= Bardeen-Brattain point-contact transistor experiment
{c}
{title2=1947}
In December 1947, <John Bardeen> and <Walter Houser Brattain> obtained transistor action at Bell Telephone Laboratories. <William Shockley> proposed the distinct <bipolar junction transistor> within the following month.<Walter Houser Brattain by John Bardeen>[Printed pp. 78–79]{ref}
Their 1948 paper describes ground, etched <germanium>, initially also anodized and rinsed. Surface contacts separated by $50$–$250\,\mathrm{\mu m}$ share a broad base contact. Positive emitter <biasing> injects <electron holes>; negative collector <biasing> draws them toward the collector. An input <signal> changes emitter <electric current>, and a collector <electrical load> converts the resulting change into voltage.<The Transistor, A Semi-Conductor Triode by John Bardeen and Walter Houser Brattain>[pp. 230–231, Fig. 1]{ref}
\Image[/point-contact-transistor-bias-and-readout.png]
{title=Functional <point-contact transistor> experiment}
{description=Original functional schematic; geometry and current arrows are explanatory, not measured data.<Walter Houser Brattain by John Bardeen>[Printed pp. 78–80]{ref}}
{source=point-contact-transistor-bias-and-readout.py}
Early <electric field> control experiments gave unexpectedly weak responses. <John Bardeen> proposed that <surface states> screened the interior. Surface charge could create a barrier and a hole-rich <semiconductor inversion layer>. Replacing an <electrolyte> control with a <gold> contact reversed the response sign, suggesting <electron hole> <carrier injection>. Closely spaced contacts made from a <gold>-coated wedge split with a razor then enabled <power gain>. The relative importance of surface and bulk transport depended on preparation and collector forming.<Semiconductor research leading to the point contact transistor by John Bardeen>[Printed pp. 333–340, Figs. 6–12]{ref}
The 1948 authors found that collector-current increments could exceed emitter-current increments; their <current gain> depended on <biasing>.<The Transistor, A Semi-Conductor Triode by John Bardeen and Walter Houser Brattain>[p. 231, amplification-factor definition]{ref} A terminal-current ratio alone cannot identify the microscopic path. <Surface states>, the local barrier, <carrier injection> and contact preparation prevent a single ideal modern <bipolar junction transistor> model from explaining every <point-contact transistor>.
For the conventional <electric current> directions drawn in the figure, $I_E$ enters the emitter, $I_C$ leaves the collector and signed $I_B$ enters the base. <Kirchhoff's current law> requires
$$
I_E+I_B=I_C,\qquad
\alpha=\left(\frac{\partial I_C}{\partial I_E}\right)_{V_C},\qquad
\delta I_B=(\alpha-1)\delta I_E\quad\text{at fixed }V_C.
$$
Thus a local <current gain> $\alpha>1$ requires an accompanying base-current change. The collector increment is not simply a count of injected <electron holes>, and it does not violate <electric charge> conservation.
For an experiment following the functional arrangement, first establish each contact's separate <current-voltage characteristic> relative to the base. Then hold collector <biasing> fixed, vary a small emitter <signal>, and compare the collector <electrical load> voltage with the input. A decisive <power gain> check measures both input <electric current> and voltage and the output <power> in a known <electrical load>; voltage enlargement alone is insufficient. These are proposed measurement controls, not additional claims about the 1947 laboratory record.
Repeating a setting tests drift; reversing or removing emitter <biasing> tests whether the response requires <carrier injection>. Contact motion, surface contamination, <temperature>, direct electrical coupling and <noise> can all affect the inference. A <small-signal model> describes measured local terminal response without deciding every microscopic mechanism.
Back to article page
