtained on steel substrates by the reactive PLD at various pressures of H2S gas. For comparison, X-ray diffraction pattern for the bare steel substrate is shown.

HRTEM studies of the Mo–S–C–H thin films confirmed their amorphous structure. Only in the Mo–S–C–H\_5.5 films obtained at the lowest H2S, pressure, MoS<sup>2</sup> nanocrystallites with laminar packing of atomic planes were found in some local regions. The size of these crystallites did not exceed 10 nm, and they were surrounded by an amorphous matrix. The concentration of MoS<sup>2</sup> nanocrystallites in the Mo–S–C–H\_5.5 film was not high, and their structure probably had a turbostratic character with a high degree of disordered local atomic packing. This was confirmed by the SAED pattern, consisting only of diffusely broadened rings (Figure 5), as well as by the results of the Raman studies of Mo–S–C–H films.

**Figure 5.** High-resolution TEM image of the Mo-S-C-H thin film obtained by reactive PLD at an H2S pressure of 5.5 Pa.

The Raman spectrum for the Mo–S–C–H\_9 coating in the frequency range of 100–600 cm−<sup>1</sup> was in many respects similar to the spectrum of the Mo–S–C–H\_5.5 coating (Figure 6). Broad peaks at 350 and 402 cm−<sup>1</sup> indicated the formation of the MoS<sup>2</sup> nanophase with a disordered atomic packing. The appearance in the spectrum of the Mo–S–C–H\_9 coating of weak-intensity and broad peaks at ~200 and ~500 cm−<sup>1</sup> suggested that, along with the MoS*<sup>x</sup>* nanophase, Mo3–S clusters could form. When such clusters are combined into a polymer-like network, MoS*<sup>x</sup>* compounds are formed, in which ≥ 3 (Mo3S12/Mo3S13-type). The composition of Mo3-S clusters includes three Mo atoms connected in the Mo3–S triangle through monomers and/or dimers of S atoms (S2−/S<sup>2</sup> <sup>2</sup>−). With a sufficiently ordered packing of atoms in such clusters, narrow peaks are observed in the indicated frequency range; they correspond to various sulphur ligands [35,43–45]. In the Raman spectrum for the Mo–S–C–H\_18 coating, the Mo3–S clusters corresponded to peaks at the following vibration modes: ν(Mo-Mo) at

~210 cm−<sup>1</sup> , ν(Mo-S)coupled at ~330 cm−<sup>1</sup> , ν(Mo-Sapical) at ~450 cm−<sup>1</sup> , ν(S-S)terminal at ~520 cm−<sup>1</sup> , and ν(S-S)bridging at 550 cm−<sup>1</sup> . In addition to these peaks, the spectrum of this coating exhibited peaks at 360, 380, and 401 cm−<sup>1</sup> , which could be due to atomic vibrations in the defective MoS<sup>2</sup> nanophase.

**Figure 6.** Raman spectra for the Mo–S–C–H thin-film coatings obtained by reactive PLD in the H2S gas at pressures of (**a**) 5.5, (**b**) 9, and (**c**) 18 Pa. The regions of Raman shifts corresponding to resonance light scattering by MoS*x*- and a-C(S,H)-based nanophases are shown at the top and bottom, respectively. The model of spectrum decomposition into the indicated peaks is discussed in the text. Inserts show the Raman spectra in the region from 100 to 2000 cm−<sup>1</sup> that allows a correct comparison of the peak intensities for different nanophases.

When choosing a model for the decomposition of the Raman spectra for C-based nanophase in Mo–S–C–H films, we took into account the changes in the Raman spectra for a-C(S,H) films with increasing hydrogen sulphide pressure. The Raman spectra for a-C(S,H) films are shown in Figure S3 (Supplementary Materials). Figure S3 shows that, as the H2S pressure grows, i.e., with an increase in the concentration of sulphur and hydrogen in a-C(S,H) films, the contribution to the Raman spectra of the two peaks at frequencies of ~1220 cm−<sup>1</sup> and ~1440 cm−<sup>1</sup> rises as well. The intensity of these peaks in films having the highest concentration of S exceeded the intensity of the D (at ~1340 cm−<sup>1</sup> ) and G (at ~1530 cm−<sup>1</sup> ) peaks characteristic of pure a-C films. In this case, with an increase in the sulphur concentration, the *I*D/*I*<sup>G</sup> ratio grew, which indicated an increase in the disordering (defectiveness) of atomic packing in graphite clusters.

Comparative analysis of the Raman spectra for the C-based nanophase in Mo–S–C–H and a-C (S,H) films showed (Figure 6) that the addition of Mo atoms to the depositing flux did not cause significant changes in the Raman spectra for the C-based nanophase. This was confirmed by the fact that the spectra of Mo–S–C–H coatings and a-C (S, H) films in the frequency range of 1000–1800 cm−<sup>1</sup> were similar in many respects. An increase in the H2S pressure during RPLD of the Mo–S–C–H coatings led to an increase in the contribution to the spectrum of lines at ~1220 and ~1440 cm−<sup>1</sup> . Our analysis of the published data on Raman studies of Mo–S–C films formed by codeposition under magnetron sputtering (including the reactive one in an Ar/CH<sup>4</sup> mixture) showed that the spectra of these films did not have properties characteristic of the spectra of the Mo–S–C–H films produced by RPLD. In the spectra of Mo–S–C films for the C-based nanophase, the positions of the D and G peaks, as well as the ratio of their intensity, tended to change, but new peaks did not appear [14,15,38,39,46,47]. Changes in the Raman

spectra were caused by the influence of the MoS<sup>2</sup> nanophase on the sp<sup>2</sup> /sp<sup>3</sup> ratio (graphitization) and the level of mechanical stresses in the a-C(H) nanophase. More significant changes in the Raman spectrum of the carbon component in the a-C(S,H) films were found when using chemical vapour deposition in H2S, as well as during magnetron sputtering and pulsed laser ablation of composite targets made of a mixture of powders (MoS2, sulphur, graphite) [48–50]. Unfortunately, these works do not contain a sufficiently detailed analysis of the Raman spectra. Therefore, to investigate the C-based nanophase in the a-C(S,H) and Mo–S–C–H films obtained by RPLD, we used the approach proposed by Takeuchi et al. [51] for organic carbon sulphur materials.

Takeuchi et al. [51] have identified a class of organic carbon sulphur materials, the Raman spectrum of which has peaks at ~1250, 1350, 1440, and 1590 cm−<sup>1</sup> . The position of each peak has a tolerance of <sup>±</sup>50 cm−<sup>1</sup> . The structure of such materials depends on the ratio of the intensities of these peaks. If the peak at 1400 cm−<sup>1</sup> is the most intense, there is a large amount of the sp<sup>3</sup> component of the G-band, and the majority of the carbon component form an undeveloped graphene (C-C) skeleton. Other peaks correspond to the sp<sup>3</sup> component of the D band (~1250 cm−<sup>1</sup> ), the sp<sup>2</sup> component of the D band (~1350 cm−<sup>1</sup> ), and the sp<sup>2</sup> component of the G band (~1590 cm−<sup>1</sup> ). The S-S bond stretching vibration should peak at ~480 cm−<sup>1</sup> . This peak is present in the Raman spectrum of the Mo–S–C–H\_18 coating (Figure 6c). The low intensity of this peak shows that RPLD was not effective for the formation of sulphur clusters. The process of dispersing sulphur in the carbon nanophase turned out to be more productive and caused a change in the local packing of carbon atoms and in the structure of the carbon skeleton. With a rise in the H2S pressure, i.e., with an increase in the concentration of sulphur, the contribution from the sp<sup>3</sup> states caused by the introduction of sulphur in the structure of the carbon skeleton grew. At the same time, an increase in the intensity of the *I*<sup>D</sup> peak at 1350 cm−<sup>1</sup> (compared to *I*<sup>G</sup> at 1540 cm−<sup>1</sup> ) was indicative of a growing number of defects in the atomic packing of pure graphite clusters. Low intensity peak at 1080 cm−<sup>1</sup> should be introduced for better fitting of the Raman spectrum.

The RBS technique made it possible to determine the concentration of sulphur in the nanocomposite Mo–S–C–H coatings. This technique nevertheless does not distinguish between the sulphur content in MoS*<sup>x</sup>* and a-C(S,H) nanophases. Therefore, XPS measurements were carried out. Figure 7 shows the XPS spectra, revealing chemical bonds in the surface layer of the Mo-S-C-H coatings formed by RPLD at various H2S pressures. Decomposition of the Mo 3d spectrum showed that the chemical state of Mo atoms did not undergo significant changes with the increasing pressure of H2S. The Mo 3d spectra contained Mo3d5/2-Mo3d/<sup>2</sup> doublets, corresponding to Mo2+, Mo4+, Mo5+, and Mo6+. The electron binding energies for the Mo3d5/<sup>2</sup> peaks at such valences of molybdenum were 228.6, 229.2, 230.3, and 232.6 eV, respectively. The dominance of the Mo4<sup>+</sup> doublet indicated the effective formation of MoS<sup>2</sup> and/or MoS*<sup>x</sup>* compounds (with packing Mo3-C), in which <sup>≥</sup> 3 [35,43,52,53]. The Mo5<sup>+</sup> doublet may have been a result of the binding in the MoS<sup>3</sup> compound with a linear packing of atoms into Mo-S<sup>3</sup> clusters [35,53,54]. The presence of a Mo6<sup>+</sup> doublet with a low relative peak intensity indicated weak surface oxidation and the formation of Mo–O compounds [35,55]. With increasing H2S pressure, the intensity of the Mo6<sup>+</sup> doublet weakened even more due to the chemical properties of hydrogen sulphide, which is a strong reducing agent. An increase in the H2S pressure caused a decrease in the contribution of the Mo2<sup>+</sup> doublet, which corresponded to the Mo–C (Mo2C) bonds [55,56]. This was due to both an increase in the total sulphur concentration in the Mo–S–C–H films with a rise in the H2S pressure and, probably, due to an increased chemical activity of radicals formed upon activation of H2S by a laser plasma, compared with carbon atoms in a laser plasma from a graphite target. Considering the small contribution of the Mo-C states, the effect of the carbide nanophase on the properties of M–S–C–H coatings was not considered in this work.

In the decomposition of the C 1s spectra for Mo–S–C–H coatings, it was assumed that C atoms could form chemical bonds with each other (C=C binding energy 284.6 eV and C-C binding energy 285.5 eV), with S atoms (C-S energy bonds 286.5 eV), and Mo atoms (C-Mo binding energy 283.6 ÷ 284.2 eV) [36,37,56,57]. The peak with the highest binding energy (~289 eV) is usually attributed to C-O

bonds [57]. The analysis of the C 1s spectra showed that, at all H2S pressures, the peak corresponding to the sp<sup>2</sup> bonds of C atoms dominated. As the H2S pressure grew, the contribution from the peak corresponding to C-S bonds increased too. In this case, the contribution of the peak at 285.5 eV, corresponding to sp<sup>3</sup> bonds of carbon atoms, slightly decreased. A thin film of organic contaminants containing CH*<sup>x</sup>* molecules may form on the surface of Mo–S–C–H coatings after being blown out of the PLD chamber. The presence of this film could have a definite effect on the results of studying the chemical state of carbon in Mo–S–C–H coatings, first of all, it could have increase the intensity of the XPS peak binding energy of ~284.5 eV.

**Figure 7.** XPS spectra of C 1s, Mo 3d, and S 2p which were measured on the surface of the Mo–S–C–H thin-film coatings obtained by reactive PLD at H2S pressures of (**a**) 5.5, (**b**) 9, and (**c**) 18 Pa. For Mo 3d spectra, the S 2s spectra of different S species are indicated.

The deconvolution of the S 2p spectra for Mo–S–C–H coatings allowed us to assume that S atoms can form chemical bonds with Mo atoms, which are characteristic of MoS*<sup>x</sup>* compounds with different values of *x*, as well as of chemical bonds with C atoms. The most common approach to analysing the chemical state of S atoms in molybdenum sulphides is the separation of two doublets S 2p3/2-S 2p1/<sup>2</sup> having "low" and "high" binding energies. A doublet with a low binding energy (the binding energy of the S 2p3/<sup>2</sup> peak does not exceed 162.3 eV, and the S 2p1/2—163.5 eV) usually corresponds to the S <sup>2</sup><sup>−</sup> species characteristic of the MoS*<sup>x</sup>* compound [35,58]. A doublet with a high binding energy (the binding energy of the S 2p3/<sup>2</sup> peak is ~162.8 ÷ 163.4.4 eV) corresponds to the S<sup>2</sup> 2 -species, which are characteristic of MoS*<sup>x</sup>* compounds, where clusters of Mo-S<sup>3</sup> and/or Mo3-S are formed due to a high concentration of sulphur ( > 2), [35,43,52,53]. S atoms in the chemical bond with C atoms (-S-C-S-) correspond to the doublet S S 2p3/2-S 2p1/2, in which the binding energy of the S 2p3/<sup>2</sup> peak is 163.6 eV, and the S 2p1/<sup>2</sup> peak equals 165.2 eV [57]. The choice of a model for describing the chemical state of S in a carbon matrix seems to be quite problematic since the binding energies of S atoms can strongly depend on the configuration of the nearest atoms. Thus, for a certain configuration of chemical bonds, the spin-orbit splitting of the S 2p state does not occur. Our analysis of the published results of XPS studies of sulphur-doped carbon materials has shown that even in the absence of spin-orbit splitting, a band at ~163.5 eV can dominate in the XPS spectra of S 2p (for example, a configuration of the C-S1÷2-C type), together with which a band at 165.0 eV (for instance, a configuration of the -C=S-type) appears [59,60].

The application of the chosen model of the decomposition of the S 2p spectra showed that an increase in the H2S pressure resulted in a decrease in the concentration of S2<sup>−</sup> states, and the contribution to the S<sup>2</sup> <sup>2</sup><sup>−</sup> states increased. The contribution of the species corresponding to the C-S bonds increased as well. The calculation of the ratio *x* = S/Mo, taking into account S species (S2<sup>−</sup> + S<sup>2</sup> <sup>2</sup>−) associated with Mo, and Mo species (Mo4<sup>+</sup> + Mo5+) associated with S, showed that it was approximately 1.8, 2.5, and 4.0 for the Mo–S–C–H coatings obtained at pressures of 5.5, 9, and 18 Pa respectively. The composition of the C component in these coatings was described by the approximate formulas C0.78S0.08H0.14, C0.73S0.11H0.16, and C0.62S0.18H0.2. We assumed that H atoms are concentrated mainly in the a-C(S,H) nanophase. The calculated composition of the a-C(S,H) nanophase for Mo–S–C–H composite films differed from the composition of a-C(S,H) coatings obtained by RPLD at similar H2S pressures. The concentration of S atoms in the nanophase was lower than in the monophase thin-film coating. This could be attributed to the fact that S atoms deposited on the surface of the growing layer from the gas phase during ablation of a graphite target can be captured during the formation of the MoS nanophase in the course of the subsequent ablation of the Mo target. Our calculations have shown that, as a result of the codeposition of Mo and C in reactive gas, the ratio = S/Mo exceeds the ratio obtained earlier for MoS*<sup>x</sup>* films produced by RPLD of molybdenum in H2S at the same gas pressures.

#### *3.3. Tribological Properties of Mo–S–C–H Films Obtained by RPLD*

Figure 8 shows the results of measuring the average coefficient of friction as a function of the sliding cycle number of the steel counterbody over the Mo–C–S–H coatings in humid air. The Mo–C–S–H\_5.5 coating obtained at the lowest H2S pressure turned out to be the most wear-resistant. The endurance of this coating exceeded 10<sup>3</sup> cycles. For other coatings, the wear resistance did not exceed 200 cycles. The analysis of the wear tracks and the wear scar showed (Figure 9) that the low wear resistance of the Mo–C–S–H\_9 and Mo–C–S–H\_18 coatings is mainly due to the weak adhesion of the coatings to the substrate. The sliding of the counterbody caused the cracking of these coatings accompanied by the separation of microplates. Microplates accumulated around the track and adhered to the counterbody as well (Figure 9b,c).

The minimum value of the friction coefficient for the Mo–S–C–H\_5.5 coating was 0.08; it was achieved after 10 sliding cycles. After 100 cycles, the coefficient of friction rapidly increased to 0.22 (±0.05), and this value remained constant throughout the entire testing period. A profilometric study of the wear crater showed that the wear rate of Mo–S–C–H\_5.5, when tested in a humid atmosphere, was ~9 <sup>×</sup> <sup>10</sup>−<sup>7</sup> mm<sup>3</sup> /N m.

**Figure 8.** Characteristic evolution of the friction coefficient as a function of the cycle number for the Mo–S–C–H thin-film coatings obtained by reactive PLD at the pressures of H2S gas of 5.5, 9 and 18 Pa. Pin-on-disk tribometer testing was conducted in wet friction conditions (RH ~58%) at 22 ◦C.

**Figure 9.** Optical images of wear tracks and wear scars formed on the steel substrates and steel balls for the Mo–S–C–H thin-film coatings obtained by reactive PLD at the different pressures of H2S gas: (**a**) 5.5, (**b**) 9, and (**c**) 18 Pa. Pin-on-disk tribometer testing was conducted in wet friction conditions (RH ~58%) at 22 ◦C. The test durations are indicated in Figure 8.

The Mo–S–C–H\_5.5 coating showed better tribological properties compared to Mo–S–C–H\_9 and Mo–S–C–H\_18 coatings when tested in a dry atmosphere at room and low temperatures. Figure 10 demonstrates that, in dry friction conditions at 22 ◦C, the average coefficient of friction after the running-in period gradually increased from 0.03 (±0.05) to 0.05 (±0.05) with an increase in the test duration from 10 to 4 <sup>×</sup> <sup>10</sup><sup>3</sup> cycles. The coating showed good adhesion to the substrate (Figure 11a). The wear rate of this coating was ~3 <sup>×</sup> <sup>10</sup>−<sup>7</sup> mm<sup>3</sup> /N m. The Mo–C–S–H\_9 coating also had fairly good antifriction properties and durability despite its poor adhesion to the substrate. For this coating, the average coefficient of friction did not exceed 0.08 during the entire testing period (i.e., 4 <sup>×</sup> <sup>10</sup><sup>3</sup> cycles). Weak adhesion of the coating to the substrate manifested itself in the formation of coating delamination areas in the track area. Coating separation caused the formation of micro-scales, which accumulated at the edges of the track (Figure 11b). The Mo–S–C–H\_18 coating had poor tribological properties: it began to deteriorate immediately after the testing had started because of its weak adhesion to the substrate. Microscopic analysis showed that the loose fragments of the coating effectively adhered to the counterbody (Figure 11c).

**Figure 10.** Characteristic evolution of the friction coefficient as a function of the cycle number for the Mo-S-C-H thin-film coatings obtained by reactive PLD at the pressures of H2S gas of 5.5, 9, and 18 Pa. Pin-on-disk tribometer testing was conducted in dry friction conditions (air + Ar mixture, RH ~8%) at 22 ◦C.

**Figure 11.** Optical images of wear tracks and wear scars formed on the steel substrates and steel balls for the Mo-S-C-H thin-film coatings obtained by reactive PLD at the different pressures of H2S gas: (**a**) 5.5, (**b**) 9, and (**c**) 18 Pa. Pin-on-disk tribometer testing was conducted in dry friction conditions (air + Ar mixture, RH ~8%) at 22 ◦C. The test durations are indicated in Figure 10.

During testing in dry friction conditions at −100 ◦C, the average coefficient of friction for the Mo–S–C–H\_5.5 coating did not exceed 0.08 (±0.1) over 10<sup>3</sup> sliding cycles (Figure 12). A shallow track formed on the coating surface, and the wear rate of the coating did not exceed 1.6 <sup>×</sup> <sup>10</sup>−<sup>7</sup> mm<sup>3</sup> /N m (Figure 13a). The sliding of the counterbody over the Mo–C–S–H\_9 and Mo-C-S-H\_18 coatings was accompanied by noticeable changes in the average coefficient of friction in the range from 0.05 to 0.25 (Figure 12). In this case, the coatings retained their continuity, but they could deform and crack (Figure 13b,c). Sliding of the ball on the Mo–C–S–H\_9 and Mo–C–S–H\_18 coatings caused more intensive wear of the counterbody than sliding on the Mo–S–C–H\_5.5 coating.

**Figure 12.** Characteristic evolution of the friction coefficient as a function of the cycle number for the Mo-S-C-H thin-film coatings obtained by reactive PLD at the pressures of H2S gas of 5.5, 9, and 18 Pa. Pin-on-disk tribometer testing was conducted in dry friction conditions at −100 ◦C.

**Figure 13.** Optical images of wear tracks and wear scars formed on the steel substrates and steel balls for the Mo-S-C-H thin-film coatings obtained by reactive PLD at the different pressures of H2S gas: (**a**) 5.5, (**b**) 9, and (**c**) 18 Pa. Pin-on-disk tribometer testing was conducted in dry friction conditions at −100 ◦C. The test durations are indicated in Figure 12.

#### **4. Discussion**

Our comparison of the tribological properties of the Mo–S–C–H coatings obtained at different pressures of hydrogen sulphide showed that an increase in pressure negatively affected both the average coefficient of friction and the wear resistance of the coatings under various tribological testing conditions. To explain this result, it is necessary to assess the possible effect of the nanophases of these coatings on the tribological properties. A change in the conditions of RPLD caused significant changes in both the MoS*<sup>x</sup>* and C-based nanophase. An increase in the H2S pressure caused a rise in the *x* = S/Mo ratio from ~1.8 to ~4.0. Fominski et al. [35] found that increasing *x* to 4 can significantly worsen the tribological properties of MoS*<sup>x</sup>* coatings. For this reason, the generally unsatisfactory properties of the Mo–S–C–H\_18 coatings could be caused by inclusions of the MoS<sup>4</sup> nanophase. At 2 ≤ ≤ 3, the properties of MoS*<sup>x</sup>* coatings depend on the conditions of tribological tests. In a humid atmosphere, after 400 cycles of sliding, MoS<sup>2</sup> and MoS<sup>3</sup> monophase coatings had the friction coefficient of ~0.2 and ~0.12, respectively. Additional studies of MoS*<sup>x</sup>* thin film coatings formed by the RPLD at H2S gas pressure of 5.5 Pa revealed unsatisfactory tribological properties. Under various tribotest conditions, these coatings began to break down almost immediately after the start of the counterbody sliding (Supplementary Materials, Figure S4). The Mo–S–C–H\_9 coating, containing MoS~2.5 nanophase, did not show good antifriction properties in a humid atmosphere, which indicates the possibility of a negative effect of another nanophase, which is part of this coating—the a-C(S,H) component formed at H2S pressure of 9 Pa. The antifriction properties of the Mo–S–C-H\_5.5 coating containing the MoS1.8 nanophase correlated rather well with the properties of a single-layer MoS<sup>2</sup> coating.

For comparing the tribological properties of the Mo–S–C–H coatings produced by RPLD with those of Mo–S–C coatings obtained by more traditional deposition methods (mainly, magnetron sputtering), it is necessary to take into account a number of important factors, such as the composition of the coatings, air humidity and the load applied on the counterbody [15,61]. In the friction tests of the Mo–S–C–H\_9 and Mo–S–C–H\_18 coatings in a humid atmosphere with an increased load on the ball (5 N), the average friction coefficient was ~0.066, and the endurance did not change with an increase in the load and was 100 cycles. The friction coefficient for the MoS<sup>2</sup> and Mo–S–C coatings at an increased carbon concentration (≥30 at.%) in a humid atmosphere (RH ≥ 50%) varied from ~0.1 to 0.3, and the wear rate varied from ~5 <sup>×</sup> <sup>10</sup>−<sup>7</sup> mm<sup>3</sup> /N m to 20 <sup>×</sup> <sup>10</sup>−<sup>7</sup> mm<sup>3</sup> /N m. The best performance is achieved only by alloying these coatings with metals (Ti, Pb, and others) [62–64]. Thus, it can be assumed that the Mo–S–C–H\_5.5 nanocomposite coating, when tested in a humid atmosphere (RH ≥ 50%), is inferior in its tribological properties only to the best samples of the MoS<sup>2</sup> and Mo–S–C coatings doped with metals.

During friction testing in dry friction conditions at room temperature, the coefficient of friction for the MoS<sup>2</sup> and MoS<sup>3</sup> coatings was 0.08 and 0.1 respectively after 400 cycles of sliding [35]. To compare, the Mo–S–C–H\_9 coating containing the MoS~2.5 nanophase also provided a fairly stable and low coefficient of friction (~0.08). However, the Mo–S–C–H\_5.5 coating containing the MoS1.8 nanophase provided a more effective decrease in the friction coefficient (down to 0.03), which was probably due to the positive effect of the a-C(S,H) component formed at an H2S pressure of 5.5 Pa. The friction coefficient value of ~0.05 has been noted in many studies of pure TMD and nanocomposite TMD+C coatings when tested under dry friction conditions (~5% ≤ RH ≤ ~30%) at moderate loads on the counterbody (for example, [15,47,61,65–67]). The wear rate can be reduced to ~2 <sup>×</sup> <sup>10</sup>−<sup>7</sup> mm<sup>3</sup> /N m, which is achieved by alloying with metals. Fundamentally lower values of the friction coefficient (~0.005 ÷ 0.01) during sliding in dry friction conditions are achieved by creating nanoscale layers of MoS<sup>2</sup> and a-C(H) [17,68]. The friction coefficient for the Mo–S–C–H\_5.5 coating equal to 0.03 at a humidity of RH ~8% turned out to be slightly lower than the values for the MoS*<sup>x</sup>* and Mo–S–C coatings formed by magnetron sputtering. Yet this coating was clearly inferior to the MoS2/a-C() nanolayer coatings exhibiting superlubricity properties. This could be caused by both the suboptimal composition of the coating nanocomponents and by the fact that the a-C(S,H) nanophase formed during RPLD did not provide ultralow friction in combination with the MoS1.8 nanophase.

For the MoS<sup>2</sup> and MoS<sup>3</sup> monophase coatings, the average coefficient of friction under extreme test conditions (−100 ◦C) was 0.18 and 0.08 respectively after 400 cycles of sliding [35]. The Mo–S–C–H\_9 coating containing the MoS~2.5 nanophase provided sliding with a higher coefficient of friction (0.2–0.3). A lower and stable coefficient of friction (~0.08) was determined for the Mo–S–C–H\_5.5 coating containing the MoS1.8 nanophase. At this stage, it is difficult to do a comparative analysis of the tribological properties of Mo–S–C–H coatings obtained by RPLD and coatings of the same type obtained by other techniques since there is no information on tribotests of various coatings at −100 ◦C in the literature. A comparison of the tribological properties of the monophase MoS*<sup>x</sup>* and nanocomposite Mo–S–C–H coatings indicates a significant effect of the a-C(S,H) phase on the tribological properties of the Mo–S–C–H nanocomposite coatings under friction at low temperatures.

A comparative analysis of the tribological properties of the Mo–S–C–H nanocomposite coatings obtained by RPLD with the properties of MoS*<sup>x</sup>* coatings (also obtained by RPLD) and TMD+C coatings prepared by more traditional techniques showed the importance of collecting additional information on the tribological properties of a-C (S,H) coatings formed by RPLD. A review of the literature shows that sulphur can significantly change the tribological and mechanical properties of a-C(S,H) coatings, and the effect of the introduction of sulphur depends on its concentration and the concentration of hydrogen to a considerable degree [69–71]. The tribological properties of a-C(S,H) coatings formed by RPLD require further research. Our results of tribological studies of a-C(S,H) thin-film coatings obtained by RPLD on steel substrates are presented in Supplementary Materials, Figures S6–S11.

It is important to note that pure a-C coatings prepared by the RPLD in a vacuum delaminated from the substrate one to two days after the sample had been taken out into the air from the deposition chamber. The SiC sublayer failed to provide sufficient adhesion of the a-C coating to the substrate, which can be explained by a high level of mechanical stress in the carbon film. Tribotests of a-C(S,H) thin-film coatings under various friction conditions showed that an increase in the concentration of sulphur due to an increase in H2S pressure had a negative effect on both antifriction properties and fracture resistance. Even if the MoS2.5 nanophase in the Mo-S-C\_9 coating could provide good antifriction properties under certain conditions, the a-C(S,H)\_9 phase, having an approximate composition of C0.73S0.11H0.16, would not let it happen. The degradation of the a-C(S,H)\_9 and a-C(S,H)\_18 coatings occurred by cracking and pilling off from the substrate. It can clearly be seen in the shape of the wear debris formed after friction test. These were mainly microplates accumulating at the edges of the track and adhering to the counterbody.

When under dry friction conditions, only the a-C(S,H)\_5.5 coating demonstrated good antifriction properties. These coatings were produced from graphite by reactive PLD at an H2S pressure of 5.5 Pa. The friction coefficient did not exceed 0.03 during 10<sup>3</sup> cycles of the sliding of the ball. In that case, the surface of the a-C(S,H)\_5.5 coating underwent a slight wear, and the wear scar on the surface of the steel counterbody was just incipient. Obviously, the qualitative tribological properties of the Mo–S–C–H\_5.5 nanocomposite coating under dry friction conditions were due to the influence of the a-C(S,H) phase. Under other friction conditions, the tribological properties of the Mo–S–C–H\_5.5 coating depended on the synergistic effect of the formation of a composition of the MoS and a-C(S,H) nanophases. Under friction in a humid atmosphere, the MoS~1.8 + a-C(S,H) combination provided the coefficient of friction characteristic of both phases, but changed the wear mechanism of the a-C(S,H) phase, preventing its cracking and delamination from the substrate. At low temperatures (−100 ◦C), the synergy effect of the MoS~1.8 and a-C(S,H) phases caused a rather low coefficient of friction and high wear resistance. The friction coefficient was found to be lower than the values typical for the MoS*x*~2 and a-C(S,H)\_5.5 thin-film coatings.

The Raman studies of the wear track on the nanocomposite Mo–S–C–H\_5.5 coating showed (Figures 14–16) that sliding friction of steel counterbody caused subtle changes in the Raman spectra measured in a middle region of the track. This indicates that the triboinduced changes occurred in a very thin near-surface layer of the coating. These changes manifested in an increase in the contribution of the peak at 1433 cm−<sup>1</sup> . This could be due to the accumulation (an increase in the concentration) of S atoms in the surface layer of the coating during friction. No changes in the structure of the MoS*<sup>x</sup>* nanophase were found. Comparison of Raman peaks for the Mo–S–C–H\_5.5 coating before (Figure 6a) and after the friction testing showed no noticeable changes in the spectra in the frequency range of 800–1000 cm−<sup>1</sup> . This indicated a high resistance of Mo–S–C–H\_5.5 coatings to oxidation in a humid atmosphere.

RS analysis of the wear debris accumulated near the counterbody reversal points showed (Figures 14 and 15) that the chemical state of the wear debris, and hence the chemical state of the tribofilm, depended on the tribotest conditions. Under friction in a humid atmosphere, the wear debris contained the crystalline phase of MoS2. This was indicated by the appearance of rather narrow peaks at ~370 and 405 cm−<sup>1</sup> caused by first-order reflection for the 2H-MoS<sup>2</sup> phase. Formed during triboinduced crystallization, this phase can cause peaks at ~520 and 650 cm−<sup>1</sup> , which appear due to the second-order vibration modes of MoS<sup>2</sup> [64]. Another peak at ~950 cm−<sup>1</sup> may have occurred due to the formation of a Fe-Mo-O compound (for example, FeMoO4) as a result of the tribochemical reaction of the surface of the steel counterbody with the surface of the Mo–S–C–H\_5.5 coating in the presence

of adsorbed water molecules. The Raman spectrum of the wear debris contained peaks that could be attributed to the Fe-S phase. The Raman spectrum of the Fe-S nanoparticulate phase has the most intense peaks at ~215, 323, and 463 cm−<sup>1</sup> [72]. Further research is needed to confirm the formation of the Fe-S phase. Figure 14 demonstrates the peaks for the Fe-S nanophase, but they are indicated by a question mark (?).

**Figure 14.** SEM image and Raman spectra for the Mo-S-C-H\_5.5 thin-film coating subjected to pin-on-disk test at 22 ◦C in wet friction conditions (RH ~58%).

**Figure 15.** SEM image and Raman spectra for the Mo-S-C-H\_5.5 thin-film coating subjected to pin-on-disk test at 22 ◦C in dry friction conditions (RH ~8%).

In addition to the MoS2, FeMoO<sup>4</sup> and, possibly, Fe-S phases, the wear debris contained a graphite-like phase, which corresponded to a doublet containing broadened G (at ~1563 <sup>÷</sup> 1587 cm−<sup>1</sup> ) and G (at 1355 <sup>÷</sup> 1360 cm−<sup>1</sup> ) peaks. The formation of such a C-based phase is typical of triboinduced changes in TMD+C coatings (see, for example, [15,24,46,73]). An important factor influencing friction is the local environment and atom packing in that phase. The Raman spectra for the wear particles formed on the Mo-SC-H\_5.5 coating after friction in a humid and dry atmosphere, differ in both the position of the G peak and the width of the G and D peaks. Under friction in a dry atmosphere, these peaks turn out to be narrower, and the G peak shifts to higher frequencies up to 1587 cm−<sup>1</sup> . It shows that during friction in a humid atmosphere, graphitization of the surface layer of the coating was insufficient, and the tribofilm structure was highly disordered. Probably, this was due to the fact that,

1

at increased air humidity, the contact of the coating with the counterbody was modified because of the adsorption of water molecules and slowed graphitization. This resulted a relatively high coefficient of friction (~0.2) for the Mo–S–C–H\_5.5 coating in a humid atmosphere.

The analysis of published data on the triboinduced graphitization of the interface layer for TMD+C and a-C(H) coatings and their comparison with the results of this work showed that S atoms incorporated into the a-C(H) phase during the RPLD of Mo–S–C–H\_5.5 films did not have a significant effect on the formation of tribofilms during friction in a dry atmosphere. The relatively narrow G and D peaks located at 1355 and 1580 cm−<sup>1</sup> respectively, correspond to the graphite/graphene-like packing of atoms with a laminar structure [46,71,73–75]. The incorporation of S atoms into the graphene structure may not cause noticeable changes in the Raman spectrum of graphene [76]. Still, under certain conditions when graphene interacts with sulphur, a band may appear at ~1440 cm–1 in the graphene Raman spectrum [77], the nature of which was considered in the above analysis of the Raman spectra of Mo–S–C–H and a-C(S,H) films with a high sulphur concentration. During sliding friction against the Mo–S–C–H\_5.5 coating in a dry atmosphere, along with the graphitization of the a-C(S,H) phase, crystallization of the MoS<sup>2</sup> nanophase occurred; therefore, this phase could not compromise the positive antifriction properties of the a-C(S,H) phase.

During the tests of the Mo–S–C–H\_5.5 coating at −100 ◦C, the formation of wear debris and their accumulation on the sample surface did not occur in an intensive way. This may be attributed to the fact that these particles were very small and effectively adhered to the counterbody surface (Figure 13a). When measuring the Raman spectrum on a single small particle, light was registered; it was resonantly scattered by both the particle and the coating (Figure 16). The main contribution to the spectrum from the particle was the appearance of weak peaks at ~1340 and 1579 cm−<sup>1</sup> , that indicated weak graphitization of the material in the wear debris. No signs of crystallization of the MoS*<sup>x</sup>* nanophase were found. It shows that that the mechanism of friction and wear of the Mo–S–C–H\_5.5 coating at −100 ◦C differed from that for the MoS<sup>2</sup> coatings, which underwent effective crystallization under similar test conditions [35]. This could be due to the fact that inclusions of the MoS*<sup>x</sup>* nanophase into the amorphous a-C(S,H) matrix could reduce the level of mechanical stresses in the coating. The nanocomposite Mo–S–C–H\_5.5 coatings wear out by the mechanism of layer-by-layer removal of the surface layer while the a-C(S,H) coatings showed a tendency to cracking. The presence of the MoS*<sup>x</sup>* phase in the nanocomposite coating could have an effect on the adsorption of water molecules on the coating surface at low temperatures, and possibly on the formation of water microcrystals. The influence of these factors on the tribological properties of Mo–S–C–H coatings at low temperatures require a further in-depth study.

**Figure 16.** SEM image and Raman spectra for the Mo-S-C-H\_5.5 thin-film coating subjected to pin-on-disk test in dry friction conditions at −100 ◦C.

#### **5. Conclusions**

Pulsed laser deposition of Mo–S–C–H thin-film nanocomposite coatings from Mo and graphite targets in H2S reactive gas ensures effective saturation of the formed layers with S atoms. The penetration of S atoms causes the formation of the MoS*<sup>x</sup>* nanophase, but also significantly changes the chemical state of the a-C(S,H) phase. In this work, we chose specific RPLD conditions under which the ablation time of the Mo target was twice as long as the ablation time of the C target. With an increase in the H2S pressure from 5.5 to 18 Pa, the concentration of S in the MoS*<sup>x</sup>* nanophase increased from *x*~1.8 to *x*~4. Significant changes were observed in the chemical composition of the a-C(S,H) phase. At 5.5 Pa, the composition was described by the formula C0.78S0.08H0.14, and at 18 Pa, an increase in the concentration of sulphur caused the formation of C0.62S0.18H0.2. For the coatings obtained at 5.5 and 18 Pa, the ratio of the number of atoms in the MoS*<sup>x</sup>* and a-C(S,H) phases was ~40/60 and 60/40, respectively. At the lowest S concentration, the local packing of atoms in the MoS nanophase was close to the laminar packing characteristic of the turbostratic MoS<sup>2</sup> structure. In this case, the a-C(S,H) nanophase was amorphous with a predominance of sp<sup>2</sup> bonds between C atoms. An increase in the concentration of S atoms caused the formation of MoS*<sup>x</sup>* clusters, in which the Mo3-S packing began to dominate. In this case, the configuration of the local packing of atoms in the a-C(S,H) phase was significantly modified due to the efficient formation of C–S bonds.

The best tribological properties were found for the Mo–S–C–H\_5.5 nanocomposite coatings obtained at an H2S pressure of 5.5 Pa. At higher H2S pressures, an increase in the concentration of S atoms both in the MoS*<sup>x</sup>* nanophase and in the a-C(S,H) nanophase caused a noticeable deterioration in the tribological properties of the Mo–S–C–H\_9 and Mo–S–C–H\_18 nanocomposite coatings. The tribological properties of the Mo–S–C–H\_5.5 thin-film coatings were superior to those of the MoS<sup>2</sup> coatings under various friction test conditions. However, in a humid atmosphere, the antifriction properties of the Mo–S–C–H\_5.5 coating turned out to be worse than the properties of the MoS<sup>3</sup> coating. MoS<sup>2</sup> and MoS<sup>3</sup> coatings were also obtained by RPLD [35]. The friction and wear of Mo–S–C–H\_5.5 coatings in a humid atmosphere at 22 ◦C and in a dry atmosphere at −100 ◦C were due to the synergy effect of the MoS*<sup>x</sup>* and a-C(S,H) nanophases. Under dry friction conditions, the sufficiently high-quality tribological properties of Mo–S–C–H\_5.5 resulted from the dominant influence of the a-C(S,H) phase, which is the most suitable for these conditions and has a relatively optimal chemical composition.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2079-4991/10/12/2456/s1, Figure S1: Experimental and simulated RBS (left) and ESDA (right) spectra of the films prepared on Si substrates by PLD of carbon under vacuum conditions and in H2S gas with different pressures. The RBS spectrum of the carbon film deposited in vacuum contains (residual gas pressure was ~10−<sup>3</sup> Pa) a peak at the channel number 200. This peak is due to scattering of ions by sulfur atoms that have been adsorbed on the surface of the Si substrate before the carbon film deposition, Figure S2: Typical SEM images (two magnifications) of a-C (S, H) film obtained on a steel substrate by reactive PLD in H2S, Figure S3: Raman spectra for (a) a-C(H) and (b-d) a-C(S,H) films obtained by PLD on Si substrate (a) under vacuum conditions and in reactive H2S gas at pressures of (b) 5.5, (c) 9, and (d) 18 Pa. The model of spectrum decomposition into the indicated peaks is discussed in the text of the article, Figure S4: The friction force (in relation with the normal force) evolution during reciprocate sliding of the counterbody over the MoS*x* coating under different environmental conditions. The MoS*x* coating was obtained on the steel substrate by reactive PLD at H2S pressure of 5.5 Pa, Figure S5: Optical images of wear tracks formed on the steel substrates for MoS*<sup>x</sup>* thin-film coating obtained by reactive PLD at H2S pressure of 5.5 Pa. Pin-on-disk tribometer testing was conducted in (a) wet friction conditions (RH ~50%) at 22 ◦C, (b) dry friction conditions (RH ~8%) at 22 ◦C, and (c) dry friction conditions at −100 ◦C. The test durations are indicated in Figure S4, Figure S6: Characteristic evolution of the friction coefficient as a function of the cycle number for a-C(S,H) thin-film coatings obtained by reactive PLD at the pressures of H2S gas of 5.5, 9 and 18 Pa. Pin-on-disk tribometer testing was conducted in wet air (RH ~58%) at 22 ◦C, Figure S7: Optical images of wear tracks and wear scars formed on the steel substrates and steel balls for a-C(S,H) thin-film coatings obtained by reactive PLD at the different pressures of H2S gas: (a) 5.5, (b) 9, and (c) 18 Pa. Pin-on-disk tribometer testing was conducted in wet air (RH ~58%) at 22 ◦C. The test durations are indicated in Figure S6, Figure S8: Characteristic evolution of the friction coefficient as a function of the cycle number for a-C(S,H) thin film coatings obtained by reactive PLD at the pressures of H2S gas of 5.5, 9, and 18 Pa. Pin-on-disk tribometer testing was conducted in dry friction conditions (air+Ar mixture, RH ~8%) at 22 ◦C, Figure S9: Optical images of wear tracks and wear scars formed on the steel substrates and steel balls for a-C(S,H) thin film coatings obtained by reactive PLD at the different pressures of H2S gas: (a) 5.5, (b) 9, and (c) 18 Pa. Pin-on-disk tribometer testing was conducted in dry friction conditions at 22 ◦C. The test durations are indicated in Figure S8, Figure S10: Characteristic evolution of the friction coefficient as a function of the cycle

number for a-C(S,H) thin-film coatings obtained by reactive PLD at the pressures of H2S gas of 5.5, 9 and 18 Pa. Pin-on-disk tribometer testing was conducted in dry friction conditions at −100 ◦C, Figure S11: Optical images of wear tracks and wear scars formed on the steel substrates and steel balls for a-C(S,H) thin-film coatings obtained by reactive PLD at the different pressures of H2S gas: (a) 5.5, (b) 9, and (c) 18 Pa. Pin-on-disk tribometer testing was conducted in dry friction conditions at −100 ◦C. The test durations are indicated in Figure S10.

**Author Contributions:** Conceptualization and writing—original draft preparation, V.F.; methodology, M.D. and A.G.; validation, R.R.; investigation, D.F., M.G., K.M., and P.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Russian Science Foundation, grant number 19-19-00081.

**Acknowledgments:** Sample characterization by Rutherford backscattering spectroscopy of ions and Raman spectroscopy has been done in REC "Functional Nanomaterials" with support from Ministry of Science and Higher Education of the Russian Federation (project FZWN-2020-0008).

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Laser Printing of Plasmonic Nanosponges**

#### **Sergey Syubaev 1,2, Stanislav Gurbatov 1,2, Evgeny Modin <sup>3</sup> , Denver P. Linklater 4,5 , Saulius Juodkazis 4,6 , Evgeny L. Gurevich <sup>7</sup> and Aleksandr Kuchmizhak 1,2,\***


Received: 15 November 2020; Accepted: 1 December 2020; Published: 4 December 2020

**Abstract:** Three-dimensional porous nanostructures made of noble metals represent novel class of nanomaterials promising for nonlinear nanooptics and sensors. Such nanostructures are typically fabricated using either reproducible yet time-consuming and costly multi-step lithography protocols or less reproducible chemical synthesis that involve liquid processing with toxic compounds. Here, we combined scalable nanosecond-laser ablation with advanced engineering of the chemical composition of thin substrate-supported Au films to produce nanobumps containing multiple nanopores inside. Most of the nanopores hidden beneath the nanobump surface can be further uncapped using gentle etching of the nanobumps by an Ar-ion beam to form functional 3D plasmonic nanosponges. The nanopores 10–150 nm in diameter were found to appear via laser-induced explosive evaporation/boiling and coalescence of the randomly arranged nucleation sites formed by nitrogen-rich areas of the Au films. Density of the nanopores can be controlled by the amount of the nitrogen in the Au films regulated in the process of their magnetron sputtering assisted with nitrogen-containing discharge gas.

**Keywords:** laser ablation; noble-metal films; magnetron sputtering; nanosecond laser pulses; porous nanostructures; plasmonics; nanosponges

#### **1. Introduction**

Three-dimensional (3D) percolated porous nanostructures made of noble metals and having large surface-to-volume ratio have drawn significant attention due to their remarkable physicochemical properties allowing to use them for various important applications ranging from the photo- or electro-catalysis, water splitting, hydrogen storage to bio- and chemosensing via surface-enhanced effects [1–5]. For most of the suggested applications, the surface-to-volume ratio defined by the distribution, density and size of the pores within the 3D nanostructure is of crucial importance.

Recently, optical properties of 3D porous nanostructures have become a hot topic [6–10]. Specifically, the porous Au nanoparticles (also referred to as nanosponges) were shown to demonstrate polarization-dependent scattering as well as to support long-lived electron emission associated

with localized and propagating surface plasmon modes having remarkably high quality factors. These optical properties make such structures appealing for various optical and nonlinear optical applications including random lasing, enhanced photo-emission, harmonic and supercontinuum light generation, as well as single-molecule biosensing based on metal-enhanced fluorescence, surface-enhanced Raman scattering (SERS), infrared absorption (SEIRA), etc. Multiple sensing applications are benefited from the random (but highly dense) arrangement of the plasmon-mediated electromagnetic (EM) hot spots within the structure allowing to obtain the spectrally broadband signal enhancement over the entire visible and near-IR spectral range [11–14]. Indeed, incorporation of the nanopores into the bulk plasmonic nanostructures with a sub-wavelength overall size provides more intense SERS signal due to multiple hot spots and enlarged surface area increasing probability for analyte molecules to reach these hot spots [15–18]. Nevertheless, upon excitation with EM radiation, both the arrangement of the nanosized pores and the overall nanostructure geometry govern the resulting response [19,20]. From this point of view, the plasmonic properties of the resulting 3D porous nanostructures, their general geometric shape as well as porosity are to be adjusted simultaneously that still remains challenging.

State-of-the-art methods for porous nanostructure fabrication generally require complicated multi-step fabrication protocols as dealloying and soft- or hard-template synthesis [21–27], where accurate management of the reaction conditions (temperature, composition, precursors, etc.) is crucial. Furthermore, the minimization of the surface free energy typically leads to generally spherical-shaped nanostructures. Seed-mediated growth provides simple and versatile method allowing to produce arbitrary-shaped porous nanostructures [28]. However, it's often problematically to obtain only one desired geometry because of internal structural variations of the seeds as well as local variations of the reaction environment. Alternatively, liquid-free lithography-based approaches as electron- or ion-beam milling [29] are suitable for high-precision formation of geometrically-diverse nanostructures. However, the need for upscaling of the fabrication procedure creates an economically justified barrier for lithography-based techniques being applied for porous nanostructure fabrication and large-area replication. In addition, post-processing is also required to impart porosity into the nanostructures.

Herein, we applied scalable easy-to-implement nanosecond (ns) laser ablation of nitrogen-rich Au films to fabricate parabola-shaped nanobumps containing multiple nanopores inside. The nanopores were found to originate from laser-induced explosive boiling and coalescence of the nucleation sites formed by nitrogen-rich areas of the Au film, while the nanopore density can be controlled by amount of nitrogen used as discharge gas for magnetron sputtering of the Au films. Most of the nanopores hidden beneath the nanobump surface can be further uncapped using gentle etching of the nanobumps by an Ar-ion beam to form 3D plasmonic nanosponges promising for various nonlinear optical and sensing applications.

#### **2. Materials and Methods**

#### *2.1. Deposition of Au Films Assisted with Various Discharge Gases*

Au films with a thickness of 150 ± 5 nm were deposited onto silica glass substrates without any adhesion sub-layer using a custom-built magnetron sputtering system and three discharge gases: argon, nitrogen and purified air. Deposition was performed at 10−<sup>2</sup> mbar and fixed applied voltage of 2.5 kV. At the same time, the current was maintained at a constant value of 25 mA by dynamically adjusting the discharge gas pressure that allowed to fix the sputtering rate at ≈1 nm·s −1 for all gases.

#### *2.2. Characterization of Au Films*

The actual thickness and average roughness of the films were controlled by an atomic-force microscopy (AFM, Nano-DST, Pacific Nanotechnology, Santa Clara, CA, USA) . Optical spectroscopic measurements performed with an integrating sphere spectrometer confirmed the identical reflectance for all Au films evaporated with different discharge gases (Cary 5000, Agilent Technologies, Santa Clara, CA, USA). The surface chemical composition of the Au films was carefully studied with X-ray Photoelectron Spectroscopy (XPS). XPS spectra were collected using a Kratos Axis Nova instrument (Kratos Analytical Inc., Manchester, UK) with a monochromatic Al K*α* source (source energy 1486.69 eV) at a power of 150 W. Elemental identification was carried out using survey spectra collected at a pass energy of 160 eV with 1 eV steps. A Shirley algorithm was used to measure the background core-level spectra, and chemically distinct species in the high-resolution regions of the spectra were fitted with synthetic Gaussian Lorentzian components after removing the background (using the CasaXPS software, v. 2.3.15). High-resolution XPS scans were performed in the N1*s* and Au4 *f* regions.

#### *2.3. Fabrication of Porous Nanostructures and Nanosponges*

Precise pointed ablation of the Au films was performed with second-harmonic (wavelength of 532 nm), ns (pulse duration of 7 ns) laser pulses generated by an Nd:YAG laser system (Brio GRM Gaussian, Quantel, France). The laser radiation was focused into a sub-micrometer spot on the sample surface using a dry objective (Nikon, 50x Plan Fluor, Tokyo, Japan) with a numerical aperture NA of 0.8 (optical spot diameter of 1.22*λ*(NA)−<sup>1</sup> <sup>≈</sup> 0.8 <sup>µ</sup>m on the sample surface). The sample was mounted onto a PC-driven nanopositioning platform (ANT series, Aerotech Gmbh., Nurnberg, Germany) allowing spot-by-spot laser printing of the computer-generated patterns with the movement repeatability better than 100 nm. The laser fluence was monitored by a pyroelectric photodetector (Ophir Optronics, Jerusalem, Israel) and adjusted by a PC-driven attenuator (Standa, Vilnius, Lithuania). All nanostructures were produced under identical ambient conditions upon single-pulse laser irradiation.

Additionally, to reveal the nanopores hidden beneath the surface, the laser-printed nanobumps were also post-processed via etching with an accelerated Ar-ion beam (IM4000, Hitachi, Tokyo, Japan) at acceleration voltage of 3 kV, gas flow of 0.15 cm3/min and discharge current of 105 µA. Such parameters were previously calibrated to provide relatively slow removal rate of ≈1 nm/s [15,30], allowing to avoid excessive heating, melting or deformation of the Au film and laser-printed nanostructures.

#### *2.4. Characterization of Laser-Printed Nanostructures*

Scanning electron microscopy (SEM) was performed with a Helios Nanolab 450 FIB-SEM (Thermo Fisher Scientific, Waltham, MA, USA). High-resolution surface characterization was conducted at an accelerating voltage of 5 kV and electron beam current 100 pA. Signal channels with secondary (SE) and back-scattered electrons (BSE) were simultaneously collected and analysed. Despite the lack of topographical information, the escape depth for BSE is greater than that for SE improving material/density sensitivity. This allows visualisation of the nanopores under the surface with high contrast and spatial resolution.

To shed light onto the internal structure of the nanobumps (nanosponges), we involved a focused ion beam technique to prepare single cross-sectional cuts and serial cuts that were subsequently combined into a 3D reconstruction. After defining the area of interest, deposition (starting from electron beam-induced deposition to prevent surface damage and followed ion beam-induced deposition) of the Pt protective layers was performed. The thickness of the layers was chosen taking into account the morphology and smoothness of the certain laser-printed structure and varied between 150 nm (for nanobumps) and 1000 nm (for through holes). Slicing was carried out using an Ga-ion beam at an accelerating voltage of 30 kV and beam current of 30 pA. After producing subsequent FIB cut, high-resolution SEM image was automatically acquired at 5 kV and 50 pA. The resulting image had a field of view of 2.5 × 1.6 µm at a 1536 × 1024 pixel resolution, which corresponds to the pixel size of 2.4 × 1.6 nm. A series of 111 slices were acquired and measured slice thickness was 12 nm with standard deviation of 6.54 nm. Further data processing (including alignment, filtration, and visualization) was performed using Aviso 8.1 software (Thermo Fisher Scientific, Waltham, MA, USA).

Three-dimensional finite-difference time-domain (FDTD) calculations were undertaken to reveal local structure of the EM fields excited near the isolated plasmonic nanosponge by linearly-polarized laser pump at 532, 632 and 1030 nm. The pump wavelengths within rather broad spectral range were chosen to highlight potential applications, where the enhanced and localized plasmon-mediated fields are highly demanding including SERS substrates and enhancement of nonlinear optical effects. Representative 3D model of the nanosponge was reconstructed using high resolution top- and side-view SEM images. The linearly polarized laser radiation was modeled to pump the nanosponge from the top. Elementary cell size was <sup>1</sup> <sup>×</sup> <sup>1</sup> <sup>×</sup> <sup>1</sup> nm<sup>3</sup> , while the computational volume was limited by the perfectly matched layers. Dielectric permittivity of Au was modeled according to the data from [31].

#### **3. Results and Discussion**

In this paper we considered three types of glass-supported Au films produced via magnetron sputtering in various discharges gases—argon, nitrogen and purified air (≈80% of nitrogen). Produced films had the same thickness (150 ± 5 nm) and showed identical diffuse reluctance as well as AFM verified surface roughness of about 2 ± 0.7 nm. This guarantied identical coupling of the incident ns-laser pulse energy to all types of metal films under study (Figure 1a). Such laser pulse induces thermalisation of charge carriers in the metal film that results in its local melting accompanied by detachment from the substrate via relaxation of the thermal-generated stress or evaporation at the interface between the film and the substrate [32,33]. At a pulse energy that is smaller than the ablation threshold (F*th* <sup>≈</sup> 0.17 J/cm<sup>2</sup> [30,33–35]), detached metal shell resolidifies before its rupture forming parabola-shaped surface protrusion (also referred to as nanobump; Figure 1b,c). Typical laser-printed nanobump on the surface of 150-nm thick Au film sputtered with nitrogen discharge gas is illustrated by corresponding top- and side-view SEM images. The former image combines the signals from the SE and BSE detectors giving useful information regarding either a surface morphology or difference in chemical composition/density. The latter allowed to reveal nanoscale pores under the surface of the nanobump that can be also visualized by producing its FIB cross-sectional cuts (Figure 1d).

At elevated laser fluence (*F* > 0.23 J/cm<sup>2</sup> ), rupture of the nanobump led to formation of the through hole in the metal film. In this case, the multiple nanopores can be identified within the resolidified rim surrounding the microhole as revealed by SEM visualization of the FIB cuts (see Figure 2a). Generally, for the fixed composition of the Au film, higher laser fluences produced the nanopores of the larger size. Similarly, the larger nanopores typically appeared closer to the nanobump center (see Figure 2d–e). Taking into account the Gaussian-shaped intensity profile of the irradiating laser beam, the nanopore formation appears to be driven by the local temperature (that will be discussed later in the text) that is higher in the metal film section coinciding with the beam center. To enrich information regarding the density and geometry of the nanopores, multiple FIB cuts were merged to build exact 3D model of through hole and the surrounding rim (Figure 2b,c). This 3D model clearly indicates broad size distribution of the nanopores ranging from 10 to 150 nm. Also, average size of the nanopores in the rim increases towards the center of the microhole. The larger nanopores can have irregular shape and reach the size ≈150 nm (Figure 2c), while the smaller nanopores far from the rim walls preserve spherical-like geometry.

In part, broad size distribution of the nanopores could be explained by merging (coalescence) of the closest nanopores growing from neighbouring randomly distributed nucleation centers. Origin of such nucleation centers will be discussed somewhat later in the paper in the context of chemical composition of the metal film fabricated with different discharge gases. Earlier coalescence was suggested as a leading mechanism of the water bubbles growing on dispersed gold nanoparticles heated by incident laser radiation [36,37]. As two bubbles with radii *r* merge, the gain in the surface energy *E*+ = 4*πr* <sup>2</sup>*σ*(<sup>2</sup> <sup>−</sup> <sup>2</sup> 2/3) compensates the work of viscous forces needed to move the melt over a distance ∼ *r* at a velocity, which can be estimated as *v* ∼ *r*/*t*. Here, *t* is the time needed for this melt relocation and *σ* = 1.1 *N*/*m* is the surface tension [38]. Using Newton's law of viscosity to calculate the energy dissipation force *F* = *ηvA*/*δ<sup>h</sup>* , where *A* = 4*πr* <sup>2</sup>—the surface area of the bubble, *<sup>η</sup>* <sup>=</sup> <sup>4</sup> <sup>×</sup> <sup>10</sup>−<sup>3</sup> Pa s—viscosity of liquid gold [38], and *<sup>δ</sup><sup>h</sup>* <sup>∼</sup> *<sup>r</sup>*—the characteristic length scale of the flow, we estimate the energy loss as *E*<sup>−</sup> = 4*πηr* <sup>3</sup>/*t*. Equalizing *<sup>E</sup>*<sup>+</sup> and *<sup>E</sup>*<sup>−</sup> we estimate the time for two bubbles to merge *t* ∼ *ηr*/*σ* ≈ 10–100 ps (estimated for *r* = 3–30 nm), which is less than the time gold film remains liquid.

**Figure 1.** (**a**) Sketch illustrating direct laser printing of porous nanostructures on a glass-supported Au film. Single-pulse laser ablation at a near-threshold fluence produces a parabola-shaped nanobump (nanosponge), while the explosive boiling of randomly distributed nitrogen-rich sites create a nanoscale pores inside irradiated area. High-intense laser pulse drills a through hole where the nanopores can be found in the surrounding resolidified rim. (**b**,**c**) Representative top- and side-view SEM images of the isolated laser-printed nanosponge produced at *F* = 0.15 J/cm<sup>2</sup> in the Au film evaporated with nitrogen discharge gas. The top-view image is divided into two parts (recorded at different e-beam acceleration voltage) to illustrate multiple nanopores hidden beneath the nanobump surface. (**d**) False-color SEM images of the cross-sectional FIB cuts made through the center of such nanobump.

This time can be estimated assuming that the heat accumulated by the illuminated spot is removed mostly by the thermal conductivity. The heated volume can be estimated as a cylinder of the radius *r* ≈ 0.5 µm (the typical lateral size of the nanobump) and the height *h* = 0.15 µm (the film thickness). The areas of the side *Sside* = 2*πrh* and the bottom *Sbottom* = *πr* 2 surfaces are comparable, but the thermal conductivity of gold *λAu* = 300 W/(Km) is larger than that of glass *λglass* = 0.5 W/(Km), so that only the heat flow through the side wall decreases the temperature of the molten surface. This heat flow can be estimated as <sup>d</sup>*<sup>E</sup>* d*t* <sup>=</sup> *<sup>λ</sup>AuSside*∇*<sup>T</sup>* <sup>≈</sup> <sup>2</sup>*πhλ*(*T<sup>m</sup>* <sup>−</sup> *<sup>T</sup>*0), where *<sup>T</sup><sup>m</sup>* <sup>≈</sup> 1.4 <sup>×</sup> <sup>10</sup><sup>3</sup> K is the melting point of Au and *T*<sup>0</sup> = 300 K. This heat flow should remove the energy accumulated by the surface in one pulse *E<sup>p</sup>* = *Fπr* 2 , hence the characteristic cooling time scale is *t* ∼ *Fr*<sup>2</sup> 2*hλT<sup>m</sup>* <sup>≈</sup> <sup>5</sup> · <sup>10</sup>−<sup>9</sup> *s*, i.e., is also comparable to the laser pulse duration. This estimation also agrees with previously reported studies [39]. Hence, coalescence of the bubbles provides a plausible explanation of the observed distribution of the pore sizes.

Noteworthy, different growth times of the pores can be also considered as alternative way to explain the large size variation of the nanopores. In particular, the pores started growing short after the onset of the melting had more time to develop than that ones started just before the resolidification. This assumption cannot be accepted because of the low growth velocity of the bubbles. Wang et al. [37] reported that during the initial bubble nucleation phase the bubble size *R* grows with time as *R* ∝ *t* 1/6 (we notice that at later stages in degassed water this dependency drops to *R* ∝ *t* 0.07, whereas in air-equilibrated water *R* ∝ *t* 1/3). Hence, to get one order of magnitude difference in the nanopore radii, the times should differ by six orders magnitude. If the earliest possible nucleation starts after ten picoseconds after the start of the laser pulse (time comparable with the electron-phonon coupling times in metals), then the smaller pores formation must start several microseconds later, which is impossible because the resolidification happens on a time scale of several nanoseconds.

In a similar way, the broad size distribution can not be explained by local temperature fluctuations (pores at hot spots grow quicker) as the *R*(*t*) ∝ *T* 1/3 [37]. Hence, three-orders-of-magnitude fluctuation of the local temperature in gold separated by a distances much less than one micrometer is obviously not possible. Spontaneous merging (coalescence) of several nanopores into the larger one can also stimulate rupture of the nanobumps and can explain formation of the through nanoholes previously reported for AuPd films processed with ns laser pulses [40]. The average pore size was observed to grow with the laser intensity, as it was shown in [15]. This can be explained by cooperative action of two mechanisms: (1) each pore grows more rapidly, since the temperature is higher and the pore radius *R*(*t*) ∝ *T* 1/3. (2) The metal film remains molten for a longer time and the coalescence-based growth is stopped by the resolidification after a longer time interval *t* ∝ *T*, so that more pores can merge together. Assuming that the surface temperature is proportional to the intensity, we expect the average observed pore size to grow with the intensity.

Remarkably, the density and the average size of the nanopores started to grow from randomly distributed nucleation centers were found to depend on the type of discharge gas used for the film fabrication. More specifically, negligible amount of nanopores was found for the nanobumps produced on the surface of Au films sputtered with Ar (see Figure 2d). The maximal density of the nanopores was observed for the Au film evaporated with nitrogen discharge gas, decreasing for the Au film produced in purified air (Figure 2e,f). Taking into account same morphology and light absorbing characteristics of all mentioned Au films produced with different discharge gases, the amount of nitrogen in the film could be considered as a driving parameter that allows to control density of the nanopores. The molecular nitrogen can be bounded to the metal film surface as well as form chemically stable AuN phase upon magnetron sputtering as it was reported by several previous studies [41–43].

XPS measurements were performed to understand the effect of the discharge gas used for magnetron sputtering on the resulting chemical composition of the Au films (see Methods for details). Analyses of the deconvolved high resolution spectra taken in the N1*s* region revealed near-zero signal for films produced with Ar discharge gas. However, a pronounced signal for Au films fabricated with purified air and nitrogen (Figure 3b–d) were also observed. Deconvolution of the latter two spectra revealed several peaks with their binding energies at ≈401, 399 and 397.3 eV. The first high-energy peak can be assigned either to carbonitride [43] or to molecular nitrogen (that can be adsorbed on the metal film surface having multiple grains and cracks as well as be trapped beneath the surface [44,45]). The two remaining peaks can be attributed to chemically stable gold nitride phases AuN*<sup>x</sup>* as well as to oxynitrides [41–43]. All of the different Au films demonstrated similar Au4 *f* XPS spectra with low-intense shoulders (marked by blue areas in the Figure 3a) shifted towards larger binding energies near both low- (Au4 *f*7/2) and high-energy (Au4 *f*5/2) peaks. These shoulders can be indicative of stable chemical compounds like surface bonded carbon (AuC) or nitrogen (AuN*x*) as shown in previous studies [43]. However, the similarity of the signals obtain from different Au films does not clarify the exact chemical nature of these low-intense shoulders where signals from different compounds could overlap. To confirm such peak assignment, the Au film sputtered with nitrogen was further annealed at 200 ◦C for 2 h. Thermal annealing was expected to remove thermodynamically unstable compounds

(such as oxynitrides) as well as most of the molecular nitrogen from the near surface layer probed by XPS. N1*s* core-level spectra of the annealed Au film demonstrates the only remaining peak with a binding energy of 398.8 eV that can be presumably attributed to AuN*x*.

**Figure 2.** (**a**) False-color SEM images of cross-sectional FIB cuts made through the center of the microhole. The hole was produced at *F* = 0.25 J/cm<sup>2</sup> in the Au film evaporated with nitrogen discharge gas. (**b**,**c**) Distribution of the nanopores in the rim around a through hole visualized by tomographic 3D model reconstructed using serial FIB cuts. Several elongated irregular-shaped nanopores are highlighted in the figure. (**d**–**f**) Top-view SEM images of the nanobumps produced under single-pulse irradiation of the Au films (*<sup>F</sup>* <sup>≈</sup> 0.15 J/cm<sup>2</sup> ) produced with argon (**d**), purified air (**e**) and nitrogen (**f**) discharge gases, respectively.

Considering the chemical composition of the Au film produced with different discharge gases we can suggest the following scenario regarding formation of the nanopores upon ns-pulse laser ablation. Such laser pulse with a near-threshold fluence rapidly heats up the Au film to temperatures above 10<sup>3</sup> K. Such temperature jump is expected to remove all molecular nitrogen from the exposed area as well as induce an explosive boiling of the N- or AuN*x*-rich areas of the film. The data regarding melting/boiling temperatures of gold nitrides is weakly discussed in a literature. For example, annealing at 90 ◦C was shown to remove AuN*<sup>x</sup>* from the film [41], while present studies clearly showed signature of AuN*<sup>x</sup>* remaining even after annealing at 200 ◦C. Anyway, both melting/boiling temperatures of gold nitrides are expected to be much lower comparing to those for the pure gold. The light absorption should be initiated at the Au grain boundaries where also nitrogen adsorption should occur. The ionisation potential of Au is 9.23 eV while that of nitrogen is 14.53 eV (for O—13.62 eV). The laser pulse driven avalanche ionisation of gold is seeding the energy deposition which is evolving into run away ablation (melting, evaporation, ionisation).

**Figure 3.** XPS characterization of the Au films produced using different discharge gases. (**a**) Representative Au4 *f* photoemission spectra measured from the glass-supported Au film produced with nitrogen discharge gas. (**b**–**e**) N 1*s* core-level photoemission spectra of the Au films produced using magnetron sputtering assisted with various discharge gases: argon (**b**), purified air (**c**), nitrogen (**d**), nitrogen followed by thermal annealing at 200 ◦C for 2 h (**e**). Deconvolution of the obtained spectral signal allowed for identification of several characteristic peaks highlighted by the colored areas in (**a**,**c**–**e**).

The general geometry of the surface structure produced by direct laser ablation is defined by the laser irradiation parameters (fluence, pulse width and beam profile [46]) as well as by the thickness/composition of the metal film [33,47]. Our results clearly show that advanced chemical engineering of the metal film composition gives additional degree of freedom allowing to modify morphology of the laser-printed structures at the nanoscale, namely, incorporate the nanopores and control their density. However, a large amount of the nanopores is typically hidden beneath the top metal shell of the nanobump that limits the potential range of practically relevant applications. To make morphology of the laser-printed nanobumps more functional, we applied gentle etching of the laser-printed nanobumps with unfocused Ar-ion beam schematically illustrated on the Figure 4a. The processing parameters were calibrated to ensure gradual removal of the Au film without melting and deformation of the nanobumps geometry (see Materials and Methods). Two representative SEM images in the Figure 4b,c compare nanoscale surface morphology of the nanobump before and after its etching with Ar beam for 20 min. As can be seen, etching reveals the multiple hidden nanopores making the nanobump shell surface perforated with multiple through nanoholes. Besides the rather random arrangement of the nanopores (nanoholes), general geometry of the produced nanosponges reproduces well from pulse to pulse (see Figure 4d). From the simple geometrical consideration it is clear that the incident electromagnetic radiation can be efficiently absorbed by such nanosponges that produce enhanced electromagnetic fields via coupling to propagating and localized surface plasmons. Being combined with the mentioned broad size distribution of the nanopores, the 3D nanosponges with their lateral size of about visible-light wavelength are expected to be efficient for local EM field enhancement within rather broad spectral range spanning from visible to near-infrared. To illustrate this, we calculated squared electromagnetic field amplitude E2/E<sup>2</sup> 0 (E<sup>0</sup> is the amplitude of the incident EM field) near the isolated plasmonic nanosponge pumped at 532, 632 and 1030 nm (see Materials and Methods). The cross-sectional E2/E<sup>2</sup> <sup>0</sup> maps shown in the

Figure 4e indicate that the nanosponge supports densely arranged EM hot spots upon broadband EM excitation. Multi-fold enhancement of the EM field amplitude comes from coupling of the incident radiation to the localized plasmon resonances of the isolated nanoscale surface features on the top side of the Au shell (like nanopores, spikes and cracks clearly seen in the Figure 4c) as well as plasmons propagating within overall micro-scale sponge geometry. Propagating plasmons can couple to or re-excite the localized ones appearing as the localized EM hot spots on the opposite size of the opaque 150-nm thick Au shell. Owing to weak radiate decay, the certain localized plasmon modes in such plasmonic nanosponges were found to survive for a long time producing substantial field localization and enhancement [8]. Alternatively, most of the modes can dissipate via heating of the nanosponge. These features will potentially allow to use such nanosponges for various applications as thermo and nonlinear plasmonics (enhanced higher harmonic and white light generation), bio-imaging and visualization, photothermal conversion, sensing based on nonlinear optical effects as well as SEIRAand SERS-based sensing [48–52].

**Figure 4.** (**a**) Schematic illustration of the nanosponge fabrication. (**b**,**c**) Representative side-view SEM images comparing typical morphology of the nanobump before and after its etching with Ar-ion beam for 20 min. The nanobump was produced at *F* = 0.165 J/cm<sup>2</sup> on the surface of Au film evaporated with nitrogen discharge gas. (**d**) SEM image of the nanosponge array showing how the nanosponge morphology reproduces from pulse to pulse. (**e**) Calculated E2/E<sup>2</sup> <sup>0</sup> maps near the isolated Au nanosponge pumped at 532, 632 and 1030 nm.

## **4. Conclusions and Outlook**

Here, we propose a simple approach allowing fabrication of porous plasmonic nanostructures using direct ns-laser ablation followed by Ar-ion beam etching. The nanopores were found to form through the explosive evaporation/boiling of the nitrogen-rich metal film areas exposed by a ns laser pulse. Detailed XPS and SEM analysis confirmed that the density of the nanopores correlates with the initial amount of nitrogen in the Au films that were fabricated by magnetron sputtering assisted with different discharge gases. Demonstrated unique porous nanostructures, parabola-shaped nanobumps perforated with multiple nanoholes, are expected to be useful for various applications where the plasmon-mediated EM fields are of mandatory importance as nonlinear plasmonic and chemo-/biosensing.

In a broader context, advanced chemical engineering of the metal film composition suggested in this paper being combined with an adjustment of the laser ablation process (as optimization of fluence, pulse width, laser beam shaping, etc.) is expected to provide facile way for fabrication of unique nanostructures. Also, elaborated strategy being applied to more complicated material combinations (like metal alloys or metal-dielectric materials) will further enrich the potential types of nanostructures as well as their application range [53–55]. For example, in a similar way nanosponges can be produced from co-sputtered noble-metal films forming nano-alloys with engineered permittivity as demonstrated for Au-Cu-Ag [56]. Along with plasmonic hot spot engineering, local potential defined by the nearest atomic composition was shown to modify physical and chemical adsorption of the analyte molecules affecting their characteristic SERS and SEIRA signals and sensor performance [57]. Insights into formation of metallic glasses and high entropy alloys with even large number of constituent materials [58] will benefit from a nanoscale control of re-melting and phase-explosions demonstrated in this work.

**Author Contributions:** Conceptualization, A.K. and S.S.; methodology, S.G., E.M. and D.P.L.; software, E.M.; validation, A.K. and S.J.; formal analysis, A.K., E.L.G. and S.S.; investigation, S.S., E.M. and D.P.L.; resources, A.K.; data curation, A.K.; writing—original draft preparation, S.S., A.K., E.M., E.L.G. and S.J.; writing—review and editing, A.K., E.L.G. and S.J.; visualization, S.G.; supervision, A.K.; project administration, A.K.; funding acquisition, A.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by Russian Foundation for Basic Research (projects nos. 20-32-70056) and Ministry of Science and Higher Education (0262-2019-0001).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
