Final Report Summary - CAMELIA (Monolithic above in ultra high value capacitors for mobile and wireless communication systems)
The project addressed the challenges of manufacturing thin film, high capacitance, above chip (hence low temperature) components by developing a novel low-cost ferroelectric multicomponent oxide based materials science and processing technology, enabling both the integration of decoupling capacitors with a very small form factor in an above-IC strategy for system on chip and wireless microsystems and, the development of novel piezoelectric switches.
The starting point was the development of novel materials and processes and the end point was the fabrication of high value capacitors for integration into a test platform. The piezoelectric properties were also investigated. The project adopted an integrated approach starting with the development of novel materials and processing routes, the characterisation of the materials, in particular the electrical properties, and the integration of the materials into test structures.
The PVD and chemical solution deposition methods developed at CEA-LETI, JSI and TNI produced high dielectric constant / low loss films. PVD was used not only to grow films but to provide seed layers. Mg deficiency was a problem in the PVD process and it was found that a PZT buffer layer was important in promoting the perovskite phase. The CVD important safety considerations associated with lead compounds and, although films were grown, the cost was found to be prohibitive at this stage. The addition of nanocrystals did not have the desired effect of promoting the perovskite phase as initially envisaged.
Using the standard film deposition and annealing approach, good dielectric properties were obtained at high annealing temperatures. The results from the visible laser program showed that laser annealing could be used to produce high dielectric constant (> 1000) films of PZT and CCTO. However, the short-pulse UV trials did not produce high dielectric constant materials. The ability to process at temperatures below 400 degrees Celsius could be developed to make it compatible with microelectronic processes such as CMOS.
It was shown that improvements to the CCTO properties could be made by growing a more columnar microstructure. This gave more order to the crystalline structure. It was also found that the annealing environment significantly affects the resulting dielectric properties with nitrogen giving the highest dielectric constant.
Within CAMELIA only flat structures were processed. Deposition onto a high aspect ratio structure was demonstrated using atomic layer deposition but only as a proof-principle using non-CAMELIA material. This would be worth pursuing in the future. Evaluation of the film growing techniques showed that although capacitance densities of 300 nF/mm2 were achievable, the breakdown voltage required for the SORIN application was high and for a capacitor dielectric thickness required to give a breakdown voltage of 28 V, the capacitance density reduces to 40 nF/mm2. This high voltage requirement was not foreseen at the outset.
The capacitor test structure was designed and fabricated along with a modified existing piezo-electric test rig. The detailed reliability study was not pursued and the resources used to support the laser annealing trials. Measurements on the capacitor test structures showed that sol-gel PMNT had a dielectric constant of 2 000 at 10 kHz and an Intrinsic effective piezoelectric coefficient of 3.1 C/m2.
The end point of the project was the successful integration of high dielectric capacitors with the specially designed rebuilt wafer stack by 3DPlus to produce a pacemaker. The chosen route was to use the PMNT sol-gel from TNI but there was a problem in introducing the chemical precursors into the LETI fabrication facility. The PZT route was used as it was the best method in the time available. The sol-gel can easily be substituted. In essence the integration platforms produced did not use the developed CAMELIA material but post CAMELIA CEA and 3DP will integrate PMNT capacitors in the rebuilt wafer. In addition, Sorin, 3DP and CEA will collaborate in a French-funded project to develop the CAMELIA ideas further.
The CAMELIA consortium represented a world-class interdisciplinary research team with leading experts in the fields of thin film growth and low temperature chemical vapour deposition reaction pathways, chemically-synthesised and size-controlled nanoparticles. The consortium also included two end-users who benefited from the novel capacitor technology developed. The consortium was vertically integrated with the Electronic Ceramics Department from JSI performing research related to chemical solution deposition of thin films. The work within CAMELIA allowed the group to gain new knowledge in the design and synthesis of solution precursors for thin films.
Specifically, research addressed the development of a promising lead-free material CCTO (CaCu3Ti4O12). Tyndall developed nanoparticle systems and thin film growth. CEA-LETI compared the properties the novel oxide material systems and multilayers developed in the CAMELIA project with his current PVD technique in terms performance and integrability on 200 mm wafers. 3D Plus developed new architectures and related prototypes for a multipurpose 3D technology and product platform. Sorin Group's business is in ICD (implantable cardioverter defibrillator) and pacemakers and they integrated the technology into one of their demonstrator platforms.
The starting point was the development of novel materials and processes and the end point was the fabrication of high value capacitors for integration into a test platform. The piezoelectric properties were also investigated. The project adopted an integrated approach starting with the development of novel materials and processing routes, the characterisation of the materials, in particular the electrical properties, and the integration of the materials into test structures.
The PVD and chemical solution deposition methods developed at CEA-LETI, JSI and TNI produced high dielectric constant / low loss films. PVD was used not only to grow films but to provide seed layers. Mg deficiency was a problem in the PVD process and it was found that a PZT buffer layer was important in promoting the perovskite phase. The CVD important safety considerations associated with lead compounds and, although films were grown, the cost was found to be prohibitive at this stage. The addition of nanocrystals did not have the desired effect of promoting the perovskite phase as initially envisaged.
Using the standard film deposition and annealing approach, good dielectric properties were obtained at high annealing temperatures. The results from the visible laser program showed that laser annealing could be used to produce high dielectric constant (> 1000) films of PZT and CCTO. However, the short-pulse UV trials did not produce high dielectric constant materials. The ability to process at temperatures below 400 degrees Celsius could be developed to make it compatible with microelectronic processes such as CMOS.
It was shown that improvements to the CCTO properties could be made by growing a more columnar microstructure. This gave more order to the crystalline structure. It was also found that the annealing environment significantly affects the resulting dielectric properties with nitrogen giving the highest dielectric constant.
Within CAMELIA only flat structures were processed. Deposition onto a high aspect ratio structure was demonstrated using atomic layer deposition but only as a proof-principle using non-CAMELIA material. This would be worth pursuing in the future. Evaluation of the film growing techniques showed that although capacitance densities of 300 nF/mm2 were achievable, the breakdown voltage required for the SORIN application was high and for a capacitor dielectric thickness required to give a breakdown voltage of 28 V, the capacitance density reduces to 40 nF/mm2. This high voltage requirement was not foreseen at the outset.
The capacitor test structure was designed and fabricated along with a modified existing piezo-electric test rig. The detailed reliability study was not pursued and the resources used to support the laser annealing trials. Measurements on the capacitor test structures showed that sol-gel PMNT had a dielectric constant of 2 000 at 10 kHz and an Intrinsic effective piezoelectric coefficient of 3.1 C/m2.
The end point of the project was the successful integration of high dielectric capacitors with the specially designed rebuilt wafer stack by 3DPlus to produce a pacemaker. The chosen route was to use the PMNT sol-gel from TNI but there was a problem in introducing the chemical precursors into the LETI fabrication facility. The PZT route was used as it was the best method in the time available. The sol-gel can easily be substituted. In essence the integration platforms produced did not use the developed CAMELIA material but post CAMELIA CEA and 3DP will integrate PMNT capacitors in the rebuilt wafer. In addition, Sorin, 3DP and CEA will collaborate in a French-funded project to develop the CAMELIA ideas further.
The CAMELIA consortium represented a world-class interdisciplinary research team with leading experts in the fields of thin film growth and low temperature chemical vapour deposition reaction pathways, chemically-synthesised and size-controlled nanoparticles. The consortium also included two end-users who benefited from the novel capacitor technology developed. The consortium was vertically integrated with the Electronic Ceramics Department from JSI performing research related to chemical solution deposition of thin films. The work within CAMELIA allowed the group to gain new knowledge in the design and synthesis of solution precursors for thin films.
Specifically, research addressed the development of a promising lead-free material CCTO (CaCu3Ti4O12). Tyndall developed nanoparticle systems and thin film growth. CEA-LETI compared the properties the novel oxide material systems and multilayers developed in the CAMELIA project with his current PVD technique in terms performance and integrability on 200 mm wafers. 3D Plus developed new architectures and related prototypes for a multipurpose 3D technology and product platform. Sorin Group's business is in ICD (implantable cardioverter defibrillator) and pacemakers and they integrated the technology into one of their demonstrator platforms.
