Chapter - 1
INTRODUCTION AND REVIEW
Superhard materials are generally recognized as being of the utmost importance for the development of industries such as aerospace, automotive, opto-electric technology and others. Particularly in the quest to reduce friction and wear, the ever-increasing need for thin, wear-protective films is driving researchers to develop new superhard materials which can be expected to play an important role in the future. For example, the new kinds of material can be used on recording media and on read-and-write heads which are capable of magnetic recording operation with increased packing capacities and faster access times. It is also of great interest to increase the performance of medical components like implants and surgical tools by use of hard, low-friction and biologically inert coatings.
The research has continuously provided the impetus to push the developments of fringe technologies which are tolerant of increasing high temperatures and hostile environments. The relatively shorter bond length (~ 1.47 ) and low bond ionicity (~7%) the C-N bond suggest its high-temperature applications. Carbon nitride may prove to be a superior semiconductor for device applications in these purposes.
In this project attempt was made to synthesized carbon nitride tin film with the help of pulsed laser deposition. Graphite was used to produce graphite plasma with the help of excimer (248 nm) laser in the atmosphere of N2 gas at atmospheric pressure. p-type (111) silicon was the substrate to deposit the films of carbon nitride.
Effect of deposition pressure and laser intensity on the mechanical properties of carbon nitride (CNx) thin films prepared by plasma assisted pulsed laser deposition was studied by M. Tabbal et al. Films prepared at high pressure increased the growth rate and nitrogen content but the films becomes soft and mechanical properties decreased but at low pressure Young's modulus and mechanical properties increased .
Non-hydrogenated amorphous carbon nitride (a-CxN1-x) films were prepared by rf magnetron sputtering source by O. Durand-Drouhin et al. Optical properties and the microstructure of the synthesized films were studied by Raman Spectroscopy in the IR and UV range .
Structured carbon nitride (CNx) thin solid films, also known as fullerene-like, consist of, upon nitrogen substitution, bent and cross-linked graphene planes. J. Neidhardt et al. synthesized fullerene-like carbon nitride thin films by unbalanced reactive magnetron sputtering and analyzed with high-resolution transmission electron microscopy (HRTEM) in combination with X-ray photoelectron spectroscopy (XPS) .
Mingwu Bai et al. deposited three kinds of ultra-thin amorphous carbon nitride films by an ion beam assisted deposition method with different internal stress on silicon (111) substrates and post-annealed to minimize the internal stress. As-deposited and annealed films were studied for microstructure and nano-mechanical properties by using micro Raman spectroscopy, nanoindentation measurement, and nanoscratch test in atomic force microscopy .
Rusop et al. prepared carbon films by pulsed laser deposition at different temperature ranging from 20 to 500oC. the prepared thin film were studied the Raman scattering analysis that shows that the structure of carbon (C) films can be changed with varying substrate temperatures .
Crystalline n-Si <100> wafers have been used as substrates to prepare carbon films by direct ion beam deposition by S. Tamulevicius et al. Mechanical stress and chemical structure was observed by laser interferometry and Raman spectroscopy respectively .
Amorphous carbon nitride films were analyzed for the relationship between the local microstructure evolution versus nitrogen content deposited by radio frequency (RF) magnetron sputtering by A. Lagrini at al. TRIM code, XPS measurements and Raman scattering spectroscopy were used to investigate the microstructure of a-CNx, films .
Effects of varying intensity of KrF excimer laser on the bonding and hardness of the carbon nitride thin films, prepared by plasma assisted pulsed laser deposition at room temperature, were studied by Philippe Merel at el. .
High-purity graphite targets were used to prepare carbon nitride thin films on Si (l00) substrates with pulsed Nd:YAG laser deposition technique by Y. Suda et al. Morphological and compositional information was gathered by using scanning electron microscopy and Auger electron spectroscopy respectively .
Aoi et al. used pulsed laser deposition combined with nitrogen r.f. radical beam source to deposit amorphous carbon nitride thin films. For structural analysis X-ray photoelectron and Raman-scattering spectroscopy was performed .
R.F. magnetron sputtering was used to prepare carbon nitride thin films and studied by B. Bouchet-Fabre with help of FTIR, Raman Scattering, NEXAFS and XPS .
M. Lejeune et al. studied the stress relaxation caused by post-deposition thermal annealing of carbon nitride thin films (CNx) deposited onto Si substrates has been carried out. The intrinsic stress values were correlated with Fourier transform spectrometer (FTIR) and thermal desorption mass spectroscopy (TDMS) results. .
Pulsed laser ablation of graphite target was done to prepare carbon nitride films under a nitrogen atmosphere at room temperature by Y. H. Cheng et al. A direct current discharge apparatus was used to supply active nitrogen species during the deposition of carbon nitride films. The composition and bonding structure of carbon nitride films were determined by Fourier transform infrared (FTIR) spectroscopy and X-ray photoelectron spectroscopy. The incorporation of nitrogen atoms in the films is greatly improved by the using of a dc glow discharge .
G. K. Gueorguiev et al. used first-principles calculations to observe the formation mechanisms and structural details of fullerene-like carbon nitride (FL CNx). Cohesive energy comparisons show the energy cost for different defects arising from substitution of C for N as a function of the nitrogen concentration. In FL CNx, combinations of pentagons and heptagons compete in causing graphene sheet curvature during the addition of CN-precursors and single species. Also, cross-linkage between graphene layers in FL CNx can be explained by the bond rotation due to incorporated N atoms. The computational results agree with recent experimental observations from the growth of FL CNx thin films .
Rutherford-backscattering spectroscopy, x-ray photoelectron spectroscopy, electron-energy loss spectroscopy, Raman and Fourier transform infrared spectroscopy and nanoindentation techniques were sued to study carbon nitride films, deposited by reactive dc magnetron sputtering in Ar/N2 discharges, with respect to composition, structure, and mechanical properties by Niklas Hellgren et al .
Y.H. Cheng et al. used graphite target to prepare carbon nitride films by direct current plasma assisted pulsed laser ablation under nitrogen atmosphere. Characterization of the samples was done by Atomic force microscope (AFM), Fourier transforms infrared (FTIR), Raman and X-ray photoelectron spectroscope (XPS) to study the surface morphology, bonding structure and composition of the deposited films .
Pulsed laser deposition (PLD) is a thin-film deposition method, which uses short and intensive laser pulses to evaporate target material. The ablated particles escape from the target and condense on the substrate. The deposition process occurs in vacuum chamber to minimize the scattering of the particles. In some cases, however, reactive gases are used to vary the stoichiometry of the deposit. As a materials processing technique, laser ablation has been utilized since the 1960's, when the laser was invented. Nevertheless, this deposition method based on laser ablation did not get much research interest until the first high-temperature superconductive films were deposited in 1987. After that the development of PLD has been more rapid, since the research on this topic has increased dramatically. PLD has offered an interesting alternative for other physical vapor deposition (PVD) methods, because of the low deposition temperature. Most of the PVD methods require the substrate temperature to be at least 300-400C to achieve a proper adhesion between the substrate and the coating. In PLD, however, it is possible to get an excellent adhesion, although the substrate is at room temperature. So, PLD gives an opportunity to coat heat sensitive materials like polymers, as well.
Further advantages of the PLD are a wide variety of coating materials, a good controllability of the film composition and the simplicity and flexibility of the equipment. The disadvantages of this method are the problems with the uniformity of the coating thickness, when the substrate area is large and "splashing", i.e. the ablation of large particulates from the target, which destroys the smoothness of the microstructure of the coating.
Basics of the pulsed laser deposition
Figure (1.1) illustrates the principle of PLD. The laser beam is generated outside the chamber. Also the optical instruments, like lenses, mirrors and apertures whose objective is to steer and focus the laser beam, are placed before the port of the deposition chamber. In the chamber, the laser beam is directed towards the target. To ensure a uniform wearing of the target, its surface can be rastered by the laser beam, or the target itself can be rotated and/or translated, as the laser beam remains stationary.
The absorption of the laser radiation is followed by breaking of chemical bonds in the target material and ablation of atoms, ions, electrons, molecules, atomic clusters and even bigger (m-sized) particulates (providing that the laser energy density exceeds the ablation threshold of the target material). These evaporated species form a plasma plume, which expands in the vacuum and flows towards the substrate to form the deposit. As an opposite to the simplicity of the deposition equipment, the physical phenomena related to the laser-target-interaction and plasma formation are very complicated. These topics will be further discussed in the following sections.
The conventional PLD setup being quite straightforward, it is possible to install additional elements to the chamber and enhance the versatility of the coating process. Deposition in the presence of a background gas (reactive deposition) and substrate heating during deposition are examples of modified approach of PLD .
The mechanisms related to the energy transfer from the laser beam to the target depend strongly on the properties of the laser, like fluence, frequency, pulse duration and wavelength. Since the deposition using ultrashort (ps and fs) pulses has not been reported as widely as nspulsed laser deposition in previous papers, the interaction mechanisms discussed here are concerning mainly the deposition using ns-pulses.
Collisional sputtering is related to the transfer of momentum: the species hitting the target lose their momentum and enable the ablation of the particles of the target material. The momentum of the energetic photons being very low (compared to the momentum of massive particles, like electrons or ions of equivalent energy), the effect of collisional sputtering in PLD is negligible. However, indirect collisional sputtering is also possible. As a consequence of the scattering in the plasma plume, part of the emitted ions may return to the target and cause ejection of particles [18, 26].
Thermal sputtering means vaporization of the target material as a result of local heating caused by the laser beam. Thermal sputtering often requires the temperature of the target to be well above the boiling point until the ablation (and reasonable rate of sputtering) becomes possible .
Electronic sputtering consists of various processes, all of which involve some kind of excitation or ionization. As the photons hit the target, they produce electron-hole-pairs and electronic excitations. As a consequence of the electron-lattice interactions, the temperature of the lattice increases strongly. This is followed by desorption of particles. Lattice defects and loose interatomic bonding facilitate desorption.
In the case of PLD, electronic sputtering is the main mechanism of laser-target-interaction. The heating of the lattice may, however, cause simultaneous vaporization via thermal sputtering. With the help of the electronic sputtering model, particle emission rates and the temperature of ablated particles can be estimated, as well as the threshold fluence for particle emission .
Exfoliational sputtering is caused by repeated thermal shocks, when the laser pulses heat the target but its surface temperature does not reach the melting point. As the thermal stresses can't be relieved by melting, the thermal cycling finally leads to cracking of the target and ejection of flakes. Target materials having high thermal expansion coefficient, high Young's modulus and/or high melting point are usually sensitive to exfoliational sputtering. Some examples of these materials are refractory metals, like tungsten and oxides such as Al2O3 .
Processes leading to the melting of the target surface and ejection of molten droplets are termed hydrodynamic sputtering. This mechanism is a special feature of processes, which use photons for target bombardment. Hydrodynamic sputtering does not occur when electrons or ions are used as incident species.
As a conclusion, the principal sputtering mechanism in PLD is electronic sputtering. The role of thermal sputtering is also quite important, but collisional, exfoliational and hydrodynamic sputtering usually has negligible significance. The occurrence of different sputtering mechanisms can often be deduced from the topography of the target after sputtering. Characteristic changes in target surface as a result of some sputtering mechanisms are illustrated in Figure 2. Collisional sputtering causes nonuniform erosion of the target, i.e. formation of cones and craters, as illustrated in Figure (1.2a). Figure (1.2b) shows the detachment of flakes as a result of exfoliational sputtering. Figure (1.2c) illustrates hydrodynamic sputtering and the formation of droplets .
As a consequence of laser-induced ablation, a layer of high-pressure vapor is produced near the surface of the target. This pressure gradient acts as a driving force, which expels the vaporized particles away from the target. First the jet of the particles moves straight in the direction of the normal of the target, but the scattering changes the motion from unidirectional to more isotropic. The jet of particles also expands and forms a plasma plume in accordance with the cosine law .
The plume contains a variety of particles: atoms, ions, electrons and atomic clusters, for example. The temperature of the particles is high and atoms in the plume are easily ionized, because there is plenty of thermal energy available. As the plume interacts with the laser beam, the particles in the plasma gain kinetic energy and their temperature is increased, as well as the degree of ionization [19, 20].
The behavior of the plasma depends on whether the process is carried out in vacuum or presence of background gas. The profile of the plume is described by cos n f curve, where 4 <n< 15, n depends particularly on gas pressure. The presence of the background gas has also an influence on scattering and kinetic energy of the particles in the plasma, as well as the rate of deposition [19, 20].
As a result of fluorescence and recombination processes occurring in the plasma, the plume emits visible light. This phenomenon offers a means to monitor the deposition process. Laser ion mass spectroscopy and optical emission spectroscopy are examples of techniques used for plume monitoring. Further methods for plasma characterization include measurements of the velocity, temperature and energy of plasma species [20, 21].
Formation of the deposit
In the next subsection, theoretical nucleation and growth models of the deposit are considered. Thereafter, the following subsections discuss the most common problems related to the deposition, namely, the particulates generated in the deposition and the thickness profile of the coating.
Film nucleation and growth
In general, when considering the theoretical models of film nucleation and growth in PLD, there are three conventional modes: three-dimensional growth of islands, two-dimensional growth of monolayers and the formation of full monolayers followed by the growth of separate three-dimensional islands. The thermodynamics related to the surface energies of the film and the substrate determines, which mode dominates the growth of the film. The three-dimensional growth of islands (also called Volmer-Weber nucleation and growth) involves a number of atomic scale processes taking place after the particles have arrived on the substrate. Figure 3 shows schematically the possible mechanisms of atomic interaction. The total free energy of each atomic cluster determines the balance between the growth and dissolution processes. By decreasing the temperature or increasing the deposition rate the free energy of the cluster nucleation and growth can be made more negative, i.e. the nucleation rate of the clusters can be increased. The lowering of the temperature may, however, decelerate the formation of the equilibrium structure of the deposit, which increases the possibility of metastable or amorphous phases [18, 19].
Depending on the size and shape of the nuclei and the interface energies of the cluster, substrate and vapor it can be energetically more favorable for the film to grow as complete monolayers instead of three-dimensional islands. This growth mode (also called Frank-van der Merwe nucleation and growth) is more possible in the conditions, where the film surface energy is low and the substrate surface energy is high. However, the atomic processes during the Frank-van der Merwe nucleation and growth are essentially the same as in Volmer-Weber-mode, but the thickness of the islands corresponds only one monolayer. Similarly, the nucleation rate can be controlled by changing the temperature or the deposition rate, as described earlier .
The third possible mode is called Stranski-Krastinov nucleation and growth. The film formation begins with complete monolayers, but after 1-5 monolayers the growth continues by island formation. The reason for this kind of behavior is probably the lattice stress, which is higher on the surface of the deposited monolayers than on the bare substrate .
In the preceding description of the nucleation and growth modes it is assumed that the nucleation occurs on random sites, homogeneously on the whole substrate surface. In practice, the surface of the substrate is not uniform enough to ensure the homogeneous nucleation: there are defects and dislocation intersections providing more favorable nucleation sites. Thus, for example the preparation of the substrate can have a remarkable effect on the nuclei density and nucleation rate .
Particulates generated by PLD
The generation of particulates is one of the major drawbacks of PLD and it usually is the main limiting factor in the application field of this deposition method. For example high performance electronic, optical and optoelectronic devices require absolutely particulate-free films. The size of the particulates ranges between nanometers and microns. The shape and the microstructure can also vary depending on the material and process parameters [18, 22].
The emission of the particulates is based on various physical phenomena, for example: dislodgement of uniformities protruding from the target surface, explosive dislocation of the substance caused by the subsurface overboiling of the target, gas phase clustering of the evaporated material due to supersaturation (especially in high gas pressures), generation of liquid phase droplets under the action of the recoil pressure of the ablated material. The particulates ejected from the target can therefore be in solid, liquid or vapor state. The shape usually depends on the state of the particulate: the shape of the solid particulates is often irregular, while the liquid ones tend to be spherical [18, 22].
A number of different solutions to the particulate problem have been developed. They can be classified into four groups according to the approach: Optimization of laser parameters, for example by changing the laser pulse duration, fluence or wavelength, making the energy profile of the laser beam more uniform or rastering the target with the beam (avoiding the inhomogeneous wearing of the target). Optimization of the target properties, for example by making the target denser or enhancing the surface quality by polishing. Modification of the deposition geometry, for example by setting the substrate in off-axis position with respect to the target. Figure (1.4) illustrates three different deposition geometries: on-axis (1.4a), off-axis (1.4b) and shaded off-axis, where a shutter is installed between the target and the substrate to catch part of the particulates (1.4c).
Modification of the deposition apparatus, for example magnetic field-assisted deposition or using second laser directed into the plasma plume to split the particulates.
The first three approaches are usually easy to implement, but the elimination of the particulates is not as effective as in modification of the deposition apparatus. On the other hand, the last-mentioned approach is often more complex and more expensive [22, 23].
The thickness profile of the coating
Another factor preventing the PLD process from getting success in commercial applications is the coating thickness nonuniformity. This is because the material distribution in the plasma plume is inhomogeneous and growth of epitaxial film is possible only in the center of the plasma plume. Improvement of the coating thickness uniformity is usually based on following approaches or combination of them:
- Rotation and/or translation of the target
- Rotation and/or translation of the substrate
- Off-axis position of the substrate (see Figure (1.4))
Rastering across the target surface with the laser beam
Increasing the target-substrate-distance may also be useful when coating large area substrates, since the plasma plume becomes wider as it gets further away from the target. At best, with the help of the mentioned techniques the uniform coating thickness has been achieved for substrates, the diameter of which has been 150 mm. One should take into consideration, however, that some of these procedures (especially off-axis position of the substrate and increasing the target-substrate-distance) may decrease the deposition rate drastically. Some applications, e.g. tribological coatings for ball bearings, require coating of nonplanar surfaces. In these situations, it is a common practice to use higher process gas pressures. This enables more intense scattering of plasma particles and the deposition of the surfaces perpendicular to the expansion direction of the plasma plume, as well [19, 25, 26].
This chapter introduces the basic components of the PLD-system. The laser and the optical components required to manipulate the beam are discussed first. The following sections consider the components inside the deposition chamber, namely, target, substrate and also the vacuum system.
Conventionally two basic types of laser are used in PLD, excimer and Nd:YAG. This section discusses the function and features of these lasers. In addition, one more modern technique, namely, fiber laser, is introduced.
The useful wavelength range used in PLD is usually 200-400 nm, since most target materials exhibit strong absorption in this part of spectrum. Typically, the shorter the wavelength, the higher is the absorption coefficient. As a consequence, when the wavelength of the laser beam gets closer to 200 nm, the penetration depth on the target surface decreases and thus, the ablated layer is thinner compared to the situation, where higher wavelength is used. In addition, the ablation fluence threshold is typically lower because of the stronger absorption. This is why the wavelengths close to 200 nm are favorable in PLD. However, wavelengths below 200 nm result in practical problems, since absorption of the molecular oxygen becomes strong in this spectral region. These short UV-wavelengths set also special requirements on the optical components used in the system. It is also worth noticing, that in the case of ultrashort laser pulses the ablation is not dependent on the laser wavelength, because of the greater intensity of the beam.
A good quality of laser beam, that is, uniform distribution of intensity across the beam cross section, is important. Deviations from homogeneity (so called hot spots) may result in droplet formation, as well as deviations in stoichiometry of the deposit. Thus, the uniformity of the beam has a critical role especially, when multicomponent targets are used .
The name "excimer" is a shortening for "excited dimer", the chemical gain medium of the laser. The first excimer laser was based on Xenon dimer, however, today the most excimers use noble gas halides as the gain medium, therefore the nomination is a bit misleading. The function of the excimer laser is based on the inertness of the noblegases. As they do not tend to form chemical compounds, the molecules used in excimers (e.g. ArF, KrCl, KrF, XeCl) have a repulsive ground state and a bound (associative) excited state. As a result of electrical discharge, the noble gases can achieve excited state and form temporary molecules (excimers) with themselves or with halogens. The excited compound will rapidly decay back into unbound atoms via spontaneous emission, since the lifetime of the excited molecule is on the order of a few ns [18, 19].
In order to support the lasing action, the gas mixture in laser chamber is pumped by a fast avalanche electric discharge. Basic components of an excimer laser discharge circuit are: capacitors, inductor coils and a pair of electrodes. To achieve a proper lasing action, the formation rate of the excimers must exceed 1023 cm-3s-1. To meet this requirement, the discharge volume must have a total gas pressure of 2-4 atm, electron density on the order of 1015 cm-3 and current density of 10 A cm-2 .
The wavelength emitted by the excimer laser depends on the molecules of the gain medium.
Table 1 lists the operating wavelengths of some commercial excimer lasers.
Nd:YAG is a solid state laser, in which the lasing medium consists of Nd+3 ions as a dopant in yttrium-aluminium-garnet (YAG:Y3Al5O12) host material. The dopant content of the host crystal is typically on the order of 1 weight%. Nd:YAG lasers are optically pumped, typically with the help of flashlamps or laser diodes, to achieve the excitation of the Nd+3 ions and lasing effect.
Pulsed Nd:YAG lasers are usually operated in a Q-switching mode. An optical switch is inserted in the laser cavity to wait for the maximum population inversion of the Nd+3 ions before it opens. Q-switching allows high energy outputs (up to 2J/pulse) and pulse durations less than 10ns. The pulse repetition rate of the Nd:YAG laser is limited to 30Hz. This is one of its drawbacks when compared to excimer lasers, which can be operated at pulse repetition rates of several hundred Hz.
Basically, the Nd:YAG lasers are emitting light with the wavelength of 1064nm (infrared). This is well above the desired wavelength range of the PLD applications. However, the light may be frequency doubled or tripled in order to achieve the useful spectral range: for example wavelengths of 355nm and 266nm can be produced. Unfortunately, frequency doubling and tripling decrease the efficiency of the beam to 15-20% relative to the original 1064nm emission. In addition to the pulse repetition rate, the wavelength is another reason why the excimer is more commonly used laser type, as far as the PLD systems are concerned [18, 27].
Fiber lasers also belong to solid state lasers. The active medium is the core material of a fiber, usually silica doped with rare earth metal ions, such as Yb+3 or Er+3. The structure of a laser fiber is typically a single mode fiber, which is illustrated in Figure (1.6). Single mode fiber consists of the active core material, inner cladding layer, a buffer layer and a jacket, or outer cladding, the purpose of which is to protect the fiber from mechanical and chemical damage. The refractive index of the core material is slightly higher than the refractive index of the cladding, to ensure the total internal reflection for the light beam proceeding inside the fiber. The buffer layer enhances the mechanical properties of the fiber, thus it acts as an additional protective component. The buffer layer is not necessary in laser fibers [27, 28].
The lasing effect is accomplished with the help of a pump beam, which is generated e.g. by a diode laser. The pump beam is launched longitudinally along the fiber length and it may be guided by the core material, which is usually the case for single mode fibers, or by the inner cladding [27, 28].
The advantages of fiber lasers (when compared to e.g. Nd:YAG-lasers) are e.g. the compact size of the system and the possibility to produce ultra short pulses at high frequency. However, the use of fiber lasers in the PLD applications is still rare and thus far, there are practically no reports available considering investigations in this area.
Pulse shaping refers to a method, which allows the tailoring of the duration and intensity of the laser pulse. This method is used especially with ultrashort laser pulses. With the help of pulse modulation e.g. a Gaussian-shaped pulse can be split into a pre-pulse with relatively low intensity and a second pulse of a longer duration and higher energy. When the ablation mechanisms of a target are known, the pulse shaping allows an accurate control on deposition, e.g. the generation of particulates can be strongly reduced .
In a PLD-system various optical components are needed to guide, shape and focus the laser beam. Since the laser as well as the optical elements steering the beam are located outside the deposition chamber, the system is really flexible. The beam can be manipulated without altering the arrangement or conditions inside the chamber. On the other hand, the optical components are not affected (e.g. by the reactive gases) during the deposition.
To ensure the good quality of the beam, it is extremely important to keep the optical elements clean. Dust and dirt interfere with the beam path; they cause e.g. scattering and attenuation of the light beam. In addition, contamination may cause permanent damage to the optics. The attenuation of the beam depends also on the length and complexity of the optical path between the laser and the target. Thus, there is a compromise between optimization of the beam quality and energy losses occurring during the optimization [18, 27].
The principal function of the lens is concentrating the light beam emitted by the laser. This is necessary to exceed the threshold energy density of the ablation as the beam hits the target surface. The final energy density of the beam affects e.g. the stoichiometry of the deposit, thus the quality of the film depends also on the properties of the lens. Two types of lenses are typically used in PLD, namely, spherical and cylindrical. With the help of a cylindrical lens, the shape of the beam can be modified, in addition to the focusing.
There are several factors affecting the function of the lens, thus, when optimizing the energy density of the beam at least following parameters should be taken into account: The focal length of the lens
- The transmittance range of the lens material
- The lens aberrations (especially spherical aberration and coma) 
Apertures are used to filter away the most inhomogeneous parts of the beam and to limit its diameter. With the help of the apertures the earlier mentioned lens aberrations can be corrected. As a result of the aperturing, the total energy of the light beam is decreased but the uniformity of the energy distribution is enhanced. The apertures can be placed before mirrors or lenses [18, 27].
Mirrors are used to steer the laser beam onto the target surface. With the help of the mirrors the beam path can be directed to different deposition chambers, thus, the process can be made very cost effective, since one laser can be shared by several chambers. An example of this kind of arrangement is illustrated in Figure (1.7). This figure shows also the correct mutual order of the optical elements: the mirrors, as well as the other components, have to be placed before the focusing lens. The high energy density of the focused beam could deteriorate the surface of the mirror. PLD applications typically use dielectric multilayer mirrors, which are tailored for the wavelength emitted by the laser. The mirror surface is coated with up to 13 layers of different refractive indexes to achieve a maximal reflectance at a specific wavelength .
A beam splitter is a device, which splits one laser beam into two (or more). The resulting beams may be of equal or unequal energy. Figure 7 demonstrates an application, where the original beam is split into two separate beams of equal energy: a multichamber deposition system. The beam splitter can also be used in power monitoring: before the beam enters the deposition chamber, a small amount of the beam is reflected off the splitter and the power is measured. The rest of the beam is transmitted through the splitter and directed into the chamber. The beam splitter is usually a plate made of fused silica. The function of the splitter is based on transmission and reflection, the proportions of which depend on the incidence angle of the beam .
Diffractive optical element
Diffractive optical element (DOE) is a computer generated reflective or transmissive component, which is used for laser beam shaping. Most of the previously mentioned optical elements needed for laser beam shaping and focusing may be replaced with a DOE. This makes the beam manipulating system simpler. When the optical path needed for beam shaping becomes shorter, the losses in the beam intensity are reduced. With the help of the DOE, the intensity distribution of the beam can be tailored. In addition, the shape of the beam can be modified very freely, e.g. the profile can be transformed into a line or an array of spots. Thus, the use of a DOE would make the control of the laser beam both easier and more versatile compared to the conventional arrangement [19, 30, 31].
The shape of the deposition chamber may be e.g. spherical, cylindrical or rectangular. Since the PLD system does not necessarily require ultra-high vacuum, the geometry of the chamber can be designed quite freely. The chamber has typically a large number of ports, e.g. for pumping system, gas inlets, pressure monitoring, target, substrate, laser beam and viewports. When designing a chamber, at least following aspects should be taken into account:
- The arrangement of the components inside the chamber should not disturb the path of the laser beam.
- Access to the target and to the substrate should be straightforward, since these components will be changed frequently.
- The target-substrate distance should be adjustable.
- The deposition of the laser window should be eliminated as well as possible .
The target can be made of e.g. pressed powders, sintered pellets, cast materials, single crystals and metal foils. It should be as dense and homogeneous as possible to ensure a good quality of the deposit. In addition, it is recommended to resurface (grind and/or polish) the target frequently to avoid the generation of the particulates during the deposition. The target is typically disk-shaped and it is often rotated and/or translated during the deposition to achieve uniform erosion. Alternatively, the target can be a cylinder rotating around its axis, although this kind of design is not common. Rastering with the laser beam is another choice to erode the target uniformly, but these results in plume moving, which may be problematic in some cases. If multilayer coatings are to be deposited, a target carousel with various targets can be installed in the chamber. In this case, the targets should be well shielded to avoid the cross contamination of the materials. Automated multitarget coating systems are commercially available .
The most important features of a substrate holder are its movement relative to the target and plasma plume and, in some cases, ability to heat the substrate. Even with the simplest substrate holders, one should be able to adjust the distance between the target and the substrate. (This is usually performed by changing the position of the substrate, not the target.) The adjustment of the target-substrate-distance provides an opportunity to control the deposition rate, as well as the energy of the particles hitting the substrate.
The uniformity of the coating thickness may require rotative and/or translative motion of the substrate during the deposition. In some systems, the angle of the substrate relative to the target may be adjustable, too [18, 32, 33].
Although the low temperature of the substrate (typically <50C) during the deposition is often considered as an advantage of the PLD method, some applications require the heating of the target. An example of this kind of coatings are high temperature superconductors (HTS), the deposition of which is often carried out in temperatures as high as 700-800 C. For the heating of the substrate, e.g. radiative or resistive elements can be used. The temperature and its distribution across the substrate surface can be monitored with the help of thermocouple wires [19, 34].
Generally, the pumping and gas-flow systems required by a PLD setup are simple and economical compared to other thin film deposition techniques. Most of the applications of PLD do not require ultra high vacuum (UHV) deposition conditions (<10-7Pa), so there are various alternatives for the pumping systems. However, since clean and oil-free atmosphere is important for thin film deposition, vacuum pumps with the risk of oil diffusion into the chamber (e.g. diffusion pumps) are not favored. For example turbomolecular pumps are a common choice for PLD-systems. Scroll pumps and rotary vane pumps can be used as backing pumps.
The working pressure during the deposition can vary on a wide range (10-2...10-7Pa) depending on whether a reactive background gas is used or not. This is important to take into account when choosing the pumps and vacuum gauges. A PLD system may contain two or three types of vacuum gauges, e.g. ionization-based gauge (such as cold-cathode gauge) for the lowest pressures, a pirani gauge (based on measurement of thermal conductivity of the gas) for the middle pressure level and a capacitive sensor for the pressures near the atmospheric level [18, 19, 33].
Carbon nitride thin films were prepared by Pulsed Laser Deposition (PLD). Graphite target was exposed to KrF (248 nm) excimer laser. Silicon was used substrate and nitrogen was as ambient atmosphere. excimer laser was employed to graphite for ablation. Ablation took place in the presence of nitrogen atmosphere depositing carbon nitride on the silicon substrate. Carbon nitride thin films were prepared at atmospheric pressure with temperature room temperature, 100, 200, 300 and 350oC for 6000 laser shots.
Following steps were involved in the experiments.
Target Preparation and Substrate Preparation
8 cm diameter disc of graphite was used in the experiment. Scratches on the graphite were removed by grinding. The silicon was in the form of 6 cm diameter disc was cut in the appropiate dimesion according to the requirement.
Abraham Gottlob Werner in 1789 discovered graphite as an allotropic form of carbon. It was named as graphein (??afe?? in Greek) means "to draw, to write" due its usages in pencils. The different properties of graphite are
- Graphite is one of the softest material
- Graphite is a good conductor of electricity
- Graphite is opaque to light
- Graphite is one of the most stable allotrope of carbon
- Graphite is used as a lubricant
An Excimer Laser (KrF) was used for ablation of the graphite target. The beam was focused on the target with the help of the UV lens with focal length 20 cm and passes through a quartz window. The pulse energy was kept constant at 45mJ for all the samples (experiments).
The vacuum chamber is equipped with a rotating multi-target system (rpm =20) with variable rotating speed and also have horizontal motion arrangement with a step of 1mm. A sample holder with heating arrangement is placed at a distance of 10 mm from the target with range from room temperature to 350oC.
Scanning Electron Microscopy
A scanning electron microscope is analogous to an optical microscope, but instead of light it uses electrons to image. The advantage of electrons is that their de Broglie wavelength (=h/p) is on the order of 10 , much less than that of light. This allows smaller features to be imaged. A scanning electron microscope takes pictures of the surface of a specimen by sending an electron beam across its surface. A detector measures the current from the electron beam that is reflected, and also the current from the secondary electrons excited off of the specimen by the incident electrons. The current is interpreted into a visual picture on a monitor. Characterization of the roughness of the films is crucial, since this affects the magnetic and electronic properties of the multilayers.
The X-ray diffraction techniques have been employed in the present research work. The x-ray diffractometer is an outgrowth of the Bragg ionization spectrometer applied in 1913, which disperse a spectrum of x-ray wavelengths by means of a crystal grating.
The x-ray diffractometer is an instrument for studying crystalline materials by measurements of the diffract x-rays of known wavelength. The intensity of the diffracted beam is measured directly by an electronic counter.
Diffractometer system consists of main three parts:
- X-ray Generator.
- Counting and Recording Method
A specimen is mounted in the goniometer at the center. The focusing geometry generally used to obtain high resolution and high-diffracted ray intensities. The specimen makes q-rotation and the X-ray detector rotates through "2q". The focal point (X-ray source), the specimen center and receiving section are always on the same circle so that when X- rays are made to fall on the specimen surface from the focal point at a certain divergent angle, diffracted rays concentrated on the receiving slit in the "2q" direction.
Counting and Recording Methods
This section consists of three components:
- X-rays Detector
- Counting rate meter/fixed time meter,
The counting and recording technique is available in two modes. In the one mode variations of X-ray intensity are recorded continuously on the chart paper together with "2" angle mark of the goniometer. This is the rate meter mode. In the second mode (fixed-time mode) the "2" values and counts in a fixed time are printed on a diffractograph.
A scintillation counter for detection of X-rays diffracted beam is normally used in the diffractometer. This counter consists of a single crystal (NaI) and a photo multiplier tube to whose face the single crystal is cemented. When a diffracted beam falls on the surface of the single crystal, flashes of tube light are produced. This light passes into the Photomultiplier tube and ejects a number of electrons from the photo cathode, which is a photosensitive material generally made of cesium-antimony inter metallic compound.
The emitted electrons at a potential 100V are more positive than the proceeding one, the last one being connected to the measuring circuit. The electrons falling on the first dynode emit more electrons from the second dynode. In this way more electrons are emitted from the last one by absorption of X-rays by the first dynode. Thus a very large number of electrons produce a pulse of an order of volts.
The whole process takes about a less than a microsecond, so that scintillation counter can operate at rate as high as 105 counts per-second without loss.
The counting rate meter is a device that indicates the average counting rate directly. It does this by a circuit that smoothes out the succession of randomly spaced pulses from the counter into a steady current, whose magnitude is proportional to the average rate of pulse production in the counter.
Marvin L. Cohen prediction about the carbon nitride as the harder material than diamond is the recent surge of activity in carbon nitride material. The challenge was accepted to prepare carbon nitride by the scientist and achieved their goal by preparing it.
In my research I have prepared carbon nitride thin film under the following conditions:
Figure (3.a) shows the surface morphology of the same prepared at room temperature in 2D. It is evident from the sample we can differentiate two types of sub-species of CNx associated with the larger agglomerate. The subspecies embedded on the surface of the bigger particle shown by light grey colour and subspecies (light grey colour) attached to the bigger one. Such kind of species is reported in the previous literature .
Figure (3.b) shows film morphology of the sample prepared at 373K. Small agglomerates attached with each other and spread over the micrograph with definite boundary and clustering and well organized shape. Crystalline facets of various grain sizes of CNx films are in accordance to the literature .
In the figure (3.c) we can differentiate two main portions: thin part (light grey) and thick (white) prepared at 473K. These sub-species areproposed as CNx reported by the earlier researchers .
Figure (3.d) shows single agglomerate with sub-species details in the form of the flakes. These flakes are seems to be piled over each other .
Figure (3.e) shows round like flakes are piled upon reach other to give single agglomerate. The microstructure observed under SEM reveals development of typical rounded/globular grain structure due to the effect of nitrogen incorporation to give C3N4 film.
CNx films which are composed of multiphase, especially deposited on the Si substrate. During deposition process Si diffused into the film. SiN, SiC and CNx phases must be formed due to electronegativity difference between C, N and Si. Structure analysis of different phases becomes very difficult due to overlap of the films by X-ray diffraction.
Teter and Hemley [xrd1] have proposed five different structural varieties for this carbon nitride:
- The graphitic C3N4 which consists of an ABAB stacking of graphitic form of carbon nitride.
- Two varieties (called a and ) derived from the a and -Si3N4.
- A ''pseudo-cubic'' structure deduced from (zinc blend type structure with carbon vacancies).
Fourteen peaks are identified for the CNx film deposited at room temperature shown in the above XRD graph and values are shown in table (3.1) [38,39]. Different phases present in the CNx system are a-C3N4, -C3N4, Pseudocubic C3N4, Cubic C3N4 and Graphitic C3N4. The dominated phases in this CNx film are a-C3N4 and -C3N4. Seven peaks of a-C3N4, three peaks of -C3N4, two peaks of Graphitic C3N4, one peak of Pseudocubic C3N4 and Cubic C3N4 are identified. Si and other nitride compounds peaks are omitted for the simplicity.
a-C3N4, -C3N4, Cubic C3N4 and Graphitic C3N4 phases are identified from the XRD graph for the CNX film at 373 K and the values of different parameters are given in the above table (3.2). From the table (3.2) it is evident that Pseudocubic C3N4 is missing in this CNx film prepared at 100oC. Out of nine peaks six peaks of a-C3N4 and one peak for each -C3N4, Graphitic C3N4 and Cubic C3N4. a-C3N4 phase again dominates in this XRD graph[38,39].
Only three a-C3N4, -C3N4 and Graphitic C3N4 phases are identified in this XRD graph and evident from the values given in the table (3.3). Cubic C3N4 and Pseudocubic C3N4 are not present. All the five phases of CNx system are present in the film prepared at room temperature but Pseudocubic C3N4 is missing in the film prepared at 100oC and Pseudocubic C3N4 is not present in the 473 K film. Among the three phases a-C3N4 dominates with eight peaks while -C3N4 and Graphitic C3N4 has only one peak each as shown in the table (3.3) [38, 39].
CNx film prepared at 573K has relatively small peaks values which is clear from the XRD graph. From the table (3.4), it is clear that only two phases exist that are a-C3N4 and -C3N4. a-C3N4 again dominates between the two[xrd5,xrd6]. Twelve peaks are identified for this XRD graph with eight peaks of a-C3N4 and four peaks of -C3N4.
Four phases of CNX films has been identified in CNX film prepared at 623K. a-C3N4 is the dominant phase among the four phases of the CNX film with seven peaks. -C3N4 and Cubic C3N4 one peak each while Graphitic C3N4 has two peaks. Different values of 2Theta and d-spacing is given in the table (3.5) [38, 39].
The temperature versus refractive index graph is shown in figure (3.3). In the spectral range of .3-0.7m the refractive index of the five films increases from 1.5-2.1 with the increase in the temperature from room temperature to 623K. The increase in the refractive index is due to the fact that increase is in temperature causes to increase the amorphous behavior of the films which is supported by XRD graphs.
Raman spectroscopy is widely used to study the vibrational properties of carbon related films. Fig (3.4) shows the Raman spectra between 800 and 2200 cm-1for carbon nitride films deposited at (a) Room Temperature, (b) 373, (c) 473, (d) 573, (e) 623K. The excitation wavelength was 514.52 nm. All the spectra exhibit a broad asymmetric Raman intensity distribution in the range 1000-1700 cm-1 centered at 1578 cm-1. After closer inspection, a shoulder band at 1400 cm-1 can be clearly observed and becomes more pronounced with increasing deposition temperature. The main feature and the shoulder peak have been widely observed in the Raman spectra of diamond-like carbon films, and are assigned to G and D peaks, respectively. The G peak corresponds to the symmetric E2g vibrational mode in graphite-like materials, while the D band arises from the limitation in the graphite domain size, induced by grain boundaries or imperfections, such as substitutional N atoms, sp3 carbon, or other impurities. Due to the overlap, the Raman vibration of C-N and C=N bonds at 1150 and 1650 cm-1 may also exist in the Raman spectra .
Eximer laser (KrF) was employed to produce graphite plasma from the motor driven graphite target. Five single crystal Si samples were taken as substrate to irradiate with 6000 laser induced plasma shots. Si substrates were kept at room temperature, 373, 473, 573 and 623 K. The target to substrate distances was adjusted to 1 cm for each irradiation and nitrogen gas flowing at atmospheric pressure in the vacuum chamber. Graphite plasma reacts with N2 gas to form CNx thin films on the Si substrates.
Characterization of the prepared CNx films is done with Scanning Electron Microscopy (SEM) and X-Ray Diffraction Spectroscopy (XRD) and Raman spectroscopy.
From the SEM micrograph it is evident that there is accumulation of material on the surface of the films in the form of agglomerates and flakes like structures. These structures are reported as CNx in the literature which is supported by the XRD and Raman spectroscopy. From the XRD graph we clearly see that the films amorphous behavior becomes more pronounced as the temperature increase which results in the value of D shown by Raman spectra and refractive index graphs.
- M. Tabbal et al., Effect of process parameters on the mechanical properties of carbon nitride thin films synthesized by plasma assisted pulsed laser deposition, Appl. Phys. A 79 (2004) 1365-1367.
- O. Durand-Drouhin et al., Comparative study of microstructure in a-CxN1-x films deposited by radiofrequency magnetron sputtering, Diamond and Related Materials 10 (2001) 1156-1159.
- J. Neidhardt et al., Correlated high resolution transmission electron microscopy and X-ray photoelectron spectroscopy studies of structured CNx (0 < x < 0.25) thin solid films, Carbon 42 (2004) 2729-2734.
- Mingwu Bai et al., Dependence of microstructure and nanomechanical properties of amorphous carbon nitride thin films on vacuum annealing, Thin Solid Films 376 (2000) 170-178.
- M. Rusop et al., Investigation of structural properties of amorphous carbon nitride thin films prepared by xenon chloride pulsed laser deposition of camphoric carbon precursor, Journal of Materials Science: Materials in Electronics 16 (2005) 367- 375.
- S. Tamulevicius et al., Mechanical properties of ion beam deposited carbon films, Carbon 42 (2004) 1085-1088.
- A. Lagrini et al., Microstructure and electronic investigations of carbon nitride films deposited by RF magnetron sputtering, Thin Solid Films 482 (2005) 41- 44.
- Philippe Merel et al., Phase segregation in pulsed laser deposited carbon nitride thin films, Diamond and Related Materials 12 (2003) 1075-1078.
- Suda et al., Pulsed laser deposition of carbon nitride thin films from graphite targets, Carbon 36 (1998) 771-774.
- Y. Aoi et al., Pulsed laser deposition of amorphous carbon nitride thin films and their electrical properties, Appl. Phys. A 79 (2004) 1533-1536.
- B. Bouchet-Fabre et al., Spectroscopic study using FTIR, Raman, XPS and NEXAFS of carbon nitride thin films deposited by RF magnetron sputtering, Thin Solid Films 482 (2005) 167- 171.
- M. Lejeune et al., Structural relaxation of sputtered amorphous carbon nitride films during thermal annealing, Diamond & Related Materials 17 (2008) 29-35.
- Y. H. Cheng et al., Synthesis of carbon nitride films by direct current plasma assisted pulsed laser deposition, Appl. Phys. A 74 (2002) 225-231.
- G. K. Gueorguiev et al., First-principles calculations on the curvature evolution and cross-linkage in carbon nitride, Chemical Physics Letters 410 (2005) 228-234.
- Niklas Hellgren et al., Role of nitrogen in the formation of hard and elastic CNx thin films by reactive magnetron sputtering, Physical Review B 59 (1999) 5162-5169.
- Y. H. Cheng et al., Dependence of the composition and bonding structure of carbon nitride films deposited by direct current plasma assisted pulsed laser ablation on the deposition temperature, Diamond and Related Materials 11 (2002) 1511-1517.
- D. B. Chrisey and G. K. Hubler (editors), Pulsed Laser Deposition of Thin Films, John Wiley & Sons Inc., New York, 1994.
- W. Boyd, Thin Film Growth by Pulsed Laser Deposition, Ceramics International 22 (1996) 429-434.
- Lenk et al., Diagnostics of laser ablation and laser induced plasmas, Applied Surface Science 106 (1996) 473-477.
- E. Gyrgy et al., Particulates-free Ta thin films obtained by pulsed laser deposition: the role of a second laser in the laser-induced plasma heating, Applied Surface Science 195 (2002) 270-276.
- E. Agostinelli et al., Great reduction of particulates in pulsed laser deposition of Ag-Co films by using a shaded off-axis geometry, Applied Surface Science 156 (2000) 143-148.
- J. M. Lackner et al., Pulsed laser deposition: a new technique for deposition of amorphous SiOx thin films, Surface and Coatings Technology 163-164 (2003) 300-305.
- J. M. Lackner et al., Industrially-scaled large-area and high-rate tribological coating by Pulsed Laser Deposition, Surface and Coatings Technology 200 (2005) 1439-1444.
- N. Pryds et al., Thickness determination of large-area films of yttria-stabilized zirconia produced by pulsed laser deposition, Applied Surface Science 252 (2006)4882-4885.
- O. Svelto, Principles of Lasers, Plenum Press, New York, 1998.
- C. Ristoscu et al., Femtosecond pulse shaping for phase and morphology control in PLD: Synthesis of cubic SiC, Applied Surface Science 252 (2006) 4857-4862.
- J. Stigwall, J. Bengtsson, Design of array of diffractive optical elements with interelement coherent fan-outs, Optics Express 12 (2004) 5675-5683.
- J. M. Lackner, Industrially-styled room-temperature pulsed laser deposition of titanium-based coatings, Vacuum 78 (2005) 73-82.
- J. M. Lackner, Deposition of TiN thin films on three dimensional shaped tools by pulsed laser deposition, Oral Presentation at Materials Week 2002, 30 September- 2 October 2002, Munich (ICM), Paper No. 378.
- W. Biegel, Pulsed laser deposition and characterization of perovskite thin films on various substrates, Applied Surface Science 168 (2000) 227-233.
- M. Mirkowska, E. Wierzbinski, K. Zdunek,Growth of nanopillar CNx layer during impulse plasma deposition, Surface & Coatings Technology 200 (2006) 4448 - 4455.
- Z.Q. Lia, J.Y. Zhoua, J. Zhanga, T.B. Chena, J. Yuan, Carbon nitrides synthesized by glow discharge method, Journal of Alloys and Compounds 346 (2002) 230-234.
- S. Matsumoto, E.-Q. Xie, F. Izumi, On the validity of the formation of crystalline carbon nitrides, C3N4, Diamond and Related Materials 8 (1999) 1175-1182.
- Y.P. Zhang, Y.S. Gu, X.R. Chang, Z.Z. Tian, D.X. Shi, X.F. Zhang, On the structure and composition of crystalline carbon nitride films synthesized by microwave plasma chemical vapor deposition, Materials Science and Engineering B78 (2000) 11-15.
- S. Chowdhury and M. T. Laugier, Evidence of nanodomes in carbon nitride thin films, phys. stat. sol. (a) 201, No. 2, R1-R4 (2004)