Surface acoustic wave devices incorporating thin layers of metallic materials

Abstract

The present invention relates to a surface wave device (200) comprising a substrate (210), a piezoelectric layer (240) above a top surface of the substrate, a pair of electrodes (250A, 250B) in contact with the piezoelectric layer, the two electrodes having fingers (252A, 252B) extending in the same direction to form a periodic structure, the fingers of the two electrodes being arranged alternately with each other and having an inter-finger distance (2p) separating the centers of two adjacent fingers of the same electrode, a metal layer (220) interposed between the substrate and the piezoelectric layer, and a dielectric layer (230) interposed between the metal layer and the piezoelectric layer, the metal layer having a thickness of 5 nm to 100 nm and the dielectric layer having a thickness of 25 nm to 600 nm.

Description

[Technical field]

[0001] The field of the invention is that of surface acoustic wave (SAW) devices having a composite structure incorporating a thin layer of piezoelectric material on a semiconductor substrate. [Background technology]

[0002] Surface acoustic wave devices, or SAW devices, are used in a wide range of applications, particularly in electronics, where they form the core elements of filters, oscillators, delay lines and transformers.

[0003] Piezoelectric materials generate a voltage when deformed by mechanical stress, and conversely, deform when a voltage is applied.

[0004] As a result, when an alternating electrical signal is applied to one or more electrodes in contact with the piezoelectric material, a corresponding mechanical signal (i.e., oscillation or vibration) is produced in the piezoelectric material, converting the electrical signal into a mechanical signal.

[0005] The mechanical signal transferred to the piezoelectric material exhibits a frequency dependence relative to the alternating electrical signal, which is based on the properties of the electrode(s), the properties of the piezoelectric material, and other factors such as the geometry of the acoustic wave device and other structures that make up the device.

[0006] Surface wave devices are increasingly being used to exploit this frequency dependence to provide one or more functions by means of surface acoustic wave (SAW) resonators or SAW transducers, forming so-called "SAW filters" implemented, for example, in the transmission and reception of RF signals for telecommunications applications.

[0007] A SAW filter comprises at least one SAW transducer potentially connected to other transducers to perform a filtering function between an input port and an output port.

[0008] A SAW filter typically comprises an input SAW transducer and an output SAW transducer formed on the same piezoelectric element, where the input SAW transducer generates surface acoustic waves from an input electrical signal and the output SAW transducer receives and converts the surface acoustic waves into an output electrical signal.

[0009] The geometry and dimensions of the transducer, as well as the type and shape of the material used, determine the characteristics of the SAW filter, such as the coupling and reflection coefficients, the Q quality factor at resonance or anti-resonance, bandwidth, parasitic response, suppression of higher order resonances, and temperature dependence.

[0010] US Pat. No. 10,938,367 (B2) discloses an interdigital SAW transducer (IDT) 100 shown in FIG. 1, where A) is a top view and B) is a cross-sectional view along the XX' plane.

[0011] The SAW transducer 100 between the fingers comprises a piezoelectric layer 140 placed on a substrate 110, a pair of electrodes 150A and 150B in contact with the surface of the piezoelectric layer 140, a metal layer 120 interposed between the substrate 110 and the piezoelectric layer 140, and a dielectric layer 130 interposed between the metal layer 120 and the piezoelectric layer 140.

[0012] Electrodes 150A and 150B respectively comprise fingers 152A and 152B extending in the same direction D so as to form a periodic structure with period 2p in a direction perpendicular to said direction D, the fingers of the two electrodes being interleaved in a conventional manner.

[0013] The dielectric and metal layers interposed between the piezoelectric layer and its substrate improve the behavior of the transducer, and more particularly limit the appearance of parasitic responses, inductive losses associated with the properties of the substrate, and interface effects within the stack.

[0014] However, the information disclosed by US Pat. No. 10,938,367 (B2) remains insufficient for practical applications requiring special features and / or high performance levels for SAW transducers. Summary of the Invention

[0015] One objective of the present invention is to characterize surface wave devices in a manner that goes beyond the simple operating principles of the prior art and provides the surface wave devices with sufficient operating parameters to implement surface wave devices in practical applications.

[0016] To this end, the invention relates to a surface acoustic wave device comprising a substrate, a piezoelectric layer above a top surface of the substrate, a pair of electrodes in contact with the piezoelectric layer, the two electrodes having fingers extending in the same direction so as to form a periodic structure, the fingers of the two electrodes being arranged alternately with respect to one another and with an inter-finger distance separating the centers of two adjacent fingers of the same electrode, a metal layer interposed between the substrate and the piezoelectric layer, and at least one dielectric layer interposed between the metal layer and the piezoelectric layer, the metal layer having a thickness between 5 nm and 100 nm and the dielectric layer(s) having a thickness between 25 nm and 600 nm.

[0017] Such devices are characterized by the phase velocity, reflection coefficient and electromechanical coupling coefficient k S 2 This represents a culmination of compromise suitable for implementation as an acoustic wave device that benefits from the positive electromagnetic shielding effect of a metal layer interposed between the piezoelectric layer and its substrate while maintaining excellent performance with respect to

[0018] According to another non-limiting feature of the present invention, either individually or in any technically feasible combination, the metal layer may have a thickness of 0.25% to 5% of the inter-finger distance; the dielectric layer may have a thickness of 250 nm to 400 nm; the dielectric layer may have a thickness greater than 5 times the thickness of the metal layer; the dielectric layer may be less than 200 nm thick, - an optional dielectric layer may be interposed between the substrate and the metal layer; the dielectric layer and the optional dielectric layer may each be less than 300 nm thick; the sum of the thickness of the dielectric layer and the thickness of the optional dielectric layer is less than 200 nm; the ratio of the thickness of the optional dielectric layer to the sum of the thickness of the dielectric layer and the thickness of the optional dielectric layer may be between 15% and 30%; -The piezoelectric layer is made of lithium tantalate (LiTaO) 3 Layer and lithium niobate LiNbO 3 and may include juxtaposition with a layer, The metal layer includes metallized surfaces separated by a distance equal to or less than the inter-finger distance.

[0019] The present invention extends to filter devices, including surface acoustic wave devices.

[0020] The invention further relates to a method of manufacturing a device comprising a direct bonding step. [Brief description of the drawings]

[0021] Other features and advantages of the present invention will become apparent from the following detailed description of the invention which refers to the accompanying drawings. [Figure 1] 1A and 1B show schematic plan and cross-sectional views of a known inter-finger SAW transducer; [Diagram 2] 1A and 1B show schematic plan and cross-sectional views of a SAW transducer between fingers according to the present invention; [Diagram 3] 1 includes two graphs showing the effect of a metal layer on a SAW transducer between the fingers. [Figure 4] 3 is four graphs showing characteristic quantities of the transducer of FIG. 2 as a function of the thickness of the metal layer 220. [Diagram 5]3 is four graphs showing the conductance and admittance, respectively, of a structure similar to that of FIG. 2 as a function of the opposing metal surface; [Figure 6] 3 includes six graphs showing harmonic admittance for different thicknesses of the dielectric layer of the transducer of FIG. 2. [Figure 7] 2 includes four graphs showing the characteristics of the transducer of FIG. 2 as a function of the thickness of the dielectric layer 230. [Figure 8] 3 shows an alternative to the transducer structure of FIG. 2 with an optional dielectric layer 330. [Figure 9] 9 shows an acoustic wave device with two transducers corresponding to the transducers of FIG. 2 or FIG. 8. [Figure 10] 9 is six graphs showing harmonic admittance for different combinations of thicknesses of the dielectric layers 230 and 330 of the transducer of FIG. 8. [Figure 11] A method for manufacturing the devices of FIGS. [Figure 12] 1 shows a discontinuous metal layer. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] The inventors of the present invention started with a typical inter-finger SAW transducer structure and performed extensive digital modeling to determine the specific parameters that critically affect the performance of inter-finger SAW transducers and to define design rules for such transducers.

[0023] Structure 1: Single dielectric layer FIG. 2 shows A) a plan view of a transducer 200 between SAW fingers according to the invention, and B) a cross-sectional view of this transducer in a plane perpendicular to the plan view and passing through line segment YY'.

[0024] The SAW transducer 200 between the fingers includes a piezoelectric layer 240 placed on a substrate 210, a pair of electrodes consisting of a first electrode 250A and a second electrode 250B, which are in contact with the upper surface of the piezoelectric layer 240 so as to interpose the piezoelectric layer 240 between the pair of electrodes and the substrate 210, a metal layer 220 interposed between the substrate 210 and the piezoelectric layer 240, and a dielectric layer 230 interposed between the metal layer 220 and the piezoelectric layer 240.

[0025] In this example, the piezoelectric layer 240 is in direct contact with the dielectric layer 230 , which is in direct contact with the metal layer 220 , which is in direct contact with the substrate 210 .

[0026] Electrode 250A and electrode 250B each have fingers 252A and fingers 252B extending in the same direction D so as to form a periodic structure with period p in a direction perpendicular to direction D, the fingers of the two electrodes being alternated, as can be seen particularly in B), so as to form a conventional inter-finger structure.

[0027] The acoustic wavelength λ of the operating mode is equal to the period 2p, and the transducer operates under the Bragg condition (the electrode period p of the transducer is half the wavelength λ) for this particular wavelength, but is not limited to this.

[0028] This period 2p is understood as the distance separating the central extension axes of two adjacent fingers of the same electrode, ie the axes forming the axis of symmetry of the corresponding finger, said axes being parallel to the extension direction D of the fingers.

[0029] Substrate 210 is preferably made of silicon, more preferably high acoustic quality silicon, but may also be made of, for example, glass or ceramic or another semiconductor material.

[0030] Preferably, metal layer 220 and electrodes 250A and 250B are independently made from a light metal that is considered to be a good electrical conductor, such as aluminum or an aluminum alloy such as Al-Cu, Al-Si, or Al-Ti, to limit mass loading effects and resistive losses on the frequency response of the transducer.

[0031] However, due to its relatively low melting point, aluminum is particularly suitable for use in alloys such as LiTaO 3 This can lead to problems during transducer manufacturing when a heat treatment is applied above this melting point to allow the piezoelectric layer to recover its piezoelectric properties.

[0032] In such circumstances, metals heavier than aluminum, i.e. molybdenum, tungsten, platinum or titanium, but also chromium, copper and nickel can be used despite their negative impact on the loading effect, or alternatively, scandium and vanadium or even conductive carbon, taking into account their favourable density.

[0033] It is understood that metal layer 220 may be composed of a different material than the materials that compose electrodes 250A and 250B. Additionally, the metal layer may optionally be in contact with a fixed potential, such as ground, which requires an additional manufacturing step but makes the layer less sensitive to environmental RF signals.

[0034] The dielectric layer 230 is made of silica, preferably silicon dioxide, SiO 2 The dielectric material is made of a material such as

[0035] ZrO 2 , Ta 2 O 5 , Si 3 N 4 and other materials such as combinations of these materials may also be considered.

[0036] SiO 2 and Si 3 N 4A graded combination with is also possible in the form of SiON for silicon oxynitride.

[0037] These dielectric materials can be in any crystalline form (eg, polycrystalline or amorphous) as may be obtained using standard microelectronic deposition techniques.

[0038] The piezoelectric layer 240 is preferably made of lithium tantalate, LiTaO 3 , lithium niobate LiNbO 3 , or lithium tantalate LiTaO 3 Layer of lithium niobate (LiNbO) 3 The piezoelectric layer 240 may also be formed using potassium niobate, gallium nitride, aluminum nitride, zinc oxide or quartz, or any other piezoelectric material.

[0039] The distance between two corresponding portions of a first finger 252A and an adjacent second finger 252B defines an electrode period p of the SAW transducer 200 between the fingers.

[0040] The ratio between the width a of the fingers 252 A and 252 B and the electrode period p defines the metallization ratio M of the transducer 200 .

[0041] Together, the electrode period p and metallization ratio M characterize the SAW transducer 200 between the fingers and can be determined along with other factors such as the properties of the piezoelectric layer 240, the dielectric layer 230, the metal layer 220, or the substrate 210, the electrode thickness h (or its relative form h / 2p), and the operating parameters of the SAW resonator.

[0042] During operation of the SAW resonator, an alternating electrical input signal applied to the first electrode 250A is converted to a mechanical signal within the piezoelectric layer 240, generating one or more acoustic waves therein.

[0043] The resulting acoustic waves converted from an electrical input signal are primarily surface acoustic waves as desired in the operation of a SAW transducer.

[0044] The amplitude and phase of the acoustic waves thus generated in the piezoelectric layer depend on the frequency of the AC input signal, the electrode period p, the relative metal thickness h / 2p, and the metallization ratio M, as well as the operating parameters of the SAW resonator 200.

[0045] This frequency dependence is often described in terms of changes in harmonic admittance, ie, harmonic conductance and susceptance, between the first electrode 252A and the second electrode 252B that vary with the frequency of the AC electrical input signal.

[0046] The acoustic waves converted from the input AC electrical signal travel through the piezoelectric layer 240 and eventually reach the second electrode 252B where they are converted into an output AC electrical signal.

[0047] Furthermore, the acoustic waves may remain confined beneath the transducer between the fingers if the transducer between the fingers consists of a network of electrodes similar to those of the transducer, preferably but not necessarily connected to electrical ground, with a mechanical period close to or identical to that of the transducer, surrounded by reflecting mirrors that reflect the incident waves in phase towards the transducer, forming a surface wave resonator. This resonator may form the basic element of a so-called "impedance element" filter, with poles and zeros combined to form the desired transfer function.

[0048] Based on the general structure disclosed above, the inventors performed a number of numerical models to determine the parameters that should be set in order to obtain a SAW transducer between the fingers that meets precise specifications, particularly the thicknesses of the metal layer 220 and the dielectric layer 230.

[0049] Other parameters such as phase velocity, reflection coefficient, dielectric constant or electromechanical coupling follow empirical laws based on the thickness of the dielectric layer between the piezoelectric layer and the metal layer according to Equation 1 below:

[0050]

number

[0051] Coefficient a 0 ~a 3 and b 0 ~b 2 depends on the operating frequency, the materials selected, and the other thicknesses of this stack apart from the thickness of the dielectric layer present between the piezoelectric and metal layers.

[0052] Since the thicknesses of metal layer 220 and dielectric layer 230 are at least approximately determined, one skilled in the art would know how to select and adjust the thicknesses and materials (including their respective orientations) to optimize the operation of the device based on its technical specifications.

[0053] Model 1: Metal layer, target An initial series of models was used to verify the validity of this general structure, and in particular the advantages and potential disadvantages of metal layer 220.

[0054] For this purpose, two structures with SAW electrodes between the fingers were compared by modeling: one was a conventional structure with a piezoelectric / silicon oxide / silicon stack, and the other was a structure similar to the conventional "metal layer structure", but where a metal layer was introduced at the interface between the silicon and silicon oxide, resulting in a piezoelectric / silicon oxide / metal / silicon stack.

[0055] This metal layer structure is shown in FIG. 2B), the conventional structure corresponding to that of FIG. 2 with the metal layer 220 omitted.

[0056] More specifically, the conventional structure consists of a silicon wafer with a residual conductivity of m-1 and 3 oriented (100) and 100S, through a 500-nm-thick layer of silicon oxide (modeled as fused quartz) corresponding to layers 230, 210, and 240 in FIG. 2, respectively, and a 600-nm-thick piezoelectric layer of LiTaO m-1 / (YXl) / 42°. 3 The electrodes were modeled as consisting of an infinite array of 2% copper-aluminum alloy electrodes with an electrode period p of 1 μm, a metallization ratio M of 50%, and a relative thickness h / λ (thickness normalized to the acoustic wavelength λ).

[0057] The metal layer structure is the same as the conventional structure, except that it further includes the metal layer 220 shown in FIG. 2, which consists of a 50% copper-aluminum alloy with a thickness of 50 nm between the silicon substrate and the silicon oxide layer, and the structure corresponds to the cross-sectional view of FIG. 2.

[0058] The harmonic admittances of these two structures were calculated using the finite element method and the boundary element method, more specifically, the finite element method-boundary integral method (FEM-BIM).

[0059]

[0060] ​For the calculation method, reference can be made to the publications "Numerical Method for SAW propagation characterization" (P. Ventura, J. M. Dede, M. Solal, J. Desbois, J. Ribbe: Proc. of the IEEE Ultrasonics Symposium, pp. 175-186, 1998) and "Simulations of surface acoustic wave devices built on stratified media using a mixed finite element / boundary element integral formulation" (S. Ballandras, A. Reinhardt, V. Laude, A. Soufyane, S. Camou, W. Daniau, T. Pastureaud, W. Steichen, R. Lardat, M. Solal, P. Ventura: Journal of Applied Physics, vol. 96, No. 12, pp. 7731-7741, 2004).

[0061] Figure 3 shows the harmonic conductance and susceptance (respectively in dB / S on the vertical axis on the left side of each graph) of A) the conventional structure and B) the metal layer structure. m-1 The solid line G is harmo , and S on the vertical axis on the right side of each graph. m-1 The dotted line B harmo susceptance).

[0062] In conventional structures, the conductance G is always greater than 0, and the main and parasitic resonances are seen to have finite G values, as opposed to true (lossless) modes, where G can be likened to a Dirac function.

[0063] The losses reflected by the finite conductance (less than -150 dB in Fig. 3) are due only to parasitic conductance effects in the silicon substrate. The electric field associated with the mechanical displacement penetrates the structure to a depth of more than 2λ (typical values ​​can reach more than 10λ for pure shear waves), thus generating leakage currents that capture the excitation energy provided by the input signal and convert it into heat.

[0064] Indeed, it may be demonstrated that silicon substrates, even with their high resistivity, have charge carriers with a sufficiently long lifetime to induce the creation of parasitic capacitance at the interface between the silicon substrate 210 and the silicon oxide dielectric layer 230.

[0065] FIG. 3 shows in B) that the introduction of a metal layer 220 between the silicon substrate 210 and the silicon oxide dielectric layer 230 keeps the harmonic conductance values ​​below the noise level until the frequency reaches the so-called Surface Skimming Bulk Wave (SSBW) branch point, where the acoustic wave is no longer confined to the upper layers of the structure but penetrates the volume of the silicon substrate in the form of a radiation wave (the model assumes a silicon volume that is unlimited with depth, see FIG. 2B).

[0066] The reason for the harmonic conductance attenuation is that the metal layer forms an electromagnetic shield between the piezoelectric layer and the substrate, eliminating any electric field penetration into the silicon substrate volume and thus suppressing the appearance of parasitic conductance.

[0067] This minimizes or even cancels the mean free path of the charges and significantly limits or even prevents the existence of leakage currents due to coherent charge propagation at the interface.

[0068] Although the suppression of leakage by a metal layer is effective for any type of metal and any thickness of metal layer, it is preferable to limit this thickness in order to avoid the appearance of new modes that would degrade the spectral purity of the structure confining the surface acoustic waves.

[0069] In addition, using high density metals and increasing the thickness of the metal layer causes a mass loading effect of the layer, slowing down the acoustic wave propagation velocity and modifying the frequency response of the transducer.

[0070] Therefore, in order to limit the drawbacks mentioned in the previous two paragraphs, it is recommended to use for the metal layer (i) light metals, in particular aluminum or alloys with more than 90% aluminum, such as Al-Cu, Al-Si or Al-Ti, and (ii) a metal thickness of 5 nm to 100 nm, even 5 nm to 10 nm.

[0071] Other modeling studies have shown that the presence of a metal layer does not significantly change the velocity of the mode and its ability to be reflected by the structure, and that the electromechanical coupling coefficient k s 2 It should be noted that was found to be only slightly lower for the metal layer than for the structure without the metal layer.

[0072] It should also be noted that metal layer 220 is not limited to a continuous layer, but may consist of a pattern of separate metallized surfaces that provide the same shielding role as a continuous layer, but allow mass loading effects to be limited.

[0073] In this regard, the distance d separating two adjacent metallization surfaces should ideally be less than λ, preferably less than λ / 2, more preferably less than λ / 4, as shown in Figure 12, and the metal layer 220 consists of a two-dimensional matrix of metallization surfaces 222 separated from each other by a distance d. This condition is by no means limiting, but it helps to reduce the mean free path of charges induced at the oxide / silicon interface (or more generally at the dielectric / semiconductor interface).

[0074] Without limitation, any kind of geometric shape and distribution (square, triangle, hexagon, etc.) can be used for the elements of this matrix.

[0075] Model 2: Metallic layer and surface waves Using the second model series, the influence of the thickness of the metal layer 220 on the characteristic quantities of surface waves and permittivity was evaluated.

[0076] For this purpose, the inventors examined a structure similar to that shown in FIG. 2 corresponding to the metal layer structure in the first model series, except that this time the metal layer consisted of a titanium layer having a thickness varying between 1 nm and 50 nm.

[0077] FIG. 4 shows four graphs showing the evolution of characteristic surface wave quantities and permittivity in such a structure as a function of the thickness of the metal layer shown on the abscissa varying from 1 nm to 50 nm, where A) is in m.s -1 represented and having a phase velocity varying between 4036 and 4048, B) is represented as a percentage and having a reflection coefficient varying between 8.86 and 9.06, C) is represented as a percentage and having an electromechanical coupling coefficient k 2 s, D) is dimensionless and having a relative permittivity varying between 48.499 and 48.5006.

[0078] These graphs show that increasing the thickness of the metal layer causes only a very slight deterioration in the magnitude characteristics of the surface waves (phase velocity, reflection coefficient, electromechanical coupling coefficient).

[0079] The relative permeability is hardly affected by this increase, with a variation of less than 1%, and the variations in the reflection coefficient and the electromechanical coupling coefficient remain on the order of 1%, so all these variations can be considered negligible.

[0080] Regarding the phase velocity, the mass loading effect of the metal layer is more significant, with a variation of about -2 m.s -1 .nm -1 or -50 ppm.nm -1 accompanied.

[0081] From these findings it can be deduced that the thickness of the metal layer has an effect and that it is advantageous to limit this effect by keeping the thickness of the metal layer below 100 nm, even below 5% of the acoustic wavelength λ, more preferably below 2.5%.

[0082] However, in order to maintain the electromagnetic shielding effect of the metal layer, it is preferable that the thickness of the metal layer is more than 0.25% of the acoustic wavelength λ.

[0083] Model 3: Metal Layers and Parasitic Capacitances In the third series of models, a model based on Green's functions (detailed in A. Reinhardt, "Simulation, concept et realisation de filtres a ondes de volume dans des couches minces piezoelectriques" [Simulation, design and realisation of bulk-wave filters in piezoelectric thin films] (Doctoral Thesis at the Universite de Franche-Comte in Engineering Sciences, 2005)) was used to model bulk-wave filters on a piezoelectric thin film with 100 nm of aluminum, 600 nm of LiTaO 3 , 500 nm SiO 2 We calculated the admittance, i.e., conductance and susceptance, of a stack of materials including (modeled as fused quartz), 50 nm of molybdenum, 650 μm of silicon, and a semi-infinite polyimide support (Kapton) to attenuate radiation waves from the surface of the stack to the core.

[0084] For the calculation, the quality factor Q of the metal layer is set to 100, and the SiO 2 About 1000, LiTo 3 and for silicon it was set to 10,000. This quality factor is the acoustic constant C ij is used to calculate the imaginary part of (the non-conserved part of the problem) as imaginary part(Cij) = real part(Cij) / Q.

[0085] For the electrical aspect, the loss angle tg(δ) is calculated by the following equation: 2 About 10 -2 , 10 for silicon -3 , LiTaO 3 About 10 -4 As mentioned above, this coefficient allows the imaginary parts of the dielectric constants to be calculated from their real values, according to processes well known to those skilled in the art.

[0086] The aluminum layer is configured to represent the metallization of a SAW filter in the frequency range of 1-2 GHz, i.e. 1 mm 2 It was believed to occupy the following surface area:

[0087] FIG. 5 shows four graphs A) to D), each of which is 500 μm 2 , 250μm 2 , 100μm 2 , and 50 μm 2 The conductance G, admittance B, and capacitance C of a structure similar to that of Figure 2, corresponding to the aluminum surface of 0 Shows.

[0088] In addition to demonstrating the High Overtone Bulk Acoustic Resonator (HBAR) effect above 3 GHz, the capacitance C formed by the opposing conductive elements (metal layer 120 and electrodes 150A and 150B) and the dielectric layer 130 separating them is 0 We also note the linearity of the effect on the conductance of

[0089] Therefore, to limit the parasitic capacitance, the metallization of the piezoelectric layer is, for example, 10 times smaller than the surface area occupied by the layers defining the resonator electrodes and their respective connection pads. 4 μm 2 should be limited by restricting it to a value less than

[0090] Preferably, the cumulative surface area of ​​the connection pads alone is 10 4 μm 2 should remain less than 100 nm, with each connection pad defined as an area of ​​the layer defining an electrode forming a rectangle of sufficient surface area designed to allow microwelding by wedge bonding and ball bonding, respectively.

[0091] Dielectric Layer First, it should be noted that the presence of the dielectric layer 230 is necessary to reduce the High Overtone Bulk Acoustic Resonator (HBAR) effect due to the presence of the metal layer 220, thereby reducing the parasitic response induced in the SAW resonator 200 between the fingers.

[0092] Model 4: Dielectric layers and admittance Furthermore, in the metal layer structure shown in FIG. 2, a fourth model series evaluated the effect of the thickness of the dielectric layer 230 interposed between the piezoelectric layer 240 and the metal layer 220 on the characteristics of the SAW resonator 200 between the fingers.

[0093] For this evaluation, we used a 1000-mV GaN-based ... 2 We considered structures similar to the one shown in FIG. 2, which correspond to the metal layer structure in the first model series, except that the layers were made of 10 ...

[0094] For the calculation, LiTaO 3 10,000 for the SiO layer and silicon substrate 2 A quality factor Q of 1,000 for ZnO and 100 for molybdenum were considered.

[0095] FIG. 6 shows the results of the measurements of SiO2 at 0 nm, 20 nm, 50 nm, 100 nm, 200 nm, and 500 nm on the graphs shown in A) to F), respectively. 2For thicknesses of 100 mm, the coefficients of harmonic admittance and susceptance (in dB / S.μm on the vertical axis to the left of each graph) are plotted as a function of frequency shown on the horizontal axis. -1 The solid line harmonic coefficients |Y| harmo and Sm on the vertical axis on the right side of each graph -1 The dotted line susceptance B is represented by harmo ) are shown in six graphs.

[0096] From these graphs, the distance between the resonant frequency (maximum harmonic conductance) and the anti-resonant frequency (minimum harmonic conductance) is SiO 2 It can be seen that the thickness of

[0097] Both frequencies increase as the mass effect decreases and the coupling between the metal and piezoelectric layers decreases with increasing SiO 2 decreases as the thickness of

[0098] Therefore, SiO 2 It may be advantageous to maximize the layer thickness.

[0099] Parasitic responses near 2.7 GHz are seen in graphs E) and F), but not in graphs A) to D), as seen for SiO greater than 200 nm. 2 It can also be seen that the .lambda.

[0100] Therefore, in order to avoid the appearance of such parasitic responses, it may be advantageous to use an oxide layer having a thickness of less than 200 nm, for example between 50 nm and 200 nm, preferably between 100 nm and 200 nm.

[0101] Model 5: Dielectric layers and surface waves A fifth series of models was used to evaluate the effect of the thickness of the dielectric layer 230 on the surface wave and dielectric constant properties.

[0102] For this evaluation, the inventors considered structures similar to those shown in FIG. 2, which correspond to the metal layer structures in the first model series, except that aluminum alloy Al-Cu with 2% copper, titanium Ti, molybdenum Mo, copper Cu and nickel Ni were used for the metal layers and the thickness of the dielectric layer varied between 0 and 400 nm.

[0103] FIG. 7 shows four graphs illustrating the evolution of the characteristic surface wave volume and dielectric permittivity in such a structure as a function of the thickness of the dielectric layer shown on the abscissa, which varies from 0 nm to 0.4 μm; A) in ms -1 A) the phase velocity, expressed as a percentage, which varies between 4050 and 4140; B) the reflection coefficient, expressed as a percentage, which varies between 8 and 9.6; C) the electromechanical coupling coefficient, k, expressed as a percentage, which varies between 6.5 and 10. 2 s, D) is the relative dielectric permittivity, which is dimensionless and varies between 51 and 48.5.

[0104] Each graph contains five curves, each corresponding to the nature of the metal layer under consideration (Al-Cu, Ti, Mo, Cu, Ni).

[0105] From these graphs, the phase velocity, reflection coefficient, and electromechanical coupling coefficient k S 2 are SiO 2 less than 600 nm and SiO 2 less than 400 nm, respectively. 2 For thickness, in particular for the phase velocity, for a thickness of about 50 nm, for the reflection coefficient, for a thickness of 80 nm to 180 nm depending on the properties of the metal layer, the electromechanical coupling coefficient k S 2 It can be seen that the maximum value is observed for a thickness of 250 nm to 400 nm.

[0106] Therefore, SiO 2 To balance the interest in maximizing thickness with the interest in maximizing one or more of the parameters mentioned in the previous paragraph, a SiO thickness of 25 nm to 2 μm, preferably 50 nm to 600 nm, more preferably 50 nm to 400 nm, and even more preferably 250 nm to 400 nm, is preferred.2 It is advantageous to select the thickness.

[0107] Alternatively, SiO 2 The preferred value of the thickness can be expressed relative to the thickness of the metal layer, in which case the dielectric layer thickness 230 can be advantageously selected to be 0.5 to 40 times, preferably 10 to 40 times, and even more preferably 20 to 40 times, the thickness of the metal layer 220.

[0108] Furthermore, for example, SiO of 250 nm to 400 nm 2 By selecting the thickness of the electromechanical coupling k S 2 To maximize the 2 or 80 nm to 180 nm SiO to maximize the phase velocity. 2 One can also try to take advantage of the existence of a maximum in order to maximize the reflection coefficient by choosing the thickness of

[0109] Structure 2: Double dielectric layer An alternative to placing the metal layer 220 in direct contact with the substrate 210 is to interpose an optional dielectric layer 330 therebetween, to obtain the layered structure shown in FIG. 8, known as the “optional dielectric layer structure”.

[0110] The mass loading effect is associated with the increase in mass per unit volume through which the wave propagates.

[0111] The wave can be considered to be more sensitive to the properties of the medium as it coincides with the location of its maximum energy density.

[0112] Therefore, the heavier the metal layer is (e.g., 10.22 g.cm for a molybdenum metal layer), the greater the -3 vs. 2.65g.cm for silica dielectric layer -3 ), the closer to the piezoelectric layer where the wave energy maximum is located, the slower the phase velocity becomes due to the mass loading effect mentioned above.

[0113] The removal of this high density metal layer from the piezoelectric layer by the presence of a less dense dielectric layer actually results in an increase in phase velocity.

[0114] In the case of the optional dielectric layers, the mass loading effect naturally takes into account the two dielectric layers (e.g., silica) always present in the stack, whose sum of thicknesses is constant from calculation to calculation; only the position of the metal layer changes, and therefore its effect on the phase velocity changes as well.

[0115] For elements common to Figures 2 and 8, please refer to the previous description.

[0116] Model 6 For this optional dielectric layer structure shown in FIG. 8, a sixth series of models was used to evaluate the effect of the thickness of the dielectric layer 230 interposed between the piezoelectric layer 240 and the metal layer 220, and the thickness of the optional dielectric layer 330 interposed between the substrate 210 and the metal layer 220, on the characteristics of the SAW transducer 200 between the fingers.

[0117] For this evaluation, the metal layer was now made of molybdenum, the optional dielectric layer 330 was present, and the dielectric layer 230 and the optional dielectric layer 330 were each silicon oxide SiO 2 of varying thickness. 2 We considered a structure similar to that shown in FIG. 8, which corresponds to the metal layer structure of Model Series 1, except that it is made of layers of

[0118] For the calculation, LiTaO 3 Quality factors Q of 10,000 for the layer and silicon substrate, 1,000 for SiO2 and 100 for molybdenum were considered.

[0119] FIG. 10 shows the results of the measurements of 0 nm, 20 nm, 50 nm, 100 nm, 200 nm, and 500 nm of SiO for the dielectric layer 230 on the graphs designated A) through F), respectively. 2and optional dielectric layer 330 thicknesses of 520 nm, 500 nm, 470 nm, 420 nm, 320 nm, and 20 nm SiO 2 For each thickness, the harmonic admittance (dB / S.μm on the vertical axis on the left side of each graph) -1 The solid line G is harmo Conductance and Sm on the vertical axis on the right side of each graph -1 The dashed line B is harmo The six graphs show the evolution of the susceptance.

[0120] It can be seen that the combined effect of the two oxide layers 230 and oxide layer 330 gives results similar to those of a single oxide layer, possibly including the effect of the thicknesses of these two layers.

[0121] For example, it may be advantageous to interpose dielectric layer 230 and optional dielectric layer 330 each having a thickness of less than 300 nm, preferably less than 100 nm.

[0122] It is also possible for these layers 230 and 330 to be interleaved such that the sum of their respective thicknesses remains less than 600 nm.

[0123] Furthermore, one of the parasitic responses disappears, corresponding to the second mode at about 2.7 GHz in the structure shown in graph D).

[0124] Therefore, in order to eliminate this second mode, it is advantageous to select the thickness of the oxide layer such that the ratio of the thickness of the optional dielectric layer to the sum of the thickness of the dielectric layer and the thickness of the optional dielectric layer may be between 15% and 30%.

[0125] Surface Acoustic Wave Devices and Filters An application of the surface wave device according to the present invention is a filtering apparatus including a pair of transducers as shown in FIG.

[0126] One skilled in the art can combine such resonators to obtain filtering functions.

[0127] FIG. 9 is a non-limiting depiction of a quadrupole device 900 incorporating two SAW transducers 200_1 and 200_2, such as those shown in FIG. 2 or FIG. 8, in contact with the same piezoelectric layer 940 supported by a common substrate.

[0128] As mentioned above, one of the two SAW transducers is an input SAW transducer and the other is an output SAW transducer.

[0129] Manufacturing method The devices shown in Figures 2 and 8 can be fabricated using methods that each include one or more direct bonding steps, in particular molecular bonding, between two of the layers that make up the device structure.

[0130] Thus, FIG. 11 shows a method 1100 for manufacturing any of the devices shown in FIGS. 2 and 8 by direct bonding 1150 of two elements A and B, where A can be the dielectric layer 230 and B the metal layer 220, A can be the piezoelectric layer 240 and B the dielectric layer 230, A can be the metal layer 220 and B the substrate 210, A can be the metal layer 220 and B the optional dielectric layer 330, A can be the optional dielectric layer 330 and B the substrate 210.

[0131] Each of element A and element B may be an individual element or may already be bonded to other layers, for example a first assembly including piezoelectric layer 240 and dielectric layer 230 (the latter representing element A) may be bonded to a second assembly including substrate 210 and metal layer 220 (the latter representing element B).

[0132] Of course, the invention is not limited to the disclosure herein and modifications may be made without departing from the scope of the invention as defined by the claims.

Claims

1. A surface acoustic wave device (200), The device described above, - Substrate (210) and, - The piezoelectric layer (240) above the upper surface of the substrate (210), - A pair of electrodes (250A, 250B) in contact with the piezoelectric layer, wherein the two electrodes have fingers (252A, 252B) extending in the same direction (D) to form a periodic structure, the fingers of the two electrodes are arranged alternately with respect to each other, and the pair of electrodes (250A, 250B) has an inter-finger distance (2 × p) that separates the centers of two adjacent fingers of the same electrode, - A metal layer (220) interposed between the substrate and the piezoelectric layer, - comprising at least one dielectric layer (230) interposed between the metal layer and the piezoelectric layer, - The metal layer (220) has a thickness of 5 to 100 nm. - A surface acoustic wave device (200) characterized in that the at least one dielectric layer (230) has a thickness of 25 nm to 600 nm.

2. The surface acoustic wave device according to claim 1, wherein the metal layer has a thickness of 0.25% to 5% of the finger-to-finger distance.

3. The surface acoustic wave device according to claim 1 or 2, wherein the dielectric layer has a thickness of 250 nm to 400 nm.

4. The surface acoustic wave device according to claim 1 or 2, wherein the dielectric layer has a thickness of more than five times the thickness of the metal layer.

5. The surface acoustic wave device according to claim 1 or 2, wherein the dielectric layer has a thickness of less than 200 nm.

6. The surface acoustic wave device according to claim 1 or 2, further comprising an optional dielectric layer (330) interposed between the substrate (210) and the metal layer (220).

7. The surface acoustic wave device according to claim 6, wherein the dielectric layer (230) and the optional dielectric layer (330) each have a thickness of less than 300 nm.

8. The surface acoustic wave device according to claim 6, wherein the sum of the thickness of the dielectric layer (230) and the thickness of the optional dielectric layer (330) is less than 200 nm.

9. The surface acoustic wave device according to claim 6, wherein the ratio of the thickness of the optional dielectric layer (330) to the sum of the thickness of the dielectric layer (230) and the thickness of the optional dielectric layer (330) is 15% to 30%.

10. The piezoelectric layer is lithium tantalate LiTaO 3 Layer and lithium niobate LiNbO 3 A surface acoustic wave device according to claim 1 or 2, including juxtaposition with layers.

11. The surface acoustic wave device according to claim 1 or 2, wherein the metal layer (220) includes a metallized surface separated by a distance less than or equal to the finger distance.

12. A surface acoustic wave filter device (900) comprising the surface acoustic wave device according to claim 1 or 2.

13. A method (1100) for manufacturing the device according to claim 1 or 2, comprising a direct bonding step (1150).

Images