As components, packages and systems continue to shrink in size, the heat generated in these dense electronic systems can be quite large and can lead to a significant rise in temperatures that in turn can cause device and system-level failures.

Heat has always been an issue for system designers, but only recently has the problem become so severe that thermal management solutions can no longer be introduced as an afterthought. Thermal management must be looked at from the beginning of the design process in order to avoid causing severe problems at the system level. This is an important consideration because at the system level, the thermal operating space is more limited (i.e. the temperatures that can be tolerated) and any solution employed at the system is likely to be more expensive than one implemented at the chip level.

A new approach to thermal management involves embedding thermal management functionality deep inside an electronic component at the source of the heat using thin-film thermoelectric devices.

Temperature Control with Thermoelectrics
TECs (thermoelectric coolers) are solid state heat pumps that operate using the Peltier effect. When an electric current is driven through a circuit containing two dissimilar materials, heat is absorbed at one junction (the cold side) and released at the other junction (the hot side). The design of mostPeltier devices requires the use of both an n-type semiconductor and a p-type semiconductor. Since heat naturally flows down a temperature gradient from hot to cold, a TEC’s ability to move heat from cold to hot in a solid-state fashion is unique. By reversing the polarity of the applied DC current, heating is also possible. This property is especially useful for applications requiring both cooling and heating for precise temperature control.

Conventional TEC solutions (sometimes referred to as “bulk” TECs) have been used for years to control the temperature of electronics. However, as the size and power density requirements of new applications are changing, conventional bulk thermoelectric technology has not kept pace. In some instances, designers choose to place the cooling device outside the package if it is too large to be placed inside. Cooling the device by cooling the entire package is at best an inefficient method for thermal management and often leads to over sizing of the TEC that requires more drive power and more waste heat in the system.

Advantages of Thin-Film Thermoelectrics
A relatively new development is the manufacture of thin-film TECs that use semiconductor processing techniques to create a nano-structured thin film used for the P and N legs. Thermoelectric thin-film TECs are typically 5 µm to 20 µm thick, versus 200 µm for the thinnest pellets used in bulk TECs, resulting in several differences. Heat flux, which is inversely proportional to the thickness of the thermoelectric material, is 20+ times greater than bulk TECs. Thin-film TECs pump a maximum heat flux of 100 W/cm² to 400 W/cm² versus less than 10 W/cm² for typical bulk TECs. Thin-film TECs can operate in a high coefficient of performance (COP) regime and still pump a reasonably high heat flux (20 W/cm² to 40 W/cm²). COP is a measure of efficiency defined as cooling power divided by input power.

Figure 1. Thin-film TEC pumps as much heat as a conventional (upper right) module 20x its size.

Depending on the design, thin-film TECs may have thermal response times as low as milliseconds enabling very rapid cooling and heating to maintain precise temperature control. They are known to have higher heat pumping capability than standard bulk TECs, but for temperature control applications, the superior switching speed of the devices may ultimately prove to be their most valuable asset.

Thermoelectric Cooling
The most basic representation of the operational space for a thermoelectric cooling device is a load line shown in Figure 2.

Figure 2. The load line.

The load line represents the ?T and power pumped conditions possible for a given TEC drive current. At the maximum drive current for the module, the load line is generated from two key parameters: 1) the maximum power the device can pump, Qmax; and, 2) the maximum temperature difference that the device can sustain between its top and bottom plates, ?Tmax.

The load line defines the operational space for TECs and is the best and most usual way to illustrate TEC performance.

System Level Considerations
The TEC, being an active thermal device, creates a thermal inversion that dramatically changes the thermal profile inside the package. Figure 3 shows a comparison of the thermal profile through the cross section of the module in two cases, a) with no TEC, or in other words, a passive solution only, and b) with a TEC actively cooling the junction. It can be clearly seen that the introduction of the TEC provides a substantial net cooling benefit. 

Figure 3. Temperature profiles through the cross section of a package from the junction to the case (?JC), and case to ambient (?CA) without (a) and with (b) a TEC. The temperature inversion created b

The heat that is pumped by the device and the additional heat created by the TEC in the course of pumping that heat will need to be rejected into the system. Since the performance of the module can be improved by providing a good thermal path for the rejected heat, it is beneficial to provide high thermally conductive pathways. For small packages, this is typically accomplished through the electrical connections themselves, and depending on the operating characteristics, this level of thermal management might be sufficient. For packages with higher heat densities, thermally conductive feed-throughs or posts may be needed to remove the heat.

Thin-film thermoelectrics offer the promise for improved product performance and increased product lifetime due to their speed and ability to pump large amounts of heat. These performance advantages will have an impact in electronics, industrial process control and other applications where precise temperature control is essential.