Laser wavelength centerlocking using photonic integrated circuits.

Abstract

To provide an apparatus for locking a wavelength of a laser.SOLUTION: An apparatus 100 includes: an optical splitter configured to receive an optical signal and to split the input optical signal to provide a first optical signal and a second optical signal; and an interferometer 110 comprising a first arm 106 and a second arm 108. A portion of the first arm 106 is exposed to a reference gas that attenuates light of a characteristic wavelength. The apparatus further includes: an optical coupler 114 configured to receive an output optical signal from the first arm and an output optical signal from the second arm and to provide a third optical signal; and a photodetector configured to receive the third optical signal and to provide a photocurrent. The photocurrent increases when the difference between the characteristic wavelength and the wavelength of the optical signals increases. The apparatus also comprises a feedback control circuit 122 configured to change properties of a laser to be locked until an error signal indicative of the difference between the characteristic wavelength and the wavelength of the laser is substantially zero.SELECTED DRAWING: Figure 1

Description

[Background technology]

[0001] Wavelength stabilized lasers are used in a variety of applications. For example, such lasers can be used as reference lasers in interferometers used for either length or wavelength measurements using a scanning Michelson interferometer. In the absence of such stabilization, the wavelength of the signal output from the laser may drift during operation. This change, or drift, can be caused by one of several factors, such as a change in the temperature of the laser during operation. Regardless of the source of drift, maintaining the laser's accuracy requires continuous monitoring and adjustment of the laser to maintain its output at the desired wavelength and tolerance.

[0002] Helium-neon (He-Ne) gas lasers are often used as wavelength-stabilized lasers. However, He-Ne lasers are relatively expensive and have a relatively short lifetime, which further increases their cost of use. Other wavelength reference lasers available include tunable semiconductor distributed feedback (DFB) lasers. Wavelength control of the output of such lasers is often based on a dither free locking scheme called side-lock, which has high short-term stability. However, such lasers have lower absolute stability.

[0003] Alternatively, wavelength control of the output of such lasers is often based on a center-lock scheme that requires a small dithering of the actual wavelength. The dithering makes these lasers impractical for some applications because the dithering combined with the finite velocity of scanning creates side modes. Thus, these lasers are impractical for use in a scanning Michelson interferometer.

[0004] Other devices and methods for providing wavelength-stabilized lasers include the Pound-Drever-Hall approach, which avoids modulating the laser but requires expensive high-speed modulators, detectors and electronics.

[0005] Therefore, what is needed is a center-locked, wavelength-stabilized semiconductor laser that is relatively inexpensive and overcomes the shortcomings of known lasers such as those discussed above.

[0006] Exemplary embodiments are best understood from the following detailed description when read in conjunction with the accompanying drawing figures. It is emphasized that the various features are not necessarily drawn to scale. In fact, dimensions may be arbitrarily increased or decreased for clarity of discussion. Wherever appropriate and practical, like reference numerals refer to like elements. [Brief description of the drawings]

[0007] [Figure 1] 1 is a schematic diagram of a laser and an apparatus for locking the center wavelength of the laser according to a representative embodiment. [Figure 2A] 4 is a graph showing the optical transmittance of standard gases of an apparatus for locking the central wavelength of a laser in accordance with a representative embodiment. [Figure 2B] 4 is a graph showing the output of a photodiode used in an apparatus for locking the center wavelength of a laser in accordance with a representative embodiment. [Figure 2C] 4 is a graph illustrating the dependence of an error signal on the value of a phase shifter used in an apparatus for locking the center wavelength of a laser in accordance with a representative embodiment. [Diagram 3] 1 is a perspective view of a portion of a photonic integrated circuit used to lock the wavelength of a laser in accordance with a representative embodiment. [Figure 4] FIG. 1 is a perspective view of a slot optical waveguide for use in a photonic integrated circuit according to a representative embodiment. [Diagram 5]FIG. 2 is a perspective view of a housing package for an apparatus for locking the center wavelength of a laser according to a representative embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] In the following detailed description, for purposes of explanation and not limitation, example embodiments disclosing specific details are set forth to provide a thorough understanding of one embodiment according to the present teachings. However, it will be apparent to those skilled in the art having access to this disclosure that other embodiments according to the present teachings that depart from the specific details disclosed herein remain within the scope of the appended claims. Moreover, descriptions of known devices and methods may be omitted so as not to obscure the description of the example embodiments. Such methods and devices are clearly within the scope of the present teachings.

[0009] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. Defined terms have the technical and scientific meaning of the defined terms as well as those commonly understood and accepted in the art of the present teachings.

[0010] Unless otherwise specified, when we say that a first element (e.g., an optical waveguide) is connected to a second element (e.g., another optical waveguide), this includes the case where one or more intermediate elements or intervening devices can be used to connect the two elements to each other. On the other hand, when we say that a first element is directly connected to a second element, this only includes the case where the two elements are connected to each other without any intermediate or intervening devices. Similarly, when we say that a signal is coupled to an element, this includes the case where one or more intermediate elements can be used to couple the signal to the element. On the other hand, when we say that a signal is directly coupled to an element, this only includes the case where the signal is directly coupled to the element without any intermediate or intervening devices.

[0011] As used herein and in the appended claims, terms without a specified quantity include both singular and plural referents unless the context clearly indicates otherwise. Thus, for example, "device" includes one device and multiple devices. The terms "substantial" or "substantially" as used herein and in the appended claims mean, in addition to their ordinary meaning, within an acceptable limit or amount. The term "about" as used herein and in the appended claims means, in addition to their ordinary meaning, within an acceptable limit or amount to one of ordinary skill in the art. For example, "approximately the same" means that one of ordinary skill in the art would consider the items being compared to be the same.

[0012] Relative terms such as "above," "below," "top," and "bottom" may be used to describe the relationship of various elements to one another as depicted in the accompanying figures. These relative terms are intended to include different orientations of those elements in addition to the orientation depicted in the figures. For example, if a device depicted in a figure (e.g., a photonic IC) is viewed upside down, an element described as, for example, "above" another element would then be "below" that element. Similarly, if the device is viewed rotated 90 degrees, an element described as "above" or "below" another element would then be "adjacent" to that other element. Here, "adjacent" means either in contact with the other element or having one or more layers, materials, structures, etc. between the elements.

[0013] Various embodiments of an apparatus for locking the wavelength of a laser are disclosed below. In one exemplary embodiment, the apparatus comprises an optical splitter configured to receive an optical signal and split the input optical signal to provide a first optical signal and a second optical signal. The apparatus further comprises an interferometer comprising a first arm and a second arm, the first arm configured to receive the first optical signal and the second arm configured to receive the second optical signal. In particular, a portion of the first arm is exposed to a reference gas that attenuates light of a characteristic wavelength.

[0014] The apparatus further comprises an optical coupler configured to receive the output optical signal from the first arm and the output optical signal from the second arm to provide a third optical signal, and a photodetector configured to receive the third optical signal to provide a photocurrent. The photocurrent increases as the difference between the characteristic wavelength and the wavelength of the optical signal increases. The apparatus also comprises a feedback control circuit configured to modify a characteristic of the laser until an error signal indicating the difference between the characteristic wavelength and the wavelength of the laser to be locked (the "wavelength difference") is substantially zero. In one exemplary embodiment, "substantially zero" refers to the error signal being 0±0.01. However, as will be appreciated from the present teachings, the wavelength difference is desirably zero. That is, ideally, the error signal is zero if and only if the wavelength difference is zero. The scaling between the error signal and the wavelength difference depends on many factors. More generally, an error signal capable of producing a wavelength difference of about 10 ppm is useful.

[0015] In another representative embodiment, a photonic integrated circuit (IC) for wavelength locking of a laser is disclosed. The photonic IC comprises a substrate, an optical splitter disposed on the substrate and configured to receive an optical signal and split the input optical signal to provide a first optical signal and a second optical signal, and an interferometer disposed on the substrate and comprising a first arm and a second arm. In particular, the first arm is configured to receive the first optical signal and the second arm is configured to receive the second optical signal. Moreover, the cladding layer of the first arm has a thickness selected such that a portion of the wave from the first arm is exposed to a standard gas that attenuates light of at least one characteristic wavelength. In particular, as will become more clear as the present description continues, the portion of the wave from the first arm that is exposed to the standard gas depends on the type of waveguide used. For example, in a slab waveguide, the portion of the wave from the first arm that is exposed to the standard gas is an evanescent wave. In contrast, in a slot waveguide, the maximum portion of the Gaussian mode is exposed to the standard gas, while the portion of this mode that is attenuating perpendicular to the propagation direction is not exposed. The photonic IC also includes an optical coupler disposed on the substrate and configured to receive the output optical signal from the first arm and the output optical signal from the second arm and provide a third optical signal. The photonic IC can also include a photodetector disposed on the substrate and configured to receive the third optical signal and provide a photocurrent. The photonic IC can be connected to a control circuit or a feedback control circuit can be integrated within the photonic IC. In either case, the feedback control circuit is configured to modify a characteristic of the laser until an error signal indicative of a difference between a characteristic wavelength and the wavelength of the laser to be locked is zero.

[0016] FIG. 1 is a schematic diagram of a laser 101 and an apparatus 100 for locking the center wavelength of the laser 101 according to a representative embodiment.

[0017] Laser 101 is illustratively a semiconductor laser, such as a distributed feedback (DBF) laser. As mentioned above, laser 101 is susceptible to drift due to various causes, such as operating temperature and ambient temperature.

[0018] Laser 101 may be used in a variety of applications, such as metrology or optical communications, where maintaining the center wavelength within a relatively precise wavelength range is useful, if not essential, to the proper operation of the device in which laser 101 is deployed. Typically, lasers, such as He-Ne lasers, are specified to have an absolute stability of about 0.01 ppm. The present teachings are designed to provide laser 101 with such stability.

[0019] Laser 101 provides an output signal that is incident on a first optical splitter 102 that splits off a small portion, typically less than 1 / 10, of the output optical signal from laser 101 to provide an optical signal to a second optical splitter 104 and another optical signal to an apparatus in which laser 101 is deployed (e.g., an optical wavemeter), not shown.

[0020] The second optical splitter 104 is a 50:50 splitter that outputs the first optical signal (OS1) to the first arm 106 of the interferometer 110. In one exemplary embodiment, the first optical signal (OS1), which may be referred to as the reference optical signal (OSref), passes through a first phase shifter 112 that imparts a predetermined phase shift to the reference optical signal OSref relative to the first optical signal OS1.

[0021] Similarly, the second optical splitter 104 outputs a second optical signal (OS2) to the second arm 108 of the interferometer 110. The second optical signal (OS2) passes through a second phase shifter 111, which imparts a predetermined phase shift to the reference optical signal OSref relative to the second optical signal OS2. The phase shifts from the first phase shifter 112 and the second phase shifter 111 are used to equalize the optical path lengths of the first arm 106 and the second arm 108 due to unavoidable manufacturing tolerances of the two arms.

[0022] As will become more apparent as this teaching continues, in order to effect proper control of the output of laser 101, the optical path lengths of first arm 106 and second arm 108 should be substantially identical.

[0023] According to an exemplary embodiment, the interferometer 110 can be a Mach-Zehnder Interferometer (MZI). In one of several ways, some or all of the first arm 106 of the interferometer 110 is exposed to a standard gas having an absorption line at a characteristic wavelength or eigenmode. The standard gas is selected to absorb light having a wavelength that is substantially the same as the desired output wavelength of the laser 101. In particular, according to the present teachings, in center lock, the output wavelength from the laser 101 is equal to the wavelength of the selected absorption line of the standard gas. Thus, to lock to an absorption line in the region of interest, a gas having an absorption line in the region of interest is chosen as the standard gas. For application of the present teachings as a reference laser, any wavelength will work, provided that the wavelength is well known and does not change significantly (e.g., when the temperature or pressure of the absorbing gas is changing). Thus, the accuracy of the absorption of the standard gas provides an accuracy of locking of the laser 101 of about 0.01 ppm or less.

[0024] Any absorption of light is coupled to dispersion by the Kramers-Kronig relation. As a result, dispersion is proportional to the effective refractive index of the waveguide (Δn r ), resulting in a change in the optical path length. The change in the optical path length results in a change in the phase of the first optical signal when the wavelength of the first optical signal is substantially equal to the characteristic wavelength of laser 101.

[0025] In particular, the absorption and dispersion by the standard gas are mathematically related. The change in effective index is proportional to the derivative of the absorption of light by the standard gas. At wavelengths relatively far from the absorption line, the change in effective index is substantially zero. Similarly, at the absorption maximum, the change in effective index is substantially zero. To the left and right of the absorption maximum, the sign of the slope (and therefore the index of refraction proportional to the derivative of the absorption) changes. This is why, according to the present teachings, dispersion can be advantageously used to generate an error signal that drives the laser wavelength to the absorption maximum. To measure the dispersion, a (MZ) interferometer 110 is used. To generate a usable error signal, the wavelength of the laser locked to the absorption line needs to be within one or two linewidths of the absorption line selected to lock the wavelength of the laser 101. Otherwise, the error signal may be too small and therefore have no substantial effect on the integrator of the control loop (described below). However, this is not difficult to achieve. This is because the wavelength of the laser 101 can be easily tuned to be sufficiently close to the characteristic wavelength of the standard gas.

[0026] Also, the fabrication process used in fabricating the device 100 may adversely affect the optical path length of the first and second arms, which may adversely affect the performance of the interferometer 110. As mentioned above, the optical path length is the product of the geometric length of the waveguide and the group index. To fabricate a balanced Mach-Zehnder interferometer in which the first arm 106 and the second arm 108 have substantially the same optical path length, the group index of the waveguide design must be known with high precision if different waveguide designs are used for the "gas-sensing" waveguide (in the first arm 106) and the "non-sensing" waveguide (in the second arm 108). It is difficult to fabricate a balanced MZI with different waveguide designs for the first arm 106 and the second arm 108. In addition, typically the "gas-sensing" waveguide (in the first arm 106) is fabricated by a partial "oxide-opening" (i.e., removal of silicon oxide cladding across the arm). The size of this oxide-opening also affects the optical path length of the sensing arm, because it separates the overall length into a portion with cladding having one group index and another portion without cladding having a second, different index. Therefore, even small size variations due to fabrication contribute to imbalances and adversely affect the performance of the interferometer 110. The present teachings overcome these potential imbalances by providing substantially identical first and second arms 106, 108, as discussed shortly.

[0027] Referring again to FIG. 1, the first arm 106 has a first aperture 107 thereon and the second arm 108 has a second aperture 109 thereon, as will be more fully described in connection with FIG. 3 below. To provide the optical path lengths in each of the first arm 106 and the second arm 108, the interferometer 110 of the present teachings is fabricated to ensure that the first aperture 107 and the second aperture 109 are provided on the first arm 106 and the second arm 108 and are substantially identical. For this purpose, "substantially identical" means that the first arm 106 and the second arm 108 have substantially the same geometric shape. This is achieved by forming the components (e.g., waveguides) of the first arm 106 and the second arm 108 in a single / common etching process, whereby manufacturing tolerances that shift the outline or change the dimensions or etch depth of the waveguide elements of the first arm 106 and the second arm 108 have the same effect on the sensing (first arm 106) and non-sensing (second arm 108) waveguides. Stated somewhat differently, to avoid undesirable differences in optical path length, the first arm 106 waveguide and the second arm 108 waveguide as well as the first and second apertures 107 and 109 across both the first and second arms 106 and 108 are manufactured during the same fabrication process. This provides a substantially perfectly symmetrical waveguide structure in which the oxide (cladding) layer is removed from over both the first arm 106 and the second arm 108. In particular, as described below in connection with Figure 3, an inert "fill" gas having substantially the same refractive index as the standard gas (e.g., n = 1) is provided over some or all of the second arm 108. Moreover, as described more fully below, a cover (not shown) is provided over the first opening 107 and the second opening 109 that substantially hermetically seals the respective gases within the volumes of the first opening 107 and the second opening 109 (standard gas in the first opening 107, inert gas in the second opening 109) to prevent contamination of these gases.After traversing each arm of the interferometer 110, the output optical signal from the first arm 106 and the output optical signal from the second arm 108 are provided to an optical coupler 114, which may be a 1×2 optical coupler or a 2×2 optical coupler. The optical coupler 114 combines the output optical signals from the first and second arms of the interferometer 110 to provide a third optical signal (OS3) and, optionally, a fourth optical signal (OS4) if a 2×2 coupler is required. Thus, the third optical signal and the fourth optical signal are based on the combination of the output optical signals from the first and second arms of the interferometer 110. The optical coupler 114 may be, for example, but is not limited to, a 3 dB coupler.

[0028] The apparatus 100 further comprises a first photodetector 116 connected to one output of the optical coupler 114 and an optional second photodetector 118 connected to the other output of the optical coupler 114. The first photodetector 116 is configured to detect the third optical signal OS3 from one output of the optical coupler 114 and provide a corresponding first photocurrent PI1. The first photocurrent PI1 is indicative of the difference between a characteristic wavelength and the wavelengths of the output optical signal from the first arm 106 and the output optical signal from the second arm 108. The optional second photodetector 118 is configured to detect the fourth optical signal OS4 from the other output of the optical coupler 114 and provide a corresponding second photocurrent PI2. The second photocurrent PI2 is indicative of the difference between a characteristic wavelength and the wavelengths of the output optical signal from the first arm 106 and the output optical signal from the second arm 108. The photocurrent PI1 is a constant value (PI1 const ) and wavelength-dependent value (PI1 λ ) and PI2. On the other hand, PI2 is a constant value (PI2 const ) and the wavelength-dependent value PI2 λ If the second optical splitter 104 and the optical coupler 114 are 50:50 couplers and both photodetectors have the same conversion coefficient from optical power to current, then the two constant values ​​PI1 const and PI2 const is equal to the wavelength-dependent value PI1 λ is PI2 λ1. When the wavelength of the optical signal in the first arm 106 of the interferometer 110 is changing relative to the characteristic wavelength, both the first photocurrent PI1 and the second photocurrent PI2 are changing. However, while the first photocurrent PI1 is increasing near a wavelength equal to 1, the second photocurrent PI2 is decreasing, and thus the difference (PI1-PI2) increases approximately twice as fast and is substantially zero at the characteristic wavelength (assuming the first and second photocurrents behave as shown in Figures 2A-2C, described below).

[0029] The first photocurrent PI1 and the second photocurrent PI2 are input to a summer 120. The summer 120 is configured to determine the difference between the first photocurrent PI1 and the second photocurrent PI2 and provide an error signal ε=PI1-PI2 indicative of this difference. It is beneficial to normalize this error signal with the sum of the two photocurrents. In other words, an alternative error signal ε n When the characteristic wavelength of the laser 101 is equal to the wavelength of the first optical signal OS1, it is advantageous to use the error signal ε and the alternative error signal ε output by the summer 120. n is 0. For example, Figure 2B shows an example first photocurrent PI1 and second photocurrent PI2 as a function of wavelength.

[0030] If only one photodetector is used, the output of the second photodetector is replaced by the product of the total optical power emitted by the laser 101 and a previously recorded and stored coefficient. The total optical power can be measured by an independent photodetector (not shown in FIG. 1) or can be inferred from the control parameters of the laser 101. For example, in the case of a DFB laser, the total optical power can be calculated from the pump current and a current to optical power conversion coefficient.

[0031] The output of the summer 120 is the error signal ε or ε na feedback control circuit 122 configured to provide a control signal CS in response to the first and second photocurrents PI1 and PI2, respectively. That is, the feedback control circuit 122 outputs the control signal CS based at least in part on the first and second photocurrents PI1 and PI2 and the comparison thereof. The control signal CS controls the bias of the laser 101, which allows automatic tuning of the laser 101 to a desired output wavelength, which is substantially equal to a characteristic wavelength of the standard gas. Advantageously, by adjusting the bias point with the control signal CS, the laser 101 is controlled to adjust its output to be substantially equal to the characteristic wavelength of the standard gas.

[0032] In one embodiment, the feedback control circuit 122 is a proportional-integral (PI) controller configured to receive and process an error signal ε indicative of the difference between the wavelength of the output signal of the laser 101 and the characteristic wavelength, and to output a control signal CS in response thereto. The PI controller includes an amplifier 124 that receives and amplifies the error signal ε and outputs an amplified error signal ε1. The PI controller further includes an integral logic device 126 that performs an integral function on the amplified error signal ε1 over time to provide an integrated error signal ε2, and an adder 448 that sums the amplified error signal ε1 and the integrated error signal ε2 to output the control signal CS. In one embodiment, the PI controller can be replaced with a PID controller that also includes differential logic, or an integral (I) controller that does not include a proportional portion. Of course, the feedback control circuit 122 can include other types of control circuits, such as an on-off controller, or a controller using fuzzy logic, without departing from the scope of the present teachings.

[0033] As can be seen above, a single photodetector could be used, but using the difference in photocurrents of PD1 116 and PD2 118 simplifies determining the error signal, at least because it is not necessary to know the output power of laser 101, and there is no need to store the expected difference between the ratio of photocurrent and optical power of PD1 116 (or PD2 118) and the expected ratio (previously determined and stored value), since this expected difference is zero at the characteristic wavelength and is therefore independent of the actual power.

[0034] FIG. 2A shows a graph 202 of the light transmission by a standard gas according to a representative embodiment. In particular, the peak absorption at point 204 indicates the absorption of light at the desired characteristic frequency of the laser 101. In particular, the standard gas is transparent to the incident light except for a few wavelength values ​​(called characteristic values / eigenvalues). These values ​​depend on the type of gas used for the standard gas. To a much smaller extent, the characteristic values ​​also depend on the temperature and pressure of the standard gas. The much smaller degree of dependence on temperature and pressure is why the characteristic values ​​can be used as a dominant wavelength reference and can be used to stabilize the wavelength of a laser (e.g., laser 101).

[0035] As described above, the first phase shifter 112 in the first arm 106 of the interferometer 110 shifts the phase of the first optical signal OS1 (i.e., the reference optical signal) relative to the phase of the second optical signal OS2 (i.e., the input optical signal). Figure 2B shows an example of the first photocurrent PI1 206 and the second photocurrent PI2 208 when the first optical signal OS1 is phase shifted by an appropriate amount. The appropriate amount of phase shifting performed by the phase shifter 424 is reached when the error signal versus wavelength is an antisymmetric function with respect to the characteristic wavelength of the absorption line of the standard gas.

[0036] As described above, the summer 120 is configured to determine the difference between the first photocurrent PI1 and the second photocurrent PI2 and provide an error signal ε indicative of this difference. The characteristic wavelength of the peak absorption of the standard absorbing gas is equal to the wavelength of the input optical signal OS1 when the error signal ε output by the summer 120 is zero. For example, FIG. 2B shows an exemplary first photocurrent PI1 206 and a second photocurrent PI2 208 when the error signal ε is equal to zero. In order for the difference between these photocurrents to be zero at the characteristic wavelength, the constant portion of PI1 and the constant portion of PI2 need to be substantially the same. According to an exemplary embodiment, this is achieved when the second optical splitter 104 and the optical coupler 114 are each 50:50 couplers and both photodetectors (PD1 and PD2) have substantially the same sensitivity.

[0037] FIG. 2C is a graph illustrating an error signal resulting from subtraction of photocurrents from a first photodiode and a second photodiode, according to a representative embodiment.

[0038] Referring to FIG. 2C, a "0" on the vertical axis indicates that both photocurrents are equal, which can only happen when the laser wavelength is equal to the characteristic wavelength of the standard gas absorption, and a "1" on the horizontal axis indicates that the optical signal OS1 and the absorption line / characteristic wavelength of the absorbing gas are equal. The different traces 210, 212, 214 show the dependence of the error signal ε on the phase shift introduced by the first phase shifter 112 in the first arm 106 of the interferometer 110. For a suitable phase range, the error signal ε changes sign at the characteristic wavelength / absorption line of the absorbing gas. The traces 210, 212 and 214 correspond to three different phase shift values. The middle trace 212 is the "ideal" phase selection, whereas the other two traces 210 and 214 correspond to phase changes slightly above and slightly below the ideal value, respectively. FIG. 2C shows that the actual value of the phase shift does not affect the "operating point" of the feedback control circuit 122. In other words, the error signal will be zero when the wavelength of the laser 101 and the characteristic wavelength of the standard gas are equal, regardless of the exact value of the phase shift. Thus, the error signal should be zero when the desired condition for the feedback control circuit 122 (i.e., the wavelength of the laser 101 and the characteristic wavelength of the standard gas) is met. Even if the phase shift from the first phase shifter 112 is not exactly the ideal value, meeting this condition helps to achieve the desired wavelength lock even if the setting of the first phase shifter 112 is imperfect.

[0039] 3 is a perspective view of a portion of a photonic integrated circuit (IC) 300 used to lock the wavelength of a laser, according to a representative embodiment. In particular, many aspects of the various components of the apparatus 100 described above with respect to FIGS. 1-2C are common to aspects of the photonic IC 300 described herein. These details may not be repeated for the sake of brevity and clarity.

[0040] As will be appreciated as the present description continues, the photonic IC 300 allows the absorption cell containing the absorbing gas, the interferometer 110, the second optical splitter 104, the optical coupler 114, the first photodetector 116 and the second photodetector 118, as well as other components described above, to be integrated onto a relatively small photonic integrated chip. Advantageously, the photonic IC 300 provides superior stability compared to bulk optical solutions and can be fabricated at a relatively low cost. Because the locking scheme does not require high speed electronics, the control electronics can also be fabricated at a relatively low cost and with very low noise, if desired.

[0041] Referring to FIG. 3, the photonic IC 300 comprises a series of connected optical waveguides as described herein. According to a representative embodiment, the optical waveguides of the photonic IC 300 are slab waveguides consisting of a channel (core). The channel is made of a material with a higher refractive index that is disposed in / surrounded by a material (cladding) with a lower refractive index than the channel. According to a representative embodiment, the waveguides of the photonic IC 300 can comprise a core 340. The core is illustratively silicon and disposed in a silicon dioxide cladding 330 that can be disposed on a silicon substrate. In general, the photonic IC 300 of various embodiments can be implemented in any material system in which a waveguide and ideally a phase modulator can be fabricated. For example, the core disposed in the cladding for a slab waveguide can be silicon (Si) in silicon dioxide (SiO2), silicon nitride (Si3N4) in silicon dioxide (SiO2), or doped indium in undoped indium (InP). However, other material systems can be incorporated without departing from the scope of the present teachings.

[0042] Input from a laser (e.g., laser 101) is provided to a first end of photonic IC 300. The laser is coupled to input waveguide 301 by known optical coupling methods. For some material systems (notably InP), the laser can be integrated into the same photonic IC.

[0043] An input signal is incident on a first optical splitter 302. By way of example, typically less than 1 / 10 of the output optical signal from the laser provides an optical signal to a second optical splitter 304, which provides another optical signal to a device in which the laser is deployed (e.g., an optical wavemeter), not shown.

[0044] The second optical splitter 304 is a 50:50 splitter and outputs the first optical signal (OS1) to the first arm 306 of the interferometer 310. In an exemplary embodiment, the first optical signal (OS1), sometimes called the reference optical signal (OSref), passes through a phase shifter (not shown) integrated in the waveguide. This phase shifter can be realized to change the effective refractive index of the waveguide thermo-optically, stress-optically, or electro-optically, and for symmetry, may be integrated in the second arm 308 after the second optical splitter 304 and before the input to the first arm 306 or the second arm 308 of the interferometer 310. As mentioned above, the phase shifter provides the reference optical signal OSref with a predetermined phase shift with respect to the first optical signal OS1. This phase shift is used to equalize the optical path lengths of the first arm 306 and the second arm 308 due to the unavoidable manufacturing tolerances of the two arms, as mentioned above.

[0045] The second optical splitter 304 also outputs a second optical signal (OS2) to a second arm 308 of the interferometer 310.

[0046] As mentioned above, the first arm 306 of the interferometer is exposed to a standard gas that induces dispersion, which is measured by the change in the effective refractive index of the waveguide (Δn r), resulting in a change in the optical path length. The change in the optical path length results in a change in the phase of the first optical signal OS1 when the wavelength of the first optical signal OS1 is substantially equal to the characteristic wavelength of the laser, as described above. According to an exemplary embodiment, this exposure to the selected standard gas can be achieved through the evanescent tail or a small portion of the guided optical signal, depending on the type of waveguide used. In the exemplary embodiment described here, the cladding layer of the slab waveguide of the first arm 306 of the interferometer 310 is thinned to ensure that the evanescent mode of the optical signal has enough power to be absorbed in a measurable amount to change the dispersion of the first arm 306 and ultimately affect the optical path length of the optical signal OS1 traveling in the first arm 306 of the interferometer 310.

[0047] Typically, the cladding is greater than 3 μm thick when operating in the 1.5 μm near infrared wavelength band, but to ensure sufficient energy is transmitted through the evanescent modes, the first region 320 of the cladding layer (e.g., SiO2) can be thinned or formed to a thickness in the range of about 0 μm (i.e., the core layer is removed so as to expose it to a standard gas) to about 1 μm. Similarly, as described below, the cladding layer of the waveguide that constitutes the second arm 308 of the interferometer 310 is reduced by substantially the same amount as the cladding layer of the waveguide that constitutes the first arm 306.

[0048] More generally, the thickness of the cladding layer (and the removed or originally cladded-free portion of the first region 320) is a design parameter that depends on various factors that can be selected for the desired absorption level / magnitude. For example, a standard gas that has a relatively strong absorption in the characteristic values ​​(and therefore exhibits a larger dispersion) can allow the exposed portion of the first arm to have a relatively thick cladding. In addition, the higher the pressure of the standard gas, the thicker the cladding that remains in the exposed portion of the first arm can be because a higher pressure provides more molecules per unit length of the waveguide, which creates more absorption. However, there is a limit to the requirement of a higher pressure of the standard gas, since the higher the pressure of the standard gas, the higher the possibility of undesirable absorption line broadening. Yet another parameter that can be used to improve absorption is the length of the waveguide in the exposed portion of the first arm. Thus, by increasing the length of the waveguide in the exposed portion of the first arm, rather than increasing the pressure of the standard gas, the absorption increases according to the Beer-Lambert Law. Of course, increasing the length of the waveguide in the first region 320 of the first arm can have a detrimental effect on the overall size of the photonic IC 300, since this increase in the length of the first region 320 results in an increase in the overall size of the device or photonic IC 300. Finally, the gain in the feedback control circuit 122 can also reduce the extent to which cladding is removed from the waveguide in the first region 320, since a higher gain allows for a relatively thicker cladding layer in the exposed portion of the first arm in the first region 320.

[0049] As alluded to above, to provide the second arm 308 with substantially the same optical path length as the first arm 306 (and substantially symmetrical to the first arm 306), the second region 321 of the cladding layer (e.g., SiO2) can be thinned or formed to a thickness in the range of about 0 μm (i.e., the core layer is removed to expose to an inert gas) to about 1 μm, while typically the cladding is greater than 3 μm thick when operating in the near infrared wavelength range of 1.5 μm. Furthermore, as explained above, the waveguides of the first arm 306 and second arm 308 and the first region 320 and second region 321 are formed in the same fabrication sequence to provide substantially identical waveguides and apertures in each of the first and second arms and first and second regions. Thereafter, to facilitate providing substantially identical first and second arms 306, 308, the area above the second arm 308 is filled with an inert gas having substantially the same refractive index as the standard gas provided across the first arm 306.

[0050] To ensure exposure of the first arm 306 only to the standard gas and the second arm 308 only to the inert gas, respective covers (not shown) of glass, metal or other suitable material are placed over the first region 320 and the second region 321. Thus, both the volumes of the first region 320 and the second region 321 have respective covers over them, and the respective gases (e.g., infrared-active standard gas over the first region 320 and infrared-inactive "fill" gas over the second region 2) are enclosed and separated from each other to avoid contamination of the respective gases. The covers can be placed over the first region and the second region and sealed using one of several known methods. The covers can also be added using a reflow soldering process with different melting points, allowing different gases (and pressures) to be used. Beneficially, the covers and sealing methods used to enclose the standard gas in the first region 320 and the inert gas in the second region 321 substantially seal the respective gases in their respective volumes within the first region 320 and the second region 321.

[0051] After traversing each arm of the interferometer 310, the output optical signal from the first arm 306 and the output optical signal from the second arm 308 are provided to an optical coupler 314, which is a 2×2 optical coupler. The optical coupler 314 combines the output optical signals from the first and second arms of the interferometer 310 to provide a third optical signal (OS3) and optionally a fourth optical signal (OS4). Thus, the third optical signal and the fourth optical signal are based on a combination of the output optical signals from the first and second arms of the interferometer 310. The optical coupler 314 can be, for example, but is not limited to, a 3 dB coupler.

[0052] As shown, the outputs to a first photodetector (not shown in FIG. 2) and an optional second photodetector (not shown in FIG. 3) are coupled to these photodetectors, which may be fabricated integrally with the photonic IC 300. Similarly, the various components required to determine the error signal and the correction signal (e.g., summer 120, feedback control circuit 122) and the electrical connections required to enable this function may be formed during the manufacturing process used to fabricate the photonic IC 300.

[0053] Finally, the photonic IC 300 is fabricated, by way of example, using one of a variety of known semiconductor processing methods commonly used in photonics. By way of example only, according to a representative embodiment in which the core 340 is silicon and the cladding 330 is SiO2, the photonic IC 300 can be formed by forming a first layer of SiO2 on a silicon substrate (not shown). The first layer of SiO2 is relatively flat and has a thickness of about a few micrometers (μm). A thin layer of silicon is then formed over the first layer of SiO2, such as by deposition of silicon. This layer has a thickness of about 500 nm. Photoresist is then applied and an etching mask (not shown) is created. The silicon is then partially removed using known etching techniques to form the core 340. After formation, the core 340 is covered by a second layer of SiO2, which is planarized to complete the waveguide structure of the photonic IC 300.

[0054] The first and second regions 320 and 321 can be formed by etching the SiO2 to the desired cladding thickness in the first and second regions, as described above, after deposition of the second layer of SiO2. Alternatively, during the manufacture of the second layer of SiO2, the first and second regions 320 and 321 can be masked during the formation of the second layer of SiO2, such that a portion of the deposit provides the desired cladding thickness in the first and second regions 320 and 321.

[0055] As mentioned above, one type of optical waveguide contemplated by the present teachings is a slab waveguide (e.g., core 340) made, for example, of a core layer of silicon and a cladding layer of silicon dioxide. Figure 4 is a perspective view of a slot optical waveguide 400 contemplated for use in a photonic IC according to a representative embodiment.

[0056] The slot optical waveguide 400, also known as a slot waveguide, comprises a first waveguide section 401, a second waveguide section 403, which together form the actual waveguiding structure for a single mode waveguide, and a space 402 disposed between them. In particular, the first waveguide section 401 and the second waveguide section 403 have a relatively high refractive index with respect to the refractive index of the material in the space. Thus, the light waves are guided along the z direction in the coordinate system shown in FIG. 4, with the maximum part of the guided mode being in the space 402.

[0057] According to an exemplary embodiment, a slot waveguide such as slot optical waveguide 400 is implemented in the region of a first / reference arm (e.g., first arm 106, 306) of an MZ interferometer (e.g., interferometer 110, 310). A selected standard gas (not shown) is placed in space 402 (and other portions of slot optical waveguide 400 surrounding first waveguide portion 401 and second waveguide portion 403). This standard gas is used for the relatively low index material required to support eigenmodes in slot optical waveguide 400 as well as absorption of evanescent modes of slot optical waveguide 400.

[0058] 5 illustrates a butterfly package 500 that may be used with the photonic ICs of the present teachings. In particular, many aspects of the various components of the apparatus 100, photonic IC 300, and slot optical waveguide 400 described above with respect to FIGS. 1-4 may be applicable to the butterfly package 500 described herein. These details may not be repeated for the sake of brevity and clarity.

[0059] The butterfly package comprises a cavity 501 that, when sealed, provides a hermetic enclosure for electronic / optoelectronic components disposed therein. According to an exemplary embodiment, a photonic IC 502 is disposed within the cavity. The photonic IC 502 can be, for example, the photonic IC 300 described above, and components not shown in FIG. 3 (e.g., the first and second photodetectors, the summer 120, the feedback control circuit 122) can also be integrally formed with the photonic IC 502 disposed within the cavity.

[0060] A standard gas (not shown) selected to provide an absorption line at the desired output wavelength of a laser (eg, laser 101 ) can be provided within the hermetically sealed cavity 501 .

[0061] Finally, a laser (e.g., laser 101) that is desirably tuned by the device / photonic chip of the present teachings can also be integrated inside the butterfly package, with its output through photonic IC 300 connected to an optical connector 503 located outside the butterfly package 500.

[0062] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustrations and descriptions are to be considered illustrative or exemplary and not restrictive, and the invention is not limited to the disclosed embodiments.

[0063] While exemplary embodiments are disclosed herein, those skilled in the art will recognize that many variations are possible in accordance with the present teachings and are within the scope of the appended claims. Accordingly, the invention is not to be limited except as within the scope of the appended claims. The claims as originally filed are as follows: Claim 1: An apparatus (100) for locking the wavelength of a laser (101), comprising: an optical splitter configured to receive an optical signal and split the input optical signal to provide a first optical signal and a second optical signal; an interferometer (110) comprising a first arm (106) and a second arm (108), the first arm (106) configured to receive the first optical signal and the second arm (108) configured to receive the second optical signal, a portion of the first arm (106) exposed to a standard gas that attenuates light of a characteristic wavelength; an optical coupler (114) configured to receive the output optical signal from the first arm (106) and the output optical signal from the second arm (108) and provide a third optical signal; a photodetector configured to receive the third optical signal and provide a photocurrent, the photocurrent increasing as a difference between the characteristic wavelength and the wavelengths of the output optical signal from the first arm (106) and the output optical signal from the second arm (108) increases; a feedback control circuit (122) configured to modify a characteristic of the laser (101) until an error signal indicative of the difference between the characteristic wavelength and the wavelength of the laser (101) to be locked becomes zero; An apparatus comprising: Claim 2: 2. The apparatus of claim 1, wherein the photodetector is a first photodetector, the photocurrent is a first photocurrent, the optical coupler is configured to provide a fourth optical signal, and the apparatus further comprises a second photodetector configured to receive the fourth optical signal and provide a second photocurrent, the second photocurrent decreasing as the difference between a characteristic wavelength and a wavelength of the optical signal increases. Claim 3: 3. The apparatus of claim 2, further comprising a summer configured to receive the first photocurrent and the second photocurrent and provide an error signal indicative of the difference between the wavelength of the first optical signal and the wavelength of the second optical signal. Claim 4: 4. The apparatus (100) of claim 3, wherein the feedback control circuit (122) further comprises a proportional-integral (PI) controller configured to receive and process the error signal and, in response, output a control signal to the laser (101). Claim 5: 5. The apparatus (100) of claim 4, wherein in response to the control signal, the laser (101) emits radiation at the characteristic wavelength. Claim 6: 6. The apparatus (100) of claim 5, wherein the PI controller comprises an amplifier (124) that receives and amplifies the error signal and provides an amplified error signal. Claim 7: 7. The apparatus of claim 6, wherein the PI controller further comprises an integral logic device that performs an integral function on the amplified error signal over time to provide an integrated error signal. Claim 8: 8. The apparatus of claim 7, wherein the summer is a first summer, and the feedback control circuit further comprises a second summer configured to sum the amplified error signal and the integrated error signal to provide the control signal. Claim 9: The apparatus (100) of claim 1, wherein the characteristic is one of a bias, a temperature, a phase shift, or a mechanical shift. Claim 10: 10. The apparatus of claim 1, wherein the second arm is exposed to another gas having substantially the same refractive index as the standard gas. [Explanation of symbols]

[0064] 100 devices 101 Laser 102 First optical splitter 104 Second Optical Splitter 106 First Arm 108 Second Arm 110 Interferometer 111 Second Phase Shifter 112 First Phase Shifter 114 Optical Coupler 114 116 First photodetector 118 Second Photodetector 120 Adder 122 Feedback Control Circuit 124 Amplifier

Claims

1. A device (100) for locking the wavelength of a laser (101), comprising: a first optical signal and a second optical signal; an optical splitter configured as an interferometer (110) comprising a first arm (106) and a second arm (108), the first arm (106) configured to receive the first optical signal and the second arm (108) configured to receive the second optical signal, a portion of the first arm (106) exposed to a standard gas that attenuates light at a characteristic wavelength, the characteristic wavelength being a wavelength at which the transmittance of light in the standard gas exhibits an absorption peak; an optical coupler (114) configured to receive the output optical signal from the first arm (106) and the output optical signal from the second arm (108) and provide a third optical signal; a photodetector configured to receive the third optical signal and provide a photocurrent, the photocurrent increasing as a difference between the characteristic wavelength and a wavelength of an optical signal in which the output optical signal from the first arm (106) and the output optical signal from the second arm (108) are combined by the optical coupler; a feedback control circuit (122) configured to modify a characteristic of the laser (101) until an error signal indicative of the difference between the characteristic wavelength and the wavelength of the laser (101) to be locked becomes zero; An apparatus comprising:

2. 2. The apparatus of claim 1, wherein the photodetector is a first photodetector, the photocurrent is a first photocurrent, the optical coupler configured to receive the output optical signal from the first arm and the output optical signal from the second arm is further configured to provide a fourth optical signal, and the apparatus further comprises a second photodetector configured to receive the fourth optical signal and provide a second photocurrent, the second photocurrent decreasing as the difference between the characteristic wavelength and the wavelength of the optical signal into which the output optical signal from the first arm and the output optical signal from the second arm are combined by the optical coupler increases.

3. 3. The apparatus of claim 2, further comprising: a summer configured to receive the first photocurrent and the second photocurrent and provide an error signal indicative of the difference between the wavelength of the first optical signal and the wavelength of the second optical signal.

4. 4. The apparatus (100) of claim 3, wherein the feedback control circuit (122) further comprises a proportional-integral (PI) controller configured to receive and process the error signal and, in response thereto, output a control signal to the laser (101).

5. The apparatus (100) of claim 4, wherein in response to the control signal, the laser (101) emits radiation at the characteristic wavelength.

6. The apparatus (100) of claim 5, wherein the PI controller comprises an amplifier (124) that receives and amplifies the error signal and provides an amplified error signal.

7. 7. The apparatus (100) of claim 6, wherein the PI controller further comprises an integral logic device (126) that performs an integral function on the amplified error signal over time to provide an integrated error signal.

8. 8. The apparatus of claim 7, wherein the summer is a first summer, and the feedback control circuit further comprises a second summer configured to sum the amplified error signal and the integrated error signal to provide the control signal.

9. 2. The apparatus (100) of claim 1, wherein the alteration of the laser characteristics is by controlling at least one of the following: a bias of the laser; a temperature of the standard gas; a phase shift induced in the first optical signal in the first arm of the interferometer; a size of an oxide aperture in a waveguide in the first arm; and a length of the waveguide in the first arm.

10. The apparatus (100) of claim 1, wherein the second arm (108) is exposed to another gas having substantially the same refractive index as the standard gas.

Images