The most memorable destruction of a lighter-than-air airship was the burning of the hydrogen-inflated Hindenburg as it approached the mooring tower (mast) at Lakehurst Naval Air Station, New Jersey, on 6 May 1937. A theory that has gained wide popular acceptance proposes that the paint on the fabric covering of the Hindenburg, rather than its highly flammable hydrogen gas, was both the source of ignition and the driving force behind the conflagration. This hypothesis is not supported by the experimental combustion of various samples of cotton cloth treated with aircraft doping paint to simulate the fabric covering of the Hindenburg.

One of the most puzzling technological mysteries of modern history is the origin of the airship Hindenburg fire. Many past inquiries have concluded that free hydrogen used to inflate the dirigible’s gas cells was ignited in some manner, and that the rapid spread of the fire was due to the burning of the hydrogen. This and various other theories about the cause of the Hindenburg disaster are discussed or linked at many web sites, including John Dziadecki’s Zeppelin web site.

One popular theory is that the doping paint used to coat the airship was responsible both for the start of the fire and its rapid progression. (1) This hypothesis, which was first proposed by a German investigator in 1937, has recently received considerable attention in the media and by advocates of hydrogen-fueled power generating devices and vehicles.

A paper by A. J. Dessler of the Lunar and Planetary Laboratory at the University of Arizona contests the paint theory. (2)

To further the understanding of this issue, experiments were conducted using a close reproduction of the Hindenburg covering material to test the flammability of the painted cloth cover. Although it is assumed, based on historical information, that the doping paint on the airship was cellulose acetate butyrate (2), cellulose nitrate was also included in this study in order to replicate a worst-case material.

Unbleached, 100% cotton muslin sheeting was coated with various doping paint regimes. Five scenarios were used for each of two types of doping paint, cellulose nitrate and cellulose acetate butyrate. A suspension of iron oxide or aluminum powder was added to some of the doping paint. There were six test samples of muslin per each combination and type of doping paint. The drying time between the application of each coat was approximately one day.

Sample 1 was handled like all the other samples except no doping paint was applied.

Sample 2 received four coats of plain doping paint.

Sample 3 received one coat of doping paint with an iron oxide suspension and three coats of plain doping paint, applied in that order.

Sample 4 received one coat of plain doping paint and three coats of doping paint with an aluminum suspension, applied in that order.

Sample 5 received one coat of doping paint with an iron oxide suspension and three coats of doping paint with an aluminum suspension, applied in that order.

All of the samples were 18 cm (7 in) by 28 cm (11 in) pieces of muslin mounted in embroidery hoops. A 10-cm by 12-cm rectangle was drawn on each muslin sample, and a square was drawn inside the rectangle. Each side of the square was 10 cm.

The doping paint combinations were painted inside the rectangle using a brush. Brush strokes were allowed to extend outside the rectangle. Adding three percent by volume of the metal powders created the metal suspensions, which were mixed with flat wooden sticks. For each coat of metal- treated paint, only one metal powder was mixed into suspension.

The burn tests were conducted outdoors. The samples were placed flat on a well-ventilated, open cooking grill in a location with little or no breeze. The flame from a butane soldering torch was applied to the narrow side of the rectangle on the sample to ignite the fire. The time required for the flame front to cross the 10-cm portion of the square (along the long side of the rectangle) was measured.

Figure 1. This is a diagram of the sample setup with markings and painting. The yellow represents the cloth piece. The oval represents the embroidery hoop in which the cloth is clamped. The rectangle with the included square is where the doping paint was applied. The short side has a dot that indicates the general place where the fire was started. The dark blue line is where the timer was activated to record the time required for the flame front to reach the left side of the rectangle.

Figure 2. This photograph shows the burning of a fabric sample coated with cellulose acetate butyrate.

The results of the burn tests are listed in Tables 1 and 2. Six burn test attempts were performed on each material combination. In some cases the count is less than six due to mishandling of the timing device. These defective burn tests were not included in the experimental results.