Biosensor with inverted E-shaped heterogeneous gate vertical tunneling field effect transistor and preparation method of biosensor
Through the design of inverted E-type heterogate vertical tunneling field effect transistor structure and L-type nanocavity, the existing biosensors have low sensitivity and complex preparation problems, and the biological with high conduction current and low static power consumption has been achieved. Sensors significantly improve detection capabilities.
Patent Information
- Application Number
- CN202411115912.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-11-26
AI Technical Summary
Existing biosensors have problems with low sensitivity, high static power consumption and complex preparation processes when detecting biomolecules. Especially the leakage current of vertical stacked nanosheet structures is large, and the conduction current is insufficient, making it difficult to integrate at high speed detection and low cost. Implemented on circuit.
Using an inverted E-type heterogate vertical tunneling field effect transistor structure, the gate dielectric layer with high dielectric constant and different work functions are used to increase the carrier tunneling area and form an L-type nanocavity to fix biomolecules. Metal gate improves detection area and sensitivity.
A biosensor with high conduction current, low static power consumption and high sensitivity is realized, which simplifies the preparation process, reduces costs, and significantly improves the detection ability of biomolecules.
Smart Images

Figure CN119029040A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of semiconductor devices, and in particular relates to a biosensor with an inverted E-type heterogate vertical tunneling field effect transistor and a preparation method thereof, which are used for detecting unknown biological molecules. Background Art
[0002] Biosensors are an advanced technology that can monitor and measure physiological parameters, biochemical indicators, and environmental factors in living organisms. Their application range covers disease diagnosis, health monitoring, treatment tracking, and many other aspects. Field effect transistor (FET) biosensors are an emerging biosensing technology that effectively solves the defects of high requirements for equipment and operation, and the cost and difficulty of preparing markers. The ion-sensitive field-effect transistor (ISFET) is an innovative development of the metal oxide field-effect transistor (MOSFET). When target ions are present in biological samples, they undergo specific chemical reactions with the sensitive layer, causing the conductivity of the transistor to change, thereby achieving highly sensitive detection of ion concentrations.
[0003] Dielectric modulation (DM) technology is an innovative technology for detecting biomolecules. The dielectric constant and charge density of biomolecules in the nanocavity can affect the degree of coupling between the gate and the channel, thereby causing changes in the electrical characteristics of the device. This type of biosensor has the characteristics of low cost, small size, and high sensitivity, and has become one of the most widely studied biosensors recently. However, with the continuous development of integrated circuits, the miniaturization of device size has made the common MOSFET short channel effect more and more serious, and the high power consumption it generates is also difficult to allow integration on the system on chip (SOC). At the same time, due to the working principle of hot carrier injection itself, the subthreshold swing is difficult to get rid of the 60mv / dec limit at room temperature, which also makes it difficult to detect at high speed. We need to find another FET device based on a new working principle to make up for these shortcomings.
[0004] The Chinese patent application with publication number CN112736142A discloses a biosensor based on a nanosheet stacked field effect transistor and a preparation method. Three vertically stacked nanosheets form a channel, and a nanocavity is etched on the side close to the source end. However, the channel of the three vertically stacked nanosheets makes the leakage current of the structure larger, there is a higher static power consumption, and the sensitivity in detecting biomolecules is also lower.
[0005] S.Shreya et al. (S.Shreya, AHKhan, N.Kumar, SIAmin, and S.Anand. Core-shell Junctionless Nanotube Tunnel Field Effect Transistor: Design and Sensitivity Analysis for Biosensing Application [J]. IEEE Sensors Journal, 2020, pp (2). 672–679) disclosed a biosensor with a core-shell junctionless nanotube tunneling field effect transistor structure, using metals with appropriate work functions to create source, drain and gate. However, due to its use of charge plasma technology, it has the defect of low on-current and low sensitivity when detecting biomolecules.
[0006] Devi WV et al. (Devi WV, Bhowmick B, Devi PP. N+ Pocket Doped Vertical TFET Based Dielectric-Modulated Biosensor Considering Non-Ideal Hybridization Issue: A Simulation Study [J]. IEEE Transactions on Nanotechnology, 2020, PP (99): 1-1. DOI: 10.1109 / TNANO.2020.2969206.) disclosed a biosensor with an embedded nanocavity based on an n+ pocket doped vertical TFET (VTFET). The device has a nanocavity near the source region, but the tunneling effect of charged molecules at the source-channel junction is improved to a very low degree, and the carrier concentration of tunneling in the source region is almost unaffected by it. Therefore, it has a lower sensitivity when detecting charged molecules. Summary of the invention
[0007] In order to overcome the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide a biosensor with an inverted E-type heterogate vertical tunneling field effect transistor and a preparation method thereof, which increases the area for carrier tunneling by adopting an inverted E-type source region 3 and a channel 2, and forms a U-shaped nanocavity 5 for fixing biological molecules through a vertical metal gate 7 and an adjacent source region 3 and a corresponding horizontal metal gate 6 and a channel 2, providing a larger detection area and improving the detection speed and sensitivity.
[0008] In order to achieve the above object, the technical solution adopted by the present invention is:
[0009] A biosensor with an inverted E-type heterogeneous gate vertical tunneling field effect transistor comprises a drain contact 8, a drain region 1 is arranged on the upper surface of the drain contact 8, a channel 2 is arranged on the upper surface of the drain region 1, three source regions 3 are arranged on the upper surface of the channel 2 to form an inverted E-type structure, a source contact 9 is arranged on the upper surface of each source region 3, a gate dielectric layer 4 with a high dielectric constant is arranged on the upper surface of the channel 2 between every two adjacent source regions 3, a horizontal metal gate 6 with a length greater than the gate dielectric layer 4 itself is arranged on the upper surface of each gate dielectric layer 4, vertical metal gates 7 are arranged at both ends of the upper surface of each horizontal metal gate 6, and each vertical metal gate 7 forms an L-shaped nanocavity 5 for fixing biomolecules with the adjacent source region 3, the corresponding horizontal metal gate 6, the channel 2 and the gate dielectric layer 4.
[0010] The channel 2 is a lightly doped N-type with a doping concentration of 1×10 15 cm -3 ~1×10 17 cm -3 .
[0011] The source region 3 is heavily doped P-type with a doping concentration of 1×10 19 cm -3 ~1×10 20 cm -3 .
[0012] The drain region 1 is heavily doped N-type with a doping concentration of 1×10 18 cm -3 ~1×10 19 cm -3 .
[0013] The source region 3 , the drain region 1 and the channel 2 all use silicon as a semiconductor material, and the gate dielectric layer 4 uses hafnium oxide (HfO 2 ) or aluminum oxide (Al 2 O 3 ) as a material.
[0014] The work function of the horizontal metal gate 6 is 3.6 eV to 3.9 eV, the work function of the vertical metal gate 7 is 4.2 to 4.5 eV, and the materials of the source contact 9 and the drain contact 8 are both polysilicon.
[0015] The drain region 1 and the channel 2 are both 80-110 nm long and 10-20 nm thick; the three source regions 3 are both 10-15 nm long and 20-30 nm thick; the gate dielectric layer 4 is 3-5 nm long and 3-5 nm thick; the horizontal thickness of the nanocavity 5 is consistent with the thickness of the gate dielectric layer 4, the vertical length of the nanocavity 5 is consistent with the horizontal thickness, the L-shaped nanocavity 5 has the same size, and two adjacent L-shaped structures are mirror-imaged between the three source regions, the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region 3, and the sum of the bottom length of the three source regions 3, the bottom length of the four L-shaped structures and the bottom length of the two gate dielectric layers 4 is consistent with the length of the channel 2.
[0016] A method for preparing a biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor comprises the following steps:
[0017] Step 1: Use a heavily doped N-type semiconductor as drain region 1;
[0018] Step 2: epitaxially grow a lightly doped N-type semiconductor as a channel 2 on the upper surface of the drain region 1 prepared in step 1;
[0019] Step 3: epitaxially grow a heavily doped P-type semiconductor on the upper surface of the channel 2 prepared in step 2;
[0020] Step 4: etching the heavily doped P-type semiconductor prepared in step 3 to form three independent source regions 3 and two deposition layers;
[0021] Step 5: depositing a layer of dielectric material with a high dielectric constant in the two deposition layers prepared in step 4;
[0022] Step 6: etching the dielectric material with a high dielectric constant deposited in step 5 to form four L-shaped nanocavities 5, and retaining a portion of the dielectric material with a high dielectric constant as a gate dielectric layer 4;
[0023] Step 7: Two metal materials with different work functions are deposited in the horizontal and vertical directions of the L-shaped nanocavity 5 prepared in step 6 as a horizontal metal gate 6 and a vertical metal gate 7, respectively. At the same time, a layer of semiconductor material is deposited on the surface of the drain region 1 formed in step 1 and on the surfaces of the three source regions 3 formed in step 4 as a drain contact 8 and a source contact 9, respectively.
[0024] The heavily doped N-type semiconductor in step 1 is prepared by ion implantation, wherein phosphorus (P) is used as a dopant with a doping concentration of 1×10 18 cm -3 ~1×10 19 cm -3, control the injection energy to 50keV to 200keV, the entire ion injection process is maintained at 20°C to 27°C, after the ion injection is completed, perform thermal annealing to activate the phosphorus ions, the thermal annealing temperature is 600°C to 1000°C, and the thermal annealing time is controlled to be 30 minutes to 60 minutes;
[0025] In the step 2, a lightly doped N-type semiconductor is epitaxially grown. First, the silicon substrate for epitaxial growth is cleaned, including pickling and ultrasonic cleaning. Then, the cleaned silicon substrate is placed in an epitaxial growth chamber for subsequent operations. The epitaxial growth chamber is a high vacuum or inert gas environment. The growth temperature is set at 1100° C. to 1300° C., and the pressure is set to a low pressure environment of 90 Pa to 100 Pa. The flow rate of the silicon source gas is set at 15 sccm to 20 sccm, and the flow rate of hydrogen as a carrier gas is set at 180 sccm to 200 sccm. During the growth process, phosphine (PH3) is introduced as a dopant gas with a doping concentration of 1×10 15 cm -3 ~1×10 17 cm -3 The growth time is set at 30 minutes to 120 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 700°C to 900°C, and the thermal annealing time is controlled at 20 minutes to 50 minutes.
[0026] In the step 3, a heavily doped P-type semiconductor is epitaxially grown. First, the silicon substrate for epitaxial growth is cleaned, including pickling and ultrasonic cleaning. Then, the cleaned silicon substrate is placed in an epitaxial growth chamber for subsequent operations. The epitaxial growth chamber is a high vacuum or inert gas environment. The growth temperature is set at 1100° C. to 1300° C., and the pressure is a low-pressure environment of 90 Pa to 100 Pa. The flow rate of the silicon source gas is set at 15 sccm to 20 sccm, and the flow rate of hydrogen as a carrier gas is set at 180 sccm to 200 sccm. During the growth process, dimethyl borane (DMBH) is introduced as a dopant gas with a doping concentration of 1×10 19 cm -3 ~1×10 20 cm -3 The growth time is set at 30 minutes to 120 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 800°C to 1100°C, and the thermal annealing time is controlled at 30 minutes to 50 minutes.
[0027] The etching in step 4 is performed by dry etching, using a reactive ion etching (RIE) device for etching, selecting hydrogen fluoride (HF) as the etching gas, setting its flow rate to 20 sccm to 25 sccm, setting the temperature in the etching chamber to 20° C. to 30° C., setting the working pressure of the etching chamber to 60 mTorr to 80 mTorr to reduce unexpected side reactions, setting the RF power to 300 W, the frequency to 13.56 MHz, and the etching time to 10 to 40 minutes, using spectroscopy endpoint detection to determine when the etching process reaches the target depth, and after etching, the residues need to be thoroughly removed to avoid potential impact on device performance;
[0028] The dielectric material deposited in step 5 adopts an atomic layer deposition process (ALD). The atomic layer deposition process (ALD) first performs chemical cleaning or hydrogen treatment on the semiconductor material to remove oxides and organic residues to ensure optimal surface acceptance. Trimethoxyhafnium or trimethoxyaluminum is used as a precursor for depositing a dielectric layer. The precursor is introduced into a reaction chamber to adsorb the precursor on the silicon surface. Oxygen or nitrogen is introduced to remove unreacted precursors and by-products. Water vapor is introduced as an oxidant to react with the adsorbed precursor to generate a dielectric layer film. The purification step is repeated to remove by-products and unreacted gases generated during the reaction. During this process, the temperature is controlled at 200° C. to 400° C. and adjusted according to the characteristics of the ALD precursor and the reaction conditions. The ambient pressure is maintained at 2 to 10 Torr. The thickness of the dielectric layer film is increased by repeating the ALD cycle, and the thickness increases by 5 to 20 angstroms per cycle.
[0029] In step 6, the U-shaped nanocavity and the gate dielectric layer in the middle thereof are dry-etched, and an inductively coupled plasma (ICP) etcher and hydrogen fluoride (HF) are selected as etching gases for operation. The overall ambient temperature needs to be maintained at 20° C. to 25° C., the inductively coupled plasma (ICP) etcher is started to generate plasma, and then HF etching gas is input to ensure that the flow rate is 40 sccm to 50 sccm, and the pressure of the reaction chamber is adjusted to 40 mTorr to 50 mTorr to maintain a stable etching atmosphere. At a discharge power of 200 W to 300 W and a frequency of 13.56 MHz, HF reacts with the surface of the gate dielectric layer to generate fluoride, thereby achieving etching. After the etching is completed, cleaning is performed to remove the etching residues;
[0030] In step 7, the horizontal metal grid 6 and the vertical metal grid 7 are deposited on the cavity by a reactive sputtering process. The reactive sputtering process uses a magnetron sputtering system for deposition. First, the target material is heated to 300°C to 600°C to increase the evaporation rate of the metal. At the same time, the temperature of the substrate is controlled at 100°C to 300°C to avoid thermal damage to the film on the substrate. The pressure of the sputtering chamber is set at 10 -6~10 -8 Torr, start the sputtering system, heat the target materials of the horizontal metal gate 6 and the vertical metal gate 7 and apply an electric field to sputter the target materials from the surface of the target gun to the silicon substrate, set the sputtering power of the material of the horizontal metal gate 6 to 50-200W, and the sputtering power of the material of the vertical metal gate 7 to 30-150W, so as to form the required metal film on the surface of the semiconductor. After the sputtering process is completed, release the vacuum chamber gas and take out the semiconductor substrate, and use characterization technology to evaluate the morphology, composition and bonding state of the metal film. The characterization technology includes SEM and XPS to ensure the quality and stability of the gate electrode; use chemical vapor deposition (CVD) to form the source contact 9 and the drain contact 8. First, preheat the CVD reaction chamber to 500-700°C and control the pressure in the chamber to 10 2 ~10 3 Torr, to promote subsequent deposition, dimethylsilane (SiH2(CH3)2) is introduced as a silicon source gas, and the flow rate is set to 50sccm~100sccm, hydrogen (H2) is used as a diluent gas, and the flow rate is set to 200sccm~500sccm, the silicon source gas is decomposed in the reaction chamber and an electrode film is deposited on the surface of the substrate, and then the deposited electrode film is annealed to improve its crystallinity and electrical properties, the thermal annealing temperature is 500℃~1000℃, and the thermal annealing time is controlled at 20 minutes to 40 minutes.
[0031] The heavily doped N-type semiconductor in step 1 is silicon, and the doping concentration is 1×10 18 cm -3 ~1×10 19 cm -3 , length is 80-110nm, thickness is 10-20nm;
[0032] The lightly doped N-type semiconductor in step 2 is silicon, and the doping concentration is 1×10 15 cm -3 ~1×10 17 cm -3 , length is 80-110nm, thickness is 10-20nm;
[0033] The heavily doped P-type semiconductor in step 3 is silicon, and the doping concentration is 1×10 19 cm -3 ~1×10 20 cm -3 , length is 80-110nm, thickness is 20-30nm;
[0034] Each independent source region 3 in step 4 has the same size and a doping concentration of 1×10 19 cm -3 ~1×1020 cm -3 , with a length of 10 to 15 nm and a thickness of 20 to 30 nm. The two deposited layers have the same size. The length of the two deposited layers plus the length of the three source regions is consistent with the length of channel 2, and the thickness is 20 to 30 nm.
[0035] The dielectric material in step 5 is hafnium oxide (HfO2) or aluminum oxide (Al2O3);
[0036] The gate dielectric layer 4 in step 6 has a length of 3 to 5 nm and a thickness of 3 to 5 nm, the thickness of the nanocavity 5 in the horizontal direction is consistent with the thickness of the gate dielectric layer 4, the length of the nanocavity 5 in the vertical direction is consistent with the thickness in the horizontal direction, the sizes of the L-shaped nanocavities 5 are the same, and two adjacent L-shaped structures are mirror-imaged between the three source regions, the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region 3, and the sum of the bottom lengths of the three source regions 3, the bottom lengths of the four L-shaped structures, and the bottom lengths of the two gate dielectric layers 4 is consistent with the length of the channel 2;
[0037] The material of the horizontal metal gate 6 in the step 7 is metal hafnium, metal chromium or metal titanium, and its length is consistent with the sum of the lengths of the nanocavities 5 on both sides and the bottom length of the L-shaped structure and the gate dielectric layer 4, and its thickness is 2 to 3 nm. The material of the vertical metal gate 7 is metal aluminum, metal silver or metal platinum, and its length is 2 to 3 nm. The top of the vertical metal gate 7 is flush with the top of the source region 3, and the bottom is flush with the top of the horizontal metal gate 6. The materials of the drain contact 8 and the source contact 9 are both polysilicon, and the length of the drain contact 8 is consistent with the length of the drain region 1, and the thickness is 2 to 3 nm. The length of the source contact 9 is consistent with the length of the source region 3, and the thickness is 2 to 3 nm.
[0038] Compared with the prior art, the beneficial effects of the present invention are:
[0039] 1. Each vertical metal gate 7 of the present invention forms an L-shaped nanocavity 5 for fixing biomolecules with the adjacent source region 3, the corresponding horizontal metal gate 6, the channel 2 and the gate dielectric layer 4, which provides a larger detection area. At the same time, the horizontal part of the nanocavity 5 has a greater influence on the modulation of the channel 2, while the vertical part of the nanocavity 5 has a greater influence on the modulation of the source region 3. Therefore, the electrical characteristics of the biosensor are more easily affected by biomolecules.
[0040] 2. The present invention adopts an inverted E-type source region 3 and a channel 2 to increase the area for carrier tunneling. A horizontal metal gate 6 and a vertical metal gate 7 with different work functions are applied to the nanocavity 5 in two different directions. For the gate close to the source region 3, a vertical metal gate 7 with a larger work function is used to slow down the bending of the gate to the source region 3 energy band, while for the gate close to the channel 2, a horizontal metal gate 6 with a smaller work function is used to intensify the bending of the gate to the channel energy band. The significant reduction in the barrier width for carrier tunneling effectively makes up for the deficiency of the traditional TFET device in the on-current.
[0041] 3. The present invention forms a channel 2 and a source region 3 on a heavily doped N-type semiconductor by conventional growth techniques, then forms a gate dielectric layer 4 and an L-shaped cavity by deposition and etching processes, and finally forms a horizontal metal gate and a vertical metal gate on the gate dielectric layer 4 and the L-shaped cavity by deposition techniques, and forms a drain contact 8 and a source contact 9 on the surface of the source region 3 and the drain region 1. Compared with the preparation process of complex structures such as the existing vertically stacked nanosheet biosensors described in the background technology, the preparation process is simpler and the process cost is reduced.
[0042] In summary, the present invention has the advantages of high conduction current, good sensitivity in detecting biomolecules, simple preparation process and low cost. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 It is a schematic structural diagram of the biosensor of the inverted E-type hetero-gate vertical tunneling field effect transistor of the present invention.
[0044] Figure 2 The present invention is a flow chart of a method for preparing a biosensor of an inverted E-type hetero-gate vertical tunneling field effect transistor.
[0045] Figure 3 The transfer characteristic curve diagram of the inverted E-type hetero-gate vertical tunneling field effect transistor biosensor of Example 1 is simulated by computer-aided design software.
[0046] Figure 4 The on-current sensitivity comparison of the inverted E-type hetero-gate vertical tunneling field effect transistor biosensor of Example 1 and the traditional dual-gate TFET biosensor for biomolecules with different dielectric constants is shown.
[0047] Figure 5 The transfer characteristic curve diagram of the inverted E-type hetero-gate vertical tunneling field effect transistor biosensor of Example 2 is simulated by computer-aided design software.
[0048] Figure 6The on-current sensitivity comparison of the inverted E-type hetero-gate vertical tunneling field effect transistor biosensor of Example 2 and the traditional dual-gate TFET biosensor for biomolecules with different dielectric constants is shown.
[0049] Figure 7 The transfer characteristic curve diagram of the inverted E-type hetero-gate vertical tunneling field effect transistor biosensor of Example 3 is simulated by computer-aided design software.
[0050] Figure 8 The on-current sensitivity comparison of the inverted E-type hetero-gate vertical tunneling field effect transistor biosensor of Example 3 and the traditional dual-gate TFET biosensor for biomolecules with different dielectric constants is shown.
[0051] Markings in the figure: 1 is the drain region, 2 is the channel, 3 is the source region, 4 is the gate dielectric layer, 5 is the nanocavity, 6 is the horizontal metal gate, 7 is the vertical metal gate, 8 is the drain contact, and 9 is the source contact. DETAILED DESCRIPTION
[0052] The present invention will be described in detail below in conjunction with the accompanying drawings.
[0053] Example 1
[0054] A biosensor with an inverted E-type heterogeneous gate vertical tunneling field effect transistor comprises a drain contact 8, a drain region 1 is arranged on the upper surface of the drain contact 8, a channel 2 is arranged on the upper surface of the drain region 1, three source regions 3 are arranged on the upper surface of the channel 2 to form an inverted E-type structure, a source contact 9 is arranged on the upper surface of each source region 3, a gate dielectric layer 4 with a high dielectric constant is arranged on the upper surface of the channel 2 between every two adjacent source regions 3, a horizontal metal gate 6 with a length greater than the gate dielectric layer 4 itself is arranged on the upper surface of each gate dielectric layer 4, vertical metal gates 7 are arranged at both ends of the upper surface of each horizontal metal gate 6, and each vertical metal gate 7 forms an L-shaped nanocavity 5 for fixing biomolecules with the adjacent source region 3, the corresponding horizontal metal gate 6, the channel 2 and the gate dielectric layer 4.
[0055] The channel 2 is a lightly doped N-type with a doping concentration of 1×10 15 cm -3 .
[0056] The source region 3 is heavily doped P-type with a doping concentration of 1×10 19 cm -3 .
[0057] The drain region 1 is heavily doped N-type with a doping concentration of 1×10 18 cm -3 .
[0058] The source region 3 , the drain region 1 and the channel 2 all use silicon as a semiconductor material, and the gate dielectric layer 4 uses hafnium oxide (HfO 2 ) material.
[0059] The work function of the horizontal metal gate 6 is 3.6 eV, the work function of the vertical metal gate 7 is 4.2 eV, and the materials of the source contact 9 and the drain contact 8 are both polysilicon.
[0060] The drain region 1 and the channel 2 are both 80 nm long and 10 nm thick; the three source regions 3 are both 10 nm long and 20 nm thick; the gate dielectric layer 4 is 3 nm long and 3 nm thick; the horizontal thickness of the nanocavity 5 is consistent with the thickness of the gate dielectric layer 4, the vertical length of the nanocavity 5 is consistent with the horizontal thickness, the L-shaped nanocavities 5 have the same size, and two adjacent L-shaped structures are mirror-imaged between the three source regions, the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region 3, and the sum of the bottom length of the three source regions 3, the bottom length of the four L-shaped structures and the bottom length of the two gate dielectric layers 4 is consistent with the length of the channel 2.
[0061] A method for preparing a biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor comprises the following steps:
[0062] Step 1: Use ion implantation to implant a heavily doped N-type semiconductor as drain region 1. The heavily doped N-type semiconductor is silicon with a doping concentration of 1×10 18 cm -3 , with a length of 80 nm and a thickness of 10 nm. The ion implantation process uses phosphorus (P) as a dopant with a doping concentration of 1×10 18 cm -3 , the injection energy is controlled to be 50keV, the entire ion injection process is maintained at 20°C, and after the ion injection is completed, a thermal annealing treatment is performed to activate the phosphorus ions, the thermal annealing temperature is 600°C, and the thermal annealing time is controlled to be 30 minutes;
[0063] Step 2: A lightly doped N-type semiconductor is newly grown on the upper surface of the drain region 1 prepared in step 1 by using an epitaxial growth process as a channel 2. The lightly doped N-type semiconductor is silicon with a doping concentration of 1×10 15 cm -3, length is 80nm, thickness is 10nm, the epitaxial growth of lightly doped N-type semiconductor, firstly, the silicon substrate for epitaxial growth is cleaned, including pickling and ultrasonic cleaning, and then the cleaned silicon substrate is placed in an epitaxial growth chamber for subsequent operations, the epitaxial growth chamber is a high vacuum or inert gas environment, the growth temperature is set at 1100°C, the pressure is a low-pressure environment of 90Pa, the flow rate of the silicon source gas is set at 15sccm~, the flow rate of hydrogen as a carrier gas is set at 180sccm, during the growth process, phosphine (PH3) is introduced as a dopant gas, and the doping concentration is 1×10 15 cm -3 The growth time is set at 30 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 700°C and the thermal annealing time is controlled at 20 minutes.
[0064] Step 3: A heavily doped P-type semiconductor is grown on the upper surface of the channel 2 prepared in step 2 by using an epitaxial growth process. The heavily doped P-type semiconductor is silicon with a doping concentration of 1×10 19 cm -3 , length is 80nm, thickness is 20nm, the epitaxial growth of heavily doped P-type semiconductor, firstly, the silicon substrate for epitaxial growth is cleaned, including pickling and ultrasonic cleaning, and then the cleaned silicon substrate is placed in the epitaxial growth chamber for subsequent operations, the epitaxial growth chamber is a high vacuum or inert gas environment, the growth temperature is set at 1100°C, the pressure is a low-pressure environment of 90Pa, the flow rate of silicon source gas is set at 15sccm, and the flow rate of hydrogen as carrier gas is set at 180sccm. During the growth process, dimethyl borane (DMBH) is introduced as a dopant gas, and the doping concentration is 1×10 19 cm -3 The growth time is set at 30 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 800°C and the thermal annealing time is controlled at 30 minutes.
[0065] Step 4: The heavily doped P-type semiconductor prepared in step 3 is etched by dry etching. Reactive ion etching (RIE) equipment is used for etching. Hydrogen fluoride (HF) is selected as the etching gas, and its flow rate is set to 20 sccm. The temperature in the etching chamber is set to 20°C. The working pressure of the etching chamber is set to 60 mTorr to reduce unexpected side reactions. The RF power is set to 300 W, the frequency is set to 13.56 MHz, and the etching time is set to 10 minutes. Spectroscopic endpoint detection is used to determine when the etching process reaches the target depth. After etching, the residue needs to be thoroughly removed to avoid potential impact on device performance. Three independent source regions 3 and two deposition layers are formed. Each independent source region 3 has the same size and a doping concentration of 1×1019 cm -3 , with a length of 10nm and a thickness of 20nm. The two deposited layers have the same size. The length of the two deposited layers plus the length of the three source regions is consistent with the length of channel 2, and the thickness is 20nm;
[0066] Step 5: using an atomic layer deposition process (ALD) to deposit a layer of dielectric material with a high dielectric constant in the two deposition layers prepared in step 4, the dielectric material is hafnium oxide (HfO2), the atomic layer deposition process (ALD) first performs chemical cleaning or hydrogen treatment on the semiconductor material to remove oxides and organic residues to ensure optimal surface acceptance, using trimethoxy hafnium as a precursor for depositing the dielectric layer, introducing the precursor into the reaction chamber, allowing the precursor to be adsorbed on the silicon surface, introducing oxygen to remove unreacted precursors and byproducts, introducing water vapor as an oxidant to react with the adsorbed precursor to generate a dielectric layer film layer, repeating the purification step to remove byproducts and unreacted gases generated during the reaction, in this process, controlling the temperature at 200°C, and adjusting according to the characteristics of the ALD precursor and the reaction conditions, the ambient pressure is maintained at 2Torr, and the thickness of the dielectric layer film is increased by repeating the ALD cycle, and the thickness is increased by 5 angstroms per cycle;
[0067] Step 6: Use dry etching to etch the dielectric material with a high dielectric constant deposited in step 5. Select an inductively coupled plasma (ICP) etcher and hydrogen fluoride (HF) as the etching gas for operation. The overall ambient temperature needs to be maintained at 20°C. Start the inductively coupled plasma (ICP) etcher to generate plasma, then input HF etching gas to ensure a flow rate of 40sccm. Adjust the reaction chamber pressure to 40mTorr to maintain a stable etching atmosphere. At a discharge power of 200W and a frequency of 13.56MHz, HF reacts with the surface of the gate dielectric layer to generate fluoride, thereby achieving etching. After the etching is completed, clean and remove the etching residue. Retained materials are formed to form four L-shaped nanocavities 5, and a portion of dielectric material with a high dielectric constant is retained as a gate dielectric layer 4, the two gate dielectric layers 4 have the same size, a length of 3nm, and a thickness of 3nm, the horizontal thickness of the nanocavity 5 is consistent with the thickness of the gate dielectric layer 4, the vertical length of the nanocavity 5 is consistent with the horizontal thickness, the L-shaped nanocavity 5 has the same size, and two adjacent L-shaped structures are mirror-imaged between the three source regions, the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region 3, and the sum of the bottom length of the three source regions 3, the bottom length of the four L-shaped structures, and the bottom length of the two gate dielectric layers 4 is consistent with the length of the channel 2;
[0068] Step 7: Using a reactive sputtering process, two metal materials with different work functions are deposited in the horizontal and vertical directions of the L-shaped nanocavity 5 prepared in step 6 as a horizontal metal gate 6 and a vertical metal gate 7, respectively. The material of the horizontal metal gate 6 is metal hafnium, metal chromium or metal titanium, and its length is consistent with the sum of the lengths of the nanocavities 5 on both sides and the bottom length of the L-shaped structure and the gate dielectric layer 4, and its thickness is 2nm. The material of the vertical metal gate 7 is metal aluminum, metal silver or metal platinum, and its length is 2nm. The top of the vertical metal gate 7 is flush with the top of the source region 3, and the bottom is flush with the top of the horizontal metal gate 6. At the same time, in step 1 A layer of semiconductor material is deposited on the surface of the formed drain region 1 and the surfaces of the three source regions 3 formed in step 4 as drain contacts 8 and source contacts 9, respectively. The materials of the drain contacts 8 and the source contacts 9 are both made of polysilicon. The length of the drain contact 8 is consistent with the length of the drain region 1 and the thickness is 2nm. The length of the source contact 9 is consistent with the length of the source region 3 and the thickness is 2nm. The reactive sputtering process uses a magnetron sputtering system for deposition. First, the target material is heated to 300°C to increase the evaporation rate of the metal. At the same time, the temperature of the substrate is controlled at 100°C to avoid thermal damage to the film on the substrate. The pressure of the sputtering chamber is set at 10 -6 Torr, start the sputtering system, heat the target materials of the horizontal metal gate 6 and the vertical metal gate 7 and apply an electric field to sputter the target materials from the surface of the target gun to the silicon substrate, set the sputtering power of the material of the horizontal metal gate 6 to 50W, and the sputtering power of the material of the vertical metal gate 7 to 30W, so as to form the required metal film on the surface of the semiconductor. After the sputtering process is completed, release the vacuum chamber gas and take out the semiconductor substrate, and use characterization technology to evaluate the morphology, composition and bonding state of the metal film. The characterization technology includes SEM and XPS to ensure the quality and stability of the gate electrode; use chemical vapor deposition (CVD) to form the source contact 9 and the drain contact 8. First, preheat the CVD reaction chamber to 500°C and control the pressure in the chamber to 10 2 Torr, to promote subsequent deposition, dimethylsilane (SiH2(CH3)2) is introduced as the silicon source gas with a flow rate of 50sccm, and hydrogen (H2) is used as the diluent gas with a flow rate of 200sccm. The silicon source gas is decomposed in the reaction chamber and an electrode film is deposited on the surface of the substrate. The deposited electrode film is then annealed to improve its crystallinity and electrical properties. The thermal annealing temperature is 500°C and the thermal annealing time is controlled to 20 minutes.
[0069] Example 2
[0070] A biosensor with an inverted E-type heterogeneous gate vertical tunneling field effect transistor comprises a drain contact 8, a drain region 1 is arranged on the upper surface of the drain contact 8, a channel 2 is arranged on the upper surface of the drain region 1, three source regions 3 are arranged on the upper surface of the channel 2 to form an inverted E-type structure, a source contact 9 is arranged on the upper surface of each source region 3, a gate dielectric layer 4 with a high dielectric constant is arranged on the upper surface of the channel 2 between every two adjacent source regions 3, a horizontal metal gate 6 with a length greater than the gate dielectric layer 4 itself is arranged on the upper surface of each gate dielectric layer 4, vertical metal gates 7 are arranged at both ends of the upper surface of each horizontal metal gate 6, and each vertical metal gate 7 forms an L-shaped nanocavity 5 for fixing biomolecules with the adjacent source region 3, the corresponding horizontal metal gate 6, the channel 2 and the gate dielectric layer 4.
[0071] The channel 2 is a lightly doped N-type with a doping concentration of 1×10 17 cm -3 .
[0072] The source region 3 is heavily doped P-type with a doping concentration of 1×10 20 cm -3 .
[0073] The drain region 1 is heavily doped N-type with a doping concentration of 1×10 19 cm -3 .
[0074] The source region 3 , the drain region 1 and the channel 2 all use silicon as a semiconductor material, and the gate dielectric layer 4 uses aluminum oxide (Al 2 O 3 ) material.
[0075] The work function of the horizontal metal gate 6 is 3.9 eV, the work function of the vertical metal gate 7 is 4.5 eV, and the materials of the source contact 9 and the drain contact 8 are both polysilicon.
[0076] The drain region 1 and the channel 2 are both 110 nm long and 20 nm thick; the three source regions 3 are both 15 nm long and 30 nm thick; the gate dielectric layer 4 is 5 nm long and 5 nm thick; the horizontal thickness of the nanocavity 5 is consistent with the thickness of the gate dielectric layer 4, the vertical length of the nanocavity 5 is consistent with the horizontal thickness, the L-shaped nanocavities 5 have the same size, and two adjacent L-shaped structures are mirror-imaged between the three source regions, the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region 3, and the sum of the bottom length of the three source regions 3, the bottom length of the four L-shaped structures and the bottom length of the two gate dielectric layers 4 is consistent with the length of the channel 2.
[0077] A method for preparing a biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor comprises the following steps:
[0078] Step 1: Use ion implantation to implant a heavily doped N-type semiconductor as drain region 1. The heavily doped N-type semiconductor is silicon with a doping concentration of 1×10 19 cm -3 , with a length of 110 nm and a thickness of 20 nm. The ion implantation process uses phosphorus (P) as a dopant with a doping concentration of 1×10 19 cm -3 , the injection energy is controlled to be 200keV, the entire ion injection process is maintained at 27°C, and after the ion injection is completed, a thermal annealing treatment is performed to activate the phosphorus ions, the thermal annealing temperature is 1000°C, and the thermal annealing time is controlled to be 60 minutes;
[0079] Step 2: A lightly doped N-type semiconductor is newly grown on the upper surface of the drain region 1 prepared in step 1 by using an epitaxial growth process as a channel 2. The lightly doped N-type semiconductor is silicon with a doping concentration of 1×10 17 cm -3 , length 110nm, thickness 20nm, the epitaxial growth of lightly doped N-type semiconductor, first of all, the silicon substrate for epitaxial growth is cleaned, including pickling and ultrasonic cleaning, then the cleaned silicon substrate is placed in the epitaxial growth chamber for subsequent operations, the epitaxial growth chamber is a high vacuum or inert gas environment, the growth temperature is set at 1300 ° C, the pressure is 100Pa low pressure environment, the flow rate of silicon source gas is set at 20sccm, the flow rate of hydrogen as carrier gas is set at 200sccm, during the growth process, phosphine (PH3) is introduced as a dopant gas, the doping concentration is 1×10 17 cm -3 The growth time is set at 120 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 900°C and the thermal annealing time is controlled at 50 minutes.
[0080] Step 3: A heavily doped P-type semiconductor is grown on the upper surface of the channel 2 prepared in step 2 by using an epitaxial growth process. The heavily doped P-type semiconductor is silicon with a doping concentration of 1×10 20 cm -3 , with a length of 110 nm and a thickness of 30 nm. The epitaxial growth of a lightly doped N-type semiconductor first cleans the silicon substrate for epitaxial growth, including pickling and ultrasonic cleaning, and then places the cleaned silicon substrate in an epitaxial growth chamber for subsequent operations. The epitaxial growth chamber is a high vacuum or inert gas environment, and the growth temperature is set at 1300°C, the pressure is a low-pressure environment of 100 Pa, the flow rate of the silicon source gas is set at 20 sccm, and the flow rate of hydrogen as a carrier gas is set at 200 sccm. During the growth process, dimethyl borane (DMBH) is introduced as a dopant gas, and the doping concentration is 1×10 20cm -3 The growth time is set at 120 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 1100°C and the thermal annealing time is controlled at 50 minutes.
[0081] Step 4: Dry etching is used to etch the heavily doped P-type semiconductor prepared in step 3. Reactive ion etching (RIE) equipment is used for etching. Hydrogen fluoride (HF) is selected as the etching gas, and its flow rate is set to 25 sccm. The temperature in the etching chamber is set to 30°C, and the working pressure of the etching chamber is set to 80 mTorr to reduce unexpected side reactions. The RF power is set to 300 W, the frequency is set to 13.56 MHz, and the etching time is set to 40 minutes. Spectroscopic endpoint detection is used to determine when the etching process reaches the target depth. After etching, the residue needs to be thoroughly removed to avoid potential impact on device performance. Three independent source regions 3 and two deposition layers are formed. Each independent source region 3 has the same size and a doping concentration of 1×10 20 cm -3 , with a length of 15nm and a thickness of 30nm. The two deposited layers have the same size. The length of the two deposited layers plus the length of the three source regions is consistent with the length of channel 2, and the thickness is 30nm;
[0082] Step 5: using an atomic layer deposition process (ALD) to deposit a layer of dielectric material with a high dielectric constant in the two deposition layers prepared in step 4, the dielectric material is aluminum oxide (Al2O3), the atomic layer deposition process (ALD) first performs chemical cleaning or hydrogen treatment on the semiconductor material to remove oxides and organic residues to ensure optimal surface acceptance, using trimethoxyaluminum as a precursor for depositing the dielectric layer, introducing the precursor into the reaction chamber, allowing the precursor to be adsorbed on the silicon surface, introducing nitrogen to remove unreacted precursors and by-products, introducing water vapor as an oxidant to react with the adsorbed precursor to generate a dielectric layer film layer, repeating the purification step to remove by-products and unreacted gases generated during the reaction process, in this process, controlling the temperature at 400°C, and adjusting according to the characteristics of the ALD precursor and the reaction conditions, the ambient pressure is maintained at 10 Torr, and the thickness of the dielectric layer film is increased by repeating the ALD cycle, and the thickness is increased by 20 angstroms per cycle;
[0083] Step 6: Use dry etching to etch the dielectric material with a high dielectric constant deposited in step 5. Select an inductively coupled plasma (ICP) etcher and hydrogen fluoride (HF) as the etching gas for operation. The overall ambient temperature needs to be maintained at 25°C. Start the inductively coupled plasma (ICP) etcher to generate plasma, then input HF etching gas to ensure a flow rate of 50 sccm. Adjust the reaction chamber pressure to 50 mTorr to maintain a stable etching atmosphere. At a discharge power of 300 W and a frequency of 13.56 MHz, HF reacts with the surface of the gate dielectric layer to generate fluoride, thereby achieving etching. After the etching is completed, clean and remove the etching residue. Retained materials are formed to form four L-shaped nanocavities 5, and a portion of dielectric material with a high dielectric constant is retained as a gate dielectric layer 4, the two gate dielectric layers 4 have the same size, a length of 5nm, and a thickness of 5nm, the horizontal thickness of the nanocavity 5 is consistent with the thickness of the gate dielectric layer 4, the vertical length of the nanocavity 5 is consistent with the horizontal thickness, the L-shaped nanocavity 5 has the same size, and two adjacent L-shaped structures are mirror-imaged between the three source regions, the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region 3, and the sum of the bottom length of the three source regions 3, the bottom length of the four L-shaped structures, and the bottom length of the two gate dielectric layers 4 is consistent with the length of the channel 2;
[0084] Step 7: Using a reactive sputtering process, two metal materials with different work functions are deposited in the horizontal and vertical directions of the L-shaped nanocavity 5 prepared in step 6 as a horizontal metal gate 6 and a vertical metal gate 7, respectively. The material of the horizontal metal gate 6 is metal hafnium, metal chromium or metal titanium, and its length is consistent with the sum of the lengths of the nanocavities 5 on both sides and the bottom length of the L-shaped structure and the gate dielectric layer 4, and its thickness is 3nm. The material of the vertical metal gate 7 is metal aluminum, metal silver or metal platinum, and its length is 3nm. The top of the vertical metal gate 7 is flush with the top of the source region 3, and the bottom is flush with the top of the horizontal metal gate 6. At the same time, in step 1 A layer of semiconductor material is deposited on the surface of the formed drain region 1 and the surfaces of the three source regions 3 formed in step 4 as drain contacts 8 and source contacts 9, respectively. The materials of the drain contacts 8 and the source contacts 9 are both made of polysilicon. The length of the drain contact 8 is consistent with the length of the drain region 1 and the thickness is 3nm. The length of the source contact 9 is consistent with the length of the source region 3 and the thickness is 3nm. The reactive sputtering process uses a magnetron sputtering system for deposition. First, the target material is heated to 600°C to increase the evaporation rate of the metal. At the same time, the temperature of the substrate is controlled at 300°C to avoid thermal damage to the film on the substrate. The pressure of the sputtering chamber is set at 10 -8Torr, start the sputtering system, heat the target materials of the horizontal metal gate 6 and the vertical metal gate 7 and apply an electric field to sputter the target materials from the surface of the target gun to the silicon substrate, set the sputtering power of the material of the horizontal metal gate 6 to 200W, and the sputtering power of the material of the vertical metal gate 7 to 150W, so as to form the required metal film on the surface of the semiconductor. After the sputtering process is completed, release the vacuum chamber gas and take out the semiconductor substrate, and use characterization technology to evaluate the morphology, composition and bonding state of the metal film. The characterization technology includes SEM and XPS to ensure the quality and stability of the gate electrode; use chemical vapor deposition (CVD) to form the source contact 9 and the drain contact 8. First, preheat the CVD reaction chamber to 700°C and control the pressure in the chamber to 10 3 Torr, to promote subsequent deposition, dimethylsilane (SiH2(CH3)2) is introduced as the silicon source gas with a flow rate of 100sccm, and hydrogen (H2) is used as the diluent gas with a flow rate of 500sccm. The silicon source gas is decomposed in the reaction chamber and an electrode film is deposited on the surface of the substrate. The deposited electrode film is then annealed to improve its crystallinity and electrical properties. The thermal annealing temperature is 1000°C and the thermal annealing time is controlled at 40 minutes.
[0085] Example 3
[0086] A biosensor with an inverted E-type heterogeneous gate vertical tunneling field effect transistor comprises a drain contact 8, a drain region 1 is arranged on the upper surface of the drain contact 8, a channel 2 is arranged on the upper surface of the drain region 1, three source regions 3 are arranged on the upper surface of the channel 2 to form an inverted E-type structure, a source contact 9 is arranged on the upper surface of each source region 3, a gate dielectric layer 4 with a high dielectric constant is arranged on the upper surface of the channel 2 between every two adjacent source regions 3, a horizontal metal gate 6 with a length greater than the gate dielectric layer 4 itself is arranged on the upper surface of each gate dielectric layer 4, vertical metal gates 7 are arranged at both ends of the upper surface of each horizontal metal gate 6, and each vertical metal gate 7 forms an L-shaped nanocavity 5 for fixing biomolecules with the adjacent source region 3, the corresponding horizontal metal gate 6, the channel 2 and the gate dielectric layer 4.
[0087] The channel 2 is a lightly doped N-type with a doping concentration of 1×10 16 cm -3 .
[0088] The source region 3 is heavily doped P-type with a doping concentration of 5×10 19 cm -3 .
[0089] The drain region 1 is heavily doped N-type with a doping concentration of 5×10 18 cm -3 cm -3 .
[0090] The source region 3 , the drain region 1 and the channel 2 all use silicon as a semiconductor material, and the gate dielectric layer 4 uses hafnium oxide (HfO 2 ) material.
[0091] The work function of the horizontal metal gate 6 is 3.7 eV, the work function of the vertical metal gate 7 is 4.4 eV, and the materials of the source contact 9 and the drain contact 8 are both polysilicon.
[0092] The drain region 1 and the channel 2 are both 100 nm long and 15 nm thick; the three source regions 3 are both 12 nm long and 25 nm thick; the gate dielectric layer 4 is 4 nm long and 4 nm thick; the horizontal thickness of the nanocavity 5 is consistent with the thickness of the gate dielectric layer 4, the vertical length of the nanocavity 5 is consistent with the horizontal thickness, the L-shaped nanocavities 5 have the same size, and two adjacent L-shaped structures are mirror-imaged between the three source regions, the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region 3, and the sum of the bottom length of the three source regions 3, the bottom length of the four L-shaped structures and the bottom length of the two gate dielectric layers 4 is consistent with the length of the channel 2.
[0093] A method for preparing a biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor comprises the following steps:
[0094] Step 1: Use ion implantation to implant a heavily doped N-type semiconductor as drain region 1. The heavily doped N-type semiconductor is silicon with a doping concentration of 5×10 18 cm -3 , with a length of 100 nm and a thickness of 15 nm. The ion implantation process uses phosphorus (P) as a dopant with a doping concentration of 5×10 18 cm -3 , the injection energy is controlled to be 100keV, the entire ion injection process is maintained at 25°C, and after the ion injection is completed, a thermal annealing treatment is performed to activate the phosphorus ions, the thermal annealing temperature is 800°C, and the thermal annealing time is controlled to be 50 minutes;
[0095] Step 2: A lightly doped N-type semiconductor is newly grown on the upper surface of the drain region 1 prepared in step 1 by using an epitaxial growth process as a channel 2. The lightly doped N-type semiconductor is silicon with a doping concentration of 1×10 16 cm -3, with a length of 100 nm and a thickness of 15 nm. The epitaxial growth of a lightly doped N-type semiconductor first involves cleaning the silicon substrate for epitaxial growth, including pickling and ultrasonic cleaning, and then placing the cleaned silicon substrate in an epitaxial growth chamber for subsequent operations. The epitaxial growth chamber is a high vacuum or inert gas environment, and the growth temperature is set at 1200°C, the pressure is a low-pressure environment of 100 Pa, the flow rate of the silicon source gas is set at 18 sccm, and the flow rate of hydrogen as a carrier gas is set at 190 sccm. During the growth process, phosphine (PH3) is introduced as a dopant gas, and the doping concentration is 1×10 16 cm -3 The growth time is set at 60 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 800°C and the thermal annealing time is controlled at 40 minutes.
[0096] Step 3: A heavily doped P-type semiconductor is grown on the upper surface of the channel 2 prepared in step 2 by using an epitaxial growth process. The heavily doped P-type semiconductor is silicon with a doping concentration of 5×10 19 cm -3 , with a length of 100 nm and a thickness of 25 nm. The epitaxial growth of heavily doped P-type semiconductors first cleans the silicon substrate for epitaxial growth, including pickling and ultrasonic cleaning, and then places the cleaned silicon substrate in an epitaxial growth chamber for subsequent operations. The epitaxial growth chamber is a high vacuum or inert gas environment, and the growth temperature is set at 1200°C, the pressure is a low-pressure environment of 100 Pa, the flow rate of the silicon source gas is set at 18 sccm, and the flow rate of hydrogen as a carrier gas is set at 190 sccm. During the growth process, dimethyl borane (DMBH) is introduced as a dopant gas, and the doping concentration is 5×10 19 cm -3 The growth time is set at 70 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 1000°C and the thermal annealing time is controlled at 40 minutes.
[0097] Step 4: Dry etching is used to etch the heavily doped P-type semiconductor prepared in step 3. Reactive ion etching (RIE) equipment is used for etching. Hydrogen fluoride (HF) is selected as the etching gas, and its flow rate is set to 22 sccm. The temperature in the etching chamber is set at 25°C, and the working pressure of the etching chamber is set to 70 mTorr to reduce unexpected side reactions. The RF power is set to 300 W, the frequency is set to 13.56 MHz, and the etching time is set to 30 minutes. Spectroscopic endpoint detection is used to determine when the etching process reaches the target depth. After etching, the residue needs to be thoroughly removed to avoid potential impact on device performance. Three independent source regions 3 and two deposition layers are formed. Each independent source region 3 has the same size and a doping concentration of 5×10 19 cm -3 , with a length of 12nm and a thickness of 25nm. The two deposited layers have the same size. The length of the two deposited layers plus the length of the three source regions is consistent with the length of channel 2, and the thickness is 25nm;
[0098] Step 5: using an atomic layer deposition process (ALD) to deposit a layer of dielectric material with a high dielectric constant in the two deposition layers prepared in step 4, the dielectric material is hafnium oxide (HfO2), the atomic layer deposition process (ALD) first performs chemical cleaning or hydrogen treatment on the semiconductor material to remove oxides and organic residues to ensure optimal surface acceptance, using trimethoxy hafnium as a precursor for depositing the dielectric layer, introducing the precursor into the reaction chamber, allowing the precursor to be adsorbed on the silicon surface, introducing nitrogen to remove unreacted precursors and by-products, introducing water vapor as an oxidant to react with the adsorbed precursor to generate a dielectric layer film layer, repeating the purification step to remove by-products and unreacted gases generated during the reaction, in this process, controlling the temperature at 300°C, and adjusting according to the characteristics of the ALD precursor and the reaction conditions, the ambient pressure is maintained at 5 Torr, and the thickness of the dielectric layer film is increased by repeating the ALD cycle, and the thickness is increased by 10 angstroms per cycle;
[0099] Step 6: Use dry etching to etch the dielectric material with a high dielectric constant deposited in step 5. Select an inductively coupled plasma (ICP) etcher and hydrogen fluoride (HF) as the etching gas for operation. The overall ambient temperature needs to be maintained at 22°C. Start the inductively coupled plasma (ICP) etcher to generate plasma, then input HF etching gas to ensure a flow rate of 45sccm. Adjust the reaction chamber pressure to 45mTorr to maintain a stable etching atmosphere. At a discharge power of 250W and a frequency of 13.56MHz, HF reacts with the surface of the gate dielectric layer to generate fluoride, thereby achieving etching. After the etching is completed, clean and remove the etching residue. Retained materials are formed to form four L-shaped nanocavities 5, and a portion of dielectric material with a high dielectric constant is retained as a gate dielectric layer 4, the two gate dielectric layers 4 have the same size, a length of 4nm, and a thickness of 4nm, the horizontal thickness of the nanocavity 5 is consistent with the thickness of the gate dielectric layer 4, the vertical length of the nanocavity 5 is consistent with the horizontal thickness, the L-shaped nanocavities 5 have the same size, and two adjacent L-shaped structures are mirror-imaged between the three source regions, the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region 3, and the sum of the bottom lengths of the three source regions 3, the bottom lengths of the four L-shaped structures, and the bottom lengths of the two gate dielectric layers 4 is consistent with the length of the channel 2;
[0100] Step 7: Using a reactive sputtering process, two metal materials with different work functions are deposited in the horizontal and vertical directions of the L-shaped nanocavity 5 prepared in step 6 as a horizontal metal gate 6 and a vertical metal gate 7, respectively. The material of the horizontal metal gate 6 is metal hafnium, metal chromium or metal titanium, and its length is consistent with the sum of the lengths of the nanocavities 5 on both sides and the bottom length of the L-shaped structure and the gate dielectric layer 4, and its thickness is 2 to 3 nm. The material of the vertical metal gate 7 is metal aluminum, metal silver or metal platinum, and its length is 2 to 3 nm. The top of the vertical metal gate 7 is flush with the top of the source region 3, and the bottom is flush with the top of the horizontal metal gate 6. At the same time, in step A layer of semiconductor material is deposited on the surface of the drain region 1 formed in step 1 and the surfaces of the three source regions 3 formed in step 4 as drain contacts 8 and source contacts 9, respectively. The materials of the drain contacts 8 and the source contacts 9 are both made of polysilicon. The length of the drain contact 8 is consistent with the length of the drain region 1 and the thickness is 3nm. The length of the source contact 9 is consistent with the length of the source region 3 and the thickness is 3nm. The reactive sputtering process uses a magnetron sputtering system for deposition. First, the target material is heated to 400°C to increase the evaporation rate of the metal. At the same time, the temperature of the substrate is controlled at 200°C to avoid thermal damage to the film on the substrate. The pressure of the sputtering chamber is set at 10 -7Torr, start the sputtering system, heat the target materials of the horizontal metal gate 6 and the vertical metal gate 7 and apply an electric field to sputter the target materials from the surface of the target gun to the silicon substrate, set the sputtering power of the horizontal metal gate 6 material to 60W, and the sputtering power of the vertical metal gate 7 material to 100W, so as to form the required metal film on the semiconductor surface. After the sputtering process is completed, release the vacuum chamber gas and take out the semiconductor substrate, and use characterization technology to evaluate the morphology, composition and bonding state of the metal film. The characterization technology includes SEM and XPS to ensure the quality and stability of the gate electrode; use chemical vapor deposition (CVD) D) is used to form a source contact 9 and a drain contact 8. First, the CVD reaction chamber is preheated to 600°C and the pressure in the chamber is controlled at 10 Torr to promote subsequent deposition. Dimethylsilane (SiH2(CH3)2) is introduced as a silicon source gas with a flow rate of 70 sccm. Hydrogen (H2) is used as a diluent gas with a flow rate of 400 sccm. The silicon source gas is decomposed in the reaction chamber and an electrode film is deposited on the surface of the substrate. The deposited electrode film is then annealed to improve its crystallinity and electrical properties. The thermal annealing temperature is 700°C and the thermal annealing time is controlled at 30 minutes.
[0101] Simulation experiment
[0102] The present invention detects biomolecules based on a tunnel field effect transistor (TFET) biosensor. Its working principle is: when biomolecules with different dielectric constants enter the nanocavity 5, they will have different effects on the coupling between the horizontal metal gate 6, the vertical metal gate 7 and the channel 2, thereby affecting the carrier concentration of the source region 3 tunneling to the channel 2, which can be observed from the electrical characteristic curve of the device.
[0103] Figure 3 The transfer characteristic curve of the biosensor in Example 1 when detecting biomolecules with different dielectric constants (K) at a drain-source voltage of 1.0 V, where K = 1 represents air (i.e., no biomolecules enter the cavity), and K = 3, 7, 12 represent biomolecules with different dielectric constants. It can be seen that the saturation current of the structure is relatively large when it is turned on due to the presence of three source regions 3, which is about 1.01×10 -5 A / μm, when detecting biological molecules with K=7, it is about 1.08×10 -6 A / μm, when detecting biological molecules with K=3, it is about 8.85×10 -11 A / μm. At the same time, the leakage current I OFF It is also relatively low, at 1.2×10 -16A / μm, indicating that the device has lower static power consumption. At the same time, the L-shaped nanocavity 5 provides a larger detection area, which has a greater influence on the modulation of the energy bands of the channel 2 and the source region 3. The difference between the saturation current when detecting different dielectric constants and the saturation current when K=1 is more obvious, which improves the sensitivity of the TFET biosensor.
[0104] Figure 4 The figure shows the comparison of the on-current sensitivity of the biosensor in Example 1 and the conventional TFET biosensor when detecting neutral biomolecules with different dielectric constants. As can be seen from the figure, with the increase of dielectric constant, the sensitivity of both gradually increases, but the structure proposed by us has better sensitivity. The sensitivity of the biosensor in Example 1 when detecting neutral biomolecules with K=3 is about 5.1×10 3 The sensitivity of detecting neutral molecules with K=7 is about 6.3×10 7 The sensitivity of detecting neutral molecules with K=12 is about 5.8×10 8 Compared with the traditional TFET biosensor, the sensor is improved by about two orders of magnitude when detecting biological molecules with K=3, and by about six orders of magnitude when detecting K=7 and K=12, which shows that the structure proposed in the present invention has excellent detection performance.
[0105] Figure 5 The transfer characteristic curve of the biosensor in Example 2 when detecting biomolecules with different dielectric constants (K) at a drain-source voltage of 1.0 V, where K = 1 represents air (i.e., no biomolecules enter the cavity), and K = 3, 7, 12 represent biomolecules with different dielectric constants. It can be seen that the saturation current of the structure is relatively large when it is turned on due to the presence of three source regions 3, which is about 1.08×10 -5 A / μm, when detecting biological molecules with K=7, it is about 1.68×10 -6 A / μm, when detecting biological molecules with K=3, is about 3.99×10 -10 A / μm. At the same time, the leakage current I OFF It is also relatively low, at 1.03×10 -17 A / μm, indicating that the device has lower static power consumption. At the same time, the L-shaped nanocavity 5 provides a larger detection area, which has a greater influence on the modulation of the channel and source bands. The difference between the saturation current when detecting different dielectric constants and the saturation current when K=1 is more obvious, which improves the sensitivity of the TFET biosensor.
[0106] Figure 6The figure shows the comparison of the on-current sensitivity of the biosensor in Example 2 and the conventional TFET biosensor when detecting neutral biomolecules with different dielectric constants. As can be seen from the figure, with the increase of dielectric constant, the sensitivity of both gradually increases, but the structure proposed by us has better sensitivity. The sensitivity of the biosensor in Example 2 when detecting neutral biomolecules with K=3 is about 5.3×10 3 The sensitivity of detecting neutral molecules with K=7 is about 2.2×10 7 The sensitivity of detecting neutral molecules with K=12 is about 1.4×10 8 Compared with the traditional TFET biosensor, the sensor is improved by about two orders of magnitude when detecting biological molecules with K=3, and by about six orders of magnitude when detecting K=7 and K=12, which shows that the structure proposed in the present invention has excellent detection performance.
[0107] Figure 7 The transfer characteristic curve of the biosensor in Example 3 when detecting biomolecules with different dielectric constants (K) at a drain-source voltage of 1.0 V, where K = 1 represents air (i.e., no biomolecules enter the cavity), and K = 3, 7, 12 represent biomolecules with different dielectric constants. It can be seen that the saturation current of the structure is relatively large when it is turned on due to the presence of three source regions 3, which is about 2.81×10 -5 A / μm, when detecting biological molecules with K=7, is about 3.97×10 -6 A / μm, when detecting biological molecules with K=3, is about 8.46×10 -10 A / μm. At the same time, the leakage current I OFF It is also relatively low, at 2.1×10 -16 A / μm, indicating that the device has lower static power consumption. At the same time, the L-shaped nanocavity 5 provides a larger detection area, which has a greater influence on the modulation of the channel and source bands. The difference between the saturation current when detecting different dielectric constants and the saturation current when K=1 is more obvious, which improves the sensitivity of the TFET biosensor.
[0108] Figure 8 The figure shows the comparison of the on-current sensitivity of the biosensor in Example 3 and the conventional TFET biosensor when detecting neutral biomolecules with different dielectric constants. As can be seen from the figure, with the increase of dielectric constant, the sensitivity of both gradually increases, but the structure proposed by us has better sensitivity. The sensitivity of the biosensor in Example 3 when detecting neutral biomolecules with K=3 is about 4.8×10 3 The sensitivity of detecting neutral molecules with K=7 is about 2.4×10 7The sensitivity of detecting neutral molecules with K=12 is about 1.6×10 8 Compared with the traditional TFET biosensor, the sensor is improved by about two orders of magnitude when detecting biological molecules with K=3, and by about six orders of magnitude when detecting K=7 and K=12, which shows that the structure proposed in the present invention has excellent detection performance.
Claims
1. A biosensor with an inverted E-type heterogate vertical tunneling field effect transistor, It is characterized in that The invention comprises a drain contact (8), a drain region (1) is arranged on the upper surface of the drain contact (8), a channel (2) is arranged on the upper surface of the drain region (1), three source regions (3) are arranged on the upper surface of the channel (2), forming an inverted E-type structure, a source contact (9) is arranged on the upper surface of each source region (3), a gate dielectric layer (4) with a high dielectric constant is arranged on the upper surface of the channel (2) between every two adjacent source regions (3), a horizontal metal gate (6) with a length greater than the gate dielectric layer (4) itself is arranged on the upper surface of each gate dielectric layer (4), vertical metal gates (7) are arranged at both ends of the upper surface of each horizontal metal gate (6), and each vertical metal gate (7) forms an L-shaped nanocavity (5) for fixing biological molecules with the adjacent source region (3), the corresponding horizontal metal gate (6), the channel (2) and the gate dielectric layer (4).
2. A biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor according to claim 1, It is characterized in that The channel (2) is a lightly doped N-type semiconductor with a doping concentration of 1×10 15 cm -3 ~1×10 17 cm -3 .
3. A biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor according to claim 1, It is characterized in that The source region (3) is heavily doped P-type, with a doping concentration of 1×10 19 cm -3 ~1×10 20 cm -3 .
4. A biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor according to claim 1, It is characterized in that The drain region (1) is heavily doped N-type with a doping concentration of 1×10 18 cm -3 ~1×10 19 cm -3 .
5. A biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor according to claim 1, It is characterized in that The source region (3), the drain region (1) and the channel (2) all use silicon as a semiconductor material, and the gate dielectric layer (4) uses hafnium oxide (HfO 2 ) or aluminum oxide (Al 2 O 3 ) as material.
6. A biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor according to claim 1, It is characterized in that The work function of the horizontal metal gate (6) is 3.6 eV to 3.9 eV, the work function of the vertical metal gate (7) is 4.2 to 4.5 eV, and the materials of the source contact (9) and the drain contact (8) are both polysilicon.
7. A biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor according to claim 1, It is characterized in that The drain region (1) and the channel (2) are both 80-110 nm long and 10-20 nm thick; the three source regions (3) are both 10-15 nm long and 20-30 nm thick; the gate dielectric layer (4) is 3-5 nm long and 3-5 nm thick; the thickness of the nanocavity (5) in the horizontal direction is consistent with the thickness of the gate dielectric layer (4); the length of the nanocavity (5) in the vertical direction is consistent with the thickness in the horizontal direction; the dimensions of the L-shaped nanocavities (5) are all the same; and two adjacent L-shaped structures are arranged in a mirror image between the three source regions; the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region (3); the sum of the bottom length of the three source regions (3), the bottom length of the four L-shaped structures and the bottom length of the two gate dielectric layers (4) is consistent with the length of the channel (2).
8. A method for preparing a biosensor having an inverted E-type heterogate vertical tunneling field effect transistor according to any one of claims 1 to 7, It is characterized in that The following steps are involved: Step 1: A heavily doped N-type semiconductor is used as a drain region (1); Step 2: epitaxially growing a lightly doped N-type semiconductor as a channel (2) on the upper surface of the drain region (1) prepared in step 1; Step 3: epitaxially growing a heavily doped P-type semiconductor on the upper surface of the channel (2) prepared in step 2; Step 4: etching the heavily doped P-type semiconductor prepared in step 3 to form three independent source regions (3) and two deposition layers; Step 5: depositing a layer of dielectric material with a high dielectric constant in the two deposition layers prepared in step 4; Step 6: etching the dielectric material with a high dielectric constant deposited in step 5 to form four L-shaped nanocavities (5), and retaining a portion of the dielectric material with a high dielectric constant as a gate dielectric layer (4); Step 7: Two metal materials with different work functions are deposited in the horizontal and vertical directions of the L-shaped nanocavity (5) prepared in step 6 as a horizontal metal gate (6) and a vertical metal gate (7), respectively. At the same time, a layer of semiconductor material is deposited on the surface of the drain region (1) formed in step 1 and the surfaces of the three source regions (3) formed in step 4 as a drain contact (8) and a source contact (9).
9. A method for preparing a biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor according to claim 8, It is characterized in that The heavily doped N-type semiconductor in step 1 is prepared by ion implantation, wherein phosphorus (P) is used as a dopant with a doping concentration of 1×10 18 cm -3 ~1×10 19 cm -3 , control the injection energy to 50keV to 200keV, the entire ion injection process is maintained at 20°C to 27°C, after the ion injection is completed, perform thermal annealing to activate the phosphorus ions, the thermal annealing temperature is 600°C to 1000°C, and the thermal annealing time is controlled to be 30 minutes to 60 minutes; In the step 2, a lightly doped N-type semiconductor is epitaxially grown. First, the silicon substrate for epitaxial growth is cleaned, including pickling and ultrasonic cleaning. Then, the cleaned silicon substrate is placed in an epitaxial growth chamber for subsequent operations. The epitaxial growth chamber is a high vacuum or inert gas environment. The growth temperature is set at 1100° C. to 1300° C., and the pressure is a low-pressure environment of 90 Pa to 100 Pa. The flow rate of the silicon source gas is set at 15 sccm to 20 sccm, and the flow rate of hydrogen as a carrier gas is set at 180 sccm to 200 sccm. During the growth process, phosphine (PH 3 ) as dopant gas, with a doping concentration of 1×10 15 cm -3 ~1×10 17 cm -3 The growth time is set at 30 minutes to 120 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 700°C to 900°C, and the thermal annealing time is controlled at 20 minutes to 50 minutes. In the step 3, a heavily doped P-type semiconductor is epitaxially grown. First, the silicon substrate for epitaxial growth is cleaned, including pickling and ultrasonic cleaning. Then, the cleaned silicon substrate is placed in an epitaxial growth chamber for subsequent operations. The epitaxial growth chamber is a high vacuum or inert gas environment. The growth temperature is set at 1100° C. to 1300° C., and the pressure is a low-pressure environment of 90 Pa to 100 Pa. The flow rate of the silicon source gas is set at 15 sccm to 20 sccm, and the flow rate of hydrogen as a carrier gas is set at 180 sccm to 200 sccm. During the growth process, dimethyl borane (DMBH) is introduced as a dopant gas with a doping concentration of 1×10 19 cm -3 ~1×10 20 cm -3 The growth time is set at 30 minutes to 120 minutes. After the growth is completed, the epitaxial layer is thermally annealed to optimize the crystal structure and electrical properties. The thermal annealing temperature is 800°C to 1100°C, and the thermal annealing time is controlled at 30 minutes to 50 minutes. The etching in step 4 is performed by dry etching, using a reactive ion etching (RIE) device for etching, selecting hydrogen fluoride (HF) as the etching gas, setting its flow rate to 20 sccm to 25 sccm, setting the temperature in the etching chamber to 20° C. to 30° C., setting the working pressure of the etching chamber to 60 mTorr to 80 mTorr to reduce unexpected side reactions, setting the RF power to 300 W, the frequency to 13.56 MHz, and the etching time to 10 to 40 minutes, using spectroscopy endpoint detection to determine when the etching process reaches the target depth, and after etching, the residues need to be thoroughly removed to avoid potential impact on device performance; The dielectric material deposited in step 5 adopts an atomic layer deposition process (ALD). The atomic layer deposition process (ALD) first performs chemical cleaning or hydrogen treatment on the semiconductor material to remove oxides and organic residues to ensure optimal surface acceptance. Trimethoxyhafnium or trimethoxyaluminum is used as a precursor for depositing a dielectric layer. The precursor is introduced into a reaction chamber to adsorb the precursor on the silicon surface. Oxygen or nitrogen is introduced to remove unreacted precursors and by-products. Water vapor is introduced as an oxidant to react with the adsorbed precursor to generate a dielectric layer film. The purification step is repeated to remove by-products and unreacted gases generated during the reaction. During this process, the temperature is controlled at 200° C. to 400° C. and adjusted according to the characteristics of the ALD precursor and the reaction conditions. The ambient pressure is maintained at 2 to 10 Torr. The thickness of the dielectric layer film is increased by repeating the ALD cycle, and the thickness increases by 5 to 20 angstroms per cycle. In step 6, the U-shaped nanocavity and the gate dielectric layer in the middle thereof are dry-etched, and an inductively coupled plasma (ICP) etcher and hydrogen fluoride (HF) are selected as etching gases for operation. The overall ambient temperature needs to be maintained at 20° C. to 25° C., the inductively coupled plasma (ICP) etcher is started to generate plasma, and then HF etching gas is input to ensure that the flow rate is 40 sccm to 50 sccm, and the pressure of the reaction chamber is adjusted to 40 mTorr to 50 mTorr to maintain a stable etching atmosphere. At a discharge power of 200 W to 300 W and a frequency of 13.56 MHz, HF reacts with the surface of the gate dielectric layer to generate fluoride, thereby achieving etching. After the etching is completed, cleaning is performed to remove the etching residues; In step 7, the horizontal metal grid (6) and the vertical metal grid (7) are deposited on the cavity by a reactive sputtering process. The reactive sputtering process uses a magnetron sputtering system for deposition. First, the target material is heated to 300° C. to 600° C. to increase the evaporation rate of the metal. At the same time, the temperature of the substrate is controlled at 100° C. to 300° C. to avoid thermal damage to the film on the substrate. The pressure of the sputtering chamber is set at 10 -6 ~10 -8 Torr, start the sputtering system, heat the target materials of the horizontal metal gate (6) and the vertical metal gate (7) and apply an electric field to sputter the target materials from the surface of the target gun onto the silicon substrate, set the sputtering power of the material of the horizontal metal gate (6) to 50-200W, and the sputtering power of the material of the vertical metal gate (7) to 30-150W, so as to form a desired metal film on the semiconductor surface, after the sputtering process is completed, release the vacuum chamber gas and remove the semiconductor substrate, and use characterization technology to evaluate the morphology, composition and bonding state of the metal film, the characterization technology includes SEM and XPS, to ensure the quality and stability of the gate electrode; Chemical vapor deposition (CVD) is used to form the source contact (9) and the drain contact (8). First, the CVD reaction chamber is preheated to 500-700° C. and the pressure in the chamber is controlled at 10 2 ~10 3 Torr, in order to facilitate the subsequent deposition, dimethylsilane (SiH 2 (CH 3 ) 2 ) as the silicon source gas, with a flow rate of 50 sccm to 100 sccm, and hydrogen (H 2 ) is used as a diluent gas with a flow rate set at 200 sccm to 500 sccm. The silicon source gas is decomposed in the reaction chamber and an electrode film is deposited on the surface of the substrate. The deposited electrode film is then annealed to improve its crystallinity and electrical properties. The thermal annealing temperature is between 500°C and 1000°C, and the thermal annealing time is controlled at 20 minutes to 40 minutes.
10. The method for preparing a biosensor having an inverted E-type hetero-gate vertical tunneling field effect transistor according to claim 8, It is characterized in that The heavily doped N-type semiconductor in step 1 is silicon, and the doping concentration is 1×10 18 cm -3 ~1×10 19 cm -3 , length is 80-110nm, thickness is 10-20nm; The lightly doped N-type semiconductor in step 2 is silicon, and the doping concentration is 1×10 15 cm -3 ~1×10 17 cm -3 , length is 80-110nm, thickness is 10-20nm; The heavily doped P-type semiconductor in step 3 is silicon, and the doping concentration is 1×10 19 cm -3 ~1×10 20 cm -3 , length is 80-110nm, thickness is 20-30nm; Each independent source region (3) in step 4 has the same size and a doping concentration of 1×10 19 cm -3 ~1×10 20 cm -3 , with a length of 10 to 15 nm and a thickness of 20 to 30 nm. The two deposited layers have the same size, and the length of the two deposited layers plus the length of the three source regions is consistent with the length of the channel (2), and the thickness is 20 to 30 nm; The dielectric material in step 5 is hafnium oxide (HfO 2 ) or aluminum oxide (Al 2 O 3 ); The gate dielectric layer (4) in step 6 has a length of 3 to 5 nm and a thickness of 3 to 5 nm, the thickness of the nanocavity (5) in the horizontal direction is consistent with the thickness of the gate dielectric layer (4), the length of the nanocavity (5) in the vertical direction is consistent with the thickness in the horizontal direction, the dimensions of the L-shaped nanocavities (5) are the same, and two adjacent L-shaped structures are arranged in a mirror image between the three source regions, the bottom and top of the L-shaped structure are both flush with the bottom and top of the source region (3), and the sum of the bottom length of the three source regions (3), the bottom length of the four L-shaped structures and the bottom length of the two gate dielectric layers (4) is consistent with the length of the channel (2); The material of the horizontal metal gate (6) in step 7 is metal hafnium, metal chromium or metal titanium. The length of the horizontal metal gate (6) is equal to the sum of the lengths of the nanocavities (5) on both sides and the sum of the bottom length of the L-shaped structure and the gate dielectric layer (4), and the thickness is 2 to 3 nm. The material of the vertical metal gate (7) is metal aluminum, metal silver or metal platinum, and the length is 2 to 3 nm. The top of the vertical metal gate (7) is flush with the top of the source region (3), and the bottom is flush with the top of the horizontal metal gate (6). The materials of the drain contact (8) and the source contact (9) are both polysilicon. The length of the drain contact (8) is equal to the length of the drain region (1), and the thickness is 2 to 3 nm. The length of the source contact (9) is equal to the length of the source region (3), and the thickness is 2 to 3 nm.
