Daniel Pantel

Pulsed laser deposition of piezoelectric lead zirconate titanate thin films maintaining a post-CMOS compatible thermal budget

Integration of lead zirconate titanate (Pb[Zrx,Ti1-x]O3 – PZT) thin films on complementary metal-oxide semiconductor substrates (CMOS) is difficult due to the usually high crystallization temperature of the piezoelectric perovskite PZT phase, which harms the CMOS circuits. In this work, a wafer-scale pulsed laser deposition tool was used to grow 1 μm thick PZT thin films on 150 mm diameter silicon wafers. Three different routes towards a post-CMOS compatible deposition process were investigated, maintaining a post-CMOS compatible thermal budget limit of 445 °C for 1 h (or 420 °C for 6 h). By crystallizing the perovskite LaNiO3 seed layer at 445 °C, the PZT deposition temperature can be lowered to below 400 °C, yielding a transverse piezoelectric coefficient e31,f of −9.3 C/m2. With the same procedure, applying a slightly higher PZT deposition temperature of 420 °C, an e31,f of −10.3 C/m2 can be reached. The low leakage current density of below 3 × 10−6 A/cm2 at 200 kV/cm allows for application of the post...

Application of DoE methods to establish a model for the pulsed laser deposition of PZT thin-films

Controlling the growth of piezoelectric lead zirconate titanate (Pb[ZrxTi1-x]O3 - PZT) thin-films is a key issue for their application in sensor and actuator devices. The pulsed laser deposition (PLD) process which is used in this work to grow PZT on platinized 150 mm silicon wafers can be easily modified by changing deposition parameters as the deposition pressure, deposition temperature, laser spot area on the target or laser pulse energy. In order to investigate the influence of these PLD parameters on the deposition rate and properties of the PZT thin-film as the leakage current density or the transverse piezoelectric coefficient e31,f, statistical Design of Experiment methods were applied. Empirical models could be derived, describing the thin-film properties as function of the deposition parameters. It is observed that not only the PZT deposition parameters but also the parameters which are modified for the deposition of the LaNiO3 (LNO) template need to be controlled carefully. The parameters with the highest effect on the e31,f are shown to be the combination of deposition pressure and deposition temperature of the PZT layer, where either both high or low values lead to good PZT properties.