According to the U.S. Energy Information Administration,

About 19.6 pounds [8.89 kg] of CO2 are produced from burning a gallon of gasoline that does not contain fuel ethanol. About 12.7 pounds [5.76 kg] of CO2 are produced when a gallon [3.8 litre] of pure ethanol is combusted.

I am not a climate change denier, but this just seems like an unreasonable statement without explanation. How is it possible for 5-10 pounds (2.3—4.6 kg) of liquid to produce 20 pounds (9 kg) of gas?

  • 3
    I added SI conversions to your units but something is missing in the conversion of one gallon / 3.8 litre to 5–10 pounds / 2.3–4.6 kg.
    – gerrit
    Commented Jan 24, 2018 at 12:54
  • 2
    Please note: theoretical answers will be deleted on this site.
    – Sklivvz
    Commented Jan 25, 2018 at 0:40
  • 3
    This question can be easily answered "theoretically" in many valid ways on Skeptics. The fact that no one bothers putting in the effort is the issue! (1) the calculations are valid if they are a citation of a competent authority; (2) there are many sites already verifying this claim or similar claims, an answer can cite many of those; (3) it's probably enough to cite basic science in order to answer the question, without calculations, like the law of conservation of mass and the definition of combustion.
    – Sklivvz
    Commented Jan 27, 2018 at 8:37
  • 4
    I've given three CW answers below for the unconvinced. It was trivial to do.
    – Sklivvz
    Commented Jan 27, 2018 at 9:22

5 Answers 5


This answer provides an example on how to address this question by citing basic science. Please ignore any voting on it.

I won't perform any calculation here as they are not on topic on Skeptics, but the confusion seems to arise from the fact that in combustion a fuel (in this case, gasoline) and an oxidant (in this case, oxygen in the air) combine to produce other compounds, among which is carbon dioxide (CO2).

Combustion or burning is a high-temperature exothermic redox chemical reaction between a fuel (the reductant) and an oxidant, usually atmospheric oxygen, that produces oxidized, often gaseous products, in a mixture termed as smoke.

So as you can see, we have two substances combining. Applying the principle of conservation of mass, the output needs to have the same mass as the input, which is more than only the fuel.

The conservation of mass was obscure for millennia because of the buoyancy effect of the Earth's atmosphere on the weight of gases. For example, a piece of wood weighs less after burning; this seemed to suggest that some of its mass disappears, or is transformed or lost. This was not disproved until careful experiments were performed in which chemical reactions such as rusting were allowed to take place in sealed glass ampoules; it was found that the chemical reaction did not change the weight of the sealed container and its contents.

Of course, this only shows that the total mass of the combustion products can be (and usually is, except for some special cases like solid rocket fuels that contain their own oxidizer) greater than the mass of the fuel alone.

Showing that this is also true for just the CO2 (not counting any other combustion products, such as water vapor or soot) produced when gasoline is burned in a car engine would require either direct experimental measurement or non-trivial calculations based on the chemical composition of gasoline and knowledge of the temperatures and air and fuel ratios found in typical modern internal combustion engines. Such calculations would be off-topic for this site, so I will not provide them; if desired, they may be found in textbooks on chemistry or automotive engineering.

Still, even without determining exactly how much CO2 burning a gallon of gasoline will produce, we can at least see that it can be more than the mass of the fuel, since some of the mass will come from oxygen in the air.

  • 3
    I know this is just an example answer, but it best isolates the confusion I had when first asking the question and directly addresses it, apparently within the allowable formats of this SE site. Commented Jan 28, 2018 at 19:38
  • +1 for distinguishing pounds of mass from pounds of weight. Discounting Carbon Monoxide production we are really just counting number of Carbon atoms to arrive at the number of Carbon Dioxide molecules as a product.
    – mckenzm
    Commented Jan 29, 2018 at 18:56
  • So we do not have the same discussion for the millionth time, comments about the site's rules are off topic and they are going to be nuked ruthlessly.
    – Sklivvz
    Commented Jan 30, 2018 at 13:55

This is directly addressed by the US Department of Energy at FuelEconomy.gov:

It seems impossible that a gallon of gasoline, which weighs about 6.3 pounds [2.9 kg], could produce 20 pounds [9.1 kg] of carbon dioxide (CO2) when burned. However, most of the weight of the CO2 doesn't come from the gasoline itself, but the oxygen in the air.


Therefore, to calculate the amount of CO2 produced from a gallon of gasoline, the weight of the carbon in the gasoline is multiplied by 44/12 or 3.7.


We can then multiply the weight of the carbon (5.5 pounds [2.5 kg]) by 3.7, which equals 20 pounds [9.1 kg] of CO2!

So the answer is: Yes.

An article at Slate draws similar conclusions.

  • 3
    Comments are not for extended discussion; this conversation has been moved to chat.
    – Oddthinking
    Commented Jan 25, 2018 at 3:33
  • 1
    Please do not continue discussions here. See the chat link above.
    – Oddthinking
    Commented Jan 26, 2018 at 5:10

It is possible to verify this idea from multiple sources that are independent of the U.S. Government. I'm going to leave verifying the exact numbers in the question out of scope of this answer, for reasons to be explained shortly, and stick to how you can verify from other sources the underlying claim being taken issue with: that the mass of the liquid is less than the mass of the gas.

Some non U.S. Government sources for this are multiple textbooks on thermodynamics. The combustion of petrochemicals is, understandably, a well-documented process in science and engineering, and there are a lot of sources on this.

I pick An introduction to thermodynamics by Y. V. C. Rao (Universities Press, 2004, ISBN 9788173714610) pretty much at random. Professor Rao teaches chemical engineering at IIT Kanpur in Uttar Pradesh, and seems to have little connection to the U.S. government. In chapter 6 ("First Law Analysis of Processes") we can find several slightly indirect confirmations that this is indeed the physical process:

In combustion process, the mass of each element remains the same even though the reactants disappear and new products are formed. […]

In most of the combustion processes, the required oxygen is not supplied as pure oxygen but is supplied as air. […]

An important parameter in the study of fuels is the air–fuel ratio, the ratio of the mass of the air to the mass of the fuel. […]

One of the more basic ideas that we need to verify can also be found stated by Jeff Hartman, an author on motor vehicles from Austin, Texas. In Nitrous Oxide Performance Handbook (MotorBooks International, 2009, ISBN 9781616730529) on page 24 he says:

When something burns, it is combining with oxygen, i.e. oxidizing.

Rao gives the actual chemical equations for such oxidization, and goes into detail about the purity of the fuel and some generalizations that descriptions of the process tend to make.

For more, we turn to Associate Professor C. Lon Enloe at James Madison University, who tells us on page 13 of Physical Science: What the Technology Professional Needs to Know (John Wiley & Sons, 2001, ISBN 9780471360186) about an important physical principle:

You may know that the conservation of mass and energy (E=mc2) is one of the basic laws of nature. The amount of matter and energy never really change. […] When you burn gasoline, you not only release its chemical energy as heat and light, but you also release matter in the form of carbon dioxide, carbon monoxide, carbon, and water. Every bit of the original matter is conserved.

Let's go to Associate Professor of Mechanical Engineering Gregory Nellis of the University of Wisconsin for another part of the whole. On page 854 of Thermodynamics (Cambridge University Press, 2011, ISBN 9781139498180) he tells us:

The reactants consist of fuel and an oxidizer (usually oxygen). The fuel in a combustion can be gaseous […], liquid (e.g. gasoline or ethane) or solid […].

So what the various professors have told us so far is that there is an underlying physical conservation principle, that there are two inputs (the reactants), and that the mass-energy of the the inputs must equal the mass-energy of the outputs.

But you might challenge this on the principle that carbon dioxide is only one of the outputs, and that the conservation principle is the conservation of mass-energy not of mass alone. Maybe the mass of one of the other products is greater, and thus the mass of carbon dioxide is not necessarily greater than the mass of gasoline that has been burned. Maybe some of the mass turns to energy.

Nellis deals with the latter. This is a chemical reaction, not a nuclear one.

Atoms must be conserved even though the mass of each substance involved in the reaction may change.

This leaves the possibility that the masses of the other products (water, carbon monoxide, and carbon, remember) are such that the mass of carbon dioxide alone is less than the mass of the gasoline. For this, let us turn to Professor of Mechanical Engineering Zellman Warhaft of Cornell University, who gives a more detailed introduction to the process in An Introduction to Thermal-Fluid Engineering: The Engine and the Atmosphere (Cambridge University Press, 1997, ISBN 9780521589277) that shows how the chemical equation that preserves atoms inevitably leads to this result, and that starts off with one of the assumptions that Rao calls out:

Let us assume that C8H8 is representative of gasoline. (In fact, gasoline consists of many different kinds of hydrocarbon molecules, but its average molecular weight is close to that of C8H8.) Equation 1.2 shoes that one molecule of C8H8 produces eight molecules of CO2. Because the molecular weight of C8H8 is 114 and that of CO2 is 44, 1kg of our gasoline produces (8 × 44) ÷ 114 = 3.1 kg of CO2. The density of gasoline is about 70 percent that of water […] and because the density of water is 1000 kg/m3, 1 litre (0.001 m3) of gasoline produces 3.1 × 0.7 = 2.17 kg of CO2, or in the U.S. system 1 gallon (3.785 l) of gasoline produces 18.1 lb of carbon dioxide.

A similar worked example, coming to the figure of 8.196kg of CO2 for 1 gallon of gasoline, is given by Professor Tom Holme of Iowa State University and Senior Lecturer Larry Brown of Texas A&M on page 132 of Chemistry for Engineering Students (Cengage Learning, 2005, ISBN 9780534389741).

Why the discrepancy, though? These are not the U.S. government figure.

The answer lies in the textbooks, again. Internal combustion is complex. There are entire books on this specific subject, after all.

As Rao notes, these explanations make simplifying assumptions. They assume that gasoline is a pure substance. They assume complete combustion. In the real world, gasoline is, as Warhaft says, a mixture. Combustion is not complete. After all, one of the subjects that these textbooks go into is the notion of stoichiometric mass and the ideas of "rich" mixtures where there is too little air and not all of the gasoline burns, and "lean" mixtures where there is too much air. Rao again:

The combustion of a hydrocarbon fuel is considered to be complete if all of the carbon and hydrogen contains in the fuel are converted into carbon dioxide and water, respectively. The minimum amount of air required for the complete combustion of fuel is called the stoichiometric or theoretical air. Usually, the combustion process does not go to completion unless the amount of air supplied is more than the theoretical air. The amount of air actually used in a combustion process is expressed in percent theoretical air or percent excess air. The amount of air supplied over and above the theoretical requirement is called excess air.

And indeed Holme and Brown:

We'll need to know the mass of a gallon of gasoline. As you might guess, this depends on the exact composition of the gasoline in question.

So there is a set of assumptions, about how close one's gasoline is to the theoretical pure octane that the textbook calculations (explicitly) assume, and about how much fuel is actually burned in real automobiles, that can affect the answers. However, the underlying point, that the mass of the carbon dioxide is greater than the mass of the gasoline because mass-energy is conserved, there is another reactant in the reaction, and the chemical equation forces the ratios of reactants to products, is there in these and many other sources.

Interestingly, your skepticism has been studied as one of the things that children do not initially understand about chemistry, and start off with the wrong mental models for. Graham Peacock of Sheffield Hallam University, John Sharp of Bishop Grosseteste University College, Rob Johnsey of the University of Warwick, Debbie Wright of the University of Plymouth, and Keira Sewell of Southampton University noted this in their book Primary Science: Knowledge and Understanding (Learning Matters, 2017, ISBN 9781526422736):

Research into the idea that energy is contained in fuels […] argued that the conservation of matter was one of the most important aspects of understanding energy from fuels and food. The key question was: Suppose we trap all of the fumes from the exhaust pipe of a car and were then able to weigh them — we should be able to compare the amount of petrol used with the amount of exhaust […]. Would the mass of exhaust be:

a. much larger than the petrol?

b. about the same as the mass of the petrol?

c. much smaller than the petrol?

Scientists would answer "a" as the process of burning fuel is constructive because the atoms of the fuel are combined with oxygen. It was found that many people believe (incorrectly) that the amount of exhaust will be smaller than the petrol.

For what they are referring to, which itself cites further works and which again is yet another source for verifying the claim being disputed in the question, see pp. 82 et seq. of Michael Littledyke's, Liz Laikin's, and Keith Ross's book Science Knowledge and the Environment (2000, London: Fulton).

Further reading

  • Richard Stone (2012). "Combustion and fuels". Introduction to Internal Combustion Engines. Palgrave Macmillan. ISBN 9781137028297.
  • Kalyan Annamalai and Ishwar K. Puri (2006). Combustion Science and Engineering. CRC Press. ISBN 9780849320712.
  • Kenneth W. Ragland, Kenneth M. Bryden (2011). Combustion Engineering. CRC Press. ISBN 9781420092516.
  • D. Bradley (2012). "Combustion in Gasoline Engines". In P.M. Weaving. Internal Combustion Engineering: Science & Technology. Springer Science & Business Media. ISBN 9789400907492.

… and so on. Entire books on this subject. ☺

  • 1
    FWIW, I had a similar argument with someone about how their statement that "eating a pound of food can't cause you to gain a pound in weight after digestion because it can't add any more mass than it has" with my argument being that they would have to include zero-calories substances such as water into the equation since it would add to the possible weight the food could hold (a classic example being the various protein and diet replacement shakes where the weight of the powder is much less than the resulting foodstuff). Commented Jan 27, 2018 at 21:30

This answer provides an example on how to address this question by citing calculations performed by competent authorities. Please ignore any voting on it.

The calculation is somewhat involved because it requires estimating a number of factors. Fortunately the EPA has done this for us

Gasoline carbon content per gallon: 2,421 grams


The Intergovernmental Panel on Climate Change (IPCC) guidelines for calculating emissions inventories require that an oxidation factor be applied to the carbon content to account for a small portion of the fuel that is not oxidized into CO2

For all oil and oil products, the oxidation factor used is 0.99 (99 percent of the carbon in the fuel is eventually oxidized, while 1 percent remains un-oxidized.)

Finally, to calculate the CO2 emissions from a gallon of fuel, the carbon emissions are multiplied by the ratio of the molecular weight of CO2 (m.w. 44) to the molecular weight of carbon (m.w.12): 44/12.

CO2 emissions from a gallon of gasoline = 2,421 grams x 0.99 x (44/12) = 8,788 grams = 8.8 kg/gallon = 19.4 pounds/gallon

According to Slate the calculation performed by the EIA gives a slightly different number: 19.564 pounds/gallon

Darthmouth university gives us a nice table as referenced from the EIA:

emissions table

Unfortunately the EIA source seems to be really hard to find (404) and I can't verify their claim exactly, but given the other numbers, they are at least in the ballpark.

  • The data in the table states the life-cycle carbon emissions and not the amount of CO2 generated in the moment of combustion. Otherwise the column for B100 (pure biodiesel) wouldn't state "0.00". This means at least some of the figures are derived from theoretical calculations. Indeed, the original source states that the "Emissions factors [are] calculated ..." This might be true also for the figure for "motor gasoline", i.e. it could include life-cycle factors such as transportation, production etc. Commented Jan 30, 2018 at 12:40
  • If that is so, then the citation (although from a "competent authority") doesn't provide the data that we need for a precise answer. We have to take the theoretical context into account to understand what the data means. Commented Jan 30, 2018 at 12:42

This answer provides an example on how to address this question by citing multiple reputable fact checkers and sites. Please ignore any voting on it.

According to many sources, this is true, allowing for some imprecision:

It seems impossible that a gallon of gasoline, which weighs about 6.3 pounds, could produce 20 pounds of carbon dioxide (CO2) when burned. However, most of the weight of the CO2 doesn't come from the gasoline itself, but the oxygen in the air.


[a gallon of gasoline] equals 20 pounds of CO2!

How can a gallon of gasoline produce 20 pounds of carbon dioxide?

Actual weight ratios will of course vary, since gasoline is not purely octane. In general, however, when you burn carbon fuels they are in "reduced" form, that is, the carbons in the molecules are attached mostly to hydrogen atoms. When they are burned, the carbon becomes "oxidized" (literally, combined with oxygen) to make CO2. Since oxygen is far heavier than hydrogen, the product is heavier than what is burned.

What is the weight ratio of CO2 released to fuel burned?

We started by asking people to think about the effects their cars have on the environment: "For each gallon of gas your car burns, it releases about 19 pounds of carbon dioxide." Explainer readers wondered about this statistic: If a gallon of gasoline weighs about 6 pounds, how can it produce three times that much greenhouse gas?

The carbon from the gasoline mixes with oxygen from the air. Gasoline consists mostly of hydrocarbons—chains of carbon encircled by atoms of hydrogen. When the hydrocarbons burn, they break apart and recombine with the air. This reaction produces heat, as well as two chemical byproducts: water and carbon dioxide.

How Gasoline Becomes CO2

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .