CN219163404U - Semiconductor device for radio frequency power amplifier - Google Patents

Abstract

The utility model provides a semiconductor device for a radio frequency power amplifier, comprising: a substrate layer configured to be formed by GaAs; a sub-collector layer configured on the substrate layer and configured to be formed by n+ -doped GaAs; a collector layer configured on the sub-collector layer and configured to be formed by an n-doped wide bandgap material; a collector metal layer disposed on the collector layer; a base layer configured on the collector layer and configured to be formed by p+ -doped GaAs; a base metal layer disposed on the base layer; an emitter layer configured on the base layer and configured to be formed by n-type doped InGaP; an emitter cap layer configured on the base layer; and an emitter metal layer disposed on the emitter cap layer.

Description

Semiconductor device for radio frequency power amplifier

Technical Field

The present utility model relates to the field of wireless communications, and more particularly, to a semiconductor device of a radio frequency power amplifier for wireless communications.

Background

4G and/or 5G radio frequency Power Amplifiers (PA) used in mobile terminals such as cell phones are typically fabricated using gallium arsenide heterojunction bipolar transistor (GaAs HBT) technology. In the early GaAs HBT process, an AlGaAs/GaAs heterojunction interface is mainly used, and along with the development of an InGaP/GaAs heterojunction technology, the InGaP/GaAs heterojunction technology gradually becomes the main stream of the GaAs HBT technology.

Disclosure of Invention

An aspect of the present utility model is to provide a semiconductor device for a radio frequency power amplifier, comprising: a substrate layer configured to be formed by GaAs; a sub-collector layer configured on the substrate layer and configured to be formed by n+ -doped GaAs; a collector layer configured on the sub-collector layer and configured to be formed by an n-doped wide bandgap material, wherein a forbidden bandwidth of the wide bandgap material is greater than a forbidden bandwidth of GaAs; a collector metal layer disposed on the collector layer; a base layer configured on the collector layer and configured to be formed by p+ -doped GaAs; a base metal layer disposed on the base layer; an emitter layer configured on the base layer and configured to be formed by n-type doped InGaP; an emitter cap layer configured on the base layer; and an emitter metal layer disposed on the emitter cap layer, wherein the collector layer is configured such that the collectorThe maximum current and voltage swing at the layer is within a safe operating region of the semiconductor device, wherein the safe operating region of the semiconductor device is defined as a safe operating region when the semiconductor device is in a forward active region and the operating current thereof is in the order of milliamperes (mA) to amperes (a) of operating region. The breakdown voltage of the safe operating region is different from the conventional collector-base reverse breakdown voltage (BV _CBO ) Or collector-emitter reverse Breakdown Voltage (BV) _CEO ) In general, the breakdown voltage of the safe operating region is smaller than the collector-base reverse breakdown voltage (BV _CBO ) And a reverse Breakdown Voltage (BV) with the collector-emitter _CEO ) Proximity.

An aspect of the present utility model is to provide a semiconductor device for a radio frequency power amplifier, wherein the wide bandgap material includes In _x Ga _(1-x) P and In _x Ga _(1-x) As _y P _(1-y) Wherein x and 1-x represent the ratio of In and Ga elements, and y and 1-y represent the ratio of As and P elements.

An aspect of the present utility model is to provide a semiconductor device for a radio frequency power amplifier, wherein when the wide bandgap material is In _x Ga _(1-x) At P, the thickness of the collector layer is configured to be less than a first threshold, preferably 50nm.

An aspect of the present utility model is to provide a semiconductor device for a radio frequency power amplifier, wherein when the wide bandgap material is In _x Ga _(1-x) As _y P _(1-y) When the thickness of the collector layer is configured to be smaller than a second threshold value, preferably, the second threshold value is 50nm.

An aspect of the utility model is to propose a semiconductor device for a radio frequency power amplifier, wherein x is in a first ratio range of 0.45 to 0.55.

An aspect of the utility model is to propose a semiconductor device for a radio frequency power amplifier, wherein y is in a second ratio range of 0.45 to 0.55.

An aspect of the utility model is to propose a semiconductor device for a radio frequency power amplifier, wherein the semiconductor device is configured for a 5G radio frequency power amplifier, the 5G radio frequency power amplifier having an operating voltage with a maximum power of 4.4V-5.5V.

An aspect of the present utility model is to propose a semiconductor device for a radio frequency power amplifier, wherein the semiconductor device is applied in a power stage amplifying unit of the 5G radio frequency power amplifier.

Drawings

Fig. 1 is a cross-sectional view showing a semiconductor device for a radio frequency power amplifier;

fig. 2 is a schematic diagram showing the band structure of a semiconductor device employing an InGaP/GaAs SHBT process;

fig. 3 is a schematic diagram showing the difference between the open emitter, open base and normal operation conditions of a semiconductor device of a radio frequency power amplifier;

fig. 4 is a cross-sectional view of a semiconductor device for a radio frequency power amplifier according to an embodiment of the present utility model;

fig. 5 is a schematic diagram showing an energy band structure of a semiconductor device for a radio frequency power amplifier according to an embodiment of the present utility model;

fig. 6A shows a breakdown schematic of an energy band structure of a semiconductor device for a single heterojunction of a radio frequency power amplifier, and fig. 6B shows a breakdown schematic of an energy band structure of a semiconductor device for a double heterojunction of a radio frequency power amplifier;

fig. 7 is a comparative diagram showing the safe operating area of a semiconductor device of a radio frequency power amplifier fabricated using a GaAs SHBT process and a GaAs DHBT process;

fig. 8 is a cross-sectional view showing a semiconductor device for a radio frequency power amplifier according to another embodiment of the present utility model;

fig. 9 is a schematic diagram showing an energy band structure of a semiconductor device for a radio frequency power amplifier according to another embodiment of the present utility model; and

fig. 10 is a circuit frame diagram showing a radio frequency power amplifier including a semiconductor device according to an embodiment of the present utility model.

Detailed Description

Before proceeding with the following detailed description, it may be advantageous to set forth definitions of certain words and phrases used throughout this patent document. The terms "coupled," "connected," and derivatives thereof, refer to any direct or indirect communication between two or more elements, whether or not those elements are in physical contact with one another. The terms "transmit," "receive," and "communicate," and derivatives thereof, encompass both direct and indirect communication. The terms "include" and "comprise," as well as derivatives thereof, mean inclusion without limitation. The term "or" is inclusive, meaning and/or. The phrase "associated with … …" and its derivatives are intended to include, be included in, interconnect with, contain within … …, connect or connect with … …, couple or couple with … …, communicate with … …, mate, interleave, juxtapose, approximate, bind or bind with … …, have attributes, have relationships or have relationships with … …, etc. The term "controller" refers to any device, system, or portion thereof that controls at least one operation. Such a controller may be implemented in hardware, or a combination of hardware and software and/or firmware. The functionality associated with any particular controller may be centralized or distributed, whether locally or remotely. The phrase "at least one," when used with a list of items, means that different combinations of one or more of the listed items may be used, and that only one item in the list may be required. For example, "at least one of A, B, C" includes any one of the following combinations: A. b, C, A and B, A and C, B and C, A and B and C.

Definitions for other specific words and phrases are provided throughout this patent document. Those of ordinary skill in the art should understand that in many, if not most instances, such definitions apply to prior as well as future uses of such defined words and phrases.

In this patent document, the application combinations of circuit blocks and the division of sub-circuit blocks are for illustration only, and the application combinations of circuit blocks and the division of sub-circuit blocks may have different manners without departing from the scope of the disclosure.

Figures 1 through 9, discussed below, and the various embodiments used to describe the principles of the present disclosure in this patent document are by way of illustration only and should not be construed in any way to limit the scope of the disclosure. Those skilled in the art will appreciate that the principles of the present disclosure may be implemented in any suitably arranged system or device.

In the current mainstream terminals such as 4G mobile phones, the 4G radio frequency power amplifier used is usually manufactured by InGaP/GaAs HBT process. The InGaP/GaAs heterojunction is characterized in that an emitter adopts an InGaP structure, and a base and a collector adopt GaAs structures. The emitter and base are heterojunction, the base and collector are homojunctions, and thus the above process is known as the gallium arsenide single heterojunction process (GaAsSHBT). For the 4G communication system, the radio frequency power amplifier uses an operating voltage of about 3.4V to meet the requirements of the 4G communication system on linearity and transmitting power. For GaAs SHBT processes, the corresponding safe operating area (Safe Operation Area, SOA) can be satisfied for safe operation of the radio frequency power amplifier at 3.4V. It will be understood by those skilled in the art that 3.4V refers to the operating voltage at which the rf power amplifier emits maximum power, 3.4V is not a particular value, and may include 3.0V-3.5V.

In the 5G application scenario, for example, in the 5G mobile phone terminal, the 5G module terminal, the 5G small base station or the 5G internet of vehicles C-V2X, the requirement of the 5G communication system on the indexes such as linearity and transmitting power is greatly improved compared with the 4G LTE communication system. The 5G rf power amplifier requires an operating voltage of about 5V for high linearity and high power output. For example, a DC-DC voltage conversion circuit with Boost function is required in a 5G mobile phone to Boost the voltage of the power supply battery to 5V to provide a 5V working power supply for the radio frequency power amplifier. The SOA corresponding to the GaAs SHBT process is difficult to meet the requirement of safe operation of the radio frequency power amplifier under 5V, the situation that linearity or output power does not reach standards often occurs, and the radio frequency power amplifier is burnt out when serious. It will be understood by those skilled in the art that 5V refers to the operating voltage at which the rf power amplifier emits maximum power, 5V is not a particular value, and may include 4.4V-5.5V.

Some current GaAs SHBT processes are improved and improved on key technical parameters of breakdown voltages (BVCEO and BVCBO), and the design of a 5G radio frequency power amplifier can be nominally supported, but in practical application, the situation of linearity deterioration or chip burning still occurs.

Therefore, whether the requirements of the 5G radio frequency power amplifier are met or not is judged by using the technical indexes of breakdown voltages (BVCEO and BVCBO), and the 5G radio frequency power amplifier cannot be ensured to work normally.

Fig. 1 is a cross-sectional view showing a semiconductor device for a radio frequency power amplifier.

Referring to fig. 1, from the epitaxial layer structure, the semiconductor device includes, from top to bottom: an emitter metal layer 101, an emitter cap layer 102, an emitter n-type doped layer (InGaP) 103, a base metal layer 104, a base p+ -type doped layer (GaAs) 105, a collector metal layer 106, a collector n-type doped layer (GaAs) 107, a sub-collector n+ -type doped layer (GaAs) 108, and a substrate layer (GaAs) 109. In fig. 1, the emitter and base are heterojunction structures, and the base and collector are homojunction structures. For a radio frequency power amplifier prepared by using a GaAsHBT process, an internal transistor is of an up-down vertical structure, and a physical performance analysis method is generally used for carrying out physical performance analysis.

Fig. 2 is a schematic diagram showing the band structure of a semiconductor device employing an InGaP/GaAs SHBT process.

Referring to fig. 2, in the band structure of a 4G radio frequency power amplifier fabricated using an InGaP GaAs Single Heterojunction (SHBT) process, since minority electrons migrate faster, transistors typically use NPN structures, i.e., emitter E (Emitter InGaP) and collector C (Collector GaAs) are doped n-type, and Base B (Base GaAs) is doped p-type. The power core unit in the radio frequency power amplifier adopts a common-emitter amplifying structure, namely, an emitter is connected to a common-ground node. The power core unit is in the energy band structure of the forward active region. In an NPN structure in a forward active region, a B-E junction is forward conducted, electrons of an emitter can cross the B-E junction and are injected into a base region by the emitter region, the B-C junction is reverse biased, and electrons diffused into a B-C junction space charge region by the base region are all swept into a collector region by reverse bias voltage.

In the case of an open emitter, the reverse bias applied at the B-C junction generates a reverse current consisting essentially of the electron current that diffuses into the barrier region at the base region and the hole current that diffuses into the barrier region at the collector region. Since electrons are minority carriers in the base region and holes are minority carriers in the collector region, the total current is relatively weak, typically in the microampere (uA) order. When the reverse bias voltage is increased to a certain value, the electric field in the barrier region is strong, and electrons and holes in the barrier region generate larger kinetic energy due to the drifting action of the strong electric field. The electrons and holes with high energy excite electrons in full band to generate new electron and hole pairs; the new electron and hole pairs continue to excite other electrons in the full band … …. The multiple impact ionization caused by the above process generates a large number of electron and hole pairs, so that the number of carriers increases dramatically, and the larger reverse current generated by the electron and hole pairs will cause the breakdown of the B-C junction barrier region, and the B-C junction reverse bias voltage is the collector-base reverse Breakdown Voltage (BVCBO).

In the case of an open base, a voltage is applied across the C-E junction, while the B-C junction is still under reverse bias, which requires additional consideration of the contribution of electron current injected by a portion of the emitter into the base region, then transited by the base region to the collector barrier region, and finally reaching the collector region. Since the base is in an open state and the B-E junction voltage has not yet opened, the current developed by this injection is relatively weak, typically in the microampere (uA) order. In the case of an open emitter, there is no additional electron current injected by the emitter into the base region. In the case of an open base, this partial electron current accelerates the formation of a large current in the collector barrier region, accelerating the breakdown of the B-C junction barrier region, where the voltage across the C-E junction is the collector-emitter reverse Breakdown Voltage (BVCEO). Clearly, BVCEO is smaller than BVCBO.

However, the rf power amplifier is not in both cases described above, but in the forward active region during normal operation. The normal operating current of the power core unit is now on the order of milliamperes (mA) to amperes (a), which is much greater than microamperes (uA), see fig. 3. Fig. 3 is a schematic diagram showing the difference between the open emitter, open base, and normal operation conditions of the semiconductor device of the radio frequency power amplifier.

From the structural analysis of the forward active region energy band, when the radio frequency power amplifier is in high-power emission, the B-E junction is additionally forward biased, so that the energy of the ECN of the conduction band bottom of the emitter is close to that of the ECP of the conduction band bottom of the base, and a large amount of electrons are injected into the base region by the emitter under the condition that the B-E junction is forward conducted. The large electron current formed by the injection of the emitter is not present in the open emitter or open base condition, and it is apparent that the breakdown of the B-C junction barrier is further accelerated after the electrons are swept by the accelerating electric field in the base region to the collector barrier. Obviously, when the emitter Injection Current (ICE) is large, for example, in the order of a, the corresponding Breakdown Voltage (BVCE) is reduced; when the Injection Current (ICE) becomes smaller, for example, in the mA order, the corresponding Breakdown Voltage (BVCE) increases; when the Injection Current (ICE) is small, e.g., of the uA order, its corresponding Breakdown Voltage (BVCE) will be close to BVCEO, similar to the base open circuit case. This creates a correspondence between ICE and BVCE, referred to herein as a Secure Operating Area (SOA). Within this range the rf power amplifier is safe, whereas exceeding this range results in degraded rf power amplifier performance or pn junction breakdown burnout. Referring to fig. 3, since the effect of the forward active region on the device operation is not considered in the conventional analysis, BV using the transistor _CEO Or BV _CBO It is not objective to evaluate whether the rf power amplifier can safely operate. The utility model provides a technical scheme for judging whether a semiconductor device meets the operation index of a 5G radio frequency power amplifier or not based on a Safe Operating Area (SOA) of a transistor, and provides the semiconductor device meeting the operation index of the 5G radio frequency power amplifier. Referring to fig. 3, when the rf power amplifier is operating normally, the rf power amplifier is formed ofThe semiconductor device therein is in the forward active region, and thus the normal operating current of the semiconductor device (transistor) of the frequency power amplifier is in the order of milliamp (mA) to ampere (a), which is much larger than microampere (uA) level. Thus, the safe operating region of a semiconductor device is defined as an operating region whose operating current is on the order of milliamperes (mA) to amperes (a), the breakdown voltage of which is different from the conventional collector-base reverse Breakdown Voltage (BVCBO) or collector-emitter reverse Breakdown Voltage (BVCEO), and in general, the breakdown voltage of which is smaller than the collector-base reverse Breakdown Voltage (BVCBO) and close to the collector-emitter reverse Breakdown Voltage (BVCEO).

When the 4G radio frequency power amplifier works in the forward active region, the maximum current and the voltage swing at the collector of the semiconductor device in the 4G radio frequency power amplifier are both within GaAs SHBT SOA under the power supply voltage of 3.4V, for example, so that the 4G radio frequency power amplifier can still be ensured to work normally when a transistor with a GaAs SHBT structure is adopted. When the 5G rf power amplifier is operating in the forward active region, the maximum current and voltage swing at the collector of the semiconductor device in the 5G rf power amplifier has exceeded the SOA range provided by the GaAs SHBT process at a supply voltage of, for example, 5V. Since the B-C junctions of GaAs SHBT process are homojunction, they are GaAs structures, and the forbidden band width Eg (GaAs) =1.43 eV for GaAs. Compared with the heterojunction of InGaP/GaAs, the narrow bandgap collector region of the structure is easier to collide and ionize, so that the SOA range is reduced, and the semiconductor device based on the GaAs SHBT process cannot meet the working requirement of the 5G radio frequency power amplifier. If the B-C junction is changed into a heterojunction and the collector region is changed into a wide bandgap structure, the occurrence of collision ionization of carriers in the collector region can be reduced, and the scope of the SOA is correspondingly increased.

Aiming at the situation that the linearity or the output power of the 5G radio frequency power amplifier prepared by adopting the GaAs SHBT technology is insufficient in the 5G application scene, and the problem that the radio frequency power amplifier burns out can also occur under the 5V working voltage. The utility model provides a 5G radio frequency power amplifier prepared by a gallium arsenide double heterojunction (GaAsDHBT) epitaxial process, wherein a collector region of the GaAs DHBT process adopts a wide bandgap structure, so that a power core unit SOA of the GaAs DHBT process has a larger range. The current and voltage swing range of the 5G radio frequency power amplifier prepared by the process under the 5V working voltage is still within the SOA so as to avoid or reduce the occurrence of the problems.

Fig. 4 is a cross-sectional view of a semiconductor device for a radio frequency power amplifier according to an embodiment of the present utility model.

Referring to fig. 4, the semiconductor devices respectively include from top to bottom: an emitter metal layer 401, an emitter cap layer 402, an emitter n-type doped layer (InGaP) 403, a base metal layer 404, a base p+ type doped layer (GaAs) 405, a collector metal layer 406, a collector n-type doped layer (In) _x Ga _(1-x) P) 407, a sub-collector n+ -doped layer (GaAs) 408, and a substrate layer (GaAs) 409. Wherein the emitter metal layer 401, the base metal layer 404, and the collector metal layer 406 are used to form a connection with an external circuit wire; the emitter cap layer 402 allows the emitter metal layer 401 to obtain good ohmic contact with the epitaxial layer and relieves stress between the emitter metal layer 401 and the emitter layer 403 due to lattice mismatch. Compared with a collector GaAs structure, the collector of the semiconductor device according to the embodiment of the utility model adopts In with a wide band gap _x Ga _(1-x) P structure, wherein x and 1-x represent the proportion of In and Ga elements.

Compared with a GaAs SHBT process, the GaAs DHBT according to the embodiment of the utility model changes the epitaxial layer of the collector from an n-doped GaAs structure to n-doped In _x Ga _(1-x) P, at the same time, compared with the n-type doped GaAs structure, the In is properly reduced _x Ga _(1-x) The P-epi layer thickness is, for example, such that the collector layer is configured such that its thickness is less than a first threshold, preferably 50nm. The base epitaxial layer is still a GaAs structure, so that the B-C junction becomes a heterojunction, forming a GaAs DHBT structure.

Fig. 5 is a schematic diagram showing an energy band structure of a semiconductor device for a radio frequency power amplifier according to an embodiment of the present utility model. Referring to FIG. 5, due to In _x Ga _(1-x) The forbidden band width of P is larger than that of GaAsThe width, therefore, forms a barrier spike in the collector barrier layer that blocks electrons from entering the collector from the base. By appropriate control of In _x Ga _(1-x) The concentration and thickness of In and Ga elements In the n-type doped layer In P can effectively reduce the energy band peak of the barrier region, so that the electrical performance of the GaAs DHBT structure is not affected.

When the power stage transistor works in the forward active region, electrons of the collector barrier layer are accelerated to move under a large electric field formed by the applied voltage of the collector. With the increase of the applied voltage (3.4V to 5V, until the voltage reaches a larger transient swing), electrons with larger energy can cross the forbidden band to the valence band, and the valence electrons in the full band are excited to form more carriers generated by collision ionization. And after the collector region adopts a wide-bandgap InGaP structure, electrons need more energy to cross the forbidden band. By adopting the GaAs DHBT technology to form the semiconductor device, when the external voltage of the 5G radio frequency power amplifier is increased to 5V, the maximum current and the voltage swing at the collector are both within the GaAs DHBT SOA, so that the 5G radio frequency power amplifier can be ensured to normally operate.

When the gallium arsenide GaAs heterojunction epitaxial layer is prepared by growth, the lattice matching degree among different materials needs to be considered, so that the stress problem and the crystal structure defect problem between the upper layer and the lower layer of the heterojunction are reduced. If there are ternary or quaternary combinations, the percentages of the elements of groups III and V need to be considered to increase the degree of lattice matching between the upper and lower layers of the heterojunction.

The lattice constant of GaAs was 0.560 nm, that of GaP was 0.545nm, and that of InP was 0.586nm.

According to an embodiment of the present utility model, a GaAs structure is used In the base and In is used In the collector _(x) Ga _(1-x) P structure, x=0.49 In terms of lattice matching degree, i.e., 49% In composition and 51% Ga composition. It is claimed here that the value of x is not a fixed value and that the x component may fluctuate between 0.45 and 0.55 when actually prepared. Wherein, when the In composition is 49% and the Ga composition is 51%, in _(0.49) Ga _(0.51) The bandgap of P is Eg_InGaP (1.91 eV), and the bandgap of GaAs isEg_GaAs(1.43eV)。

When the transistor is operated in the forward active region, the base and collector are in a reverse bias state. With the increase of the reverse bias voltage, electrons in the conduction band of the depletion region and holes in the valence band have larger energy due to the action of an electric field, and when the energy reaches a certain value, electrons in the valence band can be separated from the original electron hole pairs and reach the conduction band across the forbidden band; meanwhile, holes in the conduction band can be separated from the original electron hole pairs, and reach the valence band across the forbidden band. Thus, under the action of a strong electric field, the number of carriers increases sharply. Hereinafter, the differential case of single heterojunction and double heterojunction will be discussed in the case where the same energy and the crossover distance (Xn) are the same, i.e., the same mechanism.

In the case of single heterojunction, both the base and the collector are GaAs structures, and when the reverse bias voltage increases to a first breakdown value (vbc_s), electrons in the valence band reach the conduction band across the forbidden band; at the same time, holes in the conduction band cross the forbidden band to reach the valence band, and the crossing distance is Xn.

In the case of double heterojunction, the base is a GaAs structure and the collector is In _(x) Ga _(1-x) P structure, electrons in the valence band reach the conduction band across the forbidden band when the reverse bias voltage increases to the second breakdown value (vbc_d); at the same time, holes in the conduction band cross the forbidden band to reach the valence band, and the crossing distance is still Xn. Due to In _(x) Ga _(1-x) The forbidden bandwidth of P is larger than GaAs, and to keep the electron or hole at the same crossing distance, the reverse bias value (vbc_d) needs to be increased to narrow the forbidden bandwidth in the depletion region, so that the width distance is equal to Xn. Fig. 6A shows a breakdown schematic of an energy band structure of a semiconductor device for a single heterojunction of a radio frequency power amplifier, and fig. 6B shows a breakdown schematic of an energy band structure of a semiconductor device for a double heterojunction of a radio frequency power amplifier. Referring to fig. 6A and 6B, when the collector in the semiconductor device is changed to a wide bandgap structure, the reverse bias voltage value vbc_d is greater than vbc_s.

The relation between Xn and the reverse bias voltage Vbc and the forbidden bandwidth Eg is shown in the following formula 1.

Wherein Xn is the crossing distance, eg is the forbidden bandwidth, N is the doping concentration, q is the charge size, vbc is the reverse bias voltage, and ε _r ε ₀ Is a dielectric constant. To maintain the same Xn, the Eg value and Vbc value need to be increased or decreased simultaneously according to equation 1.

When the 5G radio frequency power amplifier works normally, the B-C PN junction of the semiconductor device works under reverse bias voltage, and the reverse bias voltage can be gradually increased or reduced along with the change of output impedance. When the reverse bias voltage is gradually increased to vbc_s or vbc_d, a PN junction large current is formed, and at this time, PN junction breakdown is highly likely to occur. At this time, since the transistor is normally operated in the forward active region, the B-C PN junction is under critical reverse bias, and the output current and voltage range at this time are called the safe operating region (SOA) of the transistor, in which the safe operating region of the semiconductor device is defined as an operating region whose operating current is in the order of milliamp (mA) to ampere (a), and the breakdown voltage of the safe operating region is different from that of the conventional collector-base reverse breakdown voltage (BV _CBO ) Or collector-emitter reverse Breakdown Voltage (BV) _CEO ) In general, the breakdown voltage of the safe operating region is smaller than the collector-base reverse breakdown voltage (BV _CBO ) And a reverse Breakdown Voltage (BV) with the collector-emitter _CEO ) Proximity.

Fig. 7 is a comparative diagram showing the safe operating area of a semiconductor device of a radio frequency power amplifier fabricated using a GaAs SHBT process and a GaAs DHBT process.

Referring to fig. 7, a 5G radio frequency power amplifier is designed and fabricated when a GaAs Double Heterojunction (DHBT) process is used; meanwhile, under the condition that a Single Heterojunction (SHBT) process with the same GaAs flow is used for designing and preparing the 5G radio frequency power amplifier, the 5G radio frequency power amplifier with the same design parameters and the same layout structure and different processes are prepared based on the same design and layout structure. The Safe Operating Area (SOA) ranges of the power stage transistors therein were tested separately, and the actual ranges of GaAs SHBTSOA and GaAs DHBT SOA are shown in fig. 7.

Referring to FIG. 7, since the collector region In GaAs DHBT employs a wide bandgap In _x Ga _(1-x) The value of Vbc_d of the P structure is larger than the value of Vbc_s of the GaAs SHBT, namely, the range of the GaAs DHBT SOA is larger than that of the GaAs SHBT SOA, so that the 5G radio-frequency power amplifier can safely work under the voltages of 3.4V and 5V.

In addition, compared with a semiconductor transistor formed by a GaAs SHBT process, the semiconductor transistor formed by the GaAs DHBT introduces thinner InGaP and InGaAsP layers with wide forbidden band structures at a collector, so that the transition time of carriers in a collector region can be further effectively reduced, and the characteristic frequency of a device is remarkably improved, therefore, the semiconductor transistor has better high-frequency characteristics, and the semiconductor transistor formed by the GaAs DHBT can be better suitable for the application of a 5G radio-frequency power amplifier in high frequency (N41, N77 and N79 frequency bands).

Since the semiconductor transistor formed by GaAs DHBT employs InGaP and InGaAsP layers of a wide bandgap structure, it expands the safe operating area range of the radio frequency power amplifier. Compared with a semiconductor transistor formed by GaAs SHBT, the GaAs layer of the transistor with the DHBT structure in the sub-collector can properly reduce the doping concentration of the transistor, the mobility of carriers in the sub-collector can be greatly improved, and the transit time of the carriers in the sub-collector is reduced. According to one embodiment of the utility model, the doping concentration of the GaAs layer of the sub-collector may be configured to be below a first doping concentration threshold, e.g., below 1x10e16/cm ³ . According to another embodiment of the present utility model, the sub-collector layer may be further configured to include: a lower sub-collector layer having a concentration of 5x10e18/cm ³ The lower sub-collector layer is configured as a highly doped region to reduce contact resistance of the collector region and the GaAs substrate, and is configured to have a thickness of 500 nm; a middle sub-collector layer having a concentration of 1x10e16/cm ³ And which is configured to have a thickness of 500nm, by configuring the middle sub-collector layer to be doped at a low concentration to reduce impact ionization of the collector region, to improve mobility of the collector region, and thereby to improve high frequency performance of the semiconductor device; an upper sub-collector layer having a concentration of 5x10e18/cm ³ And is configured to have a thickness of 5 nm. It will be appreciated by those skilled in the art that the GaAs concentration of the other layers may also be suitably reduced so that the minimum doping concentration layer is less than 1x10e16/cm in order to increase the electron mobility of the collector region ³ . According to the configuration, the characteristic frequency of the semiconductor device is improved, so that the GaAs DHBT formed semiconductor transistor can be better suitable for the application of the 5G radio frequency power amplifier in high frequency (N41, N77 and N79 frequency bands).

Fig. 8 is a cross-sectional view illustrating a semiconductor device for a radio frequency power amplifier according to another embodiment of the present utility model.

Referring to fig. 8, the semiconductor devices respectively include from top to bottom: an emitter metal layer 701, an emitter cap layer 702, an emitter n-type doped layer (InGaP) 703, a base metal layer 704, a base p+ -type doped layer (GaAs) 705, a collector metal layer 706, a collector n-type doped layer (In) _x Ga _(1-x) As _y P _(1-y) ) 707, a sub-collector n+ -doped layer (GaAs) 708, and a substrate layer (GaAs) 709. In this embodiment, the collector uses In with a wide band gap _x Ga _(1-x) As _y P _(1-y) Structure is as follows. By adding As element In the collector epitaxial layer, an n-type doped InGaAsP group IV element combination is formed, wherein x and 1-x represent the proportion of In and Ga elements, and y and 1-y represent the proportion of As and P elements.

Fig. 9 is a schematic diagram showing an energy band structure of a semiconductor device for a radio frequency power amplifier according to another embodiment of the present utility model. Referring to FIG. 9, due to In _x Ga _(1-x) As _y P _(1-y) The forbidden band width of (2) is larger than that of GaAs, and a barrier peak is formed on the barrier layer of the collector region. By appropriate control of In _x Ga _(1-x) As _y P _(1-y) The concentration and thickness of the n-type doping, for example, the collector layer, is configured such that its thickness is less than the second threshold, while controlling the ratio of the elements therein, the energy band spike of the barrier region can be effectively reduced, so as not to affect the electrical performance of the GaAs DHBT structure. As the As element is introduced, the barrier region can form two energy band peaks, but the GaAs DHBT electrical property is not affected. According to an embodiment of the utility model, in _x Ga _(1-x) As _y P _(1-y) The epitaxial layer thickness (collector layer thickness) may be configured to be less than a second threshold value, preferably 50nm.

Since the collector region In GaAs DHBT uses wide bandgap In _x Ga _(1-x) As _y P _(1-y) The GaAs DHBT SOA has a larger range than the GaAs SHBT SOA, so that the structure can also meet the requirement that the 5G radio frequency power amplifier safely works under the voltages of 3.4V and 5V.

Due to the limitation of the lattice constant. When the base electrode uses GaAs structure and the collector electrode uses In _(x) Ga _(1-x) In the P structure, x needs to be within the first scale to maintain a good lattice match. The adjustment of the forbidden bandwidth of the collector is limited due to the limitation of the x component ratio, for example, in the range of 0.45 to 0.55. If the forbidden bandwidth of the collector structure needs to be further increased or reduced, an As element can be introduced, and In can be formed by adjusting the proportion of the elements _x Ga _(1-x) As _y P _(1-y) A collector of four elements. Wherein the y component ratio may be set, for example, in the range of 0.45 to 0.55. Thus, the GaAs structure of the base electrode has a matched lattice constant by adjusting the x and y values, and the required forbidden bandwidth value of the collector electrode structure is also adjusted.

Fig. 10 is a circuit frame diagram showing a radio frequency power amplifier including a semiconductor device according to an embodiment of the present utility model.

Referring to fig. 10, a two-stage rf power amplifier will be described. In fig. 9, the 5G rf power amplifier is mainly composed of a driver stage amplifying unit, a power stage amplifying unit, a driver stage bias circuit and a power stage bias circuit, and input, inter-stage and output matching networks. The amplifier circuit can be formed by using a semiconductor device manufactured by GaAs DHBT process in a power stage amplifying unit (output stage amplifying unit) in the 5G radio frequency power amplifier, and can ensure safe operation of the 5G radio frequency power amplifier at 5V voltage because the SOA range of the semiconductor device formed by GaAs DHBT process is larger. It will be appreciated by those skilled in the art that although an example of applying the GaAs DHBT process formed semiconductor device to a power stage amplifying unit in a 5G radio frequency power amplifier is shown in the present utility model, it may be applied to other parts of the circuit, for example, to a driver stage amplifying unit in a 5G radio frequency power amplifier, without departing from the scope of the present utility model.

Although in the above examples a power cell of a radio frequency amplifier is shown employing a cascode configuration, it will be appreciated by a person skilled in the art that the inventive concept may also be applied to power cells of amplifiers of other configuration types, e.g. the power cell of a radio frequency amplifier of the utility model may also be a common-set amplifier.

According to the embodiment of the utility model, the semiconductor device is designed according to the Safe Operating Area (SOA) of the transistor so as to meet the operating requirement of the 5G radio frequency power amplifier, thereby avoiding breakdown of the transistor of the radio frequency power amplifier in an operating state.

Meanwhile, in order to meet the requirement that the 5G radio frequency power amplifier works at 5V voltage, the SOA range of the GaAs HBT process needs to be enlarged, so that the SOA completely covers all working areas of voltage and current swing of the 5G radio frequency power amplifier, and the chip is ensured to safely work and not burn. According to the embodiment of the utility model, the wide-gap collector region structure is adopted, so that the forbidden bandwidth of the collector barrier region is increased, and the collision ionization of carriers in the collector barrier region is restrained. In addition, by properly reducing the thickness of the collector barrier region, the acceleration distance of carriers in the electric field is reduced, so that the total kinetic energy generated by the carriers is reduced, and the semiconductor device is in the SOA range.

Although the present disclosure has been described with exemplary embodiments, various changes and modifications may be suggested to one skilled in the art. The disclosure is intended to embrace such alterations and modifications that fall within the scope of the appended claims.

Any description of the present utility model should not be construed as implying that any particular element, step, or function is a necessary element to be included in the scope of the claims. The scope of patented subject matter is defined only by the claims.

Claims (10)

1. A semiconductor device for a radio frequency power amplifier, comprising:

a substrate layer configured to be formed by GaAs;

a sub-collector layer configured on the substrate layer and configured to be formed by n+ -doped GaAs;

a collector layer configured on the sub-collector layer and configured to be formed by an n-doped wide bandgap material having a bandgap width greater than a bandgap width of GaAs;

a collector metal layer disposed on the collector layer;

a base layer configured on the collector layer and configured to be formed by p+ -doped GaAs;

a base metal layer disposed on the base layer;

an emitter layer configured on the base layer and configured to be formed by n-type doped InGaP;

an emitter cap layer configured on the base layer; and

an emitter metal layer disposed on the emitter cap layer,

the collector layer is configured such that a maximum current and voltage swing at the collector layer is within a safe operating region of the semiconductor device, the safe operating region of the semiconductor device being defined as a safe operating region when the semiconductor device is in a forward active region.

2. The semiconductor device for a radio frequency power amplifier of claim 1, wherein the safe operating area of the semiconductor device is defined as an operating area having an operating current on the order of milliamperes (mA) to amperes (a).

3. The method for radio frequency power amplification as set forth in claim 1A semiconductor device of a amplifier is characterized In that the wide bandgap material comprises In _x Ga _(1-x) P and In _x Ga _(1-x) As _y P _(1-y) ，

x and 1-x represent the ratio of In and Ga elements, and y and 1-y represent the ratio of As and P elements.

4. A semiconductor device for a radio frequency power amplifier according to claim 3, wherein when the wide bandgap material is In _x Ga _(1-x) At P, the thickness of the collector layer is configured to be less than a first threshold of 50nm.

5. A semiconductor device for a radio frequency power amplifier according to claim 3, wherein when the wide bandgap material is In _x Ga _(1-x) As _y P _(1-y) When the thickness of the collector layer is configured to be less than a second threshold of 50nm.

6. A semiconductor device for a radio frequency power amplifier according to claim 3, wherein x is in a first ratio range of 0.45 to 0.55.

7. A semiconductor device for a radio frequency power amplifier according to claim 3, wherein y is in a second ratio range of 0.45 to 0.55.

8. The semiconductor device for a radio frequency power amplifier of claim 1, wherein the semiconductor device is configured for a 5G radio frequency power amplifier, the 5G radio frequency power amplifier having an operating voltage with a maximum power of 4.4V-5.5V.

9. The semiconductor device for a radio frequency power amplifier according to claim 8, wherein the semiconductor device is applied to a power stage amplifying unit of the 5G radio frequency power amplifier.

10. The semiconductor device for a radio frequency power amplifier of claim 1, wherein the sub-collector layer is configured to have a doping concentration less than the first doping concentration.

Images