Two Photon Laser Techniques for 3D Miniaturized Structures: Resistive-Pulse Sensing and Potential Bio- Uses

Two Photon Laser Techniques for 3D Miniaturized Structures: Resistive-pulse Sensing and Potential Bio- uses

Caizhi LIAO Bachelor of Engineering (B. Eng) Master of Philosophy (M. Phil)

A thesis submitted for the degree of Doctor of Philosophy at The University of Queensland in 2020 Australian Institute for Bioengineering and Nanotechnology

Abstract Reliable analysis of individual nanoparticles is becoming increasingly critical to a broad range of application scenarios. Such measurements provide a gateway to understand the physical, chemical, and biological processes in complex systems. For various nanoparticle analysis purposes, one need to not only detect the size of nanoparticles, but track the dynamics of individual analyte in selected space. Fluorescence microscopy techniques and dynamic light scattering (DLS) serve as the most-commonly explored methods in nanoparticle studies. However, these measurements have been limited by the observation time and the measurement rate.

Rapid advances in label-free strategies are transforming the way sensing systems are studied, especially with a view on developing novel devices for single nanoparticle analysis. Due to their inherently simple operating principle, which is based on recording the changes in the ionic current through a miniaturized pore that is separated by two electrolyte-filled reservoirs, resistive-pulse sensing (RPS) pore sensors have been gaining prominence for a wide range of analysis: from performing the sequence of DNA/RNA to unravelling the underlying mechanisms of bio-systems. In particular, considerable effects have been devoted to the RPS investigation of nanoparticles, providing profound implications for uses in analysing the size, structure and surface feature of nanoparticles.

For a typical RPS system, micro-/nano-pore plays a central role in nanoparticle analysis, through which characteristic resistive pulses are generated due to their different ion transport properties. To fully understand the information contained within the signal, it is critical to prepare accurate pores with full control over pore’s three-dimensional (3D) geometrical features. However, the current state-of-the-art of RPS pore fabrication technologies are still not able to achieve reliable control over the 3D internal geometry of the produced pores. Hence, novel techniques enabling the generation of geometrically well- defined micro-/nano-pores are currently of immense interest.

Recently, two-photon polymerization (TPP) based techniques have become a powerful tool for rapid and accurate 3D prototyping of microscopic structures. In TPP fabrication, a femtosecond pulsed laser beam is tightly focused on a photosensitive resist material consisting of a mixture of monomers and photo-initiators. This laser-activated reaction leads to localized crosslinking of photosensitive resist material, and by moving the focal point in through the liquid material, arbitrary 3D structures can be formed. By taking advantage of

I precision optics, tailored 3D structures with up to 100 nm resolution can be realized, suitable for the fabrication of micro-/nano-pores with configurable internal geometries.

This PhD project aims to develop robust TPP-based fabrication techniques for accurate RPS pores for the first time. Firstly, a brand-new 3D ablation process based on two-photon femtosecond laser is introduced to generate accurate micro-hole structures in plastic substrates (Chapter 3). This strategy can create arbitrary shaped micro-well structures in plastics through a rapid ablation process without any masks, providing suitable substrate that is required to load the sensing pore construct. Next, this pre-treated plastic substrate is utilized for loading the miniaturized hollow 3D constructs, which is to be realized through the TPP fabrication system (Chapter 4). Systematic investigations are performed to achieve optimized fabrication conditions, including laser power level, structural design, and post- processing. As a proof-of-concept, classical 3D hollow micro-devices, including micro-pore, micro-needle, micro-pump and micro-electrode, are demonstrated and characterized.

As TPP platform has been experimentally evaluated for the preparation of tailored 3D micro- constructs, accurate miniaturized RPS pores are then formulated for robust nanoparticle analysis (Chapter 5). TPP based nanolithography is introduced for reliable preparation of customizable RPS pores. For the first time, accurate micro- and nano-pores with different cone angles have been successfully prepared for experimental studies. Subsequently, accurate 3D pores were studied for selected RPS analysis: cis- and trans-conical pores for the investigation of pore’s preferential transport capability; symmetrical pores for the electrical tracking of nanoparticle position; and cylindrical pores for the surface charge analysis of chemically distinct nanoparticles of the same size.

Lastly, a promising new strategy based on vertically arranged double-pore construct was introduced to analyse and control single molecular transportation (Chapter 6). TPP nanolithography affords to generate accurately controlled dual-pore systems that have enhanced sensing and controlling capabilities in nanoparticle analysis. By modifying the geometric features of dual-pore systems, the translocating nanoparticles can be analysed both collectively and individually. Moreover, this vertically stacked dual-pore system can be deployed as a modulator to control each nanoparticle event, and even trap/release single nanoparticle within a confined space.

In summary, the proposed two-photon laser based fabrication platform serves as robust toolkit for the preparation of accurate RPS pore constructs. The knowledge and methodologies developed in this Thesis can serve as a solid foundation for the further developments of novel/enhanced types of RPS analysis platforms, thus will contribute the development of whole single-molecular analysis systems and help to inspire the next stages in this exciting field of research.

III

Declaration by author

This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text. I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, financial support and any other original research work used or reported in my thesis. The content of my thesis is the result of work I have carried out since the commencement of my higher degree by research candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution. I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the policy and procedures of The University of Queensland, the thesis be made available for research and study in accordance with the Copyright Act 1968 unless a period of embargo has been approved by the Dean of the Graduate School.

I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis and have sought permission from co- authors for any jointly authored works included in the thesis.

Publications included in this thesis

Caizhi Liao, Fiach Antaw, Alain Wuethrich, Matt Trau. Stacked Dual-Pore Architecture for Deciphering and Manipulating The Dynamics of Individual Nanoparticles. Advanced Materials Technology, 2020, 2000701.

Caizhi Liao, Fiach Antaw, Alain Wuethrich, Will Anderson, Matt Trau. Configurable Miniaturized Three-Dimensional Pores for Robust Single Nanoparticle Analysis. Small Structures, 2020, DOI: 10.1002/sstr.202000011.

Caizhi Liao, Alain Wuethrich, Matt Trau. A Material Odyssey For 3D Nano/Micro- Structures: Two Photon Polymerization Based Nanolithography In Bioapplications. Applied Materials Today, 2020, 4, 1, 1401-1409.

 Caizhi Liao, Fiach Antaw, Will Anderson, Matt Trau. Two-Photon Nanolithography Of Tailored Hollow Three-Dimensional Microdevices For Biosystems. ACS Omega, 2019, 4,1, 1401-1409.

Caizhi Liao, Will Anderson, Fiach Antaw, Matt Trau. Maskless 3D Ablation Of Precise Microhole Structures In Plastics Using Femtosecond Laser Pulses. ACS Applied Materials & Interfaces, 2018, 10, 4, 4315-4323. ----- *Featured by dozens public media, including Material Views, Sci.Org, etc.,

Submitted manuscripts included in this thesis

No manuscripts submitted for publication.

Other publications during candidature

Conference abstracts (Oral)

Caizhi Liao, Fiach Antaw, Will Anderson, Matt Trau. Two-Photon Nanolithography for Biosystems. ECBA5: The 5th European Congress of Applied Biotechnology Sept.14 to Sept.21, 2019 Florence, Italy

Caizhi Liao, Fiach Antaw, Will Anderson, Matt Trau. Additive Manufacturing in Bio- system structures. EAIT Postgraduate Conference Jun.6 to Jun 7, 2018 Brisbane, Australia

Caizhi Liao, Fiach Antaw, Will Anderson, Matt Trau. Maskless 3D Ablation of Precise Microhole Structures in Plastics Using Femtosecond Laser Pulses. Australian-Asian Regional Material Conference Jul.21 to Jul.25, 2018 Cairns, Australia

Caizhi Liao, Will Anderson, Fiach Antaw, Matt Trau. Laser Ablation in Manufacturing of Nano/micro-structures. AIBN International Postgraduate Student Conference Nov.15 to Nov.14, 2017 Brisbane, Australia

 Caizhi Liao, Will Anderson, Fiach Antaw, Matt Trau. Simultaneous subtractive and additive fabrication of three-dimensional structures for bio-nano resistive pulse analysis. 8th International Postgraduate Symposium in Biomedical Sciences 31 Oct to 1 Nov Brisbane, Australia

Contributions by others to the thesis

The experimental work being described in this thesis has led to a succession of published peer-reviewed papers, in which I served as the first leading author. As clarified in the previous section “Publications included in this thesis”, my fellow contributing-authors on these publications have contributed in part to the experimental design, simulation protocol, data collection and analysis, and manuscript preparation processes. The details of contributing authors can also found below:

Mr. Fiach Antaw: Contributed to the conception, design, characterisation and interpretation of data, and critically revising the thesis.

Dr. Will Anderson: Contributed to the conception, design, characterisation and interpretation of data, and critically revising the thesis.

Dr. Alain Wuethrich: Contributed to the conception, and interpretation of data, and critically revising the thesis.

Prof. Matt Trau: Contributed to the conception, design, and interpretation of data, and revising the thesis.

Statement of parts of the thesis submitted to qualify for the award of another degree

No works submitted towards another degree have been included in this thesis.

Research Involving Human or Animal Subjects

No animal or human subjects were involved in this research.

VII

Acknowledgements

First and foremost, I would like to express my sincere gratitude to the fantastic “Nano Fury” members of Prof. Matt Trau group, Australian Institute for Bioengineering and Nanotechnology (AIBN), The University of Queensland, for their continued support, contributed advice and inspired knowledge throughout my PhD pursuing route.

In particular, I wish to express my heartfelt gratitude and appreciation to my supervisor Prof. Matt Trau. Being so supportive and enthusiastic in nurturing us as qualified researchers, Matt provides an open and unrestricted research environment for me to accomplish PhD projects. For me, Prof Matt is like the Northern Star guiding me through the thrilling PhD adventure. Also, I would like to express my gratitude to Dr Will Anderson for supervising me over 2 years. Will guided me entering a brand-new field by providing me in-time helps and suggestions. Furthermore, it is a great privilege to thank Dr Alain Wuethrich, for his endless support and constructive suggestions. As my associate supervisor, Dr Alain patiently helped to polish and publish the papers.

Also, I like to acknowledge many other colleagues in Trau group: Dr Yuling Wang- Thank you for providing the opportunity to initiate the research in an exciting new field; Mr. Fiach Antaw- Thank you for supporting the project with your deep expertise, and generous helps and inspirational comments; Dr Wang Jing- Thank you for the insightful discussions in a wide spectrum of topics; Ms Junrong Li- Thank you for the useful suggestions and comments; Dr Kevin Koo- Thank you for organizing the memorial gatherings .

In addition, I would like to acknowledge my gratitude for all of the technical and academic staffs in AIBN and UQ. Throughout my PhD project, I have received critical amount of training and support from the Australian National Fabrication Facility Queensland Node- ANFF. In particular, Mr Doug Mair- Thank you for providing the training on Nanoscribe, SEM, soft-lithography; Mr Kai-Yu Liu- Thank you for showing the use of DRIE; Dr Kinnari Shelat- Thank you for helping to perform AFM training and measurements; Dr Elliot Cheng ( Centre for Microscopy and Microanalysis )- Thank you for providing insightful information on fabrication process.

VIII

Last but not the least, I would like to express my sincere thanks to my family, for their unconditional and unreserved love and support throughout my PhD adventure.

Financial support This research work is supported by the Australian Government Research Training Program Scholarship, which previously was granted as the International Postgraduate Research Scholarship (IPRS).

Keywords Two Photon Polymerization, Laser Ablation, Additive Manufacturing, Bio-micro-device, Resistive-pulse-sensing (RPS), Resist Material, Dual-pore structure, Single Nanoparticle Analysis, Surface Modification, Surface Charge.

Australian and New Zealand Standard Research Classifications (ANZSRC)

ANZSRC code: 030107 Sensor Technology (Chemical aspects), 30%

ANZSRC code: 100402, Medical Biotechnology Diagnostics (incl. Biosensors), 30%

ANZSRC code: 100703, Nanobiotechnology, 40%

Fields of Research (FoR) Classification FoR code: 0912, Materials Engineering, 30% FoR code: 1004, Medical Biotechnology, 30% FoR code: 1007, Nanotechnology, 40%

Table of Contents

Abstract ...... I

Declaration by author ...... IV

Publication included in this thesis ...... V

Acknowledgements ...... VIII

List of abbreviations ...... XII

Chapter 1: Thesis Introduction ...... 1

1.1 Background ...... 1 1.2 Research aims and milestones ...... 2 1.3 Significance of project ...... 3 1.4 Structure of the thesis ...... 4 1.5 References ...... 7

Chapter 2: Thesis Literature Review ...... 8 2.1 Introduction ...... 11 2.2 Two-photon polymerization (TPP) ...... 13 2.3 Materials ...... 18 2.4 TPP for bio-uses ...... 25 2.5 Challenges and perspective ...... 38

2.6 Conclusions ...... 40

Chapter 3: Maskless 3D Ablation Of Precise Micro-Hole Structures In Plastics Using Femtosecond Laser Pulses ...... 49 3.1 Introduction ...... 51 3.2 Experimental methods ...... 54 3.3 Results and discussions ...... 56 3.4 Conclusions ...... 67 Supplemental information ...... 72

Chapter 4: Two-Photon Nanolithography of Tailored Hollow Three-dimensional Microdevices for Biosystems ...... 81 4.1 Introduction ...... 83

4.2 Experimental section ...... 85 4.3 Results and discussions ...... 88 4.4 Conclusions ...... 98 Supplemental information ...... 104

Chapter 5: Configurable Miniaturized Three-Dimensional Pores for Robust Single Nanoparticle Analysis ...... 122 5.1 Introduction ...... 124 5.2 Experimental section ...... 126 5.3 Results and discussions ...... 128 5.4 Conclusions ...... 141 Supplemental information ...... 147

Chapter 6: Stacked Dual-Pore Architecture Deciphering the Dynamics of Single Nanoparticle ...... 167 6.1 Introduction ...... 169 6.2 Results and discussions ...... 171 6.3 Conclusions ...... 184 Supplemental information ...... 187

Chapter 7: General Conclusions and Future Work ...... 207 7.1 Conclusions ...... 207 7.2 Future work ...... 209

List of abbreviations used in the thesis

Atomic Force Microscope AFM Aspect Ratio AR Blood-brain Barrier BBB Bovine Serum Albumin BSA Computer-aided Design CAD Current I Direct Laser Writing DLW Dynamic Light Scattering DLS Electrochemical Drilling ECD Electrical-discharge Machining EDM Extracellular Matrix ECM Fourier-transform Infrared Spectroscopy FTIR Finite Element Method FEM Femtosecond fs General Writing Language GWL Hyaluronic Acid HA Heat Affected Zone HAZ Hours, Minutes, Seconds hr, min, s Nanometer, Micrometer nm, µm Laser Scanning Microscopy LSM Multi-photon Ionization MPI Nanoparticles NPs Near Infrared Regions NIR Nanoparticle Tracking Analysis NTA Phosphate Buffer Saline PBS Polylactic Acid PLA Polycaprolactone PCL Polyglycolic Acid PGA Poly(ethylene glycol-co-peptide) Diacrylate PEG-DA Pentaerythritol Triacrylate PETA Polyethylene Terephthalate PET Poly(methyl methacrylate) PMMA Polycarbonate PC

XII

Polyethylene PE Resistive Pulse Sensing RPS Scanning Electron Microscope SEM Signal To Noise Ratio S/N Two Photon Polymerization TPP Two Photon Absorption TPA Three Dimensional 3D Trimethylene Carbonate TMC Volt V

XIII

Chapter 1

Thesis Introduction

1.1 Background

In Chapter 2 “Thesis Literature Review ” section, a detailed research background of this thesis will be presented. The following is a summary of Chapter 2.

Resistive-pulse sensing (RPS) is an experimental technique used to characterise colloidal particles ranging from approximately 50 nm in diameter up to the size of cells. Resistive pulse sensors trace their origins back to the Coulter counter, which was created in late 1940s to count and size biological cells and microorganisms. RPS has enabled the high throughput particle-by-particle sensing and analysis of cells since the 1950s, submicron particles including viruses since the 1970s, and particles as small as single molecules over the past decade or so, with particularly intense interest in possible nanopore-based DNA sequencing analysis recently.

The sensing principle is remarkably simple; two reservoirs separated by a pore structure are filled with conductive solutions, each containing an electrode. When an appropriate voltage is applied between the two electrodes, a formulated ionic current will pass through the aperture structures. A transient ionic current decrease caused by the translocation of analytical elements is known as a “blockade event”, whose magnitude is primarily determined by the size of analytes. Importantly, the geometry of the pore structure ultimately determines the sensitivity of the technique, as a significant occlusion event can only be observed when the size of the analyte passing through the pore is comparable to that of the pore structure.

To date, numerous pore structures have been utilised for RPS investigation, including solid- state pores (e.g., silicon-based membranes) via photolithographic etching, protein pores (e.g. α-hemolysin) via self-assembly and polymer pores (e.g., PET) via track etching. The breadth of the field can be inferred by the utilization of varied pore materials including carbon

Chapter 1 nanotubes, micropipettes, silicon nitride, polymers, and polydimethylsiloxane, as well as the application of diverse fabrication techniques to make pores, such as ion beam sculpting, track etching and laser melting, electron beam, and soft lithography. However, none technique among the aforementioned can precisely control the pore geometry in three dimensions (3D). Since the size and shape of a pore significantly influence both the detectable size range and measurement sensitivity, any fabrication technique able to precisely tailor the 3D geometrical features of pore would be in great demand.

Since its introduction in 1997, two photon polymerization (TPP) based nanolithography has become an enabling technology in many different fields, including photonics, microfluidic systems, tissue engineering and biomedical engineering. As reported, TPP is based on the two-photo absorption (TPA) mechanism: the polymerization process is triggered by a tightly focused laser pulse applied within the focal region of the photosensitive materials. With the aid of such solid freeform fabrication technique, 3D structures can be prepared by simply washing and removing the uncross-linked materials. The advent of femtosecond lasers integrated in TPP nanolithography systems improved the fabrication resolution to sub-100 nm, which is three orders of magnitude smaller than that of structures prepared by current 3D printing techniques. More importantly, the introduction of new construction materials has further driven the development of this field. In particular, when combined with bio-compatible photosensitive materials, TPP could be used to fabricate 3D nano/micro-RPS pores that possess superior stability and excellent compatibility with biological environments, serving as a versatile toolkit poised to drive advances for RPS (bio)-nanoparticle analysis.

1.2 Research Aims and Milestones

Two photon polymerization (TPP) has the potential to accurately prepare robust 3D nano/micro-pore structures for RPS implementations. This PhD thesis explores the TPP process for fabricating 3D nano/micro-pore structures with well defined geometries and assesses their applicability for RPS. To achieve this goal, the main research aims/milestones were as follows.

Milestone 1. Generation of precise micro-holes in plastic substrates. The creation of controllable micro-holes in a supporting plastic plate is the pre-requisite for the TPP fabrication of robust 3D RPS pores. This milestone involves the 2

Chapter 1

implementation of a new laser nano/micro-processing methodology which is amenable for the maskless generation of micro-holes in plastics.

Milestone 2. Development of 3D hollow micro-constructs for bio-environments. 3D hollow mico-constructs play a significant role in bio-applications. By using the TPP fabrication of 3D hollow nano/micro-pore as a model, this research milestone demonstrates the design and preparation of a group of classical micro-structures for bio-uses.

Milestone 3. Preparation of configurable miniaturized three-dimensional pores for robust single nanoparticle analysis. This milestone builds upon the achievements of Milestone 2. Systematic work attempts to generate 3D accurate pores for tuned single nanoparticle analysis.

Milestone 4. Attainment of stacked dual-pore system for deciphering and Controlling individual nanoparticles. With the aim of uncovering the dynamics of translocating nanoparticle, this milestone explores a novel type of dual-pore system for nanoparticle analysis in confined space. Adjusted stacked dual-pore constructs could effectively regulate the dynamic motions of individual nanoparticle.

1.3 Significance of project

To address the fabrication hurdles in RPS pore preparation, we here introduce the nano/micro-3D additive manufacturing technique based on two-photon polymerization (TPP). The attainment of the stated research aim and milestones in this thesis could have a profound impact in advancing the preparation of robust RPS pores. With tuned settings, the TPP nano/micro-fabrication platform would allow us to generate customizable 3D constructs for bio-applications. As a proof-of-concept, TPP prepared 3D pores are used to demonstrate a reliable and robust performance nanoparticle analysis. Furthermore, it is demonstrated that the 3D structuring capability of the TPP platform enables the fabrication of a brand-new 3

Chapter 1 type of dual-pore system, which could unleash massive opportunities for the unmet needs in RPS analysis field. From the perspective of the candidate, the studies in this thesis could communicate the vast potential of TPP fabrication platform for RPS analysis, as well as help to inspire the next stages in this exciting field of research.

1.4 Structure of The Thesis This thesis consists of seven chapters that are presented as a combination of published peer-reviewed papers and submitted manuscripts under revision. The details of the thesis structure are as follows:

Chapter 1: Thesis Introduction

This chapter briefly introduces the background for this thesis work, research aims and milestones, and project significance.

Chapter 2: Thesis Literature Review

This chapter provides detailed background knowledge for this thesis work. In this section, fundamentals of two photon polymerization (TPP) triggered by a laser and its practical uses in bio-related fields will be thoroughly discussed.

Chapter 2 is based on a published review paper:

Caizhi Liao, Alain Wuethrich, Matt Trau. “A Material Odyssey for 3D Nano/ Microstructures: Two Photon Polymerization Based Nanolithography in Bioapplications”. Applied Materials Today, 2020, 19, 100635.

Chapter 3: Maskless 3D Ablation Of Precise Micro-Hole Structures In Plastics Using Femtosecond Laser Pulses

To achieve Milestone 1, this chapter details the micro-hole preparation process in plastic substrates. By using two photon laser based ablation, we can prepare well-defined micro- hole structures in different types of plastics, providing massive opportunities for the current unmet need in micro-fabrication fields. Importantly, these processed plastic substrates with

Chapter 1 tailored micro-hole constructs are the essential components for building the 3D miniaturized pores that have been explored in the work towards Milestone 2 to Milestone 4.

Chapter 3 is based on a published paper:

Caizhi Liao, Will Anderson, Fiach Antaw, Matt Trau. “Maskless 3d ablation of precise micro-hole structures in plastics using femtosecond laser pulses”. ACS Applied Materials & Interface, 2018, 10, 4, 4315-4323

Chapter 4: Two-Photon Nanolithography of Tailored Hollow Three-dimensional Microdevices for Biosystems

Milestone 2 required the design and fabrication of tailored 3D hollow constructs that are suitable for versatile bio-applications. In this chapter, a group of typical miniaturized hollow 3D micro-devices, including micro-needle, micro-pump and micro-robot, are prepared for potential bio-uses. By achieving Milestone 2, we developed the TPP fabrication system that allows us to prepare 3D controlled micro-pore constructs, which are to be discussed in Milestone 3. Chapter 4 is based on a published paper:

Caizhi Liao, Will Anderson, Fiach Antaw, Matt Trau. “Two-Photon Nanolithography of Tailored Hollow three-dimensional Microdevices for Biosystems”. ACS Omega, 2019, 4,1, 1401-1409

Chapter 5: Configurable Miniaturized Three-Dimensional Pores for Robust Single Nanoparticle Analysis

In this chapter further explores the TPP fabrication platform, applying it for the preparation of customizable 3D RPS pores, the cornerstone to fulfil Milestone 3. As a proof-of-concept, high-performance pore structures (e.g., Cis-Conical pore, Trans-Conical pore, Symmetrical pore and Cylindrical pore) are successfully realized for RPS analysis. To further look into this robust fabrication platform, we then propose to build more powerful RPS pores, such as the brand-new type of 3D dual-pore constructs sought by Milestone 4.

Chapter 1

Chapter 5 is based on a published paper:

Caizhi Liao, Fiach Antaw, Alain Wuethrich, Will Anderson, Matt Trau. “Two- Photon Nanolithography of Tailored Hollow three-dimensional Microdevices for Biosystems”. Small Structure, 2020. DOI: 10.1002/sstr.202000011

Chapter 6: Stacked Dual-Pore Architecture Deciphering the Dynamics of Single Nanoparticle

Based on the previous Milestone achievements, complicated yet controlled 3D RPS pores could be realized through the proposed TPP nanolithography system. To further look into this issue, this chapter examines a brand-new type of dual-pore sensing platform for robust RPS analysis by using TPP fabrication scheme. The stacked dual-pore construct enables the study of individual nanoparticle dynamics within a confined space.

Chapter 6 is based on a published paper:

 Caizhi Liao, Fiach Antaw , Alain Wuethrich, Matt Trau. “Stacked Dual-Pore Architecture Deciphering and Manipulating the Dynamics of Individual Single Nanoparticle”. Advanced Materials Technology, 2020, 2000701.

Chapter 7: General Conclusions and Future Work

This chapter concludes the work discussed in this thesis, and provides an outlook based on the current progress introduced through this thesis.

Chapter 1

1.5 References

1. Xiong, W.; Liu, Y.; Jiang, L. J.; Zhou, Y. S.; Li, D. W.; Jiang, L.; Silvain, J.-F.; Lu, Y. F. Adv. Mater. 2016, 28, 2002– 2009. 2. Diddams, S. A.; Hollberg, L.; Mbele, V. Nature 2007, 445, 627– 630. 3. Le Harzic, R.; Huot, N.; Audouard, E.; Jonin, C.; Laporte, P.; Valette, S.; Fraczkiewicz, A.; Fortunier, R. Appl. Phys. Lett. 2002, 80, 3886– 3888. 4. Ni, J.; Wang, Z.; Li, Z.; Lao, Z.; Hu, Y.; Ji, S.; Xu, B.; Zhang, C.; Li, J.; Wu, D.; Chu, J. Adv. Funct. Maters., 2017, 27, 1701939. 5. Lao, Z.-X.; Hu, Y.-L.; Pan, D.; Wang, R.-Y.; Zhang, C.-C.; Ni, J.-C.; Xu, B.; Li, J.- W.; Wu, D.; Chu, J.-R. Small 2017, 13, 1603957. 6. Kozak, D.; Anderson, W.; Vogel, R.; Trau, M. Nano Today 2011, 6, 531– 545. 7. Schizas, C.; Melissinaki, V.; Gaidukeviciute, A.; Reinhardt, C.; Ohrt, C.; Dedoussis, V.; Chichkov, B. N.; Fotakis, C.; Farsari, M.; Karalekas, D. Int. J. Adv. Manuf. Technol. 2009, 48, 435– 441 8. B. Kaehr and J. B. Shear, Proc. Natl. Acad. Sci. USA, 2008, 105, 8850-8854. 9. D. Wu, Q. D. Chen, L. G. Niu, J. N. Wang, J. Wang, R. Wang, H. Xia and H. B. Sun, Lab Chip, 2009, 9, 2391-2394. 10. J. K. Hohmann and G. von Freymann, Adv. Funct. Mater., 2014, 24, 6573-6580. 11. R. Di Giacomo, S. Krodel, B. Maresca, P. Benzoni, R. Rusconi, R. Stocker and C. Daraio, Sci. Rep., 2017, 7, 45897.

Chapter 2

Candidate’s contribution to the authorship

I declared that I have obtained permission from all co-authors to include following publication directly into this chapter.

Caizhi Liao, Alain Wuethrich, Matt Trau. “A Material Odyssey for 3D Nano/ Microstructures: Two Photon Polymerization Based Nanolithography in Bioapplications” Appl. Mater. Today, 2020. 4, 1, 1401-1409. - Incorporated as Chapter 2.

Contributor Statement of contribution

Conception and design (80%) Caizhi Liao (Candidate) Drafting and production (80%)

Conception and design (20%) Alain Wuethrich Drafting and production (15%) Matt Trau Drafting and production (5%)

Chapter 2

Thesis Literature Review

Two Photon Nanolithography for 3D Miniaturized Constructs

Summary

Chapter 2 represents a detailed literature thesis review of the two photon polymerization (TPP) based fabrication platform, and the role that TPP can play in the preparation of miniaturized three dimensional (3D) devices for a wide spectrum of uses. In this chapter, we firstly describe the basics of the TPP process, including polymerization mechanism, experimental set-up and material aspects. To further illustrate the potential uses of the TPP technique, an inclusive table summarising the bio-uses of the TPP-generated miniaturized structures is provided. Next, we present a comprehensive account on the latest advancements in TPP nanolithography for bio-applications, including cell engineering, tissue engineering, drug delivery, bio-robot and bio-sensing, etc,. Following this, challenges and perspectives toward the utilization of TPP platform for preparation of 3D (bio) nano/micro-devices are presented. The candidate believes that this chapter will help to crystallise the key attributes of the TPP technique for the preparation of 3D structures, particularly the miniaturized hollow pore constructs that will be discussed in following chapters.

Chapter 2 is based on a published review paper:

Caizhi Liao, Alain Wuethrich, Matt Trau. “A Material Odyssey for 3D Nano/ Microstructures: Two Photon Polymerization Based Nanolithography in Bioapplications”. Applied Materials Today, 2020. 19, 100635.

Chapter 2

Two Photon Nanolithography for 3D Constructs

in Bioapplications

KEYWORDS

Two-Photon Polymerization (TPP); Nanolithography; Photosensitive material; 3D Nano/Micro-structures; Bio-applications

ABSTRACT

Functional three-dimensional (3D) microscopic and nanoscopic structures have the potential to significantly drive the development of innovative bioapplications. As the only 3D prototyping technique for precise and controllable additive manufacturing of sub-100 nm structures, two-photon polymerization (TPP) nanolithography has attracted significant interest. Central to the TPP fabrication process is the cross-linking of the photosensitive material that builds the construct of the functional 3D micro/nanostructures. Driven by the development of biocompatible photosensitive materials for TPP fabrication, significant progress has been achieved in cell engineering, tissue engineering, drug delivery, microfluidics, and biosensing. This review features a comprehensive and material-focused account on the latest advancements in TPP nanolithography for bioapplications. After a brief introduction of TPP, the review discusses critically photosensitive materials including SU-8, IP resists, PEG-based hydrogels, protein-based biomolecules, hybrid inorganic–organic materials, and others. Finally, current achievements, fabrication limitations, and opportunities for the advancement of TPP-based functional 3D microscopic and nanoscopic materials are summarised.

Chapter 2

2.1 INTRODUCTION

Three dimensional (3D) nano/micro-structures represent a promising paradigm for the complex implementations in bio-systems.1 To achieve these robust 3D architectures, many different types of manufacturing schemes have been established. Among these approaches, in particular, photolithography-based schemes have afforded to transfer geometrical patterns and have dominated the nano/microscale manufacturing fields for over six decades.2 However, efforts to further shrink the feature sizes down to several nano-meter and precisely control 3D patterns still pose tremendous technological challenges for typical photolithography processes.3 Firstly, conventional photolithography essentially is a two- dimensional (2D) planar technique. Although 3D patterns can be achieved through integrated approaches, the precision on vertical dimension of the fabricated structure could not be strictly controlled as that of other two dimensions.4 Secondly, the harsh processing conditions, such as UV deep exposure and HF/reactive ions etching, lead to a limited candidate material pool for processing. It is therefore urgent to develop fabrication techniques that allow accurate preparation of true 3D nano/micro-objects.

Promisingly, additive manufacturing based techniques have emerged as feasible platforms for the creation of 3D nano/micro-structures. First developed by Charles Hull at 3D Systems, stereolithography (SLA) has become a powerful 3D additive manufacturing method since 1986. Under the exposure of light, the photosensitive material is cross-linked to form a single layer of particular 3D structures, a process that is repeated for each layer of the design until the 3D object is completed. SLA can be used to fabricate any designs with high accuracy.5 However, conventional SLA 3D printing system cannot directly write a structure with strictly- controlled feature sizes well below one micrometer.6 Also, the layer-by-layer fabrication scheme of SLA technique puts some constraints for the complex constructs to be realized with sophisticatedly designed supporting structures.

To address the above limitations and further improve the resolution of 3D nano/microstructures, Satoshi et al. developed nano/micro-SLA printing via two-photon polymerization (TPP).7 Since its introduction in 1997, TPP based nanolithography has become an enabling

Chapter 2 technology in many different fields, including photonics8, microfluidic systems9, tissue engineering10 and biomedical engineering11(Figure 1). As reported, TPP is based on the two-photo absorption (TPA) mechanism: the polymerization process is triggered by a tightly focused laser pulse applied within the focal region of the photosensitive materials.12 With the aid of such solid freeform fabrication technique, 3D structures can be prepared by simply washing and removing the uncross-linked materials. The advent of femtosecond lasers integrated in TPP nanolithography systems improved the fabrication resolution to sub-100 nm, which is three orders of magnitude smaller than that of structures prepared by current 3D printing techniques.13 More importantly, the introduction of new construction materials has further driven the development of this field. In particular, when combined with biocompatible photosensitive materials, TPP can be used to fabricate 3D nano/microstructures that possess superior stability and excellent compatibility with biological environments, serving as a versatile toolkit poised to drive advances for many biological implementations.14

Figure 1. (a) Schematic graph of the TPP process. Different types of photosenstive materials, including SU-8, IP-Series, protein hydrogel and hybrid materials, can be utilized for TPP constructions. (b) The versatile uses of TPP generated structures for wide range of bioapplications, including cell engineering, tissue engineering, drug delivery, nano/micro- bio-robots, bio/chemical sensing and micro-fludic system.

Chapter 2

In this review, we will feature the latest advancements in materials for TPP fabrication system, through which robust 3D nano/micro-structures are realized for versatile uses in biological fields. It is not the intention to fully cover all-relevant past works, but rather a selection of highlighted work that can potentially signify the future trends of this particular field. In Section 2, we will briefly introduce the fundamental aspects of the TPP process, including two-photo absorption mechanism, photoinitiators and experimental set up. In Section 3, we will discuss critically and systematically the bio-implementations of TPP based on the explored application and potential in cell engineering, tissue engineering, drug delivery and microfluidic system, nano/micro-robots, biosensors and biomedical devices. In Section 4 and Section 5, we will summarise key findings, point out challenges that remain to be addressed, and provide an outlook on the future development of TPP materials portfolio for bioapplications in the laboratory and industrial settings. Hopefully, this comprehensive review would help to inspire the next stages in this exciting field of research.

2.2 TWO-PHOTON POLYMERIZATION (TPP)

2.2.1 Two-photon Absorption (TPA)

In 1931, Maria Goeppert-Mayer theoretically described the process of multiphoton absorption (TPA), in which interactions between multi-photons and one atom/molecule take place during a single quantum event. As an essential part of TPA process, the collective action of two photons must be presented simultaneously to impart enough energy to induce the transition.15 Due to the requirement of high photon intensities, TPA had not been experimentally demonstrated until the advent of ultrafast lasers. Thirty-years later, Kaiser and Garrett firstly observed the phenomenon of two-photon absorption event.16 In their pioneering work, the fluorescent blue light (425 nm in wavelength) was irradiated by 2+ illuminating a CaF2: Eu crystal with the pulse emitted from ruby laser (694 nm in wavelength). In TPA polymerization, simultaneous absorption of two photons within a small focal volume in the photosensitive materials triggers the chemical reactions between photon initiator molecules and monomers.17, 18 The instantaneous photon intensity within a pulse is considerably high, facilitating the TPA polymerization reaction. Yet, as the repetition period of ultra-fast pulses are five to six orders of magnitude shorter than that of a conventional laser, the average generated power is extremely low. Today, a laser capable of generating femtosecond pulses is normally used for the generation of pulses with high photon intensity 13

Chapter 2 restricted in a focal volume.19 Considering its short pulse width and high peak power, femtosecond Ti: sapphire lasers operated with 800 nm wavelength are widely utilized in TPP fabrication process (Figure 2a).

The atomic or molecular absorption of two photons can proceed by sequential absorption (Figure 2b) and simultaneous absorption (Figure 2c). In sequential absorption, the absorbing species transfers into a real intermediate states by the absorption of the first photon, then the second photon is absorbed which further brings the species into a excited state.15 Typically, this absorption process has a well-defined lifetime within the range of 10- 9 to 10-4 s. In the simultaneous absorption process, the intermediate energy state disappears as the species escalate into the excited states. The absorption process can be considered as an initial interaction of a photon with the atom/molecule to form a temporary virtual energy state above the ground state, which can only exist for an extremely short time interval. If the second photon arrives and interacts with the atom/molecule within this virtual state lifetime, it then can be adsorbed and excited to the higher state. Based on the uncertainty principle of the photon in visible and near-IR (NIR) ranges, the virtual state lifetime is estimated on the orders of 10-16 to 10-15 s.

Chapter 2

Figure 2. (a) Schematic picture of a typical TPP fabrication system consisting of a femtosecond laser, CCD camera, 3D movable stage and controlling system. Reproduced with permission from ref.20 (b) Schematic diagram of a sequential absorption process. (c) Schematic diagram of a simultaneous absorption process.

2.2.2 Polymerization Process

Enabled by the TPA process, the photon-sensitive materials undergo polymerization reactions to form 3D nano/micro-structures. To facilitate two photon polymerization, a mixture of reactive components, including photonsensitizer chromophore and photoinitiator molecule, is added into the bulk photonsenstive material. In polymerization, a high-intensity laser beam is precisely focused into the tiny focal volume of photosensitive material. The chromophore excited by the simultaneous absorption of two photons emits fluorescent light in the UV-vis regime, leading to the enhancement of two-photon activation. Highly photochemical-active photoinitiators then absorb the fluorescent light to generate radicals (Initiation). The radicals serve as the activator for the monomers or oligomers, producing monomer radicals to expand in a chain reaction (Propagation) until two radicals meet (Termination).21, 22 The overall polymerization procedure is described as follows:

Initiation:

 MM Propagation: RMRMRMMRM  n

Termination:

in which S is the photosensitizer, I is the photoinitiator, M is the monomer, and R* is the active radicals. I * and S * represent the excited state of photosensitizer and photoinitiator, hυ is the Planck’s Equation energy for each adsorbed photon, and FL is the fluorescent light emitted by the excited state photosensitizer, respectively. Importantly, eligible photosensitizer plays a central role in the high-precision TPP fabrication of 3D nano/micro- objects.23, 24 With the aim of realizing full potential of TPP-enabled fabrication scheme, highly active photosensitizer chromophore molecules are in great demand. Typically, the activity of chromophore can be enhanced by increasing the TPA cross section ( δ TPA ) inside these molecules. To definitely display the activity of photosensitizer molecules, the molecular two-

Chapter 2 photon cross-section is usually quoted in the units of Goeppert-Mayer (GM, 1 GM = 10-50 cm4 s photon−1) 25, 26

Recently, the design and synthesis of novel chromophores with enhanced δ TPA has garnered extensive attentions (Figure 3). To fully understand the relationships between molecular structural factors and two-photon absorption properties, Perry, et al concluded several molecular design strategies to improve the cross-section of TPA chromophore: ① Extend the π-conjugation length; ② Increase the planarity using fused aromatic rings π-bridge; ③ Increase the strength of acceptors and donors; ④ Introduce chemical functionality with high efficiency of initiation.23

Although significant advances have been achieved to produce enhanced active TPA chromophores, very few of them are suitable photoinitiators for the TPP process. Firstly, most TPP photoinitiators are hydrophobic, which strictly prohibit the fabrication of 3D hydrogels structure containing residue organic solvents. Additionally, the TPP photoinitiators should demonstrate high biocompatibility with periphery environments for biological applications. Therefore, the absence of high-efficiency TPP initiators compatible with the aqueous medium still remains as great challenge for the fabrication of 3D nano/microstructures that designed for bio-implementations. To circumvent the problem, Duan, et al. proposed a universal method to prepare the high efficient water soluble TPP initiators using the host-guest chemical interaction strategy.27

Chapter 2

Figure. 3 Molecular structures of different types of two-photon absorption dyes with distinct cross-section values. I: π-conjugated benzenoid donor-bridge-acceptor dyes fluorene derivatives dyes. II: Push-pull dipolar dyes. III: Double bonds enhanced π-linked dyes.

2.2.3 Experimental Setup As schematically described in Fig. 2a. the TPP fabrication system typically consists of a femtosecond-pulsed laser source that emits focused laser beams and a piezoelectric 3D scanning stage that controls the focus of the beam within the photosensitive materials. In 3D nano/micro-fabrication, the TPP platform can be performed either by moving the laser beam inside the photosensitive materials (Galvo-mode) or by moving the sample across the fixed laser beam (Piezo-mode). For the first scenario, a set of Galvo mirrors is used to scan the focused laser beams in the x and y dimensions, and a z-direction piezo stage to control the sample/objective up and down. In the second scheme, the sample controlled by a three- axis piezoelectric stage can be precisely moved in all three dimensions. Both techniques have proven to be effective for the fabrication of 3D nano/microstructures. Due to its high writing speed, the Galvo-mode is typically used for high throughput fabrication, while the Piezo-mode offers the fabrication of larger 3D structures by increasing the stage travel distance. Importantly, real-time monitoring of the TPP process can be achieved by the integration of a charge-coupled device (CCD) camera.28

The interaction volume of the laser with the photosensitive material is depending on the laser wavelength and described by λ3, where λ is the laser wavelength. The two-photon absorption process then decays quadratically with distance away from the interaction volume. To further decrease the interaction volume, several TPP systems equipped with high numerical aperture (NA) oil immersion mode lenses are established to achieve high resolution for focusing and fabricating.29 However, the height of structures prepared by oil lenses mode are limited to 1mm. To resolve the height limitation, air lenses can be used instead of oil lenses and structures of up to 30 mm can be prepared. Currently, TPP platform reportedly has shown to fabricate 3D structures with a resolution down to 100 nm, which

Chapter 2 can be realized through the commercial available TPP platform (e.g., Nanoscribe Professional GT system) by using designated resist materials20, 30.

3D models for TPP direct writing process could be originally generated by the computer- aided design (CAD) program, e.g., SolidWorks®, AutoCAD®, AutoDesk® and Creo Parametric®. The sketched 3D design from a CAD program then is converted into a STL (Standard Tessellation Language) file that stores the information of each surface of the 3D model in the form of triangulated sections. By increasing the number of triangles defining the surface, more data points used to spatially describe the surface parts will be stored in the text file.31 Afterwards, the STL file is further processed by the slicer software to create the interpreted G-file. The G-file divides the 3D models into a sequence of 2D horizontal cross sections, allowing the 3D structures to be sequentially printed from the base layer. The processed 3D model file is then imported into the TPP controlling system for the DLW process.

2.3 MATERIALS

Currently, the classical photosensitive materials pool utilized for TPP-based nanolithography are similar to those used in conventional lithographic applications. With negative photosensitive materials, two-photon exposure results in solidified 3D structure via the cross-linking of monomer chains, allowing the unexposed resist to be washed away in the development stage. Negative-type photosensitive materials are well-suited for high- resolution fabrications because of broad material portfolio and superior chemical stability. Example of widely-used and commercially-available TPP resists are SU-8 (MicroChem)-the most popular UV lithographic photoresist32, and the hybrid sol-gel ORMOCER® (Micro Resist Technology)33. For the positive-type photosensitive materials, laser light beam exposure leads to chain scission, creating shorter units that can be dissolved and washed away by the development solutions. Importantly, positive tone materials are more efficient for the fabrication of hollow structures that can be machined by removing only a small fraction of the total material in the original designs34. However, as most aforementioned photo-sensitive materials are proprietary materials, they cannot be easily modified to add active components for customized functionality, a disadvantage severely prohibiting the use

Chapter 2 of TPP technique in universal fields, particularly for biological implementations that require targeted chemical/biological functionalization.

The development of novel candidate materials for effective TPP processing, therefore, has become an active field of research. Photosensitive materials for bio-related applications should generally possess the properties as follows18, 35-37:

I. Transparent in visible and near infrared (NIR) regions: The laser pulses can be precisely focused within the voxel volume of photosensitive material without inducing any single- photon interactions.

II. Fast curing speed with controllable shrinkage and geometrical distortion during TPP: The polymerized part needs to confine to the focal spot and demonstrates high stability in development solvents.

III. Refractive index of materials before fabrication should be close to 1.5: The laser beam can be focused deeply into the resist materials without aberrations.

IV. Suitable viscosity with adequate adhesion force to substrates: TPP nanolithography generated 3D structures can tightly attach to substrates for further uses.

V. High mechanical strength: The formed solid structure should maintain its shape under shrinkage effects.

Recently, a number of novel photosensitive materials have been developed for the TPP nano/micro-fabrication, some of which have also been commercialized. For example, Greer et al. presented a facile approach for fabricating functional 3D structures for use in the biomedical setting by precisely-controlled polymerisation of acrylate monomer via the thiol- Michael addition reaction.38 For some bioapplications such as tissue engineering, the biocompatibility is an additional important consideration for the selection of the photosensitive material. A wide range of biocompatible and bioresorbable materials have been investigated, including polylactic acid (PLA)39, polycaprolactone (PCL)40, polyglycolic

Chapter 2 acid (PGA)12, trimethylene carbonate (TMC)41, and biomolecular protein structures42. Hydrogels scaffolds with superior biocompatibility are also exploited as candidate materials for TPP-based direct structuring in aqueous medium.43, 44 This section systematically discusses photosensitive materials for TPP fabrication in bio-relevant fields, including cell engineering, tissue engineering, drug delivery, bio-microfluidic system, bio-nano/micro robots, bio/chem-sensors and bio-inspired surface. Table 1 provides an overview of the reviewed works grouped according to photosensitive material used into photoresists, hydrogels, protein molecules, hybrid materials and others.

Table 1. Photosensitive materials utilized in bio-relevant applications for TPP fabrication of 3D nano/micro-structures.

Materials Structure Application Performance Referenc e Reversible movements 45 Micro-valve between “ ON ” to “ OFF ” states controlled by water flow Micro- Microfluidic Versatile platform for bio- 46 component system related uses Micro-fluidic On-device trapping and 47, 48 device imaging of individual yeast Micro- Microfluidic channels with 49 SU-8 channel different aspect ratios and cross-sections within a single device Hexahedral Multifunctional microrobots 50 microrobot for targeted cell delivery 3D Micro- Nano/micro- Spatiotemporally controlled 51 transporter bio-robot delivery of therapeutic agents Microrobot A magnetic microrobot for 52 carrying and delivering targeted cells Wireframe Cell Capable of probing the 53 structures engineering interplay of different types of cells 3D tetrapod Drug Sperm-driven micro-motor 54 delivery structure for controlled drug release 3D scaffold 3D imaging of polymeric 55 freestanding architectures for cell colonization 3D scaffold Cell 3D spatially defined 56 engineering proteinaceous environment

Chapter 2

Square Enhanced control over cell 57 grids morphology and proliferation

Woodpile Directed cell migration 58 structure analysis Microtube Provides high osteogenic 59 Tissue capability for the cells Porous tube engineering A 3D biomimetic, biohybrid 30 IP-Series: model of blood-brain barrier IP-L, IP-S, IP-Dip Helical Magnetic helical micro- 60 micro- swimmers functionalized with swimmers Nano/micro- lipoplexes for targeted gene bio-robot delivery Helical Surface-chemistry-mediated 61 micro- control of individual magnetic swimmers helical Micro-swimmers in a swarm Hydrophilic/ Rapid prototyping of 62 hydrophobic biocompatible surfaces with surface designed wetting properties Lotus type Bioinspired surfaces for 63 biomimetic Bio-inspired controlling the wettability of surface surface materials and devices Super Superhydrophobic surfaces 64 hydro- based on fractal and phobic films hierarchical microstructures Micro-traps Bio/chem- Deployable micro-traps 65 sensor significantly reduce biofouling Micro- Bio-medical Direct 3-D printed 66 needles device microneedles allow in-vitro perforation Micro- Drug An implantable micro-caged 67 caged delivery device device for direct local delivery of agents 3D scaffold Human dermal fibroblast cells 68 were seeded with selective bioactivity High quality structures with 40 3D scaffold little cytotoxicity for cell proliferation Cell 3D engineering Enable guidance of cell 69 scaffold migration along pre-defined 3D pathways 3D scaffold Scaffold for 3D culturing of 70 PEG-based (Porous and human mesenchymal stem Hydro-gel non-porous) cells 3D cube Full control over cell 71 array adhesion and cell shape in 3D for the first time Highly Preparation of multicellular 72 porous 3D tissue scaffolds constructs and artificial ECM

Chapter 2

3D scaffold Significantly reduced long- 73 term material toxicity 3D Tissue Provide a favourable 74 scaffold engineering environment for neuronal cell growth 3D Tailor-made 75 scaffold mechanochemical properties of HA bioscaffolds Micro- Gentamicin-doped 76 needles microneedles array Drug with antimicrobial delivery functionality 3D Controlled drug release 36 woodpile profile Micro- Responsive bacterial 42 enclosures cages capable of harvesting cell colonies Cube; Cell 77 engineering High-resolution protein μ-chess microstructures for cell board; growth investigations

Hole; BSA Pyrimidal Engineering of complex 78 micropillar protein microstructures array mimicking the native matrix niche Free- Chemical sensing in specified 79 floating 3D microenvironments object Möbius Bio/Chem- 80 strips; sensing Tuneable and spatially heterogeneous mechanical Paddlewhe properties and environmental els; sensitivity Protei n – free-floating based probes materi Stepped Drug Sustained drug release 81 als pyramids delivery systems Controllable cell orientation 77 3D scaffold Cell by cell embedding in engineering accurate 3D structures 3D scaffold New dynamic caging culture 82 systems Gelatin- 3D Support porcine 83 based scaffold mesenchymal stem cell materials adhesion and subsequent Tissue proliferation 3D engineering Support human primary 84 scaffold adipose-derived stem cell (ASC) adhesion, proliferation and differentiation Patterned Control the orientation of a 85 plugs cell-synthesized 22

Chapter 2

extracellular matrix Porous Contiguous fabrication of 86 woodpile structures with varying structure mechanical and biochemical Collagen Tissue properties 3D engineering Tissue reconstruction and 3D 87 hydrogel cultivation scaffolds Fibrin 3D Cell Uniform seeding of 88 protein scaffold engineering endothelial cells on the structure Lego-like Acceptable cell viability and 89 inter-locking cell growth profiles against 3D scaffold B35 cells

Controlled morphologies of 90 Aligned Cell fibroblast cells and epithelial fiber engineering cells under the effect of high curvature 91 Ridge and groove Controlled cell alignments on patterned printed patterns Ormocer surface ® 92 Woodpile 3D cell culture for tissue

structure engineering of artificial cartilage Micro- Medical devices with 89 needle enhanced efficiency for drug delivery Drug Fabrication of various 93 Micro- delivery microneedle structures for needle transdermal drug delivery Micro- Fabrication of microneedles 94 needle with antimicrobial array functionality Ossicular Structures provided 95 replacemen Medical acceptable cell viability and t prostheses device cell growth profiles; structure

3D vascular 96 Hybri micro- Tissue Enhanced formation d capillary engineering capability of materi cell–cell junctions als Micro-fluidic Realize fluidic channels of as 47 channel Microfluidic small as 110 nm in width

3D Bio/chem- Biotin anchored to the 3D 97 woodpile sensor structures for specific cubic detection structure

Chapter 2

Aligned Submicrometric 98 ridges structures for the study of the axon mechanobiology 3D scaffold Precise control of cell matrix 99 Cell adhesions Trabecula- engineering Enhance the differentiation of 100 like bone-like cell Ormo- structure Comp® Patterned Bioinspired 3D structures 101 surface resembling trabeculae of sponge bone Micro- Drug Mediate the creation of tissue 102 needle delivery with precisely controlled structure Micro-fluidic Microfluidic Simple, fast and inexpensive 103 channel system process for creating enclosed fluidic channels for bio- analysis Nano- Highly reproducible self-affine 104 structured Bio-surface structures facilitate the Brownian investigation of stem cell surfaces responses 3D Favoured cell invasion and 105 scaffold cell adhesion Zr- Cell containin 3D engineering Cells adhered to the 106 g hybrid scaffold scaffolds, produced neurites, material elongated and differentiated into adult neurons Micro-valve Microfluidic Replacement of natural 107 system check valve in human veins OMR83 108 containin g carbon Glucose nanotube sensor Highly sensitive chip-based and electrode glucose sensor glucose enzyme IP-s 3D Printed Bio/Chem 3D-printed carbon electrodes 109 /carbon carbon sensor for neurotransmitter detection electrodes 3D structure 110 IP-s enhanced A bio-inspired 3D micro- IP-S /Graphen graphene structure for graphene based based e sensor bacteria sensing comp IP-s/ 111 osite cobalt- Robotically controlled micro- materi nickel Microprey prey to resolve initial attack als micro- modes preceding structure Nano/Micro- phagocytosis s robot IP-s/ 112 nickel- Manage to capture, transport, titanium Cellular and release single immotile micro- cargo 24

Chapter 2

structure live sperm cells in fluidic s channels PVA 3D scaffold Allow precise control of cells 113 and their 3D microenvironments TPETA 3D scaffold Guiding cell attachment in 3D 114 Cell microscaffolds Microporou engineering Micro-structured scaffolds 115 PETTA s with adjustable pore size and 3D scaffold diffusible chemical gradients Triacrylat 3D extra- Enable the study of cell 116 e resist cellular adhesion and migration Other matrix behaviours in controlled way s Accura® 3D Primary hepatocytes 10 SI10 scaffold uniformly seeded with higher levels of liver-specific function Polylacti 3D Tissue Suitable for peripheral neural 39 c acid scaffolds engineering tissue engineering (PLA)

Suitable compressive 117 eShell Micro- strength for use in polymer needle transdermal drug delivery

Drug Acrylate- delivery Effectively inject quantum 118 based Micro- dots into porcine skin polymer needle Polyacryl Quick response to external 119 amide Micro-pump stimulating factors Hydrogel Micro-fluidic Micro- Movable micromachines 120 SCR-701 machines force-driven by optical stimulation

2.4 TPP For BIO-USES

2.4.1 Cell Engineering

Cells in body are typically supported by a complex 3D extracellular matrix (ECM), which extensively interacts with cells and regulates cell growth, proliferation, and differentiation. The cell behaviours are dominantly manipulated both by the molecular composition of the cell-ECM contact sites and the spatial distribution at the interfaces.121 . Because of the sophisticated ECM structure, traditional two-dimensional (2D) cell culture plate (e.g., Petri dish) can not be used to accurately mimic the real biological environments in vivo.

Chapter 2

In the field of cell engineering, the realization of reliable 3D architectures is of critical importance for the investigation of cell growth, proliferation, and differentiation. To generate potential 3D structures for cell colonization, Vieu et al. reported the fabrication of IP resist polymeric freestanding scaffolds for cell studies. To evaluate the efficiency of the 3D scaffold in terms of cell colonization, they seeded neuroblastoma N2A cells and characterized these with advanced fluorescence imaging techniques to clearly identify the position of cells within the scaffolds. Promisingly, their work opens appealing opportunities for the use of the developed TPP fabrication protocols in neuro-research fields.55

Proliferation and differentiation of cells are strongly influenced by the 3D topographies. TPP was explored for preparation of a 3D multi-component micro-environment consisting of IP resists and proteinaceous networks with submicron-sized features.56 These 3D proteinaceous microstructures demonstrated high performance for NIH/3T3 fibroblast cells culturing. Significantly, the ability to integrate proteinaceous elements for 3D spatial control of microenvironment in cell culture provides the possibility to control behaviour at single cell level. To increase the biocompatibility of well-defined templates fabricated by TPP 3D nanolithography, Hohmann and colleagues coated a titanium dioxide layer on the surface of 3D structures made of IP resist (Figure 4a).57 A significantly higher proliferation (170%) was observed on the coated 3D structures compared to the untreated surfaces (100%), rendering the surface treatment as an efficient strategy to improve cell proliferation.

The study of cell migration through connective tissue has garnered increasing interests in the biological areas. Traditionally, cell migration studies have been primarily performed on planar glass or plastics substrates. However, 2D culture environment cannot fully mimic the 3D extracellular matrix environment of the living body. With the aim of creating ideal constructs for cell migration analysis, Larsen et al. developed an in-chip TPP process for the preparation of 3D microporous constructs.58 This novel TPP fabrication scheme provides accurate control of both the pore size and the topology, posing no negative effects on the behaviour of migrating cells. More importantly, the source/drain design of chip provided complete control of the cell concentration and thus the steepness of the chemokine gradient within the built constructs. Hopefully, this 3D IP resist construct integrated with chip would

Chapter 2 be usefully for the validation of the migratory potential of each patient's cells in immunotherapy.

PEG-based resins have been widely used for the preparation of hydrogel structures using stereolithography.68, 122 These hydrogel structures demonstrate excellent biocompatibility and even allow the encapsulation of living cells during the fabrication process.69, 72 Cell culture scaffolds interactions serve as the key regulators of cell survival, proliferation, and differentiation. To investigate the 3D cell behaviours in in vivo environments, Lee et al. reported the use of TPP photolithographic technique to micro-pattern cell adhesive ligand (RGDS) in collagenase-sensitive poly(ethylene glycol-co-peptide) diacrylate hydrogels (PEG-DA) to guide cell migration along pre-defined 3D pathways.69 Furthermore, the cell migration behaviours could be delicately guided by providing appropriate bioactive cues in highly defined geometries.72

Figure. 4 (a) Confocal laser scanning images of fluorescent stained cells on different structures prepared by TPP systems. Reproduced with permission from ref.57; (b) Cell growth in 3D composite-polymer scaffolds prepared by sequential TPP of PEG- DA/Pentaerythritol triacrylate (PETA) materials. Reproduced with permission from ref.71;

Chapter 2

To control the spatial, mechanical, temporal and biochemical architectures of scaffolds, Arcauten et al. fabricated the PEG- diacrylate (DA) hydrogel scaffolds structure using TPP nanolithography system.68 The scaffolds were modified with controlled concentrations of fluorescently labelled dextran and bioactive PEG in different regions of the scaffold. Human dermal fibroblast cells were seeded on the top of fabricated scaffolds with selective bioactivity. Phase contrast microscopy were used to show specific localization of cells in the regions patterned with bioactive PEG. This strategy provided a promising approach for the precise spatial control over cell bioactivity. Similarly, Klein et al. reported the use of PEG- based scaffolds for 3D cell culturing investigation: cell adhesion and cell shape can be fully controlled in three dimensions, see Figure 4b.71

2.4.2 Tissue Engineering

Tissue engineering is recognized as an interdisciplinary field that combines research on cells, materials, and biochemical factors to create artificial organs and tissues.123 Despite the recent advances in the development of functional scaffolds to support cell activity and potentially promote the repair of different tissues, the development of cost-effective approaches for tissue engineering still remains a great challenge.124 Since the scaffold serves as a carrier for cells to grow, it is vital for the scaffold to mimic the nature of human tissue to direct the macroscopic process in tissue formation. Following design rules should be considered when fabricating a tissue scaffold: (1) Suitable biocompatibility for cells to attach and proliferate, appropriate mechanical stability to support the attached cells, and controllable biodegradation profile in tissue regeneration stage; (2) Extensive network of interconnecting pores in scaffold allow the cells to adhere, growth, differentiate and migrate deep within the structure; (3) Well-defined channels through which nutrient elements and oxygen can be delivered to cells deep inside the scaffold, while cellular waste products can be easily excreted. 92, 125, 126

As a novel direct laser writing technique allowing the fabrication of any computer-designed 3D structure from a photosensitive material, TPP can precisely define 3D construct geometry, which enables the direct vascularization for patient-specific tissue fabrication. To facilitate the cell penetration into the inner 3D implantable structures, Luculescu et al. 28

Chapter 2 presented an innovative 3D hierarchical, honeycomb-like structure (HS) for improved tissue engineering implementations using IP-L resist.59 These fully interconnected porous structure demonstrated high osteogenic capability for the cells, a feature that could significantly enhance the biological performance. Recently, the investigation of the crossing of exogenous substances through the blood-brain barrier (BBB) has garnered increasing interest in biomedicine research. With the aim of developing realistic models for this type of barrier structure, Ciofani et al. proposed the TPP fabrication of a 3D real-scale, biomimetic, and bio-hybrid model for investigation,30 as shown in Figure 5a. Built from IP resist material, these constructs have been successfully exploited as a robust in vitro model for the study of nanomaterials and drugs in BBB crossing process. Therefore, IP resist 3D scaffolds generated from TPP platform serve as a robust model for the 3D neuronal tissue engineering studies.

Figure. 4 (a) TPP fabrication of the microfluidic system containing porous tubular-shaped 30 micro-capillaries. Reproduced with permission from ref. ; (b) Neuro2A cells colonized on the 3D hydrogel constructs. Reproduced with permission from ref.74

In the last few years, PEG-based scaffolds have attracted increasing interests in tissue engineering fields. For instance, to further elucidate the cell behaviour phenomenon on artificial scaffolds with controlled three-dimensional topologies, Ovsianikov et al. fabricated the PEG-DA based scaffolds by means of TPP.73 An extensive study of the cytotoxicity of material formulations with respect to photo-initiator type and photo-initiator concentration indicated that the presence of water soluble molecules was toxic to seeded fibroblasts. However, sample aging in aqueous medium could significantly reduce the cytotoxicity of 29

Chapter 2

PEG-DA based 3D scaffolds, providing a route for tissue engineering applications of structures generated by TPP fabrication process.

As a crucial component of human ECM, hyaluronic acid (HA) represents an attractive candidate material for the fabrication of tissue scaffolds. Kufelt et al. created modified PEG- based materials with tailor-made mechanochemical properties for the TPP writing.75 HA was covalently cross-linked with PEG-DA to construct well-defined 3D HA−PEG-DA microstructures with different geometries and pore sizes. PEG scaffolds were also prepared for the efficient colonization of neuro cells and 3D confocal imaging (Figure 5b).74

2.4.3 Bio-microfluidic Systems

Defined as the handling and analysing of fluids at a micrometre scale, microfluidics has become an enabling technology for the study of many biological entities. Considering the ability to seamlessly combine several laboratory functions onto a single chip, microfluidic devices pose a significant advantage over traditional assays used in wide biological areas.127 Different microfluidic systems have been developed for biomedical research, from those with a singular function to integrated multiple functional analytical systems. Currently there are many different techniques for fabricating microfluidic devices, such as micromachining, soft lithography, embossing, in situ construction injection moulding and laser ablation.128 However, some of those techniques suffer from several critical problems, like requirement of highly sophisticated fabrication system, time consuming, labour intense and limited material pool available for the device preparation. Recently, advancements in TPP 3D direct writing have helped to simplify the formation of microfluidic devices into a single step. Microfluidics fabricated by TPP technique have several notable advantages attractive for bio-relevant uses, including the integration of tissue scaffold with high porosity, high resolution and well-defined structures within the device, and wider range of fabrication material choices.129, 130

Chapter 2

Promisingly, TPP allows highly efficient microfluidic device fabrication and easy integration of multiple components for complex microfluidic analytical systems. Stoneman et al. reported an improved method for the production of SU-8 photoresist based microfluidic master structures using TPP.48 The device masters served as negative relief structures for PDMS-based microfluidic devices. Upon the exposure of focal laser beam, SU-8 photoresist emitted fluorescence in the visible range, allowing the production progress of the microfluidic master to be monitored in real-time. Interestingly, fluorescence monitoring of photoresist material provides a feasible pathway for the simultaneous monitoring of biological cells and preparation of microfluidic devices for controlling cell microenvironment. Jariwala et al. developed a simple, fast and repeatable method for fabricating microfluidic channels by TPP assisted ablation.47, 103 Compared with the conventional direct ablation, TPP assist ablation requires significantly less energy and has greater control over desired feature size and spatial resolution. The channel width and depth could be precisely tuned through the manipulation of laser beam parameters and fluidic channels as small as 110 nm in width between polymerized parallel ribs could been realized.

Serving as the traffic light of microfluidic systems, microvalves precisely control the flow within the device and play a crucial part for automation of the bio-analysis process. With the aid of TPP technique, Wu et al. reported the accurate preparation of a microscale valve in a single step using SU-8 photoresist.45 The 25 µm length functional microvalve integrated in a microfluidic channel could reversibly control the fluidic movements between “ ON ” to “ OFF ” states. Light responsive microscale structures can also be produced using TPP. For example, Rodrigo et al. proposed a novel optical micro-assembly platform using multiple real-time adaptive counter-propagating-beam (CB) traps to construct reconfigurable microenvironments.46 The feature-rich microstructure building blocks were fabricated via the direct laser writing process of SU-8 based biocompatible materials.

SU-8 exposure to near-infrared beam can lead to laser-induced damage and deform the generated 3D patterns. To optimize the use of SU-8 in TPP process, Kumi et al. employed a novel PAG, namely the 9,9-diethyl-[bis-(styrytriphenyl sulfonium hexafluoro phosphonate)] fluorene (FLUOR), in an SU8 resist for the fabrication of microfluidic devices.49 The high TPA cross-section of this new PAG effectively eliminated laser-induced damage concerns. Uniform high-aspect-ratio masters with arbitrary, non-rectangular cross-sections were 31

Chapter 2 created. Microfluidic channels with different aspect ratios and cross-sections could even be realized within a single device. Furthermore, high-speed fabrication over centimetre-scale distances could be performed by using a high writing speed, which enabled the fabrication of large-scale microfluidic masters within a practical time scale. In order to better control the formation of fluidic channels in TPP process, an ultrafast laser was employed to polymerize parallel ribs, and subsequently two adjacent ribs were overlapped via the polymerization to form a closed channel.

Zr-containing hybrid material also find its application in micro-fluidic fields. Enabled by TPP fabrication process, Schizas et al. reported on the conceptual design and fabrication of a complex shaped micro check valve with internal moving components.107 The Zr containing organic–inorganic hybrid photosensitive sol-gel material used for the valve fabrication exhibited negligible distortion during photon induced polymerization. A preliminary computational fluid dynamics study simulating the blood pressures of healthy human veins was carried out to evaluate the flow performance of the fabricated valve. The results indicated that the valve device could replace the natural check valve controlling the blood flow in human veins.

2.4.4 Nano/micro Bio-robots

As an emerging field, artificial nano/micro-robots have attracted substantial interests of researchers from many different fields. Mimicking the biological machines in nature, these man-made nano/micro-robots serve as robust toolkit to tackle the thorny problems encountered in the intricate biosystem, leading to an achievement that may solely exist in Hollywood blockbuster films. For instance, tailored miniaturized robots can monitor bio- processes131, modulate bio-environments132 or even cure diseases133, 134. Promisingly, with recent advances in TPP technology having led to the implementation of 3D customizable, miniaturized structures, these benefits of realizing accurate fabrication of robust miniaturized robots are stronger than ever.

In the last few years, Nelson group and their collaborators presented a series of seminal works on the TPP-prepared magnetic nano/micro-robots structures. Early in 2013, they reported the use of photocurable SU-8 for the fabrication of porous micro-niches as a

Chapter 2 transporter in 3D cell culture and targeted transportation.50 In preparation, nickel (Ni) layer was coated onto the surface of SU-8 based object to achieve magnetic actuation, and to ensure biocompatibility for in-vivo applications. To achieve spatiotemporally controlled delivery of therapeutic agents, they later created a 3D SU-8 based compound micro- transporter that enabled to actively collect, encapsulate, transport, and controllably release of target agents51, as shown in Figure 6a. Similarly, Sun et al. developed a magnetic microrobot system for targeted cell delivery.52 The structural design of the microrobot design was optimized and shown to enhance magnetic driving capability, promote cell-carrying capacity, and benefit cell viability (Figure 6b).

Figure. 6 (a) SU-based micro-transporter structures prepared by TPP system. Reproduced with permission from ref. 51; (b) 3D magnetic microrobots for 3D cell culturing and targeted transportation/delivery uses. Reproduced with permission from ref. 50, 52

The operation of miniaturized obstructs interacting with living cells has increasingly attracted attentions in the last few years. With the aid of TPP fabrication platform, Schmidt et al. 33

Chapter 2 prepared a hybrid cargo vesicle made from a sperm cell modified with magnetic micro- helices for targeted cellular cargo delivery. The IP-S polymeric micro-helices were fabricated by TPP nanolithography and then coated with a NiTi soft-magnetic bilayer. Their findings showed that controllable 3D motion with speeds comparable to fast swimming sperms-like microorganisms could be realized under the influence of rotating magnetic fields generated by a set of Helmholtz coils. Promisingly, these metal-deposited IP-S based micro-helices are suitable for the capture, transport, and release of single immotile live sperm cells in mimicked physiological systems. Furthermore, through proper means of sperm selection and oocyte culturing, important steps toward fertilization can be addressed.112

Accurate control of engineered microprobes is essential for the study of the underlying dynamics and mechanics of phagocytosis at the cellular level. Recently, the Nelson group introduced a non-spherical, micrometer-sized magnetic prey obtained using 3D-printed templates.111 This novel system provides unconstrained and accurate control of both position and orientation of arbitrarily shaped particles, opening doors for 3D twisting experiments under dynamic/constant force and torque. As a proof-of-concept study, the proposed robotically controlled micro-prey was utilized to probe the translational and rotational attack modes of phagocytes with high spatial and temporal resolution.

2.4.5 Drug Delivery

For transdermal drug delivery, one risk involving the use of microneedles is the infection associated with formation of channels through the stratum corneum layer of epidermis. To reduce the infection risk, microneedle arrays with antimicrobial properties were demonstrated.76 The agar plating assay revealed that microneedles fabricated with polyethylene glycol 600 diacrylate containing 2mg mL-1 gentamicin sulfate could inhibit the growth of Staphylococcus aureus bacteria. Additionally, scanning electron microscopy proved that no platelet aggregation was observed on the surface of either platelet rich plasma-exposed un-doped polyethylene glycol 600 diacrylate microneedles or the gentamicin-doped polyethylene glycol 600 diacrylate microneedles. By manipulating a number of parameters during the TPP 3D printing process, including slicing and hatching distance, Salem et al. successfully realized PEG based woodpile devices for controlled drug 34

Chapter 2 delivery. These devices were formed with varying porosity and morphology, representing a precise model for controlled drug delivery.36 Due to biocompatible properties of PEG, the prepared devices were suited for a range of common-used cell lines.

Recently, TPP direct laser writing technique also has been explored as novel fabrication platform to create Ormocer based microneedle arrays for effective transdermal drug delivery.135 Ovsianikov et al. firstly fabricated microneedle arrays produced by TPP of Ormocers US-S4 material.93, 102 A large range of in-plane and out-of-plane hollow microneedles were prepared and shown to be compatible with human epidermal keratinocyte growth on their surface. Importantly, the microneedle arrays could penetrate into porcine tissue surfaces without needle fracture. The microneedles could also be integrated with control systems for the continuous monitor, analysis and treatment of chronic diseases such as diabetes mellitus. To reduce the infection risks, the surface of the microneedles was covered with antimicrobial silver coating to prevent microorganisms penetrating the stratum corneum layer of the epidermis often causing systemic infections.94

Mechanical stability of micro-needles is a key parameter in drug delivery. Micro-needles without adequate stiffness will experience damage when penetrating the skin to deliver the drug. The introduction of robust materials could lift the mechanical properties of microneedles built. For example, Gittard et al. used a combination of TPP technique and polydimethyl siloxane micromolding to fabricate microneedle arrays out of the eShell 200 polymer resist.117 The microneedle arrays were able to withstand a ten Newton axial load without fracture, and were capable to penetrate human stratum corneum and epidermal. The fabricated microneedles provided appropriate structural, mechanical, and biological properties that are promising for the high performance delivery of protein-based pharmacologic agents. Later, Gittard et al. further fabricated hollow microneedles made of acrylate-based photoactive polymer to deliver quantum dots to the stratum corneum layer of cadaveric porcine skin.118 The quantum dots with diameters between 2 and 10 nm were found to be uniformly distributed in the deep epidermis and dermis within fifteen minutes. In order to image neoplastic tissue, the fluorescent quantum dots could be conjugated with peptides, antibodies, aptamers, pharmacologic agents, and other tumour-specific molecules, a property that could widen the clinical application of microneedles.

Chapter 2

2.4.6 Bio/chemical Sensing

Kaehr et al. reported the use of a low-cost Q-switched laser as a source for direct-write fabrication of functional BSA based protein matrixes.79 Physically robust and chemically responsive microstructures can be prepared rapidly with feature sizes smaller than 0.5 μm, and cross-linking can be realized both using biologically benign sensitizers (e.g.,flavins) and the proteins themselves to induce cross-linking. Later, the same group created the BSA based stimuli-responsive biomaterials for the 3D sub-micrometer structures.80 Rapid and reversible volume changes in response chemical environments can be achieved using protein blocks with differing hydration properties. Protein matrices enable the precise mechanical manipulations by means of changes in hydrogel size and shape. Protein-based objects that retain rotational and translational degrees of freedom have also been developed by the use of a high-viscosity BSA protein-based reagent. Examples of successful complex 3D microstructures include chains of Möbius strips, paddlewheels, and unconstrained probes for bacterial motility characterizing42, 80.

The advancements of neuroscience research have been driven by the progress of implantable neural micro-sensors. However, the geometries of those microsensors are limited by the available fabrication methods and materials. To resolve this critical issue, Venton et al. developed a novel IP-S/carbon system for fabricating free-standing carbon microelectrodes that were suitable for implantable neurochemical detection. As-proof-of- concept, micro-sensors with different geometric designs were fabricated and utilized for highly sensitive detection of dopamine. The electrodes showed a limit of detection for dopamine of 11±1 nM (sphere electrode) and 10±2 nM (cone electrode). Additionally, these micro-electrodes could also be applied for the detection of other neurochemicals, including ascorbic acid, serotonin, epinephrine, and norepinephrine.109

Graphene is an ideal material for sensors: Every atom in graphene is exposed to its environment allowing it to sense changes in its surroundings, featuring a promising prerequisite for micrometre-size sensors preparation capable of detecting individual events on a molecular level. Inspired by this, Yang et al. fabricated a venous-valve-inspired 3D micro-structure to enhance the graphene sensor performance for the detection of motile

Chapter 2 bacteria.110 Because of the asymmetry features of 3D printed IP-S structure, motile cells could readily migrate from outside towards the centre within the trap, but not vice versa – hence acting as a valve. Confirmed by fluorescence analysis, the micro-valve constructs could enrich cells inside the trap by a factor of up to 3.5 for motile bacteria and up to 5.4 for normal cells in culture, leading to an improved sensing performance.

2.4.7 Medical Devices

Ormocers-based TPP has been explored for the preparation of patient-specific medical prostheses. Current ossicular replacement prostheses prepared by conventional materials, including Teflon®, titanium and Ceravital®, are suffering from many limitations that prevent the practical uses. To improve accessibility of this platform, Ovsianikov et al. utilized TPP to fabricate ossicular replacement prostheses for sound restoration in the middle ear,95 see Figure 7a. The demonstrated middle-ear bone replacement prostheses exhibited acceptable cell viability and cell growth profiles. Similar to the design of commercially available ossicular replacements, conical structures were fabricated on the head of the prosthesis, which could reduce device migration, improve cell adhesion and decrease the likelihood of tympanic membrane perforation. The prosthesis was successfully inserted and removed from the implantation site of a frozen human head showing the capability of TPP for robust and patient-tailored fabrication of complex medical implants with a greater range of materials, sizes and shapes.

Chapter 2

Figure 7. (a) TPP fabrication of the microfluidic system containing porous tubular-shaped micro-capillaries. Reproduced with permission from ref.30; (b) Insertion of an Ormocer-based ossicular replacement prosthesis into an un-frozen human head. Reproduced with permission from ref.95

2.4.8 Bio-surface

TPP nanolithography based fabrication scheme enables the controlled preparation of tuned 3D structures that are suitable for the investigations of surface properties, i.e., wettability, biocompatible response and surface energy. Eberl et al. successfully demonstrated the rapid 3D prototyping of well-defined surface structures with a wide range of highly controllable wetting states by combining the TPP technique with plasma polymerization process.62 Inspired by the super hydrophobicity of the lotus leaf, TPP has been applied to control the wettability of materials and to prepare bio-inspired re-entrant surface structures for achieving liquid super repellence. In this work, biomimetic surfaces mimicking the morphology and size of the micrometric features of lotus leaf were achieved by cross linking of the IP resist. 63 To further explore TPP for lotus leaf-like repellence, Xu et al. developed super hydrophobic flexible films by the creation of well-defined 3D fractal and hierarchical microstructures (see Figure 7b).64

2.5 CHALLENGES AND PERSPECTIVE

As a powerful 3D designable nano/micro-prototyping approach, TPP nanolithography technique has experienced an exponential growth in the last few years. Up to now, several cross-disciplinary research hubs in Asia, Europe and America have engaged intensively with TPP fabrication and have sparked scientific collaborations between chemists, materials scientists, physicists, engineers, and biologists.

Compared with conventional miniaturization fabrication technologies, TPP holds many advantages for the direct fabrication of complex 3D nano/microstructures, including high resolution, single-step process, tuneable parameter setting and ample choice of candidate 38

Chapter 2 materials. TPP technique has dramatically lowered the barrier for creating sophisticated 3D structures and enabled true rapid 3D prototyping with attendant benefits. Because of these desirable properties and due to the advancements in material science, substantial progress has been achieved within the biological application field. Remarkable examples of TPP nanolithography prepared 3D structures for practical bio-implementations include microscaffolds for cell and tissue engineering, microneedles for drug delivery, microfluidics components for bio-fluidics control, sensors for biochemical analysis, micro-robots for therapy treatment, and prostheses for organ implants. However, several challenging issues should be addressed, particular from the perspective of materials, and research should be geared towards the development of soluble photo-initiators, structural design of new materials, performance improvement of current materials, and platform optimization of processing parameters.

In preparation, the provision of high-quality photosensitive materials with desirable properties is the prerequisite for high-performance TPP nanolithography process. Future research into suitable candidate materials for TPP fabrication will dramatically accelerate the advancement of this emerging 3D prototyping field. As we discussed in this review, to initiate polymerization reactions, it is essential to provide highly soluble photo-initiators with a large TPA cross-section, for the efficient generation of reactive radicals. Therefore, systematic investigations into the synthesis of novel photo-initiators would be a promising approach to further improve TPP direct writing technique.

On the other hand, most materials currently explored in TPP fabrication have been initially developed for conventional photolithographic approaches. This has led to a dearth in available TPP materials suitable for biological applications. Such stagnation of material discovery also brings tremendous opportunities for chemists to synthesize candidate materials by taking advantage of the versatile toolkit of organic chemistry. Considering the stringent requirements of bio-environments, further improvements to the material pool should be centred on the development of non-cytotoxicity biodegradable materials that enable the direct TPP fabrication of implantable bio-devices. Another material challenge is the shrinkage-induced deformation of the formed solid 3D structures occurred during development and washing step. To overcome these shrinkage issues, highly cross-linked

Chapter 2 photosensitive materials with neglectable volume changes during processing should be developed.

TPP process offers the possibility to create accurate 3D structures with high spatial resolution at the microscopic and nano-scopic scale. The fabrication of such 3D structures was a success for cell investigations and tissue engineering, where bioactive 3D scaffolds afforded to seamlessly mimic the native 3D cellular environments. In the future, selective modification of chemical-physical properties of the resist materials at submicron resolution might help to create 3D bio-active scaffolds with tunable functional sites that further support the study of the intricate mechanism involved in cell and tissue engineering. Another development trajectory of TPP techniques would be the exploration for novel type of biological implementations, covering neural stimulation/recording, internet of things (IoT) in healthcare and even the futuristic machine to human interface.

2.6 CONCLUSION

TPP nanolithography has matured in the past decade and become a pivotal technology to create 3D nano/micro-structures for bio-applications. As highlighted by the works reviewed, TPP fabricated 3D nano/micro-structures have taken on a central role for the preparation of biocompatible materials in application such as cell engineering, tissue engineering, drug delivery, bio-micro-fluidic device, bio-micro-robot and bio-medical devices. The purpose- built 3D nano/micro-structures were made from photon sensitive materials based on SU-8 resist, IP-series resist, PEG-based hydrogel materials, protein biomolecules and hybrid inorganic–organic materials (HIOM). Significant advancements were achieved in preparation of bio-compatible TPP materials that have further expanded the capabilities of TPP-generated 3D nano/micro-structures for practical bio-applications. For instance, 3D scaffold constructs fabricated from gelatin-based protein materials have been investigated for cell profile controlling in cell engineering.

Despite these significant achievements, the material odyssey for TPP nanolithography also faces several issues that should be addressed. The main concerns are the reliability, cost, 40

Chapter 2 reproducibility, and mass production capability of TPP fabricated 3D nano/micro-structures which remain to be thoroughly examined and improved. Undoubtedly, TPP-enabled 3D nano/micro-structures innovations have high potential to realize versatile uses in bioapplications and we look forward what the next decade will bring as the material pool further develops.

REFERENCE

1. N. Guo and M. C. Leu, Front. Mech. Eng., 2013, 8, 215-243. 2. R. F. Pease and S. Y. Chou, Proc. IEEE, 2008, 96, 248-270. 3. George M. Whitesides, Emanuele Ostuni, Shuichi Takayama, Xingyu Jiang and D. E. Ingber, Annu. Rev. Biomed. Eng., 2001, 3, 335-373. 4. B. -H. Jo, L.M. Van Lerberghe, K.M. Motsegood and D. J. Beebe, J. Microelectromech. Syst., 2000, 9, 76 - 81. 5. Busaina Dhariwala, Elaine Hunt and T. Boland, Tissue Eng. , 2004, 10, 1316-1322. 6. Shoji Maruo and K. Ikuta, Sens. Actuators A Phys., 2002, 100, 70–76. 7. Shoji Maruo, Osamu Nakamura and S. Kawata, Opt. Lett., 1997, 22, 132-134. 8. J.-M. Lourtioz, Nat. Mater., 2004, 3, 427-428. 9. S. Maruo and J. T. Fourkas, Laser Photon. Rev., 2008, 2, 100-111. 10. T. M. Hsieh, C. W. Ng, K. Narayanan, A. C. Wan and J. Y. Ying, Biomaterials, 2010, 31, 7648-7652. 11. S. D. Gittard and R. J. Narayan, Expert Rev. Med. Devices, 2010, 7, 343-356. 12. M. T. Raimondi, S. M. Eaton, M. M. Nava, M. Lagana, G. Cerullo and R. Osellame, J. Appl. Biomater. Funct. Mater., 2012, 10, 55-65. 13. J. K. Hohmann, M. Renner, E. H. Waller and G. von Freymann, Adv. Opt. Mater., 2015, 3, 1488-1507. 14. B. C. Gross, J. L. Erkal, S. Y. Lockwood, C. Chen and D. M. Spence, Anal. Chem., 2014, 86, 3240-3253. 15. Guang S. He, Loon-Seng Tan, Qingdong Zheng and P. N. Prasad, Chem. Rev., 2008, 108, 1245–1330. 16. W. Kaiser and C. G. B. Garrett, Phys. Rev. Lett., 1961, 7, 229-231. 17. T. Kogej, D. Beljonne, F. Meyers, J.W. Perry, S.R. Marder and J. L. Brédas, Chem. Phys. Lett., 1998, 298, 1-6. 41

Chapter 2

18. Guang S. He, C. Xu, Paras N. Prasad, Bruce A. Reinhardt, Jay C. Bhatt, Robert McKellar and A. G. Dillard, Opt. Lett., 1995, 20, 435-437. 19. H. K. Tsang, C. S. Wong, T. K. Liang, I. E. Day, S. W. Roberts, A. Harpin, J. Drake and M. Asghari, Appl. Phys. Lett., 2002, 80, 416-418. 20. C. Liao, W. Anderson, F. Antaw and M. Trau, ACS Omega, 2019, 4, 1401-1409. 21. S. M. Kuebler, K. L. Braun, W. Zhou, J. K. Cammack, T. Yu, C. K. Ober, S. R. Marder and J. W. Perry, J. Photochem. Photobiol. A, 2003, 158, 163-170. 22. S. M. Kuebler, M. Rumi, T. Watanabe, K. Braun, B. H. Cumpston, A. A. Heikal, L. L. Erskine, S. Thayumanavan, S. Barlow, S. R. Marder and J. W. Perry, J. Photopolym. Sci. Technol. , 2001, 14, 657-668. 23. B. H. Cumpston, S. P. Ananthavel, S. Barlow, D. L. Dyer, J. E. Ehrlich, L. L. Erskine, A. A. Heikal, S. M. Kuebler, I.-Y. Sandy Lee, D. McCord-Maughon, J. Q. Qin, H. Röckel, M. Rumi, X. L. Wu, S. R. Marder and J. W. Perry, Nature, 1999, 398, 51-54. 24. N. Pucher, A. Rosspeintner, V. Satzinger, V. Schmidt, G. Gescheidt, J. r. Stampfl and R. Liska, Macromolecules, 2009, 42, 6519-6528. 25. M. Albota, D. Beljonne, J. Brédas, J. E. Ehrlich, J. Y. Fu, A. A. Heikal, S. E. Hess, T.y Kogej, M. D. Levin, S. R. Marder, D. McCord-Maughon, J. W. Perry, H. Röckel, M. Rumi, G. Subramaniam, W. W. Webb, X. L. Wu and C. Xu, Science 1998, 281, 1653- 1656. 26. S. Das, A. Nag, D. Goswami and P. K. Bharadwaj, J. Am. Chem. Soc., 2006, 128, 402-403. 27. J. Xing, J. Liu, T. Zhang, L. Zhang, M. Zheng and X. Duan, J. Mater. Chem. B, 2014, 2, 4318-4323. 28. Y. Shi, W. H. Steier, L. Yu, M. Chen and L. R. Dalton, Appl. Phys. Lett., 1991, 58, 1131-1133. 29. E. Blasco, J. Muller, P. Muller, V. Trouillet, M. Schon, T. Scherer, C. Barner-Kowollik and M. Wegener, Adv. Mater., 2016, 28, 3592-3595. 30. A. Marino, O. Tricinci, M. Battaglini, C. Filippeschi, V. Mattoli, E. Sinibaldi and G. Ciofani, Small, 2018, 14, 1702959. 31. I. Stroud and P. C. Xirouchakis, Adv. Eng. Softw., 2000, 31, 83–95. 32. W. H. Teh, U. Dürig, G. Salis, R. Harbers, U. Drechsler, R. F. Mahrt, C. G. Smith and H. J. Güntherodt, Appl. Phys. Lett., 2004, 84, 4095-4097.

Chapter 2

33. A. Doraiswamy, C. Jin, R. J. Narayan, P. Mageswaran, P. Mente, R. Modi, R. Auyeung, D. B. Chrisey, A. Ovsianikov and B. Chichkov, Acta Biomater., 2006, 2, 267-275. 34. Wenhui Zhou, Stephen M. Kuebler, Kevin L. Braun, Tianyue Yu, J. Kevin Cammack, Christopher K. Ober, Joseph W. Perry and S. R. Marder, Science, 2002, 296, 1106- 1109. 35. R. J. Narayan, A. Doraiswamy, D. B. Chrisey and B. N. Chichkov, Mater. Today, 2010, 13, 42-48. 36. A. V. Do, K. S. Worthington, B. A. Tucker and A. K. Salem, Int. J. Pharm., 2018, 552, 217-224. 37. A. Tudor, C. Delaney, H. Zhang, A. J. Thompson, V. F. Curto, G.-Z. Yang, M. J. Higgins, D. Diamond and L. Florea, Mater. Today, 2018, DOI: 10.1016/j.mattod.2018.07.017. 38. D. W. Yee, M. D. Schulz, R. H. Grubbs and J. R. Greer, Adv. Mater., 2017, 29, 1605293. 39. A. Koroleva, A. A. Gill, I. Ortega, J. W. Haycock, S. Schlie, S. D. Gittard, B. N. Chichkov and F. Claeyssens, Biofabrication, 2012, 4, 025005. 40. Frederik Claeyssens, Erol A. Hasan, Arune Gaidukeviciute, Demetra S. Achilleos, Anthi Ranella, Carsten Reinhardt, Aleksandr Ovsianikov, Xiao Shizhou, Costas Fotakis, Maria Vamvakaki, Boris N. Chichkov and M. Farsari, Langmuir, 2009, 25, 3219–3223. 41. I. K. Kwon and T. Matsuda, Biomaterials, 2005, 26, 1675-1684. 42. B. Kaehr and J. B. Shear, Proc. Natl. Acad. Sci. USA, 2008, 105, 8850-8854. 43. J. F. Xing, M. L. Zheng and X. M. Duan, Chem. Soc. Rev., 2015, 44, 5031-5039. 44. A. I. Ciuciu and P. J. Cywiński, RSC Adv., 2014, 4, 45504-45516. 45. D. Wu, Q. D. Chen, L. G. Niu, J. N. Wang, J. Wang, R. Wang, H. Xia and H. B. Sun, Lab Chip, 2009, 9, 2391-2394. 46. Peter John Rodrigo, Lóránd Kelemen, Darwin Palima, Carlo Amadeo Alonzo, Pál Ormos and J. Glückstad, Opt. Express 2009, 17, 6578-6583. 47. S. Jariwala, B. Tan and K. Venkatakrishnan, J. Micromech. Microeng., 2009, 19, 115023. 48. M. Stoneman, M. Fox, C. Zeng and V. Raicu, Lab Chip, 2009, 9, 819-827. 49. G. Kumi, C. O. Yanez, K. D. Belfield and J. T. Fourkas, Lab Chip, 2010, 10, 1057- 1060.

Chapter 2

50. S. Kim, F. Qiu, S. Kim, A. Ghanbari, C. Moon, L. Zhang, B. J. Nelson and H. Choi, Adv. Mater., 2013, 25, 5863-5868. 51. T. Y. Huang, M. S. Sakar, A. Mao, A. J. Petruska, F. Qiu, X. B. Chen, S. Kennedy, D. Mooney and B. J. Nelson, Adv Mater, 2015, 27, 6644-6650. 52. J. Li., X. Li., T. Luo., R. Wang., C. Liu., S. C. D. Li., J. Yue., S.-h. Cheng. and D. Sun, Sci. Robot., 2018, 3, eaat8829. 53. M. R. Gullo, S. Takeuchi and O. Paul, Adv. Healthc. Mater., 2017, 6, 1601053. 54. H. Xu, M. Medina-Sanchez, V. Magdanz, L. Schwarz, F. Hebenstreit and O. G. Schmidt, ACS Nano, 2018, 12, 327-337. 55. A. Accardo, M. C. Blatche, R. Courson, I. Loubinoux, C. Thibault, L. Malaquin and C. Vieu, Small, 2017, 13, 1700621. 56. D. Serien and S. Takeuchi, ACS Biomater. Sci. Eng. , 2017, 3, 487-494. 57. J. K. Hohmann and G. von Freymann, Adv. Funct. Mater., 2014, 24, 6573-6580. 58. M. H. Olsen, G. M. Hjorto, M. Hansen, O. Met, I. M. Svane and N. B. Larsen, Lab. Chip, 2013, 13, 4800-4809. 59. I. A. Paun, R. C. Popescu, C. C. Mustaciosu, M. Zamfirescu, B. S. Calin, M. Mihailescu, M. Dinescu, A. Popescu, D. Chioibasu, M. Soproniy and C. R. Luculescu, Biofabrication, 2018, 10, 025009. 60. F. Qiu, S. Fujita, R. Mhanna, L. Zhang, B. R. Simona and B. J. Nelson, Adv. Funct. Mater., 2015, 25, 1666-1671. 61. X. Wang, C. Hu, L. Schurz, C. De Marco, X. Chen, S. Pane and B. J. Nelson, ACS Nano, 2018, DOI: 10.1021/acsnano.8b02907. 62. M. F. Berwind, A. Hashibon, A. Fromm, M. Gurr, F. Burmeister and C. Eberl, Microfluid. Nanofluidics., 2017, 21, 144. 63. A. D. Lantada, S. Hengsbach and K. Bade, Bioinspir. Biomim., 2017, 12, 066004. 64. Y. Lin, R. Zhou and J. Xu, Adv. Mater. Interf., 2018, DOI: 10.1002/admi.201801126, 1801126. 65. R. Di Giacomo, S. Krodel, B. Maresca, P. Benzoni, R. Rusconi, R. Stocker and C. Daraio, Sci. Rep., 2017, 7, 45897. 66. A. Aksit, D. N. Arteaga, M. Arriaga, X. Wang, H. Watanabe, K. E. Kasza, A. K. Lalwani and J. W. Kysar, Biomed. Microdevices., 2018, 20, 47. 67. A. I. Son, J. D. Opfermann, C. McCue, J. Ziobro, J. H. Abrahams, 3rd, K. Jones, P. D. Morton, S. Ishii, C. Oluigbo, A. Krieger, J. S. Liu, K. Hashimoto-Torii and M. Torii, Sci. Rep., 2017, 7, 17624.

Chapter 2

68. K. Arcaute, B. Mann and R. Wicker, Acta Biomater., 2010, 6, 1047-1054. 69. S. H. Lee, J. J. Moon and J. L. West, Biomaterials, 2008, 29, 2962-2968. 70. T. M. Seck, F. P. Melchels, J. Feijen and D. W. Grijpma, J. Control. Release, 2010, 148, 34-41. 71. F. Klein, B. Richter, T. Striebel, C. M. Franz, G. von Freymann, M. Wegener and M. Bastmeyer, Adv. Mater., 2011, 23, 1341-1345. 72. A. Ovsianikov, M. Gruene, M. Pflaum, L. Koch, F. Maiorana, M. Wilhelmi, A. Haverich and B. Chichkov, Biofabrication, 2010, 2, 014104. 73. A. Ovsianikov, M. Malinauskas, S. Schlie, B. Chichkov, S. Gittard, R. Narayan, M. Lobler, K. Sternberg, K. P. Schmitz and A. Haverich, Acta Biomater., 2011, 7, 967- 974. 74. A. Accardo, M.-C. Blatché, R. Courson, I. Loubinoux, C. Vieu and L. Malaquin, Biomed. Phys. Eng. Express., 2018, 4, 027009. 75. O. Kufelt, A. El-Tamer, C. Sehring, S. Schlie-Wolter and B. N. Chichkov, Biomacromolecules, 2014, 15, 650-659. 76. S. D. Gittard, A. Ovsianikov, H. Akar, B. Chichkov, N. A. Monteiro-Riviere, S. Stafslien, B. Chisholm, C. C. Shin, C. M. Shih, S. J. Lin, Y. Y. Su and R. J. Narayan, Adv. Eng. Mater., 2010, 12, B77-B82. 77. S. Engelhardt, E. Hoch, K. Borchers, W. Meyer, H. Kruger, G. E. Tovar and A. Gillner, Biofabrication, 2011, 3, 025003. 78. B. P. Chan, J. N. Ma, J. Y. Xu, C. W. Li, J. P. Cheng and S. H. Cheng, Adv. Funct. Mater., 2014, 24, 277-294. 79. Bryan Kaehr, Nusret Ertaş, Rex Nielson, Richard Allen, Ryan T. Hill, Matthew Plenert and J. B. Shear, Anal. Chem., 2006, 78, 3198–3202. 80. E. C. Spivey, E. T. Ritschdorff, J. L. Connell, C. A. McLennon, C. E. Schmidt and J. B. Shear, Adv. Funct. Mater., 2013, 23, 333-339. 81. Jonathan D. Pitts, Paul J. Campagnola, Gary A. Epling and S. L. Goodman, Macromolecules, 2000, 33, 1514–1523. 82. F. A. Pennacchio, C. Fedele, S. De Martino, S. Cavalli, R. Vecchione and P. A. Netti, ACS Appl. Mater. Interfaces, 2018, 10, 91-97. 83. A. Ovsianikov, A. Deiwick, S. Van Vlierberghe, P. Dubruel, L. Moller, G. Drager and B. Chichkov, Biomacromolecules, 2011, 12, 851-858. 84. A. Ovsianikov, A. Deiwick, S. Van Vlierberghe, M. Pflaum, M. Wilhelmi, P. Dubruel and B. Chichkov, Materials, 2011, 4, 288-299.

Chapter 2

85. F. A. Pennacchio, C. Casale, F. Urciuolo, G. Imparato, R. Vecchione and P. A. Netti, Biomater. Sci., 2018, 6, 2084-2091. 86. A. Bell, M. Kofron and V. Nistor, Biofabrication, 2015, 7, 035007. 87. M. Gebinoga, J. Katzmann, U. Fernekorn, J. Hampl, F. Weise, M. Klett, A. Läffert, T. A. Klar and A. Schober, Eng. Life Sci., 2013, 13, 368-375. 88. A. Koroleva, S. Gittard, S. Schlie, A. Deiwick, S. Jockenhoevel and B. Chichkov, Biofabrication, 2012, 4, 015001. 89. J. Serbin, A. Egbert, A. Ostendorf, B. N. Chichkov, R. Houbertz, G. Domann, J. Schulz, C. Cronauer, L. Fröhlich and M. Popall, Opt. Lett., 2003, 28, 301-303. 90. H. Hidai, H. Jeon, D. J. Hwang and C. P. Grigoropoulos, Biomed Microdevices, 2009, 11, 643-652. 91. H. Jeon, H. Hidai, D. J. Hwang and C. P. Grigoropoulos, J. Biomed. Mater. Res. B 2010, 93, 56-66. 92. T. WeiÃŸ, G. Hildebrand, R. Schade and K. Liefeith, Eng. Life Sci., 2009, 9, 384-390. 93. A. Ovsianikov, B. Chichkov, P. Mente, N. A. Monteiro-Riviere, A. Doraiswamy and R. J. Narayan, Int. J. Appl. Ceram. Tec, 2007, 4, 22–29. 94. S. D. Gittard, R. J. Narayan, C. Jin, A. Ovsianikov, B. N. Chichkov, N. A. Monteiro- Riviere, S. Stafslien and B. Chisholm, Biofabrication, 2009, 1, 041001. 95. A. Ovsianikov, B. Chichkov, O. Adunka, H. Pillsbury, A. Doraiswamy and R. J. Narayan, Appl. Surf. Sci., 2007, 253, 6603-6607. 96. A. Ovsianikov, S. Schlie, A. Ngezahayo, A. Haverich and B. N. Chichkov, J. Tissue. Eng. Regen. Med., 2007, 1, 443-449. 97. T. S. Drakakis, G. Papadakis, K. Sambani, G. Filippidis, S. Georgiou, E. Gizeli, C. Fotakis and M. Farsari, Appl. Phys. Lett., 2006, 89, 144108. 98. A. Marino, G. Ciofani, C. Filippeschi, M. Pellegrino, M. Pellegrini, P. Orsini, M. Pasqualetti, V. Mattoli and B. Mazzolai, ACS Appl. Mater. Interfaces, 2013, 5, 13012- 13021. 99. A. C. Scheiwe, S. C. Frank, T. J. Autenrieth, M. Bastmeyer and M. Wegener, Biomaterials, 2015, 44, 186-194. 100. A. Marino, C. Filippeschi, G. G. Genchi, V. Mattoli, B. Mazzolai and G. Ciofani, Acta. Biomater., 2014, 10, 4304-4313. 101. A. Marino, J. Barsotti, G. de Vito, C. Filippeschi, B. Mazzolai, V. Piazza, M. Labardi, V. Mattoli and G. Ciofani, ACS Appl. Mater. Interfaces, 2015, 7, 25574-25579.

Chapter 2

102. A. Ovsianikov, A. Ostendorf and B. N. Chichkov, Appl. Surf. Sci. , 2007, 253, 6599- 6602. 103. S. Jariwala, K. Venkatakrishnan and B. Tan, Opt. Express 2010, 18, 1630-1636. 104. M. Attilio, D. Andrea, P. Mario, P. Monica, F. Carlo, M. Barbara, M. Virgilio and C. Gianni, ACS Nano, 2014, 8, 11869-11882. 105. M. T. Raimondi, S. M. Eaton, M. Lagana, V. Aprile, M. M. Nava, G. Cerullo and R. Osellame, Acta Biomater., 2013, 9, 4579-4584. 106. P. S. Timashev, M. V. Vedunova, D. Guseva, E. Ponimaskin, A. Deiwick, T. A. Mishchenko, E. V. Mitroshina, A. V. Koroleva, A. S. Pimashkin, I. V. Mukhina, V. Y. Panchenko, B. N. Chichkov and V. N. Bagratashvili, Biomed. Phys. Eng. Express, 2016, 2, 035001. 107. C. Schizas, V. Melissinaki, A. Gaidukeviciute, C. Reinhardt, C. Ohrt, V. Dedoussis, B. N. Chichkov, C. Fotakis, M. Farsari and D. Karalekas, Int. J. Adv. Manuf. Tech., 2009, 48, 435-441. 108. J. M. Kim, J. J. Park, H. J. Lee, W. S. Kim, H. Muramatsu and S. M. Chang, Bioprocess. Biosyst. Eng., 2010, 33, 47-53. 109. C. Yang, Q. Cao, P. Puthongkham, S. T. Lee, M. Ganesana, N. V. Lavrik and B. J. Venton, Angew. Chem. , 2018, 57, 1-6. 110. B. Li, H. Tan, S. Anastasova, M. Power, F. Seichepine and G.-Z. Yang, Biosens. Bioelectron., 2018, DOI: 10.1016/j.bios.2018.09.087. 111. S. Schuerle., I. A. Vizcarra., J. Moeller., M. S. Sakar., B. Özkale., A. M. Lindo., F. Mushtaq., I. Schoen., S. Pané., V. Vogel. and B. J. Nelson, Sci. Robot., 2017, 2, eaah6094. 112. M. Medina-Sanchez, L. Schwarz, A. K. Meyer, F. Hebenstreit and O. G. Schmidt, Nano. Lett., 2016, 16, 555-561. 113. X. H. Qin, X. Wang, M. Rottmar, B. J. Nelson and K. Maniura-Weber, Adv. Mater., 2018, 30, 1705564. 114. B. Richter, V. Hahn, S. Bertels, T. K. Claus, M. Wegener, G. Delaittre, C. Barner- Kowollik and M. Bastmeyer, Adv. Mater., 2017, 29, 1604342. 115. A. M. Greiner, M. Jackel, A. C. Scheiwe, D. R. Stamow, T. J. Autenrieth, J. Lahann, C. M. Franz and M. Bastmeyer, Biomaterials, 2014, 35, 611-619. 116. P. Tayalia, C. R. Mendonca, T. Baldacchini, D. J. Mooney and E. Mazur, Adv. Mater., 2008, 20, 4494-4498.

Chapter 2

117. Shaun D. Gittard, Aleksandr Ovsianikov, Nancy A. Monteiro-Riviere, Jason Lusk, Pierre Morel, Paola Minghetti, Cristina Lenardi, Boris N. Chichkov and R. J. Narayan, J. Diabetes Sci. Technol., 2009, 3, 304-311. 118. S. D. Gittard, P. R. Miller, R. D. Boehm, A. Ovsianikov, B. N. Chichkov, J. Heiser, J. Gordon, N. A. Monteiro-Riviere and R. J. Narayan, Farad. Discuss., 2011, 149, 171- 185. 119. Z. Xiong, X.-Z. Dong, W.-Q. Chen and X.-M. Duan, Appl. Phys. A, 2008, 93, 447-452. 120. S. Maruo, K. Ikuta and H. Korogi, J. Microelectromech. Syst., 2003, 12, 533-539. 121. R. A. Barry, R. F. Shepherd, J. N. Hanson, R. G. Nuzzo, P. Wiltzius and J. A. Lewis, Adv. Mater., 2009, 21, 2407-2410. 122. E. Kapyla, T. Sedlacik, D. B. Aydogan, J. Viitanen, F. Rypacek and M. Kellomaki, Mater. Sci. Eng. C Mater. Biol. Appl., 2014, 43, 280-289. 123. N. E. Fedorovich, J. Alblas, W. E. Hennink, F. C. Oner and W. J. Dhert, Trends Biotechnol., 2011, 29, 601-606. 124. F. P. W. Melchels, M. A. N. Domingos, T. J. Klein, J. Malda, P. J. Bartolo and D. W. Hutmacher, Prog. Polym. Sci., 2012, 37, 1079-1104. 125. T. Billiet, M. Vandenhaute, J. Schelfhout, S. Van Vlierberghe and P. Dubruel, Biomaterials, 2012, 33, 6020-6041. 126. W. Y. Yeong, C. K. Chua, K. F. Leong and M. Chandrasekaran, Trends Biotechnol., 2004, 22, 643-652. 127. E. K. Sackmann, A. L. Fulton and D. J. Beebe, Nature, 2014, 507, 181-189. 128. J. Cooper McDonald and G. M. Whitesides, Acc. Chem. Res., 2002, 35, 491–499. 129. C. M. Ho, S. H. Ng, K. H. Li and Y. J. Yoon, Lab Chip, 2015, 15, 3627-3637. 130. K. Venkatakrishnan, S. Jariwala and B. Tan, Opt. Express 2009, 17, 2756-2762. 131. M. Medina-Sánchez, V. Magdanz, M. Guix, V. M. Fomin and O. G. Schmidt, Adv. Funct. Mater., 2018, 28, 1707228. 132. M. J. Villangca, D. Palima, A. R. Bañas and J. Glückstad, Light Sci. Appl., 2016, 5, e16148. 133. A. V. Singh, Z. Hosseinidoust, B. W. Park, O. Yasa and M. Sitti, ACS Nano, 2017, 11, 9759-9769. 134. Veronika Magdanz, Samuel Sanchez and O. G. Schmidt, Adv. Mater. , 2013, 25, 6581–6588. 135. S. D. Gittard, A. Ovsianikov, B. N. Chichkov, A. Doraiswamy and R. J. Narayan, Expert Opin. Drug Deliv., 2010, 7, 513-533.

Chapter 3

Candidate’s contribution to the authorship

I declared that I have obtained permission from all co-authors to include following publication directly into this chapter.

Caizhi Liao, Will Anderson, Fiach Antaw, Matt Trau. “Maskless 3d ablation of precise micro-hole structures in plastics using femtosecond laser pulses”. ACS Appl. Mater. Interfaces, 2018, 10, 4, 4315-4323. - Incorporated as Chapter 3.

Contributor Statement of contribution

Conception and design (70%) Experiments (95%) Caizhi Liao (Candidate) Analysis and interpretation (70%) Drafting and production (60%)

Conception and design (10%) Will Anderson Analysis and interpretation (10%) Drafting and production (20%) Conception and design (10%) Experiments (5%) Fiach Antaw Analysis and interpretation (10%) Drafting and production (10%) Conception and design (10%) Analysis and interpretation (10%) Matt Trau Drafting and production (10%)

Chapter 3

Maskless 3D Ablation of Precise Microhole Structures in Plastics Using Femtosecond Laser Pulses

Summary Resistive-pulse sensing (RPS) has triggered attention as a robust platform for single molecular/nanoparticle analysis. One pivotal component of a RPS system is the miniaturized pore structure used for sensing, the preparation of which is being studied intensively. For typical RPS pore structures, a suitable substrate with through hole structure is required to load the sensing pore construct: the vital element to be realized for Milestone 1. Promisingly, two photon femtosecond (fs) laser can be a robust tool for the fabrication of micro-hole structures. In this work, we introduce a brand-new 3 dimensional (3D) ablation technique to generate accurate micro-hole structures in plastic substrates. The results show that a two photon fs laser based ablation scheme can create arbitrary shaped micro-well structures in plastics through a rapid single-step ablation process without the need for any masks. These processed plastic substrates could be the essential building components for the 3D miniaturized constructs/pores to be explored in Milestone 2 to Milestone 4.

Chapter 3 is based on a published paper:

 Caizhi Liao, Will Anderson, Fiach Antaw, and Matt Trau. “Maskless 3D Ablation of Precise Microhole Structures in Plastics Using Femtosecond Laser Pulses” ACS Appl. Mater. Interfaces 2018, 10, 4315−4323.

Chapter 3

ABSTRACT

Femtosecond laser ablation is a robust tool for the fabrication of micro-hole structures. This technique has several advantages compared to other micro-fabrication strategies for reliably preparing micro-hole structures of high quality and low cost. However, few studies have explored the use of femtosecond laser ablation in plastic materials, due to the lack of controllability over the fabrication process in plastics. In particular, the depth profile of micro- hole structures prepared by conventional laser ablation techniques in plastics cannot be precisely and reproducibly controlled. In this paper, a novel, three-dimensional (3D), femtosecond laser ablation technique was developed, for the rapid fabrication of precise micro-hole structures in multiple plastics in air. Using a three-step fabrication scheme, micro- holes demonstrated extremely clean and sharp geometric features. This new technique also enables the precise creation of arbitrary shaped micro-well structures in plastic substrates through a rapid single-step ablation process, without the need for any masks. As a proof-of- concept for practical application, precise micro-hole structures prepared by this novel femtosecond laser ablation technique were exploited for robust resistive-pulse sensing of micro-particles.

3.1 INTRODUCTION

Precise fabrication of micro-hole structures has a broad range of applications, - from aero- engine turbines and automotive fuel filters, to surgical/biomedical devices and biochemical microfluidic sensing systems. To obtain micro-hole structures, a myriad of machining techniques have been developed in the past few years, such as mechanical ultrasonic drilling, electrochemical drilling (ECD), electrical-discharge machining (EDM) and integrated photolithographic-chemical wet etching1. However, these multi-step micro-hole fabrication techniques are usually time/labour consuming, inflexible, expensive, and can be environmentally hazardous. In comparison to aforementioned approaches, laser ablation methods have emerged as a much more effective and environmentally friendly technique for the preparation of precise micro-hole structures2-5.

Chapter 3

Since the mid-1970s, pulsed solid-state lasers have proven to be an efficient ablation tool, which has several advantages over other micro-fabrication techniques for reliably preparing micro-hole structures of high quality and low cost6. This has lead to the development of ultrafast femtosecond laser pulsing that has enabled direct ablation and removal of target substrate with minimal disturbance of the peripheral materials. This control allows the fabrication of precise structures with micrometer to sub-micrometer resolution2, 7-8. As applied to micro-hole fabrication, the femtosecond laser ablation process can be divided into two distinct schemes: percussion drilling and trepanning. In percussion drilling, the laser pulse, shaped to a particular sized voxel, is focused on the desired location for a required time to ablate the material with a geometric size that is identical to the size of the voxel9. For trepanation processing, the laser pulses or the target sample is moved in a circular pathway to the desired diameter of micro-hole10. Percussion drilling is more applicable for the fabrication of smaller nano/micro-holes, while trepanning can be employed to create circular micro-hole structures of analysis. To manipulate the size and depth of circular micro-holes, integrated optimization of laser parameters11(e.g., pulse energy, pulse duration and pulse repetition rate) and system operation conditions12-14 (e.g., beam modification, laser pathway and assistant-gas environment) have been widely explored.

Although femtosecond laser ablation techniques enable micro-hole fabrication with minimized horizontal peripheral heating effects through optimisation of laser parameter, controlling the hole depth profile in the vertical direction is difficult. Currently, most femtosecond laser created hole structures are funnel-shaped, with resolidified debris anchored on hole sidewalls, and adherent materials at the hole entrance, because of the plasma constraints imposed on the laser ablation process15-16. The degree of the taper effect encountered in direct laser ablation is typically dependent on the thickness of the target materials and is caused by the energy loss of pulses reflected on the sides of the hole wall9, 17. These constraints significantly prohibit the use of femtosecond laser ablation in the fabrication of high-quality micro-holes.

Currently, glass, ceramic and metal films are the most suitable materials for conventional femtosecond laser ablation18-20. Unfortunately for conventional femtosecond laser ablation processes, plastic materials are easily thermally decomposed, which leads to the formation of soot that is hard to remove along the sidewalls of micro-holes21-22. Consequently, no studies have systematically explored the use of femtosecond laser pulses in precise

Chapter 3 fabrication of micro-holes in commonly used plastic materials, such as polyethylene terephthalate (PET), polycarbonate (PC) and polyethylene (PE). Due to their low-cost, lightweight, flexibility, manipulability, and a multitude of other desirable properties, plastic materials have attracted increasing attention in the micro-fabrication field. The ability to form accurate micro-hole structures within plastic materials would enable remarkable developments in many fields, such as biomedical devices and flexible electronics23-26. Therefore, micromachining technique capable of precise fabrication of micro-holes in varied plastic materials is greatly demanded.

To overcome the problem of creating precise micro-holes in plastic substrates, we hereby propose a new methodology for 3-dimensional (3D) femtosecond laser ablation, capable of open-air operation. The geometric features of micro-holes can be accurately pre-defined using the computer-aided design (CAD) models of the micro-holes, which are then ablated from the target material. Different from that of conventional percussion and trepanning femtosecond laser ablation, virtually any designable 3D micro-hole shapes can be directly achieved by this new technique without the use of masks. In this study, we use optimized ultrafast laser pulses to ablate and vaporize the illuminated volume with a negligible heat affected zone (HAZ) of the peripheral material in plastic substrates, leading to the formation of through micro-hole and non-through micro-well structures with extremely sharp and clean geometric features. We demonstrate a three-step 3D direct laser ablation process to create a wide range of perfectly shaped through-cylindrical hole structures (Φ 10µm to Φ 200µm) in many commonly used plastic materials, including PET, medium-density polyethylene (MDPE), poly(methyl methacrylate) (PMMA), polystyrene (PS), polyphenyl ether (PPE) and PC. To accurately prepare these precise micro-hole structures, laser parameters (pulse energy, scanning speed), material aspects (substrate thickness, substrate type) and micro- hole features (hole size, geometric shape), were systematically investigated and are thoroughly discussed. Additionally, this new ablation technique enables the precise fabrication of arbitrary shaped non-through micro-well structures in a rapid single-step ablation process, without the need for any masks. Finally, as a proof-of-concept, we used precisely fabricated micro-hole structures formed in plastic substrates for resistive-pulse sensing of micro-particles, a technique that plays a crucial role particularly in the fields of biomedical engineering and pharmaceutical development.

Chapter 3

3.2 EXPERIMENTAL METHODS

3.2.1 Micro-hole Fabrication in Plastics

Micro-hole structures with distinct geometric features (hole diameter, hole depth and hole shape) were designed in the Autodesk AutoCAD program. Depending on specific processing conditions, the diameter and the depth of sketched micro-hole structures ranged from Φ 10 µm to Φ 200 µm and 150 µm to 400 µm, respectively. These AutoCAD generated micro-hole STL files were imported into the DeScribe software (Nanoscribe GmbH, Germany, Figure S1) for slicing (0.3 µm) and hatching (0.4 µm) process. The smallest voxel was set to 0.3 µm in width and 0.4 µm in length (Figure 1, right down), which guarantees sub- micrometer resolution in micro-holes fabrication. The processed General Writing Language (GWL) file was then transferred to the Nanoscribe Photonic Professional GT system (Nanoscribe GmbH) for ablation of micro-hole structures.

Figure 1. Schematic diagram of 3D femtosecond laser ablation process. The plastic material was mounted on a 3D XYZ piezo stage. The ablated plastic material is directly transformed into the gaseous plasma and removed into surrounding environment. The upper right figure shows the Gaussian intensity profile of the used Ti: sapphire femtosecond laser pulse (80 MHz, 100 fs). The lower right figure shows a single ablation unit (voxel): 0.3 µm in diameter

(dxy), 0.4 µm in height (lz). 54

Chapter 3

Before laser ablation, the surfaces of plastic substrates (both sides) were thoroughly cleaned by ultrasonic washing (10 mins), followed by a nitrogen drying treatment. The cleaned plastic substrates were mounted on the Nanoscribe-controlled XYZ piezo stage (Figure S1). The 3D ablation process was performed using the laser pulses generated by a mode-locked Ti: sapphire femtosecond laser system (780 nm central wavelength, 80 MHz repetition rate and 100 fs pulse duration) (Figure 1). The output from the femtosecond laser source had a Gaussian intensity profile that tightly focused within the target material through a 25X objective lens (0.8 NA). To precisely prepare through cylindrical micro-holes, a three-step processing scheme (two FAST ablation steps plus one SLOW polish step) was performed, and the processing details are described in Figure S2. The FAST ablation step (Coding file 1 in the Supporting Information) was used to robustly remove the majority of the plastic material in the micro-hole and to define the surface morphology of the prepared micro-hole exits on both sides. While the SLOW ablation step (Coding file 2 in the Supporting Information) was used for the eradication of debris attached to the hole sidewalls, to improve the surface quality of the inner hole wall. For PET substrate material, the optimized laser pulse scanning speed was set at 70 k µm/s in the FAST ablation mode and 8 k µm/s in the SLOW ablation mode. Corresponding suitable scanning speeds of the fast and slow modes for other plastic materials are listed in the Supporting Information (Table S1). For the preparation of complex non-through micro-well structures (triangle, hexagon and donut) with a designed hole depth of 50 µm, a single FAST ablation process was used to create these structures in plastic substrate materials without the aid of masks. No post-treatment process was required for the micro-holes/micro-wells prepared by this new 3D femtosecond laser ablation process.

3.2.2 Characterization A JEOL IT-300 scanning electron microscope (SEM) was used for the morphological characterization of micro-hole structures. The surfaces of micro-hole structures analyzed by SEM were pre-coated with 20 nm thick platinum film by a JEOL coating system. A Nikon ECLIPSE Ni-U optical system equipped with a Nikon Plan Fluor 20x/0.5 DLL len was employed for the optical imaging of micro-hole/micro-well structures. A DEKTAK stylus profiler was utilized for the measurement of sidewall roughness of micro-hole structures. For the resistive-pulse sensing of 20 µm particles (COULTER cc Standard L20), a qNano

Chapter 3 system (Izon Science, New Zealand) was utilized with measurements performed at a 0.5 V potential. Buffer solutions used in the upper and lower reservoir chambers were diluted phosphate buffered saline (PBS, 0.01X) electrolyte.

3.3 RESULTS AND DISCUSSIONS

3.3.1 Multiphoton Ablation Process

Femtosecond laser pulse micromachining tools utilise high-order, multi-photon non-linear absorption processes to ablate target material27-28. To optimise our 3D ablation technique, it is essential to understand the fundamental aspects of this technology. Ultrashort femtosecond laser pulses (<200 fs), usually with power intensities in excess of 1012 W/cm2, are capable of electrostatically breaking the chemical bonds holding the target material together via the multi-photon ionization (MPI) process. The MPI process directly transforms the target plastic materials into gaseous plasma2, 29. As schematically shown in Figure S2, free electrons in the target material absorb laser pulse energy via ion collisions. Subsequently, these free electron energy carriers transfer the energy from the laser pulses to the target materials via collisions with bonded electrons, resulting in the destabilization of intermolecular bonds in the bulk material. Accordingly, the target material volume irradiated with laser pulses is directly ablated off in a vaporized plasma form30. In long-pulse laser ablation, the absorbed energy leaves the laser focal voxel through heat conduction into the peripheral material, within a time period of 10-13 s31. However in femtosecond laser pulse ablation, the rapid generation of plasma is achieved within an extremely short time period (10-15 s), thus prohibiting thermal heating of the peripheral material31-32. Therefore, when optimised there is a significant reduction of laser-induced thermal diffusion in the bulk target material around the laser voxel, allowing precise micro-fabrication in plastic materials

Chapter 3

Figure 2. (a) Z-direction cross-section view of the ablated 3D through cylindrical micro-hole with illustrative slicing pathway (slicing distance: 0.3µm). (b) XY-direction cross-section view of the ablated 3D through cylindrical micro-hole with illustrative hatching pathway (hatching distance: 0.4µm). (c) Simulated layer-by-layer 3D direct laser ablation process. The cylinder part is removed by laser ablation process. (d) Simulated and experimental SEM image of the laser ablation process in a particular sliced interfacial layer, respectively. The red arrows indicate the movement direction of the laser voxel along the X-Y axes.

It is also important to understand that in our experiments, the target material is irradiated with laser pulses with in a defined 3D pathway, to produce accurate 3D geometry. Figure 2. provides a general overview of the 3D direct laser ablation pathways for a through cylindrical micro-hole structure produce using our new laser ablation process. Two combined structural processing parameters, slicing and hatching, are used to tune the scanning patterns of the laser pulse. In general, material mounted on the XYZ piezo stage of the Nanoscribe is directly ablated off in a layer-by-layer slicing process (Figure 2a). The interfacial morphology of the hatching process of the laser pulse beams scanning on a particular sliced layer of the ablated micro-hole structure is demonstrated in Figure 2d, in which the white line in the simulated sliced layer represents the laser ablation frontline where the plastic material ablation processes occurs, and is shown in the experimental SEM image. To achieve optimal laser ablation pathways for precise micro-hole preparation, a combination of optimized laser parameters for the 3D laser ablation process, including pulse energy, pulse duration, pulse scanning speed and pulse repetition frequency, must be ascertained11, 17, 19. Because the pulse duration (100 fs) and the pulse repetition frequency (80 MHz) are fixed, we therefore systematically investigated the effects of remaining factors (pulse energy and pulse scanning speed) on the quality of fabricated micro-holes. Geometric designs and materials aspects that significantly affect the laser ablation process are also investigated.

3.3.2 Fluence Power Level in Plastics Ablation

Fluence (F, J·cm-2), defined as laser pulse energy delivered per unit area, serves as one of the most significant factors in determining the geometric quality of laser ablated holes19, 33. Figure 3 sequentially depicts the SEM topographic images of fabricated through-cylindrical 57

Chapter 3 micro-hole (Φ 100 µm) in a 7 Mil thick PET plastic substrate, with varied fluence levels (0.05 J·cm-2 to 0.8 J·cm-2) and scanning speeds (10k μm/s to 90k μm/s). The shown results clearly indicate the interdependent effects of these two variables on the surface quality of the ablated micro-holes. For the preparation of a through Φ 100 µm cylindrical micro-hole in 7 Mil thick PET substrate, the optimal threshold fluence was around 0.4 J·cm-2, operating with a scanning speed at 70 k μm/s (Red-framed image setting). As shown in Figure 3, with low fluence level laser pulses (<0.4 J·cm-2), re-solidification of melt components after ablation led to uncontrolled geometric changes to the shape of micro-hole structures. At a low fluence level, below the threshold level (0.4 J·cm-2 for PET material), the formation of plasma fume is prohibited, leading to uncontrolled ablation in micro-holes.

Figure 3. The effects of laser pulse energy level (fluence) and laser pulse scanning speed on the quality of ablated Φ 100 µm through-cylindrical micro-holes in a 7 Mil (177.8 µm) thick PET substrate. The fluence level was varied from 0.05 J·cm-2 to 0.8 J·cm-2 and the scanning speed was increased from 10k μm/s to 90k μm/s. The red rectangular-framed SEM image indicated the optimal laser operation conditions (fluence: 0.4 J·cm-2, scanning speed: 70 k µm/s) for the precise fabrication of a Φ 100 µm through-cylindrical micro-hole in a 7 Mil thick PET substrate.

Chapter 3

As the fluence level was adjusted slightly above the ablation threshold level, the localized laser pulse heats the target intensely to transform the materials directly from solid phase into vapor plasma phase with high kinetic energy (well above vaporization temperature), creating micro-hole structures with extremely sharp geometric features (Red-framed image in Figure 3). When the pulse energy was further increased (>0.5 J·cm-2 for PET plastic), mechanical stress imposed on micro-hole edges increased correspondingly because of the high plasma pressure, which eventually yielded uncontrollable processing results (Figure S3), such as deformation, cracking and fracturing34. Additionally, melting effects gradually dominate the ablation process with high fluence laser pulses, resulting in explosions of material that produces a mixture of liquid and vapor drops35. Therefore, plastic micro-holes fabricated with high fluence level laser pulse ablation have large heat affected zones, and demonstrate unpredictable features due to recast materials forming along the micro-hole profile.

3.3.3 Scanning Speed in Plastics Ablation The scanning speed of the laser pulses plays an essential role in micro-holes quality. Figure 3 clearly depicts the effects of scanning speed on the geometric profile of micro-holes. Under the condition of optimized fluence (0.4 J·cm-2), the most suitable scanning speed for micro- hole ablation in PET plastic was found to be at 70 k μm/s. For a particular laser pulse fluence level, an increment in scanning speed can dramatically facilitate the removal of ablated material and enhance the precision of the laser ablation process36. At the optimal scanning speed (70 k μm/s for PET plastic), gaseous ablation products in the plasma plume can be effortlessly vaporized from target surface into the surrounding environment. As the laser scanning speeds increase above the optimal level, uncontrolled ablations in ablated micro- holes begin to dominate. Due to the high scanning speeds (>70 k μm/s), the formulated target plasma cannot be fully removed in time, creating a high density of plasma trapped in ablated area. The rapidly increased plasma density will reduce the input energy of the laser pulse because of pulse reflection as described in introduction, and thus prohibit the material vaporization process induced by multiphoton ionization9. In summary, both laser fluence level and scanning speed can significantly impact the quality of micro-holes prepared by this 3D laser ablation technique.

Chapter 3

Figure 4. (a) SEM characterization a Φ 100 µm through-cylindrical micro-hole in 7 Mil thick PET substrate (Top-view). (b) Optical image of a Φ 100 µm through-cylindrical micro-hole in 7 Mil thick PET substrate (Top-view). (c) Cross-section SEM image of a Φ 100 µm through-cylindrical micro-hole. (d) Interfacial SEM characterization of the micro-hole sidewall surface.

3.3.4 Geometric Design in Laser Ablation

In direct laser ablation, the laser pulse becomes less effective at ablating target material as the hole depth increases. The incoming laser pulse will be restricted, both by the absorption of ablation plasma contained in the hole, and the reflection of the laser pulse at the hole sidewall. This creates micro-hole structures with a funnel-like shape (Taper effect)15. Taper effects become more pronounced as hole depth increases9, 16. Although several advanced optical methods have been developed to tailor the incoming laser beams to produce holes with desired depth profiles16, the geometric features of micro-holes generated by conventional femtosecond laser ablation techniques cannot be reproducibly formed, particularly for micro-hole structures prepared using plastic substrates7. Figure 4 demonstrates a precise Φ 100 µm through cylindrical micro-hole prepared by a three-step ablation scheme (Figure S2). The micro-hole has sharp edges and a clean sidewall, without

Chapter 3 any observable non-defined bulges that would indicate the re-solidification of heat-induced melting debris that form in classical laser ablation processes. Figure 4c demonstrates the cross-section view of a through cylindrical micro-hole. Taper effects that would cause funnel- like hole shapes as seen in standard laser ablation techniques were almost eradicated. Lateral ablation zones that would degrade the surrounding surface texture were also not observed. The hole sidewall has precisely defined features, with an averaged surface roughness around 360nm±50nm (Figure 4d). In some cases with high fluence laser pulses, micro-cracks would appear on hole sidewalls (Figure S3). The formation of plastically deformed micro-cracks inside the micro-hole was mainly attributed to the tensile stresses developed during the cooling process. Further work is required to investigate the formation and mitigation of these cracks.

Table 1. Entrance and exit geometric features of the through cylindrical micro-holes with different diameter sizes (Φ 10µm to Φ 200µm) formed in a 7 Mil thick PET plastic substrate.

Cylinder Hole Diameter 10 30 50 100 150 200 (Designed)/µm Front-side Hole Diameter 9.3 29.6 49.8 99.8 149.8 199.5 (Experimental, Mean)/µm

Front-side Hole [ -11.2%, [ -2.5%, [ -1.1%, [ -0.8%, [ -1.0%, [ -1.2%, Deviation (%) +1.5% ] +0.9% ] +0.6% ] +0.3% ] +0.5% ] +0.8% ] Backside Hole Diameter 9.4 29.8 49.7 100.1 150.2 199.8 (Experimental, Mean)/µm

Backside Hole [-10.9%, [ -1.6%, [ -1.2%, [ -0.5%, [ -0.9%, [ -1.1%, Deviation (%) +1.3%] +1.1% ] +0.4% ] +0.5% ] +0.4% ] +0.5% ]

Chapter 3

One unique aspect of this laser ablation technique is that the 3D geometric features of micro- holes, including hole diameter, hole shape and hole depth, can be accurately pre-defined by CAD software, instead of by tuning laser parameters and operating conditions as required in conventional femtosecond laser ablation methods10. Table 1 lists the geometric features of different-sized micro-holes (Φ 10 µm to Φ 200 µm) in PET substrate. Only micro-holes below Φ 10 µm deviated largely from the pre-designed size. This deviation was typically less than 10% of the hole diameter and can be attributed to the fact that smaller holes impede the plasma transportation process from moving ablated material from the target surface into the surrounding environment. As the hole diameter increased above 10 µm, these dimension-induced constraints were dramatically reduced. As shown in Table 1, the ablated holes on both sides have near-perfect circularity with a deviation below ±1%, which demonstrates the potential of this new laser ablation technique for the precise fabrication of micro-holes in plastic substrates.

It is important to note that for larger holes ablation, higher laser power is required. To fully remove the target plastic material in desired micro-holes, a laser pulse with sufficient power (near threshold fluence level) is required to generate adequate multiphoton-absorption, to induce the ionization reactions that ablate the material11, 19. In Figure 5a, the blue line demonstrates the relationship between the ablated areas (~D2) and the corresponding minimum required threshold fluence level. For optimized micro-hole fabrication, a higher level of threshold fluence was required as micro-hole size was increased. This can be attributed to the significantly increased amount of target plastic materials that needed to be ablated away, an effect which was even more apparent when the diameter of micro-holes was increased above 100 µm. Additionally, the size of ablated areas also determines the overall ablation processing time (Figure 5a, red line, FAST step). The overall processing time increased substantially as the micro-hole diameters increased from 10 µm (2.2 s) to 200 µm (158.2 s) , as shown in Table S2. The increased ablated areas exponentially extend the movement length of the laser voxel along the hatching pathway in each sliced layer, thus correspondingly increase the processing time for larger holes ablation.

Chapter 3

3.3.5 Material Aspects in Laser Ablation

Another essential parameter affecting the laser ablation process in plastics was the thickness of the substrate materials37. Figure 5b illustrates the minimum required threshold fluence level for the optimal preparation of through cylindrical micro-holes in PET plastic substrates with different thicknesses. With a fixed optimal laser pulse scanning speed at 70 k µm/s (FAST step) in the three-step ablation process (Figure S2), the laser fluence level was tuned to effectively ablate Φ 100 µm through cylindrical micro-holes in PET plastic substrates with thicknesses ranging from 5 Mil to 15 Mil. For thin plastic substrates (< 10 Mil), a relatively low threshold fluence level (~0.4 J·cm-2) was adequate to fully transform the solid plastic material directly into gaseous plasma. As the thickness of the ablated plastic material increased (> 10 Mil), the threshold fluence level required to fully remove the material had to be increased because of the increasing hole depth effect. As discussed previously, the increased hole depth will increase the energy reflection phenomenon of the laser pulses, which consequently leads to the reduction of laser pulse energy exerted on the ablated areas and prohibits the laser ablation process9, 17. As a result, to compensate for the reduced laser pulses energy in a thicker plastic substrate, a higher threshold fluence level is required.

The proposed new 3D laser ablation method was shown to be universally robust for the preparation of precise micro-hole structures in multiple other plastic materials, including PE, PMMA, PS, PC and PPE. As shown in Figure 5c, the optical band-gap energy of ablated 2 plastic material (Eg) was found to have a good linear relationship (R =0.993) to the optimal fluence level (Fth) for the ionization process induced by multi-photon absorption. For the femtosecond laser ablation process, multi-photon ionization is dominated by the generation of free electrons. Simultaneous nonlinear absorption of m photons when mhv ≥ Eg, where hv is the energy of a single photon and Eg is the optical band-gap energy of the ablated material, will lift the bonded electrons from the ground energy level to the free energy level28 (Figure S4). The low band-gap polymer material PPE-series typically has a band gap energy between 2 and 3 eV, which facilitates the multi-photon ionization process and the formation of free electrons (plasma) even under low input laser energy (~0.3 J·cm-2). As the optical band-gap energy of ablated material gradually expanded, the minimum required threshold fluence level to effectively ablate the plastic material increased linearly. To fabricate a precisely controlled Φ 100 µm through-cylindrical micro-hole in a PMMA (Eg ~ 5 63

Chapter 3 eV) plastic substrate, a threshold fluence level around 0.55 J·cm-2 was required to initiate multi-photon ionization, which was nearly two-fold larger than that of micro-holes fabrication using PPE-series plastic material. Promisingly, this result provides new insights into the optimal fluence level for the laser ablation process using different plastics.

Figure 5. (a) The interdependent relationship between ablated area (~D2) and threshold fluence level (blue line), ablated area (~D2) and overall processing time (red line), correspondingly. (b) The interdependent relationship between plastic substrate thickness and threshold fluence level for optimized laser processing. (c) Linear relationship between the optical band-gap energy of plastic materials and optimized threshold fluence level for precise micro-holes fabrication. (d) Complex micro-well structures prepared in PET substrates: triangular-shaped, hexagon-shaped and donut-shaped micro-wells with a measured depth (mean) around 48±3 µm.

3.3.6 Maskless 3D Ablation of Micro-wells

Importantly, this proposed 3D ablation technique serves as a new paradigm for the maskless preparation of precise micro-well structures. Non-through hole micro-wells hold great potential for emerging application scenarios, such as flexible electronics, tissue engineering, cell engineering and pharmaceutical analysis. To prepare non-circular-shaped micro-well structures by conventional femtosecond laser ablation approaches, masks defining the to- 64

Chapter 3 be-ablated shapes must be integrated into the laser ablation systems before processing the structures, which significantly increases the complexity of micro-hole fabrication process10, 38. Furthermore, laser pulses passing through these masks cannot be uniformly projected onto the surface of the ablated material, leading to poor controllability over the ablation process. In contrast, the ablation scheme investigated here enables the precise preparation of any desirable-shaped micro-well structures in a straightforward manner without using masks. Desired non-through micro-wells with distinct geometric features can be obtained in seconds via a single-step laser ablation process (Fast step, Figure S2). Figure 5d displays the images of the micro-wells in the shapes of a triangular, hexagon and donut formulated in a PET substrate. The geometric features were accurately controlled, with negligible geometric deviations to the CAD designs. Additionally, the depth profile of these micro-holes is highly manipulable. The averaged depth of these micro-well structures measured by a stylus profiler was around 48±3 µm, while the theoretical depth of CAD designs were 50 µm. Therefore, this new 3D femtosecond laser ablation technique affords the rapid, maskless, precise fabrication of complex-shaped micro-well structures in plastic materials.

3.3.7 Resistive-pulse Analysis Using Ablated Micro-hole

The development of nano/micro-pores for colloidal characterisation has resulted in a renaissance of “Coulter-counter” based techniques in recent years39. This has been driven by new techniques that enable the fabrication of nano/micro-sized pores structures, which when used for resistive-pulse sensing (RPS) are powerful sensors that enable high throughput particle-by-particle characterization of colloid suspensions of individual bio- macromolecules40. To validate the performance of the micro-hole structures prepared by this new 3D laser ablation technique, a through cylindrical Φ 100 µm micro-hole in a PET substrate was utilized for resistive-pulse analysis of 20 µm particles (Figure 6). As shown in Figure 6a, a voltage (0.5 V) was applied across the plastic substrate containing the micro- hole, and the ionic current flowing through the micro-pore was recorded by the Izon qNano particle analysis system. As the particle passes through the pore, there was a measurable increase in resistance caused by the occlusion of electrolyte ions, which in turn leads to a transient decrease in ionic current (Current blockade, Figure 6b). For a cylindrical micro- hole, the induced resistance change and ionic current change across the length of micro- hole is given by41:

Chapter 3

R d 3  (1) R D2l RR/ i(%) (2) 1/RR where d is the particle diameter, D is the micro-holes diameter (Φ 100 µm) and l is the micro- holes depth (177.8 µm). As shown in the above Eqns, the magnitude of ionic current changes (~0.4 nA) obtained from experimental results allows the calculation of the mean diameter of an analyzed micro-particle from the magnitude of its resistive pulse. The mean diameter of the characterized standard 20µm latex particle was calculated to be 21±0.35 µm, corresponding to the manufacture size of 20± 0.15 µm.

Chapter 3

Figure 6. (a) Schematic diagram of the resistive-pulse particle analysis system. (b) Single current blockade induced by a micro-particle passing through the prepared micro-hole structure. (c) Current trace of the resistive-pulse micro-particle analysis. The signal noise is below 30 pA, with a signal/noise ratio (S/N) larger than 10. (d) Size histogram of the current blockade distribution experimentally obtained by q-Nano system. (e) The in-house tool simulated size distribution of 20µm micro-particles. Simulation addressing voltage: 0.5 V.

The current blockades measured by the qNano system are displayed as Figure 6d. Furthermore, an in-house tool for calibration-free resistive pulse analysis was used to determine the particle size distribution42. As presented in Figure 6e, over 85 % of the micro- particles were located in the size range of 19 µm to 22 µm, with an averaged diameter size around 20.8 µm, which further validated the experimentally calculated result mentioned above (21±0.35 µm). Promisingly, robust resistive-pulse analysis of micro-particles exemplifies the versatile uses of the plastic substrates containing precise micro-hole structure fabricated by this new 3D ablation technique.

3.4 CONCLUSION

In conclusion, we have developed a new 3D femtosecond laser ablation technique, which serves as a robust ablation tool for the precise fabrication of micro-hole structures in multitude plastic materials. The dominating parameters (energy level and scanning speed) of femtosecond laser pulse ablation were systematically investigated, to accurately create micro-holes/micro-wells with extremely sharp and clean geometric features. A wide range of precisely controlled through cylindrical micro-hole structures (Φ 10µm to Φ 200µm) in multiple plastic materials, including PET, MDPE, PMMA, PS, PPE and PC, were successfully realized using our proposed three-step 3D laser ablation process. The results indicated that the optimized laser processing conditions were affected by the target plastic material properties (substrate thickness, substrate type) and pre-designed geometric features (hole size, geometric shape). More importantly, arbitrary-shaped non-through micro-well structures could be rapidly prepared using a single step ablation process without 67

Chapter 3 using any masks. Lastly, to demonstrate the practical applications of this new 3D laser ablation technique, robust resistive-pulse analysis of micro-particles using precisely ablated through-cylindrical micro-hole was performed. These findings provide a novel solution to several practical problems faced by precise macro-machining of plastic materials.

REFERENCES

(1) Judy, J. W. Microelectromechanical systems (MEMS) fabrication, design and applications. Smart Mater. Struct. 2001, 10, 1115–1134. (2) Perry, M. D.; Stuart, B. C.; Banks, P. S.; Feit, M. D.; Yanovsky, V.; Rubenchik, A. M. Ultrashort-pulse Laser Machining of Dielectric Materials. J. Appl. Phys. 1999, 85, 6803-6810. (3) Tirlapur, U. K.; König, K. Cell Biology Targeted Transfection by Femtosecond Laser. Nature 2002, 418, 290-291. (4) Xiong, W.; Liu, Y.; Jiang, L. J.; Zhou, Y. S.; Li, D. W.; Jiang, L.; Silvain, J. F.; Lu, Y. F. Laser-Directed Assembly of Aligned Carbon Nanotubes in Three Dimensions for Multifunctional Device Fabrication. Adv. Mater. 2016, 28, 2002-2009. (5) Rajput, D.; Costa, L.; Lansford, K.; Terekhov, A.; Hofmeister, W. Solution-cast High- aspect-ratio Polymer Structures from Direct-write Templates. ACS Appl. Mater. Interfaces 2013, 5, 1-5. (6) Nolte, S.; Momma, C.; Kamlage, G.; Ostendorf, A.; Fallnich, C.; Alvensleben, F. v.; Welling, H. Machining of Sub-micron Holes Using a Femtosecond Laser at 800 nm. Opt. Commun. 1995, 114, 106-110. (7) Li, Y.; Itoh, K.; Watanabe, W.; Yamada, K.; Kuroda, D.; Nishii, J.; Y.Y, J. Three- dimensional Hole Drilling of Silica Glass From The Rear Surface with Femtosecond Laser Pulses. Opt. Lett. 2001, 26, 1912-1914. (8) Diddams, S. A.; Hollberg, L.; Mbele, V. Molecular Fingerprinting with The Resolved Modes of A Femtosecond Laser Frequency Comb. Nature 2007, 445, 627-230. (9) Li, L.; Low, D. K. Y.; Ghoreshi, M.; Crookall, J. R. Hole Taper Characterisation and Control in Laser Percussion Drilling. CIRP Ann. Manuf. Technol. 2002, 51, 153-156.

Chapter 3

(10) Nikumb, S.; Chen, Q.; Li, C.; Reshef, H.; Zheng, H. Y.; Qiu, H.; Low, D. Precision Glass Machining, Drilling and Profile Cutting by Short Pulse Lasers. Thin Solid Films 2005, 477, 216-221. (11) Zhu, X.; Naumov, A. Y.; Villeneuve, D. M.; Corkum, P. B. Influence of Laser Parameters and Material Properties on Micro Drilling with Femtosecond Laser Pulses. Appl. Phys. A 1999, 69, 367–371. (12) Nolte, S.; Momma, C.; Kamlage, G.; Ostendorf, A.; Fallnich, C.; Alvensleben, F. v.; Welling, H. Polarization Effects in Ultrashort-pulse Laser Drilling. Appl. Phys. A 1999, 68, 563–567. (13) Ashkenasi, D.; Stoian, R.; Rosenfeld, A. Single and Multiple Ultrashort Laser Pulse Ablation Threshold of Al2O3 (corundum) at Different Etch Phases. Appl. Surf. Sci. 2000, 154-155, 40-46. (14) Hwang, D. J.; Choi, T. Y.; Grigoropoulos, C. P. Liquid-assisted Femtosecond Laser Drilling of Straight and Three-dimensional Microchannels in Glass. Appl. Phys. A 2004, 79, 605-612. (15) Zoubir, A.; Shah, L.; Richardson, K.; Richardson, M. Practical Uses of Femtosecond Laser Micro-materials Processing. Appl. Phys. A 2003, 77, 311-315. (16) Ghoreishi, M.; Low, D. K. Y.; Li, L. Comparative Statistical Analysis of Hole Taper and Circularity in Laser Percussion Drilling. Int. J. Mach. Tool Manufact. 2002, 42, 985–995. (17) Jiang, L.; Fang, J.; Cao, Q.; Zhang, K.; Wang, P.; Yu, Y.; Huang, Q.; Lu, Y. Femtosecond Laser High-efficiency Drilling of High-aspect-ratio Microholes Based on Free- electron-density Adjustments. Appl. Opt. 2014, 53, 7290-7295. (18) Zhang, J.; Sugioka, K.; Midorikawa, K. High-speed Machining of Glass Materials by Laser-induced Plasma-assisted Ablation Using a 532-nm Laser. Appl. Phys. A 1998, 67, 499–501. (19) Feng, Q.; Picard, Y. N.; Liu, H.; Yalisove, S. M.; Mourou, G.; Pollock, T. M. Femtosecond Laser Micromachining of a Single-crystal Superalloy. Scr. Mater. 2005, 53, 511-516. (20) Banks, P. S.; Feit, M. D.; Rubenchik, A. M.; Stuart, B. C.; Perry, M. D. Material Effects in Ultra-short Pulse Laser Drilling of Metals. Appl. Phys. A 1999, 69, 377–380. (21) Bharatish, A.; Narasimha Murthy, H. N.; Anand, B.; Madhusoodana, C. D.; Praveena, G. S.; Krishna, M. Characterization of Hole Circularity and Heat Affected Zone in Pulsed CO2 Laser Drilling of Alumina Ceramics. Opt. Laser Technol. 2013, 53, 22-32.

Chapter 3

(22) Le Harzic, R.; Huot, N.; Audouard, E.; Jonin, C.; Laporte, P.; Valette, S.; Fraczkiewicz, A.; Fortunier, R. Comparison of Heat-affected Zones Due to Nanosecond and Femtosecond Laser Pulses using Transmission Electronic Microscopy. Appl. Phys. Lett. 2002, 80, 3886- 3888. (23) Wolfe, D. B.; Ashcom, J. B.; Hwang, J. C.; Schaffer, C. B.; Mazur, E.; Whitesides, G. M. Customization of Poly(dimethylsiloxane) Stamps by Micromachining Using a Femtosecond-Pulsed Laser. Adv. Mater. 2003, 15, 62–65. (24) Baudach, S.; Bonse, J.; Krüger, J.; Kautek, W. Ultrashort Pulse Laser Ablation of Polycarbonate and Polymethylmethacrylate. Appl. Surf. Sci. 2000, 154-155, 555–560. (25) Yong, J.; Chen, F.; Yang, Q.; Du, G.; Bian, H.; Zhang, D.; Si, J.; Yun, F.; Hou, X. Rapid Fabrication of Large-area Concave Microlens Arrays on PDMS by a Femtosecond Laser. ACS Appl. Mater. Interfaces 2013, 5, 9382-9385. (26) Zheng, C.; Hu, A.; Kihm, K. D.; Ma, Q.; Li, R.; Chen, T.; Duley, W. W. Femtosecond Laser Fabrication of Cavity Microball Lens (CMBL) inside a PMMA Substrate for Super-Wide Angle Imaging. Small 2015, 11, 3007-3016. (27) Liu, X.; Du, D.; Mourou, G. Laser Ablation and Micromachining with Ultrashort Laser Pulses. IEEE J. Quant. Electron. 1997, 33, 1706-1716. (28) Gamaly, E. G.; Rode, A. V.; Tikhonchuk, A. V.; Luther-Davies, B. Electrostatic Mechanism of Ablation by Femtosecond Lasers. Appl. Surf. Sci. 2002, 197-198, 699–704. (29) Muth-Böhm, J.; Becker, A.; Faisal, F. H. M. Suppressed Molecular Ionization for a Class of Diatomics in Intense Femtosecond Laser Fields. Phys. Rev. Lett. 2000, 85, 2280- 2283. (30) Fujimoto, J. G.; Liu, J. M.; Ippen, E. P.; Bloembergen, N. Femtosecond Laser Interaction with Metallic Tungsten and Nonequilibrium Electron and Lattice Temperatures. Phys. Rev. Lett. 1984, 53, 1837-1840. (31) Luft, A.; Franz, U.; Emsermann, L.; Kaspar, J. A Study of Thermal and Mechanical Effects on Materials Induced by Pulsed Laser Drilling. Appl. Phys. A. 1996, 63, 93–101. (32) Paek, U.-C.; Gagliano, F. Thermal Analysis of Laser Drilling Processes. IEEE J. Quant. Electron. 1972, 8, 112 - 119. (33) Chen, X.; Lotshaw, W. T.; Ortiz, A. L.; Staver, P. R.; Erikson, C. E.; McLaughlin, M. H.; Rockstroh, T. J. Laser Drilling of Advanced Materials: Effects of Peak Power, Pulse Format, and Wavelength. J. Laser. Appl. 1996, 8, 233-239. (34) Mannion, P. T.; Magee, J.; Coyne, E.; O’Connor, G. M.; Glynn, T. J. The Effect of Damage Accumulation Behaviour on Ablation Thresholds and Damage Morphology in

Chapter 3

Ultrafast Laser Micro-machining of Common Metals in Air. Appl. Surf. Sci. 2004, 233, 275- 287. (35) Ganesh, R. K.; Faghri, A. A Generalized Thermal Modeling for Laser Drilling Process—I. Mathematical Modeling and Numerical Methodology. Int. J. Heat Mass Transf. 1997, 40, 3351-3360. (36) Sundaram, S. K.; Mazur, E. Inducing and Probing Non-thermal Transitions in Semiconductors using Femtosecond Laser Pulses. Nat. Mater. 2002, 1, 217-224. (37) Kim, J.; Na, S. Metal Thin Film Ablation with Femtosecond Pulsed Laser. Opt. Laser Technol. 2007, 39, 1443-1448. (38) Momma, C.; Nolte, S.; Chichkov, B. N.; Alvensleben, F. v.; Tunnermann, A. Precise Laser Ablation With Ultrashort Pulses. Appl. Surf. Sci. 1997, 109–110, 15-19. (39) Kozak, D.; Anderson, W.; Vogel, R.; Trau, M. Advances in Resistive Pulse Sensors: Devices bridging the void between molecular and microscopic detection. Nano Today 2011, 6, 531-545. (40) Anderson, W.; Lane, R.; Korbie, D.; Trau, M. Observations of Tunable Resistive Pulse Sensing for Exosome Analysis: Improving System Sensitivity and Stability. Langmuir 2015, 31, 6577-6587. (41) Anderson, W.; Kozak, D.; Coleman, V. A.; Jamting, A. K.; Trau, M. A Comparative Study of Submicron Particle Sizing Platforms: Accuracy, Precision and Resolution Analysis of Polydisperse Particle Size distributions. J. Colloid. Interface Sci. 2013, 405, 322-330. (42) Darby Kozak; Will Anderson; Robert Vogel; Shaun Chen; Fiach Antaw; Trau, M. Simultaneous Size and ζ-Potential Measurements of Individual Nanoparticles in Dispersion Using Size-Tunable Pore Sensors. ACS Nano 2012, 6, 6990–6997.

Chapter 3

SUPPLEMENTAL INFORMATION

Femtosecond Laser Ablation System

As shown below in the schematic diagram of laser ablation system, the laser pulse used for ablation process is generated by a mode-locked Ti: sapphire femtosecond laser system, with the parameter settings at 780 nm central wavelength, 80 MHz repetition rate and 100 fs pulse duration. The movement of piezoelectric 3D stage is determined by a STL file containing the geometric designs of micro-holes. The slicing and hatching processes were processed by the NanoWrite software (Nanoscribe GmbH.). Other important ablation parameters, including initial ablation direction, ablation speed, ablation power and surface define coefficient, were all settled in the NanoWrite software. CCD camera component in the system allows the real-time observation of laser ablation process.

Figure S1. Schematic diagram of the Ti: sapphire femtosecond laser ablation system (Nanoscribe), majorly consist of femtosecond laser source, CCD camera and piezoelectric 3D stage.

Chapter 3

Three-step Ablation Process for Through Cylindrical Micro-holes

To prepare precisely-controlled through cylindrical micro-hole structures in plastic substrates, a three-step direct laser ablation scheme was proposed as shown below. Firstly, the substrate was processed with the FAST ablation step, then flipped over and aligned to the right position, to perform the second FAST ablation step. Finally, a SLOW ablation step was carried out to polish the surface of hole sidewall and improve the quality of hole geometric features.

Figure S2. Schematic diagram of the three-step direct laser ablation process of a Φ 100 μm through cylindrical micro-hole on the 7 Mil (177.8 μm) thick PET substrate. The required processing times for the FAST ablation step and SLOW ablation step were ~3s and ~45s, respectively.

Chapter 3

Scanning Speeds for Different Plastics

To ablate Φ 100µm micro-hole in different types of plastic substrates (with the same thickness), the scanning speed needs to be optimized individually for the FAST step and SLOW step. Because of property difference in each type of plastic substrates, the most suitable scanning speed for ablation process is slightly differentiated. In the experiments to obtain the optimal scanning speed for each type of plastic substrate, the scanning speed is set as the sole variable parameter, all other ablation processing parameters were fixed at the same value. The ablation tests using different of plastic materials suggest that the optimal scanning speed for the FAST scanning step is typically larger than 65 k µm/s, while the scanning speed for the SLOW scanning step is usually below 8 k µm/s. The scanning speed plays a crucial role in removing the gaseous plasma generated during the femtosecond laser ablation process. Therefore, to effectively fabricate the precisely- controlled micro-hole structures in plastic materials, the scanning speed for the FAST step and SLOW step should be optimized as below.

Table S1. Optimal scanning speeds for the creation of a Φ 100µm through cylindrical micro- holes on different types of plastic substrates.

Materials PPE PC PE PET PS PMMA

Optical Image

FAST Step Scanning 70 k 65 k 70 k 70 k 75 k 90 k Speed µm/s SLOW Step 5 k 7 k 7 k 8 k 8 k 8 k Scanning Speed µm/s

Chapter 3

Micro-cracks at High Energy Level Laser Ablation

These high levels fluence ablation induced micro-cracks mostly appear in the middle sites along the micro-holes depth profile, which could be attributed to the unevenly distributed mechanical tension. It is rare to observe these micro-cracks near the two-exit positions of micro-hole. The surface mechanical tension force along the hole-wall increase as the laser pulse ablate deeper, leading to a higher possibility of the formation of micro-cracks. For a through Φ 100µm micro-hole in 7 Mil thick PET substrate, the generated micro-cracks could extend over 80 µm in length, and 3 µm to 4 µm in width.

Figure S3. Micro-cracks appear on the hole sidewalls, under the conditions of high fluence level laser ablation.

Chapter 3

Multiphoton-absorption Induced Ionization

For the ultrashort pulses (femtosecond laser) with extremely high field strength, multiphoton absorption induced ionization process can directly generate high density of free electrons, without relying on levels of seed electrons as required in the classic ablation process using long pulses laser. Normally, during the process of femtosecond laser ablation of plastic materials, the ablated material will be transformed directly into gaseous plasma. Subsequent nonlinear absorption of laser energy leads to the irreversible damage to the target material. To free bonded electrons, absorbed laser pulse energy should exceed binding energy of the target material (potential well shown below).

Figure S4. Multiphoton-absorption induced ionization process occurred in the plastic materials. Nonlinear absorption of m photons induced by the laser pulse lifts the bonded electron from the bound energy level to the free energy level.

Chapter 3

Table S2. Processing times for the femtosecond laser direct ablation process of varied micro-hole structures with differentiated diameter sizes (Φ 10µm to Φ 200µm) assembled in the 7 Mil thick PET substrate.

Cylinder Diameter /µm 10 30 50 100 150 200

SEM of Ablated Micro-

holes

FAST Step Processing Time (Experimental, 0.5 1.1 1.8 3.1 4.2 6.5 Mean)/s SLOW Step Processing Time (Experimental, 1.2 8.2 21.3 46.3 90.3 145.2 Mean)/s Overall Processing Time 2.2 10.4 24.9 52.5 98.7 158.2 (Experimental, Mean)/s

Chapter 3

Micro-particle Resistive-pulse Sensing Setup

Before resistive-pulse measurement, the edges of the PET substrate containing a laser ablated Φ 100µm through micro-hole was packaged with polyimide tape, aiming to alleviate the short-circuit issue occurs during the measurement process. The packaged micro-hole sample was then transferred into the deionized water for 10 mins ultrasonic cleaning and was dried with the nitrogen gas. To start the resistive-pulse particle sensing, the pre-treated micro-hole structure was mounted on the Izon qNano instrument as shown below. The lower chamber of measurement unit was filled with 40 µL diluted phosphate buffered saline (PBS, 0.01X) solution, while the upper chamber was filled with 80 µL 0.01X PBS solution. The voltage parameter was set to be 0.5V. When the system was stabilized (normally with noise ≤ 20pA), the software began to record the signal as particles translocated through the micro- hole.

Chapter 3

Coding file 1: FAST ablation step

% Bounding box % HatchLines: Alternate % Minimum X: -50 Y: -50 Z: 0 % ZAxis: Piezo % Maximum X: 50 Y: 50 Z: 180 % Exposure: Constant % % WritingDirection: Down % Transformation % ScanMode: Galvo % Scaling X: 1 Y: 1 Z: 1 % Translation X: -50 Y: -50 Z: 180 % Rotation X: -0.707 Y: 0 Z: 0 W: 0.707 % Writing configuration % GalvoScanMode % Slicing ContinuousMode % SlicingMode: Fixed PiezoSettlingTime 10 % Distance: 0.4 GalvoAcceleration 2 % SimplificationTolerance: 0.01 %StageVelocity 200 % Enable for more accurate but slower stitching % FixSelfIntersections: on

% % Scan field offsets % Filling XOffset 0 % ContourCount: 2 YOffset 0 % ContourDistance: 0.2 ZOffset 0 % ConcaveCornerMode: Beveled

% HatchingDistance: 0.3 % Writing parameters % HatchingAngle: 30 PowerScaling 1 % HatchingAngleOffset: 0 LaserPower 40 % ScanSpeed 70000 % Output options

Chapter 3

Coding file 2: SLOW ablation step

% Bounding box % Minimum X: -50 Y: -50 Z: 0 % Output options % Maximum X: 50 Y: 50 Z: 180 % HatchLines: Alternate % % ZAxis: Piezo % Transformation % Exposure: Constant % Scaling X: 1 Y: 1 Z: 1 % WritingDirection: Down % Translation X: -50 Y: -50 Z: 180 % ScanMode: Galvo % Rotation X: -0.707 Y: 0 Z: 0 W: 0.707 % % Slicing % Writing configuration % SlicingMode: Fixed GalvoScanMode % Distance: 0.4 ContinuousMode % SimplificationTolerance: 0.01 PiezoSettlingTime 10 % FixSelfIntersections: on GalvoAcceleration 2 % %StageVelocity 200 % Enable for more accurate but slower stitching % Filling

% ContourCount: 2 % Scan field offsets % ContourDistance: 0.2 XOffset 0 % ConcaveCornerMode: Beveled YOffset 0 % HatchingDistance: 0.3 ZOffset 0 % HatchingAngle: 30

% HatchingAngleOffset: 0 % Writing parameters % PowerScaling 1

LaserPower 40

ScanSpeed 8000

Chapter 4

Candidate’s contribution to the authorship

I declared that I have obtained permission from all co-authors to include following publication directly into this chapter.

Caizhi Liao, Will Anderson, Fiach Antaw, Matt Trau. “Two-Photon Nanolithography of Tailored Hollow three-dimensional Microdevices for Biosystems”. ACS Omega, 2019, 4, 1, 1401-1409. - Incorporated as Chapter 4.

Contributor Statement of contribution

Conception and design (80%) Experiments (95%) Caizhi Liao (Candidate) Analysis and interpretation (70%) Drafting and production (60%)

Conception and design (5%) Will Anderson Analysis and interpretation (10%) Drafting and production (20%) Conception and design (5%) Experiments (5%) Fiach Antaw Analysis and interpretation (10%) Drafting and production (10%) Conception and design (10%) Analysis and interpretation (10%) Matt Trau Drafting and production (10%)

Chapter 4

Two-Photon Nanolithography of Tailored Hollow Three Dimensional Microdevices for Biosystems

Summary After achieving Milestone 1, we were able to provide plastic substrates with tailored micro- holes, which supported the fabrication of complete 3D hollow micro-devices, as proposed by Milestone 2. To achieve this aim, two-photon polymerization (TPP) based nanolithography, a powerful tool for nano/micro-fabrication, has been investigated for accurate preparation of hollow type 3D miniaturized structures. In this Chapter, we systematically investigated the formation process of 3D hollow micro-structures using the TPP nanolithography system for the first time, and successfully established a well-studied fabrication platform for different types of hollow 3D constructs. In particular, hollow-type micro-hole structures were demonstrated for robust RPS sensing analysis of individual nanoparticles. The realization of Milestone 2 helped to shape the development of TPP- prepared resistive-pulse sensing (RPS) pores that are to be discussed in Chapter 5 and Chapter 6 (Milestone 3 and Milestone 4).

Chapter 4 is based on a published paper:

Caizhi Liao, Will Anderson, Fiach Antaw, and Matt Trau. “Two-Photon Nanolithography of Tailored Hollow Three-dimensional Microdevices for Biosystems”. ACS Omega 2019, 4, 1, 1401-1409.

Chapter 4

ABSTRACT

Functional three-dimensional (3D) microstructures incorporating accessible interiors have emerged as a versatile platform for biosystem applications. By configuring their 3D geometry features, these biosystem micro-devices can accurately evaluate and control targeted bio- environments. However, classical fabrication techniques based on photolithography-etching processes cannot precisely and programmably control the geometric of the entire hollow 3D microstructures. Here, we presented a new additive fabrication strategy based on direct laser writing (DLW) via two-photon polymerization (TPP), for the precise, straightforward, and customizable preparation of hollow 3D microstructure devices. Factors governing the formation of hollow 3D biosystem micro-devices, including material composition, laser input and (post-) development treatment, were systematically investigated. To evaluate the broad applicability of this approach, a series of tailored hollow 3D micro-devices, including a micropore, microneedle, microelectrode, microvalve and micromachine, were successfully prepared using our DLW-TPP technique. To validate the feasibility of these biosystem micro- devices in practical implementations, we demonstrated the use of hollow 3D micropore devices for the robust resistive-pulse analysis of nanoparticles.

4.1 INTRODUCTION Biosystem micro-devices are microstructures used for the analysis or handling of biological compounds, and their development has both enhanced performance and increased throughput in modern biological research and applications1-4. With recent advances in nano/micro-fabrication technology having led to the implementation of multiscale, hollow, three-dimensional (3D) microstructure devices with accessible interiors, these benefits are stronger than ever5-6. Hollow 3D microstructures engineered for both electrical and chemical functions have shown great potential for creating well-controlled micro-devices that tackle the issues encountered in intricate biosystems, including biochemical sensing7-8, pharmaceutical delivery9, neural interfacing10, biofluid manipulation11, and biomedical therapy11.

To date, fabrication strategies utilized for hollow 3D microstructures have been dominated by photolithography-etching processes12-13. However, due to the harsh processing conditions involved, candidate materials suitable for these methods are

Chapter 4 mostly limited to silicon-based inorganic substrates. Furthermore, these fabrication strategies can only generate limited geometries for hollow microstructures, resulting in a strong research focus on shapes like micropores and microneedles5. Given the geometric constraints in chemical etching processes, it is currently not possible to control the geometric features of hollow structures in all three dimensions14-15. Further geometric control of hollow 3D microstructure can be achieved using enhanced lithographic techniques, such as electron-beam lithographies for micro-pores and nanosphere lithographies for scaffolds16-17. Nonetheless, these modified lithographic methods are still not able to precisely control the full 3D geometry of hollow microstructures, yielding pseudo-3D structures with a limited range of geometries and functions12. Thus, there is a real desire to develop fabrication technologies that can accurately produce complex hollow 3D microstructures of any desired geometry.

Direct laser writing (DLW) via two-photon polymerization (TPP) is emerging as a powerful alternative to photolithographic methods for the rapid 3D fabrication of microstructures18-20. TPP reactions can be spatio-temporally controlled, a capability that provides the possibility of reliably fabricating tailored hollow 3D microstructures with nanoscale resolution21-22. During TPP process, photoresist polymerises to generate insoluble polymeric networks, and by programming the laser focal point to move in 3D space, 3D microstructures can be formed23-24. It is important to note that for TPP to create hollow 3D microstructures, un- polymerized soluble photoresist trapped inside the structure must be thoroughly removed through opening holes leading into the inner hollow space25. To our knowledge, no existing work has specifically explored the use of the DLW-TPP technique for the fabrication of tailored hollow 3D biosystem micro-devices. Therefore, factors that ultimately determine TPP’s viability for the fabrication of hollow 3D biosystem micro-device structures, such as the broad range of processing settings, including both material issues (resist type, mechanical property and biocompatibility, etc.,) and fabrication aspects (slicer parameters, laser power, development condition, etc.,), remain unassessed.

Here, we propose the use of DLW-TPP technique for the precise preparation of tailored hollow 3D microstructure devices with accessible interiors. In this work, factors that govern the ultimate structural quality, including slicer parameters, laser input, post-development 84

Chapter 4 treatment and material choice, were comprehensively investigated and are thoroughly discussed. To demonstrate the ubiquitous applicability of this DLW-TPP based fabrication technique, we created a series of model hollow 3D microstructure devices that are representative of a broad variety of biosystem applications, such as micropores for nanoparticle sensing, microneedles for pharmaceutical delivery, microelectrodes for neural recording/stimulation, microvalves for biofluid controlling and micromachines for biomedical therapy. Furthermore, to validate that this technique can create practical hollow 3D microstructures for biosystem utilizations, we fabricated hollow micropore devices with geometries tailored for resistive-pulse sensing of nanoparticles. Due to the high resolution of this fabrication process, these micropore devices demonstrate robust analytical performance, providing calibration free nanoparticle size measurements. Generally, these investigations suggest that the DLW-TPP technique serves as a promising fabrication platform for tailored hollow 3D biosystem micro-devices.

4.2 EXPERIMENTAL SECTION

4.2.1 Materials

Substrates for the loading of hollow 3D microstructure devices, including an indium tin oxide (ITO)-coated PET film (protective film on both sides, Shenzhen Jemstone Technology Co., Ltd.) and ITO-coated glass (Sigma-Aldrich Co.), were used as received. Milli-Q (Merck

Millihole) deionized (DI) H2O was used for the phosphate-buffered saline (PBS) solution preparation, nanoparticle dilution, and ultrasonic treatment medium. A PBS tablet (Sigma- Aldrich Co.) was used for the preparation of PBS solution, dissolved with DI water. Resist materials for hollow 3D microstructure fabrication, including the IP-series resist (IP-L, IP-S, and IP-Dip, Nanoscribe GmbH), ORMOCER (Micro Resist Technology GmbH) and SU-8 (MicroChem Corp.), were used as received. Different types of polystyrene (PS) nanoparticles functionalized with carboxyl group, with a diameter size range from 114 to 530 nm, were purchased from Bangs Laboratories, Inc. Raw PS nanoparticle samples were diluted with 1× PBS solution, with a dilution ratio of 1:1000 to 1:10 000, to prepare suitable nanoparticle solutions for resistive-pulse analysis.

Chapter 4

4.2.2 Direct Laser Writing

Externally accessible hollow 3D microstructures with built-in opening holes, including micropores, microneedles, microelectrodes, microvalves, and micromachines, were designed in an Autodesk AutoCAD program. These AutoCAD-generated standard triangle language (.STL) files were imported into the DeScribe software (Nanoscribe GmbH, Germany) for structural processing (slicing and hatching). This processed general writing language (.GWL) file was then transferred to the Nanoscribe Photonic Professional GT system (Nanoscribe GmbH, Germany.) for the DLW of hollow 3D microstructures. This TPP- enabled DLW fabrication system was equipped with a FemtoFiber Pro NIR laser source (Toptica Photonics AG) operating with a pulse duration (τ) ≈ 100 fs and a repetition rate (f) of 80 MHz at 780 nm (λ). Transmittance (T) of laser beam through the IP-S resist at 780 nm was ∼0.75. The laser pulse was focused by a Zeiss Plan-Apochromat 25× 0.8 NA oil DIC M27 objective, with circular polarization at the entrance of the objective plane. With the aid of an integrated charge-coupled device camera (Figure 1a), the DLW-TPP process of hollow 3D microstructures can be visualized in real time.

In the experiment, substrates coated with resist materials were mounted onto the Nanoscribe-controlled 3D XYZ piezo stage for the DLW-TPP process. For the hollow 3D microstructures fabricated with the IP-series resists (IP-L, IP-S, and IP-Dip) and ORMOCER, the liquid resist materials (∼100 μL) were directly dropped onto the loading substrate to initiate the DLW-TPP process. For the microfabrication using SU-8, prebaking (60 °C, 60 s) was required before laser irradiation. After DLW process, the SU-8-based structure was then processed with the post-exposure baking (60 °C, 2 min), followed by a third photoresist baking step (90 °C, 60 s) to harden the structure. Finally, these DLW-fabricated 3D microstructures were transferred into a corresponding development system (mixed organic solvents) for a specific period to remove the un-cross-linked resist materials existing both inside and outside the hollow microstructures entirely.

4.2.3 Characterization

A JEOL IT-300 scanning electron microscope ( JEOL Ltd.) was used for the morphological characterization of these prepared hollow 3D microstructures. The surfaces of these 86

Chapter 4 microstructures analysed by SEM were precoated with a 20 nm thick platinum film using a JEOL coating system. A Nikon Eclipse Ni–U optical system equipped with a Nikon Plan Fluor lens was employed for the optical imaging. Deposition of platinum films with different thicknesses (∼10–60 nm) on the surface of the resistive material was performed using the metal coating system, by controlling the deposition time and input current. Surface conductivity of the platinum-coated resist film was measured using a Fluke Multimeter. The surface roughness of the prepared hollow 3D micropore structures was measured and analyzed with the Dektak 150 stylus profiler (Bruker Corp.). The mechanical properties (Young’s modulus) of IP-S-based hollow microneedle structures were measured by the Asylum Research MFP-3D-Bio inverted optical atomic force microscope (Oxford Instruments), in an amplitude modulation–frequency modulation viscoelastic mapping mode. Confocal characterization imaging of the hollow 3D micropore structure was carried out by the confocal laser scanning microscopy (LSM) suite-Leica SP8 confocal laser scanning microscope (Leica AG). Ultrasonic treatment (∼10–30 s, Grant Instruments) was performed to release the hollow 3D microstructures from the loading substrate.

4.2.4 Resistive-Pulse Analysis

The prepared hollow 3D micropore structure used for resistive-pulse analysis was loaded on the PET substrate (∼1 cm × 1 cm). As demonstrated, the substrate microhole (∼Ø 100 μm) fully covered by the upper DLW-TPP-prepared micropore structure was created by our previously developed laser ablation technique. The laser ablation process details were described above. A qNano system (Izon Science, New Zealand) was utilized for the resistive-pulse nanoparticle analysis. The micropore sample was incorporated into the measurement chamber unit of qNano. PBS (1×, Sigma-Aldrich Co.) was used as the electrolyte for resistive-pulse analysis. For the study of the transient i–v curves of the hollow 3D micropore structures with a different-sized circular opening hole (1–5 μm), corresponding ionic currents were recorded at each applied voltage (−1.2 to 1.2 V, 0.6 V increment for each step). For the resistive-pulse analysis of PS nanoparticles, different types of nanoparticle (114–530 nm) solution samples (∼40 μL) were added onto the upper chamber of the measurement unit once the characterization system reached the stabilized states. In the experiment, the current blockades were collectively recorded and analyzed with the built-in qNano analysis.

Chapter 4

4.3 RESULTS AND DISCUSSION

The DLW-TPP process for fabricating hollow 3D micro-devices is illustrated in Figure 1. Here, the commercial Photonics Professional System (Nanoscribe GmbH, Germany) was used to perform the 3D DLW-TPP fabrication process. The underlying TPP mechanism and polymerization schemes are detailed in the Supporting Info. As shown in Figure 1a, the liquid photoresist was used as the immersion fluid between the microscope lens (25 ×, 0.8NA, Zeiss) and the loading substrate. During DLW, a femtosecond laser pulse (80 MHz, 100 fs) was redirected and focused inside a droplet of photoresist to trigger the TPP reactions that cross-link polymeric resist material (see Figure 1c). Significantly, the reactive resist volume is strictly restricted to a nano-sized unit (voxel)26, allowing 3D fabrication of structures with nanoscale resolution (see Figure 1a and b).

Control of micro-device surface properties in turn controls the functionality and performance of hollow 3D micro-devices, and is necessary to allow their uses in many biosystem applications27-28. To evaluate the ability of DLW to precisely control surface properties, a series of conic hollow 3D micropore structures were fabricated under different processing conditions. In preparation, two structural processing settings--slicing distance and hatching distance (see Figure 2a)--were combined to regulate the exposure area of the material used in the DLW fabrication process. Figure 2b illustrates the surface morphologic profiles of the conic hollow 3D micropores prepared using different slicing-hatching parameters, ranging from nano-scale (Slicing dxy: 100nm, Hatching lz: 200nm) to micron-scale (Slicing dxy:

1000nm, Hatching lz: 1200nm). For conical hollow micropores prepared by nano-scale (dxy:

100 nm, lz: 200 nm) slicing parameters, the structures displayed a flawless, smooth surface profile (Ra=81±10 nm). As the slicing and hatching distances were gradually increased, the hollow 3D microstructures showed a tendency toward a more rugged surface in appearance (see Figure 2c, Blue Line).

Chapter 4

Figure 1. (a) Schematic illustration of a DLW-TPP system (Nanoscribe), Oil immersion mode. Femtosecond laser pulse source: 780nm, 80 MHz, 100 fs. Peak power: ~25 kW. Highest fluence level: ~1J/cm2. Insert picture: XYZ-3D movable sample stage. (b) Schematic diagram of the two-photon absorption process (TPA). Excitation time frame: ~10-15 s. (c) Schematic graph of the direct laser writing process. Hollow 3D microstructures were sketched by a CAD program. Under femtosecond laser pulse irradiation, photoresist polymers are cross-linked to formulate the solidified hollow 3D microstructure devices.

Chapter 4

As we further increase the slicing and hatching distances in TPP fabrication, cracks of varying size could be formed on the surface of the prepared hollow 3D microstructures (Figure 2c, Black curve). These cracks can be controlled and may provide useful functions for biosystem micro-devices, such as in pharmaceutical delivery systems requiring the controlled release of drug/vaccine products29-30. Meanwhile, these parameters also affected the overall processing time of DLW process (Figure 2d). Hollow 3D microstructures prepared using finer slicing-hatching parameters to give a crack free surface required longer processing times. Taken together, hollow 3D biosystem micro-devices with accurately- controlled surface profiles can be realized via the DLW-TPP process.

Figure 2. The effects of structural processing settings on the surface morphologic profiles of prepared hollow conic 3D micropore structures. (a) Schematic slicing (up) and hatching (down) patterns for the DLW-TPP process of hollow 3D microstructure devices. (b) Surface 90

Chapter 4 morphologic properties of the hollow 3D microstructures prepared using different-sized constituent units. Scale bar: 2µm, apply to all SEM images. (c) The surface roughness of the hollow 3D microstructures vs Structural processing conditions (Black curve). Width of the cracks formulated on the surface of hollow 3D microstructures vs Structural processing conditions (Blue curve). (d) Minimum required processing time for the fabrication of hollow 3D microstructures vs Structural processing conditions.

Figure 3. (a) The effect of laser power level (fluence) on the quality of incorporated opening hole structures. Upper: SEM images of the built-in circular opening parts (Ø=3 µm) prepared under different power levels. Down: Circularity ratio of the built-in circular opening parts prepared under different power levels. Full power level (100%): ~1J/cm2. Threshold power level: ~40%, dynamic power range: 50% to 60%, damage power range: > 70%. Scale bar: 5µm, apply to all eight SEM images. (b) Gaussian distribution of the laser pulse varied with input laser power levels. (c) Schematic voxel shape inside the laser pulse. (d) Opening holes with distinct geometric shapes incorporated within fabricated hollow 3D microstructure

Chapter 4 devices, left to right: triangle, rectangle, pentagon, hexagon and pentagram. Scale bar: 3µm, apply to all five SEM images.

The opening hole of a hollow 3D micro-device is essential for the overall performance quality in many biosystem implementations, i.e, for the access of (bio)analytes into the interior functional structure. Therefore, to create high-performing hollow 3D biosystem micro- devices, it is essential to optimize the processing parameters that govern the formation process of the opening hole (e.g., laser power and development conditions. etc.,). As shown in Figure 3a, the quality of the opening hole could be dramatically influenced by the input laser power (fluence) levels. To quantitatively evaluate these laser induced effects on the quality of formed circular opening holes, we herein introduce the ratio rCircularity (0≤rCircularity≤1), with which a higher rCircularity value corresponds to a higher quality hole. The results indicated that input laser pulses with low power levels (< 40%) were not able to yield opening(s) with accurately-defined geometric features as the applied laser was not strong enough to polymerize the resist material. (Figure 3a, SEM image 1 to 3).

As the laser fluence level approached the threshold (~40% for IP-S resist), the opening holes showed clean and sharp circular shapes, exactly as CAD-designed. Within the working fluence range (50% to 60% for IP-S, see Figure 3b), the reactive resist volume was strictly confined to a single voxel (see Figure 3c. dxy: 300 nm, lz: 400 nm), creating highly controlled opening hole structures with nanoscale resolution. Significantly, accurate formation of opening hole structures facilitates the wash-away process of trapped resist during the development process, which is necessary to produce clean and accurate hollow 3D micro- devices (Figure S2). Further increments in laser fluence above the working range would lead to opening structures with degraded quality (Figure 3a, SEM image 7 and 8). At higher fluence levels, the over-exposed photoresist undergoes optical damage, which causes imperfections (e.g., cracks, surface debris, etc.,) in the formed opening holes31.

Promisingly, this technique allows opening holes to be created based on arbitrary shapes. As shown in Figure 3d, opening hole structures with distinct geometric shapes (triangle, rectangular, pentagon, hexagon and pentagram) were realized in hollow 3D micro-devices. Our results suggest that the optimum laser power level for accurate opening hole formation 92

Chapter 4 is solely determined by the material properties of the utilized resist, and is independent of the actual hole geometry (Table S1). Therefore, DLW-TPP may serve as a universal platform for creating precise hollow 3D biosystem micro-devices with tailored opening hole(s).

The suitable working temperature of biosystems varies broadly. To verify the extensive applicability of these hollow 3D micro-devices, it is essential to scrutinize their geometric stability under different temperatures. Figure S3 demonstrates the effects of increasing temperatures on the geometric properties of prepared hollow 3D micro-devices. To quantify these temperature effects, we investigated the shrinkage rate of their opening holes. The temperature-induced geometric change was still below 5% even when the temperature was increased from room temperature to a high value (70°C) then rapidly cooled (-20°C) to freeze the device in a shrunken state, indicating that these DLW-TPP prepared hollow 3D microstructures possess good geometric stability over a broad temperature spectrum.

Hollow 3D micro-devices with accessible interiors have been extensively investigated to resolve the thorny problems in biosystems2-3, 32-33. Importantly, biocompatibility of these micro-devices could be modified either through surface modification/functionalization or by using biocompatible photoresists as the bulk material34-36 to seamlessly integrate with the biosystems being analysed (Table S2). It is worthwhile to note that the design of the hollow interior can impact or even govern the fundamental functionality of the prepared hollow 3D micro-devices. Promisingly, the biosystem micro-devices fabricated in this Chapter using the DLW-TPP technique possess tailored geometric features both interior and exterior, and offered customizable functionality and performance. As shown in Figure 4b and c, hollow microneedles suitable for enhanced pharmaceutical delivery were prepared. Through the fine-tuning of geometric features (i.e. surface roughness, surface cracks and interior hollow, etc,.), drug/vaccine products loaded inside the hollow space or anchored on the micro- device surface could be released in a spatiotemporally controlled manner27, 29. Additionally, these solidified hollow 3D microneedle structures exhibited sufficient mechanical stability and rigidity to penetrate skin without inducing any structural fracture (EYoung ~2-3 GPa for IP-

Chapter 4

S resist, Figure S4). Furthermore, highly controllable hollow microneedle array patches can be readily manufactured for various bio-application purposes, as shown in Figure S5.

Figure 4. A series of CAD designed hollow 3D micro-devices were fabricated using the TPP enabled DLW technique(a). Scale bar: 50µm. (b) and (c): Hollow microneedles for enhanced pharmaceutical delivery. Left: Tilt angle-view. Right: Cross-section view. (d) Hollow microelectrode for neural recording and stimulation. Left: Tilt angle-view. Right: Cross- section view. (e) Hollow microvalve for bio-fluidic controlling. Left: Tilt angle-view. Right: Cross-section view. (f) Hollow micro-robot for medical therapy. Left: Tilt angle-view. Right: Cross-section view.

Figure 4d shows a microelectrode that could simultaneously stimulate and record neural signals by integrating the functions of chemical stimulus delivery and electrical signal recording. The conductivity of microelectrodes fabricated by DLW-TPP can be tuned to improve their electrical signal recording capability, via coating of conducting metal thin films on the surface (Table S3), or by using intrinsically conductive photoresist materials37-38. Importantly, the demonstrated hollow 3D microelectrode has a tip size smaller than 10 µm

Chapter 4 and features micropores allowing chemical delivery, sampling or perfusion through the electrode. This would enable precise manipulations and interactions at the single-cell level6, 10. Figure 4e demonstrates the possibility of a hollow microvalve structure containing sophisticated internal movable components. Such microvalves could be used to accurately control and sense bio-fluidic flow conditions in various biosystems with the advantage of being manufactured out of biocompatible materials. Figure 4f shows an example of a micromachine-inspired structure produced with the DLW-TPP technique. Similar structures have been fabricated using tradition chemical and photolithographic techniques in the literature, driven by self-propelling, chemical fuels, electric fields or surface tension39-41. The internal and external geometric features at the core of these devices could easily and accurately be produced via DLW-TPP based on simple CAD designs, greatly simplifying their production process. For usecases requiring individual, detached hollow 3D micro-devices, ultrasonic treatments (~10 to 30 s) have proven to be a robust tool that facilitates the release of these micro-devices from the loading substrate (Figure S6). Therefore, with its ability to precisely specify geometry and operate on a wide spectrum of photoresist materials, DLW-prepared hollow 3D micro-devices would shed light on an extensive range of un-explored usecases.

To validate the practical applicability of these hollow 3D micro-devices, we performed robust resistive-pulse analysis of nanoparticles using the DLW-TPP prepared conic hollow 3D micropores. Resistive-pulse sensing (RPS) has become a prominent technique for the high throughput particle-by-particle characterization of colloid suspensions of individual (bio)analytes in biosystems7, 42-43. One crucial part of this characterization system is the sensitive nano/micro-hollow hole, the preparation of which has been the subject of enormous interests in recent years. However, traditional preparation techniques suffer from several severe deficiencies: solid-state holes typically rely on etching, puncturing or heating- and-pulling methods which do not give tight control over the internal pore geometry, whilst biological pores are limited to very small size regimes and are not suitable for all chemicals or analytes.

The DLW-TPP technique presented in this paper provides a feasible resolution for these critical issues encountered in RPS pore preparation. To explore this, we successfully prepared optimized hollow 3D micropores assembled on plastic substrate (~1cm ×1cm PET, 95

Chapter 4

See Figure 5a and S7), for the high-performance resistive-pulse analysis of nanoparticles. The substrate micro-hole structure (≈ Ø100 µm) beneath the DLW prepared conic micropore device was created via a hybrid additive-subtractive fabrication process employing both the method of this Chapter and our newly developed maskless 3D laser ablation technique44 (Figure S8). Figure 5b schematically shows the two-counter electrode configuration of the resistive-pulse characterization system, in which the prepared micropore structure works as an effective sensitive component for the RPS system (See Figure S9a).

(e)

45 114 nm 330 nm 530nm % 210 nm 400 nm

15 0 Frequency/ 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Blockade Magnitude/nA

Figure 5. (a) DLW-TPP prepared hollow 3D microcone assembled on PET substrate (~1cm×1cm). (b) Schematic graph of the resistive-pulse analysis system using the DLW-

TPP prepared hollow 3D conic micro-hole structure. Working voltage (Vw) is applied on two- sides of the 3D micro-device. (c) i-v curves of hollow 3D microstructures integrated with

Chapter 4 different-sized circular openings, ranging from 1µm to 5µm. Working voltage range: -1.2V to 1.2V, voltage step: 0.06V. The down-right insert figure: The conductivity of resistive-pulse characterization system vs opening size of the hollow 3D micro-hole devices. (d) Single current blockade induced by a 300 nm PS nanoparticles passing through the hollow 3D micro-device with a 3µm-sized circular opening hole. Vw: 0.3V. (e) Current blockade magnitude (Δi) distribution of different types of PS nanoparticles (i.e.,114nm, 210nm, 300nm, 400nm and 530nm) characterized using the hollow 3D microcone structure with a 3µm-sized circular opening hole design. Vw: 0.3V. (f) Combined single current blockade of different types of PS nanoparticles (114 nm to 530 nm) passing through a hollow 3D microcone structure with a 3µm-sized circular opening design. Vw: 0.3V. (g) The relationship of current blockade magnitude vs nanoparticle volume size in resistive-pulse analysis using the hollow

3D microcone structure with 3µm-sized circular opening hole. Vw: 0.3V.

In RPS characterization, the applied external working voltage induced the transient ionic current flowing through the hollow 3D micropore structure. We prepared hollow 3D micropore structures with variable-sized (Ø: 1µm to 5µm) circular opening hole. Static i-v curves of these hollow micropore structures were recorded in Figure 5c. The results showed that, at any particular working voltage, the magnitude of transient ionic current (conductivity) linearly (R2=0.9927) increased with the circular opening hole size (See Figure S9c), a phenomenon that is modelled by Vogel et al45:

(1)

in which V is the applied voltage, DS is the small-side hole size, Dl is the large-side hole size,

LC is the conical hole length and ρ is the electrolyte resistivity(See Figure S9b). This study on transient current behavior suggests that our DLW-TPP prepared micropore devices could be used for practical RPS experiments.

To confirm this, we performed RPS-based nanoparticle size analysis to verify the robustness of these hollow 3D micropore devices. In testing, hollow 3D micropore structures with precisely-defined circular opening holes (Ø=3µm) were prepared (See Figure S7). As shown in Figure 5d, a decrease in ionic current (“current blockade”) occurs as each

Chapter 4 individual nanoparticle translocates through the hollow micropore structure. The overall current plot and other analysis results for the RPS experiment are shown in Figure S10. Figure 5e shows the combined magnitude distributions of current blockades corresponding to each type of analyzed nanoparticle solution (114 nm to 530 nm, see Figure S11). The magnitude of the mode current blockade increased remarkably as the analyzed nanoparticle size was increased from 114 nm to 530 nm (as modelled below by Eqn 2). We then calculated the volume of each type of nanoparticle based on its nominal size and recorded the corresponding current blockade magnitude (mode value) for each measurement (Table S5). These were plotted against each other, and the resulting graph (See Figure 5g) displays a highly linear relationship (R2=0.9921) between the nanoparticle volume and the magnitude of corresponding current blockade, as predicted by Kozak46-47:

(2)

(3)

(4) in which △R is the resistance change, d is the nanoparticle diameter, △i is the current blockade and V is nanoparticle volume size. As a result, these DLW-TPP prepared hollow 3D micropore devices create new opportunities for the accurate, straightforward, and low- cost differentiation of nanoparticles/analytes in bio-systems, via the use of the RPS technique (Figure 5f).

4.4 CONCLUSION In conclusion, we have successfully developed a precise, controllable, and straightforward DLW-TPP based additive fabrication strategy, for the preparation of tailored hollow 3D biosystem micro-devices. Unlike conventional fabrication techniques, the DLW-TPP 98

Chapter 4 fabrication scheme provides superior control over the full geometric properties (both interior and exterior) of the fabricated device. With a well-established fabrication process, we now can systematically and accurately modulate the processing conditions, including slicer parameters, laser input and development treatment, to tune the functionality of these customized hollow 3D biosystem micro-devices. We have demonstrated that the tailored hollow 3D microstructure devices produced by our technique hold great potential for biosystems applications, including micropores for biochemical sensing, microneedles for pharmaceutical delivery, microelectrodes for neural recording, microvalves for bio-fluidic measurement and manipulation, and micromachines for biomedical therapy. To further confirm the suitability of DLW-TPP fabricated hollow 3D micro-devices in practical biosystem implementations, we prepared hollow 3D micropores with accurately-defined opening holes, for the calibration-free resistive-pulse analysis of nanoparticles.

Chapter 4

REFERENCES

(1) Zhao, Y.; Jiang, L. Hollow Micro/Nanomaterials with Multilevel Interior Structures. Adv. Mater. 2009, 21, 3621-3638. (2) X. Lou; L. A. Archer; Yang, Z. Hollow Micro-Nanostructures Synthesis and Applications. Adv.Mater. 2008, 20, 3987–4019. (3) Nagai, M.; Oohara, K.; Kato, K.; Kawashima, T.; Shibata, T. Development and characterization of hollow microprobe array as a potential tool for versatile and massively parallel manipulation of single cells. Biomed. Microdevices 2015, 17, 41. (4) Cheng, Y.; Sugioka, K.; Midorikawa, K. Microfabrication of 3D hollow structures embedded in glass by femtosecond laser for Lab-on-a-chip applications. Appl. Surf. Sci. 2005, 248, 172-176. (5) Kim, Y. C.; Park, J. H.; Prausnitz, M. R. Microneedles for drug and vaccine delivery. Adv. Drug Deliv. Rev. 2012, 64, 1547-1568. (6) Fekete, Z. Technology of ultralong deep brain fluidic microelectrodes combined with etching-before-grinding. Microsyst. Technol. 2013, 21, 341-344. (7) Kozak, D.; Anderson, W.; Vogel, R.; Trau, M. Advances in Resistive Pulse Sensors: Devices bridging the void between molecular and microscopic detection. Nano Today 2011, 6, 531-545. (8) Blundell, E. L. C. J.; Mayne, L. J.; Billinge, E. R.; Platt, M. Emergence of tunable resistive pulse sensing as a biosensor. Anal. Methods 2015, 7, 7055-7066. (9) Ma, G.; Wu, C. Microneedle, bio-microneedle and bio-inspired microneedle: A review. J. Control. Release 2017, 251, 11-23. (10) Pongrácz, A.; Fekete, Z.; Márton, G.; Bérces, Z.; Ulbert, I.; Fürjes, P. Deep-brain silicon multielectrodes for simultaneous in vivo neural recording and drug delivery. Sens. Actuator B-Chem. 2013, 189, 97-105. (11) Schizas, C.; Melissinaki, V.; Gaidukeviciute, A.; Reinhardt, C.; Ohrt, C.; Dedoussis, V.; Chichkov, B. N.; Fotakis, C.; Farsari, M.; Karalekas, D. On the design and fabrication by two-photon polymerization of a readily assembled micro-valve. Int. J. Adv. Manuf. Tech. 2009, 48, 435-441. (12) Reznikova, E. F.; Mohr, J.; Hein, H. Deep photo-lithography characterization of SU- 8 resist layers. Microsys. Technol. 2005, 11, 282-291. (13) S. Ottow; V. Lehmann; Föll, H. Processing of Three‐Dimensional Microstructures Using Macroporous n‐Type Silicon. J. Electrochem. Soc. 1996, 143, 385-390. 100

Chapter 4

(14) Mata, A.; Fleischman, A. J.; Roy, S. Fabrication of multi-layer SU-8 microstructures. J. Micromech. Microeng. 2006, 16, 276-284. (15) Agirregabiria, M.; Blanco, F. J.; Berganzo, J.; Arroyo, M. T.; Fullaondo, A.; Mayora, K.; Ruano-Lopez, J. M. Fabrication of SU-8 multilayer microstructures based on successive CMOS compatible adhesive bonding and releasing steps. Lab Chip 2005, 5, 545-552. (16) Tseng, A. A.; Kuan, C.; Chen, C. D.; Ma, K. J. Electron beam lithography in nanoscale fabrication: recent development. IEEE Trans. Electron. Packag. Manuf. 2003, 26, 141-149. (17) Xu, X.; Yang, Q.; Wattanatorn, N.; Zhao, C.; Chiang, N.; Jonas, S. J.; Weiss, P. S. Multiple-Patterning Nanosphere Lithography for Fabricating Periodic Three-Dimensional Hierarchical Nanostructures. ACS Nano 2017, 11, 10384-10391. (18) Sun, H.-B.; Matsuo, S.; Misawa, H. Three-dimensional photonic crystal structures achieved with two-photon-absorption photopolymerization of resin. Appl. Phys. Lett. 1999, 74, 786-788. (19) Hong-Bo Sun; Takeshi Kawakami; Ying Xu; Jia-Yu Ye; Shigeki Matuso; Hiroaki Misawa; Masafumi Miwa; Kaneko, R. Real three-dimensional microstructures fabricated by photopolymerization of resins through two-photon absorption. Opt. Lett. 2000, 25, 1110- 1112. (20) Kawata, S.; Sun, H.-B. Two-photon photopolymerization as a tool for making micro- devices. Appl. Surf. Sci. 2003, 208-209, 153-158. (21) T. Watanabe; M. Akiyama; K. Totani; S.M. Kuebler; F. Stellacci; W. Wenseleers; K. Braun; S.R. Marder; Perry, J. W. Photoresponsive Hydrogel Microstructure Fabricated by Two-Photon Initiated Polymerization. Adv. Funct. Mater. 2002, 12, 611–614. (22) Montemayor, L. C.; Meza, L. R.; Greer, J. R. Design and Fabrication of Hollow Rigid Nanolattices via Two-Photon Lithography. Advanced Engineering Materials 2014, 16, 184- 189. (23) Sun, H.-B.; Maeda, M.; Takada, K.; Chon, J. W. M.; Gu, M.; Kawata, S. Experimental investigation of single voxels for laser nanofabrication via two-photon photopolymerization. Appl. Phys. Lett. 2003, 83, 819-821. (24) Lao., Z. X.; Hu., Y. L.; Zhang., C. C.; Yang., L.; Li., J. W.; Chu., J. R.; Wu., D. Capillary Force Driven Self-Assembly of Anisotropic Hierarchical Structures Prepared by Femtosecond Laser 3D Printing. ACS Nano 2015, 9, 12060-12069. (25) Li, G. Q.; Lu, Y.; Wu, P. C.; Zhang, Z.; Li, J. W.; Zhu, W. L.; Hu, Y. L.; Wu, D.; Chu, J. R. Fish scale inspired design of underwater superoleophobic microcone arrays by sucrose

101

Chapter 4 solution assisted femtosecond laser irradiation for multifunctional liquid manipulation. Journal of Materials Chemistry A 2015, 3, 18675-18683. (26) Sun, H.-B.; Takada, K.; Kim, M.-S.; Lee, K.-S.; Kawata, S. Scaling laws of voxels in two-photon photopolymerization nanofabrication. Appl. Phys. Lett. 2003, 83, 1104-1106. (27) Van der Maaden, K.; Jiskoot, W.; Bouwstra, J. Microneedle technologies for (trans)dermal drug and vaccine delivery. J. Control. Release 2012, 161, 645-655. (28) Xing, J. F.; Zheng, M. L.; Duan, X. M. Two-photon polymerization microfabrication of hydrogels: an advanced 3D printing technology for tissue engineering and drug delivery. Chem. Soc. Rev. 2015, 44, 5031-5039. (29) Prausnitz, M. R.; Langer, R. Transdermal drug delivery. Nat. Biotechnol. 2008, 26, 1261-1268. (30) Indermun, S.; Luttge, R.; Choonara, Y. E.; Kumar, P.; du Toit, L. C.; Modi, G.; Pillay, V. Current advances in the fabrication of microneedles for transdermal delivery. J. Control. Release 2014, 185, 130-138. (31) Chris B. Schaffer; Andre Brodeur; Mazur, E. Laser-induced breakdown and damage in bulk transparent materials induced by tightly focused femtosecond laser pulses. Meas. Sci. Technol. 2001, 12, 1784–1794. (32) Richter, B.; Hahn, V.; Bertels, S.; Claus, T. K.; Wegener, M.; Delaittre, G.; Barner- Kowollik, C.; Bastmeyer, M. Guiding Cell Attachment in 3D Microscaffolds Selectively Functionalized with Two Distinct Adhesion Proteins. Adv. Mater. 2017, 29, 1604342. (33) Sakellari, I.; Yin, X.; Nesterov, M. L.; Terzaki, K.; Xomalis, A.; Farsari, M. 3D Chiral Plasmonic Metamaterials Fabricated by Direct Laser Writing: The Twisted Omega Particle. Adv. Opt. Mater. 2017, 5, 1700200. (34) B. H. Cumpston; S. P. Ananthavel; S. Barlow; D. L. Dyer; J. E. Ehrlich; L. L. Erskine; A. A. Heikal; S. M. Kuebler; I.-Y. Sandy Lee; D. McCord-Maughon; J. Q. Qin; H. Röckel; M. Rumi; X. L. Wu; S. R. Marder; Perry, J. W. Two-photon polymerization initiators for three- dimensional optical data storage and microfabrication. Nature 1999, 398, 51-54. (35) M. Campbell; D. N. Sharp; M. T. Harrison; R. G. Denning; Turberfield, A. J. Fabrication of photonic crystals for the visible spectrum by holographic lithography. Nature 2000, 404, 53-56. (36) Jonathan D. Pitts; Paul J. Campagnola; Gary A. Epling; Goodman, S. L. Submicron Multiphoton Free-Form Fabrication of Proteins and Polymers Studies of Reaction Efficiencies and Applications in Sustained Release. Macromolecules 2000, 33, 1514–1523.

102

Chapter 4

(37) Blasco, E.; Muller, J.; Muller, P.; Trouillet, V.; Schon, M.; Scherer, T.; Barner-Kowollik, C.; Wegener, M. Fabrication of Conductive 3D Gold-Containing Microstructures via Direct Laser Writing. Adv. Mater. 2016, 28, 3592-3595. (38) Vyatskikh, A.; Delalande, S.; Kudo, A.; Zhang, X.; Portela, C. M.; Greer, J. R. Additive manufacturing of 3D nano-architected metals. Nat.Commun. 2018, 9, 593. (39) Tu, Y.; Peng, F.; Wilson, D. A. Motion Manipulation of Micro- and Nanomotors. Adv. Mater. 2017, 29, 1701970. (40) Yu, Y.; Fu, F.; Shang, L.; Cheng, Y.; Gu, Z.; Zhao, Y. Bioinspired Helical Microfibers from Microfluidics. Adv. Mater. 2017, 29, 1605765. (41) Veronika Magdanz; Samuel Sanchez; Schmidt, O. G. Development of a Sperm- Flagella Driven Micro-Bio-Robot. Adv. Mater. 2013, 25, 6581–6588. (42) Roberts, G. S.; Kozak, D.; Anderson, W.; Broom, M. F.; Vogel, R.; Trau, M. Tunable nano/micropores for particle detection and discrimination: scanning ion occlusion spectroscopy. Small 2010, 6, 2653-2658. (43) Blundell, E. L.; Healey, M. J.; Holton, E.; Sivakumaran, M.; Manstana, S.; Platt, M. Characterisation of the protein corona using tunable resistive pulse sensing: determining the change and distribution of a particle's surface charge. Anal. Bioanal. Chem. 2016, 408, 5757-68. (44) Liao, C.; Anderson, W.; Antaw, F.; Trau, M. Maskless 3D Ablation of Precise Microhole Structures in Plastics Using Femtosecond Laser Pulses. ACS Appl. Mater. Interfaces 2018, 10, 4315-4323. (45) Vogel, R.; Willmott, G.; Kozak, D.; Roberts, G. S.; Anderson, W.; Groenewegen, L.; Glossop, B.; Barnett, A.; Turner, A.; Trau, M. Quantitative sizing of nano/microparticles with a tunable elastomeric pore sensor. Anal. Chem. 2011, 83, 3499-3506. (46) Anderson, W.; Kozak, D.; Coleman, V. A.; Jamting, A. K.; Trau, M. A Comparative Study of Submicron Particle Sizing Platforms: Accuracy, Precision and Resolution Analysis of Polydisperse Particle Size distributions. J. Colloid. Interface Sci. 2013, 405, 322-330. (47) Anderson, W.; Lane, R.; Korbie, D.; Trau, M. Observations of Tunable Resistive Pulse Sensing for Exosome Analysis: Improving System Sensitivity and Stability. Langmuir 2015, 31, 6577-6587.

103

Chapter 4

SUPPLEMENTAL INFORMATION

Contents

1. Introduction 1.1. Two-photon Polymerization (TPP) Mechanism 1.2. Resistive-pulse Analysis 2. Experimental Methods 2.1. Development Process for 3D Hollow Microstructures 2.2. Laser Ablation of Substrate Microhole for 3D Hollow Micropore Figures: Figure S1: Setting-up of Nanoscribe system Figure S2: Development process for the DLW-TPP prepared hollow 3D microstructures Figure S3: Heating effects on prepared hollow 3D microstructures Figure S4: AFM characterization of the hollow microneedle surface coated with Pt film Figure S5: Mechanical property of the microneedle type hollow 3D microstructure Figure S6: Microneed array Figure S7: Ultrosonic treatment to detach prepared microstructure from loading substrate Figure S8: Geometric characterization of prepared hollow 3D microstructure Figure S9: SEM image of the laser ablated substrate hole Figure S10: Resisitive-pulse system setting up Figure S11: Resistive-pulse analysis of nanoparticle Figure S12: SEM characteriation of different types of nanoparticles Tables: Table S1: Development time for distinct-shaped opening structures Table S2: Hollow 3D microstructures prepared using different types of resist materials Table S3: Tunable surface conductivity of prepared hollow 3D microstructures Table S4: Processing conditions of demonstrated 3D hollow biosystem micro-devices Table S5: Volume size and ionic current magnitude for different type of nanoparticles

104

Chapter 4

1. Introduction

1.1 Two-photon Polymerization (TPP) Mechanism In 1931, Maria Goeppert-Mayer theoretically described the process of two-photon absorption (TPA), in which interactions between two photons and one atom/molecule take place during a single quantum event[S1]. As an essential part of TPA process, the collective action of two photons must be present simultaneously to impart enough energy to induce the transition[S2]. To obtain nano/micro-objects with high spatial resolution by making use of TPA polymerization, efficient emission by photosensitiseris of significant importance[S3- 4].When a high intensity laser pulse beam is precisely focused into the small focal volume of photosensitive material, the photosensitiser chromophore is excited by the simultaneous absorption of two photons and emits fluorescent light in the UV-vis regime, which is generally used to enhance two-photon activation. Photoinitiators with high photochemical activity then absorb the emitted fluorescent light and generate radicals (Initiation). The radicals serve as the activator react with monomers or oligomers, producing monomer radicals to expand in a chain reaction (Propagation) until two radicals meet (Termination)[S5-6]. The TPP procedure can be described as follows:

Initiation: SSIRhv, hv**  FL I  R 

 MM Propagation: RMRMRMMRM n

 Termination: RMRMRMRnmn m 

in which S is photosensitiser, I is photoinitiator, M is monomer, and R is active radicals. S* and I * represent the excited state of photosensitiser and photoinitiator, respectively. hv is the Planck’s Equation energy for each adsorbed photon, and FL is the fluorescent light emitted by the excited state photosensitiser.

105

Chapter 4

1.2 Resistive-pulse Analysis

Resistive-pulse sensing (RPS) is an experimental technique used to characterize colloidal particles ranging from approximately 50 nm in diameter up to the size of cells. Resistive pulse sensing can trace its origins back to the Coulter counter, which was proposed in the late 1940s to count and size biological cells and microorganisms[S7-8]. RPS has enabled high throughput micro-particle-by-particle and cells analysis since the 1950s, submicron particles including viruses since the 1970s, particles as small as single molecules over the past decades, and has inspired a particularly intense interest in nanohole-based DNA sequencing analysis recently[S9-11].

To set up the resistive-pulse sensing system, two reservoirs separated by a hole structure are filled with electrolytes (Figure 5b and Figure S10). In measurement, the resistance of hole is monitored by applying a voltage between the reservoirs, which drives a flux of electrolyte ions through it and detecting a current flowing from the voltage source. Transient ionic current changes caused by the translocation of analytes through the hole structure are denoted as “blockade events”, whose magnitude is closely related to the analyte properties[S12]. As shown in eqns (1) and (2), where 휌 is the resistivity of electrolyte, d is the analytes diameter and D is the hole diameter, the magnitude of current blockade increases significantly as the analytes expand the size.

3 4d RR R (1) i(%) (2)  D4 RRR

To date, a plethora of opening hole structures have been investigated for RPS analysis, including solid-state holes (e.g., silicon-based membranes) via photolithographic etching, protein holes (e.g. α-hemolysin) via self-assembly and polymer holes (e.g., PET) via track etching[S13]. The breadth of this field can be inferred by the utilization of varied hole materials including carbon nanotubes, micropipettes, silicon nitride and polymer materials, as well as the diverse techniques for hole fabrication, such as ion beam sculpting, track etching, laser melting, electron beam and soft lithography. However, absolute control over the 3D geometric features of hole structures still remains a significant challenge for conventional fabrication techniques.

106

Chapter 4

2. Experimental Methods

2.1 Development Process for 3D Hollow Microstructures

After DLW process, organic solvents were used to fully remove the unexposed, uncross- linked photoresist materials both trapped inside and anchored outside the 3D hollow microstructures[S14] (Figure S2a). Corresponding organic solvents development systems for each type of investigated photoresist were listed in Table S2. Determining factors affecting the development process, including development time, opening hole size and interior hollow volume, were experimentally explored and are comprehensively discussed. One significant parameter affecting the opening quality is the development time. As shown in Figure S2b, it needs at least 12 hrs to completely develop the hollow micropore structure with a 3 µm- sized circular opening. As the opening size was increased, the minimum required time to finish the development process was dramatically decreased (Figure S2c). The built-in opening components play a crucial role in the formulation of interior hollow components and facilitate the wash-away process of the un-reacted resist material trapped inside the hollow part (Green-circled area). Since the volume of the hollow part was fixed, the uncross-linked resist materials trapped inside could be fully removed within a shorter period of time when the opening size was increased, thus correspondingly reduce the minimum required development time (e.g., 1µm-sized opening: 24 hrs vs 6 µm-sized opening: 6 hrs).

Additionally, it has been shown that when the opening size was fixed, the minimum required development time to obtain workable 3D hollow microstructures would be linearly increased with the volume size of the hollow part (Figure S2d). For instance, when the volume size of 5 3 hollow part was small (~4×10 µm ) , the minimum required development (TMin) was around 5 3 9 hrs. As the hollow part was increased to 13×10 µm , the TMin increased to 28 hrs correspondingly. For the hollow microstructures, the amount of unreacted resist material trapped inside increased correspondingly as the volume size of the hollow part was enlarged. Therefore, to fully remove the uncross-linked resist materials for the microstructures with a larger-sized hollow part, the minimum required development time would be increased. This study on the post-development aspects provides significant insight into the precise fabrication of tailored 3D hollow microstructures with accurately defined hollow components. 107

Chapter 4

2.2 Laser Ablation of Substrate Microhole for 3D Hollow Micropore The 3D hollow micropore structure used for resistive-pulse analysis was assembled on the plastic PET substrate. To realize the resistive-pulse analysis function of the micropore system, substrate hole (Ø~100 µm) beneath the hollow micropore structure was firstly created using our proposed laser ablation technique. To start with, the plastic PET substrate (7 Mil thick, with protective film on both sides) was cleaned by 5 mins ultrasonic treatment. The cleaned plastic substrate was then mounted onto the Nanoscribe-controlled XYZ piezo stage. The 3D ablation process was performed using the laser pulses generated by a mode- locked Ti: sapphire femtosecond laser system (780 nm, 80 MHz, 100 fs). The output from the femtosecond laser source had a Gaussian intensity profile that tightly focused within the PET substrate through a 25X objective lens (0.8 NA). To precisely prepare the through cylindrical micro-holes in plastics, a three-step processing scheme (two FAST ablation steps plus one SLOW polish step) was performed.

In laser ablation process, the input fluence level was set to 0.4 J·cm-2, the scanning speed was at 70 k μm/s. Figure S9 shows the micro-hole structure (Ø: 100 µm) fabricated by the laser ablation technique. The micro-hole has sharp edges and a clean sidewall, without any observable non-defined bulges that would indicate the re-solidification of heat-induced melting debris that form in classical laser ablation processes. After the micro-hole fabrication in PET substrate, we then dropped the photoresist (~80 µL) to fully cover the laser ablated micro-hole. With the aid the incorporated CCD camera, the initial direct-writing position was aligned with the central position of ablated micro-hole. After this, the processed .STL file containing geometric information of the 3D hollow micropore structure was imported into the Nanoscribe system for the direct laser writing process.

108

Chapter 4

Figure S1. Setting-up of the Nanoscribe system. (a) Picture of the Nanoscribe Photonic Professional GT system. (b) Interface of the Nanowrite program controlling the direct laser writing of 3D hollow microstructures.

Table S1. Minimum required development time for openings with distinct geometric shapes

Opening Shape Triangle Rectangle Pentagon Hexagon Star Minimum ~18 hrs ~16 hrs ~16 hrs ~15 hrs ~24 hrs Development Time Dynamic Laser 50%~60% 50%~60% 50%~60% 50%~60% 50%~60% Power Range

109

Chapter 4

Figure S2. (a) Optical image of the development system used to removed the uncross- linked resist materials. (b) The effects of development time on the morphological quality of the circular opening structure (3µm). (c) Minimum development time required to fully remove uncross-linked resist materials vs the opening size. (d) Minimum development time required to fully remove uncross-linked resist materials vs the 3D inside hollow volume size.

110

Chapter 4

(b) (c) 4.2 4.2

3.9 3.9 m m   3.6 3.6 Before Heating Before Heating

After Heating, 50ºC, 1hr After Heating, 70ºC, 1hr

3.3 3.3 Hole Size/ Hole Size/ Hole Average Shrinkage Rate: 3.6 % Average Shrinkage Rate : 4.25%

3.0 3.0 0 2 4 6 8 10 0 2 4 6 8 10 Pore No. Pore No.

Figure S3. (a) Schematic figure shows the opening part size changes under heating effects . (b) The size of the opening (~4µm) before and after heating treatments, 50°C, 10 samples. (c) The size of the opening (~4µm) before and after heating treatments, 70°C, 10 samples.

111

Chapter 4

Table S2. 3D hollow micropore structures (Circular opening 3um in diameter) prepared using different types of photoresists and processing conditions. These structures show good biocompatibility.

Photoresist SU-8 IP-L IP-S IP-Dip Ormocer Threshold 30 % 40% 40% 30% 50% Energy Level Developer SU-8 PGMEA/IPA PGMEA/IPA PGMEA/IPA Ormo Solution Developer Developer Development 6 hrs 20 hrs 16 hrs 20 hrs 10 hrs Time Biocompatibility Excellent[S15] Excellent[S16] Good[S17] Good[S18] Excellent[S19]

Table S3. Tunable surface conductivity of the hollow microelectrode structure. Pt

conducting layer was uniformly coated using a metal material sputtering system. The

distance between two measured points: ~ 1cm

Coating 0s 30s 1 Mins 2 Mins 3 Mins Time Pt Layer / ~10 nm ~ 20 nm ~ 40 nm ~ 60 nm Thickness Resistivity +∞ ~50 kΩ ~30 kΩ ~16 kΩ ~10 kΩ

112

Chapter 4

Figure S4. AFM characterization of the surface roughness of the hollow micro-needle surface coated with Pt film (Thickness: ~50 nm )

Table S4. Processing conditions for the preparation of demonstrated 3D hollow biomicrosystem deviecs

Microstructure Mironeedle- Mironeedle- Microelectrode Micropump Microrobot 1 2 Printing Time 5.5 6.0 5.5 8.5 4.0 (Mins) Development 16 18 16 24 12 Time (Hrs)

113

Chapter 4

Figure S5. Mechanical properties about 3D hollow microneedle structure. The average

Young’s Modulus (EYoung) was 3.0 ± 0.7 GPa, measured by the AM-FM viscoelastic mapping mode using the Asylum Research MFP-3D-Bio Inverted Optical AFM.

Figure S6. Manipulable microneedle array structure, prepared using IP-S photoresist. Fluence level: 50%, scanning speed: 50k µm/s.

114

Chapter 4

Figure S7. Ultrosonic treatment (~1 min) to detach individual 3D hollow microstructure from the substrate. Micro-valve structure attached on the substrate (a) and detached from the substrate after sonication (b). Scale bar: 80 µm.

115

Chapter 4

Figure S8. Charactriztions of developed hollow structures (a) Top-view, cross-section view and bottom-up view of 3D hollow micro-devices used for resistive-pulse analysis. Left and middle: Describe simulated structural images; Right: Experimental SEM images. (b) Confocal image of hollow 3D micropore structure. (c) Optical characterization of interial hollow. (c) SEM characterization of half-cut 3D hollow microstructures.

Figure S9. SEM characterization of the laser-ablated micro-hole in the 7 Mil thick PET sbustrate.

116

Chapter 4

Figure S10. (a) System setting up for the resistive-pulse analysis process. (b) Picture of the resistive-pulse sensing system. (c) Screen capture of the tracking current blockades. (d) Schematic diagram of the conical-shaped micro-cone structure. (e) The magnitude of ionic current vs the size of opening hole. Working voltage: 1V.

117

Chapter 4

Figure S11. (a) The overall current tracking curve the resistive-pulse analysis of 330 nm PS nanoparticles, using the DLW fabricated 3D hollow micropore structures with 3µm-sized opening. Each current blockade represents each individual nanoparticle translocate throught the hollow strucutre. (b) The magnitude distribution of the current blockades of the 330 nm PS nanoparticles resistive-pulse analysis using the 3D hollow micropore structures with 3µm-sized opening. Working voltage: 0.3V. (c) Rate plot of the 330 nm PS nanoparticles resistive-pulse analysis, with a passing speed at 70 nanoparticles/min with 0.3V applied.

118

Chapter 4

Figure S12. SEM characterization image of different-sized PS nanoparticles. (a) to (f), correspond to 114 nm nanoparticle, 210 nm nanoparticle, 300 nm nanoparticle, 330 nm nanoparticle, 400 nm nanoparticle and 530 nm nanoparticle, respectively.

Table S5. Volume size for each type nanoparticle and recorded magnitude of corresponding current blockade (Mode value) for each individual measurement.

Nanoparticle 114 nm 210 300 330 400 530 nm nm nm nm nm 3 Volume/ퟏퟎ−ퟑµm 0.775 4.86 14.14 18.82 33.52 77.96 Ionic blockade 0.061 0.22 0.396 0.534 0.805 1.97 (Modal) /nA

119

Chapter 4

REFERENCE

2+ [S1] Kaiser, W.; Garrett, C. G. B. Two-photon excitation in CaF2: Eu Phys. Rev. Lett. 1961, 7, 229-231. [S2] Bray, R. G.; Hochstrasser, R. M. Two-photon absorption by rotating diatomic molecules. Mol. Phys. 1976, 31, 1199-1211. [S3] He, G. S.; Xu, C.; Prasad, P. N.; Reinhardt, B. A.; Bhatt, J. C.; McKellar, R.; Dillard, A. G. Two-photon absorption and optical-limiting properties of novel organic compounds. Opt. Lett. 1995, 20, 435-437. [S4] Kogej, T.; Beljonne, D.; Meyers, F.; Perry, J.W.; Marder, S.R.; Brédas, J. L. Mechanisms for enhancement of two-photon absorption in donor–acceptor conjugated chromophores. Chem. Phys. Lett. 1998, 298, 1-6. [S5] Zhou, W.; Kuebler, S. M.; Braun, K. L.; Yu, T.Y.; Cammack, J. K.; Ober, C. K.; Perry, J. W.; Marder, S. R. An efficient two-photon-generated photoacid applied to positive- tone 3D microfabrication. Science 2002, 296, 1106-1109. [S6] Das, S.; Nag, A.; Goswami, D.; Bharadwaj, P. K. Zinc(II)- and copper(I)-mediated large two-photon absorption cross sections in a bis-cinnamaldiminato schiff base. J. Am. Chem. Soc. 2006, 128, 402-403. [S7] Kozak, D.; Anderson, W.; Vogel, R.; Trau, M. Advances in resistive pulse sensors: Devices bridging the void between molecular and microscopic detection. Nano Today 2011, 6, 531-545. [S8] Blundell, E. L. C. J.; Mayne, L. J.; Billinge, E. R.; Platt, M. Emergence of tunable resistive pulse sensing as a biosensor. Anal. Methods 2015, 7, 7055-7066. [S9] Platt, M.; Willmott, G. R.; Lee, G. U. Resistive pulse sensing of analyte‐induced multicomponent rod aggregation using tunable pores. Small 2012, 8, 2436-2444. [S10] Billinge, E. R.; Broom, M.; Platt, M. Monitoring aptamer–protein interactions using tunable resistive pulse sensing. Anal. Chem. 2014, 86, 1030-1037. [S11] Billinge, E. R.; Platt, M. Aptamer based dispersion assay using tunable resistive pulse sensing (TRPS). Anal. Methods 2015, 7, 8534-8538. [S12] Kozak, D.; Kithva, P.; Bax, J.; Surawski, P. P.; Monteiro, M. J.; Trau, M. Development of encoded particle-polymer arrays for the accelerated screening of antifouling layers. Chem Commun (Camb) 2011, 47, 9687-9689.

120

Chapter 4

[S13] Willmott, G. R.; Vogel, R.; Yu, S. S.; Groenewegen, L. G.; Roberts, G. S.; Kozak, D.; Anderson, W.; Trau, M. Use of tunable nanopore blockade rates to investigate colloidal dispersions. J. Phys. Condens. Matter. 2010, 22, 454116.

[S14] Lee, S. H.; Jeong, H. E.; Park, M. C.; Hur, J. Y.; Cho, H. S.; Park, S. H.; Suh, K. Y. Fabrication of hollow polymeric microstructures for shear‐protecting cell‐containers. Adv. Mater. 2008, 20, 788-792. [S15] Nemani, K. V.; Moodie, K. L.; Brennick, J. B.; Su, A.; Gimi, B. In vitro and in vivo evaluation of SU-8 biocompatibility. Mater. Sci. Eng. C Mater. Biol. Appl. 2013, 33, 4453-4459. [S16] Tottori, S.; Zhang, L.; Qiu, F.; Krawczyk, K. K.; Franco-Obregon, A.; Nelson, B. J. Magnetic helical micromachines: fabrication, controlled swimming, and cargo transport. Adv. Mater. 2012, 24, 811-816. [S17] Worthington, K. S.; Wiley, L. A.; Kaalberg, E. E.; Collins, M. M.; Mullins, R. F.; Stone, E. M.; Tucker, B. A. Two-photon polymerization for production of human iPSC- derived retinal cell grafts. Acta. Biomater. 2017, 55, 385-395.

[S18] Hengsbach, S.; Lantada, A. D. Rapid prototyping of multi-scale biomedical microdevices by combining additive manufacturing technologies. Biomed. Microdevices 2014, 16, 617-627.

[S19] Haas, K.H.; Wolter, H. Synthesis, properties and applications of inorganic–organic copolymers (ORMOCER®s). Curr. Opin. Solid State Mater. Sci. 1999, 4, 571-580.

121

Chapter 5

Candidate’s contribution to the authorship

I declared that I have obtained permission from all co-authors to include following publication directly into this chapter.

Caizhi Liao, Fiach Antaw, Alain Wuethrich, Will Anderson, Matt Trau. “Two-Photon Nanolithography of Tailored Hollow three-dimensional Microdevices for Biosystems”. Small Structures, 2020, DOI: 10.1002/sstr.202000011. - Incorporated as Chapter 5.

Contributor Statement of contribution

Conception and design (80%) Experiments (95%) Caizhi Liao (Candidate) Analysis and interpretation (70%) Drafting and production (60%)

Conception and design (5%) Fiach Antaw Analysis and interpretation (15%) Drafting and production (10%) Conception and design (5%) Alain Wuethrich Analysis and interpretation (5%) Drafting and production (20%) Conception and design (5%) Experiments (5%) Will Anderson Analysis and interpretation (5%) Drafting and production (5%) Conception and design (5%) Analysis and interpretation (5%) Matt Trau Drafting and production (5%)

122

Chapter 5

Configurable Miniaturized Three-Dimensional Pores for Robust Single Nanoparticle Analysis

Summary The preceding Chapter (Milestone 2) described the preparation of tailored 3D hollow constructs, including miniaturized pore structures for the resistive pulse sensing (RPS) system. The 3D geometry of RPS pores has a profound impact on RPS analysis. To date, however, fabrication techniques for the RPS pore can not accurately control the 3D inside geometry, raising significant issues for RPS purposes. As achieved by the previous Milestone, the established TPP-based nanolithography platform allows us to prepare accurate 3D micro-pores. In Chapter 5, we attempt to fulfil Milestone 3 by extending the fabrication capability of TPP for more complicated (yet controlled) 3D RPS pores, which can generate finger-printing RPS signals for robust nanoparticle analysis. Four important types of micro-RPS pores, including cis-conical, trans-conical, symmetrical and cylindrical, are prepared and characterized. Significantly, the attainment of Milestone 3 helps to further advance this field by presentation of novel and powerful RPS pore constructs, as demonstrated by Milestone 4 work.

Chapter 5 is based on a published paper:

Caizhi Liao, Fiach Antaw, Alain Wuethrich, Will Anderson, Matt Trau. “Configurable Miniaturized Three-Dimensional Pores for Robust Single Nanoparticle Analysis” Small Structures 2020, 2000011.

123

Chapter 5

ABSTRACT Resistive-pulse sensing (RPS) has become a pivotal platform for single-molecule and nanoparticle analysis. Key to resistive pulse sensing is the sensing pore structure, the preparation of which is a subject of active research. Whilst existing schemes can produce pores with precise entrance diameters, producing pores with arbitrarily complex, 3- dimensional internal structures remains an open problem. Herein, two-photon polymerization (TPP) based nanolithography is introduced for reliable preparation of customizable RPS pores. For the first time, accurate micro- and nano-pores with different cone angles have been successfully prepared and their performance has been studied experimentally and by simulation. Subsequently, accurate 3D pores were studied for selected RPS analysis: cis- and trans-conical pores for the investigation of pore’s preferential transport capability; symmetrical pores for the electrical tracking of nanoparticle position; and cylindrical pores for the surface charge analysis of chemically distinct nanoparticles of the same size. The TPP nanolithography technique enables tailored 3D pore designs with openings as small as 600 nm in diameter, providing opportunities for new RPS implementations that may simultaneously investigate the physical and transport properties of translocating objects.

5.1 INTRODUCTION Resistive-pulse sensing (RPS) has attracted great attention as an analytical technique for in-solution characterisation of nano-scaled objects such as viruses1, particles2, colloids and individual bio-molecules3. In RPS, a voltage and/or pressure gradient is applied at two separated electrolyte reservoirs, forcing analyte to translocate through a micro-/nano-pore. This results in a transient change in the resistance measured across the two reservoirs, referred as a resistive pulse. Information reflecting the physical and chemical properties of particles are included in the acquired resistive pulse signals (e.g., pulse magnitude and duration).

The geometry of the micro-/nano-pore plays a central role in RPS systems4-6. So far, pores of varying depth have been explored and shown to generate characteristic resistive pulses due to their different ion transport properties.7-8 The RPS signal presents a treasure-trove of 124

Chapter 5 information, reflecting both the particle properties and the precise dynamics of the particle- pore interaction. However, without precise measurements of pore geometry it is impossible to fully resolve the information contained within the signal, and so most RPS analysis has been limited to relatively simple measurements such as maximum change-in-resistance or approximate event duration. Such measurements can be used to obtain precise per-particle size and reasonably accurate zeta potential from relatively few direct measurements of pore geometry9 or from calibration using particles of known size10. Full control over pore geometry would allow for significantly more sophisticated analysis to be performed, as it would permit direct measurements of particle position and velocity throughout the entire translocation event. This would permit not only improvements to the accuracy of particle size and charge measurements, but would enable novel analyses that take advantage of tuned fluid velocity profiles or pore surface coatings11. Hence, fabrication techniques enabling the design and generation of purpose-built, geometrically well-defined micro-/nano-pores are currently of immense interest.

RPS pores can be grouped into lipid-membrane pores and solid-state pores: Lipid- membrane pores are self-assembled from biological elements (e.g., proteins, peptides and DNA) and embedded in a lipid membrane12-14, while solid-state pores are created by forming 6, 15-16 a hole in a solid substrate (e.g., Si, Si3N4 and PET) . Artificial lipid-membrane pores have been successfully used to filter small molecules based on size and select for specific molecules12; however such pores cannot withstand large variations in external conditions (e.g., temperature, pH, ionic strength and applied voltage) limiting their usefulness in practical RPS implementations14. In comparison, solid-state pores demonstrate high chemical and mechanical stability, more easily tuned channel depth and opening size, and in many cases can be integrated directly into microfluidic devices and chips5. Despite this, current pore fabrication technologies—most notably traditional photolithography/etching techniques—are still not able to achieve reliable control over the 3D internal geometry of the produced pores16. Thus, there is a need for fabrication techniques that can reproducibly produce pores with well-defined internal geometries.

Recently, two-photon polymerization (TPP) nanolithography has emerged as a powerful tool for rapid and accurate 3D prototyping of microscopic structures17-19. In TPP, a femtosecond pulsed laser beam is tightly focused on a photosensitive resist material consisting of a mixture of monomers and photoinitiators. The photoinitiator chemistry and laser frequency 125

Chapter 5 are chosen such that the absorption of two photons is required to trigger the crosslinking of the resist, and the short pulse-width of the laser beam ensures that only at the focal point of the objective lens is there a significant probability that two photons will overlap and be absorbed by the resist simultaneously. This leads to localized crosslinking of photosensitive resist material only at the focal point, and by moving the focal point in through the liquid material, arbitrary 3D structures can be formed. By taking advantage of precision optics, tailored 3D structures with up to 100 nm resolution can be realized, suitable for the fabrication of micro-/nano-pores with configurable internal geometries20-21.

In this work, we develop a TPP technique for the construction of customizable cylindrical, asymmetric cis- and trans-conical, and symmetrical conical RPS pores. The TPP technique enabled the preparation of well-defined pores with low surface roughness (Ra<100 nm) and entrance diameters as small as 600 nm. Fundamental properties of these miniaturized pores including noise behaviour, i-v curves and surface properties were thoroughly investigated. The effect of cone angle on background resistance was systematically studied and compared to theoretical simulations. We further used the precise knowledge of pore geometry afforded to us by our technique to measure particle position directly, and applied this nanoparticle tracking system to extract accurate particle velocity and zeta potential information.

5.2 EXPERIMENTAL SECTION

5.2.1 Materials Phosphate buffered saline (PBS) tablet (Sigma-Aldrich Co.) was utilized for the preparation of PBS solution. Deionized (DI) water generated from Milli-Q (Merck Millihole) was used for the preparation of PBS solution and nanoparticle samples. Resist materials for TPP pore fabrication was purchased from Nanoscribe GmbH and was used as received. Different types of polystyrene (PS) nanoparticles functionalized with carboxyl group were obtained from Bangs Laboratories, Inc. Bovine Serum Albumin (BSA) lyophilized powder was provided by Sigma-Aldrich and was dissolved by DI water for surface modification purpose. The computer aided designs of different pores were realized by AutoCAD 2018 and Solidworks 2018, and were further processed by Nanowork to generate .GWL file for TPP fabrication. 126

Chapter 5

5.2.2 Characterizations

Scanning electron microscopic (SEM) characterization for morphological characterization of TPP generated pores was obtained from JEOL IT-300 SEM (JEOL Ltd.). The pores samples were pre-coated with ~30 nm thick Pt layer using metal coating system. The surface roughness of pore wall was characterized by the Asylum Research MFP-3D-Bio Inverted Optical AFM (Oxford Instruments). Nikon ECLIPSE Ni-U optical system equipped with a Nikon Plan Fluor lens (10X, 20X and 40X) was utilized for optical imaging. Ultrasonic treatment (~1 mins, Grant Instruments) was performed to clean the substrates loading the TPP resist materials. UV spectroscopy of IP-s resist was recorded by Nano Drop Spectrophotometer (ND-1000). Surface charge/zeta potential measurements of pore walls were realized by SurPASS electrokinetic analyser. Standardized methods for nanoparticle zeta potential measurements were enabled by ZetaSizer Nano ZS (Malvern Instrument, UK). The zeta potential analysis of pore wall surface was realized by the SurPASSTM 3 Electrokinetc Analyzer Platform. The Fourier-transform infrared spectroscopy (FTIR) characterizations of both the liquid IP-S resist and the solid cross-linked films were performed using Cary 630 FTIR Spectrometer, Agilent.

5.2.3 Resistive-pulse Sensing q-Nano RPS system is utilized for the analysis of nanoparticles. The background RPS currents of a conical pore with 3µm sized opening were recorded under different voltages (0.3V, 0.5V and 0.7V) and were analysed by Clampfit 10.7 for noise behaviour investigations. i-v curves of distinct-sized RPS pores were recorded by varying the addressed voltage from -1.2 V to +1.2 V, with a step of 0.06V. For the investigation of tilt-angle on RPS analysis, stabilized background currents obtained through pores with different designs were normalized for analysis. Cis-conical pore and trans-conical pore were utilized for the investigation of pore’s preferential transport capability at the voltage of 0.3V. Symmetrical pores were integrated with q-Nano system, and corresponding finger-printing signals as a 600 nm nanoparticle translocating through the pore at 0.5V were explored for the electrical tracking of nanoparticle position within the pore. To perform the Zeta potential analysis of distinct nanoparticles, optimized cylindrical pores were prepared and exploited for the dwell

127

Chapter 5 time measurements of 960 nm modified nanoparticles at 0.8V. Noting that electric field is crucial to optimize the nanoparticle translocating process for each pore structure.

Figure 1. (a) Schematic picture showing the TPP-DLW process for 3D configurable pore fabrication. (b) The setup up of an exemplar RPS system and its equivalent circuit diagram. Characteristic resistive pulses are generated as analytes translocate through the pore. (c) Real-time observation of TPP fabrication of pore through CCD camera.

5.3 RESULTS AND DISCUSSION

The proposed TPP nanolithography of two photons for fabricating configurable 3D miniaturized RPS pores is illustrated in Figure 1a. The photosensitive material (IP-S resist, Figure S1) was loaded to a 3D piezo-stage controlled by the Photonics Professional System (Nanoscribe GmbH, Germany). During TPP, a femtosecond pulsed laser beam (80 MHz, 100 fs) is tightly focused on the resist material to cross-link the photosensitive materials. 128

Chapter 5

RPS micro-/nano-pore structures are generated by passing the laser through the resist in a pattern generated from design files specifying the desired geometry of the pore. Due to the refractive index difference between the cross-linked resist and the uncross-linked one, the formation process of RPS pores can be observed in real-time using a CCD camera (Figure 1c). After TPP processing, the 3D pores are obtained by washing away the uncross-linked resist via standard photolithographic development techniques. Once completed, these pores were integrated with the Izon q-Nano measurement platform for RPS analysis. As shown in Figure 1b, when driven through the pore by an electric-field and/or pressure, particles cause transient interruptions of the flux of ions in the pore channel and thus induce a resistive pulse. Of significance to our work is the fact that the internal pore geometry can dramatically alter the shape of the resistive-pulse signal, providing opportunities for accurate, multipurpose and high-throughput RPS analysis by optimizing the 3D geometries of individual pore devices[14,18].

5.3.1 Noise Behaviors The capabilities of RPS implementations are determined predominantly by the system noise and pore size, which give both the range of particle sizes that can be detected and the accuracy with which those sizes can be resolved. In particular, accuracy and minimum particle size are both determined by the system noise, which in a sufficiently well-designed system is determined mainly by the noise of the pore itself. Typically, RPS experiments use electronic low-pass filters to reduce noise by reducing system bandwidth (typically down to 10-100kHz), and so low-frequency noise is the dominant type of noise in these experiments. It should be noted that noise in polymeric pores typically follows a 1/ f distribution at low frequencies15, 22-23, and the magnitude of this noise is governed by the properties of the materials comprising the pore structure. Figure 2a shows the normalized current power spectral density of an individual TPP-fabricated conical pore under different voltages. The low-frequency 1/ f noise demonstrates non-equilibrium behaviour, as the normalized power spectrum increases with increasing voltage. This observation is in agreement with Siwy’s earlier studies of noise behaviour in polymeric pores: It is likely that charged groups on the pore walls undergo a dynamic protonation/deprotonation process mediated by the trans- pore electric field, inducing voltage-dependent current fluctuations22, 24. Since our TPP- generated polymeric pore demonstrates increased noise at higher voltages, we chose low voltage (<1V) settings for the RPS analyses performed in this paper. In our TPP process,

129

Chapter 5 the system noise was further affected by the surface roughness of the pore. Increased pore surface roughness leads to fluctuations in ionic current and increased system noise. By adjusting the slicing parameters of the software used to convert CAD geometry into TPP instrument instructions, we were able to fabricate pores of low roughness (Ra ~89nm) that yielded a system noise of ~16 pA (Figure S2).

Figure 2. (a) The effects of voltage on the noise behaviour of TPP-fabricated polymeric pores. Pore opening diameter: 3 µm. Working voltages: 0.3V, 0.5V, 0.7V. (b) i-v curves of pores of different diameters. Characterization step: 0.06V (c) Schematic diagram showing the surface modification process of pore wall in acidic environments and in a protein solution. (d) Pore walls with different surface charges and bio-functionality were characterized by the SurPASSTM 3 electrokinetic analyser. The number of 훿- symbols schematically indicates the density of negative charge groups present on the pore surface with a decrease in negative charges from group 1 to 4. For each set of experimental conditions, more than 5 individual measurements were performed and the results were averaged. The error bars reflect the minimum and maximum measurements obtained for each surface modification process.

130

Chapter 5

5.3.2 I-V Curves

The current−voltage (i−v) response of a micro-/nano-pore provides detailed information about the ionic behaviour of the pore, and as a result the i-v curves of a variety of pore structures have been investigated in the literature7-8, 25. Figure 2b shows example i−v curves for conical RPS pores with diameters of 1, 2, and 3 µm, as prepared with our TPP technique. The magnitude of the ionic current increased linearly (R2=0.9913) with voltage. As the opening size of pore is significantly larger (>100 times) than the thickness of the electric double layer (EDL), it is unsurprising that no appreciable ion current rectification was observed in RPS characterization. Significantly, our nanolithography fabrication scheme could reliably generate geometry-controlled RPS pores with opening diameters down to Ø 600 nm (Figure S3).

5.3.3 Controllable Surface Property

It is generally recognized that the chemical surface properties of pores are of significance for electrokinetic fluid flow and for the molecular selectivity of RPS analysers26-27, with electroosmotic flow proportional to the zeta potential (i.e. surface charge) of the pore wall per the well-known Helmholtz-Smoluchowski equation. Since our TPP-prepared pores are made of cross-linked polymeric materials, an abundance of functionalizable groups (e.g., carboxyl and carboxylate group, etc.,) are formed upon laser exposure, as verified by the Fourier-transform infrared spectroscopy (FTIR) characterization shown in Figure S4. By exploiting these functional groups, we have developed an effective approach for the fine- tuning of pore surface charge: As shown in Figure 2c, carboxyl groups on the pore surface undergo a dynamic protonation/de-protonation process in acidic environments. By selecting appropriately acidic pH levels (1 to 3), surface charge features could be manipulated in a controlled way (Figure 2d). Modification of the pore surface with biomolecules is also possible. Protein bovine serum albumin (BSA) is a well-studied model for surface modification28; as a proof -of-concept we have successfully conjugated BSA molecules on the pore surface. Surface charge measurements indicated that BSA molecules partially conjugated with the attached reactive negative carboxyl groups, thus reducing the amount of surface charge on the pore wall surface.

131

Chapter 5

5.3.4 The Role of Cone Angle

The performance of RPS analyses utilizing classical conical pore structures is mainly dictated by the sensing zone of the pore as defined by the small opening tip27. This sensing zone is intrinsically linked to the geometric profile of the opening tip, with the angle of the cone (Figure 3a) structure dramatically influencing both the sharpness of the sensing region and the transport properties of the pore. Ion current rectification, conductivity and selectivity are all known to be affected by cone angle in conical pores7. However, the effects of cone angle on RPS analyses have not yet been experimentally studied due to the difficulty of precisely controlling the geometry of the pore tip in conventional fabrication methods.

To investigate the effect of cone angle, we first prepared a set of RPS pores with cone angles of 0°, 15°, 30°, and 45° (Figure 3c). By optimizing the slicing/hatching parameters used to generate the scanning patterns of the laser pulse in TPP process, we created pores with precisely defined openings and highly smooth surfaces (Ra: ~ 188nm, Figure S5). Smooth pore surface are important because they reduce or even eliminate ionic current fluctuations making it an ideal pore structure for reliable RPS applications29.

132

Chapter 5

Figure 3. (a) Schematic diagram showing the designs of conically-shaped pores with different cone angles. (b) Cross-section view (schematic) and top view (SEM) of a pore. For a given pore with fixed opening diameter ‘d’, the cross-sectional area is defined by the cone angle. (c) SEM characterizations of TPP-fabricated pores with different cone angle designs. The opening diameter of these pores is nominally 3 µm. (d) Comparative study of theoretical calculation (black triangle curve), simulation (red dot curve) and experimental RPS analysis (vary-coloured square curve ) using differently angled pores. For the experimental measurements, more than 6 identical structures were prepared and characterized for each type of pore and the results were averaged. The error bars reflect the minimum and maximum measurement for each type of pore.

To examine the fabrication quality of the pore with different cone angles, we measured the ionic current and compared the experimental values to the theoretical and simulated ionic current values. The dependence of ionic current on cone angle is described by 30-31

A ()(tan) ZDdZ   22 (1)

0 dz IV   L AZ () (2) in which A(Z) is the cross-section area of pore, I is the current, V is the addressed voltage, Z is the pore length, ρ is resistivity, D is the diameter (Figure 3a and b), and θ is the cone angle. The resistance of the pore in this model decreases as the cone angle increases, thus increasing the ionic current. These cone-angle induced effects were further examined by solving Poisson’s equation for electrostatics using the finite element method (FEM) for pores at different cone angles. These FEM simulations were performed using FEniCS, an open- source python toolkit for numerically solving partial differential equations. Further details of these simulations are available as supplementary information. Figure 3d shows the experimental, calculated, and simulated ionic current values for pores of different cone angle. The experimental values were very similar to the calculated and simulated ionic currents indicating that the cone angle of the physical pores has been fabricated accurately and could be used for investigating the effect of cone angle on various transport phenomena. 133

Chapter 5

5.3.5 Conical Pore for Nanoparticle Analysis

Next, we examined the capability of the TPP fabricated cis-conical pore with a cone angle of 30 degree for RPS. The pore was integrated with an Izon q-Nano RPS platform for measurement and characterization of nanoparticles. As shown in Figure 4, the ionic current underwent an abrupt drop at the moment the analysed nanoparticles entered the pore orifice. As the nanoparticle was electrophoretically driven toward the larger exit opening, the transient current gradually increased to the original undisturbed value. The size distribution of a nominally 530 nm set of nanoparticles was determined using the Izon qNano software (Figure S6). Our measurements revealed that the diameter of the analysed nanoparticles was around 578±13nm (modal value), close to the nominal 530 nm size, a result that is comparable to the one obtained by commercial pores. To investigate how the geometric profile of a pore modulates the shape of the resistive pulse, we then designed and constructed a trans-conical RPS pore and compared the resistive pulses it produced to the pulses produced by the previously introduced cis-conical pore. Figure 4a and Figure 4b show the experimentally observed resistive pulse events generated as a nanoparticle passes through the cis- and trans-conical pores, respectively. Compared to the sudden current drop and slow rise back to background current observed in RPS analysis using the cis-conical pore, the transient current of the trans-conical pore appears mirrored, gradually reducing to the peak and quickly returning to the background current.

It has been reported that the steady-state distribution of ions at the entrance of an asymmetric RPS pore is itself asymmetric, resulting in a preferential direction for the transport of ions across the pore.31 With the aim of better understanding this reported asymmetric transport behaviour in RPS, we collectively analysed and compared the resistive-pulse signals obtained from cis- and trans-conical pores (Figure 4c). Interestingly, our results suggested that RPS characterizations using different-shaped conical pores could generate similar quantitative results in macro-level analysis: For the RPS analysis using cis- conical pore, the average pulse height is 0.291 nA and the average baseline width is 2.54 ms; likewise for the RPS analysis using trans-conical pore, the average pulse height is 0.283 nA and the average baseline width is 2.68 ms. These findings were consistent with those achieved by inverting the original Izon qNano plastic pores2, 9. In spite of the complex 134

Chapter 5 trajectories that particles can follow, the average resistive pulse of the trans-conical pore was observed to be the mirror image of the average cis- pore pulse when run using the same particle sets.

Figure 4. (a) and (b) Resistive pulses captured by a cis-conical pore and trans-conical pore, respectively. Nanoparticle size: 530 nm; Operating voltage: 0.3V. The opening pore diameter for cis-conical pore and trans-conical pore is nominally 3 µm. (c) Comparative study of the collective RPS events obtained from those two conical pores. The circles (i, ii,

135

Chapter 5 iii and iv) including data groups represent the data cluster with extended translocation time. (d) Position vs Time profile for three types of pores. Result obtained by using an integrated experimental and simulation approach described in the supplementary information. The point at which the particle passed through the tightest constriction of the pore was used as the origin of both axes for all three pore types. (e) Resistive pulses captured by a symmetrical pore. Nanoparticle size: 600 nm; Operating voltage: 0.5V. Inserted right diagram: An RPS analysis system utilizing a symmetrical pore as an electrical nanoparticle positioning tracking system. The opening diameter of symmetrical pore is nominally 6 µm, and the diameter of central symetrical is nominally 3 µm.

In the case of the trans-conical pore, a number of long-duration events with extended dwell time-the circled events in Figure 4c-were observed during RPS measurements. We suspect that these clusters of long events are caused by interacting with the pore wall. For micron- scale RPS analysis using the trans-conical pore (⌀ 3µm opening), this clogging effect is especially pronounced when the size of translocating particles increases above 700 nm, or the applied voltage drops below 0.3V. Analysis of pores demonstrating this issue under SEM shows the presence of large nanoparticles occluding the pore opening (Figure S7). It is worthwhile to note that this clogging problem is not unique to our TPP-generated polymeric pores32-33. To address this issue, we took two simple approaches: First, we strengthened the electrostatic repulsion between the pore and nanoparticles by increasing solution pH to 9 to enhance the negative surface charge of the pore wall, and second we inhibited particle aggregation within the pore tip by adding Tween 20 (0.05%) surfactant into nanoparticle solutions. Based on the results of our RPS studies, we concluded that geometric variations in pore shape could modify the resistive pulse produced by a given nanoparticle and hence could be used to generate finger-printed RPS signals, opening up new opportunities for enhanced RPS implementations, including the dynamic investigation of translocating components, or even the manipulation of single nanoparticles.

136

Chapter 5

5.3.6 Symmetric Pore for Nanoparticle Analysis

Constrained by the inadequate fabrication capabilities of current techniques, micro-/nanopores with symmetric shape have been rarely reported and exploited for RPS analysis6, 16. Recent theoretical attempts to explore the potential uses of symmetric pores in RPS have demonstrated that these particular pores may be able to more precisely probe nanoparticles34. Inspired by those simulation results, we used TPP fabrication to prepare symmetric pores and performed RPS analysis using the Izon qNano system as discussed previously. Theoretically, in the presence of a suitable driving voltage, the translocation of nanoparticles along or sufficiently close to the central axis of the pore will produce a resistive pulse with a characteristically symmetric shape (Figure S8), which can be used to derive particle position along the pore axis (Figure 4e). By using the known geometry of the pore to simulate the change in resistance as a particle moves through the pore, we have been able to generate plots of particle position versus resistance for both types of conical pore and our symmetric pore design. These were then applied to data obtained from resistive pulse sensing experiments to obtain plots of average nanoparticle position inside the pore versus time, as described in detail in the supplementary information. These results demonstrated that symmetric pores could be used to track particle position for longer than standard cis- and trans-conical pores (Figure 4d), with no ambiguity if particle transport is assumed to occur in one direction. Promisingly, this suggests that the RPS platform in combination with symmetric pores serves as a novel electrical nanoparticle positioning system, with which we can readily perform real-time monitoring of a single nanoparticle’s location along its trajectory and thus observe particle dynamics as they translocates the pore. Such information could be used to measure particle shape and rotational diffusion35, and could even be exploited to observe particle-pore interactions in the case of a functionalized pore surface.

137

Chapter 5

Figure 5. (a): Resistive pulses captured by a cylindrical pore with an aspect ratio of 5. Nanoparticle size: 960 nm; Operating voltage: 0.8V (b) Dwell time distribution of an unmodified nanoparticle translocating through the cylindrical pore. (c): Diagram showing the electrokinetic and pressure forces driving nanoparticle translocation. (d): Measured Zeta potential of 530 nm diameter polystyrene particles in different solution conditions using the cylindrical pore. The particle solutions were (1) 1X PBS at pH 1, (2) 1X PBS at pH 2, (3) 1X PBS at pH 3, and (4) 5mg/ml BSA in 1X PBS at pH 7. The blue data group was analysed using our RPS based approach, whilst the yellow data group was analysed using standardized method (ZetaSizer Nano ZS system). The number of 훿- symbols schematically indicates the density of negative charge groups present on the pore surface. For each set of experimental conditions, more than 8 individual measurements were performed and the average taken. The error bars represent the minimum and maximum measurements for each group.

138

Chapter 5

5.3.7 Cylindrical Pore for Nanoparticle Analysis

Studying the event duration of resistive pulses measured during RPS allows the determination of the zeta potential of individual molecules, and hence can be used to differentiate chemically distinct particles of similar sizes36-37. Until now, however, RPS-based analysis systems have been primarily used to investigate the physical properties of particles, i.e., size or size distribution30, and studies associated with RPS characterization of the chemical properties of analytes, i.e., surface charge or Zeta potential, are scarce. Furthermore, widely applied particle size distribution analysis by dynamic light scattering is limited in its capability to measure the distribution of zeta potential in polydisperse samples.

As a proof of concept, we fabricated a set of cylindrical pores to measure per-particle Zeta potential. Cylindrical pores allow particles to pass through with negligible radial velocity, and the rectangular pulse shape allows trivial measurement of event duration. These factors contribute to a relatively narrow distribution of translocation times, and therefore we reasoned that cylindrical pores could give a good estimate of the surface charge of individual molecules based on single pulses11.

We firstly fabricated cylindrical pores of different aspect ratio (AR: ratio of pore depth to diameter) using TPP nanolithography. It was found that a high AR cylindrical pore was essential for the generation of suitable pulses for surface charge analysis: both our simulation and experimental results suggested that cylindrical pores with low aspect ratios (AR≤3) couldn’t generate sufficiently broad resistive pulses on our measurement platform (Figure S9). Conversely, the characteristic “plunged-plain” or “rectangular” shape associated with cylindrical pores was observed during the analysis of 780 nm nanoparticles translocating through a high aspect ratio (AR=5) cylindrical pore (Figure 5a). It should be noted that for a given cylindrical pore, the magnitude of the applied voltage and the size of the analysed particles must be optimized to obtain an ideal resistive pulse shape (Figure S10 and S11). We also noticed that during our optimization processes, positive/negative dual-peak RPS events were observed (Figure S12) for some combinations of particles and pores, which may be the result of transient modulation of ionic concentrations at the entrance and exit of the pore38.

Analysis of these results revealed that nanoparticle translocations through the cylindrical pores (~ 2×10-4 m/s) were notably slower than comparable translocations through conical or 139

Chapter 5

symmetric pores (normally≥10-3 m/s), as demonstrated by Figure 5b. We hypothesized that this prolonged dwell time and thus reduced translocation velocity was due partly to a less concentrated electric field within the pore, which was partially verified by our simulation results (Figure S13), and partly to competition between electroosmotic flow and electrophoretic transport. As depicted in Figure 5c, when the analyte species carries a net charge, its translocation within the electric field is driven by electrophoresis. Meanwhile, the surface of our polymeric pore walls are also negatively charged, leading to an electroosmotically driven plug flow of the whole electrolyte. In this case, the translocation velocity (V) can be modelled as a superposition of the electrophoretic and electroosmotic velocities, as reported by Willmott, et al 39 (Eqn 3).

PR4 VEVV V () Ep hEofessure Pr  Po rePartic le 8  L Area (3)

in which VEph is the velocity of electrophoresis, VEof is the velocity of electroosmosis, VPressure is the velocity due to applied pressure, ε is the dielectric constant of the electrolyte, η is the viscosity of the electrolyte, ζPore is the Zeta potential of the pore, ζParticle is the Zeta potential of the nanoparticle, E is the applied electric field, ΔP is the introduced water pressure, R, L and Area are the dimensions of pore. Since the pore and nanoparticle are both highly negatively-charged (～48 mV for nanoparticle and ～40 mV for pore-wall) in our analysed RPS system, the forward velocity of the nanoparticle through cylindrical pores is insufficient to overcome the electroosmotic flow through the pore. To facilitate this process, a slight pressure driving force (3 cm-high water, see Figure S14) was introduced to increase the total particle velocity (Eqn 3). Additionally, as moving nanoparticles within the cylindrical- type pore are largely restricted to the centre pathways, restrictions stemming from these trajectory constraints of may also have contributed to the observed long translocation durations31.

The particle velocity equation described above suggests that if a well-understood solvent is used and the zeta potential of the pore wall is known, it should be possible to derive the zeta potential of individual particles from particle velocities. Since the pore is fabricated with a known geometry, and event duration is trivial to measure in a cylindrical pore, it follows that individual particle velocity can be obtained even without complicated analysis of the event signal by simply dividing pore depth by event duration. Noting that the pores walls were pre- 140

Chapter 5 treated with acidic solution (pH~1) to obtain reduced surface charge and thus to facilitate the translocating of nanoparticles. The per-particle dwell time was measured from single translocation events, then the Smoluchoski equation and Hagen-Poiseuille equation were used in combined form as given to derive individual particle zeta potential. Details of this calculation can be found in the Sup. File.

To demonstrate that RPS analysis using cylindrical pores made by our process can be used to differentiate chemically distinct particles of the same size, we measured the dwell times of identically sized nanoparticles that had undergone different surface charge modification processes. The modification process can accurately manipulate the number of negatively- charged groups on our nanoparticles and is presented in the Sup. File. In addition to these chemically modified particles, the zeta potential of nanoparticles with artificial protein coronas (produced through exposure to BSA protein) was also measured. Our results showed that the change in surface charge of the modified nanoparticles relative to the pore charge were reflected by variations in particle dwell times. Particles possessing higher charge will have a stronger interaction with the applied electric field, meaning their translocation velocity will differ from that of particles with lower charge and hence can be differentiated due to their shorter dwell times (Figure S15). These event durations were then used to calculate the zeta potentials of the analysed particles, as summarized in Figure 5d. A comparative study using a standard zeta-potential measurement method (ZetaSizer Nano ZS) was carried out to verify the zeta potentials of nanoparticles analysed. Zeta potentials obtained through our RPS system were well-matched with those measured by the standard method, varying by less than 10% as shown in the figure.

5.4 CONCLUSION

In summary, we have reported the fabrication of precise, well-defined miniaturized 3D pores based on computer-aided design and the TPP nanolithography. The geometry of the prepared pores, i.e., opening size, cone angle and internal profile, could be accurately controlled in 3 dimensions, allowing for the fabrication of well-defined conical, symmetric and cylindrical pores with openings as small as 600 nm. Controlled surface modifications to

141

Chapter 5 the fabricated pores, including biofunctionalization and the fine-tuning of surface charge, have been demonstrated. As a proof of concept, real-time nanoparticle position tracking and studies of particle zeta-potential distribution were performed using geometric information which would be difficult or impossible to accurately obtain in pores fabricated by traditional techniques. Precise control over cone angle allowed us to measure the relationship between increasing cone angle and background current and shown to correlate well with predictions from finite-element models. We investigated the RPS performance of TPP-generated cis- conical pores, trans-conical, and symmetrical pores through which we found that symmetrical pores could be successfully used for real-time electrical tracking of single nanoparticle position. Finally, cylindrical pores of precisely known depth were explored as a tool for the investigation of the zeta potentials of analysed nanoparticles. The TPP- generated pores provide low signal noise which will enable a number of novel RPS experiments by allowing the geometry of pores to be specified in advance and reproduced with a high degree of accuracy. Furthermore, by exploiting the known geometry of these pores for RPS, it is possible to directly measure nanoparticle translocation dynamics and the particle characteristics that influence them. Such pores have potentials for developing new technologies beyond RPS, such as techniques for the manipulation and selective modification of single (bio)-nanoparticles.

142

Chapter 5

REFERENCE

(1) Zhou, K.; Li, L.; Tan, Z.; Zlotnick, A.; Jacobson, S. C. Characterization of hepatitis B virus capsids by resistive-pulse sensing. J. Am. Chem. Soc. 2011, 133, 1618-1621. (2) Anderson, W.; Kozak, D.; Coleman, V. A.; Jamting, A. K.; Trau, M. A Comparative Study of Submicron Particle Sizing Platforms: Accuracy, Precision and Resolution Analysis of Polydisperse Particle Size distributions. J. Colloid. Interface Sci. 2013, 405, 322-330. (3) Traversi, F.; Raillon, C.; Benameur, S. M.; Liu, K.; Khlybov, S.; Tosun, M.; Krasnozhon, D.; Kis, A.; Radenovic, A. Detecting the translocation of DNA through a nanopore using graphene nanoribbons. Nat. Nanotechnol. 2013, 8, 939-945. (4) Davenport, M.; Healy, K.; Pevarnik, M.; Teslich, N.; Cabrini, S.; Morrison, A. P.; Siwy, Z. S.; Letant, S. E. The role of pore geometry in single nanoparticle detection. ACS Nano 2012, 6, 8366-8380. (5) Meng-Yue, W.; Ralph, M. M. S.; Mathijs, Z.; Ulrike, Z.; Diego, K.; Philip, E. B.; Nynke, H. D.; Cees, D.; Henny, W. Z. Control of Shape and Material Composition of Solid-State Nanopores. Nano. Lett. 2009, 9. (6) Yinghua, Q.; Preston, H.; Crystal, Y.; Henriette, E. B.; Matthew, S.; Hong, W.; Dmitriy, M.; Maria, G.; Maria Eugenia, T.-M.; Arnout, I.; Siwy, Z. S. Pores with Longitudinal Irregularities Distinguish Objects by Shape. ACS Nano 2015, 9, 4390-4397. (7) Ramirez, P.; Apel, P. Y.; Cervera, J.; Mafe, S. Pore structure and function of synthetic nanopores with fixed charges: tip shape and rectification properties. Nanotechnology 2008, 19, 315707. (8) Zhang, Z.; Xie, G.; Xiao, K.; Kong, X. Y.; Li, P.; Tian, Y.; Wen, L.; Jiang, L. Asymmetric Multifunctional Heterogeneous Membranes for pH- and Temperature-Cooperative Smart Ion Transport Modulation. Adv. Mater. 2016, 28, 9613-9619. (9) Darby Kozak; Will Anderson; Robert Vogel; Shaun Chen; Fiach Antaw; Trau, M. Simultaneous Size and ζ-Potential Measurements of Individual Nanoparticles in Dispersion Using Size-Tunable Pore Sensors. ACS Nano 2012, 6, 6990–6997. (10) Somerville, J. A.; Willmott, G. R.; Eldridge, J.; Griffiths, M.; McGrath, K. M. Size and charge characterisation of a submicrometre oil-in-water emulsion using resistive pulse sensing with tunable pores. J Colloid. Interface Sci. 2013, 394, 243-251. (11) Innes, L. M.; Chen, C. H.; Schiel, M.; Pevarnik, M.; Haurais, F.; Toimil-Molares, M. E.; Vlassiouk, I.; Theogarajan, L.; Siwy, Z. S. Velocity profiles in pores with undulating

143

Chapter 5 opening diameter and their importance for resistive-pulse experiments. Anal.Chem. 2014, 86, 10445-10453. (12) Macazo, F. C.; White, R. J. Bioinspired Protein Channel-Based Scanning Ion Conductance Microscopy (Bio-SICM) for Simultaneous Conductance and Specific Molecular Imaging. J. Am. Chem. Soc. 2016, 138, 2793-2801. (13) Huang, X.; Kong, X.-Y.; Wen, L.; Jiang, L. Bioinspired Ionic Diodes: From Unipolar to Bipolar. Adv. Funct. Mater. 2018, 1801079. (14) Sanford, M. S.; Günter, B. A protein-conducting channel in the endoplasmic reticulum. Cell 1991, 65, 371-380. (15) Smeets, R. M.; Keyser, U. F.; Dekker, N. H.; Dekker, C. Noise in solid-state nanopores. Proc. Natl. Acad. Sci. USA 2008, 105, 417-421. (16) Ding, H.; Zhang, Q.; Gu, Z.; Gu, M. 3D computer-aided nanoprinting for solid-state nanopores. Nanoscale Horiz. 2018, 3, 312-316. (17) Liao, C.; Anderson, W.; Antaw, F.; Trau, M. Maskless 3D Ablation of Precise Microhole Structures in Plastics Using Femtosecond Laser Pulses. ACS Appl. Mater. Interfaces 2018, 10, 4315-4323. (18) Blasco, E.; Muller, J.; Muller, P.; Trouillet, V.; Schon, M.; Scherer, T.; Barner-Kowollik, C.; Wegener, M. Fabrication of Conductive 3D Gold-Containing Microstructures via Direct Laser Writing. Adv. Mater. 2016, 28, 3592-3595. (19) Yee, D. W.; Schulz, M. D.; Grubbs, R. H.; Greer, J. R. Functionalized 3D Architected Materials via Thiol-Michael Addition and Two-Photon Lithography. Adv. Mater. 2017, 29, 1605293. (20) Zhang, Y. L.; Chen, Q. D.; Xia, H.; Sun, H. B. Designable 3D nanofabrication by femtosecond laser direct writing. Nano Today 2010, 5, 435-448. (21) Wenhui Zhou; Stephen M. Kuebler; Kevin L. Braun; Tianyue Yu; J. Kevin Cammack; Christopher K. Ober; Joseph W. Perry; Marder, S. R. An Efficient Two-Photon-Generated Photoacid Applied to Positive-Tone 3D Microfabrication. Science 2002, 296, 1106-1109. (22) Powell, M. R.; Vlassiouk, I.; Martens, C.; Siwy, Z. S. Nonequilibrium 1/f noise in rectifying nanopores. Phys. Rev. Lett. 2009, 103, 248104. (23) Powell, M. R.; Sa, N.; Davenport, M.; Healy, K.; Vlassiouk, I.; Létant, S. E.; Baker, L. A.; Siwy, Z. S. Noise Properties of Rectifying Nanopores. J. Phys. Chem. C 2011, 115, 8775- 8783. (24) Powell, M. R.; Martens, C.; Siwy, Z. S. Asymmetric properties of ion current 1/f noise in conically shaped nanopores. Chem. Phys. 2010, 375, 529-535.

144

Chapter 5

(25) Zhang, M.; Hou, X.; Wang, J.; Tian, Y.; Fan, X.; Zhai, J.; Jiang, L. Light and pH cooperative nanofluidic diode using a spiropyran-functionalized single nanochannel. Adv. Mater. 2012, 24, 2424-2428. (26) Lundqvist, M.; Stigler, J.; Elia, G.; Lynch, I.; Cedervall, T.; Dawson, K. A. Nanoparticle size and surface properties determine the protein corona with possible implications for biological impacts. Proc. Natl. Acad. Sci. USA 2008, 105, 14265-14270. (27) Tseng, S.; Lin, S.-C.; Lin, C.-Y.; Hsu, J.-P. Influences of Cone Angle and Surface Charge Density on the Ion Current Rectification Behavior of a Conical Nanopore. J. Phys. Chem. C 2016, 120, 25620-25627. (28) Pino, P. d.; Pelaz, B.; Zhang, Q.; Maffre, P.; Nienhaus, G. U.; Parak, W. J. Protein corona formation around nanoparticles – from the past to the future. Mater. Horiz. 2014, 1, 301-313. (29) Smeets, R. M.; Dekker, N. H.; Dekker, C. Low-frequency noise in solid-state nanopores. Nanotechnology 2009, 20, 095501. (30) Kozak, D.; Anderson, W.; Vogel, R.; Trau, M. Advances in Resistive Pulse Sensors: Devices bridging the void between molecular and microscopic detection. Nano Today 2011, 6, 531-545. (31) Menestrina, J.; Yang, C.; Schiel, M.; Vlassiouk, I.; Siwy, Z. S. Charged Particles Modulate Local Ionic Concentrations and Cause Formation of Positive Peaks in Resistive- Pulse-Based Detection. J. Phys. Chem. C 2014, 118, 2391-2398. (32) Matthew, P.; Ken, H.; Maria Eugenia, T.-M.; Alan, M.; Sonia, E. L.; Zuzanna, S. S. Polystyrene Particles Reveal Pore Substructure As They Translocate. ACS Nano 2012, 6, 7295–7302. (33) Matthew, P.; Matthew, S.; Keiichi, Y.; Ivan, V. V.; Jasmine, S. K.; Kenneth, J. S.; Zuzanna, S. S. Particle Deformation and Concentration Polarization in Electroosmotic Transport of Hydrogels through Pores. ACS Nano 2013, 7, 3720–3728. (34) Chaimanatsakun, A.; Japrung, D.; Pongprayoon, P. Multiscale simulation studies of geometrical effects on solution transport through nanopores. Mol. Simulat. 2017, 44, 12-20. (35) W. J. Lan; White, S. Diffusional Motion of a Particle Translocating through a Nanopore. ACS Nano 2012, 6, 1757-1765. (36) Arjmandi, N.; Van Roy, W.; Lagae, L.; Borghs, G. Measuring the electric charge and zeta potential of nanometer-sized objects using pyramidal-shaped nanopores. Anal. Chem. 2012, 84, 8490-8496.

145

Chapter 5

(37) Eldridge, J. A.; Willmott, G. R.; Anderson, W.; Vogel, R. Nanoparticle zeta-potential measurements using tunable resistive pulse sensing with variable pressure. J Colloid. Interface Sci. 2014, 429, 45-52. (38) Cabello-Aguilar, S.; Abou Chaaya, A.; Picaud, F.; Bechelany, M.; Pochat-Bohatier, C.; Yesylevskyy, S.; Kraszewski, S.; Bechelany, M. C.; Rossignol, F.; Balanzat, E.; Janot, J. M.; Miele, P.; Dejardin, P.; Balme, S. Experimental and simulation studies of unusual current blockade induced by translocation of small oxidized PEG through a single nanopore. Phys. Chem. Chem. Phys. 2014, 16, 17883-17892. (39) Willmott, G. R.; Vogel, R.; Yu, S. S.; Groenewegen, L. G.; Roberts, G. S.; Kozak, D.; Anderson, W.; Trau, M. Use of tunable nanopore blockade rates to investigate colloidal dispersions. J. Phys. Condens. Matter. 2010, 22, 454116.

146

Chapter 5

SUPPLEMENTAL INFORMATION

Contents Experimental Details Simulation Protocols Analysis of Position vs Time Profile Surface Modification of Pore Wall 1: Pore walls with tailored negative surface charge 2: Pore walls with bio-functionalizable BSA protein Surface Modification of Nanoparticle’s Surface Charge 1: Nanoparticles with reduced negative surface charge 2: Nanoparticles modified with BSA protein corona Calculation of nanoparticles’ zeta potential

Figures & Tables

Figure S1. UV-absorption spectroscopy Figure S2. Surface roughness induced noise behaviour Table 1. Surface charges of modified pores Figure S3. SEM images of pores Figure S4. FTIR characterizations of IP-S resist Figure S5. AFM characterization of resist polymer film Figure S6. RPS results for Cis-conical pore Figure S7. SEM images of blocked pores Figure S8. SEM view (cross-section) of symmetrical pore Figure S9. The geometrical effects on cylindrical pore Figure S10. Voltage effects on cylindrical pore Figure S11. Size effects of nanoparticles passing through pore Figure S12. Dual-peak signals in RPS analysis Figure S13. FEM simulation analysis of electric field distributions Figure S14. Water pressure setting-up Figure S15. Dwell-time distributions in cylindrical pore Table S2. Calculation example of zeta potential using cylindrical pore

147

Chapter 5

Simulation Protocol

Finite element method (FEM) simulations were performed to obtain normalized background resistance for pores with different tilt angles, to obtain resistance-vs-particle-position profiles for cylindrical pores of different aspect ratios, and to obtain resistance-vs-particle-position profiles for the four types of pore disclosed in this paper (cis-conical, trans-conical, cylindrical and symmetrical). In the latter case, individual heatmaps of the electric field strength at the pore entrances were also generated.

Resistance values were obtained by solving for Possion’s equation for the electric field for each geometry at a test voltage of 1 volt across the pore, then obtaining the flux through the positive electrode and dividing that by the test voltage to obtain resistance. Resistance-vs- particle-position profiles were obtained by calculating the resistance of geometries containing a non-conducting sphere at 100 equidistant points along the central axis of the pore. All resistance profiles are normalized by dividing by the resistance of an unoccupied pore, and thus are independent of simulated media conductivity. In all resistance profiles, negative positions are within the pore structure whilst positive positions are outside.

Pore geometries for the variable-tilt-angle and variable-aspect-ratio experiments were generated procedurally, whilst the geometries of the four pore types were derived from the original design files used for 3D printing.

The variable-aspect-ratio simulations used pores with a radius of 1.5 µm, and aspect ratios of 0.5, 1.0, 1.5 and 2.0. The reservoirs adjacent to the pore had a radius of 10 µm and a length of 10um. Resistance-vs-particle-position profiles were generated using a 1µm test particle. The centre of the cylindrical pore was placed at the 0 µm intersect.

The variable-pore-angle simulations used pores with a radius of 1.5 µm, a length of 6 µm, and angles of 0, 15, 20, 30 and 45 degrees. The reservoirs were 10 µm long and had a radius of 10 µm. Resistance-vs-particle-position profiles were generated using a 600nm test particle. The small opening of the pore was placed at the 0um intersect.

The four pore types were simulated using the design files used for the 3D printing of experimental pores. The Cis-conical and Symmetrical pores had a radius of 1.5 µm, whilst

148

Chapter 5 the Trans-conical pore had a width of 1.9 µm and the Cylindrical pore had a width of 2 µm. Pores were simulated in a 200 µm by 200 µm cylinder, with the point of tightest constriction in the pore placed at the 0um intersect.

These models do not incorporate off-axis particle translocation, but experimental data suggests that off-axis translocations do not affect the pulse shape significantly enough to invalidate their use as a comparative tool. It is assumed that the pore membrane and particle are both perfect insulators, and it is assumed that the relaxation time of the system is significantly faster than the motion of the particle through the pore, so that each particle position can be simulated independent of prior particle motion.

All FEM simulations were performed using FENICS 2018.1.0 (installed via the 2018.1.0.r3 FENICS Docker image, with ezdxf 0.9 – required to parse pore geometry files – installed via pip after creation of the container), and python code used to implement these is available as Supplementary Information: As the pores used in this study are axisymmetric, 2D axisymmetric simulations were performed to reduce the computational requirements of the simulation. Heatmaps of electric field strength were generated from full FEM simulation results of the potential field by using ParaView 5.4.1 to calculate derivatives and visualize results.

Analysis of Position vs Time Profile

In the graph Figure 4e, we analysed the position vs time profile in three different types of pores. To obtain such result, we devise following combined experimental and simulated approach: First, event detection was performed on raw resistive pulse data from all three pores (Cis- conical, trans-conical, and symmetrical). Events were detected using a window method: A window was slid across the data until the centre value of the window was also the window’s minimum current value, then the difference between the maximum and minimum current (approximately equal to resistive pulse height) was compared to a threshold value (0.25 nA for cis- and trans- pores, 1.0 nA for symmetrical pores) and accepted if the event exceeded this threshold. The two edge values of the window were also checked for proximity to the maximum current value, if either edge value was further than 0.1 nA from this value the

149

Chapter 5 event was rejected. This additional step rejected spurious events due to temporary pore blockages.

Second, the events and position-resistance profiles (from the FEM simulation conducted previously) were normalized. Events were made proportional to resistance by taking their reciprocal, then events and profiles were both normalized by subtracting and scaling such that the minimum value of each event/profile was zero and the maximum value was one.

Finally, each profile was split into pre- and post-translocation regions, and each region was fit with a monotonic curve. This curve was inverted to give a function that translated normalized resistance to position, and this function was applied to the average of all events for the relevant pore to obtain the average nanoparticle position versus time during translocation.

Surface Modification of Pore Wall

1: Pore walls with tailored negative surface charge

The here proposed surface modification strategy was attempted to achieve pore wall surface with controlled negative surface charge. The underlying mechanism of surface modification is based on the protonation/de-protonation process as shown below:

푅퐶푂푂퐻 ⇄ 퐻 + + 푅퐶푂 푂 − Acid Dissociation Constant:

The surface of resist is highly negatively charged, consisting weak acids groups, like carboxyl group, -COOH, as verified by FTIR characterizations (Figure S4). Theoretically and experimentally, we found that the number of negatively charged groups will reduced as we introduce acidic hydrogen proton into the system, thus allowing us to modify the surface properties of the TPP-generated pores.

In experiment, we developed a soaking-type modification protocol to achieve chemically distinct pore wall surface:

150

Chapter 5

③ Perform surface charge ① Acid Solution ② Rinse sample Soaking (~5 hrs) & Dried measurement in 1X PBS solution

Cycled for 3~4 times

① Prepare 3 types of 1X PBS solutions; namely, original type (pH ~7), 3 acidic types (pH 1, pH 2, and pH 3) by adding HCl solution into the original type. ② Transfer the TPP-generated polymeric pores into the acidic PBS solutions for ~5hrs, dried the samples; then repeated the same surface treatment process for at least 3 times. ③ Performed the measurements of surface charge of (modified)pore wall using Surpass Analyzer system.

As summarized in Table S1, we can achieve manipulable pore wall surface with different surface charges.

2: Pore walls with bio-functionalizable BSA protein

The preparation of BSA solution: 0.5mg BSA powder (Lyophilized powder, Sigma Aldrich) was dissolved by 100 mL PBS solution (1X) to generate to a final concentration of 5.0 mg/mL.

Absorption of BSA protein to polymeric pore wall: Immerse the TPP-generated pores into the prepared BSA solution for 6 hrs, allowing BSA protein molecules to absorb onto the pore wall surface via the electrostatic force/ Van der Waals forces. The BSA solution treated pore 151

Chapter 5 structures then was rinsed with DI water and PBS solution to thoroughly remove extra un- anchored BSA molecules. These bio-functionalized pores then were dried and stored for the surface charge measurements.

Surface Modification of Nanoparticle’s Surface Charge

1: Nanoparticles with reduced negative surface charge

HCl solutions (pH=2 & pH=3) were prepared for the modification of nanoparticles (960 nm). Transfer 10 µL prepared HCl solutions into 10 µL raw nanoparticles solutions. Mix the solutions and incubate for ~3 hrs, at 40 ℃. Under the protonation reaction, the negatively charged acetate groups anchored on pore wall surface were partially combine with the hydrogen proton, leading to modified nanoparticles (modified NP-1 & modified NP-2) with reduced surface charge. These modified nanoparticle raw samples then diluted with PBS solution to form modified nanoparticle solution for RPS analysis.

2: Nanoparticles modified with BSA protein corona

1mg BSA powder was dissolved by 100 mL PBS solution (1X) to generate to a final concentration of 10 mg/mL. Use pipette to transfer 10 µL the prepared BSA solution into 10 µL nanoparticle raw solutions (960 nm, Bangs Laboratories) for 6 hrs’ incubation. Next step, this modified nanoparticle raw sample solutions were diluted by 1000 times using PBS solution for RPS analysis.

Calculation of nanoparticles’ zeta potential

By performing RPS analysis of nanoparticles using cylindrical pores, we can investigate the surface charge of analysed nanoparticles. As show in Figure 5c, the total effective velocity represents the added results of velocity of electrophoresis, velocity of electrophoresis and velocity of pressure (Vtotal = Veph + Veof + Vpressure). Analysed by the Smoluchoski equation (Eqn 2)[1] and Hagen Poiseuille equation[2, 3] (Eqn 3), we can calculate the surface charge of nanoparticles: 152

Chapter 5

(1) VtotalDwell L t  /

      r 0 (2) e E 

8LQ  P (3)  r 4

 (4) V  V V () E 1 EphEof  PoreParticle

PR4 VE 2  VV EphEofessu  e V P r r ()  PoreParticl e (5) 8 LA ear

In which L is the length of pore, tDwell is the translocating time of analysed nanoparticles, E is the electric field in the cylindrical pore,  r is relative permittivity water at 20 C,  0 is permittivity of free space,  is dynamic viscosity of water at 25 C,  is the zeta potential of pore wall, P is the addressed water pressure, r is the radius of cylindrical pore, Q is the volumetric flow rate, E is the applied electric field, R, L and Area are the dimensions of pore. By inputting the known/obtained data into these equations, we can calculate zeta potential (approximate to surface charge) of nanoparticles of analysed. To further valid the feasibility of our proposed RPS system for zeta potential characterizations, we then compared our calculated results with those obtained from conventional standardized approach (ZetaSizer Nano ZS).

153

Chapter 5

Figure S1. UV-absorption spectroscopy of high viscous IP-S resist.

Figure S2. Noise behaviour of TPP-generated polymeric pores with different surface roughness properties. Pores with a rougher wall surface lead to higher noise level in RPS characterizations

154

Chapter 5

Table S1 Pore wall with reduced negative surface charge using the proposed modifying technique.

Group pH ZP(FM) ZP(HS) Average Results [mV] [mV] Zeta[mV]

-40.62 -38.27 1 2.962 -38.38 -37.45 Retained Modifying Effects -39.95 Decreased ~16.50% at pH 3 soaking -38.29 -38.15 modifying -41.51 -38.72 -25.31 -23.24 2 2.021 -27.28 -24.80 Retained Modifying Effects -26.50 Decreased ~45.75% at pH 2 soaking -26.43 -25.63 modifying -26.96 -24.37 -19.67 -17.79 3 0.918 -22.42 -19.56 Retained Modifying Effects -21.52 Decreased ~55.46% at pH 1 soaking -21.39 -19.29 modifying -22.60 -20.48

155

Chapter 5

Figure S3. Under optimal processing conditions, TPP fabrication system can generate controllable opening hole with the smallest size down to ∽ Ø 600 nm (Left figure). As we further decrease the diameter size (∽ Ø 500 nm) of opening hole, the geometric properties of prepared hole become out of control (Right figure).

Figure S4. (a) FTIR graph of the liquid IP-S resist. (b) FTIR graph of the solid film made of cross-linked IP-S resist.

156

Chapter 5

Figure S5. Surface roughness of pore wall surface characterized by AFM

(b) 40

Frequency/

0 500 550 600 650 Size/nm

Figure S6. (a) Overall current tracking curve of RPS analysis of 530nm nanoparticles using cis-conical pores. (b) Size distribution of RPS characterized nanoparticle particles. Model size: 573 nm.

157

Chapter 5

Figure S7. SEM images show the pore blockage at a lower operating voltage/ larger analysed nanoparticles. Scale bar: 2µm

Symmetrical Central

0.2 nA 0.2 5 ms

3 µm

Figure S8. Magnified symmetrical RPS signal as a single nanoparticle (600 nm) translocate through the symmetrical pore alone the central axis.

158

Chapter 5

Figure S9. The effects of geometrical designs of cylindrical pores on RPS analysis. (a) The schematic figure shows the aspect ratio (AR) of the cylindrical pore designs. (b) Simulated current blockade curves for the cylindrical pore designs with different AR, ranging from AR: 1 to AR: 7. (c) Experimental finger-printing signal for the RPS analysis of 960 nm nanoparticles using AR: 1 designed cylindrical pore. (d) Experimental finger-printing signal for the RPS analysis of 960 nm nanoparticles using AR: 3 designed cylindrical pore. (e) Experimental finger-printing signal for the RPS analysis of 960 nm nanoparticles using AR: 5 designed cylindrical pore. (e). Experimental finger-printing signal for the RPS analysis of 960 nm nanoparticles using AR: 7 designed cylindrical pore.

159

Chapter 5

Figure S10. Finger-printing RPS analysis of 960 nanoparticles translocate through the cylindrical pore (AR: 5) at varying voltages, ranging from 0.3 V to 1 V.

Figure S11. Finger-printing RPS analysis of different-sized nanoparticles (400 nm, 600 nm, 780 nm and 960 nm) translocating through the cylindrical pore (AR: 5) at 0.8V.

160

Chapter 5

Figure S12. Dual-peak signal observed in RPS analysis using the cylindrical pores.

161

Chapter 5

162

Chapter 5

Figure S13. Finite element method (FEM) simulations of the electric field profile alone four types of pores: (a) Cis-conical pore; (b) Trans-conical pore; (c) Symmetrical pore; (d) Cylindrical pore. The simulation processes were performed using FENICS 2018.1.0 platform, with ParaView 5.4.1 to calculate derivatives and visualize results.

3cm Water

Figure S14. Setting up shows the use of added water (3 cm in height) to facilitate the translocation process of nanoparticles.

163

Chapter 5

Figure S15. Dwell time distributions of distinctly-modified nanoparticles characterized by cylindrical pores (AR: 5), at 3 cm height water pressure and 0.8 V.

164

Chapter 5

Table S2 Summarized calculation process for the Zeta potential of nanoparticle.

Calculations of ζparticle

Time 1.00E-01 s Time of pulse Total velocity (V = V + V + total eph eof V total 2.00E-04 d(m)/time(s) Vpressure)

Veph Velocity of electrophoresis minus the 1.10E-04 other velocities

Voltage -0.8 V Voltage across pore E -4.00E+04 V/m Electric field in pore 2 4 −1 −3 Er 8.85E-12 A ⋅s ⋅kg ⋅m Permittivity of free space E0 80.1 Relative permittivity water at 20 ℃ -1 -1 kg.m .s Dynamic viscosity of water at η 0.0089 P 25 °C ζ (1000* V η) / E *Er* E0 particle -34.55 mV eph*

Calculations of Veof

ζpore 2.00E-02 mv

Veof -6.37E-05 m/s Smoluchoski Eqn: (E *Er* E0* ζpore)/ η

Calculation of VPressure ∆P 35 Pa Applied with 3.5 cm high water L 2.00E-05 m Length of pore -1 -1 kg.m .s (µ is the same as η in µ 0.0089 P smoluchowski) R 2.50E-06 m Radius of pore Q 3.02E-15 m^3/s Volumetric flow rate Area 1.96E-11 m^2 Area of pore 4 Hagen Poiseuille Eqn: (π*∆P*R )/8 VPressure 1.54E-04 m/s *µ*L*Area

165

Chapter 5

REFERENCE

1. von Smoluchowski, M. (1903). "Contribution à la théorie de l'endosmose électrique et de quelques phénomènes corrélatifs". Bull. Int. Acad. Sci. Cracovie. 184.

2. Boussinesq, Joseph. "Mémoire sur l’influence des Frottements dans les Mouvements Réguliers des Fluids." J. Math. Pures Appl 13.2 (1868): 377-424.

3. Proudman, J. "IV. Notes on the motion of viscous liquids in channels." The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science 28.163 (1914): 30- 36.

166

Chapter 6

Candidate’s contribution to the authorship

I declared that I have obtained permission from all co-authors to include following publication directly into this chapter.

Caizhi Liao, Fiach Antaw, Alain Wuethrich, Matt Trau. “Stacked Dual-Pore Architecture Deciphering and Manipulating the Dynamics of Individual Single Nanoparticle”. Advanced Materials Technologies, 2020, 2000701. - Incorporated as Chapter 6.

Contributor Statement of contribution

Conception and design (80%) Experiments (90%) Caizhi Liao (Candidate) Analysis and interpretation (70%) Drafting and production (60%)

Conception and design (10%) Experiments (10%) Fiach Antaw Analysis and interpretation (10%) Drafting and production (10%) Conception and design (5%) Alain Wuethrich Analysis and interpretation (5%) Drafting and production (25%) Conception and design (5%) Analysis and interpretation (5%) Matt Trau Drafting and production (5%)

167

Chapter 6

Stacked Dual-Pore Architecture Deciphering and Manipulating Dynamics of Individual Nanoparticle

Summary The achievement of Milestones (Milestone 1 to 3) inspired further investigation of this field by the creation of novel dual-pore constructs for RPS analysis. RPS using a solid dual-pore structure holds great potential for studying the translocation dynamics of single molecular analytes. Current dual-pore designs, normally with two pores suited in a parallel manner (horizontally aligned), can only be explored for analysing molecules with long strands, e.g, DNA molecule or RNA molecule. Based on the results of previous Milestones, we learned that the TPP nanolithography system could generate accurate 3D miniaturized pores for high-performance RPS analysis. To fulfil the aims proposed by Milestone 4, this Chapter explores the preparation of vertically aligned stacked dual-pore architectures, a brand-new type of RPS pore that affords to perform nanoparticle analysis that hardly can be realized by traditional RPS platforms: The exact motion of individual nanoparticles can be examined by the finger-printing RPS signal. Also, this vertically stacked dual-pore system can be deployed as a modulator to control the nanoparticle events, and even trap/release single nanoparticle within a confined space.

Chapter 6 is based on a published paper:

 Caizhi Liao, Fiach Antaw, Alain Wuethrich, Matt Trau. “Stacked Dual-Pore Architecture Deciphering and Manipulating the Dynamics of Individual Nanoparticle” Advanced Materials Technology, 2020. 2000701.

168

Chapter 6

ABSTRACT

Striving to understand the behaviour of nanoscale objects in confined spaces represents a great challenge. To address this issue, numerous single molecular strategies have been proposed. In terms of label-free routes, resistive-pulse sensing (RPS) has emerged as one of the most promising methods. However, available RPS system is not capable enough to simultaneously sense and control single nanoparticle in motion. Here, we propose the use of two photon nanolithography for creating vertically stacked RPS dual-pore system. The prepared dual-pore structures afford to finely decipher the molecular dynamics in terms of transport velocities and inter-particle coupling effects. By further optimizing the 3D pore geometry, the analysing capability of these dual-pore system can be further extended for single nanoparticle trapping-releasing study.

6.1 INTRODUCTION

Reliable monitoring and tracking of individual nanoparticle is a gateway to understand physical, chemical, and biological processes in complex systems. Such measurements unveil critical information on molecular interactions, the property of local environment, and even the activity of composing elements1, 2. To perform this type of study, one need to not only detect the nanoparticles, but track the dynamics of individual entity in selected environment. Tracking nanoparticle in restrained space is difficult, however, as it requires fast measurement rate and long observation time3, 4.

Fluorescence microscopy based techniques serve as the most-commonly explored methods in biophysical tracking studies5, 6. However, the measurement has been impeded by the longest observation time, which is limited by the photobleaching, and the fastest measurement rate, which is limited by the low inherent fluorescent emission rate. Even without photobleaching, the observation time is still hindered by the dwell time the labelled nanoparticles spend in the illuminated volume7. Dynamic light scattering (DLS) scheme is a promising alternative for tracking particles on short-time scale8, 9. Regarding its energy- conserving process, DLS offers a nearly infinite photon budget that enables fast measurement rates; nonetheless, the established DLS illuminates a large background scattering, making it hard to eliminate the interference scattering from the surface defects

169

Chapter 6 or discernible contaminates. Again, the observation time is limited by the diffusion of nanoparticles in the imaging focal plane.

More recently, nanoparticle tracking analysis (NTA) is developed to detect and visualise particles based Brownian motion10. In NTA, individual particle positional changes are tracked in two dimensions, a process leading to the determination of particle diffusion and subsequently the particle size or particle concentration11, 12. While NTA is innovative, its potential to probe the dynamics of individual nanoparticles is somewhat limited, particular for the nanoparticles under external force to pass through or dwell within a confined space. Therefore, a clear-cut understanding of the dynamics of individual NPs in restricted space still needs further exploration.

Rapid advances in label-free strategies are transforming the way sensing systems are studied, especially with a view on developing novel single molecular sensing devices1, 13. Due to their inherently simple operating principle, which is based on recording the changes in the ionic current through a miniaturized pore that is separated by two electrolyte-filled reservoirs, resistive-pulse sensing (RPS) pore sensors have been gaining prominence for a wide range of analysis: from sequencing of DNA/RNA to unravelling the underlying mechanisms of bio-systems14, 15. In particular, considerable effects have been devoted to the RPS investigation of nanoparticles, which provides profound implications for uses in analysing the size, structure and surface feature16, 17. However, the state-of-the-art of RPS pore technology faces significant challenges partly due to their limited control over particle transport, inadequate ability to confine and study individual particle over long time scales, and poor performance in single molecular tracking18, 19.

Promisingly, RPS system with dual-pore architecture has been rising as an enabling platform to simultaneously sense and track single molecules20-22. This unique two-pore design enables repeated measurements on the same elements by distinct sensing units. Compared with its single-pore based counterpart system, RPS measurements using dual- pore structure can help to gain detailed information on the dynamics of translocating molecules and even perform single molecular trapping or manipulating3, 23, 24. Typically formed in parallel designs, these two pores can be realized through either chemical etching

170

Chapter 6 or glass pulling4, 25. However, because of current manufacturing constraints, the achievable 3D geometrical design of the two-pore system is quite limited26. Furthermore, the sensing region of current two-pore designs is opened to the external surroundings, making them extremely sensitive to environmental noises. With regard to these factors, such dual-pore architecture proved to be useful only for the analysis of long-strand type molecules, such as DNA and RNA4.

Here we propose a novel solution to these challenges based on the two-photon nanolithography, a versatile platform affords to generate a brand-new type of vertically stacked dual-pore systems, which allow individual nanoparticle to be accurately analysed within a miniaturized confined space. Our study shows that this unique dual-pore system can simultaneously detect and track individual mobile, un-labelled nanoparticles. In characterization, each nanoparticle transporting through the dual-pore system could generate a unique compound dual-peak signal, which provides information to decipher the dynamic motion behaviours of translocating nanoparticles. By modifying the 3D geometry, the prepared TRANS-conical dual-pore system can elucidate the nanoparticle transport motion behaviour in terms of inter-particle effects, including coupling and de-coupling, and transport dynamics, including trapping and releasing. The analysing capability of proposed dual-pore structure can further be stretched by introducing wave-shape sub-pore structures. To increase the single nanoparticle capture rate, we adjusted the 3D internal geometry design of the dual-pore system, allowing us to analyse the nanoparticles in a multidimensional way. In short, the ease-of-use and performance of this methodology support its potential for widespread applications in single particle analysis and manipulation.

6.2 RESULTS AND DISCUSSION

The major fabrication process-two-photon polymerization(TPP)-is performed on Nanoscribe Photonic Professional system, as shown in Figure 1a. Details on TPP working mechanism and structural formation are introduced in Supporting Information. The plastic substrate loaded with resist (IP-S, Figure 1b) is placed on a 3D piezoelectric stage. During TPP, femtosecond laser pulses (80 MHz, 100 fs) emitted from a 780 nm Ti–sapphire laser source are utilized to trigger the reactions that cross-link the polymeric resist material. Two structural processing settings—slicing distance and hatching distance (Figure 1c)—were 171

Chapter 6 combined to regulate the exposure volume of material. After the development process washing away unsolidified photoresist, the formed dual-pore structure is integrated with the resistive-pulse sensing platform (q-Nano). Once an appropriate voltage is addressed (Figure 1d), the nanoparticle translocates through the dual-pore system, resulting in unique dual-peak signals that can be utilized to simultaneously sense physical/chemical property and analyse translocating dynamics of nanoparticle. Importantly, the 3D geometric designs of dual-pore system, including individual pore size and hollow chamber space, play a vital in the performance of structure prepared, as evidenced by Figure S5 and S6.

Figure 1. (a) Schematic graph of two photon nanolithography fabrication platform. The SEM characterized structures are CIS-conical dual pore and TRANS-conical dual pore, respectively. (b) The UV spectroscopy of the utilized IP resist. (c) The hatching and slicing patterns for dual-pore structure formation. (d) Schematic experimental setup of the nanoparticle analysis systems. Electrodes are addressed on two sides of the dual-pore structure.

172

Chapter 6

173

Chapter 6

Figure 2. (a) Current-voltage (I-V) curves of a single 3µm-sized pore, a single 5µm-sized pore, and a dual-pore system consisting of a 3µm-sized entry pore and a 5µm-sized exit pore. The voltage ranges from -1V to 1V, with an increment step 0.1V. (b) Comparison results of noise behaviour of a single pore (3µm-sized ) and a dual-pore system (3µm-5µm- sized), respectively. The working voltage: 0.5V. (c) Three consecutive dual-peak signals generated using CIS-dual pore system (See inserted SEM image, scale bar: 3µm). The compound signal consists a large entry blockade and a small exit blockade. (d) Magnitude correlations between large-peak entry signal and small-peak signal in series of compound events. (e) The magnitude distribution and dwell time distribution of 330 nanoparticle sample translocating through a CIS-3µm-5µm dual pore systems. Working voltage: 0.5V.

To characterize the formed dual-pore structure, current-voltage (I-V) measurements were performed (Figure 2a, Red curve). The current linearly increased with the voltage without showing any rectification behaviour as the size of dual-pore is orders of magnitude larger than that of electrical double layer. Compared with single pore structure, the dual-pore system shows an increased total resistance because of the introduction of second pore into the electric circuit. To further interpret this result, I-V measurements from two individual single-pore structures (3µm and 5µm) recorded and graphed (Figure 2a, Blue curve and Green curve). The conductance calculated from the linear region (±1V) of I-V curves measured for each pore configuration were GS3=97.5 ± 8.3 nS, GS5=273.3 ± 19.6 nS and

GDual=74.4 ± 6.7 nS, respectively. Importantly, the conductance of measurement dual-pore structure is well matched with the total conductance of the two individual single pore structures in series (1/ GDual ≅ 1/ GS3+ 1/ GS5).

In RPS implementations, the temporal resolution for particle’ translocation is ultimately restrained by the level of ionic current noise: the 1/ f spectral character can impose a profound impact on detection27, 28. To look into this issue, the normalized current power spectral density of a dual-pore system was examined in comparison with that of its single- pore counterpart under same operating conditions, as shown in Figure 2b. In low frequency range (< 103 Hz), the dual-pore architecture exhibits significantly higher noise behaviour

174

Chapter 6 level. As the frequency magnitude extended over 104 Hz, 1/f spectral of both pore structures started to converge to each other, showing a similar noise behaviour pattern. This observation is well agreed with noise magnitude levels recorded in characterizations: 28±9 nA for dual-pore system, 16±7 nA for single-pore structure.

We then attempted to perform the nanoparticles analysis using the RPS platform based on a cis-type dual-pore structure. Driven by addressed electric field, nanoparticles translocate through the dual-pore structure, generating a unique dual-peak compound signal that consists of a large-peak blockade ( Entry signal through the first small cis-conical pore) and a small-peak blockade (Exit signal through the second large cis-conical pore), Figure 2c. In a rare case, a unique compound signal with positively-peaked blockade was observed (Figure S7), a phenomenon that might be attributed to the locally-aggregated ions profile inside the dual-pore system29.

If the nanoparticle transition remains intact through the dual-pore structure, a constant correlation coefficient is expected for the two individual current blockades(e.g., entry and exit signals) for each compound dual-peak readout. To validate this hypothesis, we calculated the ratio of large current blockade to small one in a compound signal. As shown in Figure 2d, the correlation ratio is narrowly clustered to the value of 7.0, which is well matched with the theoretical value (~7.7) derived from the Kozak model30. For details in theoretical model, please refer to the supplemental file. With the aim of better understanding this dual-peak compound signal, we collectively analysed the results using pCLAMP 10 suite. The full spectrum of peak signals is segmented into two mutually- exclusive, well-separated clusters: The green circle for small-peak blockade, the red circle for the large-peak blockade (Figure 2e). It is worthwhile to underscore the magnitude distribution result of the full collection of current blockade signals (Figure 2e, in Yellow). Impressively, the frequency of two types of peaks is evenly split, with a ratio of large-peak number to small-peak number close to 51% to 49%. The slight difference between these values might comes either from artefact signal processing or from “divergent-types” signal that not showing the complete pair feature (Figure S8 to S15). Significantly, this half-half distribution pattern is in agreement with the nature of observed dual-peak signal, in which a large blockade is paired with a small blockade. 175

Chapter 6

Figure 3, (a) The schematic picture defines the time of flight (TOF) period for the translocating nanoparticle. (b) The voltage effects on TOF. (c) The TOF distribution histogram. 15% signals signify extended TOF characteristics. (d) The velocity profile of the nanoparticle translocating through the dual pore structure.

Promisingly, our stacked dual-pore system has the capability of performing time of flight (TOF) analysis, which is essential in electrophoretic measurements, in particular enabling access to the dynamics of each translocating nanoparticle. In an experiment, the TOF 176

Chapter 6 interval begins at the moment a nanoparticle exits the first pore generating the large blockade and ends when the same nanoparticle enters the second pore generating the small blockade, see graph Figure 3a. After translocating the first pore, the nanoparticle can be electrically driven toward the second pore under the force of electrophoretic effects. Noting that the nanoparticle travels along or close the central axis of dual-pore structure, as proved 31 by our previous study . In this case, the nanoparticle exits the first pore at t1=52.870s and arrives at the second pore at t2=52.886s. The TOF for this particular event pair is t2 - t1=16ms, a critical parameter allowing accurate zeta-potential analysis of an individual nanoparticle. Example of how to perform zeta-potential analysis can be found in Supplemental File part.

It has been reported that TOF is highly depended on the magnitude of applied voltage, i.e., TOF ∝ 1/V21, 23. To elucidate this phenomenon, we measured the TOF values at different operating voltages. We find that increasing the voltage leads to a lower TOF, with a decreasing tendency resembles the one derived from theory model (Figure 3b). This result also helps to clarify that nanoparticle motion inside this dual-pore structure is dominated by the electrophoretic forces instead of random diffusion. Next, the complete set of TOF value is collectively analysed and graphed in Figure 3c. Surprisingly, an extended TOF region is observed, accounting for some 15% of the total number of trajectories. Many factors might contribute to a prolonged TOF, including the translocating nanoparticle larger in size or less surface-charged, or that the particle might follow an off-centre motion trajectory. Moreover, the pore-to-pore travel time of current spike pair can be used to investigate the pore-to-pore transient velocity profile, Figure 3d. The average measured translocation speed, v=0.63µm·ms-1, is in line with the one calculated from the cylindrical pore reported before (Chapter 5).

177

Chapter 6

Figure 4. (a) Dual peak compound signal generated as nanoparticle translocates through a TRANs-5µm-3µm dual pore system. Red star: Small-peak entry signal. Blue triangle: large- peak exit signal. Inserted image: SEM of cross-sectioned TRANS-dual-pore. Scale bar: 5 µm. (b) Finger-printing signal signifies a small-sized nanoparticle surpassing a normal-size nanoparticle in translocating through the dual-pore structure. (c) Finger-printing signal signifies the nanoparticle sticks to the entry-pore for a short period in translocating through the dual-pore structure. (d) The collective portion of normal dual-peak signals in data sets

178

Chapter 6 generated from Cis-dual structure (~ 99%) and Trans-dual structure (~ 68%). (e) The frequency of different types of abnormal multi-peak signals, including 2 entry-2 exit signal, capture signal and release signal, etc,. (f) 2 entry-2 exit signal signifies 2 nanoparticles consecutively translocate through the dual-pore system. (g) 2 entry-2 adjacent exit signal signifies two nanoparticles bounded together when exiting the dual-pore system. (h) 2 adjacent entry-2 adjacent exit signal signifies two nanoparticles bounded together when translocating through the whole dual-pore system. (i) ) 2 adjacent entry-2 exit signal signifies two nanoparticles bounded together when entering the dual-pore system, but exiting the system separately. (j) 2 entry-2 exit signal, in which 1 entry signal overlaps with 1 exit signal, signifies the first nanoparticle is exiting the dual-pore system while the second one is entering the dual-pore system. (k) A 2 entry-1 exit signal signifies a nanoparticle capture event. (m) An 1 entry-2 exit signal signifies a nanoparticle release event.

Steady distribution profile of ions alone the RPS pore could be strikingly shaped by the geometry of responding region, leading to a preferential direction for the transport of nanoparticles across the structure26, 32, 33. To get an insight into this asymmetric transport behaviour, we created and analysed a mirrored architecture of the cis-dual-pore: a Trans- dual-pore structure that starts with a large trans-conical pore and ends with a small trans- conical pore. Driven by electric force, a nanoparticle firstly passes through the large pore, generating a small current spike (Entry signal, red-starred). Then similar to the motion observed in the cis-dual-pore case, the nanoparticle travels through the hollow chamber space (TOF process), until this nanoparticle finishes its journey by translocating through the second pore structure, Figure 4a. It is unsurprising to observe such echoed finger-printing dual-peak signal with regard to the 3D designs of the pores being characterized. If the size of nanoparticle reduces to some extent, the paired nature of the compound signal disappears. Only one signal spike (exit signal), the spike generated from the nanoparticle exit the second small pore, can be observed (Figure 4b). The size of nanoparticle is too small relation to the that of the entry large pore, with a size ratio less than 1:10, resulting in no observable entry small-spike signal. Occasionally, the nanoparticle could dwell in the entry pore region for an extended period (Figure 4c), for which might be provoked by the surface interactions between the translocating nanoparticle and the functionalizable pore-wall.

179

Chapter 6

Furthermore, a direct collective comparison of the data obtained in cis-dual-pore and trans- dual-pore, for the same RPS system revealed several key insights into the geometry- induced effects on translocation pattern. For the nanoparticle analysis using cis-dual-pore, largest portion of signals (Yellow, ~99%) was the normal pair-type dual-peak signal, while for the analysis using trans-dual-pore, the ratio of the normal type compound signal droped to 68%, Figure 4d. Promisingly, the remaining atypical signals, as categorized in Figure 4e, can be used to decipher the complex motional dynamics of individual nanoparticles through the dual-pore architecture, particularly for the scenarios that twin nanoparticles simultaneously interact with the dual-pore system. Among these atypical signals, 52% are the double-entry double-exit four current spike signals (Figure 4f to 4j), from which we can assess the unique motion behaviour as two nanoparticles concurrently travel through the dual-pore.

The Figure 4f type signal, the most-frequent four-peak type signal, can be interpreted as a process in which, before a particle already in the central part of the structure reaches the exit, a second particle enters. Then the first nanoparticle exits the system through the small pore, followed by the second nanoparticle. Depending on the coupling effects of twin nanoparticles, several sibling signals of Figure 4f type were oberved. In Figure 4g, the two large-exit spikes converged into each other, showing that the two nanoparticles separately enter into the dual-pore system, then for some reason, the two nanoparticles start to bundle together and exit the structure near simultaneously. The bundling of two nanoparticles could also occur before sliding into the trans-dual-pore structure. The bundled twin nanoparticles could act as a single unit directly passing through the whole dual-pore (Figure 4h) or the particles may decouple within the chamber space between the two pores and then decoupled to exit the dual-pore system separately (Figure 4i). The forces behind this decoupling of bundled twin nanoparticles might come from surface charge distinction or physical fluid disturbance. More rarely, we can catch the exact moment that one nanoparticle enters the dual-pore system as the other one exits the dual-pore system (Figure 4j). Significantly, our developed dual-pore system allows to trap and release single nanoparticles, both of which could be accurately reflected by the finger-print blended signals. Figure 4k shows a typical nanoparticle trap event wherein the nanoparticle enters the dual- pore structure and stays inside without escaping. While for the Figure 4m type signal, a 180

Chapter 6 nanoparticle is released from the chamber space and exits the system after one other nanoparticle passes through the whole dual-pore system.

Figure 5. (a) A 3-entry-3-exit signal signifies that three nanoparticles together consecutively pass through the dual-pore system. (b) Wave-shape finger-printing signal generated using dual-pore system incorporated with wave-shape sub-pores structures. Inserted images: The design model and the experimental dual-pore structure, respectively. (c) Finger-printing signals generated using dual-pore system incorporated with three wave-shape sub-pores inside the hollow chamber space. Inserted images: The design model and the experimental dual-pore structure, respectively. (d) Overall curve consists of trapping-event signal (blue curve without black square) and releasing-event signal (blue curve with black square). (e)

181

Chapter 6

Design model and experimental dual-pore structure. (f) The frequency distribution of the trapping-event and releasing-event of two distinct dual pore structure: Blue histogram for tuned dual-pore system with extended exiting pore length, while yellow histogram for normal TRANS-conical dual-pore system. (g) A complete entry-exit signal set signifies that a nanoparticle translocates through the whole dual-pore structure.

Besides single nanoparticle and twin nanoparticles, a triple-nanoparticle group can also be investigated through the dual-pore structure. For instance, Figure 5a shows an example that can be interpreted as how three individual nanoparticles sequentially translocating through the dual-pore structure. Before any exit event (large spike, blue-triangle) occured, all three nanoparticles had already travelled into the dual-pore system through first large entry pore structure, as indicated by the three distinct preceding small spikes. Thereafter, these three nanoparticles exit the system one by one.

To further stretch the analysing capability of the proposed dual-pore structure, we introduced

5 regulated wave-shape sub-pore constructions (ØMin=3 µm) into the entry small pore and 4 similar structures (ØMin=5 µm) into the exit large pore. As nanoparticle travelled through this dual-pore, multi-wave shape signals were captured (Figure 5b). The number of wave-type spikes is aligned with the number of wave sub-pore constructs incorporated inside each pore. Surprisingly, the wave pattern of the entry signal is deviated from the one generated from theoretical model, see Figure S17. The magnitude of each wave spike embedded within the entry signal gradually reduces, and a mirrored positive spike follows its negative counterpart. The introduction of wave sub-pore might distort the electric field distribution along the entry pore, resulting in unevenly allotted potential for each sub-wave-pore and consequently, the decreasing spikes within the entry signal as nanoparticle pass through the entry pore structure. Also, the wave-shape sub-pore structures could alter the localised ions concentration profile, forming ions rich/depletion regions. When a nanoparticle passes through these regions, abnormally shaped, such as the positive one, can be observed. These irregularities in signal shape become less perceptible in the exit signal incorporating 4 sub-wave spikes, due to the fact that the enlarged size of the exit pore restricts the geometry-induced effects.

182

Chapter 6

As aforementioned, nanoparticle flies through the free chamber space between two- composing pores allows TOF analysis, which is critical for dynamic study of individual nanoparticle. It therefore is necessary to examine the role of chamber space design in nanoparticle characterizations. In study, we added 3 wave-shape sub-pore structures

(ØMin=5 µm) into the chamber space. As nanoparticle translocated through this modulated- type of dual-pore system, we recorded a more delicate dual-peak mode signal, with 3 minor- wave signals occupy the TOF signal session (Figure 5c). In short, our fabrication scheme allows us to accurately prepare and modify the dual-pore system with tuned features, delivering results for extended RPS implementations.

Spatial confinement from the nano- to the micro-scale is ubiquitous in nature34. Methods to examine the behaviour of nanoscale objects in the confined domains have received substantial attention. Here, we strive to tackle the issue of single nanoparticle trapping and releasing by using the proposed dual-pore structure. As discussed above, the cis- and trans- dual-pore system affords to capture and release nanoparticle individually. However, even for the more-capable trans-dual-pore construct, the combined occurrence rate of nanoparticle trapping/releasing is still low, with a possibility less than 10%. Our previous study implicates that nanoparticles’ motion dynamics can be radically modulated by varying the aspect-ratio (AR) of RPS pores. In attempt to increase the single nanoparticle capture rate, we therefore adjust the 3D internal geometry design of the dual-pore system: The pore depth of the exit pore (high AR cylindrical type) is extended to 5 times of that of entry pore (low AR cylindrical type), see Figure 5e. With voltage addressed, single nanoparticle interacts with the dual-pore structure in three different ways: I. Single nanoparticle only translocates through entry pore and be captured inside the chamber without exiting (Figure 5d. blue spike without black square appended or Figure 5g, In-type single signal); II. Single nanoparticle directly translocate through the whole dual-pore structure (Figure 5g, In-Out pair signal); III. The already trapped single nanoparticle released through the exit pore (Figure 5d. blue spike with black square appended or Figure 5g, Out-type single signal). Promisingly, this tuned dual-pore configuration can be expected to increase the motion resistivity of nanoparticle alone the exit pore part, thus helping to enhance the nanoparticle trapping events, as summarized in Figure 5f. By collectively analysing the signals, we find that the single nanoparticle trapping events increase to ~30%, and single nanoparticle releasing events to ~13%, over four times higher than that achieved through the normal conical dual-pore designs. The positive peaks for both Entry and Exit type signals unfold

183

Chapter 6 that un-uniform ions concentration profiles, including ions depletion and ions rich, form inside this dual-pore and interplay with the nanoparticle nearby. Additionally, single nanoparticle translocates the long cylindrical exit pore could generate extended dwell-time, which is useful for investigating the physical (e.g., size and size distribution), chemical (e.g., surface charge and functionality), and dynamic (e.g., velocity and trajectory) features of the translocating objects.

6.3 CONCLUSION To conclude, for the first time, two photon nanolithography technique was presented as a robust fabrication scheme for the accurate preparation of 3D stacked dual-pore architects. Classical dual-pore structures, including CIS-dual and TRANS-dual, and derivative dual- pore structures, including dual system incorporated with sub-pores and dual system enhanced with extended exiting designs, were successfully realized for reliable measurements. This novel family of dual-pore constructs is well suited to controllably investigate and manipulate nano-objects on the single particle/molecule level. Our CIS dual- pore platform not only allows for physical analysing of translocating nanoparticles but also unveils interesting dynamics of traveling nanoscale object. By performing the unique time- of-flight (TOF) analysis, we can finely decipher the molecular velocity profiles in terms of transport dynamics, which we corroborate with the theoretical modelling and FEniCS simulations. Promisingly, TOF measurement can help to identify chemical or physical changes the nano-object during traversal of the dual-pore system. We then fabricated trans dual-pore system, which can be exploited for the investigation of inter-nanoparticle coupling effects. The generated finger-printing compound signals can accurately reveal the nanoparticle interaction effects, including coupling and de-coupling, and transport dynamics, including trapping and releasing. Importantly, the analysing capability of this dual-pore family can further be stretched by introducing wave-shape sub-pore structures. To increase the single nanoparticle capture rate, we then adjusted the 3D internal geometry design of the dual-pore system, allowing us to electrically control the loading and manipulation of single nanoparticle into a tailored microenvironment and for the study of molecular dynamics under spatial confinement. In short, the ease-of-use and performance of this strategy support its potential for widespread applications in single particle analysis and manipulation.

184

Chapter 6

REFERENCE

1. Kozak, D.; Anderson, W.; Vogel, R.; Trau, M. Nano Today 2011, 6, (5), 531-545. 2. Liao, C.; Anderson, W.; Antaw, F.; Trau, M. ACS Appl. Mater. Interfaces 2018, 10, (4), 4315-4323. 3. Cadinu, P.; Paulose Nadappuram, B.; Lee, D. J.; Sze, J. Y. Y.; Campolo, G.; Zhang, Y.; Shevchuk, A.; Ladame, S.; Albrecht, T.; Korchev, Y.; Ivanov, A. P.; Edel, J. B. Nano Lett 2017, 17, (10), 6376-6384. 4. Zhang, Y.; Liu, X.; Zhao, Y.; Yu, J. K.; Reisner, W.; Dunbar, W. B. Small 2018, 14, (47), e1801890. 5. Lichtman, J. W.; Conchello, J. A. Nat Methods 2005, 2, (12), 910-9. 6. Huang, B.; Bates, M.; Zhuang, X. Annu Rev Biochem 2009, 78, 993-1016. 7. J. E. Hobbie; R. J. Daley; Jasper, S. Appl. Environ. Microbiol 1977, 33, (5), 1225- 1228. 8. Frisken, B. J. Appl. Opt 2001, 40, (24), 4087-4092. 9. Pecora, R. J. Nanoparticle Res 2000, 2, 123–131. 10. Filipe, V.; Hawe, A.; Jiskoot, W. Pharm Res 2010, 27, (5), 796-810. 11. Dragovic, R. A.; Gardiner, C.; Brooks, A. S.; Tannetta, D. S.; Ferguson, D. J.; Hole, P.; Carr, B.; Redman, C. W.; Harris, A. L.; Dobson, P. J.; Harrison, P.; Sargent, I. L. Nanomedicine 2011, 7, (6), 780-8. 12. Sokolova, V.; Ludwig, A. K.; Hornung, S.; Rotan, O.; Horn, P. A.; Epple, M.; Giebel, B. Colloids Surf B Biointerfaces 2011, 87, (1), 146-50. 13. Darby Kozak; Will Anderson; Robert Vogel; Shaun Chen; Fiach Antaw; Trau, M. ACS Nano 2012, 6, (8), 6990–6997. 14. Howorka, S.; Siwy, Z. Chem. Soc. Rev. 2009, 38, (8), 2360-2384. 15. Blundell, E. L. C. J.; Mayne, L. J.; Billinge, E. R.; Platt, M. Anal. Methods 2015, 7, (17), 7055-7066. 16. Billinge, E. R.; Muzard, J.; Platt, M. Nanomater. Nanosci. 2013, 1, (1), 1. 17. Matthew, P.; Ken, H.; Maria Eugenia, T.-M.; Alan, M.; Sonia, E. L.; Zuzanna, S. S. ACS Nano 2012, 6, (8), 7295–7302.

185

Chapter 6

18. Matthew, P.; Matthew, S.; Keiichi, Y.; Ivan, V. V.; Jasmine, S. K.; Kenneth, J. S.; Zuzanna, S. S. ACS Nano 2013, 7, (4), 3720–3728. 19. Willmott, G. R.; Vogel, R.; Yu, S. S.; Groenewegen, L. G.; Roberts, G. S.; Kozak, D.; Anderson, W.; Trau, M. J. Phys. Condens. Matter. 2010, 22, (45), 454116. 20. Pedone, D.; Langecker, M.; Abstreiter, G.; Rant, U. Nano Lett 2011, 11, (4), 1561-7. 21. Cadinu, P.; Campolo, G.; Pud, S.; Yang, W.; Edel, J. B.; Dekker, C.; Ivanov, A. P. Nano Lett 2018, 18, (4), 2738-2745. 22. Liu, X.; Zhang, Y.; Nagel, R.; Reisner, W.; Dunbar, W. B. Small 2019, 15, (30), e1901704. 23. Langecker, M.; Pedone, D.; Simmel, F. C.; Rant, U. Nano Lett 2011, 11, (11), 5002- 7. 24. Zhou, J.; Kondylis, P.; Haywood, D. G.; Harms, Z. D.; Lee, L. S.; Zlotnick, A.; Jacobson, S. C. Anal Chem 2018, 90, (12), 7267-7274. 25. Pud, S.; Chao, S. H.; Belkin, M.; Verschueren, D.; Huijben, T.; van Engelenburg, C.; Dekker, C.; Aksimentiev, A. Nano Lett 2016, 16, (12), 8021-8028. 26. Davenport, M.; Healy, K.; Pevarnik, M.; Teslich, N.; Cabrini, S.; Morrison, A. P.; Siwy, Z. S.; Letant, S. E. ACS Nano 2012, 6, (9), 8366-8380. 27. Powell, M. R.; Martens, C.; Siwy, Z. S. Chem. Phys. 2010, 375, (2-3), 529-535. 28. Powell, M. R.; Vlassiouk, I.; Martens, C.; Siwy, Z. S. Phys. Rev. Lett. 2009, 103, (24), 248104. 29. Menestrina, J.; Yang, C.; Schiel, M.; Vlassiouk, I.; Siwy, Z. S. J. Phys. Chem. C 2014, 118, (5), 2391-2398. 30. Kozak, D.; Anderson, W.; Grevett, M.; Trau, M. J. Phys. Chem. C 2012, 116, (15), 8554-8561. 31. Liao, C.; Anderson, W.; Antaw, F.; Trau, M. ACS Omega 2019, 4, (1), 1401-1409. 32. Qiu, Y.; Vlassiouk, I.; Hinkle, P.; Toimil-Molares, M. E.; Levine, A. J.; Siwy, Z. S. ACS Nano 2016, 10, (3), 3509-3517. 33. Yinghua, Q.; Preston, H.; Crystal, Y.; Henriette, E. B.; Matthew, S.; Hong, W.; Dmitriy, M.; Maria, G.; Maria Eugenia, T.-M.; Arnout, I.; Siwy, Z. S. ACS Nano 2015, 9, (4), 4390- 4397. 34. Yusko, E. C.; Bruhn, B. R.; Eggenberger, O. M.; Houghtaling, J.; Rollings, R. C.; Walsh, N. C.; Nandivada, S.; Pindrus, M.; Hall, A. R.; Sept, D.; Li, J.; Kalonia, D. S.; Mayer, M. Nat. Nanotechnol 2017, 12, (4), 360-367.

186

Chapter 6

SUPPLEMENTAL INFORMATION

Two Photon Polymerization (TPP) In this 3D nanolithography platform, two-photon Polymerization (TPP) is triggered by the two-photon absorption (TPA), in which the absorption of two photons with identical/different frequencies excite a molecule from a ground state to an excited state. Typically, the resist material system for TPP reaction is formulated by a mixture of bulk photosensitive molecules, photosensitiser and photoinitiators. In preparation, the photosensitiser chromophore molecule can be excited by the absorption of two photons as the intensed laser pulse beams focused within the resist system, leading to the generation of fluorescent light in the UV-vis range and susequently improve the two-photon activation process. Then, photoinitiators with high photochemical activity absorb the emitted fluorescent light from photosensitiser chromophore and generate radicals (Initiation) to activate the TPP reaaction. Be more specific, these radicals serve as the activator react with monomers or oligomers, producing monomer radicals to further expand in a chain reaction (Propagation) until two radicals meet (Termination) to stop.

Materials Two types of substrates , indium tin oxide (ITO)-coated PET film with protective film on both sides (200 µm thick, Shenzhen Jemstone Technology Co., Ltd.) and ITO-coated glass (500 µm thick, Sigma-Aldrich Co.), were used as received to load the 3D dual-pore constructs. Milli-Q (Merck Millihole) deionized (DI) water was used for the preparation of phosphate- buffered saline (1X PBS) solution, nanoparticle solution, and ultrasonic treatment medium. PBS tablet (Sigma-Aldrich Co.) and DI water were used for the preparation of PBS solution. A series of functionalized polystyrene (PS) nanoparticles, with a diameter size range from 114 to 980 nm, were purchased from Bangs Laboratories, Inc. To prepare suitable sample solutions for single nanopartilce analysis using prepared dual-pore system, raw PS nanoparticle centralized solution were diluted with 1× PBS solution, with a dilution ratio of

187

Chapter 6

1:1000 to 1:100 000, depending on characterization setting-ups. IP-s resist mateial purchased from Nanoscribe is used for the preparation of 3D dual-pore structures.

Dual-Pore System Fabrication To generate 3D dual-pore system for single nanoparticle analysis, we need to firstly design an optimize the geometrically designs in computer aided design (CAD) programs. In this work, Autodesk AutoCAD program was deployed to design and process the 3D dual-pore constructs, including CIS-dual, TRANS-dual, and a family of derivative dual-pore systems (e.g., Dual-pore system incorporated with sub-pores and dual-pores incorporated with extended exiting pore). These AutoCAD generated designs in the format of standard triangle language (.STL) files then were imported into the DeScribe software (Nanoscribe GmbH, Germany), in which structural processing parameters (e.g., slicing and hatching) were set for TPP fabrication process. Lastly, this type of processed design file in the format of general writing language (.GWL) file was imported into the Nanoscribe Photonic Professional GT system (Nanoscribe GmbH, Germany) for the customized DLW preparation of 3D dual-pore constructs.

For the Nanoscribe commerical platform, a FemtoFiber Pro NIR laser source (Toptica Photonics AG) is equipped to generate focused laser beams, with a pulse duration (τ) ≈ 100 fs and a repetition rate (f) of 80 MHz at 780 nm (λ). As the transmittance (T) of laser beam through the IP-S resist at 780 nm was high ( ∼0.75), the laser pulse beam can easily penerate through the resist and accurately focus the spot area without inducing any diffraction. A Zeiss Plan-Apochromat 25× 0.8 NA oil DIC M27 objective is used to perform the focus action. Importantly, an integrated charge-coupled device camera (CCD camera) is incorporated to visulaize the formation of 3D dual pore system in a real-time mode.

Our fabricated 3D dual-pore systems was loaded on the ITO-coated PET substrate (∼1 cm × 1 cm). As a first step, we need to create micro-sized hole (∼µm in diameter) in the PET substrate, by using a novel type of laser ablation technique we previously reported. After the laser ablation process, PET substrates were dropped with appropriate amount (50µ ML) of IP-S resist to fully cover the micro-hole, and then were mounted onto the 3D-

188

Chapter 6 controlled XYZ piezo stage for the DLW-TPP process. Once the interface is defined, the Nanoscribe system can automatically proceed to prepare the 3D dual-pore structures. The area exposed to laser pulse beam will be cross-linked to form the solid structures. These formed structures then were transferred into the development system (PGEMA/IPA) to fully remove the un-cross-linked resist materials trapped/anchored within the formed dual-pore structures. After this step, the obtained structures can be collected for further chacterizations and analysis.

Characterization A JEOL IT-300 scanning electron microscope (SEM, JEOL Ltd.) was used for the surface morphological characterizations of these prepared 3D dual-pore structures. To enhance the surface contrast in SEM characterization, the surfaces of prepared dual-pore system were precoated with platinum film using a JEOL metal coating system. Deposition of platinum films with different thicknesses (∼20nm to 50 nm) on the surface was performed using the metal coating system, by controlling the deposition time and input current. A Nikon Eclipse Ni–U optical system equipped with a Nikon Plan Fluor lens was utitilzed for the optical imaging of dual-pore structures. The surface roughness of the prepared 3D structures was analyzed by Dektak 150 stylus profiler (Bruker Corp.).

Resistive-Pulse Sensing Based Single Nanoparticle Analysis A qNano system (Izon Science, New Zealand) was utilized for the resistive-pulse analysis of single nanoparticle. To perform the characterizations, the prepared 3D dual-pore structures were integrated into the measurement chamber unit of qNano and fixed by sample holder (see Figure S6). PBS (1×, Sigma-Aldrich Co.) was used as the electrolyte for resistive-pulse analysis. For the study of the transient i–v curves of the dual-pore systems, corresponding ionic currents were recorded at each applied voltage (−1.2 to 1.2 V, 0.6 V increment for each step). In the resistive-pulse analysis of PS nanoparticles, selected types of nanoparticle solution samples (∼40 μL) were added onto the upper chamber of the measurement unit once the characterization system reached the stabilized states (with a backgroup noise less than 10 nA). Driven by the electric field, nanoparticles can translocate through the 3D dual-pore systems and generate corresponding resistive-pulse signals. These unique current blockades signal then are collectively recorded by the built-in qNano

189

Chapter 6 software. For data analysis, Axon™pCLAMP™ 10 software suite is utilized to reveal information on noise behavior of dua-pores and finger-printing resistive-pulse signals for nanoparticle translocating.

Theoretical Model For Two-Peak Corrections In measurement, the nanoparticle translocates through the dual-pore system under the drive of electric field. If the nanoparticle remians intact in passing through the dual-pore construct, a constant correlation relationship is expected for the two individual current blockades (i.e., entry and exit signals) within a compound dual-peak readout. Based on our previous work, this correlation coefficient can be easily calculated by adopting the Kozak model, as showed below:

ퟒ흆풅ퟑ 휟푹 ≃ (1) 흅푫ퟒ

ퟒ 휟푰푳 휟푹푺 푫푳 = ≃ ( ) (2) 휟푰푺 휟푹푳 푫푺

in which d is the size of translocating nanoparticle, D is the diameter size of RPS pore, DL is the diameter size of the large exiting pore and the DS is the diameter size of the small entry pore. Based on original geometrical design of our dual pore structure (DL: 5µm, DS: 3µm), the theoretical calculation result of the correlation ration between two peaks is close to 7.7.

Simulation Plots of normalized resistance versus particle position were generated for the three pore types (I: Multi-wave design in between; II: Pre-filtering design; III: Two family pore array design ) under investigation, using the FENICS FEM simulation package to solve Poison’s equation for electrostatics. The origin of the position axis is fixed at the tightest constriction of the pore, and is located at the constriction closest to the top-chamber opening for pores where multiple constrictions have the same radius. A simulated 1um particle was placed at 190

Chapter 6 different positions along the pore axis, and a resistance was calculated for each of these positions by applying a test voltage and calculating the resulting current, then applying Ohm’s law. This resistance was normalized by the resistance of the pore in the absence of a particle, making the result independent of the simulated electrolyte conductivity. Whilst this approach does not consider off-axis translocation, assumes electroneutrality and necessarily neglects dynamic effects, it replicates in silico the average pulse shape observed from experimental data and in the absence of dynamic effects allows particle position/velocity to be inferred from this pulse shape.

191

Chapter 6

Figure S1. (a). Picture of the commercial Nanoscribe Photonic Professional Platform for the two-photon polymerization (TPP) based nanolithography fabrication. (b). 3D moveable stage loading the sample. Under the irradiation of femtosecond laser source (Red in picture), resist material is cross-linked to form the dual-pore architecture designed.

Figure S2. Enhanced development system to remove the uncross-linked resist materials. (a). Sample holder for the development process generated from Mylar PET film. (b). Magnetic stirrer system with heating units controlled to improve the efficiency of development process. 192

Chapter 6

Figure S3. Mal-functional dual-pore systems formed under un-optimized fabrication setting- ups. (a). Dual-pore with pealing-off/contamination blocking the pore (External). (b). Dual- pore with contamination/uncross-linked resist material trapped inside blocking the pore (Internal). (c). Contamination alone the surface of dual-pore system distort/damage the shape of structure designed. (d). Dual-pore structure with cracks alone the edge of dual- pore constructs.

193

Chapter 6

Figure S4 (a). The dual-pore structure prepared on the PET substrate (200 µm thick, precoated with ITO). The size of PET substrate is around 1cm × 1cm. (b). Screen capture result of the resistive-pulse blockade signals as the nano-particle translocating through the dual- pore structure.

194

Chapter 6

1.06

4-4-20-Trans 3-5-5-Cis 1.04 3-5-10-Cis 3-5-15-Cis 3-5-20-Cis

1.02

Normalized Current

1.00 -40 -20 0 20 Pore Length/m

Figure S5. FEniCS simulations of dual-peak signals generated from dual-pore systems with different 3D geometric designs (entering pore diameter-exiting pore diameter-distance between two pores-pore type): Black curve: 4-4-20-Trans; Red curve: 3-5-5-Cis; Blue curve: 3-5-10-Cis; Green curve: 3-5-15-Cis; Purple curve: 3-5-20-Cis.

195

Chapter 6

Figure S6 (a). Overall current blockade signals as nanoparticles (330 nm) translocate through the TRANS-dual pore system consisting two identically-sized sub-pore structures. (b). Two consecutive compound signals generated as two nanoparticles translocate through the prepared TRANS-dual pore constructs.

196

Chapter 6

Figure S7 (a). Schematic figure of ions concentrated inside dual-pore system, through which nanoparticle is passing through. (b). Single compound dual-pore signal with positive peak (Exiting signal) as nanoparticles (330 nm) translocate through the dual pore system (5-3-5- CIS). (c). Overall current blockade signals as nanoparticles (330 nm) translocate through the dual pore system (3-5-5-TRANS). (d). Single compound dual-pore signal with positive peak (Exiting signal) as nanoparticles (330 nm) translocate through the dual pore system (3-5-5-TRANS).

197

Chapter 6

37.2

nA 37.0

Current/ 36.8

180.0 180.2 180.4 180.6 180.8 Times/s

Figure S8. Single compound dual-peak signal with extended time of flight (TOF) period.

198

Chapter 6

37.50

37.35

nA 37.20

Current/ 37.05

36.90 31.6 32.0 32.4 32.8 Times/s

Figure S9 Single compound dual-peak signal with extended time of flight (TOF) period. The current dropped a little bit during TOF as translocating nanoparticle might stick to the surface of dual-pore structure.

199

Chapter 6

Figure S10 (a). Compound dual-pore signal with smaller magnitude as smaller–sized nanoparticle translocate through the prepared dual-pore structure (CIS-Dual). (b). Nanoparticle with further shrunken size can only generate the relative-large entering signal when passing through the prepared dual-pore system (CIS-Dual).

200

Chapter 6

Figure S11 (a). Atypical 4-peak compound signal consisting two entering blockades and two exiting blockades as two nanoparticle translocate through the dual-pore structure (CIS- Dual). These two nanoparticle are passing through the dual-pore separately. (b). Atypical 4-peak compound signal consisting two entering blockades and two exiting blockades as two nanoparticle translocate through the dual-pore structure (CIS-Dual). Two coupling nanoparticles enter the dual-pore nearly simultaneously, and exits the dual-pore separately.

201

Chapter 6

37.6

36.8

36.0

Current/

35.2

134.2 134.4 134.6 134.8

Times/s

Figure S12 Irregular signal generated as debris inside the dual-pore system detached and brought disturbance to the stabilized current, causing abrupt shifts in current magnitudes.

Figure S13 Group of resist pulse signals showing that the dual-pore system has been blocked partially for a while and fully recovers as the blockage issue disappeared.

202

Chapter 6

Figure S14. (a). Atypical 4-peak compound signal consisting two entering blockades and two exiting blockades as two nanoparticle translocate through the dual-pore structure (CIS- Dual). At the moment the first nanoparticle exiting the dual-pore system, the second nanoparticle is entering the dual-pore system. (b). Atypical compound signal indicating a nanoparticle is trapped by the dual-pore system.

Figure S15. Atypical compound signal indicating the trapped nanoparticle is being released from the dual-pore system (CIS-DUAL).

203

Chapter 6

Figure S16 Half-cut cross-section view of a prepared dual-pore construct (TRANS-Dual). The red-circled area represents a debris/contamination part anchored on the surface of inside chamber surface. The detachment of debris element during measurement will cause significant changes in current signal, as showed in Figure S12 and 13. Scale bar: 5 µm.

204

Chapter 6

1.15 I II 1.10 III

1.05

1.00

Normalized Resistance Resistance Normalized -30 -20 -10 0 10 Distance/ m

Figure S17. Theoretical model for three types of modified dual-pore systems: I- Atypical compound signal indicating the trapped nanoparticle is being released from the dual-pore system (CIS-DUAL).

205

Chapter 6

Variable Value Unit Note ζ particle calculations Uses Smoluchowski equation L 5.00E-06 m Length of pore time 2.00E-02 s Time of pulse Total velocity (Vtotal = Veph + Veof + Vpressure) and subtract the values Vtotal 2.50E-04 d(m)/time(s) In the right direction of flow Veph 2.88E-04 Veph is the velocity of electrophoresis minus the other velocities Voltage -0.6 V Voltage across pore E 1.20E+05 V/m Electric field in pore Er 8.85E-12 A2⋅s4⋅kg−1⋅m−3 Permittivity of free space E0 80.1 Relative permittivity water at 20 °C η 0.0089 P kg.m-1.s-1 Dynamic viscostiy of water at 25 °C

ζparticle -30.08 mV Thing we are trying to solve

EoF Velocity Calculations Also uses Smoluchoski equation ζpore 2.00E-02 mv Veof 0.000191 m/s

Pressure-head Velocity Calculations Uses Hagen Poiseuille equation ∆P 35 Pa (3.5 cm water) L 2.00E-05 m Length of pore µ 0.0089 P kg.m-1.s-1 µ is the same as η in smoluchowski R 2.50E-06 m Radius of pore Q 3.02E-15 m^3/s Volumetric flow rate Area 1.96E-11 m^2 Area of pore

Vpressure 1.54E-04 m/s q/Area

206

Chapter 7

Conclusions & Future Work

7.1 Conclusions Reliable analysis of individual nanoparticles offers possibilities to understand the physical, chemical, and biological processes in complex systems. Due to their simple operating principle that is based on recording the changes in the ionic current, resistive-pulse sensing (RPS) based platforms have been gaining prominence for a wide range of analysis, in particular for individual nanoparticle. The performance of RPS system is strongly dependent on the miniaturized pore geometry, the preparation of which has attracted huge attentions. To remove the fabrication hurdles for accurate RPS pores, this Thesis introduces the two- photon polymerization (TPP) based platform for tailored fabrication. Promisingly, the core research being discussed throughout this Thesis will contribute to the development of whole single-molecular analysis system and help to inspire the next stages in this exciting field of research, as detailed below:

Chapter 3 realizes the invention of a maskless two-photon laser ablation technique for nano/micro-processing in plastic substrates (Milestone 1). Previously, precise micro-hole structures can be hardly realized in plastic substrates using current fabrication schemes. Here, a novel ablation process based on two-photon femtosecond (fs) laser has been developed. To accurately prepare these precise micro-hole structures, laser parameters (pulse energy, scanning speed), material aspects (substrate thickness, substrate type) and micro-hole features (hole size, geometric shape), are systematically investigated and are thoroughly discussed. The results show that this ablation process can create arbitrary shaped micro-well structures in plastics, which serve as essential substrate components to load the TPP generated micro-constructs.

Chapter 4 contributes to the preparation of accurate 3D hollow bio-micro-devices by using TPP technique (Milestone 2). 3D microstructure incorporating accessible interiors represents a versatile platform for bio-system uses. Classical fabrication techniques based on photolithography-etching processes cannot precisely control the geometric of the entire hollow 3D microstructures. The introduced new additive manufacturing system based on

207

Chapter 7

TPP enables the precise, straightforward, and customizable preparation of hollow 3D microstructure devices. To achieve this, factors governing the formation of hollow 3D biosystem micro-devices, including material composition, laser input and (post-) development treatment, have been thoroughly investigated. As a proof-of-concept, hollow 3D micropore devices are prepared and characterized for resistive-pulse analysis of nanoparticles, holding great potentials for creating robust RPS pores for various studies.

Chapter 5 strives to formulate tailored miniaturized RPS pores for robust nanoparticle analysis (Milestone 3). Internal 3D geometry of RPS pore imposes a profound impact on RPS analysis. However, current RPS pore fabrication techniques cannot accurately control the 3D inside geometry, bringing huge limitations for RPS uses. Promisingly, the proposed TPP-based nanolithography platform affords the preparation of accurate hollow miniaturized constructs. For the first time, micro-pores with precise cone angles have been successfully prepared for experimental studies. To further assess this technique, accurate 3D pores with classical internal geometric features are prepared for RPS analysis: cis- and trans-conical pores for the investigation of pore’s preferential transport capability; symmetrical pores for the electrical tracking of nanoparticle position; and cylindrical pores for the surface charge analysis of chemically distinct nanoparticles of the same size. Such TPP-generated pores have potentials for developing new technologies beyond pure RPS analysis, such as techniques for simultaneous manipulation and sensing of single (bio)-nanoparticles.

Chapter 6 further explores the creation of novel dual-pore system with extended sensing and controlling capabilities (Milestone 4). RPS represents a promising platform to understand the behaviour of translocating objects. However, conventional RPS system adopting single pore structure is not capable enough to simultaneously sense and control individual nanoparticle in motion. In this chapter, TPP platform is realized for the preparation of a vertically stacked RPS dual-pore system. These achieved dual-pore structures afford to finely decipher the molecular dynamics in terms of transport velocities and inter-particle coupling effects. By tuning the 3D internal geometric designs, this vertically stacked dual- pore system can be deployed as a modulator to control the nanoparticle events, and even trap/release single nanoparticle within a confined space.

208

Chapter 7

7.2 Future Work As summarized above, the research carried out in this thesis has significantly contributed towards advancements in the RPS analysis fields. As a powerful 3D designable nano/micro- prototyping strategy, TPP based schemes have attracted significant interests. Several cross-disciplinary research hubs across Asia, Europe and America have engaged intensively with TPP fabrication, sparking scientific collaborations between materials, chemistry, physics, engineering, and biology. However, very few groups have explored TPP platform for RPS pore fabrication. From a practical point of view, extensive evaluations of this particular field still need to be implemented to further facilitate the applications of this new technique, by carrying out research in following topics:

(1) New Materials

Most materials currently explored in TPP fabrication have been initially developed for conventional photolithographic approaches. This has led to a dearth in available TPP materials suitable for biological applications. Such stagnation of material discovery also brings tremendous opportunities for chemists to synthesize candidate materials by taking advantage of the versatile toolkit of organic chemistry. Considering the stringent requirements of bio-environments, further improvements to the material pool should be centred on the development of non-cytotoxicity biodegradable materials that enable the direct TPP fabrication of implantable bio-devices. Another material challenge is the shrinkage-induced deformation of the formed solid 3D structures occurred during development and washing step. To overcome these shrinkage issues, highly cross-linked photosensitive materials with neglectable volume changes during processing should be developed.

(2) New Structures

Up to now, the fabrication of pores was focused around RPS with single pore designs. As discussed in this Thesis, RPS system enabled by single-pore sensing components has limited utilities for certain types of analysing purposes. The demonstrated vertically stacked dual-pore constructs represent a brand new group of RPS pores for analysing the dynamic behaviour of translocating objects both collectively and individually. As a preliminary study,

209

Chapter 7 well-designed supportive studies still needed to validate the performance of the introduced dual-pore system in RPS measurements. In the future, the family of dual-pore system can be further expanded by maneuvering geometric shapes, including: pore tilt angle, pore-to- pore distance, pore tip shape and cis-trans shape reversion, etc,. Additionally, other un- explored pore shapes, regularly arranged wave shape for example, can potentially be realized through the TPP fabrication system and be utilized for designated RPS studies.

(3) New Applications

RPS system has been adopted for a wide range of analysis: from basic nanoparticle studies to complex DNA/RNA sequence examination. Although various analytes have been assessed by using RPS, these studies are primarily focused on the topic of size distributions. Due to the limited pore choice in terms of composing material and shape design, current RPS system is still not a universal approach for analysing nano/micro-objects. As discussed above, the sensing capability of RPS system can be significantly extended by adopting new materials (e.g., polymeric materials ) and new designs (e.g., dual-pore architecture) for the TPP generated 3D pores. Next, such new pores can be exploited for promising research topics, such as the investigation of molecular dynamics and the manipulation of individual small objects, that can be hardly realized through traditional RPS system. As an example, the developed RPS system could be adopted for the study of single protein aggregates by observing the folding/unfolding status transition on real-time mode, or the study of extracellular vesicle (EV) by simulating the transportation process across the blood-brain barrier (BBB).

(4) New Platforms

Integrated sensing system represents a significant strategy for the realization of next technological revolution. Currently, RPS measurement units are normally used as an individual characterization equipment for analysing the translocating objects. For analysing purposes, the RPS measurement performance can be greatly improved by combining with other examining techniques, such as micro-fluidics system and micro total analysis systems (µTAS). In such integrated sensing system, multiple investigations (identical or distinct) can

210

Chapter 7 be carried out simultaneously, dramatically improve the research efficiency in this particular field. Even more, RPS device can be integrated into the global network of machines and devices capable of interacting with each other, contributing the realization of the new technology paradigm- The Internet of Things (IoT) eco-systems.

Nevertheless, the beautiful marriage between TPP nanolithography fabrication and RPS analysis will be further developed to serve as a versatile platform for robust analysing purposes.

211