The invention relates to a micromechanical sensor using optical detection of movement. The invention is in particular but not exclusively applicable to atomic force microscopy.
Atomic force microscopy (AFM) is a scanning microscopy technique that was developed in the 80s and that allows atomic-scale resolution to be achieved. Contrary to tunneling scanning microscopy, atomic force microscopy is not limited to the formation of images of conductive surfaces, this making it suitable for insulators, semiconductors, or even for samples of biological nature. This technique is applicable to many fields of pure and applied research, but also to the microelectronics industry. A general introduction to the principles of AFM is provided in the article by F. J. Giessibl and C. F. Quate “Exploring the nanoworld with atomic force microscopy”, Physics Today, December 2006, pages 44-50.
The essential component of a conventional atomic force microscope is a probe consisting of a lever embedded at one end and provided, at the opposite end, with a tip that is oriented toward the surface of the sample to be observed. The lever generally has a length of about a few tens or hundreds of microns, and the tip a radius of curvature of a few tens or hundreds of nanometers. Such a probe, which is generally made of silicon nitride or single-crystal silicon, may be manufactured using conventional photolithography techniques, at a low cost. When the tip of the probe is brought close to a surface, it experiences the influence of attractive or repulsive forces that are of chemical nature and/or that are van der Waals, electrostatic and/or magnetic forces. By measuring these forces while the tip scans the surface of the sample to be observed, it is possible to reconstruct an image of the latter. The forces exerted between the tip and the sample may be measured in various ways. In the simplest and oldest technique (static AFM) the deflection of the embedded lever holding the tip is merely observed, in particular by optical means.
A better sensitivity may be obtained by making this lever vibrate in one of its flexural eigenmodes, and by observing the variations in the resonant frequency that are induced by the gradients of these forces (dynamic AFM). In practice, the dynamic technique is generally preferred for observations carried out in vacuum or in air. This technique is less suitable for observations in liquid medium, because the vibrations of the lever are highly damped therein, thereby decreasing the quality factor of the probe.
It is also known to use AFM probes that exploit planar vibrational modes—“longitudinal” or “vertical movement” also being spoken of—which allow very high quality factors to be achieved even with dynamic AFM in viscous media.
Whatever the chosen geometry, it is necessary to produce a transducer that allows the movements of the AFM tip to be detected. Such a transducer must have both a high sensitivity and a high dynamic range, because the movements to be detected may have an amplitude ranging from less than one picometer to several hundred nanometers. This is relatively simple in the case of a conventional probe, of the embedded lever type operating in flexion, with which the deflection of a laser beam reflected by the lever may be exploited, but it is much less so in the case of a longitudinal probe. The same problem arises in the case of micromechanical sensors other than AFM probes, accelerometers for example.
The article by Y. Liu et al. “Wide cantilever stiffness range cavity optomechanical sensors for atomic force microscopy” Optics Express Vol. 20, No. 16, pp. 18268-18280 (2012) and patent U.S. Pat. No. 8,997,258 disclose AFM probes comprising a lever that is able to be brought close to an optical resonator of ring or disk type, perturbing the resonant conditions of the latter.
The operating principle of these transducers exploits an interaction between the lever and evanescent waves in proximity to the exterior surface of the optical resonator. In this configuration, only certain optical resonant modes of the resonator are modulated. The identification of the modes is therefore very difficult, because all thereof are not shifted in frequency by the same amount in response to the variation in the distance between the lever and the resonator. In addition, the areas that face are large: the device is highly sensitive to capillary forces (in particular in liquid) which may cause the lever to bond to the optical resonator.
Moreover, AFM probes (just like other micromechanical sensors such as styluses) are subjected to asymmetric stresses. Thus, in the case of a longitudinal AFM probe, the tip is liable to experience repulsive forces that are much stronger than the attractive forces—the limiting case being abutment. However, in the case of the aforementioned prior-art systems, this may lead to the lever and optical resonator making contact, which may damage these elements or cause adhesion thereof.
The invention aims to solve all or some of these drawbacks of the prior art.
According to the invention, this aim is achieved using an optical transducer based on a resonator of ring or disk type, having at least one interruption, or slit. A movable micromechanical element, such as a beam able to be animated with a longitudinal movement, is coupled to the resonator in such a way that a movement of the micromechanical element opens or closes the interruption. This results in a modification of the length of the resonator, and therefore of its resonant wavelength, whatever the mode in question; this modification may therefore easily be detected. The facing areas are small, above all in the case of a ring resonator having a single interruption, decreasing the risk of adhesion via the effect of capillary forces.
Advantageously, a device according to the invention and intended to operate in the presence of asymmetric stresses will be designed in such a way that the movements of larger amplitude of the movable micromechanical element open the one or more interruptions, decreasing the risk of damage to the resonator.
The interruption introduces losses into the resonator, and therefore decreases its quality factor. This drawback is however broadly compensated for by the mechanical advantages of the interruption, which advantages will be used in more detail below.
One subject of the invention is therefore a micromechanical sensor comprising a movable micromechanical element and an optical resonator of disk or ring type, characterized in that the optical resonator has at least one interruption; and in that the movable mechanical element is mechanically coupled to the optical resonator in such a way that a movement of the movable micromechanical element induces a modification of the width of the interruption of said optical resonator by moving at least one edge of said interruption in a direction substantially parallel to a direction of propagation of the light in the resonator at the interruption.
According to advantageous embodiments of such a sensor:
The movable micromechanical element may extend in what is called a longitudinal direction and have a degree of freedom in translation in said longitudinal direction. More particularly, the movable micromechanical element may be borne by a planar substrate having a main surface parallel to the longitudinal direction and protrude from one edge of said substrate by extending in what is called a positive orientation of said longitudinal direction. Even more particularly, the movable micromechanical element may be mechanically coupled to the optical resonator in such a way that its movement in the longitudinal direction with what is called a negative orientation, which is opposite to the positive orientation, induces an increase in the width of the interruption of the optical resonator. Moreover, the optical resonator may be borne by the planar substrate and be substantially aligned with the micromechanical element in the longitudinal direction.
The one or more interruptions of the optical resonator may be oriented in a direction, called the transverse direction, that is perpendicular to the longitudinal direction and separate the optical resonator into two separate portions, one of which is mechanically coupled to the movable micromechanical element whereas the other of which is anchored to the substrate.
The movable micromechanical element may be directly connected to said portion of the optical resonator.
As a variant, the movable micromechanical element may be coupled to said portion of the optical resonator by way of a movement-inverting mechanical structure.
Said resonator may be of ring type and have at least one interruption oriented in the longitudinal direction, the movable micromechanical element being mechanically connected at two points of said resonator that are located on either side of said interruption by way of a mechanical structure suitable for converting a movement in the longitudinal direction into a force in the transverse direction. More particularly, said mechanical structure suitable for converting a movement in the longitudinal direction into a force in the transverse direction may comprise a fork-shaped flexible structure having two branches connected to the two said points of the resonator that are located on either side of the interruption.
The one or more interruptions of the optical resonator may separate the latter into two separate portions, one of which is mechanically coupled to the movable micromechanical element whereas the other of which is anchored to the substrate.
The optical resonator may be is of ring type, have a single interruption and form, with the movable micromechanical element, a mass-spring system.
The movable micromechanical element may bear a local-probe microscopy tip extending in the longitudinal direction.
The micromechanical sensor may also comprise at least one optical waveguide optically coupled to the optical resonator.
In a rest state in which no external force acts on the movable micromechanical element, the width of the or each interruption of the optical resonator may be comprised between 0.1 nm and 1 μm, and preferably between 10 nm and 200 nm.
Other features, details and advantages of the invention will become apparent on reading the description given with reference to the appended drawings, which are given by way of example and which show, respectively:
The invention will be described with reference to the embodiments that all relate to probes for AFM, and more particularly for longitudinal AFM. However, it is not limited to this particular application.
The probes of
a silicon substrate, noted SS in the figures, in general of a thickness of 20 μm to 1 mm, and preferably of between 50 and 600 μm;
a buried oxide (SiO2) layer (referred to as a BOX for buried oxide), generally of a thickness comprised between 0.1 μm and 5 μm, and preferably of between 0.2 μm and 2 μm; and
a thin silicon layer (thickness generally comprised between 1 nm and 100 μm,
and preferably between 0.2 μpm and 5 μm), called the “device layer”.
The mechanical elements of the probe are produced from the device layer, and are released by wet etching the subjacent BOX layer. The BOX layer remains underneath the anchors that connect the probe to the substrate SS.
It is also possible to produce the probe from a material other than SOI, for example SiN deposited on Si. It is also possible to produce certain portions of the probe in the device layer on SOI and others in a material such as Si3N4 or any other material having suitable mechanical properties. In practice, any material having a sufficiently high ratio between Young's modulus E and density p (typically
may be suitable for implementing the invention. By way of example, the silicon
and the silicon carbide
Generally, the device layer is located above a surface of the substrate, directly or via interposition of intermediate layers (BOX in the case of an SOI structure).
The probe of
The rear end (i.e. the end opposite the tip PT1) of the micromechanical element EMM is connected to an optical resonator ROP1 of ring type. The term “ring” must be understood in the topological sense, and covers the case of a circular ring (as in
Moreover, the resonator is fastened, by a fastening point located on the side thereof opposite to the slit GR1, to the anchor APT1.
A planar optical waveguide GOP is arranged behind the resonator, in immediate proximity to the latter (the spacing GC between these two components is typically about 200 nm or less), so as to allow evanescent coupling. A light wave propagates from an optical port POP1 at one end of the guide GOP to an optical port POP2 located at the opposite end. The evanescent coupling through the interval GC excites an electromagnetic resonant mode of the resonator; some of the light flux injected into the port POP1 therefore does not reach the port POP2. Preferably, the optical resonator and the waveguide are single-mode
The resonator ROP1 and the waveguide GOP may for example be produced in the device layer itself, made of silicon, this implying the use of infrared radiation and a device layer having a suitable thickness (of about a number of hundred nanometers). As a variant, they may be produced from a dielectric deposited above the device layer.
Forces exerted by the surface of a sample on the tip PT1 cause a movement of the mechanical element EMM in the longitudinal direction (y); the latter deforms the ring resonator ROP1 thereby opening the interruption GR1 (if the movement occurs in the positive orientation of the y-axis) or closing it (if the movement occurs in the negative orientation of y). GR1 is partially opened or closed by moving its edges in a direction that is substantially parallel to the direction of propagation of the light in the resonator—i.e. a tangential direction in the case of a circular resonator; in the particular case of
It is advantageous to note that the interrupted optical resonator ROP1 also has a mechanical spring function. It contributes, just like the transverse beams ET1-ET4, to the stiffness of the probe in the longitudinal direction, and therefore to its mechanical resonant frequency. One advantage of the invention is that the interrupted optical resonator ROP1 may be dimensioned quite freely so as to increase or decrease its stiffness; it therefore places few constraints on the mechanical resonant frequency of the probe. This would not be the case for an uninterrupted resonator, which would be very stiff and which would highly constrain the mechanical resonant frequency of the probe.
In these three embodiments, the interruptions GR1, GR2, GR3 extend in the transverse direction x, but this is not is not strictly necessary.
The embodiments of
In the case of
In the embodiment of
In the embodiment of
The invention has been defined with respect to a certain number of embodiments, but is not limited to the latter. For example, it is possible to envision movement-inverting/converting structures different from those illustrated in
Number | Date | Country | Kind |
---|---|---|---|
1657760 | Aug 2016 | FR | national |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/EP2017/069477 | 8/1/2017 | WO | 00 |