Point-Contact Transistor

Part of The Transistor

The point-contact transistor — invented at Bell Labs in December 1947 — was the first working transistor, using two metal wire contacts pressed onto germanium to demonstrate amplification without vacuum tubes.

Why This Matters

The point-contact transistor is the simplest transistor you can make with hand tools and basic materials. It requires no crystal-growing furnace, no diffusion tube, no photolithography. Two sharpened wires, a piece of germanium crystal, and a handful of components are enough to demonstrate transistor amplification. This was, after all, the method that changed the world.

The original device built by Bardeen and Brattain on December 16, 1947 used gold foil strips on a plastic wedge pressed against a germanium crystal. The first demonstration doubled voice signals in a speech circuit. Within months, this crude device launched the semiconductor industry, the miniaturization of electronics, and eventually computing as we know it.

For someone rebuilding electronics from scratch, the point-contact transistor represents the achievable minimum — a transistor you can build before mastering semiconductor crystal growth. It also provides insight into what transistors were before the theory of how they worked was fully understood.

Historical Background

In 1945, Bell Labs assembled a research group tasked with understanding and exploiting semiconductors. The vacuum tube had limitations: size, power consumption, fragility, and warm-up time. Solid-state alternatives were known in principle — crystal radio detectors had been used since the 1900s — but no one had made a solid-state amplifier.

The key team: William Shockley (theoretical framework), John Bardeen (surface physics theory), Walter Brattain (experimental work). They focused on germanium because it was purer than silicon (refining techniques were better) and more conductive at room temperature.

The breakthrough came from a failed experiment. Shockley predicted that a field effect should control conductivity in a semiconductor slab — but the effect was much weaker than expected. Bardeen proposed that surface states (electrons trapped at the germanium surface) were screening the applied field. Brattain, investigating these surface states, accidentally discovered that two closely spaced contacts showed current gain between them.

On December 16, 1947, Bardeen and Brattain demonstrated amplification. Shockley, excluded from the demonstration due to patent concerns, spent the next month developing the theoretical framework and the junction transistor concept that would eventually supersede the point-contact device.

Construction of a Point-Contact Transistor

Materials

• N-type germanium crystal (resistivity ~1–10 Ω·cm)
• Two pieces of phosphor-bronze wire or tungsten wire, ~0.1–0.2 mm diameter
• Thin gold or platinum foil (optional but improves performance)
• Small mechanical jig to hold wires in position
• Knife for sharpening wire tips
• Magnifying glass
• Battery and resistors for testing

The Semiconductor Base

The original transistors used n-type germanium. You need a clean, polished surface. Germanium’s advantage: it can be etched with hydrogen peroxide to clean the surface and remove oxide, creating consistent surface states.

Surface preparation:

1. Obtain or grow n-type germanium crystal (lightly phosphorus or antimony doped)
2. Cut a small piece, ~5×5×2 mm
3. Abrade one face with progressively finer abrasive: 400 grit, 1000 grit, then polish with optical polishing compound
4. Etch in 30% hydrogen peroxide for 30 seconds → gentle fizzing, then rinse with deionized water
5. Do not touch the polished surface — handle by the edges

Wire Preparation

1. Cut two ~3 cm lengths of 0.1 mm phosphor-bronze wire
2. Sharpen each end to a fine point using a sharpening stone or emery paper — drag the wire at a shallow angle while rotating it
3. The tips should be visually sharp when examined under a magnifying glass
4. Bend each wire to ~90° near the tip, leaving a 1–2 mm straight section before the bend

Assembly

The critical requirement: the two wire tips must be on the germanium surface, extremely close together — ideally 0.05–0.1 mm apart (50–100 microns). This requires a steady hand and magnification.

Original Bardeen-Brattain method:

1. Take a plastic wedge (a cut-down pencil eraser works)
2. Wrap the edge of the wedge with gold foil (gold leaf or gold from electronics)
3. Cut a thin slit through the gold at the very tip of the wedge
4. The slit separates the gold into two halves, each acting as a contact
5. Press the wedge tip against the germanium surface — the two gold strips make contact 0.05 mm apart

Wire contact method (simpler, lower performance):

1. Build a small jig from a wooden block with two grooves or guide holes
2. Thread the two sharpened wires through the guides so their tips emerge close together
3. Adjust the spacing — this is the hardest part
4. Clamp the wires in position
5. Lower the jig onto the germanium surface so both tips touch simultaneously
6. Apply very light pressure with a spring or rubber band

Electrical Connections

The three electrodes:

• Emitter: one wire contact (forward biased ~+0.1 V relative to germanium base)
• Collector: the other wire contact (reverse biased −5 to −20 V)
• Base: the n-type germanium body itself (connected via a larger contact like a metal clip or conductive paint)

Test circuit:

• 1.5 V battery (AA cell) through 10 kΩ resistor to emitter contact
• 9–22 V battery through 10 kΩ to 100 kΩ collector resistor, then to collector contact
• Germanium base connected to junction between the two batteries (common ground)
• Output: voltage across collector resistor

Expected output: if emitter current increases, collector current increases proportionally — this is amplification. Expect current gain of 2–20 for a crude point-contact device.

Why It Works

The mechanism of point-contact transistor operation differs from junction transistors and is more complex. Two effects occur:

1. Minority carrier injection: Current flowing through the emitter contact creates minority carriers (holes in n-type germanium) near the contact. These holes diffuse toward the collector contact.

2. Collector current multiplication: The collector contact, biased in reverse, collects these holes along with some electrons. A current-multiplication effect at the metal-semiconductor contact (related to avalanche multiplication) causes the collector current to exceed the emitter current — giving current gain above 1.

The point-contact transistor can actually have current gain greater than 1 (alpha > 1) in the common-base configuration — something junction transistors cannot do. This unusual property comes from avalanche multiplication at the collector contact.

Limitations

Property	Point-Contact	Junction Transistor
Current gain	2–20 (unreliable)	20–500 (reliable)
Noise	Very high	Low to moderate
Consistency	Poor — varies device to device	Good
Frequency	Moderate (MHz range)	Low to very high (varies)
Power	Very low (<1 mW typical)	Up to watts
Durability	Fragile — contacts can shift	Robust

The point-contact transistor was obsolete by the mid-1950s, replaced by alloy-junction and then diffused-junction transistors. Its value today is historical and educational: it demonstrates the principle with the absolute minimum of materials and equipment.

Building a Low-Frequency Amplifier

For a first test of a point-contact transistor:

1. Assemble the transistor as above
2. Connect emitter through 4.7 kΩ to +1.5 V
3. Connect collector through 47 kΩ to −9 V
4. Connect base to ground
5. Introduce a microphone in series with the emitter circuit
6. Connect a speaker or earphone between collector and −9 V
7. Speak into the microphone — you should hear amplified speech in the earphone

If you get no amplification:

• Adjust contact pressure (too light → no contact; too heavy → shorts between contacts)
• Check contact spacing — use a magnifier to verify points are close but not touching each other
• Try adjusting the forward bias on the emitter

Summary

Point-Contact Transistor — At a Glance

• First transistor, invented December 1947 by Bardeen and Brattain at Bell Labs using germanium and metal wire contacts
• Two sharpened wire tips pressed ~0.05 mm apart on polished n-type germanium
• Emitter contact forward biased ~0.1 V; collector contact reverse biased 5–20 V
• Current gain 2–20, high noise, unreliable — but demonstrably works
• Can be built without crystal growth or diffusion furnaces — just shaped wire and polished germanium
• Replaced by alloy-junction transistors by mid-1950s, but remains historically significant
• Building one is the fastest path from “raw germanium” to “demonstrating transistor amplification”