Advertisement

Metallurgical and Materials Transactions B

, Volume 41, Issue 4, pp 824–832 | Cite as

A Review on Die Attach Materials for SiC-Based High-Temperature Power Devices

  • Hui Shun Chin
  • Kuan Yew CheongEmail author
  • Ahmad Badri Ismail
Article

Abstract

Recently, high-temperature power devices have become a popular discussion topic because of their various potential applications in the automotive, down-hole oil and gas industries for well logging, aircraft, space exploration, nuclear environments, and radars. Devices for these applications are fabricated on silicon carbide-based semiconductor material. For these devices to perform effectively, an appropriate die attach material with specific requirements must be selected and employed correctly. This article presents a review of this topic, with a focus on the die attach materials operating at temperatures higher than 623 K (350 °C). Future challenges and prospects related to high-temperature die attach materials also are proposed at the end of this article.

Keywords

Solder Alloy Power Device Conductive Adhesive Microelectronic Packaging Silver Particle Size 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Microelectronic packaging provides interconnection, physical protection. and mechanical support for semiconductor devices for them to function in a specific condition.[1] Therefore, die attach materials, which connect the die and device to the rest of the system, play an important role to ensure the entire system works consistently. It is commonly accepted that both tin-based solder alloys (leaded [SnPb] and lead-free solder alloys [SnAgCu]) and conductive adhesives are used widely as die attach materials because of their ease of processing at temperatures below 573 K (300 °C).[2] However, with higher temperature applications (>623 K (350 °C)) in automotive, down-hole oil and gas industries for well logging, aircraft, space exploration, nuclear environment, and radars,[3] these solder alloys and conductive adhesives cannot meet their stringent requirements, and one of the obvious drawbacks is their low melting and operating temperatures.

To operate at a higher temperature, silicon carbide (SiC) has been identified as one of the potential semiconductor wafers for the future generation of power devices. This is because SiC intrinsically is a wide band gap semiconductor with superb electronic, physical, chemical, and mechanical properties that enable it to withstand high temperature, power, and voltage.[4] In fact, when using this semiconductor, a dramatic reduction in volume and weight of a power module can be achieved, as the SiC die can be shrunk, and the cooling system can be simplified.[5] Research also indicates that SiC-based power devices possess an excellent switching characteristic and stable functionality even at temperatures greater than 623 K (350 °C).[6, 7, 8] However, less attention has been paid to the development of technology and material that is related to microelectronic packaging, particularly die attach material, operating at the aforementioned temperature.

A few alternative die attach materials that can withstand high temperatures have been reported.[2,9, 10, 11, 12, 13, 14, 15, 16] Silver is a promising material, as it possesses a high melting temperature (1233 K (960 °C)), high electrical and thermal conductivity, and good reliability, but it is less preferable as a die attach material in industry because of its extremely high sintering temperature. There are two strategies to lower the sintering temperature, namely inducing pseudo-hydrostatic pressure[9] and reducing silver particle size into the nanometer range and sintering without applying any pressure.[2] Other potential die attach materials being reported are off-eutectic gold-based alloy,[10, 11, 12, 13] bismuth-based,[14] liquid-based,[15] and silver-indium-based die attach materials.[16] The progress and development of these materials employed as die attach materials in SiC-based high-temperature power devices are being reviewed in this article. The review focuses on the material properties, processing techniques, applications, and manufacturability.

Overview of Microelectronic Packaging and Die Attach Material

Generally, microelectronic packaging plays an important role in a semiconductor device aiming to connect, protect, and support the semiconductor die to a substrate.[2] Hierarchically (Figure 1), semiconductor devices in a wafer level are being interconnected (0-level interconnection), and these devices then are connected externally to a substrate via first-level interconnection technology. Subsequently, a higher level of interconnection (board-level and system-level interconnection) needs to be established for the devices to be connected externally. Microelectronic packaging technology cannot take on this role alone without considering the importance of a die attach material. The aim of this material is to attach semiconductor die to a substrate. To use a material for die attachment in a power device, ideally it should demonstrate the following properties[17,18]: (1) high thermal conductivity, (2) low coefficient of thermal expansion (CTE) between the die and the substrate, (3) good wettability and adhesion to the die and to the substrate, (4) good mechanical properties with stress relaxation behavior, (5) good fatigue resistance, (6) good corrosion resistance, (7) good reworkability, (8) high electrical conductivity and (9) good reliability.
Fig. 1

Schematic diagram of a package hierarchy

Current die attach Materials and their Challenges for High-Temperature Applications

Currently, there are two commonly used die attach materials for interconnecting power devices, which are tin-based solder alloys (leaded [SnPb] and lead-free solder alloys [SnAgCu]), and conductive adhesives. The processing temperature of the former category of die attach material is below 523 K (250 °C), which is the limitation of Si-based devices.[19,20] The maximum processing temperature of the latter category of die attach material is even lower (near 473 K (200 °C)). In comparison, the electrical and thermal properties of a conductive adhesive, such as metal-filled conductive epoxy, are poorer than in a typical tin-based solder alloy.[21] Regardless of the type of die attach materials, the low melting temperatures of those materials is considered to be a serious issue, thereby preventing them from being employed as a workable solution for high-temperature applications (>623 K (350 °C)). When these die attach materials are used at high temperatures, the tendency to form intermetallic phases is high, which may reduce the reliability of the die attach material. In addition, materials of the substrate and encapsulant may degrade considerably, as the processing temperature increases with the potential of revealing a mismatch of thermo-mechanical properties and, thus, reducing their reliability.[7]

Application of die Attach Materials in High-Temperature Power Devices

Numerous electronic applications (and, indirectly, die attach materials) require operation at higher temperatures. “High temperature” is defined as a high ambient temperature that differs greatly from a typical standard operating temperature. For a typical power electronic device, because of high local power dissipation at a specific point, an extremely high ambient temperature can be generated.[22] Listed below are a few applications (down-hole oil and gas industry for well logging, aircrafts, automotive, and space exploration) that currently require die attach materials operating at temperatures above 473 K (200 °C). Other applications include flame monitoring, high-power consumer electronics, and industrial systems. The temperature requirements of these applications are summarized in Figure 2.
Fig. 2

Indicative temperature ranges.[22]

Oil and Gas Industry for Well Logging

Currently, the oil and gas industry is one of the largest users of high-temperature electronics and is certainly the most innovative, leading some exciting developments of not only components but also systems. Fairly sophisticated sensors and data-acquisition electronic systems are used for well logging, particularly the area around drilling heads in oil, gas, and geothermal wells. Sensors used to monitor parameters such as temperature, pressure, flow rate, density, and chemical composition are installed in the vicinity of the drilling heads. Control of these parameters is needed to optimize the productivity of the well while preventing issues of well blockage. The temperature requirement is varied according to the depth of the well, ranging up to 873 K (600 °C) for the deepest geothermal wells but more typically about 523 K (250 °C) for the existing oil or gas wells, as being reported by Traeger and Lysne.[23] Another example of using high-temperature electronic components in this industry is the electric down-hole gas compressor.[24] The compressor is designed to increase the production of gas wells by putting a compressor near the gas reservoir. For this application, ambient temperature is expected to reach 423 K (150 °C), whereas for deep oil wells, the temperature may reach up to 498 K (225 °C), with the system having a 5-year lifetime.[25]

The continuous effort of developing high-temperature electronics for this application is driven by the need for accuracy and resolution improvement, excellent reliability, and reduced cooling requirements while increasing the operating temperature, which enables drilling a deeper well.[22] Even most logging activities involve short-term monitoring of the well; permanent on-the-spot monitoring is a long-term goal in which a significant increase in the lifetime of the components is one of the challenges.

Aircraft

Aircraft is a real and fairly substantial market for the 21st century to use high-temperature electronic components. The driving force for this industry comes from a better fuel efficiency aiming to increase the performance and reliability of the aircrafts. Because of the complexity in wiring and piping of commercial aircrafts, which make use of hydraulic, pneumatic, and electric actuators, more electronic devices have been used to replace the traditional systems. For example, the More Open Electrical Technology (MOET) project, which is funded by FP6—a European research program[26] is focused on “validating scalable electrical networks that can go up to 1 MW by considering new voltages and advanced concepts that include system transformation of future air, actuation and electrical systems as part of all electrical solutions.” To achieve sufficient efficiency, electrical actuators should be driven through power electronic converters in a distributed fashion[27] in which converters are placed as close as possible to the actuators that are exposed to a high-temperature condition. For example, some of them are operating near the jet engine, with ambient temperatures ranging from 218 K to 498 K (–55 °C to 225 °C).[27] Therefore, these systems need to be very reliable with a long operating life (10 to 30 years) and withstand frequent deep thermal cycles (several takeoffs and landings per day).

Automotive

The automotive industry also needs a very large volume of high-temperature electronics, with the main objective aimed at reducing emissions to comply with new legislation and to improve fuel efficiency. This problem can be solved by using hybrid electric vehicles, which combine a gasoline engine with an electric motor that is equipped with built-in inverters to control the flow of electric current in the vehicle. These inverters consist of power modules with an array of power devices that operate at temperature ~873 K (600 °C).[28] Figure 3 shows the temperatures of devices in an automobile and its related systems.
Fig. 3

Automotive temperatures and related systems (DaimlerChysler, Stuttgart, Germany).[31]

Reducing emissions also relies on more precise control of the valve timing as well as primary sensing of cylindrical pressure and camshaft positions. Consequently, sensors and signal-conditioning electronics have been built into the vehicles to improve the precision of the measurements.[29] Advanced cars now contain around 100 sensors, including sensors for engine speed and angular position, automatic brake system, power steering, and engine condition monitoring. Smart sensors and associated electronics with operation temperatures up to 523 K (250 °C) are a vital part of the new developments required to meet these goals. Some sensors even are required to withstand higher operating temperatures; for example, braking system sensors are working under an ambient temperature of 573 K (300 °C), and combustion sensors or sensors at the inlet of a catalytic converter are operating up to 1273 K (1000 °C).[3,30]

Space Exploration

Space exploration is a niche market, but there are some challenging goals to overcome.[32,33] For example, on Venus, the surface temperature can reach 733 K to 753 K (460 °C to 480 °C), whereas on Jupiter, the temperature increases with depth (and pressure), and a few hundred kilometers down, the ambient temperature reaches 673 K (400 °C) (and 100 bars), with a very aggressive atmosphere (winds around 200 m/s−1 and a hydrogen-rich chemical composition). To add to this, thermal cycling is also an issue, as ambient temperatures can be as low as 133 K (–140 °C) during the night. Therefore, the proper selection of microelectronic package materials, particularly die attach materials, is extremely crucial to determine good performance of these electronic devices.

Silicon Carbide (SiC) vs Silicon (Si) as A Semiconductor Wafer

Si is the most common semiconductor material that is used in many mainstream electronic devices.[22] Its dominant use can be attributed to the fact that its material properties are well understood. However, this device is limited to its relatively low operating temperature (<523 K (250 °C)).[34] For instance, above this temperature, Si-based semiconductor devices operate inefficiently, especially when the high temperature is combined with high power, high frequency, and a high-radiation environment.[35] This is a result of its intrinsic properties, such as narrow band gap (1.12 eV) and low thermal conductivity (1.5 Wcm−1 K−1).[36]

To overcome this issue, SiC-based semiconductor devices, with their excellent electrical and physical properties (Table I), are a potential candidate to replace Si as a semiconductor wafer for high-temperature electronic applications.[37] One of its significant attributes is its wide band gap property, which offers various advantages, namely higher temperature and chemical stability, higher thermal conductivity, and breakdown field. Therefore, electronic devices fabricated from SiC can be operated efficiently at temperatures beyond 873 K (600 °C).[38] The high breakdown electric field of SiC (3.2 MVcm−1), which is approximately 10 times higher than that of Si (0.3 MVcm−1), allows for a thinner conduction region (for constant doping). This may lead to a very low specific conduction resistance. The thermal conductivity (3.7 Wcm−1K−1) of SiC is also high (which is higher than copper) and enables a higher power density with a higher power-per-unit area to be handled in comparison with Si (1.5 Wcm−1K−1). The saturated electron velocity of SiC (2 × 107 cm/s−1) is twice that of Si (1 × 107 cm/s−1), which translates to a faster operating speed. Bonding energy between Si and C in SiC provides a high mechanical strength and radiation resistance when compared with Si-Si bonds in Si.[44,45]
Table I

Comparison Between Si and SiC Regarding Semiconductor Properties[39, 40, 41, 42, 43]

Property

Si

SiC

Band gap (eV)

1.1

3.26

Dielectric constant

11.8

9.66

Breakdown electric field (MVcm−1)

0.3

3.2

Thermal conductivity (Wcm−1 K−1)

1.5

3.7

Saturated electron velocity (cm/s−1)

1 × 107

2 × 107

Electron mobility (cm2V−1 s−1)

1400

1000

Hole mobility (cm2V−1 s−1)

600

115

Melting point (K)

1690

3100

Physical stability

Good

Excellent

Process maturity

Very high

Medium

High-Temperature die attach Materials for SiC-Based Power Devices

There are four categories of high-temperature die attach materials for SiC-based power devices that have been reported, namely off-eutectic gold-based,[13] bismuth-based,[14] liquid-based,[15] and silver-based die attach materials.[2,9,16] Properties and synthesis of the die attach material as well as application in terms of manufacturability of the die attach materials are reviewed in the subsequent sections.

Off-Eutectic Gold-Based Die Attach Material

The off-eutectic gold-tin (Au-Sn) die attach alloy has been used to attach an SiC die on an Ni/Au-plated direct-bonded copper substrate,[46, 47, 48, 49, 50] which can operate at temperatures as high as 673 K (400 °C).[13] The use of off-eutectic rather than eutectic Au as the die attach alloy is explained as follows. Eutectic Au-Sn shows excellent performance at high temperatures but is limited to 553 K (280 °C), a superior corrosion resistance, a high electrical and thermal conductivity, and a fluxless soldering process. However, because of its higher stiffness, most of the thermomechanic stresses easily are transferred to the die. As a result, breakage of the die commonly is observed if a large die size is used. For this type of die attach material to be employed in SiC-based power devices, a limited die size must be used. Therefore, it underuses superb properties of the SiC material, as the power device has a restricted wafer-level design. Another limitation of the eutectic Au-Sn die attach alloy is its low melting temperature (553 K (280 °C)). To increase the melting temperature and to lower the processing temperature, an off-eutectic[51] composition of the system has be used. By slightly increasing the concentration of Au, the melting point can be shifted to a higher temperature. Hence, a die attach material that can withstand an operating temperature as high as ~973 K (700 °C) can be created. However, in terms of manufacturability, this type of alloy requires a relatively higher processing temperature. Therefore, investment in the soldering equipment is expected to be high.

Bismuth-Based Die Attach Material

Bismuth (Bi), which has a melting point of 543 K (270 °C), is a candidate for use as a die attach material, but it is brittle and has poor bonding strength.[52] To overcome these limitations, a newly produced Bi-based composite die attach material has been developed.[14] This composite consists of 75Cu23Al2Mn particles that are prepared by the gas atomizing method and are coated with electroless Ni plating aimed at improving the wettability of the particles with Bi matrix. The coated particles are mixed with a molten Bi, and the entire system is subjected to a heat treatment at 773 K (500 °C) for 5 minutes. The mechanical strength of this die attach material is almost twice that of pure Bi.[52] It has been reported that the CTE of this die attach material is well matched with the CTE of the die and substrate.[52] Hence, this material demonstrates excellent reliability when it is subjected to a thermal cycling test over 473 K (200 °C) at 233/523 K (–40/250 °C).[52] As a result, it is a potential candidate to be used as a die-attach material for power device dies and insulated substrates by using a CTE-matched structure. One of the major concerns of using this type of alloy as a high-temperature die attach material is the processing temperature. Temperatures greater than 773 K (500 °C) must be used to melt the material to obtain a reliable bonding between the die and substrate. Because of the high processing temperature, this alloy is not suitable to be adopted in industry.

Liquid-Based Die Attach Material

Liquid-based die attach material is a combination of a metallic alloy and a polymeric underfill or glob top (Figure 4).[53, 54, 55, 56] Polymer underfill or glob top are used to retain the mechanical integrity of the die attach material. In liquid or molten form, the die attach material does not undergo strain hardening. Hence, joints of the die attach material are not subjected to fatigue failure. However, this type of die attach material fails easily when subjected to substantial electric bias. Under this condition, accelerated dendritic growth and phase changes can result. In addition, the mechanical strength of this die attach material is also questionable, as it cannot withstand even reasonable amounts of vibration. To overcome these challenges, a diffusion barrier layer is applied in between the die attach material and the metal-based substrate.
Fig. 4

Schematic diagram of liquid die attach material assembly.[15]

The eutectic indium-tin (52In48Sn) alloy, with a melting temperature as low as 391 K (118 °C), has been reported to be used as a liquid die attach material by Mannan and Clode.[15] A thermal cycling behavior of this system has been demonstrated.[57] Investigation has been carried out to understand the reaction between the liquid die attach material (52In48Sn) and various metal barrier layers.(15) A barrier layer based on niobium has shown good adhesion to die attach material.[58] The integrity of the system is good even after 200 thermal cycles (from 253 K to 453 K (–20 °C to 180 °C) for a dwell time of 30 minutes).

Silver-Based Die Attach Material

Silver-Indium Die Attach Material

High-temperature die attach material based on the alloying of silver (Ag) and indium (In) (79.75 weight pct Ag and 20.25 weight pct In) at a low-processing temperature (479 K (206 °C)) has been reported.[16,59] However, this alloy melts at a temperature >973 K (700 °C). Based on the reported work, pure silver and indium thin films were vacuum-deposited separately on a device/substrate. Then, the device/substrate assembly was heated to a temperature exceeding the melting point of indium (430 K (157 °C)) to allow the formation of Ag-In die attach material through the atomic interdiffusion method. A two-step fluxless joining process was applied, whereby the first step created the joint, and the second step was an annealing process to increase the melting temperature of the joint. By making the die attach materials with higher silver content, noneutectic reactions between Ag and In may increase the melting temperature beyond 973 K (700 °C). According to the Ag-In binary phase diagram as shown in Figure 5,[60] an Ag-In solid solution can be made by melting at temperatures greater than 673 K (400 °C). Thus, this type of die attach material can be processed at a lower temperature. A nondestructive scanning acoustic microscopy technique has been used to examine the qualities of the produced joints before and after the thermal annealing at 418 K (145 °C) for 26 hours.[61,62] Results show that the joints produced indeed were virtually void-free. Three intermetallic phases, namely, AgIn2, Ag2In, and AuIn2, were identified in the cross-section of the joint before annealing through scanning electron microscopy and energy dispersive X-ray spectroscopy evaluation. After annealing, the Ag elements in the joint spread out uniformly to form a silver-rich Ag-In die attach material (Ag2In) with an Ag composition of 62 to 68 wt pct. A nearly pure Ag phase (85 to 92 wt pct) was revealed in the region where the Ag was first deposited on the substrate.[61,62] The Ag-In die attach material also exhibited a higher joint reliability when tested under a high-temperature environment by performing a series of reliability tests at 773 K (500 °C) up to 1000 hours. This fluxless Ag-In alloy undeniably provides a potential choice as a high-temperature die attach material for high-temperature power devices.
Fig. 5

Silver-indium phase diagram.[60]

Silver Paste Film-Sintering

Silver is used widely in microelectronic packages as conductor lines and interconnections between substrate layers because of its well-known high electrical and thermal conductivity as well as its limited fatigue property.[63, 64, 65] With its high melting temperature (1233 K (960 °C)), silver is suitable for high-temperature packaging applications. But because of its extremely high processing temperature, it is not suitable for application in the electronic packaging industry. To lower the processing temperature, silver is made into a powder form. Unlike solder reflow, the sintering of silver paste will densify without melting.[2] Two strategies have been reported to lower the sintering temperature of silver, namely using a quasihydrostatic pressure to increase the sintering driving force[9] and employing nanometer-scale silver particles embedded in a paste without using any external pressure.[2]

Strategy 1: Use of Quasi-Hydrostatic Pressure

In this strategy, an external pressure is applied to compensate for the lowering of the sintering temperature. The pressure may help densify the material by eliminating some fraction of pores through compression or deformation and, meanwhile, also may help increase the contact area between the silver particles.[9] Typically, a hot press is used to apply pressure on a fixture to create a quasihydrostatic external pressure on the die/substrate assembly that is mounted in the fixture. To avoid cracking the assembly, a silicone-filled fixture is used. Commercial silver paste with silver particle sizes from 1 to 3 μms is stencil printed on the substrates to form layers with 40- to 100-μm thickness.[9] Then, the printed substrates are placed in a vacuum to remove air being trapped in the film. Next, the printed silver layers are preheated (<573 K (300 °C)) for a few hours to eliminate any organic components, solvents, and binders in the paste. Finally, the whole system is pressurized at 513 K (240 °C) for a few minutes. The sintered silver exhibits superior electrical and thermal conductivity, and adhesion strength is significantly better than those of the eutectic SnPb die attach material.[9] However, the application of this type of die attach material in industry is limited, as it requires external pressure and the process is also time consuming. The use of external pressure has a limitation because it tends to complicate automation in manufacturing process and, thus, increases the processing cost.[66] In addition, applying pressure may cause physical damage to the device, such as cracking. Therefore, an alternative strategy to lower the processing temperature needs to be developed. This strategy is reviewed in the following section.

Strategy 2: Use of Nano-Structured Material
An alternative method to lower the sintering temperature of silver is by reducing its particle size.[66] As the particle size is reduced, additional surface energy is gained, which theoretically, may increase the driving force needed for the sintering process, and thus, the sintering temperature can be lowered. Nanometer-scale silver can be obtained either commercially or be prepared by using the Carey Lea method as shown in Figure 6, which was adapted by Frens and Overbeek.[67] The silver paste has been prepared by mixing the powder with an organic binder system (dispersant, binder, and thinner). The nanoparticle silver paste has been stencil-printed on a metalized substrate then heated for a short duration to volatize the binder and solvents. Finally, it is sintered at 553 K (280 °C) on a belt reflow oven for a few minutes. This strategy showed that the die attachment can be carried out at temperatures below 573 K (300 °C) without the application of pressure.[66] Furthermore, die attach tests revealed very strong bonding between the die and the substrate. In comparison with others, this type of die attach material is the most feasible and has a higher potential to be adopted and employed in industry. However, one consideration that must be taken care of is the application of this material and technique on a large die size. As the die size increases, uniform coverage of the paste may become an issue. Releasing volatile compounds from a large area may create a problem if the process is not uniform. Problems such as cracking and reliability may prevent this strategy to be used in large die size. Table II summarizes all of the discussed die attach materials.
Fig. 6

Schematic diagram for the preparation of nanoscale silver paste.[67]

Table II

Summary of Various Die Attach Materials with Its Methods, Properties, and Manufacturability

Die Attach Materials

Die Attach Material Production Methods

Die Attach Material Properties

Manufacturability

Off-eutectic gold-based die attach material[13]

dissolution[51]

- high-temperature performance[46, 47, 48, 49, 50]

- suitable only for small die applications because of stiffness property[13]

- superior corrosion resistance[46, 47, 48, 49, 50]

- high electrical and thermal conductivity[46, 47, 48, 49, 50]

- fluxless soldering[46, 47, 48, 49, 50]

Bismuth-based high-temperature die attach material[52]

dispersing and mixing[52]

- excellent reliability property[52]

- higher processing temperature (773 K (500 °C))

- higher mechanical strength[52]

- must fabricate within CTE matched structure to gain desired properties

Liquid-based die attach material[15]

not Reported

- strong adhesion strength[58]

available in solder paste form[57]

- remains stable during temperature cycling[57]

performs well in thermal cycling[57]

- high-temperature storage when the die attach material is molten[57]

- no fatigue failure[15]

Silver-indium die attach material[16]

atomic interdiffusion[59]

- higher joint reliability[59]

- ease of processing with low-processing temperature[16]

- void free property[59]

- thermal stress reduced[61]

Silver paste film-sintering[2,9]

nanosilver purchased commercially

- superior electrical and thermal conductivity[63]

- complicated manufacturing process[66]

- limited fatigue property[64]

- higher processing cost[66]

- good adhesion strength[9]

- pressure may physically damage devices[9]

- void-free property[9]

- high joint reliability[9]

nanosilver produced by Carey Lea method[67]

- high electrical and thermal conductivity[63]

- ease of manufacturing process, which without application of pressure[66]

- limited fatigue property[64]

- lowering processing cost[66]

- strong bonding even in smaller die[2]

Future Prospect of High-Temperature Applications

In view of the current status, there are market demands of high-temperature electronic applications (e.g., automobiles [HEV-hybrid electric vehicle], oil and gas well logging, aircraft, space exploration, nuclear environments, and radars). For this reason, SiC-based devices that can operate effectively at very high temperatures have been a seriously researched topic of late. In line with this, microelectronic packaging technology, especially die attach material, must be developed. A few challenges pertain to the development of die attach material for high-temperature applications, namely the production of a die attach material that can withstand a high operation temperature but with a relatively low processing temperature, the ease of adopting the material in mainstream microelectronic packaging technology, the lowering of material and processing cost, and the improvement of reliability in die attach material. More efforts are still needed to create a high-temperature die attach material that may perform comparably with conventional Si packaging technology in terms of functionality, integrity, and reliability.

Conclusion

Various types of high-temperature die attach materials, which have been used in SiC-based high-temperature power devices, have been reviewed. The development and progress of off-eutectic gold-based, bismuth-based, liquid-based, and silver-based die attach materials systematically have been compared and reviewed. Of these, sintered Ag nanopowder is considered to be a promising die attach material to join the SiC die on a metalized substrate at temperatures below 573 K (300 °C). However, further work must be done to seek out a more optimal die attach material that can prolong the working hours in high-temperature applications.

Notes

Acknowledgments

H.S.C. would like to express her appreciation to the USM RU-PRGS grant for the scholarship and financial support on this project. K.Y.C. would like to acknowledge financial support given by USM Short Term Grant (6039038).

References

  1. 1.
    T. Wang, X. Chen, G.Q. Lu, and G.Y. Lei: J. Electron. Mater., 2007, vol. 36, pp. 1333-40.CrossRefADSGoogle Scholar
  2. 2.
    G.Q. Lu, J.N. Calata, Z.Y. Zhang, and J.G. Bai: 1st ACM Hardcopy Document Processing Workshop Proc., Washington, DC, ACM, 2004, pp. 42–46.Google Scholar
  3. 3.
    J.W. Klein: 1995 URSI International Symposium on Signals, Systems and Electronics (ISSSE’95) Proc., San Francisco, CA, 1995, pp. 157–62.Google Scholar
  4. 4.
    J.G. Bai, J. Yin, Z.Y. Zhang, G.Q. Lu, and J.D. van Wyk: IEEE Trans. Adv. Packag., 2007, vol. 30, pp. 506-10.CrossRefGoogle Scholar
  5. 5.
    H. Iwasaki, K. Hisano, and T. Takamatsu: Trans. Jpn. Soc. Mech. Eng., 2003, vol. 6, pp. 251-52.Google Scholar
  6. 6.
    W. Wondrak, R. Held, E. Niemann, and U. Schmid: IEEE Trans. Ind. Electron., 2001, vol. 48, pp. 307-08.CrossRefGoogle Scholar
  7. 7.
    T. Braun, K.F. Becker, J.P. Sommer, T. Loher, K. Schottenloher, R. Kohl, R. Pufall, M. Koch, V. Bader, R. Aschenbrenner, and H. Reichl: 55th Electronic Components & Technology Conf. (ECTC) Proc., Orlando, FL, 2005, pp. 376–83.Google Scholar
  8. 8.
    R. Singh, S.H. Ryu, D.C. Capell, and J.W. Palmour: IEEE Trans. Electron. Dev., 2003, vol. 50, pp. 774-84.CrossRefADSGoogle Scholar
  9. 9.
    Z.Y. Zhang and G.Q. Lu: IEEE Trans. Electron. Packag. Manufact., 2002, vol. 23, pp. 279-83.CrossRefMathSciNetGoogle Scholar
  10. 10.
    H. Schwarzbauer: IEEE Trans. Ind. Appl., 1991, vol. 27, pp. 93-95.CrossRefGoogle Scholar
  11. 11.
    S. Klaka and R. Sittig: 6th International Symposium Power Semiconductor Devices and ICs (ISPCD’94) Proc., 1994, pp. 259–64.Google Scholar
  12. 12.
    U. Scheuermann and P. Wiedl: Workshop Metal Ceramic Composites Functional Appl., Vienna, Austria, 1997, pp. 181-1920.Google Scholar
  13. 13.
    R.W. Johnson, C. Wang, Y. Liu, and J.D. Scofield: IEEE Trans. Electron. Packag. Manufact., 2007, vol. 30, pp. 182-91.CrossRefGoogle Scholar
  14. 14.
    Y. Yamada, Y. Takakub, Y. Yagia, Y. Nishibea, I. Ohnumab, Y. Sutouc, R. Kainumab, and K. Ishidab: Microelectronic Reliability, 2006, vol. 46, pp. 1932-37.CrossRefGoogle Scholar
  15. 15.
    S.H. Mannan, and M.P. Clode: IEEE Trans. Adv. Packag., 2004, vol. 27, pp. 508-14.CrossRefGoogle Scholar
  16. 16.
    R.W. Chuang and C.C. Lee: IEEE Trans. Compon. Packag. Technol., 2002, vol. 25, pp. 453-58.CrossRefGoogle Scholar
  17. 17.
    E. Savrun: 1 st IEEE International Conf. on Sensors Proc., 2002, pp. 1139–43.Google Scholar
  18. 18.
    K. Suganuma, S.J. Kim, and K.S. Kim: JOM, 2009, vol. 61, pp. 64-71.CrossRefADSGoogle Scholar
  19. 19.
    J.G. Bai, Z.Z. Zhang, J.N. Calata, and G.Q. Lu: IEEE Trans. Compon. Packag. Technol., 2006, vol. 29, pp. 589-93.CrossRefGoogle Scholar
  20. 20.
    J.G. Bai and G.Q. Lu: IEEE Trans. Device Mater. Rel., 2006, vol. 6, pp. 436-41.CrossRefGoogle Scholar
  21. 21.
    R. Kisiel and Z. Szczepanski: Microelectronic Reliability, 2009, vol. 49, pp. 627-29.CrossRefGoogle Scholar
  22. 22.
    R. Sharp: IEE Colloquium, 1999, vol. 1999, pp. 1/1–1/4.MathSciNetGoogle Scholar
  23. 23.
    R.K. Traeger and P.C. Lysne: IEEE Trans. Nuc. Sci., 1988, vol. 35, pp. 852-54.CrossRefADSGoogle Scholar
  24. 24.
    Corac Group: “Downhole gas compression,” product brochure, Corac Group PLC, Uxbridge, UK, 2005.Google Scholar
  25. 25.
    B.W. Ohme, M.R. Larson, J. Riekels, S. Schlesinger, K. Vignarajah, and E.M. Nance: High Temperature Electronics Conference (HiTEC) Proc., Santa Fe, NM, 2006, p. 9.Google Scholar
  26. 26.
    MOET: M. Consortium: Project year 1 executive summary report, MOET, FP6, Epernay, France, 2007.Google Scholar
  27. 27.
    I.S. Mehdi, A.E. Brockschmidt, and K. Karimi: High Temperature Electronics Conference (HiTEC) Proc., Santa Fe, NM, 2006.Google Scholar
  28. 28.
    M. Ishiko: R&D Review of Toyata CRDL, 2004, vol. 39, pp. 1-4.Google Scholar
  29. 29.
    D.S. Eddy and D.R. Sparks: Proc. IEEE, 1998, vol. 86, pp. 1747–55.Google Scholar
  30. 30.
    R.W. Johnson, J.L. Evans, P. Jacobsen, J.R. Thompson, and M. Christopher: IEEE Trans. Electron. Packag. Manufact., 2004, vol. 27, pp. 164-76.CrossRefGoogle Scholar
  31. 31.
    R. Thompson: SMTA/CAVE Workshop Harsh Environment Electronics Proc., Dearborn, MI, 2003, pp. 24–25.Google Scholar
  32. 32.
    E. Kolawa, M. Mojarradi, and T. Balint: High Temperature Electronics Conference (HiTEC) Proc., Santa Fe, NM, 2006.Google Scholar
  33. 33.
    C. Buttay, D. Planson, B. Allard, D. Bergogne, P. Bevilacqua, C. Joubert, M. Lazar, C. Martin, H. Morel, D. Tournier, and C. Raynaud: VIII Microtechnology and Thermal Problems in Electronics (Microtherm) Conf. Proc., Łódź, Poland, 2009, pp. 315–21.Google Scholar
  34. 34.
    J.B. Baliga: IEEE Electron. Dev. Lett., 1989, vol. 10, pp. 455–57.CrossRefADSGoogle Scholar
  35. 35.
    J.B. Casady and R.W. Johnson: Solid-State Electronics, 1996, vol. 39, pp. 1409-22.CrossRefADSGoogle Scholar
  36. 36.
    M.R. Werner and W.R. Fahrner: IEEE Trans. Industrial Electronics, 2001, vol. 48, pp. 249-57.CrossRefGoogle Scholar
  37. 37.
    J.A. Cooper and A. Agarwal: Proc. IEEE, 2002, vol. 90, pp. 956–68.Google Scholar
  38. 38.
    M. Willander, M. Friesel, Q. Wahab, and B. Straumal: J. Mater. Sci. Mater. Electron., 2006, vol. 17, pp. 1-25.CrossRefGoogle Scholar
  39. 39.
    M. Mehregany, C.A. Zorman, N. Rajan, and C.H. Wu: Proc. IEEE, 1998, vol. 86, pp. 1594–609.Google Scholar
  40. 40.
    M.C. Driver, R.H. Hopkins, C.D. Brandt, D.L. Barret, A.A. Burk, R.C. Clarke, G.W. Eldridge, H.M. Hobgood, J.P. McHugh, P.G. McMullin, R.R. Siergiej, and S. Sriram: 15 th GaAs IC Symposium Proc., San Diego, CA, 1993, pp. 19–21.Google Scholar
  41. 41.
    J.C. Zolper: Solid-State Electron, 1998, vol. 42, pp. 2153-56.CrossRefADSGoogle Scholar
  42. 42.
    V.E. Chelnokov and A.L. Syrkin: Mater. Sci. Eng. B, 1997, vol. 46, pp. 248-53.CrossRefGoogle Scholar
  43. 43.
    J. Camassel, S. Contreras, and J.L. Robert: Comptes Rendus de l’Acade’mie des Sciences Series IV Physics, 2000, vol. 1, pp. 5-21.CrossRefADSGoogle Scholar
  44. 44.
    Y.M. Tairov and V.F. Tsvetkov: J. Cryst. Growth. 1978, vol. 43, pp. 209-12.CrossRefADSGoogle Scholar
  45. 45.
    D.L. Barrett, R.G. Seidensticker, W. Gaida, R.H. Hopkins, and W.J. Choyke: J. Cryst. Growth, 1991, vol. 109, pp. 17-23.CrossRefADSGoogle Scholar
  46. 46.
    X.S. Liu, M.H. Hu, H.K. Nguyen, C.G. Caneau, and M.H. Rasmussen: IEEE Trans. Adv. Packag. 2004, vol. 27, pp. 640-46.CrossRefGoogle Scholar
  47. 47.
    W. Pittroff, G. Erbert, G. Beister, F. Bugge, A. Klein, A. Knauer, J. Maege, P. Ressel, J. Sebastian, R. Staske, and G. Traenkle: IEEE Trans. Adv. Packag. 2001, vol. 24, pp. 434-41.CrossRefGoogle Scholar
  48. 48.
    S.A. Merritt, K. Mobarhan, R. Whaley, S. Fox, M. Dagenais, A.R. Mickelson, N.R. Basavanhally, and Y.C. Lee: Optoelectronic Packaging, Wiley, New York, NY, 1997, pp. 59.Google Scholar
  49. 49.
    G.S. Matijasevic, C.C. Lee, and C.Y. Wang: Thin Solid Films, 1993, vol. 223, pp. 276-87.CrossRefADSGoogle Scholar
  50. 50.
    M. Nishiguchi, N. Goto, and H. Nishizawa: IEEE Trans. Comp. Hybrids. Manufact. Techno., 1991, vol. 14, pp. 523-28.CrossRefGoogle Scholar
  51. 51.
    F.P. McCluskey, M. Dash, Z. Wang, and D. Huff: Microelectronics Reliability, 2006, vol. 46, pp. 1910-14.CrossRefGoogle Scholar
  52. 52.
    V.V. Zenin, V.N. Belyaev, Yu. E. Segal, and A.A. Kolbenkov: Russian Microelectronic, 2003, vol. 32, pp. 247-56.CrossRefGoogle Scholar
  53. 53.
    R. Keelere: EP&P, 1989, vol. 29, pp. 14-18.Google Scholar
  54. 54.
    J.V. Ellerson, J. Funari, and J.A. Varcoe: Patent U.S. 5 553 769, 1996.Google Scholar
  55. 55.
    T.P. Dolbear, C.A. MacKay, and R.D. Nelson: Patent U.S. 5 170 930, 1992.Google Scholar
  56. 56.
    J.W. Smith: Patent U.S. 5 808 874, 1998.Google Scholar
  57. 57.
    I. Shohji, S. Fujiwara, S. Kiyono, and K.F. Kobayashi: Scripta Mater., 1999, vol. 40, pp. 815-20.CrossRefGoogle Scholar
  58. 58.
    H.S. Chen, D.D. Bacon, H.H. Law, G.W. Kammlott, and T.C. Wu: J. Mater. Res., 1995, vol. 10, pp. 2870-74.CrossRefADSGoogle Scholar
  59. 59.
    T. Quintero, P.O.A. Oberc, and F.P. McCluskey: High Temperature Electronics Conference (HiTEC) Proc., Albuquerque, NM, 2008, pp. 207–12.Google Scholar
  60. 60.
    T.B. Massalski: Binary Alloy Phase Diagrams, 2nd ed., ASM International, Materials Park, OH, 1990, pp. 381-83.Google Scholar
  61. 61.
    V. Simic and Z. Marinkovic: Thin Solid Films, 1979, vol. 61, pp. 149-60.CrossRefADSGoogle Scholar
  62. 62.
    R. Roy and S.K. Sen: Thin Solid Films, 1992, vol. 197, pp. 303-18.CrossRefADSGoogle Scholar
  63. 63.
    S. Rane, V. Puri, and D. Amalnerkar: J. Mater. Sci. Mater. Electron., 2000, vol. 11, pp. 67-674.CrossRefGoogle Scholar
  64. 64.
    C.R. Chang and J.H. Jean: J. Amer. Ceram. Soc., 1998, vol. 81, pp. 2805–14.CrossRefGoogle Scholar
  65. 65.
    R. Manepalli, F. Stepaniak, S.A. Bidstrup-Allen, and P.A. Kohl: IEEE Trans. Adv. Packag., 1999, vol. 22, pp. 4-8.CrossRefGoogle Scholar
  66. 66.
    Z.Y. Zhang, J.N. Calata, J.G. Bai, and G.Q. Lu: 5 th Global Symposium on Materials Processing and Manufacturing Division (MPMD) Proc. Charlotte, NC, 2004, pp. 129–35.Google Scholar
  67. 67.
    G. Frens and J.Th.G. Overbeek: Kolloid Z. Z. Polym., 1969, vol. 233, pp. 922–29.CrossRefGoogle Scholar

Copyright information

© THE MINERALS, METALS & MATERIALS SOCIETY and ASM INTERNATIONAL 2010

Authors and Affiliations

  • Hui Shun Chin
    • 1
  • Kuan Yew Cheong
    • 1
    Email author
  • Ahmad Badri Ismail
    • 1
  1. 1.Energy Efficient & Sustainable Semiconductor Research Group, School of Materials & Mineral Resources EngineeringUniversiti Sains MalaysiaNibong TebalMalaysia

Personalised recommendations