An Interview with Gradient CEO Vincent Romanin

PHA: Was there a moment that you can remember where you kind of looked around and thought to yourself, “Oh shit; we need to do something right now and I have some expertise that that can be of use”?

Romanin: It wasn't quite that punctuated. It was a series of moments. I got my PhD [from University of California, Berkeley], and then I worked in solar thermal for a while. I then ended up at this early-stage research lab, again, doing work on solar thermal. The lab was called Otherlab. They get a lot of early-stage government grants, usually for really ambitious, long-term, truly difficult problems at the intersection of energy and hardware. That was the space I was in. It was kind of my background as well. It was a lot of creative people who did exciting things and spent a lot of time brainstorming or building prototypes on the weekend.

I'm going on a tiny bit of a tangent, but one of the things that the lab did and is still doing was a grant for the Department of Energy. Otherlab said, “If you're going to invest in changing our energy infrastructure, you have to understand it and you don't really have a detailed understanding.” So, they built this maddeningly detailed Sankey diagram that shows energy flows in the US. So now you can know how much natural gas we use versus how much coal. I can tell you how much energy we use driving to church on Sundays as a country. The idea of this effort was, if we understand more, we can say, “Okay, if we switch this industry from chemicals to electricity or from coal to this other thing, what's the effect it's going to have?” A lot of interesting things came out of this project.

And so, it wasn't so punctuated where I looked at this graph and said, “HVAC's big.” It was the kind of a broader effect where we realized that the emissions in HVAC are as big of a challenge as electrifying transportation and as cleaning up the grid from a total emission standpoint. The reason it's not obvious—it actually doesn't even show up here as much as it should—is because the natural gas that we use to heat our homes occasionally leaks. It leaks at a rate of somewhere between 2% and 9%, and the higher global warming potential [GWP] of natural gas when it's methane than when it is burned means that that leakage rate roughly doubles the carbon footprint of natural gas.

That was the first, “Oh shit.” Natural gas at the time was marketed as cleaner than coal. That's true if you burn 100% of it. If you burn 97% of it, it becomes worse than coal—and it's somewhere between 94% and 98% (a recent estimate placed the figure between 96% and 97%). It also depends on the GWP horizon used—20 years or 100 years—but you get the point. At the time, people were also starting to understand that it’s hard to measure the rate at which we were leaking natural gas. The natural gas in your home has an odorant in it. Natural gas out of the ground does not smell, so it’s very hard to detect when you're leaking it.

The second “oh shit” is due to refrigerants. The standard refrigerant we use in HVAC today is R-410a, which is 2000 times stronger on global warming than CO2. The leaked refrigerant out of our air conditioners has a massive impact on climate, and not many people were talking about it. That's on top of the fact that it’s projected that there's going to be another 4 billion ACs on the planet by 2050, which is a tripling of today's number. So, we were like, “All right. Refrigerants are bad. They could potentially get way worse.” And at the time we started the company, it was after the Kigali Amendment to the Montreal Protocol had been passed, which is phasing out HFCs [hydrofluorocarbons], which are the high GWP refrigerants.

So, it's solved, right? Not quite. The replacement for HFCs that the industry was talking about has the potential to be just as bad for the environment—not for global warming, but for other things. And we've done this before. CFCs [chlorofluorocarbons] were putting a hole in the ozone layer, so we replace them with HFCs. The result is global warming. So it’s kind of out of the frying pan and into the fire, and we were on track to do it again. That was the second “oh shit.” The solution is that we have to electrify heating, and we have to use better refrigerants.

The third thing is less of an “oh shit” kind of one moment, but we realized these devices need to be smart for the grid to handle all of the electricity.

So that was why we got into HVAC. These three things need to happen to decarbonize and they're not happening, but it was also because people need more HVAC to protect themselves against climate change. It was, it was really like the intersection of adaptation and mitigation that got me excited. I thought we could do something better for the planet and help people.

PHA: So, all this led you to start Gradient? When was that?

Romanin: We incorporated on April 7th, 2017.

PHA: That’s very precise.

Romanin: <Laughter> Yes. So, because Otherlab was good at government grants and because this is a really long-term, hard problem to solve, we started with government grants, and we wrote probably 10 to 15 failed grants. This may sound bad, but I think it's a success story, right? You write a grant, and they say, “No, we're not going to fund it for these reasons.” You then take the feedback and iterate.

After those initial failures, we won four in a row, and I think we built a coalition between potential industry partners, the engineering startup community, and a couple governments to say like, “Hey, we gotta figure out how to solve this.”

The first grant we got was actually a fellowship just for me to start the company through a program that was at the time called Cyclotron Road. They've rebranded, it's now called Activate. It is for PhD scientists to take what they're doing and figure out how to turn it into a company. With their support, we received some other research grants to fund their research. A year after that, we got our first venture check from Matt Rogers, who was the former co-founder of Nest. Of the people who understand the intersection between climate and HVAC and startups, he's one of very few, especially at the time.

PHA: When did you actually have your first prototype completed?

Romanin: We had some incredibly ambitious and ugly prototypes before we even started as a company. So, Otherlab fronted a lot of the resources and very small amounts of money for myself and a couple other people to write these grants and build some very early prototypes. I was very much an engineer at the time, so I said, “All right, we need to make the components cheaper and better.”

Once we incorporated, it switched from “we want to solve this problem” to “let's talk to the people who are going to buy the thing.” So, the first thing we did when I was full-time, when I was at Activate, was talk to customers, and you learn very quickly that everyone buys the cheapest system. You look at data across countries, people pay for the cheapest HVAC.

So we said, “All right, how do we make things more efficient? Let's make them cheaper!” Very engineering answer. We then talk to more customers and realize the major cost is not in the hardware or components. It's in the installation. That is the barrier. If you buy an efficient, high-end heat pump mini split, the labor is still most of the ticket price.

We learned one other thing in that first year of talking to customers. We would talk to the person who buys HVAC for a giant skyscraper or for a single-family home when it's being built, and the person who buys the HVAC is usually the installer. They care about having familiar hardware. They care about something that they know how to install and service. They care that they are buying it from a company who can send them a replacement part in three years. And I was like, “Well, this isn't a startup sale.” They're never going to trust us as a startup.

But then we talked to people who used window ACs, and they metaphorically grabbed us by the shoulders and said, “I hate this thing. I will do anything to get rid of it. It's the bane of my existence, but it's necessary.” Okay, this sounds like a pain point. That is something that we as a startup can solve.

Another interesting thing around this time is me and a couple other early founders were reading this nerdy book called Thermal Delight in Architecture. It's written by an MIT [Massachusetts Institute of Technology] architect, and she makes this, in hindsight, amazingly obvious point that I had never thought of, which is that humans get joy and inspiration from things that they see and feel. There's art that we like looking at, but there's also a lot of inspiration you get from your thermal environment like a cool breeze on a beach, or when you're sitting in front of a warm fire when you're camping and the radiative heat feels good. So, for her, making all buildings 72 degrees at all times is like an interior decorator painting everything blue. She's like, our HVAC systems do very little. They do an absolutely bare minimum job of making us comfortable. They kind of even suck at that, and they do very little to create exciting experiences with the thermal environment. You shouldn't always be the same temperature. Your body wants a different temperature when you're about to go to bed, right? Like, if it’s summer, your home should be slightly warmer. You should be experiencing summer. We really just want our systems to provide more than just baseline comfort.

And so we realized that we should not just focus on decarbonizing things. We should make products that actually provide better experiences for people, because that's the only way it's going to scale fast enough to have an environmental impact in the first place. And so we said, all right, HVAC, the comfort of your home is a very emotional part of your home. And HVAC devices do nothing to address that. Right? It should be exciting, like that new fuzzy blanket is exciting. And we needed that inspiration to get to this first product design.

PHA: How did you improve on your initial prototype to provide that kind of experience to people?

Romanin: After we learned those things, we kind of had the tech stack. We had the problem and the thesis. We had very small amounts of initial funding, but we knew we needed to make this exciting for people and we had this user pain point that we wanted to solve. We got connected through Matt Rogers to the former head of design at Nest, a guy named Rocky Jacob. He consulted for us, and he was the person who reminded us that nobody wants to look at their AC. They want to look out their window.

So, we built this saddlebag with a shelf on top. That shelf on top of the indoor unit is something that, in hindsight, seems so obvious. This is why you get a good industrial designer. And then we put a plant on it and took a picture, and we're like, “This is awesome, right?” The focus is the plant—not the actual HVAC. That image with the plant has done a better job of pitching the company than I ever could in words. It just becomes so obvious when you look at it.

So, we recognized that this concept works, but had to make sure that it's easy to install. We made the decision that anything that required a licensed technician was the big barrier that is going to add costs. There's workforce shortages in HVAC technicians and electricians and building contractors. So we said, “Let's make the installation cheap and fast, so there's no need for extra permits skilled labor with specific licenses. That was a big engineering effort.

But there's more that we want to do. We eventually want to get to gradient temperature. The level of air movement is really important, obviously, like the convective cooling of moving air changes your comfort. Long term, we want to get into air quality management, so you don't have to decide to bring in fresh air. The same thing with natural light control. Your system should know when you want natural light and when you don't from an energy perspective or a comfort perspective.

PHA: How much energy does a Gradient unit use in a year when compared to a conventional heat pump system or when compared to an older HVAC system?

Romanin: Our flagship model is well suited for moderate climates. Its rated cooling capacity is 8600 BTU/hr, and its heating capacity range is 3000 - 8000 BTU/hr. The big difference between our flagship model and our cold climate model, currently in development, is in heating mode. Our cold climate model offers very high performance in freezing temperatures and heating mode. It's also better performing in cooling, but it is really optimized to hit those cold New York winters. Specs for our flagship model are available on our website, and we'll be announcing the specs for our cold climate very soon.

PHA: Do Gradient units work better in a humid or a drier climate?

Romanin: They're certainly designed for both. From a thermal perspective, all HVAC units work better in dry climates because removing water is harder. Think of your window AC. It condenses the humidity out of the air into water, which is an intensive process. It is not very thermodynamically efficient. Ideally, you would want that humidity to just move as vapor at the same temperature outside, but instead you're spending energy to cool it and then you dump the cold water outside.

So, you should do a couple of things. You should re-evaporate it on the condenser outside when you can. Our system does that. We take that condensate and spray it over the outside part, and it kind of cools the outdoor hot part. The second thing you should do is you should only dehumidify as much as you need to make people comfortable. A lot of times people say they get sick when they run the air conditioner because the air gets really dry. The indoor heat exchanger gets very cold and pulls out too much humidity.

What you can do when your system has a variable speed drive, which ours does, is you can to a certain extent control how much humidity you're pulling out. Right now, if you buy a Gradient, you can't set a specific humidity level, but one day you'll be able to. Because this is another area of comfort, right? It's air temperature, humidity, and radiant temperature, which are the three things that result in how you perceive the temperature in the environment. I guess air movement, too.

PHA: What types of refrigerants do the Gradient units use?

Romanin: So I've got a really long soapbox on refrigerants that I sometimes get into. I kind of hinted at it already, right? We invented CFCs in the 1950s, and they created a hole in the ozone layer. We had the Montreal Protocol that phased them out, and the industry shifted to HFCs. The part of that story that's not often told is that DuPont at first resisted the ozone science, but eventually came around and funded the Montreal Protocol. They were one of the supporters of it because they had lower margins on CFCs because their patents were expiring, and they had already patented HFCs and thought they could get higher margins with this transition.

They repeated the same trick with the Kigali Amendment. The chemical companies that own these HFC refrigerants partially funded the amendment that would phase them out because they had patents for the next generation of refrigerants, HFOs [hydrofluoroolefins], which have no ozone depletion potential and no global warming potential, but plenty of other bad things. It's still a manufactured fluorinated chemical for which we have little understanding of what happens if we dump massive quantities of it into the atmosphere. One of the things that we do know is that it breaks down into TFA [trifluoroacetic acid], which is a forever chemical that appears to have lots of issues disrupting plants and animals.

If you tried to think of two environmental disasters that threatened global ecosystems, ecosystems on a global level, you'd come up with the hole in the ozone layer and global warming, both in no small part caused by refrigerants. Do we want to phase out HFCs and replace it with another chemical and run this experiment again on a global scale?

The class of refrigerants that I believe should be the future are called natural refrigerants. They're called natural because we don't manufacture them. They are substances that exist naturally that we use as refrigerants. The three most common are CO2 (R-744), propane (R-290), and butane (R-600). CO2 is a global warming gas, but it has a GWP of 1 compared to GWP of 2000 [like R-410a].

There is a fear-mongering campaign that we can't use propane or butane as a refrigerant because it is flammable, which is true, but it's in such small quantities and that vulnerability is very controllable. If you have a mini fridge, it's probably using butane. No mini fridges ever explode. We are using propane to replace an open natural gas line in many homes—an open line that runs kilograms of natural gas through it on a regular basis in front of a pilot light. The idea that we can't switch to these other refrigerants is insane.

The good news is that there seems to be a massive movement in the industry away from HFOs and toward natural refrigerants. Europe has passed some laws outlawing the byproducts of HFOs. They also passed laws encouraging the use of natural refrigerants. Europe seems to be leaning into R-290, which I think is great, right? If Europe does it, and then a couple other regions do it, it will be enough to pull the rest of the industry along. The US does not have as good of a track record. Europe is hopefully going to drag us along.

To answer your question, though, in the US today we use R-32. R-32 is a steppingstone refrigerant. It has a GWP that is 68% lower than R-410a, the other standard. Our system also uses less charge and is hermetically sealed in the factory, not in the field, which allows us to cut down on leakage by 87% when compared to split system heat pumps. And looking ahead—while R-32 is better than R-410a, we still hope to eventually move to natural refrigerants.

PHA: What are some of the things that you’re most excited about for Gradient’s immediate and long-term future?

Romanin: We won a large contract with the New York City Housing Authority (NYCHA) to install our cold climate units in their buildings. They operate at much higher efficiency and capacity in freezing temps so that they can, even in New York City's winter, fully replace the boiler system. And with the 120-volt outlet. It's super difficult to do technically; but if you can do it, it’s going to unlock electrifying buildings and decommissioning New York City's boilers.

Local Law 97 is one of the drivers of this, but NYCHA also aimed to purchase systems that can pull the market forward. So they worked with NYSERDA and NYPA [the New York Power Authority] on what they called the Clean Heat for All Challenge that said, if someone makes this type of system, we'll buy tens of thousands of them. It was an incredibly ambitious and groundbreaking effort. We were selected as an awardee, which we're super proud of.

Long-term, we don't have anything specifically posted yet, but we’re working on this idea of how we can make your comfort multidimensional and more exciting and really lean into the emotional aspect of the comfort of your home. That’s what's exciting to us.

We also see a lot of new construction shifting more toward modular room systems. Running air ducts through a home is actually an incredibly expensive and inefficient way to do things. You have to control the temperature of the whole home instead of an individual room and ducts lose heat. The rest of the world seems to have already gone ductless. It's why you see way more mini splits in other countries besides the US and generally people like them more.

So, the modular architecture we've designed plays really well into where we see the future of HVAC going, even in new construction. We think it's going to be a per room system and we're going to lean into all of these other things I’ve mentioned: improving air quality, bringing in fresh air and natural light, and addressing different aspects of human comfort.

Currently, we’re only doing retrofits. Our system is very much designed to be easy to install for retrofits. When you look at the growth in AC across the globe, that 4 billion new ACs number comes from an IEC [International Electrotechnical Commission] report published in 2018 on the future growth of HVAC. Most of those new ACs will be in existing buildings in growing economies, where it’s people getting their first economic access to an AC. So, retrofits are the bigger problem just because building turnover is somewhere in the vicinity of around 80 years, whereas an HVAC system is shorter.

However, we want to be the platform for even new construction because this is absolutely a new construction problem as well. We want to be working with Passive House systems and we want to be integrated with the window design and the building design to provide even better user experiences. We think it is really amenable to providing great products for that market as well.

PHA: Some people have written about what is known as the “air conditioning doom loop,” where we have more people using air conditioners, which will burn more greenhouse gases, which will make global warming worse, which will make people use their ACs even more, and so on. How does Gradient address that issue and propose to solve that problem?

Romanin: The thing to know about the Doom Loop is that it is real. It is why we're here, right? I think that the first time I saw that I was like, “Yes, that is genius!” The thing is breaking the doom loop, at least to me as an engineer, seems very easy. It's technically easier to do than decarbonizing transportation or the electricity grid, but it wasn't happening.

A lot of folks early on would say, you’re going to run into Jevons Paradox. Jevons was an economist during the industrial revolution, and everyone was trying to make more efficient steam so that they could use less coal. His comment was, if we make a more efficient steam engine, we will not use less coal. We will use more coal because the service will become cheaper and we'll use more of it. This ended up being incredibly true, right? Efficiency led to more consumption of the resource. Following the same reasoning, if you make AC better, it's going to scale faster and we're going have more emissions.

I have two major problems with this. One is that there are many countries and entire regions where not having access to HVAC is a public health concern, right? If we are letting people suffer the effects of climate change and not letting them use tools to mitigate it, we are a part of the problem. Secondly, CO2 is not a fuel that we're using to create HVAC. It is a byproduct. It is a byproduct of this device that we use today that is easy to uncouple. Global warming emissions are not necessary for effective HVAC. You can build an HVAC system today that emits zero carbon. It just has to use electricity from a solar panel. It has to use a better refrigerant.

Once you do that, you will be left with the embodied energy. But those very small percentages, because you're going to get more than a 90% reduction if you do those things.

So that's the good news. The doom loop is possible to break. It's going to take a long time because building devices takes a long time to scale up. Buildings take a long time to change over. However, it's absolutely possible to have HVAC and zero carbon. That's Gradients mission. You shouldn't have to feel guilty when you run your HVAC in the summer. You shouldn't have to compromise other people or the planet when you do so.

All images courtesy of Gradient.