| rate of heat transfer to water | |
|---|---|
| rated power output according to manufacturer | |
| mass (water in this case) | |
| heat capacity of water | |
| rate of temperature change | |
| burner efficiency |
Photo credit to tcameliastoian
I had never thought about the dangers of gas stoves until my mom sent me this Atlantic article arguing that all gas stoves should be replaced because of the undetectable toxic gases they emit. My initial skeptical instinct was to brush it off as hyperbole so I did a quick google scholar search to debunk it. Much to my surprise, despite the sensational tone of the article, it appeared to be onto something.
About a week before the conversations that began this article, I replaced my broken kitchen stove/oven. It was a relief to realize I had made a good decision going with a mid-line electric model. At the time I had not realized there were any health implications to the decision so my apparent wisdom in selecting an electric stove was entirely accidental.
Are the products of combustion resulting from gas stoves dangerous? Obviously, proper functioning gas stoves do not cause acute toxicity in normal humans. However, studying and proving a chemical is toxic under low-dose chronic exposure is difficult. In this article, we will discuss the hazards of gas stove use, primarily carbon monoxide and nitrogen dioxide as well as the trade-offs between gas and electric stoves in their power output and cost of operation.
Natural gas is a combustible gas containing a number of hydrocarbons, primarily methane. Its composition depends on the location from which it is extracted [1] but it is processed to more uniform composition prior to use. Pipeline quality natural gas is predominantly methane (>85%) (\(\require{mhchem} \ce{CH4}\)), with <5% ethane (\(\require{mhchem} \ce{C2H6}\)), propane (\(\require{mhchem} \ce{C3H8}\)), butane and iso-butane (\(\require{mhchem} \ce{C4H10}\)), and pentane and iso-pentane (\(\require{mhchem} \ce{C5H12}\)). Additionally, some residential natural gas contains trace oxygen and nitrogen [2].
The combustion process of natural gas (which we will assume is 100% methane for simplicity) results in several reactions. Considering ethane and higher molecular weight hydrocarbons in natural gas does not introduce any other significant species of toxic gas. The methane, \(\require{mhchem} \ce{CH4}\), is the natural gas. Nitrogen and oxygen are the dominant parts of Earth's atmosphere (about 80% and 20%, respectively).
Of the six reaction products, the first two are most desirable. Carbon dioxide is relatively non-toxic compaired to the other reaction products. Water, \(\require{mhchem} \ce{H2O}\), is produced as steam which is also nontoxic.
The third product, carbon monoxide, is a poison with largely acute hazards. Chronic exposure to low-concentration \(\require{mhchem} \ce{CO}\) is a part of life but should be limited as much as possible. If the fuel air mixture on the stove is wrong, carbon monoxide can become a major acute problem, sometimes causing fatalities.
Of the next two products, the \(\require{mhchem} \ce{NO}_x \) compounds, \(\require{mhchem} \ce{NO2} \) (nitrogen dioxide) is the more hazardous and the subject of most of this article. (\(\require{mhchem} \ce{NO}_x \) refers to all oxides of nitrogen, \(\require{mhchem} \ce{NO},\ce{NO2},\ce{N2O},\ce{N2O3},\ce{N2O5}\), etc., but only the first two are major products of combustion under Earth's atmosphere [3].) Both gases are poisons at sufficiently high concentrations and the legal exposure limits are extremely low. Whether a fire is hazardous is a question of whether there is sufficient ventilation to prevent the build up of dangerous levels of these compounds.
The final product is elemental carbon, \(\require{mhchem} \ce{C(s)}\). This is commonly known as soot. It sticks to cookware and makes a black, sticky mess. In restaurants, it is commonly removed with Carbon-Off!, a mix of aggressive organic solvents. Soot is produced by burning hydrocarbons in the presence of insufficient oxygen. However, the presence of soot usually indicates the presence of carbon monoxide. Both are created by insufficient oxygen and \(\require{mhchem} \ce{CO}\) is much more dangerous. Soot is undesirable for practical reasons and health reasons but it is largely an indicator of other, less visible and more dangerous side products.
Carbon monoxide is a deadly, colorless, odorless gas [4,5]. It is often said by educated folks that carbon monoxide kills by binding tightly to the hemoglobin in the blood and displacing the oxygen [6]. This mechanism was first proposed in 1857 by Claude Bernard [4,7]. There has been no doubt for 100 years that this is part of the mechanism of acute fatal carbon monoxide poisoning [8,9] but it is not the only mechanism. Dogs breathing 13% carbon monoxide die within about an hour with Hb-CO concentrations between 54% and 90%. However, giving dogs a blood transfusion with Hb-CO until Hb-CO blood concentrations reach 50-60% causes no ill effects [10]. In 1975, Institutional Animal Use and Care Committees (the ethics committees who decide which studies are allowed to be conducted on animals) were less restrictive than they are today. While the complete pathogenesis is beyond the scope of this article [5], it does partially explain the hazards of chronic, low-concentration exposure.
Chronic carbon monoxide poisoning is often missed in clinical visits, or only presents as subclinical cases. It is an ongoing area of concern and study in the medical profession, particularly in the old, the young, and the pulmonarily infirm [11,12,13,14].
Accidental carbon monoxide poisoning is common, particularly among outdoorsie people. Reading the literature gives the impression that everyone is dying of carbon monoxide in their tents [15,16,17,18,19]. It is so common doctors have written review papers all about carbon monoxide in tents [17], how different fuels affect carbon monoxide in tents [20], and how to get more carbon monoxide into your tent [21]. The world probably was not improved by that last study. Stay out of tents, those things are dangerous. And keep your stove out of your tent too.
OSHA requires ambient \(\require{mhchem} \ce{CO} \) remain below 50 ppm for an 8 hour work day. Workers must evacuate immediately from spaces with more than 100 ppm of \(\require{mhchem} \ce{CO} \) [22,23]. The Environmental Protection Agency standards are more conservative at less than 9 ppm for 8 hours or 25 ppm for 1 hour [24].
Nitrogen dioxide, \(\require{mhchem} \ce{NO2} \), is a reddish-brown gas with a sharp smell reminescent of bleach [25]. However, gas stoves produce such low concentrations, it will never be detectable by smell or color. Nitrogen dioxide causes pulmonary symptoms. If you have enough, you can give emphysema to a rabbit [26], presumably other mammals too (but there are no studies on other mammals). We have been aware of its acute toxicity since the first case report in 1804 [27] when a French merchant and his dog lost their lives. It is now an infamous, widely-translated, medical case report [28].
The deleterious effect of nitrogen dioxide on pediatric asthma is well documented in randomized controlled trials [29,30], survey studies [31,32,33] and meta analyses of those trials [34]. At concentrations as low as 6 ppb, nitrogen dioxide can affect pediatric asthma [33]. Other studies indicate the effect is not limited to children [35,36] but the effect on adult asthma is less studied. In susceptible populations, exposures as low as 260 ppb for 30 minutes can compromise lung function [36].
Lung diseases fall into two broad categories: obstructive (difficulty moving air in), and restrictive (difficulty moving air out). Chronic obstructive pulmonary disease (COPD) is the more common variety, particularly among cigarette smokers. COPD is also a significant risk factor for nitrogen dioxide sensitivity [37,38].
The WHO guidelines admit that there is some conflicting evidence [39]. However, studies I have found that conclude nitrogen dioxide poses no threat are invariably myopic. They usually involve healthy young adults, use doses that are low in concentration or duration, and then measure some trivial proxy variable like blood flow to the forearm. If the blood flow to the forearm is unchanged, nitrogen dioxide is safe [40]. This is an oversimplified view of this study but certainly the view one would have to take to argue it shows nitrogen dioxide is safe. The authors would correctly say their study shows nitrogen dioxide does not affect short-term vasoconstriction, not that it proves the long-term safety of anything.
The European Union long term exposure limit for nitrogen dioxide is 200 ppb [41], usually mentioned in the context of transportation pollution. The World Health Organization recommends long term exposures not exceed 20 ppb [30]. In the US, OSHA recommends no short term exposures exceed 5,000 ppb for any length of time. Average exposures over a 40 hour week must be limited to 200 ppb [42]. The EPA rules for safe housing set 1 hour exposure limits at 100 ppb and average of 53 ppb year round [43]. Throughout this article, we will use 100 and 20 ppb as benchmark numbers for safe exposure since they are the most conservative.
Nitric oxide, \(\require{mhchem} \ce{NO}\), is considered safe at the levels generated by a gas stove. Gas stoves typically do not cause nitric oxide concentrations to rise above 1 ppm, not even briefly [44,45]. In the US, the Occupational Safety and Health Administration and the American Conference of Governmental Industrial Hygienists recommend 8 hour exposures remain under 25 ppm [46]. I have not found any studies indicating nitric oxide is hazardous at concentrations likely to result from gas stove use.
Most of the known danger from gas stoves is though to be the result of nitrogen dioxide. Low-dose carbon monoxide is somewhat hazardous but the effects of long-term low-dose carbon monoxide are not fully understood.
Gas ovens can produce carbon monoxide concentrations in excess of 100 ppm [47]. However, most model kitchen experiments show concentrations well below 10 ppm [45]. In properly functioning stoves, there is insufficient evidence to indicate the low-level carbon monoxide production is hazardous. However, many stoves are not properly functioning. Malfunctioning stove hazards are beyond the scope of this article.
[Caption] \(\require{mhchem} \ce{CO} \) vs time plot in a model residential kitchen with a gas stove. Model kitchen was 3 x 4 m with 2.3 m ceiling and a ventilation fan removing air at 90 m3/hr. Average residence time for air in the kitchen was 18 minutes. Adapted from [45]. Here peak levels are less than half the most restrictive exposure guidelines I could find. However, this experiment was done with extremely high air turnover which is unlikely to exist in residential kitchens. Other sources [47] indicate \(\require{mhchem} \ce{CO} \) can rise to dangerous levels from a gas stove but I have not been able to find time-resolved data showing this.
The primary reason carbon monoxide is less hazardous than nitrogen dioxide is because it can be economically detected near the toxic threshold. Home \(\require{mhchem} \ce{CO} \) meters, while typically not required by building codes, are widely available at home improvement stores for $50-$100. Standard smoke detectors are not carbon monoxide sensors. If you suspect for any reason, that any gas appliance is malfunctioning, a $100 carbon monoxide detector is well worth the piece of mind. Gas stoves are more vulnerable to malfunctioning than other appliances so if you have one, you should also have a carbon monoxide detector nearby.
Nitrogen dioxide levels are the primary point of concern. Since at least 1980, we have known that average, ambient \(\require{mhchem} \ce{NO2} \) levels in the home are elevated by a factor of five [48,49] (4-16 ppb to 30-120 ppb) and that it likely caused some pulmonary problems at ambient levels [49,50,51]. During as little as 15 minutes of cooking, with a ventilation fan running, levels can rise from less than 5 ppb to 120 ppb. Levels of \(\require{mhchem} \ce{NO2} \) are highly variable depending on floor plan, ventilation, cooking style (Asian restaurant workers are exposed to more nitrogen dioxide than any others [52,53]), stove model, and stove output power. All the studies I found which simulate or measure nitrogen dioxide during cooking found levels between "mildly concerning" and "extremely alarming."
[Caption] Sample \(\require{mhchem} \ce{NO2} \) vs time plots in residential kitchens with gas stoves. Plot (a) Adapted from Figure 1A in [45]. Model kitchen 3 x 4 m with 2.3 m ceiling and a ventilation fan removing air at 90 m3/hr. Average residence time for air in the kitchen was 18 minutes. Nitrogen dioxide levels in a kitchen exceeded EPA safe 1 hour exposure levels in less than 15 minutes. Most US kitchens do not have forced ventilation systems and therefore see much higher concentrations of combustion products. Plot (b) from Figure 2 in [44]. Kitchen in residential home, 117 m2, closed-floor plan and no ventilation. With only passive circulation to remove the off-gases from the stove/oven, concentrations rise to much higher levels.
There is a small seasonal variation (30-50%) in \(\require{mhchem} \ce{NO2} \) as we expect any ventilation-dependent process to have [54]. Air turn over is lower in the winter as colder temperatures cause people to shut windows and doors.
[Caption] Average \(\require{mhchem} \ce{NO2} \) levels in the kitchen microenvironment from [54]. Homes sampled were located in Northern Europe so there was a large difference in ventilation between the summer and winter months.
Ventillation systems are the least effective way to reduce average nitrogen dioxide levels in the home. Carbon filter air purifiers are more effective than ventilation but the best option by far is to replace the gas stove with an electric model [55]. Ventillation is an effective way to reduce peak concentrations [45] but not average concentrations [55].
Unfortunately, unlike carbon monoxide, nitrogen dioxide cannot be measured accurately and economically in the concentration range of interest for long-term health. Sensors which are available typically measure no less than 0.1 ppm putting the region of interest at the edge of their detection limits.
In general, gas stoves output more power than electric models. The industry has decided to list the output powers in different units to prevent direct comparison. It is more fair than it sounds because the efficiencies are also different so directly comparing the output power will not give a direct comparison of stove performance. The power output on a gas stove ranges from about 2,640 W (9,000 BTU/hr) on the cheapest stove at Lowes to 8,800 W (30,000 BTU/hr) on commercial restaurant style units. Electric stoves range from 2,100 W (7,000 BTU/hr) (cheapest unit at Lowes) to 2,650 W (9,000 BTU/hr) (Garland restaurant model). Interestingly, the electric stove in my kitchen puts out 3,000 W (10,000 BTU/hr), 10% more than the restaurant style stoves I glanced through.
Electric stoves more efficiently transfer power to the food. Gas stoves heat more of the surrounding environment since the combustion process moves heat out from under the cookware via convection. The important parameter here is the rate of heat delivery to the food. I measured this on a few friends' stoves using the same pot and volume of water. The best gas stoves performed about as well as the worst electrics.
Start with an energy balance, where \(\dot{Q}\) is the rate of heat entering the water in the pot (in J s-1 or W), \(m\) is the mass of the water in the pot (in this experiment 2 kg), \(C_p\) is the heat capacity of the water (4184 J kg-1 °C-1), and \(\frac{dT}{dt}\) is the rate of temperature change (the slope of the lines in the left graph).
The efficiency of the stove is then computed by the ratio of the measured heating rate to the heating rate claimed by the manufacturer. We expect this to be less than one because the manufacturer arrived at this number by measuring how many watts of electricity or natural gas travel through each burner.
[Caption] Heating rate on several friends' stoves. Temperature measurements every minute. Efficiency was computed using the manufacturer's stated heat output and the measured heat output in the left graph. I was unable to determine the efficiency of several stoves because I could not find model numbers. The large fast burners on gas stoves have a similar heating rate to the smaller burners on an electric stove. For me this disproves the myth that gas stoves heat faster. They may have high energy output ratings but they are less efficient. As the gases escape from under the bottom of the pot carrying heat into the kitchen air, they do nothing to move heat into the pot and food.
Keep in mind, if the regulators in an electric stove are dying, they may partially fail in a way that dramatically lowers the output power. Some negative experiences with electric stoves are caused by old and failing regulators, not the stoves themselves.
Let us assume that electricity is about $0.10 per kilowatt-hour (US EIA) and natural gas is $1.00 per Ccf (100 cubic feet) (US EIA). Natural gas has 100,000 BTUs per Ccf according to the US EIA. Clearly this calculation will change with energy prices.
[Caption] Operating cost of a single, typical, gas burner.
[Caption] Operating cost of a single, typical, electric burner.
Even if the stove was on full blast for an hour a day, there is less than $3 per month difference between the gas and electric stoves. The price difference between gas and electric stoves exists but the electric stoves are cheaper to purchase upfront ($440 vs $370 for the cheapest gas and electric stoves at Lowes, respectively).
I cannot tell you that your choice of stove is going to kill you. Lots of people have used gas stoves for a long time. It is difficult, bordering on impossible, to prove the toxicity of a low dose, chronic exposure of anything. It took many decades of study to prove that cigarettes are a protracted suicide of lifestyle. I doubt we will look at gas stoves in the future the way we look at cigarettes today. However, there is clear evidence that no one is doing themselves any favors by living with a gas stove and for those with asthma or COPD in the family, it is an active disservice.
Given there is no price difference, I am convinced I will never have a gas stove in my home ever again.
[1] "The occurrence of helium in natural gas and the composition of natural gas," Journal of the American Chemical Society, vol. 29, pp. 1523—1536, 1907.
[2] "Overview of the Natural Gas Industry," in Fundamentals of natural gas processing, 2006.
[3] "Diesel engine exhaust emissions," in Handbook of diesel engines, pp. 445, 2010.
[4] "Carbon monoxide poisoning," Journal of the Royal Society of Medicine, vol. 94, pp. 270—272, 2001.
[5] "Carbon monoxide poisoning," Emergency Medicine Clinics, vol. 22, pp. 985—1018, 2004.
[6] "Review of Carbon Monoxide Poisoning.," Public Health Bulletin, 1930.
[7] Le Cons sur les effets des substances toxiques et médicamenteuses, 1857.
[8] "The effect of carbon monoxide on the oxyhemoglobin dissociation curve," American Journal of Physiology, vol. 141, pp. 17—31, 1944.
[9] "The gasometric estimation of the relative affinity constant for carbon monoxide and oxygen in whole blood at 38C," American Journal of Physiology, vol. 90, pp. 511—2, 1929.
[10] "What is the mechanism of carbon monoxide toxicity?" Aviation, Space, and Environmental Medicine, vol. 46, pp. 1289—1291, 1975.
[11] "Epidemiology, pathophysiology, clinical evaluation, and treatment of carbon monoxide poisoning in child, infant, and fetus," Northern Clinics of Istanbul, vol. 4, pp. 100, 2017.
[12] "Chronic and occult carbon monoxide poisoning: we don't know what we're missing," Emergency Medicine Journal, vol. 19, pp. 386—390, 2002.
[13] "Carbon monoxide poisoning-a public health perspective," Toxicology, vol. 145, pp. 1—14, 2000.
[14] "Carbon monoxide poisoning: Early awareness and intervention can save lives," Postgraduate Medicine, vol. 105, pp. 39—50, 1999.
[15] "Carbon monoxide poisoning during camping in Korea," Inhalation Toxicology, vol. 28, pp. 719—723, 2016.
[16] "Silent killer-the dangers of carbon monoxide poisoning during international travel," Journal of Travel Medicine, vol. 24, 2017.
[17] "Carbon monoxide poisoning while using a small cooking stove in a tent," The American Journal of Emergency Medicine, vol. 22, pp. 204—206, 2004.
[18] "Carbon monoxide poisoning from a cooking stove in a tent," The American Journal of Emergency Medicine, vol. 23, pp. 205—206, 2005.
[19] "Carbon monoxide exposure in mountaineers on Denali," Alaska Medicine, vol. 30, pp. 85—90, 1988.
[20] "Effect of fuel type on carbon monoxide accumulation in tents of varied design," Wilderness & Environmental Medicine, vol. 24, pp. 280—284, 2013.
[21] "Inducing high levels of carbon monoxide in a tent," The American Journal of Emergency Medicine, vol. 23, pp. 204, 2005.
[22] "Title 29 Standard 1917.24 Carbon Monoxide," 1997.
[23] "Carbon monoxide poisoning fact sheet," 2012.
[24] "Revised evaluation of health effects associated with carbon monoxide exposure: an addendum to the 1979 EPA air quality criteria document for carbon monoxide. United States Environmental Protection Agency, Washington, DC," Environmental Protection Agency, 1979.
[25] "Household and Industrial Products," in Clinical Veterinary Toxicology, pp. 139—176, 2004.
[26] "Nitrogen dioxide-induced emphysema in rabbits," American Review of Respiratory Disease, vol. 95, pp. 797—805, 1967.
[27] "Sur une morte prompte occasionné par le gaz nitreux," J Med Chir Pharm, vol. 8, pp. 487—505, 1804.
[28] "The first death from nitrogen dioxide fumes: the story of a man and his dog," JAMA, vol. 229, pp. 1181—1182, 1974.
[29] "The respiratory health effects of nitrogen dioxide in children with asthma," European Respiratory Journal, vol. 38, pp. 303—309, 2011.
[30] "Effect of indoor nitrogen dioxide on lung function in urban environment," Environmental Research, vol. 138, pp. 8—16, 2015.
[31] "24-h Nitrogen dioxide concentration is associated with cooking behaviors and an increase in rescue medication use in children with asthma," Environmental Research, vol. 159, pp. 118—123, 2017.
[32] "Traffic-related pollution and asthma prevalence in children. Quantification of associations with nitrogen dioxide," Air Quality, Atmosphere & Health, vol. 7, pp. 459—466, 2014.
[33] "Household levels of nitrogen dioxide and pediatric asthma severity," Epidemiology, vol. 24, pp. 320, 2013.
[34] "Meta-analysis of the effects of indoor nitrogen dioxide and gas cooking on asthma and wheeze in children," International Journal of Epidemiology, vol. 42, pp. 1724—1737, 2013.
[35] "Repeated nitrogen dioxide exposures and eosinophilic airway inflammation in asthmatics: a randomized crossover study," Environmental Health Perspectives, vol. 122, pp. 850—855, 2014.
[36] "Nitrogen dioxide exposure enhances asthmatic reaction to inhaled allergen in subjects with asthma," American Journal of Respiratory and Critical Care Medicine, vol. 155, pp. 881—887, 1997.
[37] "Exposure to nitrogen dioxide and chronic obstructive pulmonary disease (COPD) in adults: a systematic review and meta-analysis," Environmental Science and Pollution Research, vol. 25, pp. 15133—15145, 2018.
[38] "The effects of air pollution and temperature on COPD," COPD: Journal of Chronic Obstructive Pulmonary Disease, vol. 13, pp. 372—379, 2016.
[39] "WHO guidelines for indoor air quality: selected pollutants," 2010.
[40] "Exposure to nitrogen dioxide is not associated with vascular dysfunction in man," Inhalation Toxicology, vol. 22, pp. 192—198, 2010.
[41] "Impact of primary NO2 emissions at different urban sites exceeding the European NO2 standard limit," Science of The Total Environment, vol. 646, pp. 1117—1125, 2019.
[42] "Nitrogen dioxide," 2018.
[43] "Primary National Ambient Air Quality Standards (NAAQS) for Nitrogen Dioxide," 2018.
[44] "Pollutant concentrations and emission rates from natural gas cooking burners without and with range hood exhaust in nine California homes," Building and Environment, vol. 122, pp. 215—229, 2017.
[45] "Air quality measurements in a model kitchen using gas and electric stoves," Proceedings of Healthy Buildings, vol. 1, pp. 567—572, 2000.
[46] "Nitric oxide in workplace atmospheres," 1991.
[47] "Exposure Guidelines for Residential Indoor Air Quality - A Report of the Federal-Provincial Advisory Committee on Environmental and Occupational Health," 1989.
[48] "Average N02 concentrations in dwellings with gas or electric stoves," Atmospheric Environment, vol. 11, pp. 869—872, 1977.
[49] "The relation between respiratory illness in primary schoolchildren and the use of gas for cooking: II-Factors affecting nitrogen dioxide levels in the home," International Journal of Epidemiology, vol. 8, pp. 339—346, 1979.
[50] "Health effects of inhalation of ambient concentrations of nitrogen dioxide," American Review of Respiratory Disease, vol. 120, pp. 281—292, 1979.
[51] "Health consequences of nitrogen dioxide exposure," Energy and the Environment, pp. 261, 1979.
[52] "Exposure to particles, polycyclic aromatic hydrocarbons, and nitrogen dioxide in Swedish restaurant kitchen workers," Annals of Work Exposures and Health, vol. 61, pp. 152—163, 2017.
[53] "Respiratory health and lung function in Chinese restaurant kitchen workers," Occupational and Environmental Medicine, vol. 68, pp. 746—752, 2011.
[54] "Seasonal variation of indoor and outdoor air quality of nitrogen dioxide in homes with gas and electric stoves," Environmental Science and Pollution Research, vol. 23, pp. 17784—17792, 2016.
[55] "Home interventions are effective at decreasing indoor nitrogen dioxide concentrations," Indoor Air, vol. 24, pp. 416—424, 2014.