After World War II, the U.S. military desired better strategic aerial reconnaissance to help determine Soviet capabilities and intentions. Into the 1950s, the best intelligence the American government had on facilities deep inside the Soviet Union were German Luftwaffe photographs taken during the war of territory west of the Ural Mountains, so overflights to take aerial photographs of the Soviet Union began. After 1950, Soviet air defenses aggressively attacked all aircraft near the country's borders—sometimes even those over Japanese airspace—and the existing reconnaissance aircraft, primarily bombers converted for reconnaissance duty such as the Boeing RB-47, were vulnerable to anti-aircraft artillery, missiles, and fighters. Richard Leghorn of the USAF suggested that an aircraft that could fly at 60,000 feet (18,300 m) should be safe from the MiG-17, the Soviet Union's best interceptor, which could barely reach 45,000 feet (13,700 m). He and others believed that Soviet radar, which used American equipment provided during the war, could not track aircraft above 65,000 feet (19,800 m).
The highest-flying aircraft available to America and its allies at the time was the English Electric Canberra, which could reach 48,000 feet (14,600 m). The British had already produced the PR3 photo-reconnaissance variant, but the USAF asked for English Electric's help to further modify the Martin B-57 (the American licensed version of the Canberra) with long, narrow wings, new engines, and a lighter-than-normal airframe to reach 67,000 feet (20,400 m). Air Research and Development Command mandated design changes that made the aircraft more durable for combat, but the resulting RB-57D aircraft of 1955 could only reach 64,000 feet (19,500 m). The Soviet Union, unlike the United States and Britain, had improved radar technology after the war, and could track aircraft above 65,000 feet (19,800 m).
It was thought that an aircraft that could fly at 70,000 feet (21,300 m) would be beyond the reach of Soviet fighters, missiles, and radar. Another USAF officer, John Seaberg, wrote a request for proposal in 1953 for an aircraft that could reach 70,000 feet (21,300 m) over a target with 1,500 nmi (1,700 mi; 2,800 km) of operational radius. The USAF decided to solicit designs only from smaller aircraft companies that could give the project more attention. Under the code name "Bald Eagle", it gave contracts to Bell Aircraft, Martin Aircraft, and Fairchild Engine and Airplane to develop proposals for the new reconnaissance aircraft. Officials at Lockheed Aircraft Corporation heard about the project and decided to submit an unsolicited proposal. To save weight and increase altitude, Lockheed executive John Carter suggested that the design eliminate landing gear and avoid attempting to meet combat load factors for the airframe. The company asked Clarence "Kelly" Johnson to come up with such a design. Johnson was Lockheed's best aeronautical engineer, responsible for the P-38 and the P-80. He was also known for completing projects ahead of schedule, working in a separate division of the company, informally called the Skunk Works.
Original U-2A at USAF Museum
Johnson's design, named CL-282, was based on the Lockheed XF-104 with long, slender wings and a shortened fuselage. The design was powered by the General Electric J73 engine and took off from a special cart and landed on its belly. It could reach an altitude of 73,000 feet (22,300 m) and had a 1,600 mi (1,400 nmi; 2,600 km) radius. The reconnaissance aircraft was essentially a jet-powered glider. In June 1954, the USAF rejected the design in favor of the Bell X-16 and the modified B-57. Reasons included the lack of landing gear, use of the J73 engine instead of the more proven Pratt & Whitney J57 (like the competing designs), and not using multiple engines, which, the USAF believed, was more reliable. General Curtis LeMay of Strategic Air Command (SAC) walked out during a CL-282 presentation, saying that he was not interested in an airplane without wheels or guns.
Civilian officials including Trevor Gardner, an aide to Secretary of the Air Force Harold E. Talbott, were more positive about the CL-282 because of its higher potential altitude and smaller radar cross section, and recommended the design to the Central Intelligence Agency's Office of Scientific Intelligence. At that time, the CIA depended on the military for overflights, and Director of Central Intelligence Allen Dulles favored human over technical intelligence gathering methods. However, the Intelligence Systems Panel, a civilian group advising the USAF and CIA on aerial reconnaissance, had recognized by 1954 that the RB-57D would not meet the 70,000 feet (21,300 m) requirement that panel member Allen Donovan of Cornell Aeronautical Laboratory believed was necessary for safety. The CIA told the panel about the CL-282, and the aspects of its design that the USAF saw as flaws (the single engine and light load factor) appealed to Donovan, a sailplane enthusiast who believed that a sailplane was the type of high-altitude aircraft the panel was seeking.
Edwin Land, the developer of instant photography, and another member of the panel proposed to Dulles through Dulles' aide, Richard M. Bissell, Jr., that his agency should fund and operate this aircraft. Land believed that the military operating the CL-282 during peacetime could provoke a war. Although Dulles remained reluctant to have the CIA conduct its own overflights, Land and James Killian of MIT told President Eisenhower about the aircraft; Eisenhower agreed that the CIA should be the operator. Dulles finally agreed, but some USAF officers opposed the project because they feared it would endanger the RB-57D and X-16. The USAF's Seaberg helped persuade his own agency to support the CL-282, albeit with the higher-performance J57 engine, and final approval for a joint USAF-CIA project (the first time the CIA dealt with sophisticated technology) came in November 1954. Lockheed had meanwhile become busy with other projects and had to be persuaded to accept the CL-282 contract after approval.
Bissell became head of the project, which used covert funding; under the Central Intelligence Agency Act of 1949, the CIA's director is the only federal government employee who can spend "un-vouchered" government money. Lockheed received a $22.5 million contract in March 1955 for the first 20 aircraft, with the first $1.26 million mailed to Johnson's home in February 1955 to keep work going during negotiations. The company agreed to deliver the first aircraft by July of that year and the last by November 1956. It did so, and for $3.5 million under budget. The Flight Test Engineer in charge was Joseph F Ware, Jr.
Procurement of the aircraft's components occurred secretly. When Johnson ordered altimeters calibrated to 80,000 feet (24,400 m) from a company whose instruments only went to 45,000 feet (13,700 m), the CIA set up a cover story involving experimental rocket aircraft. Shell Oil developed a new low-volatility, low vapor pressure jet fuel that would not evaporate at high altitudes; the fuel became known as JP-7, and manufacturing several hundred thousand gallons for the aircraft in 1955 caused a nationwide shortage of Esso's FLIT insect repellent. The aircraft was renamed the U-2 in July 1955, the same month the first aircraft, Article 341, was delivered to Groom Lake. The "U" referred to the deliberately vague designation "utility" instead of "R" for "reconnaissance", and the U-1 and U-3 aircraft already existed. The CIA assigned the cryptonym AQUATONE to the project, with the USAF using the name OILSTONE for their support to the CIA.
James Baker developed the optics for a large-format camera to be used in the U-2 while working for Perkin-Elmer. The new camera had a resolution of 2.5 feet (76 cm) from an altitude of 60,000 feet (18,000 m). The aircraft was so crowded that when Baker asked Johnson for six more inches of space for a lens of 240 in (610 cm) focal length, Johnson replied "I'd sell my grandmother for six more inches!"; Baker instead used a 180 in (460 cm) f/13.85 lens in a 13×13 in (33×33 cm) format for his final design.
When the first overflights of the Soviet Union were tracked by radar, the CIA initiated Project Rainbow to reduce the U-2's radar cross-section. This effort ultimately proved unsuccessful, and work began on a follow-on aircraft, which resulted in the Lockheed A-12 Oxcart.
In August 2015, the 60th-year anniversary of the U-2 program, Lockheed Martin's Skunk Works revealed they were internally developing a successor to the U-2, referred to as the UQ-2 or RQ-X, combining features from both the manned U-2 and unmanned Northrop Grumman RQ-4 Global Hawk and improving upon them. Disclosed details say the design is essentially an improved U-2 airframe with the same engine, service ceiling, sensors, and cockpit, with the main differences being an optional manning capability (something Lockheed has proposed for the U-2 to the Air Force several times but has never gained traction) and low-observable characteristics. The Air Force has no requirement or timeframe for a next-generation High-Altitude Long Endurance (HALE) platform, but Lockheed sees a future need and wants something in development early. Having the option of an onboard pilot is considered a deterrent because it can be used in peacetime situations where unmanned aircraft would more likely be engaged, since there is no possibility of killing a person. The company's last attempt to create a stealth unmanned aircraft was the RQ-3 DarkStar, which never made it past flight testing and was canceled. Plans for a U-2 replacement would not conflict with development of the SR-72, another project by the company to create a hypersonic unmanned surveillance plane, as it would be suited for missions that require greater speed for time-sensitive targets.
The company released a notional artist's impression of the TR-X aircraft at an Air Force Association conference in Washington on 14 September 2015. Its name was changed to mean "tactical reconnaissance" to reflect its purpose as an affordable peace and wartime ISR aircraft, distinguishing it from strategic, penetrating SR-71-class platforms; TR is a reference to the short-lived rebranding of the U-2 as the TR-1 in the 1980s. Size, and thus cost, is kept down by having less endurance than the Global Hawk at around 20 hours, which is still about the same time as a normal RQ-4 sortie even though it is capable of flying for 34 hours. Although originally planned to be optionally-unmanned, some Lockheed officials are leaning towards a purely unmanned aircraft, as it is expected that issues with airspace integration of UAVs will be addressed by the time it will be introduced. The TR-X concept is aimed squarely at Air Force needs, and is not currently being marketed to the CIA or other government agencies. It would have increased power and cooling to accommodate new sensors, communication equipment, electronic warfare suites, and perhaps offensive or defensive laser weapons. TR-X could be ready for service in the 2025 timeframe, with a fleet of 25–30 aircraft proposed to replace the nearly 40-aircraft mix of U-2s and RQ-4s.
Lockheed revealed more specifications about the TR-X at a 15 March 2016 media day, confirming the aircraft would be unmanned and air refuelable. Its maximum takeoff weight would be greater than either the U-2's or RQ-4's at around 54,000 lb (24,000 kg), with a 5,000 lb (2,300 kg) payload and 130 ft (40 m) wingspan. It will use the same F118-101 turbofan and generator as the U-2, but thrust could increase to 19,000 lb and power increased to 65–75 kVA; service ceiling would increase to 77,000 ft (23,000 m) with a second engine. The TR-X is meant to be "survivable, not unnoticeable," operating outside of enemy air defense bubbles rather than penetrating into them.
The U-2 has used Jet Propellant Thermally Stable (JPTS) since the aircraft's development in the 1950s. JPTS is a high thermal stability, high altitude fuel, created specifically as fuel for the U-2. JPTS has a lower freeze point, higher viscosity, and higher thermal stability than standard Air Force fuels. In 1999, the United States Air Force spent approximately $11.3 million on fuel for the U-2 aircraft and was looking for a lower cost alternative. JPTS is a specialty fuel and as such has limited worldwide availability and costs over three times the per-gallon price of the Air Force's primary jet fuel, JP-8. Research is under way to find a cheaper and easier alternative involving additives to generally used jet fuels. A JP-8 based alternative, JP-8+100LT, is being considered. JP-8+100 has increased thermal stability by 100 °F (56 °C) over stock JP-8, and is only 0.5 cents per gallon more expensive; low temperature additives can be blended to this stock to achieve desired cold performance.
Due to the small landing gear, a perfect balance in the fuel tanks was essential for a safe landing. Similarly to sailplanes, the U-2 had a yaw string on the canopy to detect slip or skid during the approach. A skid during flight with no bank was the hint of an unbalance around the longitudinal axis which could be resolved by moving the fuel to the left or right wing tank.