Open Access
Add to BibSonomy
Abstract
We present the development of a transportable laser frequency stabilization system with application to both optical clocks and a next-generation gravity mission (NGGM) in space. This effort leverages a 5-cm long cubic cavity with crystalline coatings operating at room temperature and with a center wavelength of 1064 nm. The cavity is integrated in a custom vacuum chamber with dedicated low-noise locking electronics. Our vacuum-mounted cavity and control system are well suited for space applications, exhibiting state-of-the-art noise performance while being resilient to radiation exposure, vibration, shock, and temperature variations. Furthermore, we demonstrate a robust means of automatically (re)locking the laser to the cavity when resonance is lost. We show that the mounted cavity is capable of reaching technology readiness level (TRL) 6, paving the way for high-performance ultrastable laser systems and eventually optical atomic clocks amenable to future satellite platforms.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
There is increasing interest in high-stability cavity-stabilized lasers and optical atomic clocks capable of efficient long-term operation in space. Potential applications include next-generation global navigation satellite systems (GNSS) [1], Earth Observation (e.g., Next Generation Gravity Mission, NGGM) [2], as well as fundamental physics experiments such as the Space Optical Clock (SOC) [3] and the laser interferometer space antenna (LISA) [4]. To develop robust and compact laser stabilization systems amenable to space applications, it is necessary to increase the system technology readiness level (TRL) of high-finesse and low-thermal-noise cavities such that they are capable of surviving launch and satellite deployment [5]. Similarly, advances must be made in the development of supporting subsystems including low-noise space qualified locking electronics, frequency offset tunability, automatic laser relocking and fiber path length stabilization to cover the range of space applications listed above. Various ground-based experiments focused on sub-Hz integrated linewidth cavity-stabilized lasers have previously achieved TRL 4. These efforts show clear progress towards space-applicable solutions [6,7,8,9,10,11,12,13]. In this manuscript, we provide a detailed overview of the European Space Agency (ESA) funded and Airbus Defence and Space (AirbusDS) led optical stabilizing reference cavity (OSRC) consortium. In this activity [14], we have developed an optical reference cavity based on the National Physical Laboratory (NPL) vibration-insensitive cubic spacer design [15] employing low-noise substrate-transferred crystalline coatings [16,17], targeting a minimum cavity finesse of 200 000. This high finesse value is necessary for advanced space-based optical clock capabilities, as well for tests of fundamental physics and select applications in optically-enabled satellite navigation. The assembled cavity is integrated in a thermally and mechanically decoupled vacuum chamber with custom low-noise control electronics. This compact, yet high-performance system is an ideal platform for space-based cavity-stabilized lasers for future optical clocks and also applications mentioned above, such as NGGM and LISA. We pursue a series of environmental tests on the cavity-in-vacuum-chamber assembly including proton exposure, as well as probing its acceleration, vibration, and shock resistance. TRL 6 is achieved for the mounted cubic cavity, demonstrating a clear path to sub-Hz integrated linewidth lasers in space.
2. System overview
In the subsections below we provide a detailed overview of our cavity-stabilized laser including the reference cavity and vacuum chamber, the custom locking electronics, as well as the fiber link stabilization system. Here, the emphasis is on engineering the optical local oscillator to perform in a satellite environment with phase-stabilized optical distribution.
2.1 Cavity and vacuum chamber design
The reference cavity, shown in Fig. 1, utilizes a cubic 5-cm long ultralow expansion (ULE) glass spacer operating at temperatures between 20-35°C, where the linear coefficient of thermal expansion (CTE) is zero, with GaAs/AlGaAs-based crystalline mirror coatings capable of high finesse at 1064 nm. The use of crystalline coatings enables a lower cavity thermal noise floor when compared to the original NPL cavity design with traditional amorphous mirror coatings. We note that this also requires laser light with linear optical polarization to lock to the birefringent cavity. The cavity has a theoretical thermal-noise-limited fractional frequency performance [18,19] of 3 × 10−16 (instability at 1 s averaging). The Pound-Drever-Hall (PDH) technique is used for laser frequency locking [20]. In our implementation, the cubic spacer design provides minimal acceleration sensitivity in a compact form factor, while crystalline coatings provide lower thermal (Brownian) noise than amorphous coatings, owing to their reduced elastic loss [21]. The thermal mismatch between the ULE spacer and fused silica mirror substrates results in a decrease in the effective zero-crossing temperature for the thermal expansion of the cavity assembly. This decrease is partially recovered by optically contacting ULE thermal compensation rings to the backside of the mirror substrates, elevating the CTE zero-crossing point back to near room temperature [22].
Fig. 1. Photographs of the bare and packaged cavity. The assembled cavity with cubic spacer and crystalline coatings is shown mounted in the inner frame (left), with passive thermal shielding added (middle) and, finally, contained inside the completed vacuum enclosure (right).
The cubic cavity utilizes a similar mounting arrangement as with other NPL cubic cavities [23]. In this case the cavity is mounted using spheres manufactured from a plastic with suitable compliance within a titanium frame surrounded by a layer of thermal shielding as shown in Fig. 1. Following the results of thermal modelling, the frame and thermal shields were mechanically polished and gold-coated to improve thermal isolation. Polished titanium, for example, has an emissivity of 0.19 [24] whereas the emissivity of polished gold can be as low as 0.03. To minimize the sensitivity of the cavity length to acceleration, bespoke alignment jigs were used to center the cube in its mounting frame with a tolerance of 50 µm. The cavity assembly was secured in a titanium mid-plane mounting ring. This was installed within a second set of heatshields in a custom vacuum chamber evacuated by means of an ion pump to the 1 × 10−7 mbar level. The chamber has a vacuum seal around the mid-plane and a mass below 10 kg. For evaluation of the cavity, mode-matching input optics were mounted on a breadboard that was fixed to the top surface of the vacuum chamber, with a transmission photodiode on the underside (see Fig. 1).
2.2 Locking electronics
The control electronics for locking the laser to the optical cavity are based around an FPGA clocked by an external 10 MHz reference. The system is represented by the functional block diagram depicted in Fig. 2. Direct digital synthesizers (DDS) are employed to generate the modulation signal for the electro-optic modulator (EOM) at 18.75 MHz and for the acousto-optic modulator (AOM) at 80 MHz. Analog-to-digital converters (ADCs) digitize the signals. The PDH photodiode detects the optical beam reflected from the cavity and a second photodiode detects the transmitted signal. A laser-frequency-lock error signal is generated by numerically mixing the PDH photodiode signal and a digitally controlled oscillator signal at the PDH frequency with the appropriate phase. The laser frequency control signal supplied to the piezo actuator of the laser and its temperature control are processed by three successive digital proportional-integral-derivative (PID) controllers. In addition to this laser frequency lock, the intra-cavity optical power is stabilized to mitigate drift of the cavity resonance due to laser power fluctuations. The DC signal from the photodiode downstream from the optical cavity is detected and processed by a PID. The resulting control signal is transferred to the analog domain in a digital-to-analog converter (DAC) and adjusts the RF power applied to the AOM, keeping both the optical output power incident on the cavity, and hence the intra-cavity power, constant.
Fig. 2. Functional block diagram of the overall system including control electronics. The contributions to the subunits from the consortium partners are shown with different colored backgrounds. Free space optical beams are drawn as red solid lines, connections using single mode polarization maintaining optical fibers are depicted as solid blue lines, and electronic connections as black lines. AOM: Acousto-optic modulator; BS: Beam splitter; DBM: Double-balanced mixer; EOM: Electro optic modulator; FI: Faraday isolator; FM: Fiber mirror; HW: Half wave plate; PD: Photodetectors.
The control electronics have been implemented employing standard “off-the-shelf” parts. As part of the activity, it has been assessed that all of the functions described above can be implemented with radiation hard space-qualified components. For example, FPGA solutions from NanoXplore, Xilinx and Microchip provide enough IOs, slices, and sufficiently high clocking capability. ADCs and DACs with sufficient effective resolution and noise are additionally available. The only critical components are the DDS, which are used to synthesize the necessary RF tones for the EOM and AOM modulation. However, as an alternative to DDS, radiation hard PLLs could be employed.
2.3 Laser auto (re)locking and noise performance
An important feature required for any cavity-stabilized laser in space is the ability to automatically lock and relock the laser to the desired optical resonance. This capability has been realized in the OSRC control electronics by implementing a search algorithm as demonstrated in Fig. 3. The laser is unlocked from the cavity by blocking the laser beam incident on the cavity. When the beam is blocked, no light is transmitted by the cavity, which is detected by the transmission photodetector (PDT in Fig. 2). When the signal detected by this photodetector falls below a threshold set by the user, the system switches to the unlocked state and positive and negative ramps with increasing amplitude are applied to the fast laser frequency actuator (internal piezo of the non-planar ring oscillator laser). The ramps become visible by comparing the OSRC-controlled laser frequency against a reference laser system and are apparent in the beat note frequency versus time plot in Fig. 3. While scanning the laser frequency, the signal from the PDT photodetector is continuously monitored. Once the signal exceeds the lock detection threshold, all PID controllers (laser frequency and power) of the system are re-engaged, and the laser returns to the locked state with beat note frequency again at its previously locked value.
Fig. 3. Time series of the beat note frequency between the OSRC-stabilized laser and a reference laser. When the laser light incident to the OSRC is blocked by the user, an automatic PDH lock recovery is engaged. Successive ramps with increasing amplitude are applied to the laser frequency modulation input until lock is recovered.
We tested the stabilization performance with a source similar to that planned for NGGM, i.e. a 1064 nm Nd:YAG Mephisto laser, as shown in Fig. 4 in the form of a fractional frequency amplitude spectral density (ASD). The measured performance is limited by the reference laser systems available at the time of testing at AirbusDS, but is sufficient in comparison against the NGGM specifications. For future optical clock activities, measurements using ultrastable reference lasers are planned at NPL, while improved reference laser infrastructure is being implemented at AirbusDS. The measured performance is likely to be limited by the reference laser and is not significantly different to measurements before the environmental tests. The blue line in Fig. 4 shows the ESA (NGGM) fractional frequency spectral noise specification (i.e., $\sqrt {{S_y}(f )} = \sqrt {{S_{}}(f )} $/ν0) [25]. The specification $\sqrt {{S_{}}(f )} $ is at ν0 = c/1064.6(8) nm = 281.6 THz with c the speed of light:
The two curves show the result of a measurement against another cavity-stabilized laser (red) and an atomic reference (green). Note that in all measurements, the OSRC was not placed on the vibration isolation table and was operated close to room temperature, which is ∼10 K below the CTE zero crossing, the determination of which is described in the following section.
Fig. 4. Frequency noise beat measurements carried out at AirbusDS for the OSRC against cavity-stabilized and iodine-stabilized reference lasers, respectively. The blue line is the target NGGM specification. This data was recorded after environmental testing of the system.
2.4 Cavity thermal expansion and thermal time constant
The effective cavity thermal expansion was measured over the range of 24 °C to 35 °C. At each temperature, the cavity was allowed to thermally stabilize over a period of at least two days. The change in cavity length was determined from a beat frequency measurement. This beat was between a 1064-nm reference laser locked to a 48-cm cavity with its frequency drift corrected relative to an NPL maser [26] and the Mephisto laser locked to the TEM00 mode of the cubic cavity. These beat frequency measurements allowed determination of both the thermal expansion and the thermal time constant following a typical 2 °C change in set-point temperature. The vacuum chamber temperature is controlled via a set of heater pads, one on top of the chamber, one below, and the remainder placed around the sides. The chamber is insulated within a formed polystyrene cylinder. Six temperature sensors fixed to the outside of the vacuum chamber are distributed on the top and bottom plates, as well as the sidewalls. The temperature recorded for Fig. 5 is the mean of the six sensor readings. The error bars indicate our estimated standard uncertainties (coverage factor $k$ = 1) for both the temperature and frequency values. The frequencies are an estimate (with uncertainty) of the beat as $t \to \infty $ (Fig. 6) and the temperature uncertainty is the estimated temperature stability at each set point.
Fig. 5. Measurement of the thermal expansion of the cubic cavity as a function of temperature. The temperature is a mean taken using six temperature sensors fitted to the outside of the vacuum chamber. The thermal expansion zero-crossing for the assembled cavity is determined to be 32.1(1) °C from this fit.
Fig. 6. Measurement of the cubic cavity thermal time constant following a ∼2 °C increase in set-point temperature. The effective cavity temperature is determined from the beat frequency (Fig. 5) and the thermal time constant was determined to be 16 hours.
In Fig. 5, the reference laser is higher in frequency than the laser locked to the cubic cavity and so a minimum beat corresponds to a maximum 1064-nm laser frequency. The results are fitted to a quadratic function ${f_{beat}} = {f_0} + \beta {({T - {T_0}} )^2}$; here, ${f_0}$ and $\beta $ are fitting parameters and the temperature T0 is where the linear thermal expansion is zero, which was determined to be 32.1(1) °C.
For the thermal time constant measurement (Fig. 6), the effective cavity temperature T was determined by inverting the function used to fit the thermal expansion data (Fig. 5) to give (for $T < {T_0}$):
This temperature T is evaluated as a function of time t and plotted in Fig. 6. The results are fitted to the equation:
Here, ${T_0}$ is the effective temperature as $t \to \infty $ and $\Delta T$ is the temperature change. A thermal time constant of $\tau = $ 16 hours is extracted from this dataset.
2.5 DC acceleration sensitivity
The cavity acceleration sensitivity of the OSRC was tested in all three directions by applying directional-dependent DC accelerations for roughly 10-20 s. The results are summarized in Fig. 7 below. The numbers correspond to the sides of the cubes against which 1 g is applied, where g is the acceleration due to gravity. It is found to be 8/10/0.2 × 10−11/g along directions perpendicular to the sides of the cubes. The largest acceleration sensitivity is a factor of 4 higher than the worst case obtained with a similar cavity spacer design [15], which we attribute to non-optimal tightening of the support screws in the titanium mounting frame. Future work aims to improve this performance metric.
Fig. 7. DC acceleration sensitivity of the OSRC measured by inverting the cavity along different directions and recording the beat note of the system against a long-term stable reference laser. The numbers correspond to the faces of the cubic spacer.
Following launch, the cavity will be subject to microvibrations in orbit that can be at the low μg level [27], which would contribute a few parts in 1016 to the cavity fractional frequency instability. For near-earth orbits, drag compensation is also possible to reduce the non-gravitational acceleration along the flight direction, giving a much-reduced contribution to the fractional frequency noise.
2.6 Fiber link stabilization
For high-performance ultrastable laser systems, it is necessary to compensate for frequency noise arising from temperature changes or acoustic vibrations in the several meters of optical fibers on the spacecraft. To mitigate such perturbations, a fiber stabilization system has been developed for the OSRC system. In this self-heterodyne setup [28], the phase between the signal after one round-trip in the fiber between the local and remote optical modules and the local phase at the fiber input is compared. This scheme is covered in detail in Fig. 8.
Fig. 8. Self-heterodyne fiber link stabilization setup. AOM: Acousto-optic modulator; BS: Beam splitter; C: Connector; Circ.: Circulator; DBM: Double-balanced mixer; FM: Fiber mirror; Osc: Oscillator; PD: Photo detector; PI: Proportional integral controller; PM-SMF: Polarization maintaining single mode fiber; Ref: Reference; VCO: Voltage-controlled oscillator.
At the optical input, the laser beam is split into two parts. In one branch the light passes via a circulator, is then frequency shifted by 40 MHz by an AOM and sent via a 6-m-long PM delivery fiber to the remote location where the laser signal is to be delivered. The beam is then partly reflected back through the delivery fiber, frequency shifted a second time by the AOM, transmitted via the circulator and mixed with the component of the laser signal directed into the other branch at the initial beam-splitter. The laser beat note signal detected by the photodiode occurs at twice the AOM modulation frequency and also carries the frequency noise induced by the delivery fiber and the other components between the first beam splitter and the fiber mirror. Using a PLL, the majority of the noise contribution of the fiber delivery system is compensated, as can be seen in Fig. 10.
The performance of this noise cancellation scheme has been estimated by calculating the Allan deviation of the time series of the beat note frequencies of the in-loop photodetector shown in Fig. 8 and an out-of-loop detector as indicated in Fig. 9.
Fig. 9. Schematic diagram of the test setup for measurement of the frequency instability (Allan deviation) of the stabilized fiber link. The in-loop detector is in the local module close to laser source, the out-of-loop detector is that close to the remote module.
Fig. 10. Performance of the fiber stabilization system measured at LTF. The Allan deviation (ADEV) of the laser beat note between the input and output of noise-stabilized fiber link is depicted. We note that fiber path length stabilization ensures compliance with the requirements of the most stringent optical clock and cavity applications.
The out-of-loop photodetector measures a beat note between the local phase delivered by a reference path (between first beam splitter and the beam splitter up-stream of the out-of-loop photodetector) and the signal delivered by the fiber stabilization system at the remote location. The in-loop beat note exhibits the doubled frequency shift of the AOM and twice the fiber induced frequency noise. Its frequency is at 80 MHz and is divided by 2, thus enabling comparison of in- and out-of-loop frequency instability. The measurements obtained are depicted in Fig. 10.
For the measurement without fiber noise compensation, the VCO generating the AOM signal was phase locked to the 10 MHz reference. This was done to measure the frequency instability induced by the fiber setup and not by the VCO. In this case (green curve), the relative frequency instability remains above 10−16.
With the noise-compensation activated, the frequency stability is significantly improved. The relative frequency stability measured for the out-of-loop signal is slightly degraded in comparison to the in-loop signal. The noise increase in the out-of-loop curve observed at around 1000 s integration time is related to the air-cooling system in the laboratory. Two effects can contribute to reduced out-of-loop performance. The first arises from frequency instabilities induced in the out-of-loop fiber path of the fiber noise stabilization system. The second results from the non-common path in the stabilizing interferometer, e.g., arising from the reference path.
The need for fiber length-stabilization will depend on the future space activity; for example, this will be necessary for those systems requiring high accuracy time and frequency capabilities. For space-based optical clocks intended for next-generation navigation or fundamental physics missions, an appropriate level of stabilization will likely be required to avoid potential loss of accuracy or stability in the optical system due to unstabilized fiber links between sub-systems.
2.7 Offset frequency locking
Optical clocks require the laser frequency to be tuned to an optical frequency given by the clock transition, which typically does not coincide with the relatively arbitrary cavity resonance frequency. For such applications, it is therefore desirable to lock the laser at an optical frequency offset to the cavity optical resonance. To achieve this, we implement an offset lock functionality to the control electronics. This is achieved via a different modulation scheme of the EOM with respect to standard PDH locking [29]. In standard PDH locking, the EOM is driven by the PDH frequency directly; for the offset locking phase the following signal is applied to the EOM:
Here ${V_{EOM}}$ is the voltage applied to the EOM, $\Omega_{offset}$ is the angular offset frequency of the laser with respect to the cavity resonance, and $\Omega_{PDH}$ the PDH angular frequency, at which the signal of the photodiode is demodulated to obtain the frequency lock error signal. While for the standard PDH locking applied for the results shown in Fig. 4, $\Omega_{PDH}\; $was 18.75 MHz, for the offset locking $\Omega_{PDH}$ is 12.5 MHz. This modulation is applied to the EOM with appropriate modulation index of the offset frequency tone and phase modulation index of the PDH tone ${\beta _2}$. In the upper panel of Fig. 11, the laser frequency is scanned around the cavity optical resonance at up to plus and minus twice the offset frequency foffset (75 MHz in this case). In the center of the applied laser frequency ramp, the cavity resonance is observed. PDH error signal features are then observed at ± foffset. These two discriminants have opposite slope, allowing the user to select the desired laser offset.
In Fig. 11, the beat note frequency between the OSRC-stabilized laser and a reference laser are shown when different values of foffset are set to the electronics. First the offset frequency is set to 70 MHz resulting in a beat note frequency of 217.6 MHz. When the offset frequency is varied between 65 MHz and 60 MHz, the beat note frequency changes from 212.6 MHz to 207.6 MHz, indicating that the laser achieves the lock to the cavity at those different offset frequencies. The same behavior is also confirmed for further values of offset frequency. In the experiment, the offset frequency is changed by the user in the electronics control software. The lock is then achieved automatically with the automatic (re)lock algorithm described previously.
Fig. 11. Offset locking demonstration. Top Panel: Oscilloscope trace of the offset lock error signal and cavity transmission signal; the frequency scale is indicated by the 75 MHz frequency interval shown, corresponding to the offset frequency chosen for the measurement. Bottom Panel: Beat note frequency between cavity stabilized laser and reference laser versus time for different values of offset frequency.
3. Environmental testing relevant to space qualification
In order to qualify the current TRL of our reference cavity, we pursue in-depth environmental testing including radiation exposure and vibration tests. In these efforts, the cavity finesse is monitored in order to probe any potential degradation in system performance.
3.1 Cavity finesse measurements
The cavity finesse was determined via the cavity ring-down (CRD) time both before and after irradiation. Output from a continuous-wave tunable distributed feedback (DFB) diode laser was propagated through a collimator and two biconvex lenses on a mechanical translation stage, allowing for movement along the propagation axis to permit adjustments of the beam waist inside the cavity. The transmitted light from the cavity was measured via an InGaAs detector with an intrinsic rise/decay time of 165 ns and was also mapped in two dimensions with an IR camera. The transmitted signal from the cavity was sent to an oscilloscope and analyzed.
Experiments were performed by exciting the TEM00 mode and measuring the decay time for a series of rectangular pulse trains (Fig. 12(a)). The modal profile was verified using an IR camera. In Fig. 12(c), we summarize the results of multiple measurements performed before and after the radiation exposure. The TEM00 mode shows a characteristic lifetime of 10.81(0.35) μs with a corresponding finesse of 203 700(6 600); the number in parenthesis is the estimated standard uncertainty. In addition, we have also monitored the decay times of a TEM01 mode (Fig. 12(b)), which was selected and excited by tuning the waist position by means of a translation stage. In this case, shorter decay times of 8.52(0.38) μs were observed (Fig. 12(d)) with a corresponding finesse of 160 500(7 100).
Fig. 12. Cavity ringdown characterization results. Decay of the (a) TEM00 and (b) (TEM01) modes and the corresponding mode profiles (insets). (c) Cavity finesse of the TEM00 and TEM01 modes before (left) and after (right) irradiation, showing no degradation in performance from high-energy proton exposure. (d) Lifetimes of the TEM00 and TEM01 modes before (left) and after (right) proton irradiation.
Following the initial round of CRD characterization, the cavity was unmounted and sent to the Trento Institute for Fundamental Physics and Applications (TIFPA) for radiation exposure. The details of the proton exposure tests are provided below.
3.2 Radiation exposure
To probe potential sensitivity to radiation damage of the cavity mirror crystalline coatings, the OSRC cavity (comprising the cubic spacer with optically-contacted crystalline mirrors, mounted inside the support structure with thermal shields, and held under vacuum within its custom chamber) was irradiated with protons at the Physics and Engineering Beamline of the Research Laboratory of the Proton Therapy Centre, managed by TIFPA, a research center of the Italian National Institute for Nuclear Physics (INFIN) [30] of the Autonomous Province of Trento [31]. The proton accelerator and irradiation facilities were built and are currently operated by IBA [32]. According to the test plan, the cavity in its evacuated vacuum chamber (akin to the system configuration in space) was exposed to proton fluences and energies that simulate those relevant to a 5-year Low Earth Orbit (LEO) mission operational lifetime. At the time of the experiment, not all orbit parameters were defined, and, as such, ADM-Aeolus values [33] were taken as a reference. Figure 13 (left) depicts the integral trapped proton flux for ADM-Aeolus and solar minimum conditions (AP8 min) multiplied by 1.58 × 108 s (5 years) to indicate the integrated fluence. The planned irradiation (energy, fluence) points simulating the spectrum are also shown.
Fig. 13. Radiation exposure tests of the vacuum-mounted OSRC. (Left) Integrated trapped proton flux for solar minimum conditions in 1.58 × 108 seconds. (Right) Photograph of the chamber showing the alignment laser for maintaining irradiation centered in the optical access windows.
During testing, the vacuum chamber was mounted on a rotation and translation stage with height adjustment. The linear translation stage was used for fine positioning in a direction orthogonal to the proton beam while the rotation stage allowed the chamber to be remotely rotated by 180° to irradiate both optical windows. The chamber was accurately positioned by means of a laser previously aligned with the proton beam (Fig. 13, right).
The accelerator can deliver protons with energies in the 70-220 MeV range. The 37 MeV and 46 MeV energies were obtained by passing the 70 MeV beam through RW3 (‘solid water’, white-polystyrene and polystyrene with titanium oxide admixture) plates of 20 mm and 25 mm thickness respectively. Irradiation energies of 70, 100, and 200 MeV were added to cover the irradiation spectrum in the higher energy range.
Having adopted an axial irradiation scheme, simulations were carried out to estimate the quantity of protons absorbed in the input window and compare this with the count that would be transmitted and reach the opposite mirror and window. The simulations were performed with the SRIM software package [34]. The simulations showed that the 37 and 46 MeV protons are completely absorbed in the first chamber window (5 mm thick BK7) and fused silica (6.35 mm thick SiO2) substrate of the first cavity mirror (the high-reflectivity GaAs/AlGaAs crystalline multilayer is only 6.2 µm thick and its contribution was ignored). Almost all (above 99%) of the 70, 100, and 200 MeV protons pass through the first optical window and cavity substrates, reaching the second optical window with a mean energy of 46.5, 82.7 and 189.9 MeV respectively. The simulations show that the transmitted protons exhibit minimal lateral deflection, with the exposure region still within the area of the second window. Considering the contribution of the transmitted protons (and the energy change) when compared with the total irradiation of the windows, the target 46 MeV proton fluence was reduced by the value of the 70 MeV target fluence, whereas the 82.7 MeV and 189.9 MeV protons were combined with the 100 MeV and 200 MeV protons, respectively, and the irradiation fluences for these two energies were corrected accordingly, delivering half of the fluence on each side.
To prevent irradiation of the metallic portions of the chamber (which are more easily activated), a circular collimator of 8 mm diameter (area of 50.3 mm2) was added to the beam exit. An ionization chamber was placed between the collimator and the cavity chamber to measure the fluence. Table 1 reports the expected fluence values.
Table 1. Planned and delivered fluence/dose
After irradiation, the optical windows were inspected with a high-purity germanium (HPGe) radiation detector to verify possible activation. As expected, no induced radioactivity was measured. Following this process, the cavity was re-mounted into the CRD setup, and the decay times were again characterized to probe for any radiation-induced changes in the mirror optical losses. As shown in Fig. 12(c) and 12(d), the decay times and, correspondingly, the cavity finesse show negligible degradation (within statistical errors). This indicates that the optical properties of the high-reflectivity crystalline mirrors were not adversely affected by proton irradiation. Similar results have been observed for crystalline mirrors subject to 1.173 MeV and 1.332 MeV gamma ray exposure at the NASA Goddard Space Flight Center [35]. In those experiments, no degradation in optical performance was observed for a total dosage of 40 krad, double the estimated lifetime exposure for the current LISA spacecraft design.
3.3 Vibration and thermal testing
One of the most critical environmental tests is the vibrational test for launch. For these efforts, the full OSRC optomechanical apparatus (including the optical cavity comprising the cubic spacer with optically-contacted crystalline mirrors, mounted inside the support structure with thermal shields, and held under vacuum within its custom chamber) has been tested to 10 g RMS random vibration (see Fig. 14) and up to 20 g sinusoidal excitation at 21 Hz in all spatial directions. A resonance search was applied before and after each test to check if the characteristics has changed. No significant difference was detected. The assembly was additionally tested for shock resistance by applying a peak acceleration of 360 g. No degradation in performance nor misalignment of the cavity was detected in follow-on characterization efforts.
Fig. 14. Example outcome of a random vibration test of the mounted cavity. Here, 10 g was applied along the optical axis of OSRC with a spectrum indicated with the blue reference line. The orange line is the response of the accelerometer mounted on the OSRC support, while the green line is the response of the OSRC vacuum chamber near the ion pump.
The optomechanical assembly was then subject to a non-operational thermal vacuum test, being subjected to 3 cycles from -40 °C to 60 °C with a dwell time of 3 hours and a temperature ramp rate of 1 K/min as shown in Fig. 15. Before and after this test, the system performance was measured and showed no significant difference.
Fig. 15. Thermal cycling of the OSRC. AP, IF, and CP denote thermal sensors located at the interface of the mounting frame and shroud, which is the source of the temperature variation. The next four sensors (bottom, side bottom, side top, and top) are in contact with the chamber walls. An additional sensor on the shroud monitors the applied environmental temperature. For completeness the pressure of the environment is shown as well.
4. Conclusion and outlook
The work reported here was carried out within the ESA-funded OSRC consortium. This effort targets the development of a space-compatible optical reference cavity providing low fractional frequency instabilities for future application in, for example, next generation gravity missions or satellite-based optical atomic clocks. The system pursued in these efforts relies on a low-acceleration-sensitivity cubic cavity designed by NPL incorporating GaAs/AlGaAs-based crystalline end mirrors from CMS (now Thorlabs Crystalline Solutions) for enhanced Brownian noise suppression. This cavity is mounted in a titanium retaining frame with low-emissivity shielding contributed by NPL and integrated in a custom vacuum system developed by AirbusDS and Sodern, with AirbusDS being responsible for testing the overall OSRC system performance, including mechanical and thermal resilience. Locking electronics capable of effective RAM suppression [36], as well as automatic (re)lock acquisition and offset tuning via electro-optic sideband modulation [29], have been developed and demonstrated by CSEM, with fiber link stabilization contributed by LTF. These capabilities are required for space applications where stable and automated locking to atomic frequency standards over extended time periods will be of key importance. Design analysis for vibration and thermal testing was performed by STI. Finally, environmental testing of the acceleration and radiation (proton) sensitivity of the cavity in vacuum chamber (led by the teams at AirbusDS and FBK respectively) show no degradation in performance. These results were verified by pre- and post-exposure finesse measurements, complementing previous efforts involving gamma exposure of crystalline mirrors and confirming TRL 6 has been reached for this subsystem.
The totality of these environmental tests, including irradiation, vibration, shock, and thermal cycling of the OSRC optomechanical apparatus (including the optical cavity comprising the cubic spacer with optically-contacted crystalline mirrors, mounted inside the support structure with thermal shields, and held under vacuum within its custom chamber) indicate TRL 6. The remaining custom components and subsystems that make up the full cavity stabilization system (laser, electronics, fiber- and free-space optics not attached to the chamber) remain at TRL 4. However, it is clearly the case that there are space-qualified components and subsystems that are similar to these latter items that have been flown in other space missions and in-orbit demonstrations. Combining these existing technologies with the examples shown here will yield a deployable system with a very high TRL.
By employing a low-acceleration sensitivity cubic spacer and crystalline mirror coatings, the ESA OSRC project aims to push the achievable thermal noise performance of a transportable ultrastable laser system to the mid 10−16 fractional frequency range. In follow-on efforts, additional cavities employing crystalline coatings with a center wavelength of 1397 nm will be investigated, targeting space optical lattice clocks based on Sr. Additional performance improvements may be realized by incorporating more advanced coating designs, including thermo-optic-noise optimized crystalline multilayers [37].
Funding
European Space Agency (contract 4000119670/17/NL/BJ).
Acknowledgments
This work was funded by the European Space Agency GSTP programme (ESA project Optical Stabilising Reference Cavity) with support of the respective national space agencies.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. [Online]. Available: https://www.dlr.de/kn/en/desktopdefault.aspx/tabid-17411/.
2. L. Massotti, “Next Generation Gravity Mission design,” ESA, [Online]. Available: https://indico.esa.int/event/416/contributions/7507/attachments/4913/7529/Clean%20Space%20Industrial%20Days_NGGM%20w_video_v3.pdf. [Accessed 3 May 2023].
3. K. Bongs, Y. Singh, L. Smith, et al., “Development of a strontium optical lattice clock for the SOC mission on the ISS,” C. R. Phys. 16(5), 553–564 (2015). [CrossRef]
4. “LISA factsheet,” ESA, [Online]. Available: https://www.esa.int/Science_Exploration/Space_Science/LISA_factsheet. [Accessed 3 May 2023].
5. “Technology Readiness Levels (TRL),” ESA, [Online], Available: https://www.esa.int/Enabling_Support/Space_Engineering_Technology/Shaping_the_Future/Technology_Readiness_Levels_TRL [Accessed 3 May 2023].
6. R. Pierce, M. Stephens, P. Kaptchen, et al., “Stabilized Lasers for Space Applications: A High TRL Optical Cavity Reference System,” in Conference on Lasers and Electro-Optics, paper JW3C.3 (2012).
7. B. Argence, E. Prevost, T. Lévèque, et al., “Prototype of an ultra-stable optical cavity for space applications,” Opt. Express 20(23), 25409–25420 (2012). [CrossRef]
8. D. Świerad, S. Häfner, S. Vogt, et al., “Ultra-stable clock laser system development towards space applications,” Sci. Rep. 6(1), 33973 (2016). [CrossRef]
9. Y. Luo, H. Li, Y.-Q. Li, et al., “Prototype of a monolithic cavity-based ultrastable optical reference for space applications,” Appl. Opt. 60(10), 2877–2885 (2021). [CrossRef]
10. P. Zhao, J. Deng, X. Chengwen, et al., “A Spaceborne Mounting Method for Fixing a Cubic Fabry–Pérot Cavity in Ultra-Stable Lasers,” Appl. Sci. 12(24), 12763 (2022). [CrossRef]
11. C. Xing, J. Xiong, P. Zhao, et al., “Development of space-borne transportable high-finesse Fabry–Pérot cavity and its performance in ultra-stable laser,” in Proceedings Volume 12711, Third Optics Frontier Conference (OFS2023).
12. L. Chen, X. Hu, L. Han, et al., “Theoretical and experimental study on vibration sensitivity of cubic cavity for space applications,” Opt. Laser Technol. 158(A), 108915 (2023). [CrossRef]
13. J. Liu, T. Liu, L. Chen, et al., “A compact sub-hertz linewidth Fabry Perot cavity frequency stabilized laser for space application,” Opt. Laser Technol. 136, 106777 (2021). [CrossRef]
14. C. Greve, K. Dahl, G. P. Barwood, et al., “Space based lasers for gravitational wave detection,” in Laser Congress (Optica Publishing Group, 2019), paper CM3C.4.
15. S. Webster and P. Gill, “Force-insensitive optical cavity,” Opt. Lett. 36(18), 3572–3574 (2011). [CrossRef]
16. G. D. Cole, W. Zhang, M. J. Martin, et al., “Tenfold reduction of Brownian noise in high-reflectivity optical coatings,” Nat. Photonics 7(8), 644–650 (2013). [CrossRef]
17. G. D. Cole, W. Zhang, B. J. Bjork, et al., “High-performance near- and mid-infrared crystalline coatings,” Optica 3(6), 647–656 (2016). [CrossRef]
18. K. Numata, M. Ando, K. Yamamoto, et al., “Wide-Band Direct Measurement of Thermal Fluctuations in an Interferometer,” Phys. Rev. Lett. 91(26), 260602 (2003). [CrossRef]
19. G. Xu, D. Jiao, L. Chen, et al., “Thermal Noise in Cubic Optical Cavities,” Photonics 8(7), 261 (2021). [CrossRef]
20. R. W. P. Drever, J. L. Hall, F. V. Kowalski, et al., “Laser Phase and Frequency Stabilization Using an Optical Resonator,” Appl. Phys. B: Photophys. Laser Chem. 31(2), 97–105 (1983). [CrossRef]
21. S. D. Penn, M. M. Kinley-Hanlon, I. A. O. MacMillan, et al., “Mechanical ringdown studies of large-area substrate-transferred GaAs/AlGaAs crystalline coatings,” J. Opt. Soc. Am. B 36(4), C15–C21 (2019). [CrossRef]
22. T. Legero, T. Kessler, and U. Sterr, “Tuning the thermal expansion properties of optical reference cavities with fused silica mirrors,” J. Opt. Soc. Am. B 27(5), 914–919 (2010). [CrossRef]
23. I. R. Hill, R. J. Hendricks, S. Donnellan, et al., “Dual-axis cubic cavity for drift-compensated multi-wavelength laser stabilisation,” Opt. Express 29(22), 36758–36768 (2021). [CrossRef]
24. “Emissivity Coefficients common Products,” [Online]. Available: https://www.engineeringtoolbox.com/emissivity-coefficients-d_447.html. [Accessed 9 June 2023].
25. K. Dahl, K. Nicklaus, M. Herding, et al., “High stability laser for interferometric Earth gravity measurements,” Proc. SPIE 10562, 76 (2017). [CrossRef]
26. M. Schioppo, J. Kronjäger, A. Silva, et al., “Comparing ultrastable lasers at 7 × 10^-17 fractional frequency instability through a 2220 km optical fibre network,” Nat. Commun. 13(1), 212 (2022). [CrossRef]
27. L. Li, M. Zhou, Y. Zhu, et al., “Satellite microvibration measurement based on distributed compressed sensing,” Measurement 203, 112031 (2022). [CrossRef]
28. L.-S. Ma, P. Jungner, J. Ye, et al., “Delivering the same optical frequency at two places: accurate cancellation of phase noise introduced by an optical fiber or other time-varying path,” Opt. Lett. 19(21), 1777–1779 (1994). [CrossRef]
29. J. I. Thorpe, K. Numata, and J. Livas, “Laser frequency stabilization and control through offset sideband locking to optical cavities,” Opt. Express 16(20), 15980–15990 (2008). [CrossRef]
30. “Trento Institute for Fundamental Physics and Applications,” [Online]. Available: http://www.tifpa.infn.it/. [Accessed 3 May 2023].
31. F. Tommasino, M. Rovituso, S. Fabiano, et al., “Proton beam characterization in the experimental room of the Trento, Proton Therapy facility,” Nucl. Instrum. Methods Phys. Res., Sect. A 869, 15–20 (2017). [CrossRef]
32. “IBA Industrial Solutions website,” [Online]. Available: https://www.iba-industrial.com/
33. E. S. Agency, “CEOS EO Handbook - Mission Summary - Aeolus,” 2023. [Online]. Available: http://database.eohandbook.com/database/missionsummary.aspx?missionID=377.
34. J. F. Ziegler, M. D. Ziegler, and J. P. Biersack, “The stopping and range of ions in matter,” Nuclear Instruments and Methods in Physics Research Section B 268, 93–129 (2010). [CrossRef]
35. G.-W. Truong, K. Numata, C. Nguyen, et al., “Gamma Radiation Exposure of Crystalline Coatings for Space Applications,” in Optical Interference Coatings Conference (Optica Publishing Group, 2019), paper FA.4.
36. W. Zhang, M. J. Martin, C. Benko, et al., “Reduction of residual amplitude modulation to 1 × 10^-6 for frequency modulation and laser stabilization,” Opt. Lett. 39(7), 1980–1983 (2014). [CrossRef]
37. T. Chalermsongsak, E. D. Hall, G. D. Cole, et al., “Coherent cancellation of thermo-optic noise in GaAs/Al0.92Ga0.08As Bragg mirrors,” Metrologia 53(2), 860–868 (2016). [CrossRef]
