mi17050592.pdf - 第32页

Micromachines 2026 , 17 , 592 32 of 34 array alignments and fully integrating the r esulting automated ﬂows into advanced test cells will be essential to meet the escalating manufacturing demands of next-generation silic…

Micromachines 2026, 17, 592

31 of 34

data imputation techniques [

] must be developed to maintain alignment robustness

against real-time probe drift and intermittent signal degradation.

Transitioning these methodologies to high-volume manufacturing demands broader

experimental validation across diverse foundries, device architectures, and multi-ﬁber

arrays. In such high-dimensional optimization landscapes, the algorithmic hierarchy may

shift favorably toward Bayesian surrogate models [

]. Crucially, these sub-second

alignment strategies must be seamlessly integrated into double-sided test cells to meet

the aggressive throughput demands of CPO production ﬂows [

]. Finally, coupling

operational alignment metrics with digital twin frameworks and transformer-based sensor

placement [

] will enable robust structural health monitoring, facilitating predictive

maintenance and extending probe card lifespans in demanding industrial environments.

6. Conclusions

This paper presented a comprehensive evaluation of alignment optimization tech-

niques for wafer-level photonic integrated circuit testing, utilizing the Technoprobe Eclipse

Dynamic probe card system. Through robust experimental validation and extensive simu-

lations spanning diverse operating regimes, incorporating varying actuation speeds and

hysteresis conditions, this study systematically benchmarked seven alignment algorithms.

The results unequivocally establish the Fix-Grad algorithm as the optimal solution for

high-volume manufacturing. Across all evaluated conditions, this approach demonstrated

exceptional robustness, consistently delivering a 95.8% reliability rate and 99.4% align-

ment accuracy.

Transitioning to high-speed actuator drive electronics is essential for production via-

bility, providing a 14.6–21.9

throughput improvement over low-speed operation (Table 3).

In the ideal high-speed, hysteresis-compensated regime, the Fix-Grad algorithm achieved a

rapid alignment time of 1.44 a.u. per die. While uncompensated hysteresis approximately

doubles this execution time, algorithmic mitigation strategies successfully preserve ﬁnal

measurement quality without requiring complex closed-loop positioning feedback. Alter-

native methods presented speciﬁc trade-offs—the Var-Grad technique offered the fastest

alignment (1.18 a.u.) but suffered a higher failure rate of 9.7%, rendering it suitable only for

non-critical screening. Furthermore, despite high sample efﬁciency, BO approaches exhib-

ited excessive computational overhead and heightened sensitivity to hysteresis, limiting

their viability for strict production environments.

From a production perspective, data proﬁle analysis conﬁrms that the high reliability

of the Fix-Grad method limits manual operator intervention to just 4.2% of tested dies. For a

standard 300 mm wafer containing 200 dies, this integrated architecture facilitates complete

optical characterization in approximately 288 to 664 a.u., depending on the extent of active

hysteresis compensation. By achieving highly reliable optical alignments entirely through

open-loop control, this system matches or exceeds the performance metrics reported in

recent literature [

]. Crucially, embedding this capability directly into the probe card

eliminates the need for bulky external positioning stages, signiﬁcantly reducing overall

equipment footprint and enabling the parallel multi-channel testing required for emerging

CPO architectures [1,3].

Future research will focus on advancing data-driven control strategies to further

enhance system throughput and absolute accuracy. Key initiatives include implementing

Long Short-Term Memory networks to replace analytical models for history-dependent

hysteresis compensation, alongside deploying Reinforcement Learning agents to adaptively

counteract complex mechanical irregularities. Additionally, leveraging historical wafer

data to generate predictive spatial heatmaps will heavily optimize search initialization

parameters. Expanding these methodologies to accommodate simultaneous multi-ﬁber

https://doi.org/10.3390/mi17050592

Micromachines 2026, 17, 592

32 of 34

array alignments and fully integrating the resulting automated ﬂows into advanced test

cells will be essential to meet the escalating manufacturing demands of next-generation

silicon photonics.

Author Contributions: Conceptualization, R.V., M.M. and S.M.; methodology, M.B., A.G. and R.V.;

software (algorithm writing and simulation), M.B.; investigation (research and experiment data

recording), M.B. and A.G.; resources (hardware design and build), A.G. and R.V.; writing—original

draft preparation, M.B.; writing—review and editing, M.B., A.G., M.M. and S.M.; supervision, M.M.

and S.M.; project administration, A.G. and R.V. All authors have read and agreed to the published

version of the manuscript.

Funding: Funded by the European Union (Grant Agreement No. 101072491). Views and opinions

expressed are however those of the author(s) only and do not necessarily reﬂect those of the European

Union or the European Research Executive Agency. Neither the European Union nor the granting

authority can be held responsible for them.

Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.

Data Availability Statement: The data presented in this study are available on request from the

corresponding author.

Conﬂicts of Interest: Authors Mehdi Bejani, Alessia Galli, Riccardo Vettori, and Marco Mauri were

employed by the company Technoprobe S.p.A. The authors declare that this research was conducted

as part of a collaboration between Politecnico di Milano and Technoprobe S.p.A. within the EU-

funded project (Grant Agreement No. 101072491). The remaining authors declare that the research

was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as

a potential conﬂict of interest.

References

Thomson, D.; Zilkie, A.; Bowers, J.E.; Komljenovic, T.; Reed, G.T.; Vivien, L.; Marris-Morini, D.; Cassan, E.; Virot, L.; Fédéli, J.-M.;

et al. Roadmap on silicon photonics. J. Opt. 2016, 18, 073003. [CrossRef]

Maeda, Y.; Miura, T.; Matsuo, S.; Fukuda, H. Accurate Fiber Alignment using Silicon Photodiode on Grating Coupler for

Wafer-Level Testing. In Proceedings of the 24th OptoElectronics and Communications Conference (OECC) and 2019 International

Conference on Photonics in Switching and Computing (PSC), Fukuoka, Japan, 7–11 July 2019; pp. 1–3. [CrossRef]

3. Couny, F. Seamlessly move PIC testing from lab to fab. PIC Mag. 2022, 2, 34–36.

De Coster, J.; De Heyn, P.; Pantouvaki, M.; Snyder, B.; Chen, H.; Marinissen, E.J.; Absil, P.; Van Campenhout, J.; Bolt, B. Test-station

for ﬂexible semi-automatic wafer-level silicon photonics testing. In Proceedings of the 21th IEEE European Test Symposium

(ETS), Amsterdam, The Netherlands, 23–27 May 2016; pp. 1–6. [CrossRef]

Petrini, M.; Baldi, R.; Seyfried, M.; Morichetti, F.; Melloni, A. Probe-card based automatic testing of photonic integrated ﬁlters. In

Proceedings: Optica Advanced Photonics Congress 2022; Technical Digest Series (Optica Publishing Group, 2022); Paper IM4C.2;

Optica Publishing Group: Washington, DC, USA, 2022. [CrossRef]

Gnausch, T.; Grundmann, A.; Juhasz, T.; Kaden, T.; Büttner, R.; von Freyhold, T. Novel Opto-Electronical Probe Card for

Wafer-Level PIC Testing. In Proceedings of the 2019 Optical Fiber Communications Conference and Exhibition (OFC), San Diego,

CA, USA, 3–7 March 2019; pp. 1–3.

Vettori, R.; Galli, A. A comprehensive study on improving probe card transmission lines for effective high-frequency wafer-level

testing. In Proceedings of the 2024 IEEE European Test Symposium (ETS), The Hague, The Netherlands, 20–24 May 2024.

Vettori, R.; Galli, A. Eclipse Dynamic Probe Card: A Novel Approach for Wafer-Level Photonic Testing with Automated Fiber

Array Unit Alignment. In Proceedings of the 2025 IEEE International Test Conference (ITC), San Diego, CA, USA, 20–26

September 2025; pp. 446–449. [CrossRef]

Bejani, M.; Appello, D.; Mauri, M.; Todaro, S.; Mariani, S. AI-Assisted Framework for Real-Time Monitoring and Management of

Probe Cards in Electrical Wafer Sort Applications. In Proceedings of the 2025 IEEE European Test Symposium (ETS), Tallinn,

Estonia, 26–30 May 2025; pp. 1–4. [CrossRef]

10.

Yuan, Q.; Peczek, A.; Frankel, J.; Rishavy, D.; Mai, C.; Christenson, E.; Pratap, D.; Zimmermann, L. Fully Automated Wafer-Level

Edge Coupling Measurement System for Silicon Photonics Integrated Circuits. IEEE Trans. Semicond. Manuf. 2025, 38, 168–177.

[CrossRef]

https://doi.org/10.3390/mi17050592

Micromachines 2026, 17, 592

33 of 34

11.

Jansen, R.; Pustakhod, D.; Combee, B.; Leijtens, X.; Williams, K.; Latkowski, S. Optical Alignment Optimization for High

Throughput Testing of Photonic Integrated Circuits. In Proceedings of the 2024 24th International Conference on Transparent

Optical Networks (ICTON), Bari, Italy, 14–18 July 2024; pp. 1–4. [CrossRef]

12.

Wu, C.Y.; Lee, M.C.M.; Lu, M.J.; Mu, S.Y.; Chen, J.; Cheng, C.S. Butt-Coupling Optical Probe Card for Wafer-Scale Photonic-

Integrated-Circuits Test with Polarization Control. In Proceedings of the 2023 Conference on Lasers and Electro-Optics Europe &

European Quantum Electronics Conference (CLEO/Europe-EQEC), Munich, Germany, 26–30 June 2023. [CrossRef]

13.

Fleming, A.J. A review of nanometer resolution position sensors: Operation and performance. Sens. Actuators A Phys. 2013,

190, 106–126. [CrossRef]

14.

Kuhnen, K. Modeling, Identiﬁcation and Compensation of Complex Hysteretic Nonlinearities: A Modiﬁed Prandtl-Ishlinskii

Approach. Eur. J. Control 2003, 9, 407–418. [CrossRef]

15.

Al Janaideh, M.; Su, C.Y.; Rakheja, S. Development of the rate-dependent Prandtl-Ishlinskii model for smart actuators. Smart

Mater. Struct. 2008, 17, 035026. [CrossRef]

16.

Zhou, C.; Yuan, M.; Feng, C.; Ang, W.T. A Modiﬁed Prandtl–Ishlinskii Hysteresis Model for Modeling and Compensating

Asymmetric Hysteresis of Piezo-Actuated Flexure-Based Systems. Sensors 2022, 22, 8763. [CrossRef]

17.

Chandra, A.; Daniels, B.; Curti, M.; Tiels, K.; Lomonova, E.A.; Tartakovsky, D.M. Discovery of sparse hysteresis models for

piezoelectric materials. Appl. Phys. Lett. 2023, 122, 214101. [CrossRef]

18.

Beiranvand, V.; Hare, W.; Lucet, Y. Best practices for comparing optimization algorithms. Optim. Eng. 2017, 18, 815–848.

[CrossRef]

19.

Shi, Z.; Huang, H.; Penty, R.; Cheng, Q. Photonic Hybrid Integration: Strategies and Promises of Advanced Additive Manufactur-

ing. Adv. Opt. Mater. 2025, 14, e01750. [CrossRef]

20.

Mandelli, L.; Dankwart, C.; Napoli, C. Motion Stage Precision Prediction for Photonic Integrated Circuit Assembly. J. Intell.

Manuf. 2026, 37, 171–184. [CrossRef]

21.

Trappen, M.; Blaicher, M.; Dietrich, P.I.; Dankwart, C.; Xu, Y.; Hoose, T.; Billah, M.R.; Abbasi, A.; Baets, R.; Troppenz, U.; et al.

3D-printed optical probes for wafer-level testing of photonic integrated circuits. Opt. Express 2020, 28, 37996–38007. [CrossRef]

22.

Polster, R.; Dai, L.Y.; Jimenez, O.A.; Cheng, Q.; Lipson, M.; Bergman, K. Wafer-scale high-density edge coupling for high

throughput testing of silicon photonics. In Proceedings of the 2018 Optical Fiber Communications Conference and Exposition

(OFC), San Diego, CA, USA, 11–15 March 2018; pp. 1–3.

23.

Vermeulen, D.; Selvaraja, S.; Verheyen, P.; Lepage, G.; Bogaerts, W.; Absil, P.; Thourhout, D.V.; Roelkens, G. High-efﬁciency

ﬁber-to-chip grating couplers realized using an advanced CMOS-compatible Silicon-On-Insulator platform. Opt. Express 2010,

18, 18278–18283. [CrossRef]

24.

Liao, P.; Sakib, M.; Park, J.; Wlodawski, M.; Kopp, V.I.; Neugroschl, D.; Liboiron-Ladouceur, O. Ultra-dense optical I/O interface

for Silicon photonic interconnects. In Proceedings of the 2015 IEEE Optical Interconnects Conference (OI), San Diego, CA, USA,

20–22 April 2015; pp. 116–117. [CrossRef]

25.

Barcaya, M.C.B.; Tosi, M.; Ortiz, L.L.; Guichal, G.; Rossini, L.A.B.; Caso, P.A.C. A Cost-Effective Solution for Characterizing

Photonic Integrated Circuits. In Proceedings of the XX Workshop on Information Processing and Control (RPIC), Oberá,

Argentina, 1–3 November 2023; pp. 1–6. [CrossRef]

26.

He, J.; Dong, T.; Xu, Y. Review of Photonic Integrated Optical Phased Arrays for Space Optical Communication. IEEE Access 2020,

8, 188284–188298. [CrossRef]

27.

Schneider, P.I.; Sekulic, I.; Plock, M.; Hammerschmidt, M.; Rodt, S.; Reitzenstein, S.; Burger, S. Physics-informed Bayesian

optimization of nanophotonic devices. In Proceedings: Advanced Photonics Congress (IPR, Networks, NOMA, SOLITH, SPPCom);

Optica Publishing Group: Washington, DC, USA, 2025; p. ITu1A.2. [CrossRef]

28.

Kandel, S.; Rebufﬁ, L.; Cha, W.; Harder, R.J.; Shi, X.; Highland, M.J.; Frith, M.G.; Cherukara, M.J. Bayesian optimization for

autoalignment of an X-ray focusing system. In Proceedings: Optica Imaging Congress (3D, COSI, DH, FLatOptics, IS, pcAOP); Optica

Publishing Group: Washington, DC, USA, 2023; p. JW2A.32. [CrossRef]

29.

Roussel, R.; Edelen, A.L.; Boltz, T.; Kennedy, D.; Zhang, Z.; Ji, F.; Huang, X.; Ratner, D.; Garcia, A.S.; Xu, C.; et al. Bayesian

optimization algorithms for accelerator physics. Phys. Rev. Accel. Beams 2024, 27, 084801. [CrossRef]

30.

Lajtner, K.; Koenig, C.; Rupenyan, A.; Resan, B. Machine Learning for automatic pointing alignment and spatial beam ﬁltering.

EPJ Web Conf. 2023, 287, 13014. [CrossRef]

31.

Wu, K.; Kim, K.; Garnett, R.; Gardner, J.R. The Behavior and Convergence of Local Bayesian Optimization. arXiv 2024,

arXiv:2305.15572. [CrossRef]

32.

Fan, Z.; Wang, W.; Ng, S.H.; Hu, Q. Minimizing UCB: A Better Local Search Strategy in Local Bayesian Optimization. arXiv 2024,

arXiv:2405.15285. [CrossRef]

33.

Yu, Z.; Wu, Y.; Fang, Z.; Sun, H. Modeling and compensation of hysteresis in piezoelectric actuators. Heliyon 2020, 6, e03999.

[CrossRef]

https://doi.org/10.3390/mi17050592