281 
SLOW RECOVERY AND PERMANENT SET IN COPPER, ALUMINUM, 
AND LEAD. 
ALBERT EX. WOODRUFF. 
When a copper wire is strained and the stress is removed there is an 
instantaneous recovery followed by a slow recovery with time. This slow 
recovery may be accounted for by the hypothesis that it is due to the re- 
crystallization of the metal. The fact that there is a recovery with time is 
not new. Even the magnitude of the recovery as a function of time has 
been investigated experimentally and is fairly definitely known. But so far 
no one seems to have been sufficiently interested to seek for an explanation 
of the cause of recovery. 
Andrade (1) has done much experimental work on the flow of soft 
_metals under stress. Knowing from the micro-photographic work of 
Quineke (2), Ewing and Rosenhain (8), and Beilby (4) that when a metal 
is strained there is a breaking down of the crystallar structure of the metal, 
he reasons that since the two take place simultaneously the one is due to 
the other, and he is able to substantiate his theory by experiment, both from 
observed phenomena of a strained wire and by looking into a microscope 
and seeing what actually takes place in the metal. 
But the work of Quincke, Ewing and Rosenhain, and Beilby does not stop 
with the observation that the crystallar structure of a metal breaks down 
under excessive strain. They find that as soon as the stress is removed 
that the erystals of the metal begin to reform. For copper at ordinary 
temperatures this process is very slow. But if the temperature is raised, 
to say three or four hundred degrees the process of recrystallization is ex- 
ceedingly rapid, almost instantaneous, and the specimen becomes annealed. 
The object of this paper is to present facts which show that slow recovery 
is due to recrystallization of the metal. 
Plate I shows that an unannealed specimen of copper has a much more 
decided recovery than an annealed specimen subjected to the same strain. 
Plate III shows the same thing for aluminium. Plates II, IV and VI show 
that for all three metals investigated the larger the stress applied, other 
conditions being equal, the greater the recovery. Plates I, III and V show 
that the longer the time of applying the stress the greater the recovery, the 
stresses being equal. It will be noticed from these same curves that per 
unit of length per unit of stress the amount of recovery of the different 
metals is in the same order as the temperatures at which the metals 
anneal. All these facts support the hypothesis that the recovery of a metal 
after the stress has been removed is connected with the process of recry- 
stallization. 
In the first instance the material of a drawn wire that has not been 
annealed is largely reduced to the amorphous phase. Such crystals and 
parts of crystals as remain are under strain which is the result of drawing. 
Most of the strain was relieved when the tension of the drawing process 
was released. But the fine amorphous particles fill the spaces about the 
remaining crystals leaving the metal still in a state of strain. Annealing 
or recrystallization immediately begins and the amorphous particles begin 
