19Feb 2019

Some time in August of 2013, hackers breached Yahoo! servers and stole private account information for up to 3 billion users. Verizon Communications received a $350 million discount in the price of its acquisition of Yahoo! in 2017, exemplifying the staggering costs of one single encounter with cyber risk.

The concept of risk and risk management is not new. In 1688, Edward Lloyd set up what would become today Lloyd’s of London to contain the emerging risks of the new and growing maritime trans-Atlantic trade. Since then, the business world has worked diligently to contain such risk in everything from food to the Internet.

Actually, almost everything. One such modern risk that remains inadequately addressed is battery safety, specifically the safety of lithium-ion batteries that are so ubiquitous. To be fair, industry has recognized long time ago the safety hazards surrounding the lithium-ion battery. Battery fires in the early 2000s caused expensive recalls. But they were largely treated as one-off events. These were times when the annual volume of batteries was a few hundred millions. These fires were not treated as an on-going risk. They were seen as failures in manufacturing that could be eliminated by improvements in factories or designs.

Today’s battery shipments have skyrocketed to billions of units and counting. Even a minuscule chance of battery fire becomes a real problem when multiplied by the sheer volume of batteries. Battery failures are an ongoing risk that needs to be contained.

Estimates place the risk of battery fire in the range of a few to tens of parts-per-million (or ppm). One ppm means that for every one million units shipped, there is a risk that one of them will catch fire. It does not mean that one *will* catch fire. It just means that statistically speaking, the probability of a fire is one in a million. Now that seems like a small number. You might tell a precious love that they are “one in a million.” In an industry that ships two billion smartphones annually, that translates to several thousand battery fires annually! Not acceptable! We need to bring this figure down by a factor of 100 or 1,000.

Edward Lloyd’s business was possible because it had its underpinnings in the mathematical advances of probability pioneered by Blaise Pascal and Pierre de Fermat early in the 17th century. In that same vein it is possible to make great improvements in battery safety because it leverages the advances in computation of the past 50 years.

Every smartphone is a miracle device. It contains a processor that is infinitely more powerful than the computer that landed Apollo 11 on the moon. It also contains sophisticated electronics that can measure minute voltages and currents, and in turn it is very telling of the chemical reactions inside the battery. Merge it all with intelligent software, and we can now predict what the battery’s health will be in the future.

But why can’t we just manufacture the perfect battery that will never catch fire? Simply put, it is prohibitively expensive. Consider this: nearly every person with a smartphone is also an amateur photographer. Despite the fact their camera lens is optically deficient, software allows them to take incredible photographs.

The same goes for batteries. Manufacturing batteries in large volumes means that some will have defects. That’s just the balance between quality and cost when it comes to battery manufacturing in large scale. To make matters more challenging, every person will use or abuse their battery in unpredictable ways. It becomes essential to catch and screen these few bad batteries in the field before they become a hazard. Naturally, this is not meant to supersede good manufacturing practices, but rather to complement them in our quest to reduce battery fires to zero.

So how does it work? I talked in the past about electrochemical impedance spectroscopy (EIS). It is a workhorse test instrument in battery laboratories around the world. It is capable of measuring the chemical processes that are taking place inside the battery. Now imagine if you had such a similar tool inside your device. With some expertise, you can now start making smart decisions about your battery. This is not a new concept; a similar concept, for instance, allows glucose measuring devices to save the lives of millions of diabetics.

It’s high time we get serious about battery health and safety. Let’s address this risk before it escalates. The spread between device capabilities and battery threats is only growing — let’s get smart and manage potential incidents before they blow up into something bigger.