Crowd-sourcing a solar-powered society with World Community Grid

Share this post:

Imagine if we could embed solar power generators into objects and materials around us? And what if the solar cells in these products were as cheap and easy to mass-manufacture as plastic bags, bringing the possibility of affordable solar energy to impoverished areas? These are the dreams of Harvard University’s ‘Clean Energy Project’, which has achieved huge gains in the field of organic solar science and gained White House recognition along the way. 

Alán Aspuru-Guzik, Professor of Chemistry at Harvard University

In June 2013, Harvard University’s ‘Clean Energy Project’ (CEP) launched a free, open-source database for researchers into renewable energy. The database has been cataloguing the suitability of 2.3 million compounds for use in solar energy generation – organic, carbon-based compounds which are typically cheaper to manufacture and more adaptable than the silicon-based versions commonly installed in solar panels.

This research would once have taken decades. But thanks to the ‘World Community Grid’ concept, the CEP has already scored big – crowd-sourcing millions of calculations from volunteers and bringing to light thousands of potential new solar materials.

The promise of organic solar cells
The aim of the database is to accelerate the move from silicon-based solar materials to organic ones. Silicon cells convert about 15% of sunlight into electricity whereas most organic compounds have only had a 4-5% efficiency. But silicon cells are expensive to produce, whereas carbon-based materials can be mass-manufactured. “They’re also lighter and more flexible,” explains Alán Aspuru-Guzik, Professor of Chemistry at Harvard, who is leading the effort. “They could possibly be thin and light enough to be applied to portable devices, and could even be painted or sprayed onto roofs, windows and walls. A handful of successful compounds are needed to make that happen, and that’s where we come in.”

Already, approximately 1,000 of the new structures characterized in the research have shown a potential for converting at least 11% of captured sunlight into electricity, and about 35,000 may achieve an efficiency of 10% or more.

This was made possible by IBM’s World Community Grid, which joins together the unused computing power of more than 2.3 million ordinary devices from over 600,000 volunteers, and makes it available for humanitarian research.

Bringing supercomputing power to the world
Anyone can volunteer for World Community Grid by downloading a small program onto their Windows, Mac, Linux or Android device. When they are performing lightweight tasks, their device requests data on a project, performs computations on it, sends the results back and asks the server for a new piece of work.

World Community Grid creates a distributed system with computational power equivalent to some of the most powerful supercomputers, allowing researchers to accelerate their experiments: Because the work is split into small pieces that can be processed simultaneously, research time is reduced from years to months.

That has helped CEP researchers find so many solar candidates so quickly. “Before this, researchers knew of a handful of carbon-based solar materials that could operate at efficiencies comparable to silicon,” says Professor Aspuru-Guzik. “To create a sustainable future based around renewable energy, we need to analyze vastly more than that. But each new compound takes, on average, a year to experimentally synthesize and characterize. By using the World Community Grid, we’ve been able to look at tens of thousands of compounds every day.”

So far, IBM World Community Grid has provided the CEP project with the equivalent of 19,000 years of computing time as part of the world’s most extensive quantum chemical investigation, with over 90,000 World Community Grid volunteers contributing.

At the forefront of tomorrow’s economy
The work has now been recognized at the highest levels. In June 2013, the project received official praise from The White House’s Office of Science and Technology Policy for its contribution to materials science.

But, says Professor Aspuru-Guzik, this is just the first step in the process. “The compounds we released were generated in such a way that they are harder to synthesize than we ideally want. The next generations of compounds to go on the grid improve on that, and we will release more generations of computed materials and more advanced analysis procedures as we go. This is still an exciting work in progress.”