You are likely to have come across numbers like sugars 17kJ/g (4kcal/g) as the energy supposedly available to our bodies after eating. Yet these values have been determined using very artificial means, typically lighting up food with a red hot iron in a steel chamber at high oxygen pressure.

Many historic energy values have been proven wrong in the past. For example, salad leaves had zero calories in many historic energy tables before we realized that bacteria in our gut can break them down and we in turn can break down bacterial products, gaining back part of the chemical energy of the plant.

Are you aware of any research that has determined the energy value of different foods in more natural conditions, ideally in humans? I expect some calorie values to change from those determined by the bomb calorimeter, don't you?


The device you refer to is a bomb calorimeter. Inside one of these, a sample of material is completely combusted. Typically, the heat from the combustion warms a measured mass of water and the rise in temperature is used to calculate the Calories/Joules in the material.

Of course our cells don't ignite food and burn it to completion. Even if they did, some of the heat from the process would escape collection. Cells use the process of cellular respiration to harvest the energy. The basic version of cellular respiration describes the utilization of glucose to create ATP, by far the most used substance to power the energy needs of the cell. Here's just one source that takes you through the calculations of how efficient this process is: http://www.tiem.utk.edu/~gross/bioed/webmodules/ATPEfficiency.htm

So, at least for the very common sugar glucose, only about 40% of the total Joules, as measured in the calorimeter, end up as ATP. Most other food substances we use for energy share much of the same chemical pathway as glucose. I haven't seen calculations of their efficiency but suspect it would be roughly similar.

One of the most basic and important lays of physics is that no transfer of energy is 100% efficient. (Read more here: https://en.wikipedia.org/wiki/Energy_conversion_efficiency ). The remaining energy in the glucose that was not converted to energy in ATP is "lost" as heat. By lost, it is meant that it's not avaible to do any useful work other than heat us up. Of course, being homeothermic ("warm blooded") animals, that's useful too.

  • $\begingroup$ The U of Tennessee link you sent is great. They propose a very simple calculation that gives a good first approximation. Energy captured in ATP divided by theoretically available energy from glucose combustion ~ 50% after correcting for concentrations. $\endgroup$ – SeanJ Sep 30 '16 at 19:49
  • $\begingroup$ But that's does not yet take into account the energy required to metabolize glucose. Most of us get our glucose from polymers, so we need to make enzymes to break them down. Then we need to import and ship it to the cells that do the chopping; more energy required. Then in the cell that gets that molecule of glucose, the glycolytic enzymes don't come for free, right? What would be left of the 50% efficiency estimate when all these expenses are taken into account? 25%, 10%,..? $\endgroup$ – SeanJ Sep 30 '16 at 19:49
  • $\begingroup$ It's true there are a few steps in glycolysis that require energy but, in subsequent steps, this investment is returned. The calculations include this, they are a net value of the process. As to parsing out the energy required to produce the enzymes involved, and to manufacture mitochondria to govern most of this energy harvesting, I've never heard of anyone attempting to calculate this. All of that is generally included under the category of "what we need the energy for". $\endgroup$ – bpedit Sep 30 '16 at 20:47

I think there are two different things you are asking about.

The first is the standard enthalpy of combustion, which is indeed measured more or less as you describe. There's absolutely no problem with that, because the total energy released by the metabolism of glucose to CO2 is exactly the same as released by combustion. On the way down the energy slope you might even use some of it to make ATP and actually do some "work". There might be some small corrections needed (you run at 37C and not at 25C for example) but no matter which way you take, turning glucose and oxygen in carbon dioxide and water will release the same amount of energy. And remember, we need a lot of this energy just to keep warm, so the waste heat in metabolism is not really a waste.

The second part is a bit more complex, when you're talking about food. Here the bomb calorimeter part is a bit misplaced, because most foodstuffs will be analyzed using other techniques to determine the amount of carbohydrate, fat and protein. Then they'll do a calculation to get the number of calories in your food. This should get you quite close to the real number of calories, as stuff that's not digestable by humans is not counted. Regarding the salad bacteria: I'm not sure if they put it on the labels, but we know whether or not certain fibers will be metabolized, and you could take those into account as well if you wanted. It wouldn't make much of a difference though, almost all of the fiber we consume is coming out the other end.

If you want to know more, you're getting into a swamp of definitions and standards. For example you could skim this FAO report: http://www.fao.org/docrep/006/y5022e/y5022e03.htm#TopOfPage (chapter 2 METHODS OF FOOD ANALYSIS) http://www.fao.org/docrep/006/y5022e/y5022e04.htm#TopOfPage (chapter 3 CALCULATION OF THE ENERGY CONTENT OF FOODS - ENERGY CONVERSION FACTORS)

This shows that people are working hard on the problem you're posing, using a wide variety of approaches. It shows that every method has its problems, but also that all the values are quite close to eachother.

  • $\begingroup$ Hi VonBeche, sure the enthalpy of combustion of glucose is per definition constant, but it's of course more interesting to know what fraction of that potential can be captured by our bodies for its own purposes as in ATP, NADP, and the like. It also probably varies between individuals how efficient this capture is. And it probably even varies in the same person over time. I'd venture a guess that we recover more from food if we are fasting than if we have just had too far much to eat. I'll check out your links. Thanks. $\endgroup$ – SeanJ Sep 30 '16 at 19:39
  • $\begingroup$ That's the 40% of the other answer, and also that will be roughly constant (there are only so many pathways). You might just "burn up" the ATP in some cases (in brown fat for example), but the glucose isn't going anywhere without your body getting all the products from glycolysis (or whatever pathway is used). $\endgroup$ – VonBeche Sep 30 '16 at 19:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.