Thin Film Deposition of Functional Materials by Pulsed Laser Deposition

In today’s world, functional materials are part of everyday life e.g. a protective finish on devices, object or tools to reduce wear and tear, coatings for decoration, sensors or medical applications, or magnetic coatings to store information on our hard drives. These materials are deposited as thin films using a large variety of deposition techniques and the desired film properties will determine how the film will be grown.

Also the thickness of the coating varies greatly ranging from thick films with thicknesses of tens of microns down to a few nm, termed ultra-thin films. Obviously, depending on cost, efficiency and control of the deposition process there are non-vacuum techniques available like spray pyrolysis or chemical vapour deposition. Both techniques are ideally suited for large area, thick film deposition but they depend on solvents which have to be removed subsequently by a heating step from the deposited layer. This is different for vacuum deposition techniques such as sputtering, thermal evaporation or molecular beam epitaxy. They offer a much better control over the thickness, roughness, purity, crystallinity, and physical properties, e.g. electrical conductivity.

Pulsed Laser Deposition

One of the most versatile deposition techniques in solid state physics and analytical chemistry is the vaporization of condensed matter using photons. Pulsed laser deposition (PLD) is a growth technique in which the photon, characterized by pulse duration, laser wavelength and fluence, interacts with a bulk material [1, 2]. A schematic of this deposition technique is shown in Fig. 1. A short-pulsed high-power laser beam ranging from ~100 fsec to 20 nsec is focused onto a sample surface thereby converting a finite volume of a solid instantaneously into its vapour phase constituents such as ions and neutrals, thereby forming a luminous ablation plume. Subsequently, the vapour moves away from the target at a high velocity (several km/sec) and can be sampled either to grow a film or for analysis by various spectroscopic techniques.

Fig. 1. a) Schematic of pulsed laser deposition set-ups. The incoming laser beam is focused onto a target, thereby vaporizing the material of the surface region. The ejected material is partially ionized and forms the ablation plume which is directed towards the substrate. b) Schematic of a RF-plasma enhanced pulsed laser deposition [3]. c) Schematic of a gas-pulse set-up combined with PLD also known as pulsed reactive crossed beam laser ablation [4]. The two beams merge after passing the interaction zone and expand together. Under each schematic, a photograph shows the plasma plume for the corresponding PLD technique (adapted from [1], images courtesy of M. Bator and D. Stender).

To enhance the reactivity of the background gas with the ablated species, either a RF-plasma source [3] (Fig. 1b) or a gas pulse configuration [4] (Fig. 1c) are used. The latter technique can provide more reactive species to the ablation process e.g. atomic oxygen to improve the oxygen content of an as-grown film [5], or the controlled anionic substitution in the gas phase e.g. to grow N-doped SrTiO₃ or LaTiO_3-xN_x by using N₂ or NH₃ in the gas pulse [6, 7].

Advantages of Pulsed Laser Deposition

Pulsed laser deposition for thin film growth has several advantages which are listed below:

• The flexibility in wavelength and power density allows the process to ablate many material or materials combination by selecting the appropriate laser wavelength in order to match the absorption properties of materials.
• Large pressure range to deposit materials: from <10^-7 mbar (vacuum without additional background gas) up to 1mbar.
• The laser is not part of the vacuum system. Therefore a considerable degree of freedom in the ablation geometry is possible.
• The use of a pulsed laser beam enables precise control over the growth rate (sub-monolayer per pulse).
• The congruent transfer of the composition can be achieved for many ablated material or materials combinations.
• Moderation of the kinetic energy of evaporated species to control the growth properties and growth modes of a film. In addition, a background gas can provide an appropriate reactive atmosphere using e.g. oxygen to create oxide species in the plasma, when growing oxide films.
• Controlled preparation of nano-particles by fs-PLD [8].

Disadvantages of Pulsed Laser Deposition

There are also disadvantages to associated with the PLD process. Some of them are of a technical nature; some are intrinsic to the ablation process and the electromagnetic interaction between photons and matter [2]:

• The large kinetic energy of some plume species causes re-sputtering and likewise defects in the substrate surface and growing film.
• An inhomogeneous energy distribution in the laser beam profile gives rise to an inhomogeneous energy profile and angular energy distribution in the laser plume.
• Light elements like oxygen or lithium have different expansion velocities and angular distributions in a plume as compared to heavier elements. To obtain the desired film composition, e.g. an adapted target composition or a background gas is required.
• Due to the high laser energies involved, macroscopic and microscopic particles from the target can be ejected which can be detrimental to the desired properties of films and multilayers.

The latter point can partially be overcome by working with very dense polycrystalline or even single crystalline targets, but it also depends on the absorption and mechanical properties of the target material and laser fluence used. One example where a direct emission of micron sized particles has been observed is ns-PLD of 8-YSZ as a consequence of delaminating the target surface [9]. To prevent particles reaching the film surface, the gas-pulse has been used as an “air-brush” [10]. Due to the directed nature of the laser plasma plume, PLD is often used for small scale sample preparation. Prominent examples where PLD has been successfully adopted for large scale applications can be found in [11-14].

Plasma Diagnostics

Often, PLD is used as a black-box technique since one doesn’t need to know what happens in a plasma and they can still prepare high quality thin films and multilayers. However, the laser induced plasma is the heart of the ablation process where most of the chemistry takes place which defines subsequently the properties of a film.

Figure 2 shows one example for plasma mass spectrometry of La_0.4Ca_0.6MnO₃ ablated at l=193nm, with a fluence of 1.5 J/cm² using different background conditions: vacuum, O₂ and N₂O at p=1.5x10^-1 Pa. Striking is the difference in the number of oxygen-containing species measured in vacuum, O₂ and N₂O (Fig. 2). The amount of metal-oxygen species is significantly enhanced if the ablation takes place in N₂O as compared to O₂. In addition, supplying atomic oxygen by introducing N₂O increases the amount of negative species in a plasma almost by a factor of two and the achieved oxygen stoichiometry is closer to the expected value without additional oxygen annealing steps [15].

Fig. 2. Plasma mass spectra of negative species of ablated La_0.4Ca_0.6MnO₃ measured in vacuum, O₂ and N₂O at p=1.5x10^-1 Pa. The excimer laser wavelength is l=193nm with a fluence of 1.5 J/cm².

Another useful tool for plasma diagnostics is time resolved plasma imaging [16-18] to study the expansion of excited species in a plasma. By selecting the emission wavelength for one species, it is possible to distinguish between the different expansion velocities of species and processes involved during the time of expansion. The expansion dynamics of light and heavy elements is very different and the mass ratio of the species involved will determine if a congruent transfer can be obtained. This has been studied e.g. for the battery electrode material LiMn₂O₄ which is as a PLD grown thin film inherently Li-deficient [19]. To achieve a congruent transfer of LiMn₂O₄, the Li content of the target composition needs to be adjusted [19, 20].

Thin Film Growth

The growth of oxide thin films involves a number of boundary conditions: for epitaxial growth the lattice matching to a substrate, chemical compatibility, comparable thermal expansion coefficients, as well as a thermodynamically and chemically stable substrate surfaces are important. For some applications it is also important to choose selectively the chemical termination of the surface [21, 22]. In Figure 3, there are three examples to demonstrate how growth modes and growth conditions can influence profoundly the quality and hence physical properties of the final film. La_0.4Ca_0.6MnO₃ thin films grown on (100) SrTiO₃ in vacuum (T_S = 750°C, F=1.5 Jcm^-2 and l=193nm in vaccum) are single phase, strain free with a (100) orientation but the crystallinity is poor with a rocking curve width of >2°. For La_0.4Ca_0.6MnO₃ films grown in O₂ and N₂O (p=1.5x10^-1 Pa) the crystalline orientation changes to (001), and the rocking curve width of 0.15° and <0.07° is indicative of a good to already excellent crystalline quality for the as-grown (001) oriented films [15] (see Fig. 3a).

The origin of the change in crystalline orientation and quality is due to the moderation of the kinetic energy of the plasma species and the oxidizing background. Fig. 3b) shows one example of a single crystalline like growth of a (110) LuMnO₃ thin film grown on (110) YAlO₃ substrate. Due to a good lattice match between the film and substrate a twin-free growth can be achieved as demonstrated by the reciprocal space map for the (221) film and substrate orientation, which also shows some in plane Laue oscillation [23].

To provoke the growth of a film at an angle like 45° with respect to the surface normal, a substrate is placed at the appropriate angle with respect to the incoming materials flux typically originating from a thermal source. This growth technique is called inclined substrate deposition (ISD) and has been successfully demonstrated e.g. for the growth of MgO on Ni-tapes using thermal evaporation [24]. The PLD process typically has species with a larger kinetic energy compared to thermal evaporation. However, by carefully selecting the deposition conditions of the ion conductor Y-stabilized ZrO₂ (YSZ) and tilting the substrate to 45°, likewise an inclined growth can be provoked (see Fig. 3c). Here, two layers of YSZ with a substrate inclination of ±45° have been grown on SrTiO₃ [25].

Fig. 3. a) Rocking curves of the (100) (inset) and (001) film peaks for La_0.4Ca_0.6MnO₃ grown at T_S = 750°C with a fluence of 1.5 Jcm^-2 and l=193nm in vacuum (inset), and a O₂ and N₂O background pressure of 1.5x10^-1 Pa [15]. b) Reciprocal space map of the (221) film and substrate orientation of a 70nm (110) LuMnO₃ thin film grown on a (110) YAlO₃ substrate. The LuMnO₃ thin film is twin free and shows in-plane Laue oscillations [26]. c) SEM cross section of a PLD grown YSZ thin films deposited on SrTiO₃ at 45° and -45° to obtain a zig-zag like growth structure [25].

Polymer Film Transfer

As for metals and oxides, PLD of polymers is also possible and has been successfully applied. Due to the nature of the ablation process, the process only for certain polymers such as Teflon [27] and PMMA [28]. These polymers depolymerize upon irradiation and subsequently polymerize with a different, most probably lower molecular weight structure on the substrate. To prevent this and make polymer deposition available for sensitive materials, matrix-assisted pulsed laser evaporation (MAPLE) was developed. Using a frozen solvent target containing the material to be deposited upon irradiation (the matrix), the solvent is evaporated and forms a plume together with the molecules of interest [29-32]. Polymers or other materials can also be transported by transferring a layer from a donor to a receiver substrate while conserving the shape defined by the laser spot. This technique is called laser-induced forward transfer (LIFT) where the donor material is illuminated from the back of the substrate with the laser beam, subsequently ejected as an intact flyer and re-deposited onto the receiver substrate. Ideally, the ablation conditions for the donor film are such, that the film is ejected in a compact flyer and collected in one piece on the receiver substrate. To reduce or even completely avoid thermal decomposition of ablated polymers, a sacrificial layer can be added which absorbs all the laser energy (dynamic release layer, DRL). The decomposition of the DRL and the subsequent pressure build-up transfers the polymer across a gap onto the receiver substrate. Likewise, a pattern transfer can be achieved with the receiver substrate in contact with the film. An example for a transfer of a complete three color organic polymer LED with two doped (green and red) and one undoped (blue) Polyfluorene (PFO) as the light-emitting polymer is shown in Figure 4 [33].

Fig. 4. Schematic diagram of the LIFT process in three steps: Step 1 shows the prepared substrate architectures; the dashed line separates the donor substrate from the receiver substrate. Step 2 shows the transfer process as the pressure build up from the gaseous products of triazene ablation push the overlying layers onto the receiver substrate. Step 3 shows the receiver substrate after the donor substrate has been taken away and a bias is applied across the device for light emission (adopted from [33]).

References

1. C. W. Schneider and T. Lippert, in Springer Series in Material Science, edited by P. Schaaf (Springer, Berlin, 2010), Vol. 139, pp. 89-112.
2. P. R. Willmott and J. R. Huber, Rev. Mod. Phys. 72 (1), 315-328 (2000).
3. M. Braic, M. Balaceanu, A. Vladescu, A. Kiss, V. Braic, G. Epurescu, G. Dinescu, A. Moldovan, R. Birjega and M. Dinescu, Applied Surface Science 253 (19), 8210-8214 (2007).
4. P. R. Willmott, R. Timm and J. R. Huber, Journal of Applied Physics 82 (5), 2082-2092 (1997).
5. M. J. Montenegro, K. Conder, M. Döbeli, T. Lippert, P. R. Willmott and A. Wokaun, Applied Surface Science 252 (13), 4642-4646 (2006).
6. I. Marozau, M. Döbeli, T. Lippert, D. Logvinovich, M. Mallepell, A. Shkabko, A. Weidenkaff and A. Wokaun, Applied Physics A: Materials Science & Processing 89 (4), 933-940 (2007).
7. I. Marozau, A. Shkabko, M. Döbeli, T. Lippert, M. Mallepell, C. W. Schneider, A. Weidenkaff and A. Wokaun, Acta Materialia 59 (19), 7145-7154 (2011).
8. S. Amoruso, R. Bruzzese, N. Spinelli, R. Velotta, M. Vitiello, X. Wang, G. Ausanio, V. Iannotti and L. Lanotte, Appl. Phys. Lett. 84 (22), 4502-4504 (2004).
9. S. Heiroth, J. Koch, T. Lippert, A. Wokaun, D. Gunther, F. Garrelie and M. Guillermin, Journal of Applied Physics 107 (1), 014908-014910 (2010).
10. S. Heiroth, ETH, 2010.
11. M. Lorenz, H. Hochmuth, D. Natusch, M. Kusunoki, V. L. Svetchnikov, V. Riede, I. Stanca, G. Kastner and D. Hesse, IEEE Trans. Appl. Supercond. 11 (1), 3209-3212 (2001).
12. H. W. Neumuller, W. Schmidt, H. Kinder, H. C. Freyhardt, B. Stritzker, R. Wordenweber and V. Kirchhoff, J. Alloy. Compd. 251 (1-2), 366-372 (1997).
13. https://www.coherent.com
14. J. M. Lackner, Surface and Coatings Technology 200 (5–6), 1439-1444 (2005).
15. M. Esposito, M. Bator, M. Dobeli, T. Lippert, C. W. Schneider and A. Wokaun, Appl. Phys. Lett. 99 (19) (2011).
16. S. Amoruso, R. Bruzzese, N. Spinelli and R. Velotta, Journal of Physics B: Atomic, Molecular and Optical Physics 32 (14), R131 (1999).
17. D. B. Geohegan, A. A. Puretzky, G. Duscher and S. J. Pennycook, Appl. Phys. Lett. 72 (23), 2987-2989 (1998).
18. D. B. Geohegan and A. A. Puretzky, Applied Surface Science 96–98 (0), 131-138 (1996).
19. S. Canulescu, E. L. Papadopoulou, D. Anglos, T. Lippert, C. W. Schneider and A. Wokaun, Journal of Applied Physics 105 (6), 063107-063111 (2009).
20. F. Simmen, T. Lippert, P. Novak, B. Neuenschwander, M. Dobeli, M. Mallepell and A. Wokaun, Appl. Phys. A-Mater. Sci. Process. 93 (3), 711-716 (2008).
21. M. Kawasaki, K. Takahashi, T. Maeda, R. Tsuchiya, M. Shinohara, O. Ishiyama, T. Yonezawa, M. Yoshimoto and H. Koinuma, Science 266 (5190), 1540-1542 (1994).
22. G. Koster, B. L. Kropman, G. J. H. M. Rijnders, D. H. A. Blank and H. Rogalla, Appl. Phys. Lett. 73 (20), 2920-2922 (1998).
23. Y. Hu, M. Bator, M. Kenzelmann, T. Lippert, C. Niedermayer, C. W. Schneider and A. Wokaun, Applied Surface Science (in press) (2012).
24. M. Bauer, R. Metzger, R. Semerad, P. Berberich and H. Kinder, in Fundamental Mechanisms of Low-Energy-Beam-Modified Surface Growth and Processing, edited by S. C. Moss, E. H. Chason, B. H. Cooper, J. M. E. Harper, T. D. DeLaRubia and M. V. R. Murty (Materials Research Society, Warrendale, 2000), Vol. 585, pp. 35-44.
25. D. Stender and T. Lippert, unpublished (2012).
26. Y. Hu, C. Borca, B. Delley, M. Janousch, M. Kenzelmann, M. Nachtegaal, T. Lippert, A. Wokaun and C. W. Schneider, submitted (2012).
27. G. B. Blanchet, Macromolecules 28 (13), 4603-4607 (1995).
28. E. Suske, T. Scharf, H.-U. Krebs, T. Junkers and M. Buback, Journal of Applied Physics 100 (1), 014906-014904 (2006).
29. D. B. Chrisey, A. Piqué, R. A. McGill, J. S. Horwitz, B. R. Ringeisen, D. M. Bubb and P. K. Wu, Chemical Reviews 103 (2), 553-576 (2003).
30. D. B. Chrisey, A. Pique, J. Fitz-Gerald, R. C. Y. Auyeung, R. A. McGill, H. D. Wu and M. Duignan, Applied Surface Science 154–155 (0), 593-600 (2000).
31. A. Purice, J. Schou, P. Kingshott, N. Pryds and M. Dinescu, Applied Surface Science 253 (19), 7952-7956 (2007).
32. A. Purice, J. Schou, P. Kingshott and M. Dinescu, Chemical Physics Letters 435 (4–6), 350-353 (2007).
33. J. Shaw-Stewart, T. Lippert, M. Nagel, F. Nüesch and A. Wokaun, ACS Applied Materials & Interfaces 3 (2), 309-316 (2011).