|
|
1.MINIATURIZED OPTICAL SYSTEMS FOR QUANTUM SENSINGDue to their unrivalled accuracy and precision, atomic quantum sensors and clocks operated in field or space may improve on applications in the areas of navigation, timekeeping, exploration, and can be used to address fundamental questions of modern physics.1 The transition from laboratory-based setups to field- and space-compatible devices requires a high degree of integration and miniaturization to fulfill requirements on size, weight, and power (SWaP) budgets. Moreover, a high environmental robustness must be achieved for operation on mobile platforms, i.e. with regard to mechanical and thermal loads and radiation hardness. The physics package of a cold atomic quantum sensor consists of multiple key components, usually including atom source, ultrahigh vacuum system and the surrounding optics setup. In many cases an optical dipole trap (ODT)2 is used for optical trapping and cooling of the atomic ensemble that can be realized by far detuned optical light fields tailored to the used species and foreseen application. Such an ODT relies on high light intensities and gradients, which are realized by focusing high power laser to narrow waists. A configuration to achieve strong spatial confinement of the atoms in all three dimensions is a crossed beam ODT, where two highpower laser beams are tightly focused and precisely overlapped at a specific angle. This requires laser light to be guided around the cold atom setup with mirrors, beam-splitters, waveplates, polarizers and subsequently focused and overlapped inside the ultra-high vacuum chamber. Although such a large-aperture setup yields very small optical waists and thus large trap depths, it requires frequent re-alignment, is sensitive to thermal fluctuations or mechanical disturbances and occupies a rather large volume (about 2 to 3 L). While being feasible for lab-based setups, another approach is necessary for mobile setups on earth or microgravity platforms. Within our work at the Ferdinand-Braun-Institut (FBH) we address the miniaturization and integration of the quantum sensor’s physics package to small field and space compatible setups. Advanced micro-integration methods are established for the miniaturization of narrow linewidth diode lasers,3 proved their suitability in sounding rocket missions4–7 and will be used for the diode-based laser sources for the BECCAL mission aboard the ISS.8 These methods are adapted and used to realize a compact crossed beam optical dipole trap displayed in Fig. 1. The setup is operated at 1064nm and injected by a light source via a single-mode, polarization-maintaining fiber and a collimator, generating a beam of 1 mm diameter on a ceramic bench. After the collimator, the light is polarized and split into two paths with perpendicular linear polarization. Both paths are focused to a waist radius of wR = 34 μm and precisely overlapped. The center aperture of the optical system gives clearance for one of the beams of a magneto-optical trap (MOT), a first stage to trap and cool atoms. The overall optical system has a footprint of 36 × 43 mm2 and a volume of ~25 mL including optics. The optical setup can be operated with optical powers up to 2 W cw and exhibits an exceptional thermal and mechanical stability when operated at up to 60 °C during tests. Currently the miniaturized crossed beam ODT is operated within a cold atom experiment. In summary, the micro-integration approach led to a substantial miniaturization of the optical system and simultaneously a high passive stability, both key factors for future applications on mobile and space-based platforms. Figure 1.Miniaturized optical system to be used in a cold atom-based quantum sensor. Two beams (1064 nm, 1W each) with overlapping focus (wR = 34 μm) create a so-called optical dipole trap, which can be used for trapping, cooling and manipulating rubidium atoms.  Based on this setup, we work on combining optical systems with an atom chip.9 The next generation of such chips is currently being developed at the Institute of Micro Production Technology (IMPT)10 and is based on silicon wafers with current conductor paths manufactured by micro-fabrication methods and mirror coatings. Such a chip allows to generate ultra-cold atom ensembles with high cycle times, enabled by steep magnetic fields with high trapping frequencies, forming, e.g., chip-based Ioffe-Pritchard traps. Ultimately, we aim to combine a micro-integrated optical setup with an atom chip within the ultra-high vacuum (UHV) environment of the physics package. This enables atom trapping configurations like purely optical or hybrid magneto-optical traps in the vicinity of the atom chip surface, while achieving a compact overall setup with high alignment stability. Such a setup requires miniaturized optical setups with high pointing stability and UHV compatibility. To assemble our optical systems, adhesive integration methods are used. They have the advantages of ease of application, short setup times, low cost and suitable mechanical and thermal properties. In our experience, they enable precise alignment of micro-optics, and the integration process is highly reproducible. The application of quantum sensors in field and space imposes further requirements on their e.g., mechanical, thermal, and radiative resilience. 2.MECHANICAL AND ENVIRONMENTAL QUALIFICATION OF ADHESIVESTo qualify the used adhesive integration methods for the desired applications, mechanical and environmental tests are conducted. The mechanical properties of an adhesive joint, i.e. bond strength, modulus, …, depend not only on the used adhesive. The materials involved, their surface properties, cleaning, adhesive layer thickness and the curing procedure itself have a major influence on the formation of the adhesive’s polymer network and thus the mechanical properties of the joint. In this study, test samples of adhesive joints with application-related geometries and materials are set up and environmentally aged by thermal cycling and radiation exposure. Subsequently their mechanical bond strength is characterized by shear strength tests. 2.1Sample materials and setupThe investigated samples and setup procedures are described in Ref. 11 and summarized in the following. The samples consist of precision fused silica glass blocks with a footprint of 2 × 4 mm2, a typical bond area for our miniaturized optical systems. For the presented shear strength tests the glass bodies are cuboid. In the future, also tensile strength tests with trapezoidal sample bodies are planned. The lateral surfaces are polished to ensure reproducible UV light transmission for curing the adhesive. To ensure a reproducible bond area, the glass surface forming the bond is of very low roughness (surface profile height Rt ≤ 5 nm). Simple silicon wafers are used as the substrate. Pre-tests with surface variations, e.g., by passivation or gold coating, showed similar behavior in terms of die shear strength and failure mode. For easy handling, the wafers are cut into 10 × 10 mm2 squares. The test samples are set up with different epoxy adhesive systems, cured according to the datasheets and their bond strength systematically investigated. The four different one or two component epoxy adhesives used, are common in optomechanics industry and have different curing mechanisms:
Fig. 2 shows the setup of the adhesive samples. Prior to assembly, all components are cleaned with isopropanol in an ultrasonic bath to remove contaminations e.g., by hydrocarbons and particles. Subsequently, the samples are wiped with acetone and a lint-free lens tissue, dried with dry nitrogen, inspected with a microscope and any remaining particles removed with a fiber optics cleaning swap. The glass test body is then attached to a micro-integration system to align it parallel to the substrate and to determine the zero-gap position. Next, the test body is lifted off the substrate and 240 nL of adhesive is applied with a pneumatic dispenser system which, after calibration for the specific adhesive under test, allows precise adhesive application down to 20 nL. By moving the test body back towards the substrate, the gap for the adhesive layer is set to 30 μm and the applied adhesive is cured with UV light using an intensity-calibrated mercury light source. The UV-light is guided to the sample via two optical fiber light guides fixed at a defined distance in the sample holder, compare Fig. 2, in order to ensure reproducible exposure for curing of the adhesive layer. 2.2Environmental aging and mechanical adhesive testsThe general requirements, the adhesive joints need to fulfill, are summarized in Ref. 11. Within this work, we specifically address influences of thermal and radiative stress on the shear strength of adhesive bonds. The selected test conditions follow the ESCC 23201, and the standards specified therein. Thermal cycling is performed at IMPT with a Vötsch VT3 7006 S2, which features one hot and one cold temperature chamber and thus quick cycling between temperatures. Radiative stress is applied by exposure to a cobalt-60 source at the Helmholtz Zentrum Berlin, emitting gamma rays at 1.17 MeV and 1.33 MeV. The adhesive samples are exposed to the following sets of different environmental conditions:
For each test condition and each adhesive, one sample set, consisting of six samples, is prepared. Thus, in total 120 test samples are assembled. The mechanical bond strength of the adhesive joints is evaluated by destructive shear strength testing according to MIL-STD-883L Test Method 2019.9 ’Die Shear Strength’ with a bond tester (Nordson Dage 4000 Plus), compare Fig. 2, with an accuracy ≤ 1 % for the recorded force values. The results are displayed in Fig. 3 and compared to the required shear strength to fulfill MIL-STD-883L Test Method 2019.9 for an 8mm2 die of 24.5 N (factor 1.0x) or 49 N (factor 2.0x). Additionally, on the second y-axis the shear strength is scaled to g-force, assuming a weight of 0.044 g for the fused silica test body. In general, a scattering of the measurement results can be observed, although particular care was taken to keep the sample preparation, cleaning, and set-up procedure as comparable as possible for all samples. This can be caused by multiple reasons during the set-up or test procedure, e.g., residual contamination on the interface surfaces or microscopic variations of their structure, poor alignment precision of the samples, the adhesive pot life and thus variation of adhesive viscosity during sample preparation or human error when assembling and testing the samples. However, during set-up of our micro-optical systems we expect the same influences on the bonding process and thus the study yields representative results. Another error source might be the alignment of the die shear tool to the test body in the bond tester, which can lead to a deviation from the desired shear test geometry to a shear and torsion test geometry. The bond tester first aligns the shear tool automatically to the silicon substrate via touchdown. Subsequently the shear tool is aligned optically to the glass block with an expected accuracy of ≤1°. Additionally, a low force rotation joint on the machine corrects for small deviations once the shear tool touches the glass test body. Thus, we assume the error of the test method to be significantly smaller than the observed scattering of the shear test values. A detailed discussion of the of the results displayed in Fig. 3 follows at the end of this section. Figure 3.Mechanical bond strength of the tested adhesives evaluated by shear strength testing and displayed as boxplot. Outliers from the boxplot evaluation are displayed as grey diamond. In addition, the measurement points are plotted as an intuitive guide to the results.  The failure mechanism of an adhesive joint is categorized according to DIN EN ISO 10365, compare Ref. 12, as follows:
Microscope images of substrate and glass block are recorded after shear strength test for every sample and categorized by their failure mechanism. Exemplary images for each failure mode are depicted in Fig. 4. The results of the shear strength measurements are assigned to the failure modes, as depicted in Fig. 5. Furthermore, the fracture images are evaluated whether the bond has a fillet of adhesive. This describes a partial adhesive joint at the lateral surfaces of the glass body formed by adhesive flowing out of the gap. Such a geometry is expected to increase the bond strength and influence the failure mode.13, 14 If the image shows a fillet on at least three sides, the bond is categorized as ’with fillet’ and marked accordingly in Fig. 5. Figure 4.Exemplary microscope images for the different adhesive failure modes of glass block (upper) and substrate (lower), recorded after die shear strength test. The failure mode is categorized according to the fracture pattern and their characteristics explained in the text.  Figure 5.Failure mechanisms of the tested adhesive joints categorized according to the DIN EN ISO 10365 for the four adhesives and different environmental stresses according to Section 2.2. Additionally, the marker style indicates whether the sample had an adhesive fillet. If the image of a sample after shear testing shows a fillet on at least three sides, the bond is categorized as ’with fillet’ and marked accordingly.  This evaluation gives valuable information for optimization of the adhesive bond. As shown, fracture modes vary for the chosen adhesives, the environmental stress and within each sample set. The highest bond strength is achieved by either JF or CF fracture modes, which is an expected behavior.15 In this case the bonding at the interface is stronger than the materials and either the adhesive layer (CF) or the Si wafer or the glass block (JF) fracture, but not the interfaces of the joint. As expected,13 these failure modes correlate very strongly with the occurrence of the fillet due to beneficial force application into the joint and less stress onto the interfaces of adhesive and adherents during die shear strength testing. For the samples with the lowest bond strength the failure mode is AF or ACF, a complete or partial adhesion failure at one of the interfaces between adhesive and adherent. In this study, the failure occurred mostly at the interface to the fused silica test body. This is due to insufficient bonding of the adhesive’s polymer chains to the substrate and can be caused by contamination on the glass surface or insufficient wetting. The presented shear strength dataset in Fig. 3 shows different influence of the environmental stresses, evaluated by die shear testing for four different adhesives. The fracture mode and the influence of the adhesive fillet is evaluated from microscope images of substrate and glass block and compiled in Fig. 5. Bond strength of adhesive 1 increases with exposure to environmental stresses. The reference set does not pass the MIL-STD-883L Test Method 2019.9. Thermal aging according to TC A or TC B increases the bond strength of adhesive 1. This is also the case for radiation exposure Rad and a combination of both Rad + TC B. The sample set for Ref and Rad do not have fillets and exhibit AF and ACF failure modes. Hence these sets are not fully comparable to the other environmentally aged sample sets. However, the samples aged with TC A, TC B and Rad & TC B are comparable and the bond strength increases with these environmental stresses. This could indicate that, although this is a purely UV curing adhesive, the adhesive is not fully cured by the UV light and contains uncured monomers. The fast UV-cure of the adhesive could also lead to internal stresses in the polymer network. Application of external thermal or radiative energy increases the cross-linking of the polymer network and polymers can re-arrange, leading to a better cured adhesive layer with less internal stress. The bond strength of the second purely UV curing type, adhesive 2, shows much less variation when exposed to thermal and radiative stresses and the bond strength increases after thermal exposure. The sample set for adhesive 2 shows comparable failure modes and occurrence of fillets for all assembled test samples. For this adhesive the bond strength exhibits a small decline when exposed to γ-radiation according to Rad compared to the reference set. This indicates that the cross-linking during UV cure leads to a high degree of cure, while the radiation exposure causes a partial destruction of the cross-linked polymer network and hence a reduced bond strength. Subsequent thermal stress according to TC B starts the polymer cross-linking again at the broken bonds and thus an increased bond strength is achieved.16 The results for adhesive 3 in Fig. 3 exhibit the least variation of the bond strength with respect to the investigated thermal and radiative environmental loads. This hybrid curing adhesive passes the MIL-STD-883L Test Method 2019.9 requirement with factor 2x for each sample set and the variation of the sample mean for external stresses is within the error of each measurement. Furthermore, the variance for each stress is less compared to adhesive 1 and 2, which could be caused by reduced internal stress due to the additional heat curing. During the UV cure the polymer network forms rapidly, resulting in internal stresses within the network. Heating the sample can lead to a re-arrangement of the polymers when thermal part of the cure sets in, resulting in reduced stress and increased bond strength. Adhesive 4 exhibits the highest bond strength values measured within this study, though standard deviation within each sample set is also very high (up to 39.78 N for TC B sample set). The bond strength of this hybrid curing adhesive increases when exposed to thermal or radiative stress compared to the reference set. The large scattering of the overall sample set for adhesive 4 can be explained by the more complex application procedure, cure schedule and shorter pot life for this adhesive. The shorter pot life probably also caused the sample set for Rad & TC B to not have fillets, since this set was set-up last and due to increased viscosity less adhesive is applied. Interestingly, the JF fracture mode still occurred basically only for this set. This shows a very strong interface between adhesive and glass block and an adhesive layer with high shear strength. 2.3Influence of plasma cleaning on bond strengthOne way to improve the bond strength of adhesive joints is plasma cleaning of the interface surfaces. Especially for joints that exhibit failures at the interface of adhesive and adherent (AF, ACF) plasma cleaning before assembly is expected to increase the bond strength. A partially ionized gas, which in laboratory applications is usually generated by electric fields, is accelerated onto the surface that should be cleaned. Residual molecules on this surface are first physically ablated by this incident high-energy ions. For non-oxidizing materials often oxygen plasma is used, which then also chemically reacts with residual hydrocarbons on the surface. This process creates free bond sites on the surface, resulting in a better formation of chemical bonds with the epoxy polymer and hence increased bond strength. The effect of two different plasma sources on the adhesive shear strength is investigated. The first source is a vacuum chamber plasma, where the samples are placed in a chamber and after pump-down treated with oxygen plasma for 30 s. This source has the advantage that it treats large surface areas or multiple samples at once and that the gaseous process products are directly pumped away. But since this is a rather large setup it is not available directly at the micro-integration facility. Therefore, the samples must be transported, which increases the time from surface cleaning/activation to assembly while the surface is exposed to air. This leads to an increasing occupation of the free bond sites on the surface, reducing the positive effects on the bond strength. The second plasma source is the μPQ (ultra-compact atmospheric pressure microwave plasma source, see Ref. 17), shown in Fig. 6. It is an in-house development of the FBH and generates a localized jet of atmospheric plasma (power 10 W to 20 W) and has a compact footprint (114 × 33 × 25mm3). Here, gaseous process products are transported away by a constant stream of plasma and air from the source. In contrast to the chamber source, this compact device can be used directly at our micro-integration facility. Thus, the freshly cleaned and activated sample can be assembled directly after cleaning. The plasma is confined to a thin plasma jet, allowing a selective surface treatment, e.g, on substrates for optical assemblies as shown in Section 1. Our samples were cleaned for 6.5 s at 12 mm from the source. Figure 6.Left: the compact microwave plasma source (μPQ), emitting a plasma jet at its nozzle (red circle). Right: influence of plasma surface cleaning on the shear strength of adhesive 1. A vacuum chamber plasma treatment and the compact microwave plasma source are investigated.  For this study adhesive 1 is used, since the sample set-up is best controlled. Also, this adhesive shows the lowest bond strength and also interface-related failure modes (AF, ACF) for the samples without fillet, compare to Section 2.2. Sample geometries, preparation and assembly is performed according to Section 2.1. As additional step plasma cleaning is performed after ultrasonic cleaning of the components. In addition to the reference set from Section 2.2, one sample set for each plasma cleaning method is assembled. To achieve comparable results, the samples were all prepared without fillet, which was confirmed by checking the microscope images after the die shear test. The results of the die shear strength test are compiled in Fig. 6. The reference set has a mean bond strength of 19.99 N and does not pass the MIL-STD-883L Test Method 2019.9 requirement of 24.5 N (factor 1.0x). After both plasma cleaning methods, the shear strength increases substantially and reaches a mean value of 84.68 N for the chamber plasma source and 78.67 N for the μPQ source. All samples pass the MIL-STD-883L Test Method 2019.9 requirement of 49 N (factor 2.0x), even though there is one outlier for the μPQ plasma source. In summary, the shear strength increased by a factor of approximately four compared to the reference set. Since all samples did not have fillets and interface failure modes occurred for the reference set this improvement is expected. The sample surfaces have fewer residual contaminations after the plasma cleaning and free bond sites at the interface surface allow for better chemical bonding of the polymer network. Both plasma cleaning methods increased the bond strength by a similar amount and the differences are within the margin of error. Further studies with an increased sample size, investigation of the effect for other failure modes, adhesives and environmental stresses need to be conducted in future. 3.SUMMARY AND OUTLOOKMicro-integration of the optical systems for atom trapping and manipulation is one key method to realize compact, field and space compatible cold atom quantum sensors. Using this approach, an optical setup to generate a crossed beam optical dipole trap is assembled in a compact volume of ~25 mL, compare Section 1. As a next step we plan to combine such optical systems with atom chips within the UHV system of a cold atom quantum sensor, enabling hybrid trapping geometries in a compact and stable setup with low SWaP budget. Since our micro-optical systems rely on adhesive integration methods, these are qualified by die shear testing according to MIL-STD-883L Test Method 2019.9 in application relevant geometries, in our case for fused silica test bodies with a footprint of 2 × 4 mm2. Samples are set-up for four adhesives with different curing mechanisms and environmentally aged by thermal cycling (up to –55°C to 150°C) and/or gamma-radiation (10 000 mSv). Subsequent die shear testing shows for almost all sample sets a fulfillment of the MIL-STD-883L Test Method 2019.9 and a mostly increased bond strength for the environmentally aged samples. This can be explained by a better crossed-linked polymer network and released stress of the adhesive layer after application of energy to the adhesive joint, see Section 2.2. Microscope images of the samples after testing are analyzed to evaluate failure mode and whether the joint has a fillet of adhesive. The weakest joints show a failure mode directly at the interface between adhesive layer and mostly the glass test body and a fillet increases the shear strength. For samples without fillets and exhibiting an interface failure mode the effect of two different plasma cleaning methods is investigated and shows an improvement by a factor of about four for the shear strength, compare Section 2.3. The shear strength tests showed overall a large variation, caused by the stronger influence of process parameters for small adhesive joints, e.g., interface properties, accuracy of alignment of test block and adhesive dispensing of small volumes. An increased sample size is necessary to reduce the variation. In future, tensile strength tests of the adhesive joints are foreseen, which are expected to provide valuable information regarding the modulus of the joint. This, however, has proven difficult to accurately resolve for the thin layers of ≤30 μm due to systematic errors of currently available testing setups and geometries. A new tensile strength test setup with dedicated measurement capabilities is planned. The investigated adhesives were selected, besides suitable mechanical, thermal and application properties, because they are promising regarding future application for UHV-compatible micro-optics assemblies. As next step we plan to qualify the adhesives, optical components, and the assembly procedure for UHV applications with an outgassing and mass spectrometry test bed, compare Ref. 11. For this, the accuracy of component mixing for two-component epoxies, the crosslinking of the polymer network and degree of cure is expected to be a significant variable. To further qualify these influences, parallel studies with calorimetry methods are foreseen, providing information about the glass transition temperature Tg of adhesives. ACKNOWLEDGMENTSWe thank Robert Smol, Max Schiemangk and Andreas Wicht from the Joint Lab Quantum Photonic Components at FBH for their support and the provision of the micro-integration setup. We also thank Johannes Zender and Ulrike Winterwerber from the Prototype Engineering Lab at FBH for the provision of the μPQ plasma source for this work. This work is supported by the German Space Agency (DLR) with funds provided by the Federal Ministry for Economic Affairs and Climate Action (BMWK) due to an enactment of the German Bundestag under grant number 50WM1949 and 50RK1978. REFERENCESOi, D. K. L., Ling, A., Grieve, J. A., Jennewein, T., Dinkelaker, A. N., and Krutzik, M.,
“Nanosatellites for quantum science and technology,”
Contemporary Physics, 58
(1), 25
–52
(2017). https://doi.org/10.1080/00107514.2016.1235150 Google Scholar
Grimm, R., Weidemüller, M., and Ovchinnikov, Y. B.,
“Optical Dipole Traps for Neutral Atoms,”
Advances In Atomic, Molecular, and Optical Physics, 42 95
–170 Academic Press(2000). https://doi.org/10.1016/S1049-250X(08)60186-X Google Scholar
Wicht, A., Bawamia, A., Krüger, M., Kürbis, C., Schiemangk, M., Smol, R., Peters, A., and Tränkle, G.,
“Narrow linewidth diode laser modules for quantum optical sensor applications in the field and in space,”
10085
(100850F),
(2017). Google Scholar
Döringshoff, K., Gutsch, F. B., Schkolnik, V., Kürbis, C., Oswald, M., Prübster, B., Kovalchuk, E. V., Bawamia, A., Smol, R., Schuldt, T., Lezius, M., Holzwarth, R., Wicht, A., Braxmaier, C., Krutzik, M., and Peters, A.,
“Iodine Frequency Reference on a Sounding Rocket,”
Physical Review Applied, 11
(5), 1
(2019). https://doi.org/10.1103/PhysRevApplied.11.054068 Google Scholar
Lezius, M., Wilken, T., Deutsch, C., Giunta, M., Mandel, O., Thaller, A., Schkolnik, V., Schiemangk, M., Dinkelaker, A., Kohfeldt, A., Wicht, A., Krutzik, M., Peters, A., Hellmig, O., Duncker, H., Sengstock, K., Windpassinger, P., Lampmann, K., Hülsing, T., Hänsch, T. W., and Holzwarth, R.,
“Space-borne frequency comb metrology,”
Optica, 3
(12), 1381
(2016). https://doi.org/10.1364/OPTICA.3.001381 Google Scholar
Lachmann, M. D., Ahlers, H., Becker, D., Dinkelaker, A. N., Grosse, J., Hellmig, O., Müntinga, H., Schkolnik, V., Seidel, S. T., Wendrich, T., Wenzlawski, A., Carrick, B., Gaaloul, N., Lüdtke, D., Braxmaier, C., Ertmer, W., Krutzik, M., Lämmerzahl, C., Peters, A., Schleich, W. P., Sengstock, K., Wicht, A., Windpassinger, P., and Rasel, E. M.,
“Ultracold atom interferometry in space,”
Nature Communications, 12
(1), 1
–6
(2021). https://doi.org/10.1038/s41467-021-21628-z Google Scholar
Becker, D., Lachmann, M. D., Seidel, S. T., Ahlers, H., Dinkelaker, A. N., Grosse, J., Hellmig, O., Müntinga, H., Schkolnik, V., Wendrich, T., Wenzlawski, A., Weps, B., Corgier, R., Franz, T., Gaaloul, N., Herr, W., Lüdtke, D., Popp, M., Amri, S., Duncker, H., Erbe, M., Kohfeldt, A., Kubelka-Lange, A., Braxmaier, C., Charron, E., Ertmer, W., Krutzik, M., Lämmerzahl, C., Peters, A., Schleich, W. P., Sengstock, K., Walser, R., Wicht, A., Windpassinger, P., and Rasel, E. M.,
“Space-borne Bose–Einstein condensation for precision interferometry,”
Nature, 562
(7727), 391
–395
(2018). https://doi.org/10.1038/s41586-018-0605-1 Google Scholar
Frye, K., Abend, S., Bartosch, W., Bawamia, A., Becker, D., Blume, H., Braxmaier, C., Chiow, S.-W., Efremov, M. A., Ertmer, W., Fierlinger, P., Franz, T., Gaaloul, N., Grosse, J., Grzeschik, C., Hellmig, O., Henderson, V. A., Herr, W., Israelsson, U., Kohel, J., Krutzik, M., Kürbis, C., Lämmerzahl, C., List, M., Lüdtke, D., Lundblad, N., Marburger, J. P., Meister, M., Mihm, M., Müller, H., Müntinga, H., Nepal, A. M., Oberschulte, T., Papakonstantinou, A., Perovsek, J., Peters, A., Prat, A., Rasel, E. M., Roura, A., Sbroscia, M., Schleich, W. P., Schubert, C., Seidel, S. T., Sommer, J., Spindeldreier, C., Stamper-Kurn, D., Stuhl, B. K., Warner, M., Wendrich, T., Wenzlawski, A., Wicht, A., Windpassinger, P., Yu, N., and Wörner, L.,
“The Bose-Einstein Condensate and Cold Atom Laboratory,”
EPJ Quantum Technology, 8
(1), 1
(2021). https://doi.org/10.1140/epjqt/s40507-020-00090-8 Google Scholar
Keil, M., Amit, O., Zhou, S., Groswasser, D., Japha, Y., and Folman, R.,
“Fifteen years of cold matter on the atom chip: promise, realizations, and prospects,”
Journal of Modern Optics, 63
(18), 1840
–1885
(2016). https://doi.org/10.1080/09500340.2016.1178820 Google Scholar
Kassner, A., Rechel, M., Heine, H., Herr, W., Christ, M., Krutzik, M., Rasel, E. M., and Wurz, M. C.,
“Atom Chip technology for use under UHV conditions,”
in Smart Systems Integration; 13th International Conference and Exhibition on Integration Issues of Miniaturized Systems, Barcelona, Spain,
69
–75
(2019). Google Scholar
Christ, M., Kassner, A., Smol, R., Bawamia, A., Heine, H., Herr, W., Peters, A., Wurz, M. C., Rasel, E. M., Wicht, A., and Krutzik, M.,
“Integrated atomic quantum technologies in demanding environments: development and qualification of miniaturized optical setups and integration technologies for UHV and space operation,”
CEAS Space Journal, 11
(4), 561
–566
(2019). https://doi.org/10.1007/s12567-019-00252-0 Google Scholar
Habenicht, G.,
“[Kleben],”
Springer Berlin Heidelberg, Berlin, Heidelberg
(2009). https://doi.org/10.1007/978-3-540-85266-7 Google Scholar
Adams, R. D. and Peppiatt, N. A.,
“Stress analysis of adhesive-bonded lap joints,”
Journal of Strain Analysis, 9
(3), 185
–196
(1974). https://doi.org/10.1243/03093247V093185 Google Scholar
da Silva, L. F. M.,
“Design Rules and Methods to Improve Joint Strength,”
[Handbook of Adhesion Technology, 689
–723 Springer Berlin Heidelberg, Berlin, Heidelberg
(2011). https://doi.org/10.1007/978-3-642-01169-6 Google Scholar
Habenicht, G.,
“[Kleben - erfolgreich und fehlerfrei],”
Springer Fachmedien Wiesbaden, Wiesbaden
(2016). https://doi.org/10.1007/978-3-658-14696-2 Google Scholar
Al-Attabi, A., Fouracre, R. A., Banford, H. M., and Tedford, D. J.,
“High dose neutron and gammaradiation effects in an epoxy resin,”
International Journal of Radiation Applications and Instrumentation. Part, 40
(5), 411
–419
(1992). Google Scholar
Wolf, D.,
“Resonator und Leistungsoszillator zum Aufbau einer integrierten Plasmaquelle sowie deren Verwendung,”
Patent, DE 10 2020 100 872 B4 2021.08.05,
(2021). Google Scholar
|
