Experiment to investigate the materials used to manufacture pencil sharpeners
AJ PENTON, JD HALL, C REED, DA GEORGESON, JS PIPET, University of Cambridge. June 1998.


An attempt was made to verify the hypothesis that pencil sharpeners are manufactured from magnesium metal. Test subjects were exposed to a high temperature flame, in an attempt to ignite them. Some of the subjects burned with a bright white flame that is characteristic of magnesium. It was concluded that old pencil sharpeners and the more expensive modern pencil sharpeners are made from an alloy that is principaly or wholy magnesium.


Proir to this experiement, several people had claimed that pencil sharpeners were made of magnesium. Our initial reaction to this claim was, "Never, why should they be". We were convienced that other suitable, safer and cheaper materials existed and that these would be used in preference to magnesium. Still, no discoveries were ever made with this kind of thinking, so we set out to test the hypothesis.

Most people's school science lab memories of magnesium will be of soft strips of a dull grey metal that burn with an intensely bright white light. Indeed magenesium powder was burnt in camera flashes for many years until the Xenon flash tube was introduced. A quick mechanical and visual test showed that the pencil sharpeners were quite soft (they were easily scratched by steel) and dull grey (except when new). However, many materials are both soft and dull grey, so the pencil sharpeners had to be burnt to show for definite whether they were or were not made of magnesium.


Pencil sharpeners were collected from a variety of sources, mainly bought new from local shops, but old sharpeners were also collected. The prices of the new sharpeners were noted. The sharp, and potentially dangerous blades were carefully removed and disguarded. The pencil sharpeners were then taken to a remote location, close to a large supply of water (for fire-fighting purposes) and placed on a slab of concrete. The concrete was concidered inert enough to withstand the intense flame that was expected.

The test subjects were individually exposed to the flame from a gas powered blow-torch. The reactions of the sharpeners were carefully noted. Due to the low budget of the investigation, no quantitative measurements were made.


The following table shows the type and cost of each pencil sharpener and its reaction to the blow-torch flame:
Type Cost/pence Appearance Before Reaction to flame Appearance After
Ryman 120 Low weight. Glowed orange before emitting small white sparks then burning with a bright white flame and emiting large burning bright white sparks. Excellent entertainment. Large pile of white ash
Helix 100 Shiny Melted into a shapeless blob. A shapeless blob
WHSmith 99 Heavy. Obviously cast, not machined. Melted slightly As before, but with rounded corners
WHSmith 'Wooden' 50 Clearly made of wood Burnt slowly with a yellow flame.
Old ('Made in West Germany') ? Dull grey Glowed orange before emitting small white sparks then burning with a bright white flame. A large pile of white ash.
Old (Double sharpener) ? Very dull grey Burnt as above, except for longer. As above.


A large number of pictures were taken. Unfortunately, due to a hard disk crash, these are unavailable. We hope to replicate the experiment soon to bring you some pictures. In the meantime imagine a black square with a bright white spot in the centre.


Clearly some of the test subjects burnt with an impressive white flame. This indicates that the metal was either magnesium or some other which burns in a similar manner. Since magnesium is the only common and safe element which burns like this, the metal is most likely to be either Magnesium or an alloy containing a large proportion of it. The softness of the metal and the white oxide powder formed do not contradict the hypothesis, but are not substancial evidence in support, since these are characteristics of many metals.

Further Research

If more equipment were available, qualitative work with this experiment might be done. An optical pyrometer might be used to measure the temperature of the flame and a spectrometer would allow any spectral characteristics to be detected. The spectrum of magnesium might well be distinctive enough to positively identify the metal as magnesium (by comparison).

Chemical, NMR or mass spectrometer analysis would allow a more positive identification, but this defeats the point of using this simple experimental technique.

John Penton. September 1998