This is exploratory investigation of a new-ish hypothesis, it is not intended to be a comprehensive review of the field or even a a full investigation of the hypothesis.
I’ve always been skeptical of the seed-oil theory of obesity. Perhaps this is bad rationality on my part, but I’ve tended to retreat to the sniff test on issues as charged and confusing as diet. My response to the general seed-oil theory was basically “Really? Seeds and nuts? The things you just find growing on plants, and that our ancestors surely ate loads of?”
But a twitter thread recently made me take another look at it, and since I have a lot of chemistry experience I thought I’d take a look.
The PUFA Breakdown Theory
It goes like this:
PUFAs from nuts and seeds are fine. Deep-frying using PUFAs causes them to break down in a way other fatty acids do not, and these breakdown products are the problem.
Most of a fatty acid is the “tail”. This consists of hydrogen atoms decorating a backbone of carbon atoms. Each carbon atom can make up to four bonds, of which two have to be to other carbons (except the end carbon which only bonds to one carbon) leaving space for two hydrogens. When a chain has the maximum number of hydrogen atoms, we say it’s “saturated”. These tails have the general formula CnH2n+1:
For a carbon which is saturated (i.e. has four single bonds) the bonds are arranged like the corners of a tetrahedron, and rotation around single bonds is permitted, meaning the overall assembly is like a floppy chain.
Instead, we can have two adjacent carbons form a double bond, each forming one bond to hydrogen, two bonds to the adjacent carbon, and one to a different carbon:
Trans (top) and cis (bottom) double bonds.
Unlike single bonds, double bonds are rigid, and if a carbon atom has a double bond, the three remaining bonds fall in a plane. This means there are two ways in which the rest of the chain can be laid out. If the carbons form a zig-zag S shape, this is a trans double bond. If they form a curved C shape, we have a cis double bond.
The health dangers of trans-fatty acids have been known for a long while. They don’t occur in nature (which is probably why they’re so bad for us). Cis-fatty acids are very common though, especially in vegetable and, yes, seed oils.
Of course there’s no reason why we should stop at one double bond, we can just as easily have multiple. This gets us to the name poly-unsaturated fatty acids (PUFAs). I’ll compare stearic acid (SA) oleic acid (OA) and linoleic acid (LA) for clarity:
Going from stearic acid to oleic acid to linoleic acid, the number of (cis) double bonds increases by one each time.
Linoleic acid is the one that seed oil enthusiasts are most interested in. We can go even further and look at α-linoleic acid, which has even more double bonds, but I think LA makes the point just fine.
Three fatty acids, usually identical ones, attach to one glycerol molecule to form a triglyceride.
Isomerization
As I mentioned earlier, double bonds are rigid, so if you have a cis double bond, it stays that way. Mostly. In chemistry a reaction is never impossible, the components are just insufficiently hot. If we heat up a cis-fatty acid to a sufficient temperature, the molecules will be able to access enough energy to flip. The rate of reactions generally scales with temperature according to the Arrhenius equation:
v=Aexp(−EakBT)
Where A is a general constant determining the speed, Ea is the “activation energy” of the reaction, T is temperature, and kB is a Boltzmann’s constant which makes the units work out. Graphing this gives the following shape:
Suffice to say this means that reaction speed can grow very rapidly with temperature at the “right” point on this graph.
Why is this important? Well, trans-fatty acids are slightly lower energy than cis ones, so at a high enough temperature, we can see cis to trans isomerization, turning OA or LA into something you 100% definitely do not want to be eating. As far as I’m aware nobody claims trans fats aren’t bad.
So what’s the right temperature, if it’s 1000∘C then we don’t care. The literature seems to be somewhat split on how fast this reaction is. One study claims that after 5 hours, 0.2% of LA is converted to a trans form at 140 ∘C, and about 1% is converted at 220 ∘C (284 and 428 F respectively).[1] Another finds that after 1 hour, trans isomers are negligible at temperatures below 220 ∘C.[2]
This second study also notes that OA is converted to trans isomers at a higher rate than LA which is corroborated by a study of OA breakdown.[3]
Now lets think about how much trans fat is likely to be produced during a typical deep frying episode. At-home deep frying typically lasts <1 hour in my experience (usually less than 30 minutes), although oil might be re-used several times, which is probably something to consider.
McDonald’s on the other hand… changes their frying oil every two weeks. 8 hours by 14 days = 112 hours of operations. I actually can’t get great data on their frying temperature, but let’s try the full gamut of options. Here’s the graph from the paper earlier:
I don’t know how much trans OA is harmful. Margarine is roughly 5-10% trans fats and that’s supposed to be pretty bad. It is also plausible I got these numbers wrong, since the paper insists on insane units like (g/100g)h−1 and doesn’t give any units in their table there.
The fact that OA makes trans fats faster than LA surprised me until I noticed the reason. LA is more likely than OA to be oxidized to hydroperoxides. The first study also notes up to a ~25% decrease in total un-oxidized LA after 5 hours at 220 ∘C (I’ve said “up to” since there could be other issues at play like hydrolysis creating free fatty acids).
Hydroperoxides
As well as flipping round, double bonds can react with oxygen at an adjacent carbon to form a “hydroperoxide”. The rate of hydroperoxide formation is pretty high (again, up to 25% of the LA after 5 hours at 220 ∘C) so for PUFAs, this is the major degradation product over normal trans-fatty acids.
“Singlet oxygen” means an excited state of oxygen which is formed at high temperatures (oxygen actually has this weird quirk which makes it unreactive, which is why we don’t normally catch fire spontaneously, try walking around in 20% fluorine or chlorine gas to appreciate the difference). Deep-frying can raise oxygen to this state, allowing it to oxidize lipids to hydroperoxides, which happens to the exclusion of the formation of trans-LA.
The hydroperoxides also often take the form trans-fatty acids as well as being oxidized. These don’t show up when looking for normal trans-LA but they’re presumably just as harmful.
Once a hydroperoxide has formed, it can catalyse the formation of more hydroperoxides, which is probably why LA (2 double bonds per fatty acid, so 6 per triglyceride molecule) is so eager. LA also has an especially vulnerable-looking carbon between two sets of double bonds.
So what might be the effects of peroxides in diet? The first paper I found contains the phrase “while fish oil had been considered poisonous, in fact, fresh fish oil itself was not poisonous, unlike oxidized fish oil”.[4] So presumably at high enough concentrations they are bad! This paper shows that oxidized oils are pretty bad, say it with me now, IN MICE and also IN HUMAN CELLS IN A DISH, but doesn’t actually give much causal evidence in humans beyond association studies of oxides in serum. Old and sick people have different stuff in their blood, more at 11.
Another paper suggests that dietary lipid peroxides might cause cancer, but doesn’t say much about the sorts of things that seed oil proponents often mention (weight gain, inflammation, etc.)[5]
Nonetheless, if these things are poisonous at high concentrations, they’re probably not great at low concentrations.
Final Thoughts
All of the degradation studies have been done on pure oils, and most likely they were done in an inert (glass) vessel. Many, many things can catalyse oxidation and isomerization reactions, including metal ions, so I suspect cooking with actual food in an actual deep fryer would cause more of both reactions.
Ideally I’d like someone to just sample the McDonald’s cooking oil every few days through a few cycles of oil changes, but I doubt this will happen.
Olive oil is probably much better than other oils for normal frying for two reasons. Firstly it contains very low PUFAs, secondly it contains anti-oxidant compounds which might be protective against lipid peroxide formation.
Personally, I am not too worried about trans-fat formation in my own kitchen. The amount of formation is probably extremely low at anything resembling a normal cooking time. Unless you re-use cooking oils for days at a time, or cook at extreme temperatures, you’re most likely to be fine.
I am somewhat worried about the formation of lipid peroxides in my own cooking. I normally cook with olive oil or butter anyway, but in the rare case when I deep-fry things I’ll make two changes:
Try to use lower-PUFA oils for deep frying. Peanut/groundnut oil EDIT: Mustard oil might be even better will likely be my primary choice based on the intersection of PUFAs/price/ethics/flavour/availability.
Never re-use deep-frying oil
I think that peanut oil is less likely to cause health problems than other deep-frying oils, since it’s the main deep-frying oil in east Asia, which has much less obesity than the west, even in places like the Philippines where deep fried food is most common.
I am also somewhat concerned about trans-fat formation in commercial cooking, but this is dwarfed by the significant worries about oxidation from commercial cooking which will cash out as eating much fewer shop-bought deep-fried foods like crisps (chips) and chips (fries).
I think the PUFA breakdown hypothesis is more likely than the naive seed oil hypothesis, but I haven’t spent any time looking for counter-evidence yet.
So What’s Up With PUFAs Chemically?
This is exploratory investigation of a new-ish hypothesis, it is not intended to be a comprehensive review of the field or even a a full investigation of the hypothesis.
I’ve always been skeptical of the seed-oil theory of obesity. Perhaps this is bad rationality on my part, but I’ve tended to retreat to the sniff test on issues as charged and confusing as diet. My response to the general seed-oil theory was basically “Really? Seeds and nuts? The things you just find growing on plants, and that our ancestors surely ate loads of?”
But a twitter thread recently made me take another look at it, and since I have a lot of chemistry experience I thought I’d take a look.
The PUFA Breakdown Theory
It goes like this:
PUFAs from nuts and seeds are fine. Deep-frying using PUFAs causes them to break down in a way other fatty acids do not, and these breakdown products are the problem.
Most of a fatty acid is the “tail”. This consists of hydrogen atoms decorating a backbone of carbon atoms. Each carbon atom can make up to four bonds, of which two have to be to other carbons (except the end carbon which only bonds to one carbon) leaving space for two hydrogens. When a chain has the maximum number of hydrogen atoms, we say it’s “saturated”. These tails have the general formula CnH2n+1:
For a carbon which is saturated (i.e. has four single bonds) the bonds are arranged like the corners of a tetrahedron, and rotation around single bonds is permitted, meaning the overall assembly is like a floppy chain.
Instead, we can have two adjacent carbons form a double bond, each forming one bond to hydrogen, two bonds to the adjacent carbon, and one to a different carbon:
Unlike single bonds, double bonds are rigid, and if a carbon atom has a double bond, the three remaining bonds fall in a plane. This means there are two ways in which the rest of the chain can be laid out. If the carbons form a zig-zag S shape, this is a trans double bond. If they form a curved C shape, we have a cis double bond.
The health dangers of trans-fatty acids have been known for a long while. They don’t occur in nature (which is probably why they’re so bad for us). Cis-fatty acids are very common though, especially in vegetable and, yes, seed oils.
Of course there’s no reason why we should stop at one double bond, we can just as easily have multiple. This gets us to the name poly-unsaturated fatty acids (PUFAs). I’ll compare stearic acid (SA) oleic acid (OA) and linoleic acid (LA) for clarity:
Linoleic acid is the one that seed oil enthusiasts are most interested in. We can go even further and look at α-linoleic acid, which has even more double bonds, but I think LA makes the point just fine.
Three fatty acids, usually identical ones, attach to one glycerol molecule to form a triglyceride.
Isomerization
As I mentioned earlier, double bonds are rigid, so if you have a cis double bond, it stays that way. Mostly. In chemistry a reaction is never impossible, the components are just insufficiently hot. If we heat up a cis-fatty acid to a sufficient temperature, the molecules will be able to access enough energy to flip. The rate of reactions generally scales with temperature according to the Arrhenius equation:
v=Aexp(−EakBT)
Where A is a general constant determining the speed, Ea is the “activation energy” of the reaction, T is temperature, and kB is a Boltzmann’s constant which makes the units work out. Graphing this gives the following shape:
Suffice to say this means that reaction speed can grow very rapidly with temperature at the “right” point on this graph.
Why is this important? Well, trans-fatty acids are slightly lower energy than cis ones, so at a high enough temperature, we can see cis to trans isomerization, turning OA or LA into something you 100% definitely do not want to be eating. As far as I’m aware nobody claims trans fats aren’t bad.
So what’s the right temperature, if it’s 1000 ∘C then we don’t care. The literature seems to be somewhat split on how fast this reaction is. One study claims that after 5 hours, 0.2% of LA is converted to a trans form at 140 ∘C, and about 1% is converted at 220 ∘C (284 and 428 F respectively).[1] Another finds that after 1 hour, trans isomers are negligible at temperatures below 220 ∘C.[2]
This second study also notes that OA is converted to trans isomers at a higher rate than LA which is corroborated by a study of OA breakdown.[3]
Now lets think about how much trans fat is likely to be produced during a typical deep frying episode. At-home deep frying typically lasts <1 hour in my experience (usually less than 30 minutes), although oil might be re-used several times, which is probably something to consider.
McDonald’s on the other hand… changes their frying oil every two weeks. 8 hours by 14 days = 112 hours of operations. I actually can’t get great data on their frying temperature, but let’s try the full gamut of options. Here’s the graph from the paper earlier:
If we assume a conservative 160 ∘C:
\(\ln k = −11700 \times \frac 1{433} + 20.7 = −6.3\\)
k=e−6.3=0.0018 (g/100g) h−1
Which gives around 0.2% trans OA.
If instead they do a average-ish 175 ∘C:
\(\ln k = −11700 \times \frac 1{449} + 20.7 = −5.5\\)
k=e−5.5=0.0041 (g/100g) h−1
Which gives around 0.46% trans OA
Or a high temperature 190 ∘C:
\(\ln k = −11700 \times \frac 1{463} + 20.7 = −4.6\\)
k=e−4.6=0.01(g/100g) h−1
Which gives 1.1% trans OA.
I don’t know how much trans OA is harmful. Margarine is roughly 5-10% trans fats and that’s supposed to be pretty bad. It is also plausible I got these numbers wrong, since the paper insists on insane units like (g/100g)h−1 and doesn’t give any units in their table there.
The fact that OA makes trans fats faster than LA surprised me until I noticed the reason. LA is more likely than OA to be oxidized to hydroperoxides. The first study also notes up to a ~25% decrease in total un-oxidized LA after 5 hours at 220 ∘C (I’ve said “up to” since there could be other issues at play like hydrolysis creating free fatty acids).
Hydroperoxides
As well as flipping round, double bonds can react with oxygen at an adjacent carbon to form a “hydroperoxide”. The rate of hydroperoxide formation is pretty high (again, up to 25% of the LA after 5 hours at 220 ∘C) so for PUFAs, this is the major degradation product over normal trans-fatty acids.
“Singlet oxygen” means an excited state of oxygen which is formed at high temperatures (oxygen actually has this weird quirk which makes it unreactive, which is why we don’t normally catch fire spontaneously, try walking around in 20% fluorine or chlorine gas to appreciate the difference). Deep-frying can raise oxygen to this state, allowing it to oxidize lipids to hydroperoxides, which happens to the exclusion of the formation of trans-LA.
The hydroperoxides also often take the form trans-fatty acids as well as being oxidized. These don’t show up when looking for normal trans-LA but they’re presumably just as harmful.
Once a hydroperoxide has formed, it can catalyse the formation of more hydroperoxides, which is probably why LA (2 double bonds per fatty acid, so 6 per triglyceride molecule) is so eager. LA also has an especially vulnerable-looking carbon between two sets of double bonds.
So what might be the effects of peroxides in diet? The first paper I found contains the phrase “while fish oil had been considered poisonous, in fact, fresh fish oil itself was not poisonous, unlike oxidized fish oil”.[4] So presumably at high enough concentrations they are bad! This paper shows that oxidized oils are pretty bad, say it with me now, IN MICE and also IN HUMAN CELLS IN A DISH, but doesn’t actually give much causal evidence in humans beyond association studies of oxides in serum. Old and sick people have different stuff in their blood, more at 11.
Another paper suggests that dietary lipid peroxides might cause cancer, but doesn’t say much about the sorts of things that seed oil proponents often mention (weight gain, inflammation, etc.)[5]
Nonetheless, if these things are poisonous at high concentrations, they’re probably not great at low concentrations.
Final Thoughts
All of the degradation studies have been done on pure oils, and most likely they were done in an inert (glass) vessel. Many, many things can catalyse oxidation and isomerization reactions, including metal ions, so I suspect cooking with actual food in an actual deep fryer would cause more of both reactions.
Ideally I’d like someone to just sample the McDonald’s cooking oil every few days through a few cycles of oil changes, but I doubt this will happen.
Here’s a chart of fatty acid content:
Olive oil is probably much better than other oils for normal frying for two reasons. Firstly it contains very low PUFAs, secondly it contains anti-oxidant compounds which might be protective against lipid peroxide formation.
Personally, I am not too worried about trans-fat formation in my own kitchen. The amount of formation is probably extremely low at anything resembling a normal cooking time. Unless you re-use cooking oils for days at a time, or cook at extreme temperatures, you’re most likely to be fine.
I am somewhat worried about the formation of lipid peroxides in my own cooking. I normally cook with olive oil or butter anyway, but in the rare case when I deep-fry things I’ll make two changes:
Try to use lower-PUFA oils for deep frying. Peanut/groundnut oil EDIT: Mustard oil might be even better will likely be my primary choice based on the intersection of PUFAs/price/ethics/flavour/availability.
Never re-use deep-frying oil
I think that peanut oil is less likely to cause health problems than other deep-frying oils, since it’s the main deep-frying oil in east Asia, which has much less obesity than the west, even in places like the Philippines where deep fried food is most common.
I am also somewhat concerned about trans-fat formation in commercial cooking, but this is dwarfed by the significant worries about oxidation from commercial cooking which will cash out as eating much fewer shop-bought deep-fried foods like crisps (chips) and chips (fries).
I think the PUFA breakdown hypothesis is more likely than the naive seed oil hypothesis, but I haven’t spent any time looking for counter-evidence yet.
https://pubs.rsc.org/en/content/articlelanding/2019/ra/c9ra00328b
https://www.sciencedirect.com/science/article/pii/S0023643822012282?via%3Dihub
https://core.ac.uk/download/pdf/225888553.pdf
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8062262/
https://www.mdpi.com/2072-6643/12/4/974