Patent Application
20240209496
This invention relates to low-capacitance nanopore sensors on insulating substrates fabricated with circular or polygonal etching windows.
Nanopore-based electronic DNA sequencing is promising in clinical use and personal medical care for its portability, high speed, low cost, and long DNA read length. However, despite improvements in the nanopore structure design and data analysis algorithms, it remains extremely difficult to accurately interpret the DNA sequence from the electrical signals. One challenge is associated with an insufficient signal-to-noise ratio (SNR) that limits the signal reading speed and resolving power.
Sapphire-supported (SaS) nanopore sensors for high-speed, sensitive, and low-noise DNA detection using thin membranes on insulating crystal wafers (e.g., sapphire) are described. Substrate insulation reduces conductance-based capacitance and the feasibility of controlling membrane dimension down to <10 μm allows reduced chip capacitance even for ultrathin (˜1 nm) membrane materials (e.g., low-dimensional materials). Thus, nanopore chip capacitance can be reduced by orders of magnitude compared with that of a typical silicon (Si) chip, thus reducing the background high-frequency electrical noise and improving high-bandwidth sensing. The low device capacitance is also favorable for fast circuit response. Further, the membrane creation in sapphire is batch-processing compatible, and will allow high-throughput, low-cost production for device implementation. Additionally, the single-crystalline and insulating (bandgap ˜10 eV) nature of the sapphire can suppress leakage current and photoluminescence background for high DNA signal integrity, feasible for both low-noise electrical and optical sensing. A hexagonal window design can be used to fabricate membranes shaped as irregular hexagons or equilateral triangles.
A method for reproducibly creating SaS nanopores triangular membranes with <25 μm dimensions on 2-inch sapphire wafers is also described. Completely eliminating the substrate conductivity-induced stray capacitance, these SaS nanopores produced two-order-of-magnitude smaller device capacitance (10 pF) compared to a measured SiS nanopore (˜1.3 nF), despite having a 100 times larger membrane area. Accordingly, the SaS nanopores generated ˜2.6 times smaller RMS ionic current noise than the SiS nanopore at 100 kHz bandwidth, which resulted in high-fidelity DNA sensing with two times higher SNR despite a larger nanopore size and thicker SiN membrane. The SaS nanopore sensor can also be used to interrogate a variety of other biomolecules at single-molecule level, such as RNA, protein, and extracellular vesicles, and their molecular interactions at improved speed and accuracy.
Embodiment 1 is a method of fabricating a nanopore sensor, the method comprising depositing a first oxide layer on a first side of a sapphire substrate and a second oxide layer on a second side of the sapphire substrate opposite the first side; patterning the second oxide layer to form an etch mask having a mask opening in the second oxide layer; performing a crystalline orientation dependent wet anisotropic etch on the second side of the sapphire substrate using the etch mask to form a cavity having sloped side walls through the sapphire substrate to yield an exposed portion of the first oxide layer, each of the sloped side walls being a crystalline facet aligned with a respective crystalline plane of the sapphire substrate; depositing a silicon nitride layer on the first oxide layer; removing the exposed portion of the first oxide layer in the cavity, thereby defining a silicon nitride membrane in the cavity; and forming an opening through the silicon nitride membrane, wherein the opening is a nanopore with a diameter in a range of 1 nm to 20 nm.
Embodiment 2 is the method of embodiment 1, wherein the mask opening is in the shape of a circle.
Embodiment 3 is the method of any one of embodiments 1 through 2, wherein the silicon nitride membrane is polygonal and has three-fold symmetry.
Embodiment 4 is the method of any one of embodiments 1 through 3, wherein the polygonal silicon nitride membrane has 6 edges.
Embodiment 5 is the method of any one of embodiments 1 through 4, wherein the mask opening is in the shape of a polygon,
Embodiment 6 is the method of any one of embodiments 1 through 5, wherein the polygon is an equilateral triangle, and an edge of the equilateral triangle is aligned at an offset angle α from a crystalline plane of the sapphire substrate, where 0°<α<60°.
Embodiment 7 is the method of any one of embodiments 1 through 6, wherein the silicon nitride membrane is in the shape of an equilateral triangle when 0°<α<20° and 40°<α<60°.
Embodiment 8 is the method of any one of embodiments 1 through 7, wherein the silicon nitride membrane is in the shape of a nonagon when 20°<α<40°.
Embodiment 9 is the method of any one of embodiments 1 through 8, wherein the polygon is a hexagon.
Embodiment 10 is the method of any one of embodiments 1 through 9, wherein an edge of the hexagon is aligned at an offset angle α from a crystalline plane of the sapphire substrate, where 5°<α<55°.
Embodiment 11 is the method of any one of embodiments 1 through 10, wherein the silicon nitride membrane is in the shape of an equilateral triangle when 10°<α<35°.
Embodiment 12 is the method of any one of embodiments 1 through 11, wherein an area of the silicon nitride membrane is substantially constant when 10°<α<35°.
Embodiment 13 is the method of any one of embodiments 1 through 2, wherein sides of the silicon nitride membrane are parallel to sides of the hexagon when 10°<α<35°.
Embodiment 14 is the method of any one of embodiments 1 through 13, wherein the silicon nitride membrane is in the shape of an irregular hexagon when a <10° or a >35°.
Embodiment 15 is the method of any one of embodiments 1 through 14, wherein sides of the irregular hexagon are oriented along particular crystal orientations of the sapphire substrate.
Embodiment 16 is the method of any one of embodiments 1 through 15, wherein interior angles of the irregular hexagon are between about 90° and 150°.
Embodiment 17 is the method of any one of embodiments 1 through 16, wherein a duration of the etch is in a range of minutes to hours.
Embodiment 18 is the method of any one of embodiments 1 through 17, wherein the etch is conducted at a temperature in a range between 150° C. and 450° C.
Embodiment 19 is the method of any one of embodiments 1 through 18, wherein a dimension of a surface of the sapphire substrate is in a range between about 1 mm and about 20 cm and a thickness of the sapphire substrate is in a range between about 0.1 mm and about 1 mm.
Embodiment 20 is a nanopore sensor comprising a sapphire substrate defining a first opening and second opening, wherein the first opening and the second opening are superimposed; and a silicon nitride membrane extending across the second opening and defining a nanopore therethrough, wherein the membrane is in the shape of a polygon.
Embodiment 21 is the method of embodiment 20, wherein the first opening is in the shape of a circle.
Embodiment 22 is the method of any one of embodiments 20 through 21, wherein the silicon nitride membrane is polygonal and has three-fold or six-fold symmetry.
Embodiment 23 is the method of any one of embodiments 20 through 22, wherein the polygonal silicon nitride membrane has 6 edges.
Embodiment 24 is the method of any one of embodiments 20 through 23, wherein the mask opening is in the shape of a polygon,
Embodiment 25 is the method of any one of embodiments 20 through 24, wherein the polygon is an equilateral triangle, and an edge of the equilateral triangle is aligned at an offset angle α from a crystalline plane of the sapphire substrate, where 0°<α<60°.
Embodiment 26 is the method of any one of embodiments 20 through 25, wherein the silicon nitride membrane is in the shape of an equilateral triangle when 0°<α<20° or 40°<α<60°.
Embodiment 27 is the method of any one of embodiments 20 through 26, wherein the silicon nitride membrane is in the shape of a nonagon when 20°<α<40°.
Embodiment 28 is the method of any one of embodiments 20 through 27, wherein the polygon is a hexagon.
Embodiment 29 is the method of any one of embodiments 20 through 28, wherein an edge of the hexagon is aligned at an offset angle α from a crystalline plane of the sapphire substrate, wherein 5°<α<55°.
Embodiment 30 is the method of any one of embodiments 20 through 29, wherein the silicon nitride membrane is in the shape of an equilateral triangle when 10°<α<35°.
Embodiment 31 is the method of any one of embodiments 20 through 30, wherein sides of the silicon nitride membrane are parallel to sides of the hexagon when 10°<α<35°.
Embodiment 32 is the method of any one of embodiments 20 through 31, wherein the silicon nitride membrane is in the shape of an irregular hexagon when a <10° or a >35°.
Embodiment 33 is the method of any one of embodiments 20 through 32, wherein sides of the irregular hexagon are oriented along particular crystal orientations of the sapphire substrate.
Embodiment 34 is the method of any one of embodiments 20 through 33, wherein interior angles of the irregular hexagon are between about 90° and 150°.
The details of one or more embodiments of the subject matter of this disclosure are set forth in the accompanying drawings and the description. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.
Embodiments of the present invention provide a method for forming a low-noise nanopore sensor in a thin membrane suspended on an insulating substrate, for example, a sapphire substrate. In some embodiments, the manufacturing method of such nanopore devices on sapphire includes using controlled anisotropic wet etching, in which the etch rates are dependent on the crystal orientation of sapphire and hence can precisely control the membrane size by the design of an etching mask. Insulating materials, such as silicon oxide, silicon nitride, etc., have been used as the membrane layer in which the nanopore is formed. Wet etching of sapphire substrates can have a high etching rate of 0.1 μm to 1 μm per minute, allowing etching through the thickness of the sapphire wafer at a high throughput. Compared with dry etching processes that require extensively long etching time and single-wafer processing, wet etching is compatible with large-volume batch production, hence allowing high throughput production of high-sensitivity nanopore sensors at a low cost. Additionally, the method provides a process and system that are compatible with conventional process technology without substantial modifications to conventional equipment and processes.
A first electrode 261 and a second electrode 262 are coupled to an electrically conductive fluid 270 disposed in the first and second reservoirs. The electrodes are configured to impose an electrical potential difference from a voltage supply (V) 280 to conductive fluid 270. As a biomolecule 240 passes though the nanopore channel, it partially blocks the nanopore and thus changes the effective nanopore resistance. This results in a current amplitude change and can also modify the DNA translocation time through the nanopore. These electrical signals can be measured to provide genetic information on the molecule. Nanopore device 200 can also include a current measuring circuit (I) 282 for measuring the current through the nanopore. In this case, a constant voltage can be applied and current measured, which could be an instantaneous measurement or over a period of time (e.g., with an integrating capacitor).
Alternatively, a constant current can be applied to the conductive fluid, and the voltage can be measured to determine changes in the resistance as a biomolecule 240 passes though the nanopore channel. In this case, component 282 (I) can represent a constant current supply, and component 280 (V) can represent a voltage measurement circuit. Further, both current and voltage can vary as well. As long as the current or voltage source is varied in a known or reproducible fashion, the measured electrical signals can be used to identify genetic information of the molecule. Nanopore device 200 can also have a control circuit 284 for controlling the measurement for processing the detected signal. Control circuit 122 may include amplifier, integrator, noise filter, feedback control logic, and/or various other components. Control circuit 122 may be further coupled to a computer 286 for analyzing the signals to determine the components of the molecule, e.g., bases of a DNA molecule.
Embodiments of the present invention provide a method of forming dielectric membranes on insulating sapphire substrate, which is suitable to manufacture nanopore devices. The method can also be implemented using other crystalline insulating substrates having a wet etching selectivity that is crystalline orientation dependent. Embodiments can use an anisotropic chemical etching method to create cavities on a sapphire substrate. Typically, a mixture of sulfuric acid and phosphoric acid is heated to above 250° C. and used for sapphire etching. This etching method results in selective etching c-plane sapphire much faster than other crystal planes. Using such an etching solution, a patterned material layer, e.g. made in SiO2, can be used as an etch mask to effectively protect the sapphire underneath and hence creates a cavity in the region without this protective layer.
By precisely designing the shapes and dimensions of this protective layer (etching mask), embodiments can control the crystal facets on the sapphire substrates, and thus control the lateral dimensions of membrane precisely. This anisotropic wet etching is suitable for batch processing of multiple wafers for large-scale and low cost production, and thus hundreds and even thousands of nanopore membranes can be manufactured at the same time. The mask shape and dimensions can be precisely controlled, hence allowing precise high-yield production. In addition, the sapphire substrate is compatible with Si-based micro and nanofabrication technologies and can be further integrated with other electronic components.
At step 410 of method 400 in
At step 420, as shown in
The deposition can be carried by standard processes used in the semiconductor industry, such as low-pressure chemical vapor deposition (LPCVD), plasma enhanced chemical vapor deposition (PECVD), etc. In an LPCVD process, a silicon precursor, such as silane SiH4, and an oxygen source, e.g., O2, are reacted in a low pressure system to form a layer of silicon oxide. In a PECVD process, the activation of plasma enables oxide deposition at lower temperatures. The thickness of such SiO2 mask can be in a range of 10 nm to 10 μm depending on the etching selectivity of the dielectric layers in subsequent etching steps. In the description below, the first dielectric layer 212 and the second dielectric layer 214 will be referred to as the first oxide layer 212 and the second oxide layer 214. It is understood that other dielectric materials can also be used.
At step 430, as shown in
Next, the SiO2 layer can be etched by reactive ion etching (RIE) with the patterned resist mask. As is known in the semiconductor industry, reactive-ion etching (RIE) is a type of dry etching that uses chemically reactive plasma to remove material deposited on wafers. The plasma is generated under low pressure (vacuum) by an electromagnetic field. Under a high voltage, high-energy ions from the plasma attack the wafer surface. The ions can react chemically with the materials on the surface of the wafer, and can also knock off (sputter) some material. Due to the mostly vertical delivery of reactive ions, reactive-ion etching can produce anisotropic etch profiles, such as a vertical profile. In contrast, wet etching is a material removal process that uses liquid chemicals or etchants to remove materials from a wafer. The specific patters are defined by masks on the wafer. Materials that are not protected by the masks are etched away by liquid chemicals.
Unlike dry etching, wet etching is usually isotropic, i.e., the etch rate is the same in all directions. In embodiments of the invention, wet etching of the sapphire substrate is carried out using an anisotropic etching process in which the etch rate depends on the crystalline orientation. After the conclusion of the etch process, the photoresist is stripped using a standard process, e.g., by oxygen plasma ashing. In some embodiments, the mask opening can have a triangular-shaped window, such that all sidewalls of the cavity etched in a c-plane sapphire substrate, i.e., sapphire with (0111) orientation, are defined by crystalline facets after the crystalline orientation dependent etch. This design can lead to better etch profile control. Further, the triangular shaped opening can provide more mechanical stability. In
At step 440, as shown in
At step 450, as shown in
This etching selectivity allows one of the films to be used as a masking layer or an etch stop layer during the etching of the other film. In this example, silicon nitride layer 220 has a desirable etch selectivity with respect to the first oxide layer 212, and the SiO2 can be selectively etched to leave the thin S13N4 layer suspended on sapphire. In some embodiments, a protective layer 219, e.g., a dielectric layer, can be formed on the sidewalls of the cavity. To simplify the drawings, dielectric layers 219 will be omitted in some of the figures described below.
At step 460, as shown in
At step 470, as shown in
In electron-beam lithography, a focused beam of electrons is scanned to draw desired shapes on a surface covered with an electron-sensitive resist film. The electron beam changes the solubility of the resist, enabling selective removal of either the exposed or non-exposed regions of the resist by immersing it in a solvent in a developing process. One advantage of electron-beam lithography is that it can draw patterns (direct-write) with sub-10 nm resolution without using a mask. This form of maskless lithography has high resolution and low throughput. After the mask pattern is formed on the wafer, reactive ion etching can be used to form the nanopore. Alternatively, an opening can be formed in the silicon nitride film using focused electron beam etching to form the nanopore. In this maskless process, an electron beam is used to activate a chemical reaction in selected regions on a wafer to form a nanopore in the membrane film.
As described above, in embodiments of the present invention, a method for forming a nanopore device on a sapphire substrate includes anisotropic etching of a sapphire substrate based on different etching rates of various crystalline planes in a sapphire substrate using appropriate wet etching chemicals.
The etching mask can be patterned and aligned to selected crystal orientations of the sapphire substrate for accurate determination of the membrane dimensions. For example, FIGS. SA-5C illustrate a method of using a triangular-shaped etching mask for forming a triangular-shaped membrane according to some embodiments of the present invention.
with h/L=tan a where a is the angle between the c-plane sapphire and the evolved sidewall facets during etching, and h is the depth of the cavity or the thickness of the sapphire wafer. It follows that the relationship between the etching window sizes can be expressed as follows:
In the method described above, the depth of the cavity in the sapphire substrate h is the depth of the cavity and is also the thickness of the sapphire substrate, which typically ranges from 100 μm to 1 mm. After determination of the sapphire thickness to be used and the etching angle a, the length of the side of the window in each mask Li can be completely determined based on the desired membrane size L2.
with a=60° and a sapphire thickness of 250 μm. For example, for a membrane size L2 of about 10 μm, the mask dimension Li can be determined to be about 510 μm, as indicated by data point 1010. To get a larger membrane size L2 of about 100 μm, the mask dimension L1 can be determined to be around 600 μm, as indicated by data point 1020. This method allows good control of membrane size, which is a factor to controlling the membrane capacitance and the current noise.
The mask window can be a circle or a polygon (e.g., a triangle or a hexagon). The polygon can have a regular or irregular shape.
In some embodiments, the polygon is an equilateral triangle, and an edge of the equilateral triangle is aligned at an offset angle α from a crystalline plane of the sapphire substrate, where 0°<α<60°. The silicon nitride membrane is in the shape of an equilateral triangle when 0°<α<20° and 40°<α<60°. The silicon nitride membrane is in the shape of a polygon (e.g., a nonagon) when 20°<α<40°.
In some embodiments, the polygon is a hexagon. An edge of the hexagon can be aligned at an offset angle α from a crystalline plane of the sapphire substrate, where 5°<α<55°. The silicon nitride membrane is in the shape of an equilateral triangle when 10°<α<35°. An an area of the silicon nitride membrane is substantially constant when 10°<α<35°. Sides of the silicon nitride membrane are parallel to sides of the hexagon when 10°<α<35°. The silicon nitride membrane is in the shape of an irregular hexagon when 10°>α>35°. Sides of the irregular hexagon are oriented along particular crystal orientations of the sapphire substrate. Interior angles of the irregular hexagon are between about 90° and 150°.
In some embodiments, the mask opening is in the shape of a circle. When the mask opening is in the shape of a circle, the silicon nitride membrane is polygonal and can have three-fold symmetry. The polygonal silicon nitride membrane has 6 edges.
2 inch sapphire wafers were lithographically patterned using a desktop laser writer to study the impact of etching window design on the membrane formation. The mask design included triangular etching windows with a rotational alignment angle α between one of the window sides to the A-plane sapphire facet (11
According to the results, α˜0° was chosen to create small membranes on sapphire. Additionally, the triangular mask window size L1 was varied to experimentally establish a relationship between the mask design L1 and the membrane size L2. A plasma enhanced chemical vapor deposition (PECVD) SiO2 layer was deposited and patterned into triangular cavity windows by laser writer and reactive ion etching (RIE). The SiO2 masked etching of 250 μm thick 2″ diameter sapphire wafers in a mixture of hot sulfuric and phosphoric acids (solution temperature ˜300°), with a high selectivity of 500:1. Rectangular dicing marks surrounding the cavity windows were includes during lithography, creating trenches in sapphire after acid etching that allowed the sapphire to be hand-diced into 5 mm by 5 mm square chips. This chip size was designed to fit into a fluidic jig and TEM holder for nanopore drilling and electrical characterization. The formed SiO2 membrane (3 μm thick) was found intact during the etching and chip dicing process, with the membrane size L2 tunable in a wide range from 5 to 200 μm. The correlation between L1 and the measured membrane size L2 was fitted using a theoretical model, and the effective facet angle δ˜50° was found. Thus, the fabrication of ultrasmall membranes for functional sapphire chips was demonstrated.
The membrane geometry of the resulting membranes was found to be sensitive to α, (i.e., the alignment angle between the triangular mask and the sapphire crystal flat). As a result, an error in α may result in a deviation in membrane size and accordingly adversely affect the accuracy in controlling the device capacitance. Such an effect makes it particularly challenging to create small membranes. In an effort to increase accuracy in controlling the device capacitance, hexagonal etching windows were used.
The primary facets of sapphire include the c-plane (0001, the wafer surface in these experiments), the A-plane (11
250 μm thick 2-inch c-plane sapphire wafers, purchased from Precision Micro-Optics Inc., were RCA2 cleaned (deionized water: 27% hydrochloric acid: 30% hydrogen peroxide =6:1:1, 70° C.) for 15 min. The RCA2 surface cleaning promotes film adhesion to the substrate, which otherwise can result in film cracking during high-temperature sapphire etching. One to 3 μm silicon oxide (SiO2 ) (thicker is preferred for larger membranes) was then deposited via plasma-enhanced chemical vapor deposition (PECVD, 350° C., deposition rate 68 nm/min) on both sides, followed by photolithography and reactive-ion-etching (RIE) (PlasmaTherm 790, CHF3 based chemistry, etching rate 46 nm/min) to form a triangular etching window in SiO2. Next, hot sulfuric acid and phosphoric acid (3:1, solution temperature ˜300° C.) were used to etch through the sapphire wafer (etching rate up to 12 μm/hr) and to suspend the SiO2 membrane. To ensure the safety of handling hot acids, a custom-designed a quartz glassware setup suitable for the high-temperature acid-based sapphire etching process was used. The sapphire wafer was intentionally placed vertically in a 2-inch glass boat in the etching container to minimize possible damage to the membrane from the boiling acids. Acid was added to the quartz glassware, and then loaded the 2-inch glass boat with the wafer into the quartz glassware. A clamp seal and a condenser column were then installed on top of the glassware to minimize acid vapor leakage. The etching rate was chosen to be relatively slow in this customized container to minimize wafer breakage during etching; however, further increasing the solution temperature is an option to exponentially increase the etching rate and thus allow for larger throughput.
Following sapphire etching, the SiO2 membrane was thinned down as needed by RIE to <1.5 μm. This was followed by depositing a layer of silicon nitride (SiN) (30-300 nm) onto the SiO2 membrane via low-pressure chemical vapor deposition (LPCVD) (Tystar TYTAN 4600, 750° C., deposition rate: 6 nm/min). The unintentionally deposited SiN in the back cavity of the chip was removed by RIE. Next, hydrofluoric acid (8%) was used to etch the SiO2 layer (90 nm/min) to suspend the SiN layer. The final SiN membrane was thinned down as needed by hot 85% phosphoric acid (etching rate ˜2.5 nm/min) to desired thickness.
The thicknesses of membranes were determined by optical reflectance measurement (Filmetrics F40) and by subsequent fitting. An experimentally measured refractive index of the SiN films on a Si monitor sample (Woollam Spectroscopic Ellipsometer) was used to improve fitting accuracy.
For comparison, SiS nanopore membranes were purchased from SiMPore Inc. The chips were made from 100 mm diameter, 200 μm thick, float-zone Si wafer (resistivity of 1-10 Ω·cm) with ˜100 nm thermal SiO2 and ˜20 nm LPCVD SiN films, where the thermal SiO2 from the cavity side was removed to produce an array of suspended SiN membranes of 4-5 μm in diameter. The SiO2 and SiN film thicknesses were confirmed by M-2000 ellipsometer (J.A. Woollam Co.) as 99 nm and 23 nm, respectively.
A JEOL 2010F transmission electron microscope (TEM) was used to drill the nanopores. The 5 mm by 5 mm nanopore chips were diced and placed in a customized TEM sample holder. The largest condenser aperture and beam spot size were used for maximum beam current output. After alignment, imaging magnification was maximized (1.5M), followed by 5-15 min beam stabilization. The focus was re-adjusted when beam drifting was severe; beam stabilization was then re-monitored at maximum magnification. Upon stabilization, the beam spot was reduced to ˜7 mm and rounded by adjusting the condenser astigmatism. Under the conditions of 7.01 kV anode A2 (focusing anode), 3.22 kV anode A1 (extraction anode), and 30 nm SiN membrane, it typically took 75-90 sec to drill through the membrane.
The TEM-drilled nanopore chip was treated with UV ozone cleaner (ProCleaner™, BioForce Nanosciences Inc.) for 15 min to improve hydrophilicity. The chip was then mounted onto a customized flow cell. A solution of 1:1 mixed ethanol and DI water was injected into the flow cell to wet the chip for 30 min. The solution was subsequently flushed away by injection of DI water. Next, 100 mM KCl was injected into the flow cell to test the current-voltage (IV) curve using an Axopatch 200B amplifier and a Digidata 1440A digitizer (Molecular Devices, LLC.). A 1M KCl solution was injected for characterization of the device current.
For DNA sensing, 1 k bp dsDNA (Thermo Scientific NoLimits) was diluted using 1M KCl to 5 ng/μL. Poly(A)40 ssDNA (Standard DNA oligonucleotides, Thermo Fisher Scientific Inc.) was diluted using 1M KCl to 50 nm, followed by brief vortex mixing. The DNA solution was injected into the flow cell to collect DNA signals under a 10 kHz and 100 kHz low-pass filter with a sampling frequency of 250 kHz at 50, 100, and 150 mV bias voltages. The flow cell was kept in a customized Faraday cage on an anti-vibration table (Nexus Breadboard, Thor labs) to minimize the environment noise during measurement. The DNA signals were observed and recorded with the Clampex software. Finally, an edited MATLAB program was used to convert all the .abf files to .mat files. All the collected DNA signals were then imported to an OpenNanopore program to generate the dwelling time and blockade current amplitude data of each DNA signal for subsequent analysis.
Suspended dielectric membranes were created on sapphire by anisotropic wet etching. PECVD-deposited SiO2 was used here due to its high-selectivity in masking sapphire etching, which was experimentally determined to be over 500:1. Considering the three-fold symmetry of sapphire crystal, triangular shaped SiO2 etching masks were patterned. Alignment angle of the masks and the sapphire crystal (denoted α) was varied between 0° and 60°, and membrane evolution was monitored. Triangular membranes formed when 0<α<20° and 40°<α<60°. The two sets of triangles were offset by a rotational angle of ˜30°. In contrast, complex polygon membranes with up to nine sides emerged when 20°<α<40°. Additionally, the membrane area was found sensitive to α, yielding an area of more than three orders of magnitude larger when α˜30° compared to α˜0°.
Since sapphire essentially eliminates the stray capacitance through the substrate, the membrane capacitance of the SaS chips, which is highly dependent on membrane area and thickness, largely determines the total chip capacitance and high-frequency noise. Fabrication of micrometer-sized membranes that are attractive for picofarad sensor capacitance and low-noise biosensing are demonstrated. To guide the mask layout design, theoretical calculations were performed to study the relationship between the membrane and the mask dimensions while keeping α=0°. The membrane triangle length L2 could be engineered by the mask triangle length L1 following
where h is the sapphire wafer thickness and θ is an effective angle between the exposed facets in the cavity and sapphire c-plane that can be empirically determined. By designing L1 from ˜750 μm to ˜900 μm, modulation of the SiO2 membrane size L2 was demonstrated within a wide range, from 5 to 200 μm.
By fitting the experimental data with a theoretically calculated L1-L2 relationship, it was estimated that a good empirical value for facet angle θ is ˜50°. Based on this knowledge, L1 was designed as 760, 762, 764, and 766 μm and α˜0° for wafer-scale fabrication of <20 μm size membranes, which are attractive for picofarad sensor capacitance and low-noise biosensing. Etching two 2-inch sapphire wafers in the same batch, no wafer or membrane breakage was discovered, and 116 suspended micron-sized membranes were obtained, while having 4 membranes not yet completely etched through. It was believed that further etching would eventually create these 4 membranes while slightly enlarging the existing ones due to slower lateral etching on exposed facets. From intentionally patterned rectangular dicing marks surrounding the cavity etching windows, trenches were formed in sapphire wafers after acid etching, allowing even-hand dicing of 5 mm square chips despite the hexagonal crystal structure of sapphire. Importantly, this wafer-scale demonstration strongly indicated the scalability of the membrane formation process, which is crucial to future large-scale, cost-effective sensor fabrication.
When compared to the best available low-noise SiS chips and glass-supported chips that typically have a membrane capacitance of <10 pF, SaS chips were found to be competitive in their expected small capacitance and corresponding capacitive noise. Noticeably, the low-noise SiS and glass chips all require very complicated fabrication processes. For example, the SiS membranes need to be very carefully engineered to reduce the membrane area and introduce thick insulating layers, demanding processes involving nanolithography, bonding, film deposition, etching, and even silicone painting.
A challenge in fabricating glass chips lies in the reproducibility of creating small membranes on glass, at least in part because bulk and isotropic etching of amorphous glass in hydrofluoric acids (HF) can have poor dimension control while RIE etching is typically only applicable at single-wafer or single-chip level with drastically lowered throughput and increased manufacturing cost. Although combining femto-second laser ablation with LPVCD and chemical wet etching could form glass chips with a ˜2 pF device capacitance, it remains unclear how the membrane uniformity (e.g., variation 5 to 40 μm), fabrication throughput, and yield are affected by process fluctuation in laser ablation and chemical etching. The wafer-scale SaS chip design and fabrication strategy presents a scalable manufacturing alternative to the prevalent manufacturing processes of low-noise sensors that are complicated, time-consuming, low-yield, and costly.
The observed membrane size variation within the wafer could be attributed to a few factors. First, the sapphire wafers were found slightly thinner (˜1 μm) at the edge than at the center, which could cause membrane enlargement at the edge. Second, the customized hot-plate based etching apparatus could leave a temperate gradient in the acid bath that could affect the etching rate. Further, acid convection under boiling condition may produce local variation in acid concentration and etching rate. In addition, the complex evolution of sapphire facets, currently not fully understood but thought to be due to the competition between R- and N-planes of the sapphire crystals, could be sensitive to crystal orientation alignment and the etching bath conditions. In future studies, the membrane uniformity could be improved by compensating etching window sizes over the wafer, utilizing an etching system that provides better temperature control and acid circulation, and further studying the etching mechanism and optimizing the etching window designs.
Using the triangular SiO2 membranes formed by sapphire etching, a process to create thin SiN membranes suitable for nanopore formation and DNA sensing was developed. Briefly, low-stress LPCVD SiN film was deposited on suspended SiO2 membranes, and then the SiO2 film was removed via selective dry etching and HF based wet etching from the cavity side. The use of SiN film allows precise control of the membrane thickness and minimizes the impact of SiO2 film stress on the membrane structural integrity. The SiN film can be further thinned down to desired thickness when necessary by either RIE and/or wet etching in hot phosphoric acid. RIE could cause non-uniformity and might damage the membrane, causing current leakage, as shown by current-voltage (IV) characteristics. In contrast, wet etching in hot phosphoric acid yielded uniform SiN membrane without current leakage, and thus could be preferable for a DNA sensing test. A nanopore was drilled in the SiN membranes on the sapphire chip (and the float-zone Si chip using TEM for electrical characterization and DNA sensing.
The device capacitance of the SaS and SiS nanopore chips was characterized. Noticeably, the SaS nanopore chip had a 100 times larger membrane area (L2=68 μm, or ˜2000 μm2 in area) than the SiS chip (4.2×4.7 μm square, or ˜20 μm2) and slightly thicker SiN (measured 30 nm for sapphire and ˜23 nm for Si). The membrane capacitance Cm was estimated at 3.8 pF for the SaS chip, >70 times greater than that of the SiS chip (0.05 pF), following
where ε is the permittivity of SiN, and A the membrane area and d membrane thickness. Experimentally, Cm was found ˜ 10 pF for the SaS chip, with a deviation from theoretical value possibly attributed to slightly smaller SiN thickness in reality, and much smaller than that of the SiS chip (˜1.3 nF) because of the stray capacitance from Si substrate. Considering SaS and SiS nanopores that both have only the simplest membrane structure, insulating sapphire was demonstrated to reduce or eliminate the dominant capacitance resulting from Si substrate conductivity, thus appealing to low-noise measurement.
The ionic current noise for the SaS nanopore, the SiS nanopore, and the open-headstage system (Axopatch 200B) was assessed under 10 kHz and 100 kHz low-pass filter. The root-mean-square (RMS) of the measured current of the SaS nanopore chip is ˜5 and 18 pA using 10 and 100 kHz filters, which is only slightly higher than the open-headstage system RMS noise (3 and 11 pA), and yet much better than those from the SiS nanopores (˜16 and 46 pA). In comparison, the best reported silicone-painted SiS chips that utilized a locally thinned membrane (0.25 μm2 area and 10-15 nm thick in the center) produced ˜7 and ˜13 pA noise current at 10 kHz and 100 kHz, measured by an optimally designed amplifier that outperforms Axopatch 200B in high-frequency recording. Glass-supported nanopore chips with micro-membranes (25 μm2, 70 pF and 314 μm2, ˜2 pF) demonstrated ˜13 and ˜19 pA noise current at 10 kHz bandwidth, which is approximately 3-4 times larger than the SaS chip. A comparison of noise current from glass-supported nanopore chips showed that the SaS chips are successful in suppressing the noise current.
Additionally, analysis of the power spectral density (PSD) further demonstrates that the SaS nanopores outperformed the SiS chips, particularly at high bandwidth (e.g., >10 kHz) due at least in part to the significantly reduced device capacitance. In the moderate frequency range (e.g., 100 Hz to 10 kHz), the noise power of the SaS nanopore was about one order of magnitude smaller (˜10−3 pA2/Hz) than that of the measured SiS nanopore (˜10−2 pA2/Hz) and one order of magnitude smaller or comparable to the glass chips and low-noise SiS chips (10−2˜10−3 pA2/Hz), partly attributed to lower dielectric noise and Johnson noise. Sapphire has a very small dissipation factor D (˜10−5), two to five orders of magnitude smaller than that of typical borosilicate glass (10−3 to 10−2) and Si (1-100) and comparable to that of high-purity fused silica (˜10−6). Such a small dissipation factor, together with its small device capacitance, is favorable for minimizing noise related to dielectric loss SD∝DCchipf, where Cchip is nanopore chip capacitance and f is the frequency. Additionally, the high resistivity of sapphire (>1014 Ω·cm) also served to minimize resistance-related Johnson noise.
To evaluate the performance of DNA molecule detection capability of the SaS nanopore, 1 k bp ds-DNA were translocated through the SaS and the SiS nanopores at 100 kHz) and 10 kHz low-pass filters under 50 mV, 100 mV, and 150 mV bias, respectively. By comparing representative ionic current traces of 1 k bp dsDNA, it was observed that the DNA signals collected by the SiS nanopore displayed severe signal distortion, particularly at lower bias voltages. These irregular signals, together with the high baseline noise, made it challenging to faithfully distinguish DNA signals from the background. In comparison, the SaS nanopore produced easily distinguishable DNA signals with much less distortion or noise at as high as 100 kHz bandwidth. Additionally, low-frequency recording, e.g., at 10 kHz, would result in serious data loss of fast DNA signals, thus presenting only longer and in some occasions distorted signals. Thus, the SaS nanopores were found to enable high-speed, high-throughput, and high-fidelity detection of DNA signals.
To study the DNA translocation mechanism from the SaS chip, the DNA signals were extracted using the OpenNanopore Program, and then scatter-plotted the fractional blockade current IB(=ib/i0) and the dwelling time Δt of all the DNA events under 50 mV. Here ib is the blocked-pore current and i0 is the open-pore current. The use of IB allows us to eliminate the impact of bias difference on DNA signal analysis. Two distinct populations were observed and recognized as the translocation events and the collision events. Further, the current blockade distribution was analyzed and fitted with Gaussian function, producing two distinct IB populations attributed to translocation and collisions. The dwelling time Δt of each of the two event populations was analyzed and fitted with exponential decay function. This showed that the translocation events had a longer tail (decay constant 16.19 μs) than the collision events (decay constant 8.45 μs).
This signal segregation approach was further applied to analyze all the DNA signals collected from the SiS and SaS nanopores. By scatter-plotting the normalized DNA blockade signal (1−IB=ΔI/i0) and marking the normalized current noise (IRMS/i0, dash-dot lines) at each bias voltage (50 mV, 100 mV, 150 mV), the SNR (defined here as
of the true DNA translation signals was investigated. The SaS nanopores are seen to produce slightly smaller DNA signal amplitude than SiS nanopores because of their larger pore size and thicker membrane. Noticeably, given the suppressed noise current, the SaS nanopore still evidently outperformed the SiS nanopore in SNR. For example, the SaS nanopore has a SNR of 21 at 150 mV bias, almost twice as good as the SiS nanopore with a SNR of 11. Generally smaller nanopore and thinner membrane are preferred for optimal signals and SNRs. Experimental data illustrated the impact of pore sizes on the DNA signals for both SiS (e.g., 4 nm, 7 nm) and SaS (e.g., 7 nm, 2 0 nm) nanopore chips (
Detection of short single-stranded (ss) DNA molecules using SaS nanopores was attempted. Ionic current traces of Poly(A)40 ssDNA translocation events were recorded under 100 kHz low-pass filter with the voltages from 100 mV to 150 mV. The same analysis was performed to investigate the SNR of this ssDNA, and a SNR of ˜6 was obtained for both 100 mV and 150 mV bias voltages. This provided evidence that the SaS nanopores can detect a wide range of biomolecules of different sizes. It is believed that SNR can be enhanced by using thinner membrane thickness and smaller nanopores.
Although this disclosure contains many specific embodiment details, these should not be construed as limitations on the scope of the subject matter or on the scope of what may be claimed, but rather as descriptions of features that may be specific to particular embodiments. Certain features that are described in this disclosure in the context of separate embodiments can also be implemented, in combination, in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments, separately, or in any suitable sub-combination. Moreover, although previously described features may be described as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can, in some cases, be excised from the combination, and the claimed combination may be directed to a sub-combination or variation of a sub-combination.
Particular embodiments of the subject matter have been described. Other embodiments, alterations, and permutations of the described embodiments are within the scope of the following claims as will be apparent to those skilled in the art. While operations are depicted in the drawings or claims in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed (some operations may be considered optional), to achieve desirable results.
Accordingly, the previously described example embodiments do not define or constrain this disclosure. Other changes, substitutions, and alterations are also possible without departing from the spirit and scope of this disclosure.
This application is a continuation of U.S. patent application Ser. No. 17/908,795 entitled “LOW-CAPACITANCE NANOPORE SENSORS ON INSULATING SUBSTRATES” and filed on Sep. 1, 2022, which is a National Stage Application of International Application No. PCT/US2021/020585, filed Mar. 3, 2021, which claims the benefit of U.S. Patent Application No. 62/984,381 entitled “LOW-CAPACITANCE NANOPORE SENSORS ON INSULATING SUBSTRATES” and filed on Mar. 3, 2020, all of which are incorporated herein by reference in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| 62984381 | Mar 2020 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 17908795 | Sep 2022 | US |
| Child | 18545658 | US |
