In recent years, landfill operators have encountered temperatures in landfills that are well above typical levels. These elevated temperatures are typically accompanied by changes that greatly increase management costs. Some issues include steam emissions and dark, oily surface discharges.

Several workers have proposed smoldering combustion of organic matter as the cause of these conditions. Others have proposed aluminum and aluminum reprocessing wastes reacting with water as the cause of the elevated temperatures. Which could it be?

To understand the processes occurring at these sites, it is useful to consider common symptoms at these sites and compare them with changes expected from hypothetical causes such as aluminum wastes. Several common symptoms of these sites are:

  • waste temperatures that increase convexly with depth to a maximum of 350 F or less then decrease convexly;
  • maximum waste temperatures in saturated or near-saturated waste;
  • hydrogen (H2) in landfill gas;
  • decreased leachate potential of H2 (pH), as low as pH 5;
  • high concentrations of volatile fatty acids (acetic and formic acids) in leachate;
  • very high biochemical oxygen demand (BOD) of leachate;
  • an areal extent of several acres; and
  • increased leachate flows.

We can consider these symptoms along with results expected from potential reactions or biochemical processes to see if they agree. We’ll do this first for smoldering combustion and then for the aluminum processing wastes reacting with water.


Smoldering combustion is a phenomenon that can occur at landfills because of air intrusion, and it is commonly mitigated by cutting off the air that is entering the waste, either through the cover soils or preferential pathways. We can compare the conditions known to occur from smoldering combustion with those observed at these sites to see if they agree.

Smoldering combustion produces red-hot conditions, so temperatures are above approximately 900 F. To our knowledge, the maximum temperatures reported for these sites are below approximately 350 F.

Elevated temperatures from smoldering combustion commonly affect an area of several hundred square feet in a landfill, while these sites have elevated temperatures over areas of tens of thousands of square feet.

Because it is a combustion process, smoldering combustion produces ash and smoke, which commonly are observed in the gas collection system when smoldering combustion is occurring. However, smoke and ash are not observed in the gas collection systems at these sites.

As noted previously, the maximum temperature at these sites is commonly observed in saturated or near-saturated waste. The presence of water is expected to quench the high temperatures of smoldering combustion so that the maximum temperature would be expected to occur in unsaturated waste (where air could penetrate). The graph shows a vertical temperature profile at an elevated temperature site with an inverted triangle showing the approximate depth of leachate-saturated waste well above the maximum measured temperature.

Because smoldering combustion occurs above the saturated waste, it has minimal effects on the leachate. This contradicts the major changes in leachate chemistry observed at these sites.


Clearly, a first factor to consider in assessing whether the reaction of aluminum wastes is the cause of these elevated temperatures would be whether these elevated-temperature conditions have been observed at sites with no record of receiving aluminum processing or aluminum recycling wastes. This is the case. However, a lack of recorded disposal of these substances is not proof of their absence, and other evidence should be evaluated.

Standard chemistry texts can tell us the conditions that would be created from reaction of aluminum processing wastes with water, so we can compare common symptoms of these sites with those expected from reactions of aluminum processing wastes.

The presence of elevated concentrations of H2 are consistent with aluminum oxidation and could have been the reason that aluminum wastes were first considered as a possible cause of the conditions. Aluminum can react with water to produce H2, heat and hydroxide ions.

Heat and H2 are common symptoms at these landfills. The question is whether there are other possible sources of H2 in landfills. H2 is constantly produced biochemically in normal landfills as the first step in methane production through the “carbon dioxide (CO2) reduction” pathway.

Under normal conditions, the H2 would then react with CO2 to produce methane; but, at elevated temperatures, the microbes responsible for the second step become inactive, resulting in the accumulation of H2. At more than one elevated temperature landfill, H2 isotope data have been consistent with this biochemical production of H2 and not with oxidation of aluminum. Studies have shown that H2 in waste can be triggered by elevated waste temperatures, so the H2 observed may be a symptom of elevated temperatures rather than a product of oxidation of aluminum.

The chemical changes in leachate that are typical at these sites are also not consistent with reaction of aluminum processing wastes with water. The oxidation of aluminum (to produce H2) would be expected to increase the pH rather than decrease it. In addition, the reactions of aluminum processing wastes with water would not be expected to result in the drastic increases in BOD and high concentrations of volatile fatty acids typically observed at these sites.

Aluminum processing wastes could potentially produce elevated temperatures in landfills. Waste from recycling of aluminum called “aluminum dross” can generate approximately 2,000 joules of heat per gram in a period of a week or two, which could heat waste by 450 F. So, the heat from these wastes could be more than adequate to reach the temperatures observed, which are up to 220 F higher than normal temperatures.

Considering the amount of heat produced is only part of the story. Heat generated in waste must either migrate away from or be retained in the waste and increase its temperature. It can be suggested that the heat generated by reaction of aluminum wastes migrates away and raises the temperature of the other materials present—usually municipal solid waste. In typical landfills, waste temperatures can range up to 130 F to 150 F, so that the waste is warmer than temperatures outside the waste mass.

This is consistent with slow migration of heat out of the waste in comparison to the relative slow heat generation rate from waste decomposition, so that the much faster heat generation from reaction of aluminum wastes with water would be expected to result in localized temperature increases where the aluminum wastes are located.

Heat-transport calculations suggest that more than 15 years would be required for the localized temperature increases to become more dispersed.

This suggests that the rapid heat generation from aluminum wastes reacting with water would not result in the temperature profiles observed.

The temperature profiles commonly observed at these sites are consistent with textbook equations for heat generation dispersed throughout the waste rather than with localized sources, such as one would expect for deposits of aluminum processing wastes within the waste mass.

The graph below shows an observed vertical temperature profile at an elevated temperature site that is consistent with the general pattern observed at these sites. The graph’s dashed line shows temperatures calculated for a widely distributed heating source as opposed for multiple localized sources, as one would expect from aluminum processing wastes.

Because aluminum wastes reacting with water would consume water, this hypothesis does not explain the increased leachate flows commonly observed at these sites.


A hypothesis for the conditions observed that has not received as wide an audience as smoldering combustion or aluminum processing wastes reacting with water is aqueous pyrolysis.

Biomass is known to undergo aqueous pyrolysis reactions that generate heat through a reaction called cellulose dehydration. Approximately 30 to 40 percent of municipal solid waste is cellulose, so it satisfies the condition of being widely dispersed in the waste that aluminum processing waste does not.

The inverted triangle in the graph above represents the approximate depth of leachate-saturated waste. The dashed line represents temperatures calculated for a widely distributed heating source.

Cellulose dehydration is an aqueous reaction; because of that, it aligns with the observation of the maximum temperature in saturated waste. Cellulose dehydration is expected to create as much as 30 gallons of water per cubic yard of waste.

Aqueous pyrolysis of biomass has been shown to produce an aqueous phase with low pH and high concentrations of volatile fatty acids, similar to leachate at these sites, as well as compounds with potential commercial value.

Aqueous pyrolysis reactions seem to agree with the symptoms observed at these sites. As noted, the hypothesis of aqueous pyrolysis reactions at these sites has not been well-characterized, and further work is needed to assess it.


Although each landfill is unique, for sites with the symptoms described in this article, the conditions likely are not the result of smoldering combustion or aluminum processing wastes reacting with water.

Drastic changes in leachate chemistry and maximum temperatures in saturated waste are not consistent with smoldering combustion as the cause of the conditions observed at these sites.

Decreased leachate pH and increased leachate concentrations of BOD and volatile fatty acids, as well as the vertical temperature profiles consistent with a dispersed heat source, are not consistent with aluminum reprocessing wastes as the heat source.

Aqueous pyrolysis of biomass seems to produce conditions similar to those observed at these sites, but additional work is needed to be certain.

Henry Kerfoot is a principal with Civil and Environmental Consultants Inc., Phoenix, and is an independent consultant in northeast Maryland. He can be contacted via email at hkerfoot@