The article does a good job of showing how a typical barrel of oil is converted into a dozen or more distinct usable products.
It would be helpful to also have a chart that shows how much gasoline or diesel as a percentage of each barrel is produced. It would be a bit variable, since not all crude oil is the same, but I think it would be close for most of it.
Some people think when diesel and regular gas prices diverge, that they should just be able to produce one at the expense of the other; but the distillation process shows that they are fundamentally different.
The article is quick to point out the huge role of oil in the modern energy mix. It also fails to note that most of the energy ends up us waste heat. The so called "Primary energy fallacy". Other than that, it's a great read.
To me (as someone who has worked on oil rigs, oil pipelines, oil refineries, and chemical plants), crude oil seems far more valuable as a material than as an energy source. It feels like a damned shame that we're still combusting so much of it for heat rather than reserving it for physical materials.
I understand the ways that economics are very important, and that the economics still currently favor burning a large fraction of the crude oil. But I also know that the right kinds of investments and a bit of luck can often change those economics, and that would be nice to see.
> It also fails to note that most of the energy ends up us waste heat.
I've heard the statistic that 40% of the total oil pumped out of the ground just to transporting oil. We use almost half just to move it to and fro before even using it.
As someone with no real-world petrochemistry experience, but much gaming experience, I was very surprised how familiar the crude oil processing diagram looks. Factorio and GregTech are two prime examples of fairly realistic oil processing lines (probably as accurate as any game would reasonably try to be).
This is a really good overview of oil refining. I'll add a few things.
1. The light and heavy distinction is covered by a measure called API gravity [1]. The higher the API gravity, the lighter the crude;
2. Refiners mix different crude types depending on what kind of refined products they want to produce;
3. Heavy crude tends to be less valuable although it's essential for some applications. Lighter crude produces generally more valuable products like gasoline, diesel and avgas. But heavy crude goes into construction (eg roads) and fuel for ships (ie bunkers));
4. Most refineries in the US are very old and are very polluting. They don't need to be this way. A new refiner would produce vastly less pollution but they're almost impossible to get permission to build now. One exception is the Southern Rock refinery currently being built in Oklahoma [2], which will be powered by largely renewable energy and produce a lot less emissions than an equivalent older refinery with the same capacity;
5. There are different blends of gasoline that the US produces. The biggest is so-called summer and winter blends. What's the differene? Additives are added to summer blends (in particular) to increase the boiling point so less of the gasoline is in gas form because that produces more smog;
6. California uses their own blends so in 2021-2022 when CA gas went to $8+, it wasn't just "gouging". It doesn't really work that way. CA requires a particular blend that only CA refineries produce so it's simple supply and demand as no new capacity gets added to CA refineries and demand goes up with population growth.
The reason for the CA blend goes back to the 80s and 90s when smog was a much bigger problem. Better vehicle emissions standards since then as well as improvements in the blends the rest of the country uses have largely made the CA blend obsolete so CA is really paying $1+/gallon more for literally no reason; and
7. California doesn't build pipelines so is entirely dependent on seaborne oil imports (~75%) despite the US being a net energy exporter. Last I checked, ~20% of that foreign oil comes through the Strait (from Iraq, mostly) so, interestingly, CA is more vulnerable to the Strait of Hormuz closure than the rest of the country.
I guess I'll add a disclaimer: I'm very much pro-renewables, particular solar. I think solar is the future. But we currently live in a world that has huge demand for oil and no alternatives for many of those uses (eg diesel, plastics, construction, industrial, avgas) so we should at least be smart about how we go forward.
> they're almost impossible to get permission to build now
While I do agree there's a ton of regulatory hurdle to cross to build a new refinery, lots of interviews with oil executives have stated the economics of building a new refinery aren't always great. The reasons why they aren't building isn't necessarily because the regulatory hurdles are too high, its that they don't think they'll end up making any money building them. The future demand of many refined products are uncertain, adding a lot of new capacity is quite a capital risk.
I'd love to see a lot of our ancient refineries shut down and replaced with far more modern designs, but the oil industry isn't going to do it because it probably won't be profitable.
It will be interesting to see the economics of these few new refineries coming online actually play out in the coming years.
Well-meaning legislation (eg CEQA in CA) is effectively weaponized by NIMBYs who have outsized power to add years if not a decade or more to something getting built. There is also an overly naive, even performative opposition to anything fossil fuel related without having a substitute (again, I say this as a particularly pro-solar person). This adds significantly to costs.
I'm also anti-nuclear because it's too expensive, not as safe as advocates make out and the waste problem is not even remotely solved despites all the claims to the contrary. But it's also true that the same kind of anti-development tactics used against refineries are effectively used against nuclear plants such that it takes 15+ years to build a nuclear plant and the costs balloon as a result.
But there's also strong direct evidence contrary to your claim: the new refineries in Oklahoma and Texas. Why are they getting built if "the oil industry isn't going to do it"?
I'll go even further than this: if private industry won't build new refineries, the government should. In fact, that's my preferred outcome anyway.
>While I do agree there's a ton of regulatory hurdle to cross to build a new refinery, lots of interviews with oil executives have stated the economics of building a new refinery aren't always great. The reasons why they aren't building isn't necessarily because the regulatory hurdles are too high, its that they don't think they'll end up making any money building them. The future demand of many refined products are uncertain, adding a lot of new capacity is quite a capital risk.
This is a circular statement.
The regulatory hurdles are a large part of what drive cost.
I know a venue that wants to pave a dirt lot so they can better use it for stuff. It doesn't pencil out because of stupid stormwater permitting crap that'll add $250k to the project. It'd never pay off in a reasonable timeframe. So it just continues to exist in its current grandfathered in capacity when even the most unfavorable napkin math shows that what they want is an improvement.
A few weeks ago I was party to the installation of a perimeter railing on a flat commercial roof. The railing cost more than the rest of the job it was there for. Something tells me they won't be pulling permits for petty electrical work ever again.
Oil and most other heavy industry is faced with the same sort of problems with more digits in front of the decimal.
> The regulatory hurdles are a large part of what drive cost
I agree, they are a large part. The things they have to do to meet the standards are expensive.
The claim was "impossible to get permission to build now". As in, the government won't let them build it. That the standards are just technically impossible to meet. They can get the permission to build it any day. Its possible to meet these standards. They just don't think it'll be worth it when they have to do it right.
> Most refineries in the US are very old and are very polluting
India's Reliance is also investing $300B [0] in a Texas megarefinery [1] in specifically for cleaner and more efficient shale refining.
This is deeply technical and complex but low margins work (semiconductor fabrication falls in the same boat) which saw this industry leave for abroad in the 2000s and 2010s when other states like China and India subsidized their refinery industries to build domestic capacity for a number of petroleum byproducts with industrial applications.
This is the same strategy Japan, South Korea, and Taiwan used in the 1960s-90s as well.
Cars are the most familiar to the everyday user, which is why it's the most common in perception. It's also actually one of the easier ones to solve (ie it's basically done).
Trucking is technically not to hard but logistically difficult. Aviation is extremely technically challenging. Shipping is economically difficult. Electricity generation has lots of factors, there's a lot of generation that can and has been changed easily, but some generation which is harder to switch.
If you get outside of oil into CO2 generally, there's even thornier issues. Concrete production, for example.
Oil is cooked. BYD is filing 52 patents every single day and has a 700 km in 9 minutes vehicle available TODAY ! Charging by Solar is going to be the norm. Watch : https://www.youtube.com/watch?v=vgCYYrhL-kE
Because airplanes are also a type of vehicles. So are container ships. Neither of which are very practicable with pure electric AFAICT, and are integral to modern life. (Though more marine hybrid could be practical.)
I think there should be more of a push for BEV/hybrid cars (and transport trucks), and think more home electrification would be good (though air sealing and insulation are more important, relatively speaking). But let us set reasonable expectations of what is possible at various timeframes (and not let the perfect be the enemy of the good/better).
Oil is cooked. BYD is filing 52 patents every single day and has a 700 km in 9 minutes vehicle available TODAY ! Charging by Solar is going to be the norm. Watch : https://www.youtube.com/watch?v=vgCYYrhL-kE
If you're interested in how the oil industry as a whole operates and why, Oil 101 is an interesting read.
The article does a good job of showing how a typical barrel of oil is converted into a dozen or more distinct usable products.
It would be helpful to also have a chart that shows how much gasoline or diesel as a percentage of each barrel is produced. It would be a bit variable, since not all crude oil is the same, but I think it would be close for most of it.
Some people think when diesel and regular gas prices diverge, that they should just be able to produce one at the expense of the other; but the distillation process shows that they are fundamentally different.
The article is quick to point out the huge role of oil in the modern energy mix. It also fails to note that most of the energy ends up us waste heat. The so called "Primary energy fallacy". Other than that, it's a great read.
To me (as someone who has worked on oil rigs, oil pipelines, oil refineries, and chemical plants), crude oil seems far more valuable as a material than as an energy source. It feels like a damned shame that we're still combusting so much of it for heat rather than reserving it for physical materials.
I understand the ways that economics are very important, and that the economics still currently favor burning a large fraction of the crude oil. But I also know that the right kinds of investments and a bit of luck can often change those economics, and that would be nice to see.
> It also fails to note that most of the energy ends up us waste heat.
I've heard the statistic that 40% of the total oil pumped out of the ground just to transporting oil. We use almost half just to move it to and fro before even using it.
Is this accurate?
As someone with no real-world petrochemistry experience, but much gaming experience, I was very surprised how familiar the crude oil processing diagram looks. Factorio and GregTech are two prime examples of fairly realistic oil processing lines (probably as accurate as any game would reasonably try to be).
This is a really good overview of oil refining. I'll add a few things.
1. The light and heavy distinction is covered by a measure called API gravity [1]. The higher the API gravity, the lighter the crude;
2. Refiners mix different crude types depending on what kind of refined products they want to produce;
3. Heavy crude tends to be less valuable although it's essential for some applications. Lighter crude produces generally more valuable products like gasoline, diesel and avgas. But heavy crude goes into construction (eg roads) and fuel for ships (ie bunkers));
4. Most refineries in the US are very old and are very polluting. They don't need to be this way. A new refiner would produce vastly less pollution but they're almost impossible to get permission to build now. One exception is the Southern Rock refinery currently being built in Oklahoma [2], which will be powered by largely renewable energy and produce a lot less emissions than an equivalent older refinery with the same capacity;
5. There are different blends of gasoline that the US produces. The biggest is so-called summer and winter blends. What's the differene? Additives are added to summer blends (in particular) to increase the boiling point so less of the gasoline is in gas form because that produces more smog;
6. California uses their own blends so in 2021-2022 when CA gas went to $8+, it wasn't just "gouging". It doesn't really work that way. CA requires a particular blend that only CA refineries produce so it's simple supply and demand as no new capacity gets added to CA refineries and demand goes up with population growth.
The reason for the CA blend goes back to the 80s and 90s when smog was a much bigger problem. Better vehicle emissions standards since then as well as improvements in the blends the rest of the country uses have largely made the CA blend obsolete so CA is really paying $1+/gallon more for literally no reason; and
7. California doesn't build pipelines so is entirely dependent on seaborne oil imports (~75%) despite the US being a net energy exporter. Last I checked, ~20% of that foreign oil comes through the Strait (from Iraq, mostly) so, interestingly, CA is more vulnerable to the Strait of Hormuz closure than the rest of the country.
I guess I'll add a disclaimer: I'm very much pro-renewables, particular solar. I think solar is the future. But we currently live in a world that has huge demand for oil and no alternatives for many of those uses (eg diesel, plastics, construction, industrial, avgas) so we should at least be smart about how we go forward.
[1]: https://en.wikipedia.org/wiki/API_gravity
[2]: https://www.oklahoman.com/story/news/2023/05/24/5-6-billion-...
Looking at the chart in the article I was kind of surprised at how small wind and solar are globally and that coal is still ~25%.
That’s because of the primary energy fallacy: https://medium.com/@jan.rosenow/have-we-been-duped-by-the-pr...
TL;DR: the efficiency of converting fossil energy resources into something useful is poor.
> they're almost impossible to get permission to build now
While I do agree there's a ton of regulatory hurdle to cross to build a new refinery, lots of interviews with oil executives have stated the economics of building a new refinery aren't always great. The reasons why they aren't building isn't necessarily because the regulatory hurdles are too high, its that they don't think they'll end up making any money building them. The future demand of many refined products are uncertain, adding a lot of new capacity is quite a capital risk.
I'd love to see a lot of our ancient refineries shut down and replaced with far more modern designs, but the oil industry isn't going to do it because it probably won't be profitable.
It will be interesting to see the economics of these few new refineries coming online actually play out in the coming years.
Well-meaning legislation (eg CEQA in CA) is effectively weaponized by NIMBYs who have outsized power to add years if not a decade or more to something getting built. There is also an overly naive, even performative opposition to anything fossil fuel related without having a substitute (again, I say this as a particularly pro-solar person). This adds significantly to costs.
I'm also anti-nuclear because it's too expensive, not as safe as advocates make out and the waste problem is not even remotely solved despites all the claims to the contrary. But it's also true that the same kind of anti-development tactics used against refineries are effectively used against nuclear plants such that it takes 15+ years to build a nuclear plant and the costs balloon as a result.
But there's also strong direct evidence contrary to your claim: the new refineries in Oklahoma and Texas. Why are they getting built if "the oil industry isn't going to do it"?
I'll go even further than this: if private industry won't build new refineries, the government should. In fact, that's my preferred outcome anyway.
>While I do agree there's a ton of regulatory hurdle to cross to build a new refinery, lots of interviews with oil executives have stated the economics of building a new refinery aren't always great. The reasons why they aren't building isn't necessarily because the regulatory hurdles are too high, its that they don't think they'll end up making any money building them. The future demand of many refined products are uncertain, adding a lot of new capacity is quite a capital risk.
This is a circular statement.
The regulatory hurdles are a large part of what drive cost.
I know a venue that wants to pave a dirt lot so they can better use it for stuff. It doesn't pencil out because of stupid stormwater permitting crap that'll add $250k to the project. It'd never pay off in a reasonable timeframe. So it just continues to exist in its current grandfathered in capacity when even the most unfavorable napkin math shows that what they want is an improvement.
A few weeks ago I was party to the installation of a perimeter railing on a flat commercial roof. The railing cost more than the rest of the job it was there for. Something tells me they won't be pulling permits for petty electrical work ever again.
Oil and most other heavy industry is faced with the same sort of problems with more digits in front of the decimal.
> This is a circular statement.
Its not if you get the context.
> The regulatory hurdles are a large part of what drive cost
I agree, they are a large part. The things they have to do to meet the standards are expensive.
The claim was "impossible to get permission to build now". As in, the government won't let them build it. That the standards are just technically impossible to meet. They can get the permission to build it any day. Its possible to meet these standards. They just don't think it'll be worth it when they have to do it right.
> Most refineries in the US are very old and are very polluting
India's Reliance is also investing $300B [0] in a Texas megarefinery [1] in specifically for cleaner and more efficient shale refining.
This is deeply technical and complex but low margins work (semiconductor fabrication falls in the same boat) which saw this industry leave for abroad in the 2000s and 2010s when other states like China and India subsidized their refinery industries to build domestic capacity for a number of petroleum byproducts with industrial applications.
This is the same strategy Japan, South Korea, and Taiwan used in the 1960s-90s as well.
[0] - https://www.bloomberg.com/opinion/articles/2026-03-17/ambani...
[1] - https://www.bloomberg.com/news/articles/2026-03-11/reliance-...
Cool to see how when people talk about “transitioning off oil” it's more than replacing gasoline in cars. It's replacing this entire global machine.
Cars are the most familiar to the everyday user, which is why it's the most common in perception. It's also actually one of the easier ones to solve (ie it's basically done).
Trucking is technically not to hard but logistically difficult. Aviation is extremely technically challenging. Shipping is economically difficult. Electricity generation has lots of factors, there's a lot of generation that can and has been changed easily, but some generation which is harder to switch.
If you get outside of oil into CO2 generally, there's even thornier issues. Concrete production, for example.
If you are seriously interested in these issues, I highly recommend https://www.youtube.com/c/EngineeringwithRosie
Oil is cooked. BYD is filing 52 patents every single day and has a 700 km in 9 minutes vehicle available TODAY ! Charging by Solar is going to be the norm. Watch : https://www.youtube.com/watch?v=vgCYYrhL-kE
> Oil is cooked. BYD is […]
By "vehicles" do you mean "cars"?
Because airplanes are also a type of vehicles. So are container ships. Neither of which are very practicable with pure electric AFAICT, and are integral to modern life. (Though more marine hybrid could be practical.)
I think there should be more of a push for BEV/hybrid cars (and transport trucks), and think more home electrification would be good (though air sealing and insulation are more important, relatively speaking). But let us set reasonable expectations of what is possible at various timeframes (and not let the perfect be the enemy of the good/better).
Oil is cooked. BYD is filing 52 patents every single day and has a 700 km in 9 minutes vehicle available TODAY ! Charging by Solar is going to be the norm. Watch : https://www.youtube.com/watch?v=vgCYYrhL-kE