Vertical carbon nanotube forest microradiometer, array and its fabrication method
By setting a vertical carbon nanotube forest in the central local thermal coupling region of the suspended membrane and retaining an annular discontinuity region on the periphery, combined with a lateral connecting layer and a root electrode, the problems of infrared absorption and thermal isolation in the prior art are solved, achieving a balance between high absorption rate and low heat capacity, which is suitable for uncooled infrared detection.
Patent Information
- Application Number
- CN202611247280.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-17
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies struggle to achieve high infrared absorption, low heat capacity, edge thermal isolation, and temperature conversion within the same pixel, and lack feasible processes that prevent the readout circuit from undergoing high-temperature growth.
A vertical carbon nanotube forest microradiometric calorimeter is designed, in which a vertical carbon nanotube forest is set only in the central local thermal coupling zone of a suspended membrane, and an annular discontinuity zone is formed between the periphery and the edge of the suspended membrane. A transverse connecting layer and root electrodes are set in the root region. It is prepared by local low-temperature in-situ growth or pre-fabricated forest root transfer method to avoid heat leakage.
It achieves a balance between high infrared absorptivity and low heat capacity with the same pixel projection area and effective thermistor area, reduces equivalent thermal conductivity, and has good process feasibility, making it suitable for uncooled infrared detection.
Smart Images

Figure CN123124583A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of uncooled infrared detection technology, and in particular to a vertical carbon nanotube forest microradiometer, array, and its fabrication method. Background Technology
[0002] Microbolometers typically utilize an infrared absorbing layer to absorb target radiation and generate a temperature rise, which is then converted into an electrical signal via a thermistor, thermocouple, diode, or transistor. To achieve high responsivity, pixels usually require high infrared absorptivity, low heat capacity, and low thermal conductivity of the suspended structure.
[0003] Common infrared absorbing layers include metal-dielectric composite films, porous metallic black, carbon black, planar carbon nanotube networks, or resonant cavity absorption structures. Planar films are easily limited by thickness and optical constants; while increasing thickness can improve absorption, it usually increases heat capacity simultaneously. Some porous absorbing layers also suffer from poor mechanical stability, process contamination, or incompatibility with microfabrication processes. Vertical carbon nanotube absorbers can be placed in suspended films or pits, but if the forest cover extends to the edge of the film, heat may still dissipate to the supporting beam through continuous carbon materials or root bonding layers; and when using only the forest as an absorber, it is difficult to utilize the temperature-sensitive resistance of the root tube bundle contact junctions.
[0004] Therefore, it is evident that the existing technologies mentioned above cannot simultaneously achieve high infrared absorption, low heat capacity, edge thermal isolation, and temperature conversion in the same pixel, and lack feasible processes that do not subject the readout circuit to high-temperature growth. It is necessary to propose new solutions. Summary of the Invention
[0005] The present invention aims to provide a vertical carbon nanotube forest microradiometric calorimeter, array and its preparation method to overcome the shortcomings of the prior art. The technical problem to be solved by the present invention is achieved through the following technical solution.
[0006] According to a first aspect of this application, a vertical carbon nanotube forest microbolometer is provided, comprising a substrate, a suspended membrane spaced apart from the substrate, a thermally insulating support beam connecting the suspended membrane and the substrate, and a vertical carbon nanotube forest disposed on the suspended membrane, wherein: The vertical carbon nanotube forest is only located in the central local thermal coupling area of the suspended membrane, and the projected area of the central local thermal coupling area accounts for 30% to 85% of the area of the suspended membrane. The periphery of the vertical carbon nanotube forest and the edge of the suspended membrane form a gap of 0.5 μm to 10 μm. At least two spaced-apart root electrodes and a transverse connecting layer connecting different vertical carbon nanotube bundles are provided in the root region not exceeding 20% of the height of the vertical carbon nanotube forest, so that the incident infrared radiation is absorbed by the vertical part of the vertical carbon nanotube forest, and the temperature rise is converted into an electrical signal through the equivalent resistance change of the contact junction of the transverse connecting layer and the carbon nanotube bundle.
[0007] Preferably, the average axial angle of the carbon nanotube bundles in the vertical carbon nanotube forest relative to the normal direction of the suspended membrane is no greater than 45°, and the areal density is no less than 10. 9 The forest height is 0.5 μm to 100 μm per square centimeter, and the interconnected porosity from the top of the forest to the root region is not less than 30%; the lateral connecting layer includes a carbon nanotube network layer, a conductive polymer layer, a graphene layer, or a discrete conductive nanoparticle connecting layer with a thickness of 10 nm to 500 nm.
[0008] Preferably, the vertical carbon nanotube forest is provided with at least three regions along the plane of the suspended membrane, with heights of 3μm to 8μm, 8μm to 18μm, and 18μm to 35μm, and a reflective layer is provided below the suspended membrane; the height distribution of the forest and the cavity depth from the reflective layer to the suspended membrane are configured such that the absorption peak and valley values of the two wavebands, 3μm to 5μm and 8μm to 14μm, are both not less than 85%.
[0009] Preferably, the top of the vertical carbon nanotube forest is provided with a porous limiting layer supported by several support columns. The porous limiting layer maintains a non-contact gap with the top of the forest, with an opening rate of 20% to 95% and a thickness of less than one-tenth of the forest height.
[0010] Preferably, the surface of the carbon nanotube bundle is provided with a conformal functional layer of alumina, boron nitride, fluoropolymer or discrete metal nanoparticles with a thickness of 1 nm to 20 nm; the absolute value of the temperature coefficient of resistance of the lateral resistance path in the root region is 0.5% per Kelvin to 3% per Kelvin, and the root electrode adopts a four-terminal or Kelvin connection.
[0011] Preferably, under the same pixel projection area and the same effective thermal sensitive area, the average absorption rate of the vertical carbon nanotube forest in the 8μm to 14μm band is not less than 95%, and its heat capacity per unit area is not more than 1.2 times that of the heat capacity per unit area of a 200nm thick planar carbon nanotube network.
[0012] According to a second aspect of this application, an infrared focal plane array is provided, comprising a plurality of the aforementioned vertical carbon nanotube forest microbolometers arranged in rows and columns, a readout circuit for pixel-by-pixel gating and biasing, and at least one reference pixel, wherein different pixels have different forest height distributions or microcavity depths to form a mid-wave infrared subarray and a long-wave infrared subarray.
[0013] Preferably, the forest of the reference pixel is infrared shielded, compacted, or thermally coupled to the substrate, and blank pixels without forest are set to correct for substrate temperature, forest transfer integrity, and root resistance drift, respectively.
[0014] According to a third aspect of this application, a method for preparing the above-mentioned vertical carbon nanotube forest microbolometer is provided, comprising: A root electrode, a lateral connecting layer, and a catalytic region are formed in the central local thermal coupling region of the suspended membrane; A vertical carbon nanotube forest with roots facing the suspended membrane is formed, and an annular discontinuous region is retained that does not extend to the thermally insulating support beam; wherein the vertical carbon nanotube forest is formed by one of the following methods: local low-temperature in-situ growth with the overall temperature of the substrate not exceeding 400°C and the local effective temperature of the catalytic region not exceeding 550°C, or transfer bonding with the roots of the forest facing the suspended membrane after prefabricating the vertical carbon nanotube forest on an independent growth substrate and sequentially forming a transferable support layer and an adhesive layer.
[0015] Preferably, the local low-temperature in-situ growth employs plasma-enhanced chemical vapor deposition with a plasma power of 20W to 300W, a volume ratio of carbon-containing gas to diluent gas of 1:2 to 1:50, and controls the temperature difference between the catalytic region and the substrate to be 50°C to 250°C; in the transfer bonding process, the retention rate of the transferred forest island is not less than 90%, and the effective bonding area between the roots and the suspended membrane accounts for 70% to 95% of the forest projection area.
[0016] The embodiments of the present invention have the following advantages: The vertical carbon nanotube forest is only located in the central local thermal coupling zone of the suspended membrane, and an annular discontinuity zone is retained on the periphery that does not extend to the thermal isolation support beam. The forest, the lateral connecting layer and the adhesive material do not extend directly to the support beam, thereby limiting the lateral heat leakage of the forest to the membrane edge and the support beam and reducing the equivalent thermal conductivity of the pixel.
[0017] The vertical portion of the same vertical carbon nanotube forest mainly undertakes infrared absorption, while the root lateral connecting layer and carbon nanotube bundle contact junction mainly undertake resistance temperature conversion, so that absorption and temperature conversion are achieved in the same structure in separate zones. Under the same pixel projection area and the same effective thermistor area, the exemplary design makes the average absorption rate in the 8μm to 14μm band not less than 95%, while the heat capacity per unit area is not more than 1.2 times that of the heat capacity per unit area of a 200nm thick planar carbon nanotube network.
[0018] By combining gradient forest regions with heights of 3μm to 8μm, 8μm to 18μm, and 18μm to 35μm with reflective microcavities, the absorption of the forest volume and the interference absorption of the microcavities are superimposed, so that the absorption peak and valley values of the two bands of 3μm to 5μm and 8μm to 14μm are not less than 85%, forming a dual-band pixel with structural synergistic constraint.
[0019] A porous limiting layer, supported by several pillars and with a non-contact gap between it and the top of the forest, restricts large-scale lodging of the forest while maintaining infrared incident channels and interconnected pores, unlike protective structures that directly cover or compact the forest.
[0020] The local low-temperature in-situ growth route limits the thermal exposure of the CMOS back end through a dual window of overall substrate temperature and local catalytic zone temperature; the prefabricated forest root transfer route allows high-temperature growth to be completed entirely on an independent growth substrate, and the suspended film and readout circuit do not experience the growth temperature of the carbon nanotube forest. Both routes have good process feasibility. Attached Figure Description
[0021] Figure 1 This is a schematic cross-sectional view of a vertical carbon nanotube forest suspended microbolometer according to an embodiment of the present invention. Figure 2 This is a top view of a pixel structure according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a gradient vertical carbon nanotube forest according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the root transverse thermally sensitive passage and the overhead porous limiting structure according to an embodiment of the present invention. Detailed Implementation
[0022] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.
[0023] It should be noted that the above detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.
[0024] For ease of understanding, some terms used in this article are defined first. The "vertical carbon nanotube forest 1" refers to a carbon nanotube bundle whose average axial angle relative to the normal direction of the suspended membrane 5 is no greater than 45° and whose areal density is no less than 10. 9The forest consists of an aggregate of carbon nanotubes per square centimeter, with interconnected pores from the top surface to the root. These interconnected pores are defined as pores with a porosity of at least 30% and a top-to-bottom through-path. The root region refers to a section extending upwards from the bottom of the forest, not exceeding 20% of the forest height, within which the lateral connecting layer and root electrode 6 are located. The central local thermal coupling zone refers to the central area where the forest roots are in direct thermal contact with the suspended membrane 5 or the root resistive structure, with a projected area accounting for 30% to 85% of the area of the suspended membrane 5. The annular discontinuity zone 10 refers to a continuous area without forest, with low density or significantly reduced height, extending from the forest perimeter to the membrane edge, and not extending to the thermally insulating support beam 7. The central local thermal coupling zone and the annular discontinuity zone 10 together restrict lateral heat leakage from the forest to the membrane edge and support beam, while the root lateral connecting layer converts the temperature rise into a measurable change in resistance. The above limitations are common in all embodiments of the present invention and will not be repeated below.
[0025] like Figure 1 As shown, the vertical carbon nanotube forest microbolometer of this embodiment includes a substrate 8, a suspended membrane 5 spaced apart from the substrate 8, a thermally insulating support beam 7 connecting the suspended membrane 5 and the substrate 8, and a vertical carbon nanotube forest 1 disposed on the suspended membrane 5. The substrate 8 can be a silicon-based CMOS readout chip, an insulating substrate, or an intermediate substrate with an interconnect layer. The suspended membrane 5 is disposed above the substrate 8 via the thermally insulating support beam 7, and a vacuum or low-pressure cavity is formed below the suspended membrane 5 to reduce the thermal conductivity of the suspended structure through the gas. The suspended membrane 5 can be formed of silicon nitride, silicon dioxide, alumina, polyimide, or a stack thereof, with a thickness of 50 nm to 2 μm; in applications requiring low stress and high release yield, a low-stress silicon nitride film with a thickness of 200 nm to 800 nm is preferred. The number of thermally insulating support beams 7 can be one, two, four, or more, and their length, width, and thickness are configured to provide the lowest possible thermal conductivity while ensuring mechanical stability.
[0026] The vertical carbon nanotube forest 1 is only located in the central local thermal coupling zone of the suspended membrane 5, forming an annular discontinuity zone 10 with a width of 0.5 μm to 10 μm between the periphery of the forest and the edge of the suspended membrane 5. The projected area of the central local thermal coupling zone accounts for 30% to 85% of the area of the suspended membrane 5; when the proportion is close to 30%, the forest is concentrated in the center of the membrane, the annular discontinuity zone 10 is wider, and the edge thermal insulation is the strongest, which is suitable for long-wavelength pixels that are sensitive to thermal conductivity; when the proportion is close to 85%, the effective thermally sensitive area is larger, and the responsivity is higher, which is suitable for applications that are sensitive to fill factor. The annular discontinuity zone 10 prevents the vertical carbon nanotube forest 1, the lateral connecting layer, and the adhesive material from directly extending to the thermally insulating support beam 7, thereby cutting off the lateral channel for heat dissipation to the support beam through the continuous carbon material or the root adhesive layer, and reducing the equivalent thermal conductivity of the pixel.
[0027] Vertical carbon nanotube forest 1 comprises single-walled, double-walled, or multi-walled carbon nanotube bundles with an average outer diameter of 1 nm to 100 nm, a forest height of 0.5 μm to 100 μm, and an areal density of not less than 10. 9 The interconnected porosity per square centimeter of the roots is no less than 30%. The average axial angle of the carbon nanotube bundles relative to the normal direction of the suspended membrane 5 is no greater than 45° to ensure that the forest has sufficient optical path length in the vertical direction for infrared absorption, while retaining interconnected pores at the roots for the release medium to enter and exit. Incident infrared radiation is mainly absorbed by the vertical portion of the vertical carbon nanotube forest 1. The vertical orientation of the forest and the interconnected pores create multiple scattering and optical path extension, thereby achieving a high absorption rate with a low heat capacity per unit area.
[0028] Within a root region not exceeding 20% of the height of the vertical carbon nanotube forest 1, at least two spaced-apart root electrodes 6 and a lateral connecting layer connecting different vertical carbon nanotube bundles are disposed. The lateral connecting layer comprises a carbon nanotube network layer, a conductive polymer layer, a graphene layer, or a discrete conductive nanoparticle connecting layer with a thickness of 10 nm to 500 nm, used to connect the roots of different vertical tube bundles, forming a lateral resistance path. The current path between the two root electrodes 6 is mainly located in the root lateral connecting layer and the carbon nanotube bundle contact junction, with an absolute value of the absolute value of its equivalent resistance temperature coefficient of 0.5% to 3% per Kelvin. During operation, incident infrared radiation is absorbed by the vertical portion of the vertical carbon nanotube forest 1, causing a temperature rise. This temperature rise is conducted through the root region and changes the equivalent resistance of the lateral connecting layer and the carbon nanotube bundle contact junction, which is read out as an electrical signal through the root electrodes 6. Thus, the same vertical carbon nanotube forest 1 undertakes infrared absorption in the vertical direction and resistance-temperature conversion in the lateral direction at the root, enabling absorption and temperature conversion to be achieved in the same partition within the same structure, eliminating the need for a separate planar thermistor layer and further reducing the overall heat capacity of the pixel. The root electrode 6 uses a four-terminal or Kelvin connection to reduce the influence of electrode contact resistance on temperature readout; no continuous metal cover is formed between the root electrode 6 and the top of the forest to avoid increasing heat capacity and reflection.
[0029] In another structural variant, the temperature-sensitive conversion unit 4 is disposed on the suspended film 5 and can be a serpentine carbon nanotube network thermistor, vanadium oxide, amorphous silicon, platinum resistance thermometer, thermocouple, or transistor. Electrode 6 extends along the thermally insulated support beam 7 to the substrate 8. An electrically insulating thermal coupling layer with a thickness of 1 nm to 500 nm is provided between the temperature-sensitive conversion unit 4 and the vertical carbon nanotube forest 1 to suppress bias current from entering the forest while achieving thermal coupling. In this variant, the absorption function is mainly undertaken by the vertical carbon nanotube forest 1, while the temperature conversion function is undertaken by the temperature-sensitive conversion unit 4. The two are thermally coupled but electrically isolated, making it suitable for applications where mature thermistor material processes are preferred.
[0030] like Figure 2 and Figure 4 As shown, to suppress large-scale lodging of forests during release, drying, and use, a porous confinement layer 2, supported by several pillars, can be installed at the top of the vertical carbon nanotube forest 1. The porous confinement layer 2 maintains a non-contact gap of 0.1 μm to 5 μm between itself and the top of the forest, has a porosity of 20% to 95%, and a thickness less than one-tenth of the forest height. The support pillars are located at the forest periphery or in low-density areas to avoid forming high thermal conductivity paths that penetrate the forest. The porous confinement layer 2 restricts large-scale lodging of the forest while maintaining infrared incident channels and interconnected pores, unlike protective structures that directly cover or compact the forest, and therefore does not significantly increase the heat capacity of pixels or reduce absorption rates.
[0031] To reduce resistivity drift caused by environmental adsorption, a conformal functional layer of alumina, boron nitride, fluoropolymers, or discrete metal nanoparticles with a thickness of 1 nm to 20 nm can be formed on the surface of the carbon nanotube bundles, while ensuring that the interconnected pores are not completely filled. The thickness of the conformal functional layer is lower than the average spacing between adjacent bundles to avoid transforming the forest into a dense film. By controlling the material and thickness of the conformal functional layer, selective absorption, environmental stability, or temperature response can be further adjusted while passivating environmental adsorption. Under the same pixel projection area and the same effective thermosensitive area, the vertical carbon nanotube forest 1 of this embodiment has an average absorption rate of not less than 95% in the 8 μm to 14 μm band, and its heat capacity per unit area is not more than 1.2 times that of a 200 nm thick planar carbon nanotube network.
[0032] In structural variants that require band selection, such as Figure 3 As shown, a reflective layer 9 can be disposed below the suspended membrane 5, and the vertical carbon nanotube forest 1 can be configured along the plane of the suspended membrane 5 to form at least three regions: a low-height, high-density region 11 with a height of 3μm to 8μm, a medium-height, medium-density region 12 with a height of 8μm to 18μm, and a high-height, low-density region 13 with a height of 18μm to 35μm. The forest height distribution and the cavity depth from the reflective layer 9 to the suspended membrane 5 are configured such that the absorption peak and valley values of the two wavelength bands, 3μm to 5μm and 8μm to 14μm, are both not less than 85%. The structural synergy will be further explained in conjunction with the microcavity depth in the array structure embodiment below.
[0033] like Figure 3As shown, a gradient vertical carbon nanotube forest 1 with varying forest height along the plane of the suspended membrane 5 can be set within a single pixel, including a low-height, high-density region 11 with a height of 3 μm to 8 μm, a medium-height, medium-density region 12 with a height of 8 μm to 18 μm, and a high-height, low-density region 13 with a height of 18 μm to 35 μm. This density gradient can be obtained by partitioning the catalyst thickness, local growth time, plasma etching, or mechanical shearing. A reflective layer 9 is formed below the suspended membrane 5, and the microcavity depth is defined by the thickness of the sacrificial layer; in the example design, the low, medium, and high forest regions can be combined with equivalent optical cavity depths of 0.8 μm to 1.5 μm, 1.5 μm to 2.5 μm, and 2.5 μm to 4 μm, respectively. By combining forest regions of different heights with different equivalent optical cavity depths, the forest volume absorption and microcavity interference absorption superimpose to form a double peak or broadened plateau. This ensures that the absorption peak-valley values in both the 3μm-5μm and 8μm-14μm bands are no less than 85%, thus constructing a dual-band pixel with structurally co-defined characteristics. For arrays requiring spectral identification, different pixels can also employ different forest heights, conformal functional layers, or microcavity depths to form short-wave, mid-wave, and long-wave response subarrays.
[0034] The vertical carbon nanotube forest microbolometers described in several of the aforementioned structural embodiments are arranged in rows and columns to form an infrared focal plane array, and are equipped with readout circuits for pixel-by-pixel gating and biasing. The readout circuits can employ constant voltage bias, constant current bias, pulse bias, or chopper bias; to reduce bias self-heating of the forest and the thermosensitive layer, the bias can be turned off outside the integration window. At least one reference pixel is placed every few rows or columns in the array. The forest of the reference pixel is infrared-shielded by a high-reflectivity metal shielding layer, compacted, or thermally coupled to the substrate 8 through a wide support beam, and is used to correct substrate temperature, forest transfer integrity, and root resistance drift, respectively. Blank pixels without forest can also be included in the array; by comparing the outputs of blank pixels without forest, standard-height forest pixels, and gradient forest pixels, substrate temperature, forest transfer integrity, absorber layer heat capacity, and root resistance drift can be evaluated and corrected, respectively.
[0035] In this array embodiment, different pixels have different forest height distributions or microcavity depths, thereby forming mid-wave infrared subarrays and long-wave infrared subarrays. Accordingly, the same focal plane array can simultaneously or time-divisionally image in both mid-wave and long-wave infrared bands, and on-chip temperature compensation and non-uniformity correction can be achieved using reference pixels and blank pixels, improving the array's consistency across different operating temperatures and device batches. The forest of the reference pixel is infrared-shielded, compacted, or thermally coupled to the substrate 8, and blank pixels without forests are provided to correct substrate temperature, forest transfer integrity, and root resistance drift, thereby suppressing fixed-pattern noise and temperature drift of the array without the need for an external temperature homogenizer.
[0036] The method for preparing the vertical carbon nanotube forest microradiometric calorimeter of the present invention includes two basic steps: forming a root electrode 6, a lateral connecting layer and a catalytic region in the central local thermal coupling region of the suspended membrane 5, and forming a vertical carbon nanotube forest 1 with roots facing the suspended membrane 5 and retaining an annular discontinuous region 10 that does not extend to the thermally insulating support beam 7.
[0037] Among them, the vertical carbon nanotube forest 1 can be formed through one of two routes: local low-temperature in-situ growth and pre-fabricated forest root transfer. Both routes aim to prevent the readout circuit from experiencing the high growth temperature of the carbon nanotube forest 1, and have good process feasibility. The following is combined with Figure 4 Each will be explained separately.
[0038] One preparation method: local low-temperature in-situ growth route. For example... Figure 4 As shown, before the suspended film 5 is released, a root electrode 6, a lateral connecting layer, and a catalytic region are formed in the central local thermal coupling area. Iron, cobalt, nickel, or their alloy catalyst nanoparticles are formed within the catalytic region. Subsequently, a vertical carbon nanotube forest 1 is grown using plasma-enhanced chemical vapor deposition (PECVD), keeping the overall temperature of the substrate 8 below 400°C and the local effective temperature of the catalytic region below 550°C. The plasma power is 20W to 300W, the volume ratio of carbon-containing gas to diluent gas is 1:2 to 1:50, and the forest growth rate is 0.05μm / min to 5μm / min. When establishing a local temperature field using a local microheater, pulsed laser, induction heating, or remote plasma, the temperature difference between the catalytic region and the substrate 8 is controlled to be 50°C to 250°C, ensuring the catalytic region reaches the local effective temperature required for carbon nanotube growth, while the substrate 8 and its readout circuits remain at a lower temperature, thus limiting the thermal exposure temperature and duration experienced by the CMOS backend. After growth is complete, an annular discontinuity region 10 is retained, not extending to the thermally isolated support beam 7. The local low-temperature in-situ growth limits the thermal exposure of the CMOS back end through a dual window of the overall temperature of the substrate 8 and the local temperature of the catalytic region. The overall temperature of the substrate 8 is no higher than 400°C and the effective local temperature of the catalytic region is no higher than 550°C, which together constitute the temperature constraint window.
[0039] In localized low-temperature in-situ growth routes, gaseous hydrogen fluoride, oxygen plasma, or other release methods with no liquid surface tension are preferred when releasing the sacrificial layer to avoid forest aggregation and lodging caused by liquid surface tension. When wet release is necessary, the forest pores are temporarily filled with a removable polymer, and the filler is removed in a low-damage manner after supercritical drying or freeze-drying. Before releasing the suspended membrane 5, an elevated porous confinement layer 2 can be formed by support columns to further reduce forest aggregation and lodging during release and drying.
[0040] Method Two: Pre-fabricated Forest Root Transfer Route. A vertical carbon nanotube forest 1 is grown at high temperature on a quartz, silicon, metal foil, or ceramic growth substrate and patterned as pixel islands. A transferable support layer and an ultrathin adhesive layer 3 are sequentially formed at the forest roots. The transferable support layer maintains the relative position of the root bundles during transfer. The growth substrate is flipped so that the forest roots face the suspended membrane 5, which has already formed root electrodes 6 and lateral connecting layers. The root region is fixed by hot pressing, local laser, or chemical bonding. Then, the growth substrate and transferable support layer are removed by release layer decomposition, selective etching, or mechanical peeling, ensuring that the suspended membrane 5 and readout circuit do not experience the growth temperature of the carbon nanotube forest 1. After transfer, the retention rate of the forest islands is not less than 90%, the effective bonding area between the roots and the suspended membrane 5 accounts for 70% to 95% of the forest's projected area, and a ring-shaped discontinuous region 10 with a width of 0.5 μm to 10 μm is formed around the forest perimeter. This transfer can be completed before the release of the suspended film 5, so that the suspended film 5 is mechanically supported by the sacrificial layer; alternatively, a local reinforcing frame can be formed first, followed by transfer after release. For arrays, multiple forest islands can be transferred at once using a single flexible stamp, and the height compensation structure on the stamp surface can accommodate wafer warpage. Since the high-temperature growth is performed entirely on a separate growth substrate, this process route is particularly suitable for integration with already interconnected CMOS readout chips, avoiding degradation of the readout circuitry due to the high growth temperature of the carbon nanotube forest 1.
[0041] The formation of the root electrode 6 and the lateral connection layer can be achieved through the following steps: two or four root electrodes 6 are pre-formed in the central region of the suspended membrane 5, with an electrode spacing of 2 μm to 30 μm; after the vertical carbon nanotube forest 1 is transferred or grown, conductive nanoparticles are impregnated in the root region from the bottom of the forest upwards, not exceeding 20% of the forest height, or a lateral carbon nanotube network layer with a thickness of 10 nm to 500 nm is formed to connect the roots of different vertical tube bundles, thus constituting the lateral connection layer. By controlling the material and thickness of the lateral connection layer, the resistance value and temperature coefficient of resistance of the root equivalent resistance can be adjusted to match the dynamic range of the readout circuit. In addition, a conformal functional layer of alumina, boron nitride, or fluoropolymer with a thickness of 1 nm to 20 nm can be formed on the surface of the carbon nanotube bundles, while keeping the interconnecting pores from being completely filled to reduce drift caused by environmental adsorption.
[0042] Specific application examples are as follows: Application Example 1: Wafer-level Packaging and Bias Readout. After releasing the suspended structure, the aforementioned infrared focal plane array is wafer-level packaged in a vacuum or low-pressure inert gas environment. An anti-reflective film is provided on the infrared window of the package cover. A gap is maintained between the porous limiting layer 2 and the package cover to prevent the cover from directly pressing the vertical carbon nanotube forest 1. The readout circuit applies a bias to the root electrode 6 of each pixel and reads its equivalent resistance change. Constant voltage bias, constant current bias, pulse bias, or chopper bias can be used. To reduce the bias self-heating of the forest and the thermally sensitive layer, the bias is turned off outside the integration window, and substrate temperature compensation is performed using a reference pixel or an on-chip temperature sensor.
[0043] Application Example 2: Multi-Temperature Calibration and Non-Uniformity Correction. During the calibration phase, at least two known blackbody temperatures are applied to the array to establish a correspondence between pixel output, reference pixel output, and substrate temperature. For subarrays with different forest heights or microcavity depths, gain and non-uniformity correction parameters are stored separately. By comparing the outputs of blank pixels without forests, standard-height forest pixels, and gradient forest pixels at the wafer level or after packaging, forest transfer integrity, absorber layer thermal capacity, and microcavity deviation can be evaluated, and pixel calibration parameters can be generated, further improving the consistency and stability of the array output.
[0044] Application Example 3: Dual-band Imaging. Utilizing the mid-wave infrared sub-array and long-wave infrared sub-array formed in the aforementioned array structure embodiments, the same focal plane array can simultaneously or sequentially image in two bands: 3μm to 5μm and 8μm to 14μm. Since the absorption peak and valley values of both bands are not less than 85%, dual-band radiation information can be obtained without replacing the detector, which can be used for target identification, temperature inversion, and camouflage reveal.
[0045] Application Example 4: Non-Imaging Detection Extension. This invention is not limited to imaging microbolometers; it can also be used in infrared switches, heat flow sensors, infrared spectral detection units, gas absorption measurements, and neuromorphic devices with infrared stimulus responses. The forest can be arranged in concentric rings, strips, lattices, or fractal patterns, as long as an annular discontinuity 10 is maintained at the forest perimeter without extending to the thermally insulated support beam 7. In gas absorption measurement applications, the conformal functional layer can be used to adjust the selective absorption of specific wavelengths to improve the response to the characteristic absorption peaks of the target gas.
[0046] In other alternative embodiments, the vertical carbon nanotube forest 1 can be partially replaced by a mixed forest of vertical graphene, carbon nanofibers, or carbon nanotubes with other one-dimensional nanomaterials; the forest can also be arranged in the upper layer of the double-layer suspended membrane, with the temperature-sensitive conversion unit 4 arranged in the lower layer, and the two connected by low-heat-capacity support pillars. The suspended membrane 5 can adopt a circular, rectangular, annular, or multi-island structure, and the number of thermally insulating support beams 7 can be one, two, four, or more. Those skilled in the art can make substitutions or combinations of materials, structures, dimensions, and process sequences without departing from the concept of the present invention, and all such substitutions or combinations should fall within the protection scope defined by the claims.
[0047] To verify the synergistic effect of the central local thermal coupling region and the annular discontinuity region 10 on absorptivity, heat capacity, and edge thermal insulation, three sets of pixels with the same projected area, identical suspended membrane 5 and supporting beam structure, and different only absorption layer configurations were prepared for comparison. Figure 1 , Figure 2 and Figure 4 The pixel of the present invention shown is compared with two sets of comparative pixels.
[0048] Comparative Example 1 is a planar carbon nanotube network pixel with a thickness of 200 nm. This pixel uses a planar carbon nanotube network as the infrared absorption layer. Its heat capacity per unit area increases with increasing thickness, making it difficult to achieve a sufficiently high absorptivity while maintaining a low heat capacity. Therefore, the average absorptivity in the 8 μm to 14 μm wavelength range is significantly low. Its heat capacity per unit area is used as a benchmark and denoted as 1.0.
[0049] Comparative Example 2 shows a pixel with vertical forest coverage extending to the edge of the membrane. Although this pixel has an infrared absorption rate similar to that of the pixel of the present invention (not less than 95%), the vertical carbon nanotube forest 1 and its root bonding layer extend to the thermally insulating support beam 7 without retaining the annular discontinuity region 10, thus forming an additional lateral heat leakage channel, resulting in a higher equivalent thermal conductivity along the support beam direction.
[0050] The pixel of the present invention has a central local thermal coupling region and an annular discontinuity region 10. Its average absorptivity in the 8μm to 14μm band is not less than 95%, and its heat capacity per unit area is not more than 1.2 times that of Comparative Example 1. Compared with Comparative Example 2, which covers the edge of the membrane, the pixel of the present invention does not extend to the thermal isolation support beam 7 because the forest, the lateral connecting layer and the adhesive material do not extend to the thermal isolation support beam 7, and the equivalent thermal conductivity along the support beam direction is significantly reduced.
[0051] The suspended membrane 5, supporting beam 7, and readout method are the same for all three pixel groups, differing only in the absorption layer configuration. Therefore, the differences in absorptivity, heat capacity per unit area, and equivalent thermal conductivity can be attributed to the absorption layer configuration. Absorption spectra from 3 μm to 14 μm were obtained using Fourier transform infrared spectroscopy reflectance method, and the heat capacity per unit area was inversely calculated from the known thermal conductivity and thermal time constant. Table 1 presents a performance comparison of the three pixel groups, with comparison items including average absorptivity in the 8 μm to 14 μm band, heat capacity per unit area (relative value based on a 200 nm thick planar carbon nanotube network), and equivalent thermal conductivity along the supporting beam direction.
[0052] Table 1 Performance comparison of each embodiment and comparative example.
[0053] As shown in Table 1, although Comparative Example 1 uses the unit area heat capacity as a baseline, its average absorptivity in the 8μm to 14μm band is significantly lower because the planar absorption layer cannot simultaneously achieve both absorptivity and low heat capacity. Comparative Example 2 has an absorptivity similar to that of the present invention, but its equivalent thermal conductivity is higher because the forest extends to the supporting beam. The pixels of the present invention maintain a high absorptivity of no less than 95% and a unit area heat capacity no more than 1.2 times the baseline, while significantly reducing the equivalent thermal conductivity. This demonstrates that the synergistic limitation of the central local thermal coupling zone and the annular discontinuity zone 10 is key to the present invention's ability to balance high infrared absorption, low heat capacity, and edge thermal isolation within the same pixel.
[0054] As can be seen from the above embodiments, the present invention, by confining the vertical carbon nanotube forest 1 in the central local thermal coupling region of the suspended membrane 5, retaining an annular discontinuous region 10 on the periphery that does not extend to the thermally insulating support beam 7, and setting a lateral connecting layer and at least two root electrodes 6 in the root region, enables the same vertical carbon nanotube forest 1 to undertake infrared absorption in the vertical direction and resistance-temperature conversion in the lateral direction at the root, thereby achieving a balance between high infrared absorption, low heat capacity, and edge thermal isolation in the same pixel. Combined with the synergistic design of gradient forest and reflective microcavity, the suspended non-contact porous confinement layer 2, and two fabrication routes that do not subject the readout circuit to high-temperature growth, the present invention can provide uncooled infrared detection pixels and focal plane arrays with synergistic performance and well-defined structure while ensuring process feasibility. The technical features of the above embodiments can be combined with each other without conflict to form more implementation methods.
[0055] The following provides further details regarding various embodiments of the present invention to assist those skilled in the art in implementing the present invention.
[0056] Further explanation of the structural embodiments. Regarding the suspended membrane 5, its material and thickness require a trade-off between mechanical strength, thermal conductivity, and heat capacity: a thinner thickness is beneficial for reducing heat capacity and thermal conductivity, but too thin a thickness will reduce the release yield and resistance to deformation; a thicker thickness has the opposite effect. Therefore, in applications requiring low stress and high yield, a low-stress silicon nitride film with a thickness of 200 nm to 800 nm is preferred, and silicon dioxide or aluminum oxide can be used as a stress compensation layer. Regarding the vertical carbon nanotube forest 1, the forest height, areal density, and interconnected porosity are interrelated: a higher forest height results in a longer vertical optical path, which is more beneficial for improving infrared absorption, but excessive height increases the risk of collapse and reduces mechanical stability; excessively high areal density reduces interconnected porosity, hindering the entry and exit of the release medium and increasing heat capacity, while excessively low areal density reduces absorption. Therefore, the forest height is taken as 0.5 μm to 100 μm, and the areal density is not less than 10. 9 A combination of roots per square centimeter with a connectivity porosity of not less than 30% can achieve a balance between absorption rate, heat capacity, and process feasibility. The average axial angle between the carbon nanotube bundles and the normal direction of the suspended membrane is not greater than 45°, which can ensure the vertical orientation consistency of the forest and avoid the formation of a dense film due to excessive tilting, thereby losing connectivity porosity.
[0057] The lateral connecting layer serves to connect the roots of different vertical carbon nanotube bundles laterally, forming a measurable resistance path. When a carbon nanotube network layer is used as the lateral connecting layer, its thickness ranges from 10 nm to 500 nm. A thinner thickness results in higher resistance and limited signal-to-noise ratio, while a thicker thickness increases the heat capacity of the root region. When a conductive polymer layer or graphene layer is used, its temperature coefficient of resistance can be superimposed with the temperature coefficient of resistance of the forest contact junction, thereby adjusting the temperature coefficient of resistance of the equivalent root resistance to a range of 0.5% per Kelvin to 3% per Kelvin. The root electrode 6 uses a four-terminal or Kelvin connection, which allows the contact resistance between the electrode and the lateral connecting layer to be subtracted from the measurement circuit, preventing the temperature drift of the contact resistance from being mistakenly counted as the temperature response of the forest, thus improving the accuracy and stability of temperature readout.
[0058] Further explanation of the array structure embodiment: The three height regions of the gradient forest are not isolated from each other, but are continuously distributed on the same suspended film 5. Their density gradient can be obtained by varying the thickness of the partitioned catalyst, local growth time, plasma etching, or mechanical shearing. The low-height, high-density region 11 mainly contributes to bulk absorption in the short-wavelength direction, while the high-height, low-density region 13 contributes to absorption in the long-wavelength direction through its longer optical path and larger porosity. The medium-height, medium-density region 12 lies between the two, serving as a transition and broadening platform. The microcavity depth between the reflective layer 9 and the suspended film 5 determines the position of the microcavity interference absorption peak: the greater the microcavity depth, the more the interference absorption peak shifts towards the long-wavelength direction. Therefore, by combining low, medium, and high forest regions with equivalent optical cavity depths of 0.8μm to 1.5μm, 1.5μm to 2.5μm, and 2.5μm to 4μm, respectively, the absorption peaks of the forest volume and the microcavity interference absorption peaks can complement each other in the two bands of 3μm to 5μm and 8μm to 14μm, thereby increasing the peak and valley values of the absorption spectra in both bands and ensuring that the absorption rate is not less than 85%.
[0059] At the focal plane array level, reference pixels and blank pixels perform on-chip correction. The forest of reference pixels is infrared-shielded, compacted, or thermally coupled to the substrate 8, and its output mainly reflects the substrate temperature and bias self-heating, thus it can be used to subtract temperature drift of the common mode. Blank pixels do not contain forests, and their output reflects the response of the suspended film 5 and the readout circuit itself, thus it can be used to evaluate forest transfer integrity and root resistance drift. By combining the outputs of reference pixels, blank pixels, standard height forest pixels, and gradient forest pixels, fixed pattern noise and temperature drift of the array can be suppressed without the need for an external temperature homogenizer, improving the array's consistency across different operating temperatures and device batches.
[0060] Further explanation of the preparation method embodiments. The local low-temperature in-situ growth route and the pre-fabricated forest root transfer route each have their applicable scenarios. The advantage of the local low-temperature in-situ growth route is that the forest roots are in-situ bonded to the lateral connecting layer and root electrode 6, resulting in lower interface thermal resistance and contact resistance, making it suitable for applications requiring high root resistance consistency. Its challenge lies in establishing a local high-temperature field sufficient for carbon nanotube growth within the temperature window that the CMOS backend can withstand. Therefore, it places high demands on local heating methods such as local microheaters, pulsed lasers, induction heating, or remote plasma, as well as temperature difference control (50℃ to 250℃). The advantage of the prefabricated forest root transfer route is that high-temperature growth is completed entirely on an independent growth substrate, and the suspended film 5 and readout circuit do not experience the growth temperature of the carbon nanotube forest 1 at all, thus having the best compatibility with the already interconnected CMOS readout chip. The difficulty lies in maintaining the integrity of the forest island and the relative position of the root vascular bundles during the transfer process. Therefore, a transferable support layer is required, and the retention rate of the forest island after transfer should be controlled at no less than 90%, and the effective root bonding area should be controlled at 70% to 95% of the forest projection area.
[0061] Regardless of the route used, the release and drying processes are crucial for maintaining the forest's morphology. Due to the high aspect ratio and interconnected pores of the vertical carbon nanotube forest 1, the surface tension generated during liquid evaporation within the pores easily leads to forest aggregation and lodging. Therefore, dry release methods, such as gaseous hydrogen fluoride or oxygen plasma, which have no liquid surface tension, are preferred. When wet release is necessary, the forest pores can be temporarily filled with a removable polymer, followed by supercritical drying or freeze-drying to remove the solvent, and finally, the filler can be removed in a low-damage manner. Furthermore, a porous confinement layer 2, pre-formed with support pillars, provides restraint to the forest top during release and drying, further reducing the risk of forest aggregation and lodging.
[0062] Further explanation of the comparative examples: The comparison of the three sets of pixels was conducted under controlled conditions, i.e., the suspended membrane 5, the supporting beam 7, the readout method, and the pixel projected area were all kept consistent, with only the absorption layer configuration being changed, thus attributing the performance differences to the absorption layer configuration itself. Comparative example one represents the traditional approach of increasing the thickness of the planar absorption layer to improve the absorptivity, the inherent contradiction of which is that the absorptivity and heat capacity increase synchronously with the thickness; comparative example two represents the approach of directly using a vertical forest without edge thermal isolation design, the inherent defect of which is that the forest extends to the supporting beam, forming lateral heat leakage. The pixels of the present invention, through the synergistic limitation of the central local thermal coupling zone and the annular discontinuity zone 10, avoid the above two types of defects, achieving a balance of high absorption, low heat capacity, and low equivalent thermal conductivity, the result of which is intuitively reflected in Table 1. It should be noted that the absorptivity, relative heat capacity per unit area, and equivalent thermal conductivity in Table 1 are exemplary acceptance indicators and relative comparison results, and the specific values should be based on actual measurements.
[0063] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.
[0064] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.
[0065] Furthermore, the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or apparatus.
[0066] For ease of description, spatial relative terms such as "above," "on top of," "on the upper surface of," "above," etc., are used herein to describe the spatial positional relationship of a device or feature as shown in the figures to other devices or features. It should be understood that spatial relative terms are intended to encompass different orientations in use or operation beyond the orientation of the device as described in the figures. For example, if the device in the figures were inverted, a device described as "above" or "on top of" other devices or structures would subsequently be positioned as "below" or "under" other devices or structures. Thus, the exemplary term "above" can include both "above" and "below." The device may also be positioned in other different ways, such as rotated 90 degrees or in other orientations, and the spatial relative descriptions used herein will be interpreted accordingly.
[0067] In the detailed description above, reference has been made to the accompanying drawings, which form part of this document. In the drawings, similar symbols typically identify similar parts unless the context otherwise indicates otherwise. The illustrated embodiments described in the detailed specification, drawings, and claims are not intended to be limiting. Other embodiments may be used and other changes may be made without departing from the spirit or scope of the subject matter presented herein.
[0068] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A vertical carbon nanotube forest microbolometer, characterized in that, The system includes a substrate, a suspended membrane spaced apart from the substrate, a thermally insulating support beam connecting the suspended membrane and the substrate, and a vertical carbon nanotube forest disposed on the suspended membrane, wherein: The vertical carbon nanotube forest is only located in the central local thermal coupling area of the suspended membrane, and the projected area of the central local thermal coupling area accounts for 30% to 85% of the area of the suspended membrane. The periphery of the vertical carbon nanotube forest and the edge of the suspended membrane form a gap of 0.5 μm to 10 μm. At least two spaced-apart root electrodes and a transverse connecting layer connecting different vertical carbon nanotube bundles are provided in the root region not exceeding 20% of the height of the vertical carbon nanotube forest, so that the incident infrared radiation is absorbed by the vertical part of the vertical carbon nanotube forest, and the temperature rise is converted into an electrical signal through the equivalent resistance change of the contact junction of the transverse connecting layer and the carbon nanotube bundle.
2. The vertical carbon nanotube forest microbolometer according to claim 1, characterized in that, The average axial angle of the carbon nanotube bundles in the vertical carbon nanotube forest relative to the normal direction of the suspended membrane is no greater than 45°, and the areal density is no less than 10. 9 The forest height is 0.5 μm to 100 μm per square centimeter, and the interconnected porosity from the top of the forest to the root region is not less than 30%; the lateral connecting layer includes a carbon nanotube network layer, a conductive polymer layer, a graphene layer, or a discrete conductive nanoparticle connecting layer with a thickness of 10 nm to 500 nm.
3. The vertical carbon nanotube forest microbolometer according to claim 1, characterized in that, The vertical carbon nanotube forest is arranged along the plane of the suspended membrane in at least three regions with heights of 3μm to 8μm, 8μm to 18μm, and 18μm to 35μm, and a reflective layer is disposed below the suspended membrane; the height distribution of the forest and the cavity depth from the reflective layer to the suspended membrane are configured such that the absorption peak and valley values of the two wavebands of 3μm to 5μm and 8μm to 14μm are both not less than 85%.
4. The vertical carbon nanotube forest microbolometer according to claim 1, characterized in that, The top of the vertical carbon nanotube forest is provided with a porous limiting layer supported by several pillars. The porous limiting layer maintains a non-contact gap with the top of the forest, with an opening rate of 20% to 95% and a thickness of less than one-tenth of the forest height.
5. The vertical carbon nanotube forest microbolometer according to claim 2, characterized in that, The surface of the carbon nanotube bundle is provided with a conformal functional layer of alumina, boron nitride, fluoropolymer or discrete metal nanoparticles with a thickness of 1 nm to 20 nm; the absolute value of the temperature coefficient of resistance of the lateral resistance path in the root region is 0.5% per Kelvin to 3% per Kelvin, and the root electrode adopts a four-terminal or Kelvin connection.
6. The vertical carbon nanotube forest microbolometer according to claim 1, characterized in that, Under the same pixel projection area and the same effective thermal sensitive area, the average absorption rate of the vertical carbon nanotube forest in the 8μm to 14μm band is not less than 95%, and its heat capacity per unit area is not more than 1.2 times that of the heat capacity per unit area of a 200nm thick planar carbon nanotube network.
7. An infrared focal plane array, characterized in that, It includes multiple vertical carbon nanotube forest microbolometers as described in any one of claims 1 to 6 arranged in rows and columns, readout circuits for pixel-by-pixel gating and biasing, and at least one reference pixel, with different pixels having different forest height distributions or microcavity depths to form mid-wave infrared subarrays and long-wave infrared subarrays.
8. The infrared focal plane array according to claim 7, characterized in that, The forest of the reference pixel is infrared shielded, compacted, or thermally coupled to the substrate, and blank pixels without forest are set to correct for substrate temperature, forest transfer integrity, and root resistance drift, respectively.
9. A method for preparing a vertical carbon nanotube forest microbolometer according to any one of claims 1 to 6, characterized in that, include: A root electrode, a lateral connecting layer, and a catalytic region are formed in the central local thermal coupling region of the suspended membrane; A vertical carbon nanotube forest with roots facing the suspended membrane is formed, and an annular discontinuous region is retained that does not extend to the thermally insulating support beam; wherein the vertical carbon nanotube forest is formed by one of the following methods: local low-temperature in-situ growth with the overall temperature of the substrate not exceeding 400°C and the local effective temperature of the catalytic region not exceeding 550°C, or transfer bonding with the roots of the forest facing the suspended membrane after prefabricating the vertical carbon nanotube forest on an independent growth substrate and sequentially forming a transferable support layer and an adhesive layer.
10. The preparation method according to claim 9, characterized in that, The local low-temperature in-situ growth employs plasma-enhanced chemical vapor deposition with a plasma power of 20W to 300W, a volume ratio of carbon-containing gas to diluent gas of 1:2 to 1:50, and controls the temperature difference between the catalytic region and the substrate to be 50°C to 250°C. In the transfer bonding process, the retention rate of the transferred forest island is not less than 90%, and the effective bonding area between the roots and the suspended membrane accounts for 70% to 95% of the forest projection area.
