How to Transform the Chemical Industry -- One Reaction at a Time | Miguel A. Modestino | TED

I was born and raised in Caracas, Venezuela,

back in the '80s and '90s.

Venezuela is a beautiful country,

rich with natural treasures from beaches to rainforests.

But one thing that people don't quite realize

is that we also have really tall mountains.

These are part of the Andes,

and they stretch all the way down from Chile and Argentina.

And they're also home to some of the very few tropical glaciers

in the world.

Back in 1910, when the first map was drawn,

Venezuela used to have five tropical glaciers,

which we used to call the five white eagles.

Today, unfortunately, there's only one left,

and it covers an area which is smaller than Central Park in New York City.

These glaciers have been disappearing because of the use of a resource

that we extract not too far from them:

oil.

As a Venezuelan, my relationship with oil is complicated.

On the one hand, oil is the resource that helped my country develop,

as well as human civilization.

On the other hand,

it's a resource that slowly but surely fuels an existential threat.

Climate change.

We now know that to avert the worst impacts of climate change,

we're going to have to decarbonize everything that we do

and everything that surrounds us.

And to do that within the next three decades.

These include things like transitioning from oil, natural gas

to clean sources of electricity;

electrifying all of our vehicles;

rethinking the way that we make our buildings and cities.

The good news is that technologies to do some of these things already exist.

And legislation to frame and direct this effort

is starting to be put in place.

The bad news is that even if we do all of these things,

it will still not be enough.

There are still large sectors of our economy

that will need to be decarbonized

for which we don't have the technology yet.

These include the things like manufacturing processes

that are used to make everything that surrounds us,

from the cement and steel that holds our buildings together

to the chemicals and materials that we use to make the goods

that we use in our daily lives.

Things like the chair that you're sitting in,

your shoes, the cars that you use,

and all the devices that we use every day.

The chemical manufacturing industry alone accounts for 10 percent

of the energy consumption.

And 20 percent of the carbon emissions from industry.

These emissions arise from the use of fossil fuel combustion

to generate the heat that is required to drive processes inside of our old

and inefficient chemical plants.

One way to start fixing the issue will be to use electricity

instead of heat from fossil fuels

to drive the processes inside of these plants.

Sounds easy enough, right?

Just unplug the natural gas pipelines, plug in the electrical wire,

and make sure that you source electricity from clean sources like solar,

wind or nuclear.

Unfortunately, in reality, it's much harder than that.

We're going to have to come up with new chemical reactions

that can source their energy directly from electricity.

These are called electrochemical reactions.

And it’s what my research group at New York University is focusing on:

figuring out electrochemical paths

that can create direct links between clean electricity and molecules.

In electrochemical reactions,

heat is replaced by electricity as an energy source.

This allows us to operate reactors

and carry out reactions at room temperature

with minimal energy losses.

But the problem is

that for electrochemical reactions to become viable,

they will need to be more efficient and cost competitive

than the heat-powered ones that we use today.

That means that electrochemical reactors

will need to transform raw materials into products

very efficiently, with minimal waste and very selectively.

This will then in turn incentivize chemical manufacturers

to switch to electrochemical production processes

instead of deploying new fossil fuel chemical plants.

In our research, we started by looking at a chemical reaction

that is used to make one of the most important polymers in the world:

Nylon 66.

More than 5 million tons of all kinds of nylons

are produced every year.

And out of those, two million tons are of Nylon 66.

And these are used to make things like the textile fibers

that we use in our winter jackets

or the plastic car parts

that help make our vehicles lightweight and fuel-efficient.

Through our research,

we managed to identify electrochemical reaction mixtures

with the right additives and voltages

to be able to transform raw materials into nylon precursors very selectively.

We achieved selectivity of more than 80 percent,

a value that is on par with commercial reactors.

But we didn't stop there.

While most electrochemical reactions operate

with a constant stream of electrical currents,

we knew that if we control the interactions

between the electricity input and molecules,

we could achieve a higher performance.

To do that, we decided to start exploring different electrical pulses

in the search for the right pulse sequence

that will deliver electrons at the same rate

that molecules could react.

Since there were so many pulses that we needed to test,

we decided to then use a few number of results

from different pulses to train machine-learning models

that then help us identify the appropriate pulse sequence

for this reaction.

This was the first time

that artificial intelligence was used to optimize an electrochemical reaction,

and the pulse sequence that we discovered

allows us to increase by 30 percent

the production rate of these reactors.

These large improvements in efficiency,

selectivity and production rates are very important

because they will ensure

that electrochemical reactions are competitive

with fossil-fuel-powered ones which are already highly optimized.

But hey, nylon is only one molecule

in a network of tens of thousands of chemical products,

many of which you see in this diagram,

and many of which are a common occurrence in your daily lives.

Just think of the plastics that are used to make your computers and phones,

or the pane that covers the walls around us,

or the medicines that help us live a healthy life.

We're going to have to decarbonize it all within the next 30 years.

That means that we're going to have to discover,

develop and deploy chemical processes much faster

than we did in the past century.

To that end, my group at NYU is combining artificial intelligence

with autonomous research tools

to run hundreds to thousands of experiments per day.

In this way, we hope to decrease the time from idea to discovery

by more than 100 times.

Beyond nylon,

we're also looking into production processes

to manufacture other chemicals

like ethylene and propylene.

These are the main precursors to most of the plastic in the world.

Through electrochemistry,

we have managed to electrify already the most energy intensive steps

on the production process of these chemicals:

a set of processes that alone account for a very substantial amount of energy

in the chemical manufacturing industry.

We're also rethinking the way that we make our chemicals

and the chemicals supply chain,

by developing new processes

that can use and transform food waste into the chemicals

that we make today from oil.

This can be done by extracting molecules from waste streams

and using them as precursors to things like plastics,

additives, cosmetics and pharmaceuticals,

and many more.

And in this way,

enhancing sustainability in chemical manufacture even further.

The chemical industry is central to our modern economy

and to our ways of life.

Decarbonizing it will take us several decades.

This will include things like

retrofitting our chemical plants so they can use electricity directly,

capturing and sequestering CO2 as it comes out of the gas flue stacks,

and more importantly,

developing new chemical processes that are more efficient

and can source electricity directly from clean sources.

By accelerating research in electrochemical manufacturing,

my team aims to develop some of these reactions

and in this way help to transform the chemical industry

one reaction at a time.

When we get there,

we will have been able to erase the carbon footprint

of the chemical industry.

And while we won't be able to bring our tropical glaciers back,

we will have helped to develop a sustainable path

for the future of our planet.

Thank you.

(Applause)