Design challenges for matrix assisted pulsed laser evaporation and infrared resonant laser evaporation equipment

Abstract

Since the development of the Matrix Assisted Pulsed Laser Evaporation (MAPLE) process by the Naval Research Laboratory (NRL) in the late 1990s, MAPLE has become an active area of research for the deposition of a variety of polymer, biological, and organic thin films. As is often the case with advancements in thin-film deposition techniques new technology sometimes evolves by making minor or major adjustments to existing deposition process equipment and techniques. This is usually the quickest and least expensive way to try out new ideas and to "push the envelope" in order to obtain new and unique scientific results as quickly as possible. This process of "tweaking" current equipment usually works to some degree, but once the new process is further refined overall designs for a new deposition tool based on the critical attributes of the new process typically help capitalize more fully on the all the salient features of the new and improved process. This certainly has been true for the MAPLE process. In fact the first MAPLE experiments the polymer/solvent matrix was mixed and poured into a copper holder held at LN 2 temperature on a laboratory counter top. The holder was then quickly placed onto a LN 2 cooled reservoir in a vacuum deposition chamber and placed in a vertical position on a LN 2 cooled stage and pumped down as quickly as possible. If the sample was not placed into the chamber quickly enough the frozen matrix would melt and drip into the bottom of the chamber onto the chambers main gate valve making a bit of a mess. However, skilled and motivated scientists usually worked quickly enough to make this process work most of the time. The initial results from these experiments were encouraging and led to several publications which sparked considerable interest in this newly developed technique Clearly this approach provided the vision that MAPLE was a viable deposition process, but the equipment was not optimal for conducting MAPLE experiments on a regular basis for several reasons. The first reason is that the polymer/solvent mix as well as the sample holder are both exposed to the humidity in the air which will coat the entire surface of the holder and target with water vapor. Some polymer and/or solvent materials may not react well with water vapor. Also, the layer of water vapor absorbed on the target surface may then absorb the incident laser radiation until it is removed from the surface. Thus, it may be unclear when the water vapor is fully removed from the polymer/solvent surface and the MAPLE deposition process actually occurs. This makes deposition of specific polymer thickness difficult to calculate. While it is well known that Quartz crystal microbalances do not work well for PLD of oxide materials it can be used for the deposition of MAPLE materials. However, with rastered laser beams the tooling factor becomes a dynamic number making interpretation of final thickness potentially difficult without careful pre-calibration. Another serious issue with the initial MAPLE process was related to the use of UV lasers such as an excimer operating at 193- or 248-nm or frequency tripled, Nd:YAG lasers at 355 nm. These lasers have high energy per photon (between about 6.4 to 3.5 eV) which can lead to a variety of deleterious photochemical mechanisms that can damage the polymer chains or organic structure. Such mechanisms can be direct photo-decomposition by photochemical bond breaking and photothermal effects. Alternative lasers, such as a Er:YAG laser operating at 2.9 microns produce photons with energy of ∼0.43 eV. Such longer wavelength lasers have been used for the IR-MAPLE process and may be very useful for future MAPLE systems. A third issue with the initial approach to MAPLE was that the process did not lend itself easily to growing multilayer films. Most standard pulsed laser deposition tools have "multi-target" carousels that allow for easy target changes and multilayer film growth. This is true for sputtering, MBE, and evaporation equipment as well. This multilayer feature would certainly benefit the MAPLE process for the growth of multilayer organic materials. Another more recent advancement in thin-film laser deposition is that of Resonant Infra Red Pulsed Laser Deposition (RIRPLD) of polymer materials. This process is more akin to standard PLD but uses tunable lasers with which to select the proper wavelength to couple to vibration bands of a solid polymer, or in some cases a polymer/solvent MAPLE mixture. This technique was developed under a collaboration of researchers at the Naval Research Labs and the Free Electron Laser (FEL) at Vanderbilt University. The wide tuning range of the FEL and its relatively high power make it a very attractive source for RIRPLD. However, the price of such lasers-of order several million dollars in capital costs alone-is very high and well beyond the budgets of most research institutions. Advances in RIRPLD are currently limited due to the scarcity of tunable lasers with sufficient power in the IR range of interest to obtain reasonable deposition rates. Over the past nine years commercial equipment for MAPLE has been on the market and new lasers are being developed that may significantly improve MAPLE and RIRPLD capabilities. Examples of basic single-target MAPLE equipment, as well as multiple target MAPLE systems are described. Discussion of current lasers for MAPLE and RIRPLD are given. Finally, even though these processes have been around for a significant amount of time there are still many unknowns associated with these techniques that still should be explored before these processes can be used for production of useful products. Some of these issues which need to be addressed will be discussed. © 2011 The Author(s).