Fact-checked by Grok 2 weeks ago

p–n junction

A p–n junction is the interface between a region of a semiconductor doped to produce mobile positive charge carriers (holes), known as the p-type region, and an adjacent region doped to produce mobile negative charge carriers (electrons), known as the n-type region, forming a fundamental compositional structure in semiconductor electronics.^[1] This junction arises in a single crystal where the impurity concentration varies from donor-dominated (n-type) to acceptor-dominated (p-type), typically achieved during crystal growth by controlling dopant addition.^[1] Upon formation, free electrons from the n-type region diffuse across the interface and recombine with holes in the p-type region, leaving behind immobile ionized donors (positive) on the n-side and ionized acceptors (negative) on the p-side; this creates a depletion region near the junction, characterized by a lack of mobile carriers and the presence of an electric field due to the resulting space charge.^[2] The diffusion process establishes a built-in potential (or contact potential) across the depletion region, approximately 0.7 V for silicon at room temperature, which acts as a barrier opposing further net carrier diffusion and maintaining thermal equilibrium with zero net current.^[1] The width of the depletion region depends on the doping concentrations and typically spans tens to hundreds of nanometers.^[2] The theoretical foundation for the p–n junction was developed by William Shockley in his 1949 paper, emphasizing its role in rectification and transistor action within germanium and other semiconductors.^[3] Under forward bias, where the p-side is connected to the positive terminal and the n-side to the negative, the applied voltage reduces the built-in potential barrier, narrowing the depletion region and allowing majority carriers to inject across the junction, resulting in exponential increase in current flow.^[4] Conversely, under reverse bias, the applied voltage increases the barrier, widening the depletion region and restricting current to a minimal reverse saturation current carried by minority carriers, enabling the unidirectional conduction essential for diode operation.^[4] These biasing characteristics make the p–n junction the core element in semiconductor devices including diodes for rectification, bipolar junction transistors for amplification, photovoltaic solar cells for energy conversion, and light-emitting diodes for photon emission.^[1]

Semiconductor basics

Intrinsic and extrinsic semiconductors

An intrinsic semiconductor is a pure material, such as silicon or germanium, with no significant impurities, where the electrical conductivity arises solely from thermally generated electron-hole pairs.^[5] In its energy band structure, electrons occupy the valence band at absolute zero temperature, leaving the conduction band empty; the valence band is separated from the conduction band by a bandgap energy E_g, typically on the order of 1 eV, which allows limited thermal excitation of electrons across the gap at room temperature.^[6] For silicon, E_g = 1.12 eV, while for germanium, E_g = 0.66 eV.^[7] The thermally excited electrons in the conduction band are equal in number to the holes left in the valence band, resulting in an intrinsic carrier concentration n_i of approximately $10^{10} cm^{-3} for silicon at 300 K.^[5] The generation of these charge carriers in intrinsic semiconductors follows the temperature dependence given by n_i \propto T^{3/2} \exp(-E_g / 2kT), where T is the absolute temperature, k is Boltzmann's constant, and the exponential term reflects the probability of thermal excitation across half the bandgap (since each carrier pair requires E_g).^[8] This relationship arises from the full expression n_i = \sqrt{N_c N_v} \exp(-E_g / 2kT), with N_c and N_v as the effective densities of states in the conduction and valence bands, respectively, both scaling as T^{3/2}.^[8] At room temperature, the low n_i limits conductivity compared to metals or doped materials, but increasing temperature exponentially boosts carrier density, enhancing intrinsic conduction.^[5] Extrinsic semiconductors are created by intentionally introducing impurities, known as doping, into the intrinsic material, which modifies the band structure by adding donor or acceptor levels near the band edges and shifts the Fermi level away from the midgap position. This alteration increases the concentration of majority charge carriers—electrons in n-type or holes in p-type—by orders of magnitude over the intrinsic n_i, enabling controlled conductivity essential for devices like p-n junctions.^[9] The Fermi level shift depends on the doping concentration and type, positioning it closer to the conduction band for n-type or valence band for p-type materials, while minority carriers remain near n_i^2 / N, where N is the majority carrier density.^[5]

Doping mechanisms

Doping mechanisms in semiconductors rely on the controlled introduction of impurities, known as dopants, to create an imbalance in charge carriers, forming either n-type or p-type materials crucial for p-n junctions. These impurities substitute into the crystal lattice and modify the band structure by introducing allowed energy levels within the bandgap. In silicon, the most common semiconductor for such devices, doping concentrations typically range from $10^{15} to $10^{18} \, \mathrm{cm}^{-3}, far exceeding the intrinsic carrier concentration of approximately $10^{10} \, \mathrm{cm}^{-3} at room temperature.^[5] For n-type doping, donor impurities from group V elements, such as phosphorus, are incorporated into the silicon lattice. Phosphorus atoms possess five valence electrons, forming four covalent bonds with neighboring silicon atoms and leaving one loosely bound electron that can easily enter the conduction band, acting as a free electron. This extra electron makes electrons the majority charge carriers in n-type silicon, with donor concentrations denoted as N_d. The donor level lies just below the conduction band edge, with an ionization energy E_d \approx 0.045 \, \mathrm{eV} for phosphorus in silicon, classifying it as a shallow donor that is nearly fully ionized at room temperature due to thermal energy exceeding this barrier.^[5]^[10] Other group V elements like arsenic and antimony serve similar roles as donors.^[11] In contrast, p-type doping employs acceptor impurities from group III elements, exemplified by boron in silicon. Boron atoms have three valence electrons, creating a deficiency that accepts an electron from the valence band upon incorporation, thereby generating a mobile hole as the majority carrier. Acceptor concentrations are denoted as N_a, typically in the same range as N_d. The acceptor level is positioned about 0.045 eV above the valence band edge for boron, also a shallow level that ionizes completely at room temperature. Aluminum and gallium function analogously as acceptors.^[5]^[12]^[13] The presence of these dopants shifts the Fermi level significantly: in n-type material, it approaches the conduction band, enhancing electron availability, while in p-type, it nears the valence band, favoring holes. When both donor and acceptor impurities coexist at low concentrations (e.g., N_d and N_a both below $10^{16} \, \mathrm{cm}^{-3}), compensation doping occurs, where oppositely charged impurities neutralize each other, yielding a net carrier concentration of |N_d - N_a| and potentially reducing overall conductivity due to scattering effects.^[5]^[14]

Junction formation

Interface between p-type and n-type regions

The interface between p-type and n-type semiconductor regions forms when a material with acceptor dopants (p-type, rich in holes) is brought into contact with a material rich in donor dopants (n-type, rich in electrons), creating a boundary where carrier concentrations differ sharply. In the ideal case of an abrupt junction, this boundary assumes a step-like change in doping profile, with uniform concentrations on each side up to the interface; in contrast, graded junctions feature a gradual variation in net doping across the transition region, often resulting from diffusion during fabrication.^[3]^[15] Upon contact, without any applied voltage, the system approaches thermal equilibrium through the initial diffusion of charge carriers driven by concentration gradients. Electrons, as majority carriers in the n-type region, diffuse across the interface into the p-type region where their concentration is low, while holes, majority carriers in the p-type region, similarly diffuse into the n-type region. This bidirectional diffusion occurs spontaneously due to the thermal energy of carriers, leading to a net transfer until equilibrium is established.^[2]^[3] As these minority carriers cross the interface, they encounter opposite carriers and undergo recombination, neutralizing each other and leaving behind immobile ionized dopant atoms. On the n-type side, the diffused holes recombine with electrons, exposing positively charged donor ions; conversely, on the p-type side, diffused electrons recombine with holes, exposing negatively charged acceptor ions. This process of diffusion and recombination initiates the separation of fixed charges at the interface, setting the stage for the junction's electrostatic properties in thermal equilibrium.^[2]^[3] A common example is the silicon p-n junction used in integrated circuits, where abrupt interfaces are engineered via ion implantation to achieve precise control over carrier diffusion and recombination during device fabrication.^[15]

Depletion region development

When a p–n junction is formed by joining p-type and n-type semiconductors, majority carriers diffuse across the interface due to concentration gradients, leading to recombination and the exposure of fixed ionized impurities. This charge separation creates a space charge region, known as the depletion region, where the electric field balances further diffusion, establishing equilibrium. The process results in a region depleted of mobile carriers, with positive charge from ionized donors on the n-side and negative charge from ionized acceptors on the p-side.^[2]^[16] The depletion approximation simplifies the analysis by assuming complete ionization of dopants within the depletion region and negligible mobile carrier density there, while the regions outside remain electrically neutral with uniform carrier concentrations. This approximation holds well for abrupt junctions under equilibrium or low bias conditions, enabling tractable electrostatic modeling. Under this model, the net charge density \rho is uniform on each side: \rho = q(N_d - N_a) on the n-side, where N_d and N_a are donor and acceptor concentrations, respectively, and \rho = -q(N_a - N_d) on the p-side; for one-sided junctions where one doping level greatly exceeds the other (e.g., N_a \gg N_d), the charge is dominated by the lighter doping.^[17]^[18] The electric field arises from these fixed charges of ionized donors (positive on n-side) and acceptors (negative on p-side), building up to oppose diffusion. The field peaks at the metallurgical junction and exhibits a triangular profile across the depletion region in the abrupt junction case, with the maximum value determined by the integrated charge. This field profile is derived using Poisson's equation, \frac{dE}{dx} = \frac{\rho}{\epsilon}, where E is the electric field, \epsilon is the permittivity of the semiconductor, relating the field gradient directly to the local charge density.^[16] The total depletion width W scales inversely with the square root of the doping concentration on the lower-doped side, being wider there to accommodate the lower charge density for electrostatic balance. For a one-sided abrupt junction, the width approximates W \approx \sqrt{\frac{2\epsilon V_{bi}}{q N_{low}}}, where V_{bi} is the built-in potential and N_{low} is the lower doping level, illustrating how lighter doping extends the region.^[18]

Equilibrium properties

Built-in electric field

In a p–n junction at thermal equilibrium, the built-in electric field arises from the separation of mobile charge carriers during junction formation, where electrons diffuse from the n-type region to the p-type region and holes diffuse in the opposite direction, leaving behind fixed ionized donors on the n-side and acceptors on the p-side. This creates a space charge region with positive charge on the n-side and negative charge on the p-side, establishing an electric field directed from the positively charged n-side to the negatively charged p-side. The field opposes further diffusion of majority carriers across the junction by exerting a drift force on electrons toward the n-side and on holes toward the p-side, achieving a balance between diffusion and drift currents with zero net current flow.^[19]^[20] For an abrupt p–n junction, the electric field varies linearly within the depletion region due to the uniform doping approximation, reaching its maximum magnitude at the metallurgical junction. The peak field strength is approximated as E_{\max} \approx \frac{q N_d W_n}{\epsilon}, where q is the elementary charge, N_d is the donor concentration on the n-side, W_n is the depletion width on the n-side, and \epsilon is the permittivity of the semiconductor; a symmetric expression applies for the p-side using acceptor concentration N_a and W_p. This maximum field typically ranges from $10^4 to $10^5 V/cm in silicon junctions with moderate doping levels around $10^{16} cm^{-3}, sufficient to separate photogenerated carriers in applications like solar cells. The field profile is derived from Poisson's equation, \frac{dE}{dx} = \frac{\rho}{\epsilon}, where \rho is the charge density from ionized dopants, and integrated across the depletion region.^[19]^[17]^[21] The electrostatic potential V(x) across the junction is obtained by integrating the electric field:

V(x) = -\int E(x) \, dx,

yielding a quadratic potential profile in the depletion approximation, which quantifies the built-in voltage drop. This field-induced band bending ensures alignment of the Fermi levels throughout the junction, as the differing Fermi positions in isolated p- and n-type materials equalize through charge transfer, maintaining constant electrochemical potential in equilibrium. In the band diagram, the field causes upward bending of bands on the p-side and downward on the n-side relative to the Fermi level, preventing net carrier flow.^[19]^[22]/07:_The_Crystalline_Solid_State/7.01:_Molecular_Orbitals_and_Band_Structure/7.1.05:_Semiconductor_p-n_Junctions) The built-in electric field can be experimentally measured indirectly through capacitance-voltage (C-V) profiling, a non-destructive technique that applies a small AC signal superimposed on a DC reverse bias to measure junction capacitance as a function of voltage. From the C-V data, the doping profile is extracted using N(x) = -\frac{2}{q \epsilon A^2 \frac{d(1/C^2)}{dV}}, where C is capacitance, A is junction area, and x is depletion width, allowing reconstruction of the field via integration of the charge density from Poisson's equation. This method has been widely used to verify field strengths in silicon and compound semiconductor junctions, confirming theoretical predictions with profiles accurate to within 10-20% for abrupt junctions.^[23]^[24]^[25]

Potential barrier

In a p–n junction at thermal equilibrium, the potential barrier, also known as the built-in potential V_{bi}, arises from the diffusion of majority carriers across the interface between the p-type and n-type regions, leading to charge separation and Fermi level alignment. This potential opposes further carrier diffusion, establishing a stable electrostatic barrier that governs carrier transport. The derivation of V_{bi} stems from the condition that the electrochemical potential (Fermi level) must be constant throughout the structure, requiring the electrostatic potential difference to compensate for the initial difference in Fermi levels between the isolated p-type and n-type materials.^[16] The magnitude of the built-in potential is expressed as

V_{bi} = \frac{kT}{q} \ln \left( \frac{N_a N_d}{n_i^2} \right),

where k is Boltzmann's constant, T is the absolute temperature, q is the elementary charge, N_a and N_d are the acceptor and donor doping concentrations, respectively, and n_i is the intrinsic carrier concentration. This formula reflects the exponential dependence of carrier concentrations on the Fermi level position relative to the band edges in the respective regions. In the equilibrium energy band diagram, the conduction and valence bands exhibit upward bending in the p-type region and downward bending in the n-type region, with the total band bending equal to q V_{bi}, forming the potential barrier for both electrons and holes.^[15] At equilibrium, the potential barrier ensures a dynamic balance between diffusion and drift currents: the concentration gradient drives diffusive flow of electrons from the n-side to the p-side and holes in the opposite direction, but the resulting built-in electric field generates an equal and opposite drift current, yielding zero net current across the junction. This balance is fundamental to the rectifying behavior of the device. The built-in potential decreases with increasing temperature due to the strong temperature dependence of n_i, which follows n_i \propto T^{3/2} \exp(-E_g / 2kT) where E_g is the bandgap energy; for silicon p–n junctions at 300 K, V_{bi} typically ranges from 0.6 to 0.8 V depending on doping levels.^[16]^[15] The concept of the potential barrier in p–n junctions is analogous to the contact potential in metal-semiconductor (Schottky) junctions, where the barrier height similarly originates from the alignment of the metal work function with the semiconductor's Fermi level, though the distributed space charge in p–n junctions leads to a broader depletion region compared to the abrupt interface in Schottky contacts.^[15]

Biased conditions

Forward bias effects

In forward bias, a positive voltage V is applied to the p-side relative to the n-side of the p-n junction, which reduces the effective potential barrier from the equilibrium built-in voltage V_{bi} to V_{bi} - V.^[15] This applied voltage opposes the built-in electric field, allowing majority carriers to overcome the barrier more easily and cross the junction. As a result, the depletion region, which forms due to the separation of charge carriers at the interface, begins to shrink because the external field counteracts the internal one.^[26] The width of the depletion region W decreases with increasing forward bias according to the relation W \propto \sqrt{V_{bi} - V}, leading to a weakening of the electric field within the region.^[27] This reduction in width and field strength facilitates the injection of majority carriers across the junction: electrons from the n-side into the p-side and holes from the p-side into the n-side.^[28] Once injected, these carriers become minorities in the opposite regions and diffuse away from the junction, generating diffusion currents that dominate the total current flow.^[15] Analysis of these effects typically assumes low-level injection, where the injected minority carrier concentration remains much smaller than the equilibrium majority carrier concentration, ensuring that the majority carrier distribution is not significantly perturbed.^[29]^[30] For silicon p-n junctions, significant current flow begins around a turn-on voltage of approximately 0.7 V at room temperature, beyond which the current rises exponentially with further increases in voltage due to the enhanced carrier injection.^[31]^[32]

Reverse bias effects

In reverse bias, a negative voltage is applied to the p-type region relative to the n-type region, which enhances the built-in electric field and increases the potential barrier height to V_{bi} - V, where V < 0 is the applied bias and V_{bi} is the built-in voltage.^[33] This configuration opposes the diffusion of majority carriers across the junction, effectively blocking significant current flow.^[32] The depletion region widens under reverse bias as the enhanced electric field sweeps mobile charges farther from the junction interface, reducing the effective doping density in the space-charge layer. The depletion width W increases proportionally to \sqrt{V_{bi} + |V|}, leading to a higher risk of reaching breakdown conditions at sufficiently large reverse voltages.^[15]^[32] The small current that flows in reverse bias, known as the reverse saturation current I_s, arises primarily from the drift of thermally generated minority carriers—electrons in the p-region and holes in the n-region—that are swept across the widened depletion region by the strong electric field.^[34]^[35] These minority carriers are extracted from the neutral regions adjacent to the depletion edges, maintaining a nearly constant I_s independent of the bias magnitude until breakdown occurs.^[36] At high reverse biases, typically exceeding 5 V for silicon junctions, breakdown mechanisms can initiate significant conduction. Zener breakdown dominates in heavily doped junctions (doping > 10^{18} cm^{-3}), where quantum tunneling allows electrons to cross the narrow, high-field depletion region at low voltages (< 5 V).^[37] In contrast, avalanche breakdown occurs in more lightly doped junctions at higher voltages, driven by impact ionization where accelerated carriers gain sufficient kinetic energy to generate additional electron-hole pairs, leading to a multiplicative current increase.^[15]^[32] The junction capacitance decreases under reverse bias due to the expanded depletion width, which acts as the dielectric thickness in a parallel-plate-like capacitor model, with capacitance C_j \propto 1 / \sqrt{V_{bi} + |V|}.^[28] This voltage-dependent behavior is utilized in varactor diodes for tuning applications.^[38]

Electrical models

Depletion width calculation

The depletion width in a p–n junction refers to the total extent of the space-charge region where mobile carriers are depleted, calculated under the depletion approximation for an abrupt junction. This approximation assumes that all dopants are ionized within the depletion region, free carrier concentrations are negligible there, and the doping profile changes abruptly at the metallurgical junction. The derivation begins with Poisson's equation, which relates the electric potential V(x) to the charge density \rho(x): \frac{d^2 V}{dx^2} = -\frac{\rho(x)}{\varepsilon}, where \varepsilon = \varepsilon_r \varepsilon_0 is the permittivity of the semiconductor, with \varepsilon_0 the vacuum permittivity and \varepsilon_r \approx 11.7 for silicon.^[39]^[19] In the p-type region (-x_p < x < 0), the charge density is \rho = -q N_A due to ionized acceptors, where q is the elementary charge and N_A the acceptor concentration. Integrating Poisson's equation twice yields a parabolic potential profile, with the electric field E(x) = -\frac{dV}{dx} being linear and zero at the depletion edge x = -x_p. Similarly, in the n-type region ($0 < x < x_n), \rho = q N_D from ionized donors, leading to E(x) linear and zero at x = x_n. Continuity of E and V at x = 0, along with the total potential drop V_{bi} - V across the junction (where V_{bi} is the built-in potential and V the applied bias, positive for forward), determines the edges. Charge neutrality requires N_A x_p = N_D x_n.^[19]^[40] The resulting total depletion width is

W = x_p + x_n = \sqrt{\frac{2\varepsilon (V_{bi} - V)}{q} \left( \frac{1}{N_A} + \frac{1}{N_D} \right)},

valid for both equilibrium (V = 0) and biased conditions, with V < V_{bi} to ensure W > 0. For symmetric doping (N_A = N_D = N), this simplifies to W = \sqrt{\frac{4\varepsilon (V_{bi} - V)}{q N}}. In one-sided junctions, such as when N_A \gg N_D, the depletion extends primarily into the lightly doped side, yielding W \approx \sqrt{\frac{2\varepsilon (V_{bi} - V)}{q N_D}} (or vice versa).^[19]^[17]^[40] The width W decreases under forward bias (V > 0) as the potential barrier reduces, narrowing the space-charge region, and increases under reverse bias (V < 0) to accommodate the enhanced field. This voltage dependence scales as \sqrt{V_{bi} - V}, highlighting the strong influence of bias on junction capacitance and breakdown characteristics. Material parameters like \varepsilon directly affect W, with higher \varepsilon_r leading to wider regions for given dopings.^[15]^[41] The depletion approximation holds for abrupt junctions under low-level injection conditions, where injected carriers do not significantly alter the doping profile, and for reverse biases below tunneling regimes. It incurs errors in linearly graded junctions, where the charge density varies continuously, requiring a different cubic potential solution, or in high-injection scenarios where free carriers screen the field.^[17]^[42]

Current-voltage characteristics

The current-voltage (I-V) characteristics of a p-n junction diode describe the relationship between the applied voltage V and the resulting current I flowing through the device, which is fundamentally governed by carrier transport mechanisms across the junction. In the ideal case, this relationship is captured by the Shockley diode equation, derived from the continuity of minority carrier diffusion and drift under low-level injection assumptions. The equation is given by

I = I_s \left[ \exp\left( \frac{qV}{kT} \right) - 1 \right],

where I_s is the reverse saturation current, q is the elementary charge, k is Boltzmann's constant, and T is the absolute temperature.^[16] This model assumes negligible series resistance and generation-recombination effects, focusing on diffusion-dominated transport. For forward bias (V > 0), the exponential term dominates, leading to a rapid increase in current as the potential barrier is reduced, allowing minority carriers to diffuse across the junction. In reverse bias (V < 0), the current approaches -I_s, representing the thermally generated minority carriers that diffuse to the junction and are swept away by the built-in field.^[16] The reverse saturation current I_s quantifies the thermally generated leakage current and is expressed as

I_s = q A \left( \frac{D_n n_i^2}{L_n N_A} + \frac{D_p n_i^2}{L_p N_D} \right),

where A is the junction area, D_p and D_n are the diffusion coefficients for holes and electrons, L_n and L_p are the corresponding diffusion lengths, N_A and N_D are the acceptor and donor doping concentrations, and n_i is the intrinsic carrier concentration.^[16] This expression arises from integrating the minority carrier diffusion equations in the quasi-neutral regions, assuming one-dimensional transport and Boltzmann statistics. In forward bias, the total current is primarily diffusion current due to injected minority carriers; in reverse bias, it is dominated by generation of electron-hole pairs in the depletion region, though the ideal model approximates this as diffusion of thermally generated carriers from the bulk.^[43] Real diodes deviate from the ideal equation due to non-radiative recombination and other effects, leading to the introduction of an ideality factor \eta in the generalized form:

I = I_s \left[ \exp\left( \frac{qV}{\eta kT} \right) - 1 \right].

Here, \eta = 1 for diffusion-dominated transport in long-base diodes, while \eta = 2 applies when Shockley-Read-Hall recombination in the depletion region dominates, particularly at low forward biases.^[43] At high forward currents, series resistance R_s in the neutral regions causes a deviation from exponential behavior, manifesting as a linear increase in voltage drop (I R_s) and reducing the effective turn-on sharpness.^[43] The semi-logarithmic plot of current versus voltage (\log I vs. V) provides a diagnostic tool for analyzing diode characteristics, yielding a straight line in the exponential region with slope q / (\eta kT) (approximately 60 mV/decade per \eta = 1 at room temperature) and y-intercept at V = 0 equal to -I_s. This plot allows extraction of \eta, I_s, and barrier height from experimental data.^[16] Temperature significantly influences the I-V characteristics through its effects on carrier concentrations and mobilities. The saturation current I_s increases strongly with temperature, approximately doubling for every 10°C rise near room temperature due to the exponential dependence on the intrinsic carrier concentration n_i \propto \exp(-E_g / 2kT).^[44] Consequently, the forward voltage V at a fixed current decreases by about 2 mV/°C, reflecting the reduced barrier height and increased thermal generation.^[44]

Practical aspects

Fabrication techniques

p–n junctions are fabricated using several established techniques that introduce dopants into semiconductors to create the necessary p-type and n-type regions. Diffusion involves thermally driving dopants into the substrate, typically boron for p-type doping in silicon at temperatures around 1000°C to form shallow junctions.^[45] This process relies on solid-state diffusion from a dopant source, such as a predeposited layer, to achieve controlled penetration depths, often on the order of micrometers, suitable for older device generations.^[46] Ion implantation accelerates dopant ions, such as arsenic for n-type regions, into the semiconductor lattice at energies ranging from keV to MeV, enabling precise control over doping profiles and depths as shallow as tens of nanometers.^[47] Following implantation, high-temperature annealing, typically between 800°C and 1100°C, activates the dopants by placing them on substitutional lattice sites and repairs implantation-induced lattice damage through recrystallization.^[48] This method has become dominant in modern integrated circuit production due to its compatibility with lithography for selective area doping.^[49] Epitaxial growth deposits crystalline layers with in-situ doping via chemical vapor deposition (CVD), allowing abrupt and precise p–n interfaces essential for devices like bipolar transistors.^[50] In this process, precursors such as silane and diborane are used to grow silicon layers on a substrate at temperatures of 900–1200°C, incorporating dopants during growth to form tailored profiles without post-deposition diffusion.^[51] Variants like metalorganic CVD (MOCVD) extend this to compound semiconductors, ensuring high uniformity and low defect densities.^[52] Fabrication challenges include repairing lattice damage from ion implantation, which can introduce point defects and dislocations if annealing is insufficient, potentially degrading carrier mobility and junction quality.^[53] Achieving uniformity across large wafers, such as 300 mm silicon substrates, requires precise control of temperature gradients and gas flows to minimize variations in doping concentration and junction depth.^[54] Post-2000 advances have focused on ultra-shallow junctions for sub-10 nm technology nodes, incorporating techniques like molecular monolayer doping combined with rapid thermal annealing to form junctions with depths below 5 nm while minimizing dopant diffusion.^[55] Atomic layer deposition (ALD) has enabled conformal thin films for dopant sources or barriers, facilitating precise control in advanced nodes and reducing short-channel effects in transistors.^[56] These developments, often integrated with low-temperature processes, support scaling in high-performance computing applications.^[57]

Common applications

The p–n junction forms the core of semiconductor diodes, enabling rectification in electronic circuits where alternating current is converted to direct current by permitting unidirectional current flow under forward bias while blocking it under reverse bias.^[15] Zener diodes, a specialized type of p–n junction diode, operate in the reverse breakdown region to provide stable voltage regulation, maintaining a nearly constant voltage across a load despite variations in input voltage or current, typically in the range of 2 to 200 V.^[58] Light-emitting diodes (LEDs) utilize forward-biased p–n junctions in direct-bandgap semiconductors, where injected electrons and holes recombine radiatively, emitting photons for applications in displays, indicators, and lighting.^[59] In bipolar junction transistors (BJTs), two p–n junctions—one between the emitter and base, and another between the base and collector—enable current amplification by controlling a large collector current with a small base current, forming the basis for switching and amplifying circuits in analog and digital electronics. Solar cells rely on the photovoltaic effect in p–n junctions, where absorbed photons generate electron-hole pairs separated by the built-in electric field, producing a photocurrent; single-junction silicon solar cells typically achieve efficiencies of 20–24% in commercial modules as of 2025.^[60]^[61] Varactor diodes exploit the voltage-dependent capacitance of a reverse-biased p–n junction, where varying the reverse bias adjusts the depletion width and thus the junction capacitance, allowing tuning in radio-frequency (RF) circuits such as voltage-controlled oscillators and filters.^[62] In integrated circuits (ICs), particularly complementary metal-oxide-semiconductor (CMOS) technology, billions of p–n junctions are incorporated as isolation structures in doped wells and as components in diodes and protection circuits, supporting the dense packing of transistors on modern chips.^[63]

Historical context

Early theoretical foundations

The theoretical foundations of the p–n junction emerged in the early 20th century, building on observations of rectifying behavior in metal-semiconductor contacts before the full development of quantum mechanics for solids. In the 1920s, researchers explored non-ohmic junctions as potential rectifiers, with Lars O. Grondahl and Paul H. Geiger discovering the rectifying properties of the copper-cuprous oxide interface in 1926. This junction exhibited asymmetric current flow, allowing conduction in one direction while blocking the reverse, which they attributed to electrochemical effects at the interface without invoking detailed carrier transport models. Their work laid early groundwork for understanding barrier layers in rectifier devices, though explanations remained qualitative and classical. By the 1930s, classical models began incorporating diffusion and space charge concepts to explain rectifier characteristics. J. A. Becker at Bell Laboratories analyzed copper oxide rectifiers, proposing that current flow involved diffusion of charge carriers across a space charge region near the junction, leading to rectification due to the imbalance between forward and reverse diffusion currents. This model predicted the formation of a depleted layer where mobile carriers were scarce, influencing the voltage drop and current asymmetry, and was applied to practical devices like power rectifiers. These ideas relied on drift and diffusion principles from classical physics, predating quantum descriptions of carrier behavior. The advent of quantum mechanics in the late 1920s enabled more sophisticated band theory for semiconductors, which Shockley and colleagues at Bell Labs extended to junctions in the 1930s. Alan H. Wilson's 1931 theory introduced energy bands in solids, distinguishing semiconductors from metals and insulators by partially filled bands or impurity levels that could generate mobile electrons and holes. Shockley built on this, applying band theory to rectifier junctions in his 1939 paper, where he modeled the potential barrier at the interface using quantum-derived carrier densities and Fermi levels, predicting exponential current-voltage behavior from thermal generation across the barrier. This work marked a pivotal shift from classical to quantum-informed models, providing the conceptual framework for later p–n junction theory.^[64]^[65]

Experimental milestones

The invention of the transistor in 1947 marked a pivotal experimental validation of p-n junction principles, demonstrating amplification in semiconductors for the first time. At Bell Laboratories, physicists John Bardeen and Walter Brattain constructed the initial point-contact transistor using a sliver of n-type germanium with two closely spaced gold contacts acting as point junctions to inject and collect carriers, achieving a current gain of about 18 on December 16, 1947.^[66] This device, while fragile and prone to instability, served as a crucial precursor, highlighting the role of surface states and carrier injection at metal-semiconductor contacts, which informed subsequent refinements.^[67] William Shockley, also at Bell Labs, built on this by theorizing a more stable structure in early 1948, proposing a bulk p-n-p junction transistor where carriers flow through the semiconductor volume via two back-to-back p-n junctions in germanium, avoiding the point-contact's mechanical vulnerabilities.^[68] Experimental realization of Shockley's junction transistor followed in 1951, when Gordon K. Teal and Morgan Sparks at Bell Labs grew large single crystals of germanium and fabricated the first n-p-n grown-junction devices by doping during crystal pulling, yielding transistors with power gains up to 50 dB at low currents and voltages.^[69] These germanium-based units operated reliably as amplifiers and switches, confirming the theoretical predictions of minority carrier injection across p-n junctions.^[70] By the mid-1950s, efforts shifted to silicon for enhanced thermal stability and reduced leakage, as germanium's narrow bandgap (0.66 eV) limited high-temperature performance compared to silicon's 1.12 eV. In January 1954, Morris Tanenbaum produced the first silicon p-n-p junction transistor at Bell Labs using a modified grown-junction method, followed by Calvin S. Fuller and Gerald L. Pearson's diffusion technique later that year, which formed precise p-n junctions by diffusing boron into n-type silicon wafers, enabling devices with superior reverse-bias breakdown voltages and operational reliability up to 150°C.^[71] Empirical validation of the theoretical current-voltage (I-V) characteristics for p-n junctions advanced through precise measurements in the 1950s. James L. Moll and colleagues at Bell Labs extended Shockley's 1949 diode equation—describing saturation current and exponential forward bias dependence—to transistors via the Ebers-Moll model, analyzing large-signal I-V curves of germanium and silicon junction devices to confirm parameters like ideality factor and reverse saturation current, with experimental data matching theory within 10-20% across bias ranges. These studies, involving oscilloscope-traced I-V plots under varying temperatures and doping, established the model's accuracy for predicting diode and transistor behavior, underpinning reliable circuit design.^[72] The application of p-n junctions extended to computing with the advent of integrated circuits in the late 1950s. On September 12, 1958, Jack Kilby at Texas Instruments demonstrated the first IC prototype on a germanium substrate, integrating resistors via bulk germanium, capacitors from p-n junctions, and a transistor—all interconnected without discrete wires—oscillating at 1 MHz to validate monolithic fabrication. Independently, in 1959, Robert Noyce at Fairchild Semiconductor patented a silicon-based monolithic IC using diffused p-n junctions for isolation and aluminum wiring over oxide for connections, enabling scalable production of multiple transistors on a single chip and revolutionizing digital logic circuits.^[73]

References

[1]
[PDF] William Shockley - Nobel Lecture
A p-n junction is the simplest so-called compositional structure in semicon- ductor electronics. By a compositional structure is meant a single crystal of.<|control11|><|separator|>
[2]
Formation of a PN-Junction - PVEducation
This region is called the "depletion region" since the electric field quickly sweeps free carriers out, hence the region is depleted of free carriers. A "built- ...
[3]
[PDF] shockley-pn-junction.pdf
The Theory of p-n Junctions in Semiconductors and p-n. Junction Transistors. By W. SHOCKLEY. In a single crystal of semiconductor the impurity concentration may ...
[4]
PN Junction Theory for Semiconductor Diodes - Electronics Tutorials
A PN-junction is formed when an N-type material is fused together with a P-type material creating a semiconductor diode.
[5]
[PDF] Electrons and Holes in Semiconductors
In an intrinsic Si crystal, there are few mobile electrons and holes. Their concentrations are equal to ni (~1010cm–3 for Si), the intrinsic carrier ...
[6]
[PDF] Lecture 3 Introduction to Semiconductors and Energy Bandgaps
Bandgap Energy: Energy required to remove a valence electron and allow it to freely conduct. Intrinsic Semiconductor: A “native semiconductor” with no dopants.
[7]
Types of Wafer Substrates | BYU Cleanroom
Silicon, Si - the most common semiconductor, atomic number 14, energy gap Eg = 1.12 eV - indirect bandgap; crystal structure - diamond, lattice constant 0.543 ...
[8]
[PDF] california state university, northridge
The following equation expresses intrinsic carrier concentration. ni = √𝑛 ... ni = sqrt(Nc*Nv)*exp(-(Eg*q)/(2*k*T)). % Calculating Thermal voltage. Vt ...
[9]
[PDF] Lecture 7: Extrinsic semiconductors - Fermi level - An-Najah Staff
For an n-type semiconductor with Nd donors which are fully ionized, n = Nd when Nd ni, so that the Fermi level is shifted above the center of the band gap ...
[10]
[PDF] The dopant density and temperature dependence of electron ...
where (^^(o) = 0.045 eV is the ionization energy of the phosphorus donor; with a = 3.1 x. 10~^, eq (3.15) gives a zero ionization energy for N^^ = 3 x. 10^^ cm~ ...
[11]
[PDF] Lecture-2 - Electrical Engineering
Suitable pentavalent impurities are antimony, phosphorus, and arsenic. Such impurities donate excess electron carriers and are referred to as donor, or n-type, ...
[12]
[PDF] NASA TN 0-7517
dp/dT _ -dpp/dT. (4). P. Mp. The ionization energy of boron in silicon is 0.045 eV, and most of the boron is ionized at temperatures well below room temperature ...
[13]
Electrical Properties of Silicon Containing Arsenic and Boron - ADS
Analysis of extrinsic carrier concentration indicates the ionization energy of arsenic donor levels to be 0.049 ev and of boron acceptor levels to be 0.045 ev ...<|control11|><|separator|>
[14]
[PDF] Electrons and holes in pure Silicon
Jun 4, 2012 · Intrinsic Carrier Concentration (ni) free-electron concentration is equal to hole concentration (whole pure Si crystal is charge neutral), and ...
[15]
[PDF] PN and Metal–Semiconductor Junctions
As a device, it is called a rectifier or a diode. The PN junction is. the basic structure of solar cell, light-emitting diode, and diode laser, and is present. ...
[16]
Shockley - 1949 - Bell System Technical Journal - Wiley Online Library
The theory of potential distribution and rectification for p-n junctions is developed with emphasis on germanium. The currents across the junction are carried ...
[17]
Solving for Depletion Region - PVEducation
Assumptions are: · Depletion approximation: the electric field is confined to the junction region and there is no electric field in the quasi-neutral regions.
[18]
The theory of p-n junctions in semiconductors ... - Semantic Scholar
The theory of potential distribution and rectification for p-n junctions is developed with emphasis on germanium, resulting in an admittance for a simple ...
[19]
[PDF] Lecture 4 - pn Junctions: Electrostatics - MIT OpenCourseWare
Finally, putting these in Poisson's equation, a single equation for φ(x) in ... Integrating again gives the electrostatic potential: = -qNDnxn/εSi φ(x).
[20]
[PDF] Diodes in the Dark - MIT OpenCourseWare
2. This built-in electric field is established at a pn- junction because of the balance of electron & hole drift and diffusion currents.
[21]
[PDF] pn JUNCTION DEVICES AND LIGHT EMITTING DIODES
Apr 10, 2001 · A semiconductor device structure that has junctions between different bandgap materials is called a heterostructure device (HD). The refractive ...
[22]
[PDF] Semiconductor Junctions - TU Delft OpenCourseWare
This approximation (the depletion approximation) assumes that the space-charge density, ρ, is zero in the quasi-neutral regions and it is fully determined ...
[23]
[PDF] Capacitance-voltage Profiling Techniques for Characterization of ...
Capacitance-voltage profiling is a non-destructive technique for measuring barrier capacitance of semiconductor junctions, used to characterize doping profiles.
[24]
[PDF] C-V measurement for semiconductor junctions - arXiv
C-V measurement measures junction barrier capacitance at different reverse voltage biases to infer doping concentration and is non-destructive.
[25]
4.3 Electrostatic analysis
A capacitance-voltage measurement also provides the doping density profile of one-sided p-n diodes. For a p+-n diode, one obtains the doping density from ...<|separator|>
[26]
[PDF] ECE 255 - Purdue Engineering
Aug 31, 2017 · When the voltage is forward bias, it generates an electric field that cancels the original electric field in the transition region, reducing the ...
[27]
[PDF] ISs tS - DSpace@MIT
the junction built-in voltage Vbi and the applied voltage Va because W is proportional to the square root of the difference between Vbi and Va for a step ...
[28]
[PDF] Electronics I - PN Junction Chapter 3
• The fixed ions left behind create an electric field. • The electric field opposes the diffusion of holes in the N region and free electrons in the P region.Missing: Shockley | Show results with:Shockley
[29]
[PDF] Lecture 14 - MIT
Law of the Junction is valid if minority carrier concentration is less than equilibrium majority concentration. This condition is called Low Level. Injection.
[30]
[PDF] ECE 440 Lecture 22 : Quantitative Current Flow in a PN Junction
Mar 8, 2010 · We assume low level injection – majority charge concentrations do not ... Minority carrier concentration usually varies the most and the majority.
[31]
[PDF] Semiconductors
Due to the very rapidly increasing exponential, the diode looks like it turns on at at VF ≈ 0.7 V in Silicon diodes. The IV curve is actually exponential all ...
[32]
[PDF] ECE 255, PN Junction and Diode - Purdue Engineering
Mar 14, 2018 · A pn junction is important for understanding diodes. It forms a depletion region due to diffusion, and its behavior changes with forward and ...Missing: seminal paper
[33]
Bias of PN Junctions - PVEducation
The increased diffusion from one side of the junction to the other causes minority carrier injection at the edge of the depletion region. These carriers ...Missing: breakdown | Show results with:breakdown
[34]
[PDF] Lecture 15 - The pn Junction Diode (I) IV Characteristics
Nov 1, 2005 · • If minority carrier injection: ⇒ carrier concentration above equilibrium prevails. ⇒ recom recombination n + p. Si-Si bond recombination.Missing: assumption | Show results with:assumption
[35]
[PDF] 28 pn junctions
The built-in potential is therefore the internal potential energy required to cancel the diffusive flow of carriers across the junction and should be exactly ...<|control11|><|separator|>
[36]
[PDF] ECE 440 Lecture 32 : Minority and Majority Carrier Currents
Nov 9, 2009 · Excess electrons on p-side: Excess holes on n-side: Now we understand the hole diffusion current at any point… Hole diffusion proportional to ...
[37]
[PDF] ECE 340 Lecture 26 : Reverse Bias Breakdown
Oct 21, 2011 · The Zener effect is the breakdown mechanism if the reverse bias required to force breakdown occurs at low voltages. Consider the heavily doped ...
[38]
[PDF] ECE 340 Lecture 27 : Junction Capacitance
Oct 29, 2012 · Reverse bias brings the conduction band very close to the valence band. This brings many occupied states on the p-side into energetic alignment.
[39]
General Properties of Silicon - PVEducation
General Properties of Silicon ; Relative Permittivity (εr), 11.7 ; Electron Affinity, 4.05 eV ; Electron Diffusion Coefficient (De), kT/q µ ; Hole Diffusion ...
[40]
[PDF] Depletion width approximation derivation - EE 3161 spring 2008
Derivation of Depletion Widths in Depletion Approximation for an abrupt junction. ... And the total width of the depletion region is. W= X₁ + to. W= QE Voi Na+ ...
[41]
[PDF] PN Junctions
As the voltage across a pn junction is increased, the depletion region widens and the E field increases. Eventually, a limit must be reached. When this happens, ...
[42]
[PDF] Depletion Approximation vs. “Exact” Solution of the PN Junction
Sep 29, 2005 · The depletion approximation is extremely useful, when it is valid, so it is important to understand its limitations.
[43]
https://ieeexplore.ieee.org/document/4056679
[44]
Diode Equation | PVEducation
The diode law for silicon - current changes with voltage and temperature. For a given current, the curve shifts by approximately 2 mV/°C. The light blue ...Missing: original | Show results with:original
[45]
Enhanced photo-sensitivity in a Si photodetector using a near-field ...
Jun 20, 2019 · The p-type layer was also fabricated by the thermal diffusion ... Boron doping was performed in a furnace at 1000 °C for 120 min with ...<|control11|><|separator|>
[46]
Fabrication of Ultra-Shallow P+-N and N+-P Junctions by Diffusion ...
Oct 24, 1993 · In-situ doping was achieved with B2H6 and PH3 for p-type and n-type doping, respectively. Boron and phosphorus diffusion in ion-implanted Si0.7 ...Missing: technique | Show results with:technique
[47]
Fabrication of p-n junctions in ZnO by arsenic ion implantation ...
We report the properties of p-n junctions produced in ZnO films by As+ ion implantation followed by annealing in atomic oxygen.
[48]
https://ieeexplore.ieee.org/document/10219243
[49]
ion implantation: basics to device fabrication
Ion implantation offers one of the best examples of a topic that starting from the basic research level has reached the high technology level within the ...
[50]
Rapid Thermal Chemical Vapour Deposition of Epitaxial Si and SiGe
Epitaxy of Si on Si substrates by CVD is an old technique. Developed to improve the performance of bipolar transistors and integrated circuits, ...Missing: junction | Show results with:junction
[51]
Demonstration of 1.27 kV Etch-Then-Regrow GaN p-n Junctions ...
Compared with diffusion and implantation methods, epitaxial regrowth is still regarded as the most feasible method for the selective-area doping [9]–[12].Missing: techniques | Show results with:techniques
[52]
Heteroepitaxially grown homojunction gallium oxide PN diodes ...
We present the successful fabrication of a Ga 2 O 3 pn diode utilizing ion implantation technology. The Ga 2 O 3 epilayers were grown on c-plane sapphire ...
[53]
Al-implantation induced damage in 4H-SiC - ScienceDirect.com
In this work, we explore the defects created during Al-ion implantation and post implantation annealing using low-energy muon spin rotation (LE- SR) ...
[54]
Silicon Crystal Growth and Wafer Technologies - IEEE Xplore
Commercially avail- able epitaxial and silicon-on-insulator (SOI) wafers are com- monplace and wafer diameters have increased from 0.5 in to. 300 mm in ...Missing: repair | Show results with:repair
[55]
A doping-less junction-formation mechanism between n-silicon and ...
Oct 16, 2017 · The nanometer-thin boron was used as a capping and protective layer, and as a window for the optical characterization tests. The electrical ...
[56]
Modulation Doping of Silicon using Aluminium-induced Acceptor ...
Apr 20, 2017 · Molecular monolayer doping was developed to achieve ultra-shallow p-n junctions via RTA-diffusion. Since this concept is based on ...
[57]
Low-temperature poly-Si nanowire junctionless devices with gate-all ...
Jun 22, 2012 · Subsequently, the high-κ/metal gate stack structure was formed using the atomic-layer deposition system to conformally deposit 10-nm Al2O3 and ...
[58]
[PDF] 4.4 Operation in the Reverse Breakdown Region — Zener Diodes
Feb 27, 2012 · A Zener diode is a pn junction diode meant to operate in breakdown, where voltage is nearly independent of current, acting like a voltage ...
[59]
[PDF] Chapter 7 Semiconductor Light Emitting Diodes and Solid State ...
Semiconductor light emitting diodes are forward biased pn junction diodes in which electron-hole recombination due to spontaneous emission in the junction ...
[60]
Solar Photovoltaic Cell Basics | Department of Energy
Multijunction solar cells have demonstrated efficiencies higher than 45%, but they're costly and difficult to manufacture, so they're reserved for space ...
[61]
Silicon Solar Cells - Stanford University
Nov 13, 2015 · Its energy conversion efficiency was 1%, so for every 100 Joules of solar energy that fell upon it, it created 1 Joule of electrical energy. [1] ...<|separator|>
[62]
[PDF] VCO Design Project - ECE218B Winter 2011 - UCSB ECE
Feb 18, 2011 · The varactor Q is typically 150 at 200 MHz and 5 volts reverse bias. You should avoid biasing the varactor under 1 volt reverse bias so that the ...
[63]
https://ieeexplore.ieee.org/document/6467920
[64]
The theory of electronic semi-conductors - Journals
If there is no external electric field, the number of electrons moving in any direction is equal to the number moving in the opposite direction. The action of a ...
[65]
The theory of crystal rectifiers | Proceedings of the Royal Society of ...
It is the purpose of this paper to criticize existing theories of this effect and to suggest a new one. In any crystal rectifier we have two contacts to ...
[66]
1947: Invention of the Point-Contact Transistor | The Silicon Engine
John Bardeen & Walter Brattain achieve transistor action in a germanium point-contact device in December 1947.
[67]
How the First Transistor Worked - IEEE Spectrum
Nov 20, 2022 · The first recorded instance of a working transistor was the legendary point-contact device built at AT&T Bell Telephone Laboratories in the fall of 1947.
[68]
1948: Conception of the Junction Transistor | The Silicon Engine
William Shockley conceived the junction transistor in 1948, based on the p-n junction, using a three-layer sandwich of n and p type semiconductors.
[69]
Junction Transistors | Phys. Rev. - Physical Review Link Manager
References (17). W. Shockley, Bell System Tech. J. 28, 435-489 (1949); W. Shockley, Electrons and Holes in Semiconductors (D. van Nostrand Company, Inc., New ...Missing: paper | Show results with:paper
[70]
The Nobel Prize in Physics 1956 - NobelPrize.org
The Nobel Prize in Physics 1956 was awarded jointly to William Bradford Shockley, John Bardeen and Walter Houser Brattain for their researches on ...
[71]
1954: Silicon Transistors Offer Superior Operating Characteristics
In January 1954 Bell Labs chemist Morris Tanenbaum fashioned the first silicon transistor using a variation on Morgan Sparks and Gordon Teal's grown-junction ...
[72]
Large-Signal Behavior of Junction Transistors - Semantic Scholar
Large-Signal Behavior of Junction Transistors · J. Ebers, J. Moll · Published in Proceedings of the IRE 1 December 1954 · Engineering, Physics.
[73]
1959: Practical Monolithic Integrated Circuit Concept Patented
Kilby is credited with building the first working circuit with all components formed using semiconductor material; Noyce with the metal-over-oxide ...