Low-noise nonlinear calibration process design and mirror image twin integrated capacitor manufacturing method
By using low-noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method in RF and mixed signal integrated circuits, the problems of nonlinear error and poor matching of polysilicon capacitors are solved, and high precision linearity and stability are achieved. High-precision A/D and D/A converters.
Patent Information
- Application Number
- CN202480001303.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-12
- Publication Date
- 2024-12-06
AI Technical Summary
The electrical stress, thermal stress and hysteresis effects of polysilicon capacitors in existing RF and mixed signal integrated circuits lead to nonlinear errors and mismatch errors, affecting signal processing accuracy, and the errors in metal sputtering and polysiliconization processes lead to poor precision matching of capacitors. It is applicable It has low performance and stability, making it difficult to meet the requirements of high-precision linear design.
The low-noise nonlinear calibration process is used to design and mirror twin integrated capacitor manufacturing method. By forming groove isolation and trap areas on the substrate, doping elements are injected into the doping concentration and impurity distribution of the polycrystalline film layer to form twin capacitor cells. Combined with the thermal oxygen method to grow the capacitor dielectric layer and metal interconnection, a coupling bridge structure is realized, and the voltage coefficient and temperature coefficient of the capacitor are improved.
The integrated capacitor is realized with high dielectric strength, low leakage and low power consumption, and the precise linearity of the voltage coefficient reaches <2ppm/V, which improves the uniformity and stability on-chip and batches, and the matching degree is increased from ±0.1%. To ±0.0001%, suitable for high-precision A/D and D/A converters, reducing manufacturing costs.
Smart Images

Figure CN119096725A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of semiconductor integrated circuits, and in particular to a low-noise nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method. Background Art
[0002] The continuous development of "6G" communication technology has put forward higher requirements on the low noise, precision matching, nonlinear calibration, and no dielectric hysteresis of RF and mixed-mode (such as AD / DA data converters) integrated circuits. At the same time, the accuracy of passive devices integrated in process technology is becoming more and more important, and has become a decisive factor affecting the final accuracy of circuits and systems.
[0003] With the in-depth study of the electrical properties of polysilicon materials and the continuous optimization of the carrier transport model in polysilicon, integrated polycrystalline capacitors are widely used in the design of high-performance integrated circuits such as low offset voltage op amps, voltage-controlled oscillators, voltage-controlled frequency modulators, AD / DA data converters, etc., for their good matching characteristics. However, the electrical stress, thermal stress and hysteresis effects of polycrystalline capacitors aggravate the nonlinear errors and mismatch errors in integrated circuits, seriously affecting the signal processing accuracy in integrated circuits.
[0004] On the other hand, in RF and analog mixed signal integrated circuit process technology, structural capacitors such as Metal-Insulator-Metal and PolySicide-Insulator-Metal have been widely used. Although these integrated capacitors can improve the nonlinearity of capacitors by eliminating the polysilicon depletion effect, they are limited by the process thermal budget and have many defects in the capacitor dielectric material, resulting in low device dielectric strength, large leakage, and high power consumption. They are especially unsuitable for special environments such as deep space, deep sea, and polar regions. On the other hand, the second-order voltage coefficient QVCC of these integrated capacitors is single. Due to the influence of process errors such as metal sputtering and polysiliconization, the capacitor voltage coefficient varies greatly within and between batches, which will lead to poor precision matching of on-chip integrated capacitors, low uniformity and stability between chips and batches, and ultimately seriously affect the precision matching performance of these integrated circuits.
[0005] Calibration technology that is simply optimized through circuit architecture is already difficult to meet the linear design requirements of analog and digital signals in precision mixed-signal circuits. Therefore, the use of effective integrated polysilicon capacitor nonlinear calibration process design to fundamentally improve the key performance of capacitors such as voltage coefficient and temperature coefficient has become a continuous research hotspot in industry and academia, and is also the focus and difficulty of RF, precision mixed-signal integrated circuit process innovation and device new structure development. Summary of the invention
[0006] The object of the present invention is to provide a low-noise nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method, comprising the following steps:
[0007] 1) Making trench isolation on the substrate, and then forming N-type well region and P-type well region;
[0008] 2) Field oxygen-cutoff implantation isolation or trench isolation is used between the N-type well region and the P-type well region;
[0009] 3) depositing a P0 angstrom capacitor lower polycrystalline film layer on a uniform flat area of a n kilo-angstrom thick field oxide layer;
[0010] 4) The doping concentration and impurity distribution of the polycrystalline film layer of the lower electrode of the capacitor are adjusted by injecting N-type doping elements; according to the circuit design requirements, the doping concentration and impurity distribution of the polycrystalline film layer in the twin capacitor cells C1 and C2 regions are adjusted by injecting N-type doping elements in the required regions of the polycrystalline film layer of the lower electrode of the capacitor;
[0011] A polycrystalline structure of the bottom electrode of the twin capacitor cells C1 and C2 is formed;
[0012] 5) etching the polycrystalline film layer into a centrally symmetrical cell, growing a capacitor dielectric layer of a required thickness according to process requirements, and filling the gap between the twin capacitor cells formed in step 4) to form a capacitor lower plate protection structure;
[0013] 6) depositing a P1 angstrom polycrystalline film layer on the capacitor dielectric layer and completing N-type element implantation and doping;
[0014] 7) using a photolithography process to etch the polycrystalline film layer into a centrally symmetrical cell corresponding to the capacitor lower electrode formed in step 4), and the polycrystalline film layer serves as the upper electrode plate of the polycrystalline capacitor;
[0015] 8) growing an n angstrom thermal oxide layer by a thermal oxidation method; depositing a d1 angstrom silicon nitride dielectric layer to form a capacitor upper plate protection structure;
[0016] 9) completing the base and emitter region implantation of bipolar transistors and / or the source and drain implantation of field effect transistors, and annealing and activation;
[0017] 10) Metal silicide exposure and etching, sputtering Ti metal film, capacitor cell C1 and cell C2 polysilicon, other exposed silicon, polysilicon area metal silicide and then stripping off the metal Ti that has not undergone metal silicide reaction;
[0018] 11) Photolithography dielectric layer deposition, reflow filling and planarization, contact hole etching, tungsten plug filling and planarization;
[0019] 12) The upper electrode of the capacitor cell C1 is metal interconnected with the upper electrode of the capacitor cell C2; at the same time, the lower electrode of the capacitor cell C1 is metal interconnected with the lower electrode of the capacitor cell C2, and finally a Twin-Combined-Bridge architecture unit is formed.
[0020] Further, when the base and emitter region injection of the bipolar transistor or / and the source and drain injection of the field effect transistor are completed, no additional injection doping process adjustment is performed on the gate of the MOS transistor or the emitter polycrystalline film layer of the bipolar transistor, and the conditions are kept the same as the main process;
[0021] The implantation doping process is used to match the impurity concentration and interface distribution of the polycrystalline film layer dedicated to passive devices or the base polycrystalline film layer in the double polycrystalline Bipolar process;
[0022] The implantation dose range of polycrystalline film layer of capacitor cell C1 is 1e 15 cm -2 -1e 16 cm -2 , maintain the subdegenerate state of the polycrystalline film interface;
[0023] The implantation dose range of the polycrystalline film layer of the lower electrode of the capacitor cell C2 is > 1e 16 cm -2 , maintaining the degenerate state of the polycrystalline film interface and forming a shared carrier migration mode.
[0024] Furthermore, according to the process flow design, the polycrystalline film of the integrated polycrystalline capacitor electrode is the gate polycrystalline necessary for the MOS device, or the polycrystalline layer deposited by a special passive device process.
[0025] Furthermore, the process of manufacturing the trench isolation includes etching and backfilling processes.
[0026] Furthermore, the capacitor lower polycrystalline film layer of step 3) is deposited by low pressure chemical vapor deposition.
[0027] The polycrystalline film layer in step 7) is deposited by low pressure chemical vapor deposition.
[0028] Furthermore, the method of implanting the doped N-type element includes photolithography implantation.
[0029] Furthermore, the process of etching the polycrystalline film layer into a centrally symmetrical cell to form a capacitor lower plate protection structure includes a photolithography process.
[0030] Furthermore, when growing the n angstrom thermal oxide layer, the main process temperature is ≤850° C. and the main process time is ≤30 min.
[0031] Furthermore, the process of completing the base and emitter region implantation of the bipolar transistor and / or the source and drain implantation of the field effect transistor includes Bipolar, CMOS or BiCMOS.
[0032] The integrated capacitor based on the low-noise nonlinear calibration process design and the mirror twin integrated capacitor manufacturing method is characterized by: mainly comprising a substrate, an isolation groove, an N-type buried layer, a P-type buried layer, an N-type epitaxial layer, an N-type well, a P-type well, a field oxide layer, a sacrificial oxide layer, a gate oxide layer, a polycrystalline thin film layer, a silicon dioxide dielectric layer, a silicon oxynitride dielectric layer, a metal silicide film layer, a metal interconnect film layer, and a passivation layer;
[0033] The substrate is located at the bottom; the buried layer is located at the bottom of the epitaxial layer; and the well region is located on the surface of the epitaxial layer;
[0034] The N-type well and the P-type well are isolated by an isolation trench dielectric;
[0035] The field oxide layer, the sacrificial oxide layer, and the gate oxide layer respectively cover different areas of the well;
[0036] The double polycrystalline capacitor polycrystalline film layer is located above the field oxide layer above the N-well;
[0037] The top surface and sidewall of the polycrystalline film layer are covered with different silicon dioxide and silicon nitride composite dielectric layers and a high dielectric constant capacitor dielectric layer.
[0038] The low dielectric constant filling film layer is filled in the region between silicon-metal 1, polysilicon-metal 1, field oxygen-metal 1 and the multi-layer metal.
[0039] The technical effect of the present invention is unquestionable. The present invention provides a low-noise, low-mismatch nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method, and adopts a coupled bridge nonlinear calibration process design to realize the connection of integrated polycrystalline capacitors. Within the RF and mixed signal operating voltage range, the voltage coefficient of the integrated capacitor reaches a nonlinear accuracy of <2ppm / V.
[0040] The integrated capacitor nonlinear calibration process design and integration method proposed in the present invention realizes high capacitance density of the capacitor while the capacitance density of the capacitor cell C1 is equal to that of the capacitor cell C2, and realizes high dielectric strength, low leakage and low power consumption of the integrated capacitor.
[0041] The capacitor twin cell doping proposed in the present invention realizes the mirror characteristic of the second-order voltage coefficient of the capacitor cell, that is, the second-order voltage coefficient QVCC of the capacitor cell C1 is less than 0, and the second-order voltage coefficient QVCC of the capacitor cell C2 is greater than 0, thereby finally realizing the voltage coefficient of the integrated capacitor to achieve a precise linearity of less than 2ppm / V.
[0042] The coupling-bridge architecture of the twin capacitor cell proposed in the present invention not only improves the precision matching within the chip, but also improves the uniformity and stability of the integrated capacitors between chips and batches. At the same time, the coupling-bridge architecture proposed in the present invention is adopted, and the high potential is at the upper electrode of the twin polycrystalline capacitor cell, which reduces the influence of parasitic capacitance substrate noise and crosstalk noise, and realizes the low noise characteristics of the integrated capacitor.
[0043] Through the low-noise, low-mismatch, nonlinear calibration process design and mirror-image twin integrated capacitors of the present invention, the matching degree of the integrated dual polycrystalline capacitor can be improved from a mismatch of ±0.1% to an ultra-precise matching mismatch level of ±0.0001%, which can be applied to high-precision A / D and D / A converters with a resolution of more than 20 bits, thereby improving process compatibility and device integration density, reducing the cost of precision circuit manufacturing, and enhancing product market competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 It is a cross-sectional view of the growth of the shield oxide layer after completing the conventional analog BiCMOS integrated circuit processes such as buried layer implantation, epitaxial growth, well implantation, annealing, and field oxidation;
[0045] Figure 2 It is a cross-sectional view of the lower plate of the double polycrystalline capacitor after the polycrystalline film layer is deposited and doped;
[0046] Figure 3 It is a cross-sectional view of the edge protective film layer of the polycrystalline film layer of the lower plate of the polycrystalline capacitor;
[0047] Figure 4 It is a cross-sectional view of the polycrystalline capacitor capacitor dielectric layer deposition and exposure etching structure;
[0048] Figure 5 It is a partial enlarged cross-sectional view of the edge area of the dual polycrystalline capacitor structure after the subsequent wet etching and oxidation processes are completed using the edge optimization technology solution;
[0049] Figure 6 It is a schematic diagram of the 3D structure after the polycrystalline film layer of the upper plate of the polycrystalline capacitor is deposited and doped, and then exposed and etched;
[0050] Figure 7 This is a schematic diagram of the main device structure of the analog BiCMOS process integration before the metal interconnection process: including polycrystalline emitter vertical bipolar transistor VNPN, MOS transistor NMOS, precision polycrystalline capacitor PIP-Capacitor, low temperature coefficient polycrystalline resistor Poly-Resistor;
[0051] Figure 8After completing the two-layer metal interconnection process, the main device structure diagram of the simulated BiCMOS process integration is shown: including polycrystalline emitter vertical bipolar transistor VNPN, MOS transistor NMOS, precision polycrystalline capacitor PIP-Capacitor, and low temperature coefficient polycrystalline resistor Poly-Resistor;
[0052] Fig. 9 This is a schematic diagram of the structure of the analog BiCMOS integrated precision polycrystalline capacitor of Example 12;
[0053] In the figure, 101, isolation groove (optional); 102, N-type well region (preferred); 103, LOCOS field oxide layer; 104, lower polycrystalline film layer; 105, capacitor dielectric layer; 106, polycrystalline capacitor upper plate; 201, polysilicon / field oxygen-metal 1 interlayer ILD dielectric planarization layer; 202, polycrystalline capacitor plate-metal 1 interlayer contact hole tungsten plug structure; 203, metal layer 1 structure area; 204, multi-layer metal layer interlayer IMD dielectric planarization layer; 205, multi-layer metal layer interlayer through hole tungsten plug structure area 206, top metal layer area. DETAILED DESCRIPTION
[0054] The present invention is further described below in conjunction with the embodiments, but it should not be understood that the above subject matter of the present invention is limited to the following embodiments. Without departing from the above technical ideas of the present invention, various substitutions and changes are made according to the common technical knowledge and customary means in the art, which should all be included in the protection scope of the present invention.
[0055] Embodiment 1:
[0056] See also Figures 1 to 9 , a low noise nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method, comprising the following steps:
[0057] 1) Making trench isolation on the substrate, and then forming N-type well region and P-type well region;
[0058] 2) Field oxygen-cutoff implantation isolation or trench isolation is used between the N-type well region and the P-type well region;
[0059] 3) depositing a P0 angstrom capacitor lower polycrystalline film layer 104 on a uniform flat region of the n kiloangstrom thick field oxide layer;
[0060] 4) The doping concentration and impurity distribution of the polycrystalline film layer of the lower electrode of the capacitor are adjusted by injecting N-type doping elements; according to the circuit design requirements, the doping concentration and impurity distribution of the polycrystalline film layer in the twin capacitor cells C1 and C2 regions are adjusted by injecting N-type doping elements in the required regions of the polycrystalline film layer of the lower electrode of the capacitor;
[0061] A polycrystalline structure of the bottom electrode of the twin capacitor cells C1 and C2 is formed;
[0062] 5) etching the polycrystalline film layer into a centrally symmetrical cell, growing a capacitor dielectric layer 105 of a required thickness according to process requirements, and filling the gap between the twin capacitor cells formed in step 4) to form a protective structure for the capacitor lower plate 106;
[0063] 6) depositing a P1 angstrom polycrystalline film layer on the capacitor dielectric layer and completing N-type element implantation and doping;
[0064] 7) using a photolithography process to etch the polycrystalline film layer into a centrally symmetrical cell corresponding to the capacitor lower electrode formed in step 4), and the polycrystalline film layer serves as the polycrystalline capacitor upper plate 106;
[0065] 8) growing an n angstrom thermal oxide layer by a thermal oxidation method; depositing a d1 angstrom silicon nitride dielectric layer to form a capacitor upper plate protection structure;
[0066] 9) completing the base and emitter region implantation of bipolar transistors and / or the source and drain implantation of field effect transistors, and annealing and activation;
[0067] 10) Metal silicide exposure and etching, sputtering Ti metal film, capacitor cell C1 and cell C2 polysilicon, other exposed silicon, polysilicon area metal silicide and then stripping off the metal Ti that has not undergone metal silicide reaction;
[0068] 11) Photolithography dielectric layer deposition, reflow filling and planarization, contact hole etching, tungsten plug filling and planarization;
[0069] 12) The upper electrode of the capacitor cell C1 is metal interconnected with the upper electrode of the capacitor cell C2; at the same time, the lower electrode of the capacitor cell C1 is metal interconnected with the lower electrode of the capacitor cell C2, and finally a Twin-Combined-Bridge architecture unit is formed.
[0070] When the base and emitter region implantation of the bipolar transistor or / and the source and drain implantation of the field effect transistor are completed, no additional implantation and doping process adjustment is performed on the gate of the MOS transistor or the emitter polycrystalline film layer of the bipolar transistor, and the conditions are kept the same as the main process;
[0071] The implantation doping process is used to match the impurity concentration and interface distribution of the polycrystalline film layer dedicated to passive devices or the base polycrystalline film layer in the double polycrystalline Bipolar process;
[0072] The implantation dose range of polycrystalline film layer of capacitor cell C1 is 1e 15 cm -2 -1e 16 cm -2 , maintain the subdegenerate state of the polycrystalline film interface;
[0073] The implantation dose range of the polycrystalline film layer of the lower electrode of the capacitor cell C2 is > 1e 16 cm -2 , maintaining the degenerate state of the polycrystalline film interface and forming a shared carrier migration mode.
[0074] According to the process flow design, the polycrystalline film of the integrated polycrystalline capacitor electrode is the necessary gate polycrystalline of the MOS device, or the polycrystalline layer deposited by the special passive device process.
[0075] The process of making trench isolation includes etching and backfilling processes.
[0076] The capacitor lower polycrystalline film layer of step 3) is deposited by low pressure chemical vapor deposition.
[0077] The polycrystalline film layer in step 7) is deposited by low pressure chemical vapor deposition.
[0078] The method of implanting the doped N-type element includes photolithography implantation.
[0079] The process of etching the polycrystalline film layer into a centrally symmetrical cell to form a capacitor lower plate protection structure includes a photolithography process.
[0080] When growing an n angstrom thermal oxide layer, the main process temperature is ≤850°C and the main process time is ≤30 minutes.
[0081] The process of completing the base and emitter region implantation of bipolar transistors or / and the source and drain implantation of field effect transistors includes Bipolar, CMOS or BiCMOS.
[0082] Embodiment 2:
[0083] The low noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method include the following steps:
[0084] 1) Making trench isolation on the substrate, and then forming N-type well region and P-type well region;
[0085] 2) Field oxygen-cutoff implantation isolation or trench isolation is used between the N-type well region and the P-type well region;
[0086] 3) depositing a P0 angstrom capacitor lower polycrystalline film layer on a uniform flat area of a n kilo-angstrom thick field oxide layer;
[0087] 4) The doping concentration and impurity distribution of the polycrystalline film layer of the lower electrode of the capacitor are adjusted by injecting N-type doping elements; according to the circuit design requirements, the doping concentration and impurity distribution of the polycrystalline film layer in the twin capacitor cells C1 and C2 regions are adjusted by injecting N-type doping elements in the required regions of the polycrystalline film layer of the lower electrode of the capacitor;
[0088] A polycrystalline structure of the bottom electrode of the twin capacitor cells C1 and C2 is formed;
[0089] 5) etching the polycrystalline film layer into a centrally symmetrical cell, growing a capacitor dielectric layer of a required thickness according to process requirements, and filling the gap between the twin capacitor cells formed in step 4) to form a capacitor lower plate protection structure;
[0090] 6) depositing a P1 angstrom polycrystalline film layer on the capacitor dielectric layer and completing N-type element implantation and doping;
[0091] 7) using a photolithography process to etch the polycrystalline film layer into a centrally symmetrical cell corresponding to the capacitor lower electrode formed in step 4), and the polycrystalline film layer serves as the upper electrode plate of the polycrystalline capacitor;
[0092] 8) growing an n angstrom thermal oxide layer by a thermal oxidation method; depositing a d1 angstrom silicon nitride dielectric layer to form a capacitor upper plate protection structure;
[0093] 9) completing the base and emitter region implantation of bipolar transistors and / or the source and drain implantation of field effect transistors, and annealing and activation;
[0094] 10) Metal silicide exposure and etching, sputtering Ti metal film, capacitor cell C1 and cell C2 polysilicon, other exposed silicon, polysilicon area metal silicide and then stripping off the metal Ti that has not undergone metal silicide reaction;
[0095] 11) Photolithography dielectric layer deposition, reflow filling and planarization, contact hole etching, tungsten plug filling and planarization;
[0096] 12) The upper electrode of the capacitor cell C1 is metal interconnected with the upper electrode of the capacitor cell C2; at the same time, the lower electrode of the capacitor cell C1 is metal interconnected with the lower electrode of the capacitor cell C2, and finally a Twin-Combined-Bridge architecture unit is formed.
[0097] Embodiment 3:
[0098] A low-noise nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method, the technical content is the same as in Example 2, further, when completing the bipolar transistor base region and emitter region injection or / and the field effect transistor source and drain injection, the MOS transistor gate or the bipolar transistor emitter polycrystalline film layer does not make additional injection doping process adjustments, and maintains the same as the main process conditions;
[0099] The implantation doping process is used to match the impurity concentration and interface distribution of the polycrystalline film layer dedicated to passive devices or the base polycrystalline film layer in the double polycrystalline Bipolar process;
[0100] The implantation dose range of polycrystalline film layer of capacitor cell C1 is 1e 15 cm -2 -1e 16 cm -2 , maintain the subdegenerate state of the polycrystalline film interface;
[0101] The implantation dose range of the polycrystalline film layer of the lower electrode of the capacitor cell C2 is > 1e 16 cm -2 , maintaining the degenerate state of the polycrystalline film interface and forming a shared carrier migration mode.
[0102] Embodiment 4:
[0103] The low-noise nonlinear calibration process design and the mirror twin integrated capacitor manufacturing method have the same technical content as any one of Examples 2-3. Furthermore, according to the process flow design, the integrated polycrystalline capacitor electrode polycrystalline film is the necessary gate polycrystalline of the MOS device, or the polycrystalline layer deposited by a special passive device process.
[0104] Embodiment 5:
[0105] A low-noise nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method, the technical content is the same as any one of Examples 2-4, and further, the process for making trench isolation includes etching and backfilling processes.
[0106] Embodiment 6:
[0107] A low-noise nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method, the technical content is the same as any one of Examples 2-5, further, the capacitor lower pole polycrystalline film layer in step 3) is deposited by low-pressure chemical vapor deposition.
[0108] The polycrystalline film layer in step 7) is deposited by low pressure chemical vapor deposition.
[0109] Embodiment 7:
[0110] A low-noise nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method, the technical content of which is the same as any one of Examples 2-6, and further, the method of injecting doped N-type elements includes photolithography injection.
[0111] Embodiment 8:
[0112] The low-noise nonlinear calibration process design and the mirror twin integrated capacitor manufacturing method have the same technical content as any one of Examples 2-7. Furthermore, the process of etching the polycrystalline film layer into a centrally symmetrical cell to form a capacitor lower plate protection structure includes a photolithography process.
[0113] Embodiment 9:
[0114] A low-noise nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method, the technical content of which is the same as any one of Examples 2-8, further, when growing an n-angstrom thermal oxide layer, the main process temperature is ≤850°C, and the main process time is ≤30min.
[0115] Embodiment 10:
[0116] A low-noise nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method, the technical content is the same as any one of Examples 2-9, further, the process of completing the base and emitter injection of the bipolar transistor or / and the source and drain injection of the field effect transistor includes Bipolar, CMOS or BiCMOS.
[0117] Embodiment 11:
[0118] The integrated capacitor according to the low-noise nonlinear calibration process design and the mirror twin integrated capacitor manufacturing method described in any one of embodiments 1-10 is characterized by: mainly comprising a substrate, an isolation groove 101, an N-type buried layer, a P-type buried layer, an N-type epitaxial layer, an N-type well 102, a P-type well, a LOCOS field oxide layer 103, a sacrificial oxide layer, a gate oxide layer, a polycrystalline thin film layer, a silicon dioxide dielectric layer, a silicon oxynitride dielectric layer, a metal silicide film layer, a metal interconnect film layer, and a passivation layer;
[0119] The substrate is located at the bottom; the buried layer is located at the bottom of the epitaxial layer; and the well region is located on the surface of the epitaxial layer;
[0120] The N-type well and the P-type well are isolated by an isolation trench dielectric;
[0121] The field oxide layer, the sacrificial oxide layer, and the gate oxide layer respectively cover different areas of the well;
[0122] The double polycrystalline capacitor polycrystalline film layer is located above the field oxide layer above the N-well;
[0123] The top surface and sidewalls of the polycrystalline film layer are covered with different silicon dioxide and silicon nitride composite dielectric layers 106 and a high dielectric constant capacitor dielectric layer.
[0124] The low dielectric coefficient filling film layer is filled in the area between the silicon-metal layer 1, the polysilicon-metal layer 1, the field oxygen-metal layer 1 and the multi-layer metal, including the polysilicon / field oxygen-metal 1 interlayer ILD dielectric planarization layer 201 and the multi-layer metal interlayer IMD dielectric planarization layer 204. The metal layer 1 structural area is marked as 203.
[0125] A polycrystalline capacitor plate-metal 1 interlayer contact hole tungsten plug structure 202 is formed between the polycrystalline capacitor plate and the metal 1 layer, and a multi-layer metal interlayer through hole tungsten plug structure area 205 is formed between the multi-layer metals. The top metal layer area of the integrated capacitor is marked as 206.
[0126] Embodiment 12:
[0127] The invention relates to a method for implementing a low noise, low mismatch, nonlinear calibration process design and a mirror twin integrated capacitor manufacturing method in an analog BiCMOS integrated circuit process, which mainly includes the following steps:
[0128] 1) Forming N-type buried layer, P-type buried layer, and N-type epitaxial growth on the substrate
[0129] 2) Forming an N-type well, a P-type well and a vertical NPN bipolar transistor collector N+ implant on the epitaxial layer and completing the well push junction;
[0130] For the SOI process, after the epitaxial layer growth is completed, a trench isolation process is used to form a full dielectric isolation region.
[0131] 3) Complete P-type implantation of the base region of the vertical NPN Bipolar transistor in the N-type well;
[0132] 4) depositing a silicon nitride film, and exposing and etching away the silicon nitride film layer outside the active device area, and using a thermal oxidation process to grow a field oxide layer of m angstroms;
[0133] 5) Deposit a P1 angstrom polycrystalline film layer on the field oxide layer using low pressure chemical vapor deposition, and complete N-type photolithography implantation doping according to the performance requirements of the capacitor. The polycrystalline film layer can be used as the lower plate of the polycrystalline capacitor;
[0134] The polycrystalline film layer of the lower plate of the polycrystalline capacitor is deposited on the field oxide layer above the P-type well surrounded by the dielectric isolation groove, which can reduce the influence of parasitic capacitance and substrate noise and improve the linearity of the polycrystalline capacitor.
[0135] 6) growing an n angstrom thermal oxide layer using thermal oxidation process conditions;
[0136] The main process temperature of thermal oxidation is ≤850℃, and the main process time is ≤30min, which is conducive to process compatibility and reduces the impact on the main process devices of analog BiCMOS integrated circuits.
[0137] 7) According to the circuit design, the polycrystalline film layer of the lower electrode of the capacitor cell C1 is doped with N-type elements to adjust the doping concentration and impurity distribution of the polycrystalline film layer; According to the circuit design, the polycrystalline film layer of the lower electrode of the capacitor cell C2 is doped with N-type elements to adjust the doping concentration and impurity distribution of the polycrystalline film layer;
[0138] Firstly, N-type doping elements are simultaneously injected into the polycrystalline film layers of the capacitor cells C1 and C2 to adjust the doping concentration of the polycrystalline film layers, and then N-type doping elements are injected into the polycrystalline film layer of the lower electrode of the capacitor cell C2 again to adjust the doping concentration and impurity distribution of the polycrystalline film layer;
[0139] 8) etching the polycrystalline film layer into a centrally symmetrical cell using a photolithography process; growing a silicon dioxide-silicon nitride-silicon dioxide dielectric layer of a required thickness according to process requirements and filling the cell gap formed in 7); and forming a capacitor lower plate protection structure using a photolithography process.
[0140] The capacitor dielectric film layer can be a high dielectric constant film layer to increase the unit capacitance of the dual polycrystalline capacitor.
[0141] 9) Complete the normal process steps such as sacrificial oxidation, channel injection, exposure etching, gate oxide oxidation, etc. before the gate polycrystalline deposition of the analog BiCMOS integrated circuit;
[0142] 10) using low pressure chemical vapor deposition to deposit a P2 angstrom MOS transistor gate polycrystalline film layer, and completing MOS transistor gate polycrystalline doping according to MOS device performance requirements;
[0143] 11) Next, complete the normal analog BiCMOS integrated circuit MOS transistor source and drain injection and Bipolar transistor emitter injection, and anneal activation;
[0144] No additional implantation and doping process adjustment is made to the gate polycrystalline film layer of the MOS transistor or the emitter polycrystalline film layer of the Bipolar transistor, and the conditions are kept the same as the main process;
[0145] The implantation doping process is used to match the impurity concentration and interface distribution of the polycrystalline film layer of a specialized polycrystalline passive device or the base polycrystalline film layer in a dual polycrystalline bipolar process;
[0146] The implantation dose range of the polycrystalline film layer of the lower electrode of capacitor cell C1 is 3e 15 cm -2 -1e 16 cm -2 , maintain the subdegenerate state of the polycrystalline film interface;
[0147] The implantation dose range of the polycrystalline film layer of the lower electrode of the capacitor cell C2 is > 1e 16 cm -2 , maintain the degenerate state of the polycrystalline film interface and form a shared carrier migration mode;
[0148] 12) Metal silicide exposure and etching, sputtering Ti metal film. After metal silicide of capacitor cell C1 and cell C2 polysilicon and other exposed silicon and polysilicon areas, the metal Ti that has not undergone metal silicide reaction is stripped off.
[0149] 13) Normal processes such as photolithography dielectric layer deposition, reflow filling and planarization, contact hole etching, tungsten plug filling and planarization;
[0150] 14) The upper electrode of the capacitor cell C1 is metal interconnected with the upper electrode of the capacitor cell C2; at the same time, the lower electrode of the capacitor cell C1 is metal interconnected with the lower electrode of the capacitor cell C2, and finally a Twin-Combined-Bridge architecture unit is formed.
[0151] Keeping the capacitor cell Cn in the same shape and area can reduce the impact of process errors on the nonlinear mismatch of the capacitor.
[0152] For processes that are not suitable for metal silicide, the upper electrode of capacitor cell C11 and the lower electrode of capacitor cell C12 can be metal interconnected; at the same time, the lower electrode of capacitor cell C11 and the upper electrode of capacitor cell C12 can be metal interconnected, and finally a metal-silicon-free Combined-Bridge architecture unit is formed, which improves the intra-chip matching and inter-chip and batch-to-batch uniformity of the integrated capacitor of capacitor cell C1. Capacitor cell C2 can also adopt a similar coupling-bridge method.
[0153] Embodiment 13:
[0154] For the analog BiCMOS integrated circuit process, the integrated production of precision polycrystalline capacitors is carried out based on the completion of N-type buried layer, P-type buried layer, epitaxial growth, isolation trench process, N-well injection push-junction, P-well injection push-junction, field oxide layer growth and other processes.
[0155] 1) The gate polycrystalline film layer of the MOS transistor is used as the lower plate polycrystalline film layer of the polycrystalline capacitor, and no additional injection doping process adjustment is made, maintaining the same process conditions as the main process.
[0156] 2) Depositing a polycrystalline film layer for the upper plate of the double polycrystalline capacitor, and completing doping, exposure, and etching to form the upper plate structure of the polycrystalline capacitor.
[0157] For a polycrystalline emitter Bipolar transistor, the outer emitter polycrystalline film layer can be used as the upper plate of the polycrystalline capacitor to simplify the process steps.
[0158] The subsequent steps adopt the normal analog BiCMOS integrated circuit process scheme:
[0159] Embodiment 14:
[0160] For the analog BiCMOS / Bipolar integrated circuit process integrating self-aligned double polycrystalline Bipolar transistors, the integrated production of precision polycrystalline capacitors is carried out on the basis of completing the processes of N-type buried layer, P-type buried layer, epitaxial growth, isolation trench process, N-well injection push-junction, P-well injection push-junction, field oxide layer growth, etc.:
[0161] 1) Using the external base region polycrystalline layer as the polycrystalline film layer of the lower plate of the polycrystalline capacitor, and adopting the polycrystalline plate matching injection doping adjustment described in Implementation 1.
[0162] The implantation dose range of the polycrystalline film layer of the lower electrode of capacitor cell C1 is 1e15cm -2 --1e16cm -2 , maintain the subdegenerate state of the polycrystalline film interface;
[0163] The implantation dose range of the polycrystalline film layer of the lower electrode of capacitor cell C2 is > 1e16cm -2, maintaining the degenerate state of the polycrystalline film interface and forming a shared carrier migration mode; 2) using the external emitter polycrystalline film layer as the upper plate of the polycrystalline capacitor, the dielectric reflux filling after etching the emitter polycrystalline film layer can be used to optimize the edge effect of the polycrystalline film layer on the upper plate of the polycrystalline capacitor.
[0164] The subsequent steps adopt the same process scheme of simulating BiCMOS integrated circuit in Example 12.
Claims
1. Low-noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method, characterized in that: The following steps are involved: 1) Trench isolation is made on the substrate, and then an N-type well region and a P-type well region are formed. 2) Field oxygen-cutoff implantation isolation or trench isolation is used between the N-type well region and the P-type well region; 3) depositing a P0 angstrom capacitor lower polycrystalline film layer on a uniform flat area of a n kilo-angstrom thick field oxide layer; 4) The doping concentration and impurity distribution of the polycrystalline film layer of the lower electrode of the capacitor are adjusted by injecting N-type doping elements; according to the circuit design requirements, the doping concentration and impurity distribution of the polycrystalline film layer in the twin capacitor cells C1 and C2 regions are adjusted by injecting N-type doping elements in the required regions of the polycrystalline film layer of the lower electrode of the capacitor; A polycrystalline structure of the bottom electrode of the twin capacitor cells C1 and C2 is formed; 5) etching the polycrystalline film layer into a centrally symmetrical cell, growing a capacitor dielectric layer of a required thickness according to process requirements, and filling the gap between the twin capacitor cells formed in step 4) to form a capacitor lower plate protection structure; 6) depositing a P1 angstrom polycrystalline film layer on the capacitor dielectric layer and completing N-type element implantation and doping; 7) using a photolithography process to etch the polycrystalline film layer into a centrally symmetrical cell corresponding to the capacitor lower electrode formed in step 4), and the polycrystalline film layer serves as the upper electrode plate of the polycrystalline capacitor; 8) growing an n angstrom thermal oxide layer by a thermal oxidation method; depositing a d1 angstrom silicon nitride dielectric layer to form a capacitor upper plate protection structure; 9) completing the base and emitter region implantation of bipolar transistors and / or the source and drain implantation of field effect transistors, and annealing and activation; 10) Metal silicide exposure and etching, sputtering Ti metal film, capacitor cell C1 and cell C2 polysilicon, other exposed silicon, polysilicon area metal silicide and then stripping off the metal Ti that has not undergone metal silicide reaction; 11) Photolithography dielectric layer deposition, reflow filling and planarization, contact hole etching, tungsten plug filling and planarization; 12) The upper electrode of the capacitor cell C1 is metal interconnected with the upper electrode of the capacitor cell C2; at the same time, the lower electrode of the capacitor cell C1 is metal interconnected with the lower electrode of the capacitor cell C2, and finally a Twin-Combined-Bridge architecture unit is formed.
2. The low-noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method according to claim 1, characterized in that: When the base and emitter region implantation of the bipolar transistor or / and the source and drain implantation of the field effect transistor are completed, no additional implantation and doping process adjustment is performed on the gate of the MOS transistor or the emitter polycrystalline film layer of the bipolar transistor, and the conditions are kept the same as the main process; The implantation doping process is used to match the impurity concentration and interface distribution of the polycrystalline film layer dedicated to passive devices or the base polycrystalline film layer in the double polycrystalline Bipolar process; The implantation dose range of polycrystalline film layer of capacitor cell C1 is 1e 15 cm -2 -1e 16 cm -2 , maintain the subdegenerate state of the polycrystalline film interface; The implantation dose range of the polycrystalline film layer of the lower electrode of the capacitor cell C2 is > 1e 16 cm -2 , maintaining the degenerate state of the polycrystalline film interface and forming a shared carrier migration mode.
3. The low noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method according to claim 1, characterized in that: According to the process flow design, the polycrystalline film of the integrated polycrystalline capacitor electrode is the necessary gate polycrystalline of the MOS device, or the polycrystalline layer deposited by the special passive device process.
4. The low noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method according to claim 1, characterized in that: The process of making trench isolation includes etching and backfilling processes.
5. The low noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method according to claim 1, characterized in that: The capacitor lower polycrystalline film layer of step 3) is deposited by low pressure chemical vapor deposition. The polycrystalline film layer in step 7) is deposited by low pressure chemical vapor deposition.
6. The low noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method according to claim 1, characterized in that: The method of implanting the doped N-type element includes photolithography implantation.
7. The low noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method according to claim 1, characterized in that: The process of etching the polycrystalline film layer into a centrally symmetrical cell to form a capacitor lower plate protection structure includes a photolithography process.
8. The low noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method according to claim 1, characterized in that: When growing an n angstrom thermal oxide layer, the main process temperature is ≤850°C and the main process time is ≤30 minutes.
9. The low noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method according to claim 1, characterized in that: The process of completing the base and emitter region implantation of bipolar transistors or / and the source and drain implantation of field effect transistors includes Bipolar, CMOS or BiCMOS.
10. An integrated capacitor based on the low noise nonlinear calibration process design and mirror twin integrated capacitor manufacturing method according to any one of claims 1 to 9, characterized in that: Mainly includes substrate, isolation groove, N-type buried layer, P-type buried layer, N-type epitaxial layer, N-type well, P-type well, field oxide layer, sacrificial oxide layer, gate oxide layer, polycrystalline thin film layer, silicon dioxide dielectric layer, silicon oxynitride dielectric layer, metal silicide film layer, metal interconnect film layer, passivation layer; The substrate is located at the bottom; the buried layer is located at the bottom of the epitaxial layer; and the well region is located on the surface of the epitaxial layer; The N-type well and the P-type well are isolated by an isolation trench dielectric; The field oxide layer, the sacrificial oxide layer, and the gate oxide layer respectively cover different areas of the well; The double polycrystalline capacitor polycrystalline film layer is located above the field oxide layer above the N-well; The top surface and sidewall of the polycrystalline film layer are covered with different silicon dioxide and silicon nitride composite dielectric layers and a high dielectric constant capacitor dielectric layer. The low dielectric constant filling film layer is filled in the region between silicon-metal 1, polysilicon-metal 1, field oxygen-metal 1 and the multi-layer metal.
