Exploration of Metal Composites and Carbon Nanotubes for Thermal Interfaces
Doctoral thesis, 2020
Modern microelectronics are perpetually pushing against limitations caused by inadequate heat dissipation. One of the critical bottlenecks is at the interfaces between different materials and components. Thermal interface materials (TIM) are used to improve the heat transfer at these interfaces, and to improve TIMs is one of the critical research areas in order to reduce the total thermal resistance for electronics systems.
A TIM requires both high thermal conductivity, ability to conform to mating surfaces, and the ability to absorb stress from thermal expansion mismatch during thermal cycling. Solder based TIMs utilize solder to form a strong connection between the mating surfaces with high thermal conductivity, but their stiffness prevents adequate absorption of thermal expansion mismatch. In this thesis, the solder is combined with a fiber network phase, which modifies the mechanical properties, while maintaining the continuous heat paths within the solder. This solder matrix fiber network composite TIM allows for the tailoring of the mechanical properties of solder based TIM while retaining thermal performance.
Another promising TIM candidate is based on arrays of vertically aligned CNTs. CNT arrays can achieve good thermal performance, but the reliability had not previously been investigated experimentally. A thorough investigation of the reliability of CNT array TIM revealed that reliability is not guaranteed, but requires careful matching between CNT array height, bonding method and substrate configuration.
Furthermore, we developed a new joule self-heating chemical vapor deposition (CVD) method for the synthesis of double-sided CNT arrays on thin foils, which can be used both as TIM or as supercapacitor electrodes. Double-sided arrays are challenging with conventional CNT array synthesis methods, but the Joule heating CVD method allows for rapid, scalable and uniform synthesis of large area double-sided arrays. Finally, this method was used to study the effect of heat treatment of CNT arrays on graphite. The heat treatment serves to simultaneously improve the CNT crystallinity, eliminate catalyst residues, and form a seamless connection between CNT arrays and graphite.
A TIM requires both high thermal conductivity, ability to conform to mating surfaces, and the ability to absorb stress from thermal expansion mismatch during thermal cycling. Solder based TIMs utilize solder to form a strong connection between the mating surfaces with high thermal conductivity, but their stiffness prevents adequate absorption of thermal expansion mismatch. In this thesis, the solder is combined with a fiber network phase, which modifies the mechanical properties, while maintaining the continuous heat paths within the solder. This solder matrix fiber network composite TIM allows for the tailoring of the mechanical properties of solder based TIM while retaining thermal performance.
Another promising TIM candidate is based on arrays of vertically aligned CNTs. CNT arrays can achieve good thermal performance, but the reliability had not previously been investigated experimentally. A thorough investigation of the reliability of CNT array TIM revealed that reliability is not guaranteed, but requires careful matching between CNT array height, bonding method and substrate configuration.
Furthermore, we developed a new joule self-heating chemical vapor deposition (CVD) method for the synthesis of double-sided CNT arrays on thin foils, which can be used both as TIM or as supercapacitor electrodes. Double-sided arrays are challenging with conventional CNT array synthesis methods, but the Joule heating CVD method allows for rapid, scalable and uniform synthesis of large area double-sided arrays. Finally, this method was used to study the effect of heat treatment of CNT arrays on graphite. The heat treatment serves to simultaneously improve the CNT crystallinity, eliminate catalyst residues, and form a seamless connection between CNT arrays and graphite.
Solder
Thermal interface materials
Carbon nanotubes
Author
Josef Hansson
Chalmers, Microtechnology and Nanoscience (MC2), Electronics Material and Systems
Novel nanostructured thermal interface materials: a review
International Materials Reviews,; Vol. 63(2018)p. 22-45
Review article
Fabrication and characterization of a carbon fiber solder composite thermal interface material
2017 IMAPS Nordic Conference on Microelectronics Packaging (NordPac),; (2017)p. 97-100
Paper in proceeding
Effect of Fiber Concentration on Mechanical and Thermal Properties of a Solder Matrix Fiber Composite Thermal Interface Material
IEEE Transactions on Components, Packaging and Manufacturing Technology,; Vol. 9(2019)p. 1045-1053
Journal article
Synthesis of a Graphene Carbon Nanotube Hybrid Film by Joule Self-heating CVD for Thermal Applications
Proceedings - Electronic Components and Technology Conference,; (2018)
Paper in proceeding
Reliability investigation of a carbon nanotube array thermal interface material
Energies,; Vol. 12(2019)
Journal article
Bipolar electrochemical capacitors using double-sided carbon nanotubes on graphite electrodes
Journal of Power Sources,; Vol. 451(2020)
Journal article
Effects of high temperature treatment of carbon nanotube arrays on graphite: increased crystallinity, anchoring and inter-tube bonding
Nanotechnology,; Vol. 31(2020)
Journal article
Degradation of Carbon Nanotube Array Thermal Interface Materials through Thermal Aging: Effects of Bonding, Array Height, and Catalyst Oxidation
ACS Applied Materials & Interfaces,; Vol. 13(2021)p. 30992-31000
Journal article
From the heat of a phone in your hand to the noise of an overburdened computer fan, most people are familiar with the effects of heat generation in electronics. This heat generation limits the performance improvements that our society has become used to, and for that reason improved thermal management is critically important for many applications. Within thermal management, thermal interfaces are one of the most significant bottlenecks. A thermal interface is the boundary between two surfaces in contact with each other, such as with a silicon chip and a heat sink, and since surfaces are never completely flat, the contact between the surfaces is limited, which severely restricts the heat flow.
Thermal interface materials (TIMs) are materials placed between these surfaces, and conform and fill out the gaps, allowing heat to flow better. A good TIM should have high thermal conductivity, conform well to the surfaces, and not break during the device lifetime. Within my research, I have been focused on achieving TIMs with these properties, based on two different concepts: metal composites and carbon nanotubes (CNTs).
Metals have generally high thermal conductivity, and by melting them, such as we do with solders, they can conform to surfaces very well but have high stiffness which can cause reliability issues. Part of this thesis describes our development a composite material consisting of polymer fibers in a metal matrix, combining the mechanical properties of polymers with the thermal properties of metals.
The other route is to utilize vertically aligned CNTs to bridge the interface. Carbon nanotubes have both high thermal conductivity and flexibility to conform to surfaces. In principle, these should also be soft enough for good reliability, but this had never been experimentally tested, until now. Our studies show that reliable CNT array TIM are possible, but only under certain conditions.
Thermal management is not the only application for CNTs, and during my research, I found that my work could have applications within more areas. Within this thesis, I have also utilized CNTs as supercapacitor electrodes, developed a new fabrication method and studied what happens to CNT arrays when they get really hot. In total, this work represents an explanation of the possibilities of using both metal composites and carbon nanotubes for thermal interfaces, and beyond.
Thermal interface materials (TIMs) are materials placed between these surfaces, and conform and fill out the gaps, allowing heat to flow better. A good TIM should have high thermal conductivity, conform well to the surfaces, and not break during the device lifetime. Within my research, I have been focused on achieving TIMs with these properties, based on two different concepts: metal composites and carbon nanotubes (CNTs).
Metals have generally high thermal conductivity, and by melting them, such as we do with solders, they can conform to surfaces very well but have high stiffness which can cause reliability issues. Part of this thesis describes our development a composite material consisting of polymer fibers in a metal matrix, combining the mechanical properties of polymers with the thermal properties of metals.
The other route is to utilize vertically aligned CNTs to bridge the interface. Carbon nanotubes have both high thermal conductivity and flexibility to conform to surfaces. In principle, these should also be soft enough for good reliability, but this had never been experimentally tested, until now. Our studies show that reliable CNT array TIM are possible, but only under certain conditions.
Thermal management is not the only application for CNTs, and during my research, I found that my work could have applications within more areas. Within this thesis, I have also utilized CNTs as supercapacitor electrodes, developed a new fabrication method and studied what happens to CNT arrays when they get really hot. In total, this work represents an explanation of the possibilities of using both metal composites and carbon nanotubes for thermal interfaces, and beyond.
Areas of Advance
Production
Subject Categories
Nano Technology
Condensed Matter Physics
Infrastructure
Nanofabrication Laboratory
ISBN
978-91-7905-278-2
Doktorsavhandlingar vid Chalmers tekniska högskola. Ny serie: 4745
Publisher
Chalmers