DALL·E generated image of an industrial plant.

With Heat From Heat Pumps, US Energy Requirements Could Plummet By 50%

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

Recently I had one of those embarrassing aha moments, one where something is so obvious that it not occurring to me much sooner made me cringe. Still, better late than never. And what was the epiphany? Well, that actually useful energy we have to make will drop drastically when we apply heat pumps everywhere possible in commercial, residential, and industrial heating.

This discussion has to start with Sankey energy flow diagrams. I shared a bunch of them from different organizations and countries a few weeks ago in an extended piece where I considered the requirements for an energy futures simulator. For this discussion I’ll start with the LLNL US energy flows diagram.

LLNL US Energy Flow Diagram for 2021 highlighting key energy services
LLNL US Energy Flow Diagram for 2021 highlighting key energy services

These numbers are in quads, which is to say quadrillion British Thermal Units (BTU) equivalents. The LLNL methodology converts everything into common units, which is a painful and usually thankless exercise, so I’m glad they have done it.

Two points about quads. The first is that LLNL’s methodology for converting renewable electricity into quads assumes fossil fuel plant heat rates for efficiency, which seems odd and I still haven’t gotten to the bottom of. It may be that they are understating renewables primary energy as a result, or it may be appropriate, but I haven’t found documentation or expansion on that. If anyone has a good pointer to their choices, please let me know. The second, of course, is the irony of the US persisting in using Imperial measurements.

What is the primary energy fallacy? Well, it’s the belief that we have to replace all the primary energy inputs on the left hand side of the chart above, as opposed to just the energy services including the ones I’ve called out. Take this piece for example: EIA: World energy consumption to increase 28% by 2040. That’s a US projection by the Energy Information Administration, a different organization than LLNL but part of the same government. Left hand please talk to right hand.

What we really have to replace is the energy services in light gray boxes on the right hand side of the chart. That’s right, the US doesn’t need to replace 97.3 quads of energy, it needs to replace 31.8 quads of energy. Or does it even need to do that?

Can we reduce the big wasted energy reflected in the rejected energy box? Sure, that’s why the mantra is electrify everything. (For nitpickers, the ‘possible’ in the mantra is silent and implied, not to mention vastly larger than nitpickers assume or wish to be true.) Using electricity in electric motors is vastly more efficient than using gasoline or diesel in cars or trucks for example, about four times more efficient.

As I’ve pointed out various times, when the US gets done with pretending that all ground transportation won’t be electrified, over 80% of energy requirements for transportation will be fully electrified, including rail. With most domestic aviation, all inland shipping and a lot of nearshore shipping electrifying in the coming decades, 21.2 quads of rejected energy from transportation will plummet to perhaps 2 quads.

And, of course, most of the electricity will be flowing from wind and solar fairly directly into grid-tied and battery-electric vehicles, so there will be a lot fewer losses from burning coal and gas for electricity. Wind and solar have transmission and storage losses, which are minimal, while gas and coal throw away 50% or more of the primary energy in them out of the box, and then have the transmission and storage losses on top of that.

In late 2020, I pointed out that Bill Gates’ favorite energy guru, Vaclav Smil, completely missed rejected energy, ignoring it in his published dyspeptic view of the energy transition. Gates apparently has read all 38 or so Smil’s books, and that has led to one of the world’s richest and most influential men and a founder of Breakthrough Energy Ventures missing the boat on energy for well over a decade. It was only months later in mid-2021 that Smil finally published a short piece acknowledging the primary energy fallacy. Even then his dyspeptic pessimism persisted. It is unclear if the implications of electrification had sunk in for him, never mind for the legions of people like Gates who he has misinformed. And to be clear, I agree with Smil that the transition will take decades, I just have a much smaller number of decades in mind.

The primary energy fallacy is pervasive. In the linked article, the EIA projects an increase of 161 quads of energy globally by 2050, mostly from increased use of fossil fuels. That headline was from 2017, but even their 2021 projections through 2050 reflect massive growth of primary energy demands, when the opposite will be true. Peak coal demand from 2013? Peak oil demand coming this decade most likely? Peak natural gas next decade? Not according to the EIA. Energy services will still grow even as primary energy shrinks. Or will it?

Renewables to electrified energy services will be sipping delicately at primary energy, unlike the messy gluttony around energy seen today.

Table of US 2021 residential, commercial and industrial energy services and rejected energy by author
Table of US 2021 residential, commercial and industrial energy services and rejected energy by author

Let’s look at the energy services I highlighted a bit more closely. I’ve moved numbers off the chart into a tabular form to make this a bit clearer.

Heating and cooling are the biggest energy draws in the residential category, representing 55% of energy demand per the same EIA that is confused by the primary energy fallacy above. The majority of that is for heating the building and heating water. That means that about 2.2 quads of residential energy services are for heating air and water.

And heat is the primary source of rejected energy. For these building types, the largest source of rejected energy is waste heat. There are relatively easy fixes for most of them, ones that will be added more and more over time for simple cost effectiveness reasons.

Why is the percentage of heating important to note? Heat pumps get about two-thirds of their energy from the environment and one-third from electricity. They move heat from place to place, just as a refrigerator or air conditioner does today. That means that 2.2 quads of energy services drops to about 0.7 quads. That’s 1.5 quads that don’t have to be provided by us via routes from primary energy sources. Various other things means wasted energy in residential settings will drop as well, mostly through continued electrification and more solid state technology replacing everything in the home. Can hot water be heated with a heat pump? Yes, yes it can. Modern heat pumps can provide hot-water heater temperature water for taps or radiant heating. A bit more insulation wouldn’t hurt and will likely be added at conversion to heat pumps in the coming decades to optimize the economics.

Chip in a few dollars a month to help support independent cleantech coverage that helps to accelerate the cleantech revolution!

What about commercial buildings? Well, the EIA indicates that a third of all commercial building energy is for heating, without splitting out hot water. That suggests that 1.9 quads of energy becomes 0.65 quads of demand. Wait, the same source says that electricity is already the dominant heating source for commercial buildings. Does that matter? No, it doesn’t. Electric base board heaters, forced air heaters, and hot water heaters just run electricity through resistance coils like those on stove tops, which means one unit of electricity turns into one unit of heat, just as burning natural gas does. Heat pumps turn one unit of electricity into three units of heat.

Industrial? This one is tougher. A lot of industrial heat is above 200° Celsius, the current reasonable limit for industrial heat pumps, and well above the 100° limit for current mature heat pump technologies. How much? Well, 45% of industrial heat is below 200° and 25% is below 100°. That means that with existing commoditized heat pumps we can already address 2.1 quads of industrial heat demand, turning it into 0.7 quads. And with well understood technical pathways, some engineering, manufacturing and distribution, we could address 3.8 quads. Since we are talking end game, let’s use the biggest number and keep at the simplified ratio of 1 unit of electricity for three units of heat. That turns 3.8 quads into 1.3 quads. A bunch of the rejected energy is bad insulation, creating excess heat in one place and not reusing it in another, and creating excess heat that’s out of time sequence with demand elsewhere. Better insulation, electric heat sources that are more tightly integrated with processes and so less wasteful, and some thermal storage will improve efficiency there as well.

What does this all turn into?

Table of US projected residential, commercial, and industrial energy services and rejected energy with heat pumps etc by author
Table of US projected residential, commercial, and industrial energy services and rejected energy with heat pumps etc by author

Instead of 26.2 quads of energy services, 20.5 quads, about a 38% reduction in energy services requirements. And with renewables flowing through some transmission and storage, primary energy requirements plummet from 46.64 quads to 24.6 quads, about a 47% drop in primary energy demand.

So that’s the eureka moment I had that was so long delayed. Heat pumps would actually drop energy service requirements substantially. Instead of having to replace all energy services, we only have to replace the portion of heat we can’t get from the environment for free. Our economies have enormous room to grow useful outputs without increasing energy services at all. 20% ‘energy’ services demand growth by 2050? No problem. 20% increase in primary energy? Get a grip.

The vast majority of these massive energy savings involved applying existing technologies, ones that are already scaled commodities. Like generating more low-carbon electricity with renewables, moving it with HVDC, and storing it as necessary with closed-loop pumped hydro — no invention is required, just deployment.

Let’s do one more calculation. Let’s add transportation back into the mix. Remember, 2 quads rejected energy when everything possible is electrified. That means about 8 quads of primary energy requirements. Added to the roughly 25 quads of primary energy requirements for commercial, residential, and industrial energy services, that’s about 33 quads of primary energy requirements.. The biomass needs to shift to biofuels for the hard to decarbonize segments of aviation and marine shipping, but those are small segments of overall energy demand.

So how many quads do wind, solar, hydro, nuclear, geothermal, and biomass provide the US economy today? Over 20. About 80% of the energy requirements of a fully electrified society with heat pumps everywhere possible already exist. Electrification and heat pumps radically reduce the requirement to build new wind, solar, nuclear, hydro and geothermal primary energy sources.

Is this the likely outcome? No, not at all, at least not in the US by 2050. Among other things, human nature is built on Jevons Paradox, so people will open windows in the wintertime while turning the heat up, and fixing industrial heating faces many headwinds. And to be clear, while I’m comfortable with the heat pump ratios, I’m doing a bit of arm waving on remaining rejected energy, so that might be higher.

But it is indicative of the scale of the possible, and it’s likely where things will be in 2100. It’s much closer to reality than projections by those stuck in the primary energy fallacy. It’s a clear indication that a transition to a low-carbon economy with all of the same comforts, travel, and the like that we have now is well within reach. It’s also the best path for emerging economies to take from the start, just as they went straight to wireless data and voice, bypassing data flowing over wires into every home and office. Oh, and all that wasted heat today? It’s strongly correlated with green house gas emissions and health-harming air pollution too. How many wins is that?

UPDATE: We reached out to Mark Z. Jacobson at Stanford with a follow-up question, which he was gracious enough to respond with the following information:

Q: Does the Stanford scenario modeling of 100% renewables by 2050 have a methodology for reduced energy service requirements through application of heat pump technologies, and if so, how is it calculated?

A: Yes, ever since our first study in 2009, we have accounted for reduced end-use energy needs due to electrification with heat pumps, electric vehicles, and electrified industry.

Here is a summary I posted on LinkedIn the other day:
We have been treating energy reductions like you are proposing since 2009 (e.g., initially in this article https://web.stanford.edu/group/efmh/jacobson/Articles/I/sad1109Jaco5p.indd.pdf) First, we start with end-use energy (from IEA), which is energy directly used by a consumer. It is the energy embodied in electricity, natural gas, gasoline, diesel, kerosene, and jet fuel that people use directly. It equals primary energy minus the energy lost in converting primary energy to end-use energy, including the energy lost during transmission and distribution and waste heat. We then electrify all end uses. So, for example, going from a gasoline car to an electric car reduces end-use energy by another ~75% beyond the conversion from oil to gasoline as does going from a natural gas water or air heater to a heat pump. Here is the latest summary of all the energy reductions past end-use energy between BAU and WWS. https://web.stanford.edu/group/efmh/jacobson/Articles/I/145Country/145-Timeline.pdf There are 5 major energy saving: due to EVs vs ICEs; heat pumps versus combustion; electrifying industry; eliminating energy due to mining fuels; and end-use energy improvements beyond BAU.
I am fairly sure IPCC scenarios ignore most if not all five of these reductions, though.

Mea culpa: a sharp eyed reader noticed I’d transposed some numbers in an earlier version of this assessment. It ended up being close to a wash in the end result, but the article has nonetheless been corrected.

Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.

Latest CleanTechnica.TV Video

CleanTechnica uses affiliate links. See our policy here.

Michael Barnard

is a climate futurist, strategist and author. He spends his time projecting scenarios for decarbonization 40-80 years into the future. He assists multi-billion dollar investment funds and firms, executives, Boards and startups to pick wisely today. He is founder and Chief Strategist of TFIE Strategy Inc and a member of the Advisory Board of electric aviation startup FLIMAX. He hosts the Redefining Energy - Tech podcast (https://shorturl.at/tuEF5) , a part of the award-winning Redefining Energy team.

Michael Barnard has 708 posts and counting. See all posts by Michael Barnard