On June 25, 1997 a docking mishap of a Progress supply ship caused the Progress vehicle to crash into an array of solar panels and puncture the hull of the Spektr module. The puncture was small enough to allow the crew to seal off the Spektr module and repressurize the rest of the station. The Progress vehicle struck the Spektr module several times and the exact location, size, and number of punctures in the Spektr hull was unknown. Russian cosmonauts donned space suits and went inside the Spektr module to repair some electrical power cables and look for the location of the hull breach, they could not identify the exact location of the hole (or holes). The Spektr module was pressurized with Mir cabin air twice during the STS-86 fly around in an attempt to detect leakage (in the form of ice particles) from the module. Seven particles were observed within a 36 second time span, but tracking the path of the individual particles did not pinpoint a specific leak location.
During STS-91, the last NASA Shuttle mission to the Mir in the Phase 1 program, a fluorescent tracer gas experiment was conducted. A Russian contingency Air Pressurization Unit (APU) with a total capacity of approximately 40 liters was filled with the saturation vapor pressure of acetone (38 grams) the saturation vapor pressure of biacetyl (9 grams) and 4260 psig total pressure of Nitrogen gas. When no oxygen is present, acetone and biacetyl fluoresce in sunlight and emit a bluish green glow. The plan of the experiment was to release the gas mixture into the Spektr module, use the nitrogen as a pressurant to cause the gas mixture to migrate through the hole (or holes) in the Spektr hull and leak out to space. If the leak plume were illuminated by sunlight, and an observer could view the plume against a uniformly dark background, the plume could be perceived with the unaided eye, and the location and number of holes in the Spektr hull could be identified.
The mechanics and operations of the experiment were conducted as planned, but no fluorescent plume was seen, and no fluorescent plume was photographed. The primary cause for this lack of visible plume was an insufficient amount of tracer gas, the large distance between the Spektr module and the Shuttle viewing platform, and the glare of sunlight.