Bottoms up - personal energy solutions

Materials World magazine
1 Dec 2010
A conceptual approach to bottom up energy harvesting with individual chips harvesting energy from a variety of sources, sharing this energy across single and multiple circuit boards

Professor Robert Dorey, Head of the Microsystems and Nanotechnology Centre at Cranfield University, UK, considers the future of personal energy solutions and the role that materials science will play.

The way energy is used is changing. We live in an age where electricity is available on demand from a socket in the wall. However, this situation is neither universally applicable nor necessarily required for the future. Just as fixed line telecommunication was largely bypassed in favour of mobile communication in many developing countries and remote communities, so centralised power generation and fixed power transmission could be superseded in favour of a distributed energy generation model – a local energy network (LEN), for example.

To an extent we are already witnessing this transformation with increased use of local power generation systems, such as the installation of photovoltaic (PV) solar cells in homes and businesses.

Borrowing from the world of nanotechnology and the expression ‘there’s plenty of room at the bottom’, a similar view could be adopted in the world of energy.

The terms ‘top down’ and ‘bottom up’ are often used to describe the two ways of creating structures where material is selectively removed or added respectively. These terms could equally well be used to describe two different approaches to energy gen eration. Viewed in this way, a domestic PV cell would be considered a ‘top down’ energy source with centralised power generation and distribution within a single household. So what would the features of a truly ‘bottom up’ energy solution be? Perhaps a more appropriate description would be to consider the case of energy generation at the individual component level, with multiple components sharing excess energy on a single circuit board, multiple boards sharing energy in a product, and perhaps products themselves sharing energy (see image, top left).

Flexible power supply

With flexible and scalable energy harvesting platforms, the degree of energy capture could be tailored to suit different situations. In some cases, the power harvested may just serve the immediate local needs, such as a single wireless sensor node or, in the extreme, become a pure generation system with little or no need for power within the actual device. In this way, the approach could dovetail with existing ‘local centralised’ energy solutions, such as PV cells, and form a critical part of the energy network as a whole. The form that such LEN energy harvesting systems take would be governed by the available sources of energy, the ways in which such sources can be harnessed, and the manufacturing techniques available to deliver solutions cost effectively.

If solar harvesting is considered as a local centralised solution and not suitable for embedded or enclosed systems, the most abundant and available energy sources are mechanical movement and heat. At the small-length scales envisaged for local energy networks, functional materials can provide an effective route to tap into these energy sources. Piezoelectric, thermoelectric and pyroelectric materials are all capable of transforming mechanical or thermal energy into electrical energy.

Compared to PV technology, with energy densities in the region of 2-20mWcm-2, energy densities achievable from piezoelectric, pyroelectric and thermoelectric systems only extend into the lower end of this range, with reported values ranging from a few μWcm-2 to a few mWcm-2. Despite these lower figures there is still a great potential for these technologies, thanks to the advantages of high duty cycles, the potential for miniaturisation and the robustness of the technology.

Dynamic energy source

Making use of the charge generated by the action of a stress on the material, piezoelectric materials have been extensively investigated for energy collecting applications. Many different designs of piezoelectric harvester have emerged as a result of attempts to harness either the high force, low frequency sources (human, vehicular and structural movement), or the high frequency, low force sources of movement (sound and machine vibration) that surround us.

In harnessing low frequency, high force sources, mechanical systems are used to enhance the response by transferring stress to the piezoelectric element in such a way that the charge generated is increased. These mechanical systems can take the form of an amplifier that concentrates the load on the piezoelectric element. Alternatively, as used in the Thunder system, the piezoelectric element can be placed under compressive pre-stress so that it is able to sustain a much greater deflection, and hence stress, without failure. In so doing, the level of energy generated is also increased.

In high frequency, low force cases, the main challenge is reducing the stiffness of the piezoelectric element so that it is able to respond quickly to the rapidly changing low loads. A common approach is to use a cantilever design, where the application of the load causes a deflection and a subsequent stress in the piezoelectric film on the cantilever. These devices work most effectively if they can be used at, or near, their resonant frequency, where the power harvested can be one or two orders of magnitude greater. The disadvantage is that the range of frequencies over which the cantilever resonates is typically quite narrow and so can only access a part of the available vibration energy.

Scanning electron microscope image of micro cantilever and schematic representation of two approaches to increasing the band width of resonating cantilever harvesters

Feeling the heat

The pyroelectric effect, which is weaker than piezoelectric, allows the harvesting of energy from temperature fluctuations. While the pyroelectric effect on its own is capable of producing only μW levels of power, it can be enhanced by coupling with the piezoelectric effect giving a pseudo-pyroelectric effect.

In this pseudo-pyroelectric route, temperature fluctuations lead to a thermally induced differential strain – as in a bimetallic strip – leading to a piezoelectricallygenerated electrical charge. As with vibration energy collection, the amount of power generated depends on both the amplitude of oscillation (proportional to temperature and differences in thermal expansion coefficient), as well as the frequency of oscillation.

While the performance of the pseudo-pyroelectric device is enhanced relative to the pure pyroelectric device, it is still lower than that of the vibration harvester due to the limited oscillation rate. In general, it is not the device’s ability to respond that limits performance, rather it is the rate at which thermal fluctuations occur in themajority of environments.

Despite this there may be niche applications where rapid fluctuations in temperature occur and where the pseudo-pyroelectric effect may have an application – in engines, for example.

Take the temperature

Thermoelectic, unlike pyroelectric, energy is generated when a temperature difference across a sample causes an electrical current due to the behaviour of p- and n- type semiconducting materials within the structure. One of the biggest challenges of these devices is related to thermal stability, toxicity and the cost of the functional materials found in these devices. Examples of materials used include Bi-Te, Pb-Te and Co-Sb compounds. Despite these drawbacks, power densities of up to 10mWcm-2K-1, comparable to PV cells, have been reported.

Materials research is at the heart of the evolution of all of these harvesting devices and will play a pivotal role in enhancing capability by enabling higher performance, lower cost and lower environmental impact, through elimination of toxic elements, and by reducing the amounts of materials used.

At the same time new processing techniques will allow nano- and microscale structuring of materials to yield improvements to device response and efficiencies.

While it is unlikely that a local energy network, as described above, will completely replace centralised electrical generation and fixed line distribution – particularly for high power demand applications – there are a number of applications where the LEN approach will find applications. This is particularly relevant where power is required in mobile or remote locations.

With increasing energy harvesting capabilities, reducing power requirements of electronics and reduced manufacturing costs, there is great potential for local energy networks.  

Pyroelectric currents produced by lead zirconate titanate due to temperature fluctuations of 40K at 0.2Hz (oil and water baths)

Further information

Microsystems and Nanotechnology Centre, Cranfield University, Bedfordshire, MK43 0AL, UK. Tel: +44 (0)1234 750111. Email: Website: