Spilled oil naturally disperses through storms, currents, and osmosis with the passage of time. Chemical dispersants accelerate the dispersal process, although they may have significant side-effects. Corexit EC9500A and Corexit EC9527A have been the principal dispersants employed. These contain propylene glycol, 2-butoxyethanol, and dioctyl sodium sulfosuccinate. 2-butoxyethanol was identified as a causal agent in the health problems experienced by cleanup workers after the 1989 Exxon Valdez oil spill. Warnings from the Hazardous Substance Fact Sheet for 2-Butoxyethanol include "Cancer Hazard: 2-Butoxy Ethanol may be a carcinogen in humans since it has been shown to cause liver cancer in animals. Many scientists believe there is no safe level of exposure to a carcinogen" and "Reproductive Hazard: 2-Butoxy Ethanol may damage the developing fetus. There is limited evidence that 2-Butoxy Ethanol may damage the male reproductive system (including decreasing the sperm count) in animals and may affect female fertility in animals".
Environmental groups attempted to obtain information regarding the composition and safety of ingredients in Corexit through the Freedom of Information Act but were denied by the EPA. After Earthjusticesued on behalf of the Gulf Restoration Network and the Florida Wildlife Federation, the EPA released a list of all 57 chemicals in the 14 dispersents on the EPA's National Contingency Plan Product Schedule. In August 2011, Earthjustice and Toxipedia released an analysis of the 57 chemicals, stating that "5 chemicals are associated with cancer; 33 are associated with skin irritation from rashes to burns; 33 are linked to eye irritation; 11 are or are suspected of being potential respiratory toxins or irritants; 10 are suspected kidney toxins; 8 are suspected or known to be toxic to aquatic organisms; and 5 are suspected to have a moderate acute toxicity to fish.”
Corexit manufacturer Nalco states that "[COREXIT 9500] is a simple blend of six well-established, safe ingredients that biodegrade, do not bioaccumulate and are commonly found in popular household products.... COREXIT products do not contain carcinogens or reproductive toxins. All the ingredients have been extensively studied for many years and have been determined safe and effective by the EPA". According to the OSHA-required Material Safety Data Sheets (MSDSs) for both versions of Corexit used in the Gulf, "Component substances have a low potential to bioconcentrate" (or bioaccumulate), defined by the EPA as "accumulation of a chemical in tissues of a fish or other organism to levels greater than in the surrounding medium". The data sheets further state: "No toxicity studies have been conducted on this product".
Corexit EC9500A and EC9527A are neither the least toxic, nor the most effective, among the Environmental Protection Agency approved dispersants. They are also banned from use on oil spills in the United Kingdom. Twelve other products received better toxicity and effectiveness ratings, but BP says it chose to use Corexit because it was available the week of the rig explosion. Critics contend that the major oil companies stockpile Corexit because of their close business relationship with its manufacturer Nalco.
On 1 May, two military C-130 Hercules aircraft were employed to spray oil dispersant. On 7 May, Secretary Alan Levine of the Louisiana Department of Health and Hospitals, Louisiana Department of Environmental Quality Secretary Peggy Hatch, and Louisiana Department of Wildlife and Fisheries Secretary Robert Barham sent a letter to BP outlining their concerns related to potential dispersant impact on Louisiana's wildlife and fisheries, environment, aquatic life, and public health. Officials requested that BP release information on their dispersant effects. The Environmental Protection Agency later approved the injection of dispersants directly at the leak site, to break up the oil before it reaches the surface, after three underwater tests. Independent scientists suggest that underwater injection of Corexit into the leak might be responsible for the oil plumes discovered below the surface. However, National Oceanic and Atmospheric Administration administrator Jane Lubchenco said that there was no information supporting this conclusion, and indicated further testing would be needed to ascertain the cause of the undersea oil clouds. By 12 July, BP had reported applying 1,070,000 US gallons (4,100 m3) of Corexit on the surface and 721,000 US gallons (2,730 m3) underwater (subsea). The same document listed available stocks of Corexit which decreased by over 965,000 US gallons (3,650 m3) without reported application, suggesting either stock diversion or unreported application. Under reported subsea application of 1,690,000 US gallons (6,400 m3) would account for this discrepancy. Given the suggested dispersant to oil ratio between 1:10 and 1:50, the possible use of 1,690,000 US gallons (6,400 m3) in subsea application could be expected to suspend between 400,000 barrels (64,000 m3) to 2,000,000 barrels (320,000 m3) of oil below the surface of the Gulf.
On 19 May, the Environmental Protection Agency gave BP 24 hours to choose less toxic alternatives to Corexit from the list of dispersants on the National Contingency Plan Product Schedule, begin applying the new dispersant(s) within 72 hours of Environmental Protection Agency approval or provide a detailed reasoning why the approved products did not meet the required standards. On 20 May, US Polychemical Corporation reportedly received an order from BP for its Dispersit SPC 1000 dispersant. US Polychemical said that it could produce 20,000 US gallons (76 m3) a day in the first few days, increasing up to 60,000 US gallons (230 m3) a day thereafter. Also on 20 May, BP determined that none of the alternative products met all three criteria of availability, toxicity, and effectiveness. On 24 May, Environmental Protection Agency administrator Jackson ordered the Environmental Protection Agency to conduct its own evaluation of alternatives and ordered BP to scale back dispersant use. According to analysis of daily dispersant reports provided by the Deepwater Horizon Unified Command, before 26 May, BP used 25,689 US gallons (97.24 m3) a day of Corexit. After the EPA directive, the daily average of dispersant use dropped to 23,250 US gallons (88.0 m3) a day, a 9% decline. By 30 July, more than 1,800,000 US gallons (6,800 m3) of dispersant had been used, mostly Corexit 9500.
On 31 July, Rep. Edward Markey, Chairman of the House Energy and Environment Subcommittee, released a letter sent to National Incident Commander Thad Allen, and documents revealing that the U.S. Coast Guard repeatedly allowed BP to use excessive amounts of the dispersant Corexit on the surface of the ocean. Markey's letter, based on an analysis conducted by the Energy and Environment Subcommittee staff, further showed that by comparing the amounts BP reported using to Congress to the amounts contained in the company's requests for exemptions from the ban on surface dispersants it submitted to the Coast Guard, that BP often exceeded its own requests, with little indication that it informed the Coast Guard or that the Coast Guard attempted to verify whether BP was exceeding approved volumes. “Either BP was lying to Congress or to the Coast Guard about how much dispersants they were shooting onto the ocean,” said Rep. Markey.
On 2 August, the EPA said dispersants did no more harm to the environment than the oil itself, and that they stopped a large amount of oil from reaching the coast by making the oil break down faster. However, independent scientists and EPA's own experts continue to voice concerns regarding the use of dispersants.
Dispersant use was said to have stopped after the cap was in place. Marine toxicologist Riki Ott wrote an open letter to the EPA in late August with evidence that dispersant use had not stopped and that it was being administered near shore. Independent testing supported her claim. New Orleans-based attorney Stuart Smith, representing the Louisiana-based United Commercial Fisherman’s Association and the Louisiana Environmental Action Network said he “personally saw C-130s applying dispersants from [his] hotel room in the Florida Panhandle. They were spraying directly adjacent to the beach right at dusk. Fishermen I’ve talked to say they’ve been sprayed. This idea they are not using this stuff near the coast is nonsense.”
Use of dispersants deep under water
Some 1,100,000 US gallons (4,200 m3) of chemical dispersants were sprayed at the wellhead 5,000 feet (1,500 m) under the sea. This had never previously been tried but due to the unprecedented nature of this spill, BP along with the U.S. Coast Guard and the Environmental Protection Agency, decided to use "the first subsea injection of dispersant directly into oil at the source".
Dispersants are said to facilitate the digestion of the oil by microbes. Mixing the dispersants with the oil at the wellhead would keep some oil below the surface and in theory, allow microbes to digest the oil before it reached the surface. Various risks were identified and evaluated, in particular that an increase in the microbe activity might reduce the oxygen in the water. Various models were run and the effects of the use of the dispersants was monitored closely. The use of dispersants at the wellhead was pursued and the National Oceanic and Atmospheric Administration (NOAA) estimated that roughly 409,000 barrels (65,000 m3) of oil were dispersed underwater.
Environmental scientists say the dispersants, which can cause genetic mutations and cancer, add to the toxicity of the spill and that sea turtles and bluefin tuna are exposed to an even greater risk than crude alone. According to them, the dangers are even greater for dispersants poured into the source of the spill, where they are picked up by the current and wash through the Gulf. University of South Florida scientists released preliminary results on the toxicity of microscopic drops of oil in the undersea plumes, finding that they may be more toxic than previously thought. The researchers say the dispersed oil appears to be negatively affecting bacteria and phytoplankton – the microscopic plants which make up the basis of the Gulf's food web. The field-based results were consistent with shore-based laboratory studies showing that phytoplankton are more sensitive to chemical dispersants than the bacteria, which are more sensitive to oil. On the other hand, NOAA says that toxicity tests have suggested that the acute risk of dispersant-oil mixtures is no greater than that of oil alone. However, some experts believe that all the benefits and costs may not be known for decades.
Because the dispersants were applied deep under the sea, much of the oil never rose to the surface – which means it went somewhere else, said Robert Diaz, a marine scientist at the College of William and Mary in Williamsburg, Va. "The dispersants definitely don't make oil disappear. They take it from one area in an ecosystem and put it in another," Diaz said. One plume of dipersed oil measured at 22 miles (35 km) long, more than a mile wide and 650 feet (200 m) tall. The plume showed the oil "is persisting for longer periods than we would have expected," said researchers with the Woods Hole Oceanographic Institution. "Many people speculated that subsurface oil droplets were being easily biodegraded. Well, we didn’t find that. We found it was still there". In a major study on the plume, experts found the most worrisome part to be the slow pace at which the oil was breaking down in the cold, 40 °F (4 °C) water at depths of 3,000 feet (910 m) 'making it a long-lasting but unseen threat to vulnerable marine life'. In September, Marine Sciences at the University of Georgia reported findings of a substantial layer of oily sediment stretching for dozens of miles in all directions from the capped well.