Molecular Switch for Sub-Diffraction Laser ... - ACS Publications

Jun 5, 2017 - Preparative Macromolecular Chemistry, Institut für Technische und Polymerchemie, Karlsruhe Institute of Technology (KIT),. 76128 Karlsr...
0 downloads 0 Views 3MB Size
Molecular Switch for Sub-Diffraction Laser Lithography by Photoenol Intermediate-State Cis−Trans Isomerization Patrick Mueller,†,⊥ Markus M. Zieger,‡,§ Benjamin Richter,∥ Alexander S. Quick,‡,§ Joachim Fischer,† Jonathan B. Mueller,†,⊥ Lu Zhou,†,⊥ Gerd Ulrich Nienhaus,†,⊥,# Martin Bastmeyer,∥,∇ Christopher Barner-Kowollik,*,‡,▼ and Martin Wegener*,†,⊥ †

Institute of Nanotechnology, Karlsruhe Institute of Technology (KIT), 76344 Eggenstein-Leopoldshafen, Germany Preparative Macromolecular Chemistry, Institut für Technische und Polymerchemie, Karlsruhe Institute of Technology (KIT), 76128 Karlsruhe, Germany § Institut für Biologische Grenzflächen (IBG), Karlsruhe Institute of Technology (KIT), 76344 Eggenstein-Leopoldshafen, Germany ∥ Cell- and Neurobiology, Karlsruhe Institute of Technology (KIT), 76128 Karlsruhe, Germany ⊥ Institute of Applied Physics, Karlsruhe Institute of Technology (KIT), 76128 Karlsruhe, Germany # Department of Physics, University of Illinois at Urbana−Champaign, Urbana, Illinois 61801, United States ∇ Institute of Functional Interfaces (IFG), Karlsruhe Institute of Technology (KIT), 76021 Karlsruhe, Germany ▼ School of Chemistry, Physics, and Mechanical Engineering, Queensland University of Technology (QUT), 2 George Street, Brisbane, QLD 4000, Australia ‡

S Supporting Information *

ABSTRACT: Recent developments in stimulated-emission depletion (STED) microscopy have led to a step change in the achievable resolution and allowed breaking the diffraction limit by large factors. The core principle is based on a reversible molecular switch, allowing for light-triggered activation and deactivation in combination with a laser focus that incorporates a point or line of zero intensity. In the past years, the concept has been transferred from microscopy to maskless laser lithography, namely direct laser writing (DLW), in order to overcome the diffraction limit for optical lithography. Herein, we propose and experimentally introduce a system that realizes such a molecular switch for lithography. Specifically, the population of intermediate-state photoenol isomers of α-methyl benzaldehydes generated by two-photon absorption at 700 nm fundamental wavelength can be reversibly depleted by simultaneous irradiation at 440 nm, suppressing the subsequent Diels−Alder cycloaddition reaction which constitutes the chemical core of the writing process. We demonstrate the potential of the proposed mechanism for STED-inspired DLW by covalently functionalizing the surface of glass substrates via the photoenol-driven STED-inspired process exploiting reversible photoenol activation with a polymerization initiator. Subsequently, macromolecules are grown from the functionalized areas and the spatially coded glass slides are characterized by atomic-force microscopy. Our approach allows lines with a full-width-at-halfmaximum of down to 60 nm and line gratings with a lateral resolution of 100 nm to be written, both surpassing the diffraction limit. KEYWORDS: direct laser writing, molecular switch, cis−trans isomerization, photoenol, STED lithography, super-resolution lithography, surface functionalization amously formulated by Abbe in 1873,1 the diffraction barrier withstood for more than a century as a widely accepted theoretical limit for optical microscopy. Only upon using concepts of modern physics unknown to Abbe, the diffraction barrier could finally be broken with the advent of stimulated-emission depletion (STED) microscopy.2 Alongside

F

© 2017 American Chemical Society

with stochastic localization techniques (commonly referred to as STORM3 and PALM4), the field of super-resolution Received: April 24, 2017 Accepted: June 5, 2017 Published: June 5, 2017 6396

DOI: 10.1021/acsnano.7b02820 ACS Nano 2017, 11, 6396−6403

Article

www.acsnano.org

Article

ACS Nano Scheme 1. Proposed Reaction Mechanism

microscopy opened up and found numerous applications, especially in the life sciences. The concept underlying STED microscopy includes two components: First, a molecule that can be brought into a fluorescent (“on”) state by one laser color, the excitation laser, and that can be brought back into a nonfluorescent (“off”) state by a second laser color, the depletion laser. Second, the depletion laser focus needs to be in a special spatial shape. The electric field has a zero crossing, or equally, the intensity is touching zero, at the same position where the excitation focus has its intensity maximum. By increasing the depletion laser power, more and more molecules will be switched into (and effectively kept in) the “off” state, with the exception of the molecules located at the point of zero depletion intensity. In principle, the remaining molecules in the visible “on” state can be confined to an arbitrarily narrow, i.e., diffraction-unlimited, spot by increasing the depletion laser power. For optical microscopy, both components have been realized in several ways: Switching off the fluorescence of a molecule, i.e., depleting the fluorescent state, was originally achieved by stimulated-emission depletion, but has later also been realized by photoinduced intersystem crossing into dark triplet states or by switching between bistable molecular states, such as cis−trans isomers of fluorescent dyes or proteins (see, e.g., the review by Hell5). Points or lines of zero intensity in the depletion beam have been realized by employing different kinds of dielectric or birefringent phase masks6 or by interference patterns commonly referred to as “structured illumination”.7,8 It was suggested theoretically9 and demonstrated experimentally10,11 that this concept can also be transferred from microscopy to maskless laser lithography, namely direct laser writing (DLW). Here, tightly focused laser beams induce multiphoton absorption, typically in negative-tone photoresists, causing photopolymerization of, e.g., acrylates. In its singlebeam diffraction-limited form, DLW technology allows for 3D printing of arbitrary structures on the micro- and nanoscale, which has led to numerous applications in different fields.12,13 As it remains a demanding task to fabricate very small structures on the nanoscale with this technique, considerable effort has been devoted toward developing STED-inspired schemes in order to increase the achievable resolution. Obviously, the main challenge hereby is the identification of suitable materials that offer a molecular switch to optically inhibit polymerization.14 Several different mechanisms have been reported in literature:

into an excited state from where they are likely to follow a pathway leading to their decay into radicals that can initiate free-radical polymerization of monomers. This mechanism is often used in regular, diffraction-limited DLW. It was shown, that the excited state of some initiator molecules, e.g., diethylamino-3-thenoylcoumarin (DETC), can be efficiently depleted by stimulated emission with focused light at 532 nm wavelength, suppressing radical formation. In a photoresist composed of DETC and an acrylate monomer, 2D and 3D structures were fabricated with subdiffraction-limit resolution in the lateral and axial directions.15 The achievable resolution is limited by parasitic absorption of the depletion laser that kicks in with rising laser power.16 (2) Depletion by excited-state-absorption: Similar to the previous case, other photoinitiators can absorb photons of the depletion light in their excited state by excitedstate-absorption and thus transit into higher triplet states from where they either relax nonradiatively into the ground state or form nonreactive species. In both cases, the population of the excited state is again depleted and thus radical polymerization is suppressed. In the literature, this mechanism was used to fabricate subdiffraction structures with malachite green carbinol base17 and isopropyl thioxanthone.14 Long lifetimes of the intermediate triplet states as well as irreversible consumption of initiators might pose limitations to the accessible writing speeds. (3) Photoinhibitor lithography: Instead of affecting the initiating species, the depletion laser here activates inhibitor molecules that are additionally mixed into the photoresist. Upon activation, these molecules decay into species that inhibit the free-radical polymerization by radical scavenging or chain termination. Fabrication of subdiffraction structures has been reported.18,19 However, a conceptual drawback is the irreversible consumption of the inhibitors. With increasing depletion laser power Pdepl the resolution scaling in both STED-inspired microscopy and lithography is proportional to 1/√Pdepl.20 Therefore, to increase the resolution by a factor of 10, the depletion laser power must be increased by a factor of 100. Hence, possible parasitic twophoton writing of the depletion beam increases by a factor of 10000. To avoid such undesired writing by the depletion beam, efficient photochemical switching mechanisms are highly desirable. Thus, the “on” state of a molecule should have a long lifetime, ideally much larger than the typical nanosecond

(1) Stimulated emission of two-photon-excited photoinitiator molecules: By two-photon absorption at wavelengths around 800 nm, photoinitiator molecules can be brought 6397

DOI: 10.1021/acsnano.7b02820 ACS Nano 2017, 11, 6396−6403

Article

ACS Nano

Figure 1. Sample preparation involves a four-step process. (a) Plasma-activated glass samples are silanized by immersion into a solution containing benzaldehyde-carrying triethoxysilane. (b) The sample is selectively functionalized by spatially resolved laser lithography. In areas solely exposed to the excitation laser at 700 nm, activated benzaldehyde reacts with bromine-carrying maleimide via a Diels−Alder cycloaddition. In contrast, the reaction is suppressed in areas that are simultaneously exposed to the depletion light at 440 nm. (c) Surfaceinitiated atom-transfer polymerization (SI-ATRP) is performed to grow poly(ethylene glycol) methyl ether methacrylate brushes from the functionalized areas, with the bromine acting as initiating species. (d) The sample is characterized by atomic-force microscopy (AFM).

time scale of fluorescence, because long lifetimes translate to low depletion laser intensities needed. However, very long lifetimes on the order of milliseconds might limit the writing speed, rendering the mechanism impractical for maskless lithography.10 The lifetime is therefore a trade-off. A comprehensive review in terms of resolution, writing power, depletion power, time constants, and writing speed for the discussed approaches has been given by Fischer and Wegener.10 In this article, we propose and realize such a molecular switch for STED-inspired direct laser writing, based on the cis−transphotoisomerization of intermediate photoenol species derived from α-methyl benzaldehydes.

tion at 700 nm fundamental wavelength and subsequently undergoes intramolecular hydrogen abstraction, that, via a biradical intermediate, results in the formation of the intermediate photoenol. The photoenol is formed in two isomeric states, i.e., a long-lived E-enol and short-lived Z-enol, which feature lifetimes that typically differ by 2 orders of magnitude.21 In the presence of activated enes, such as maleimide species, the E-enol can undergo a Diels−Alder cycloaddition forming a stable cycloproduct. On the other hand, the short-lived Z-enol reketonizes rapidly via reverse hydrogen transfer before a reaction with the maleimide can occur, reforming the starting product. By irradiation at 440 nm wavelength, the E-enol can be converted into Z-enol by photoisomerization, effectively depleting the population of Eenols and thereby also suppressing the formation of the cycloproduct. Hence, we have constructed a molecular switch that can be activated by one wavelength and deactivated by a second wavelength, the aforementioned prerequisite for a STED-inspired mechanism. Cis−trans photoisomerization has previously been used in super-resolution microscopy.28,29 In the literature, it was debated whether only E-enol is converted to Zenol by photoisomerization,30 or whether the inverse photoisomerization from Z-enol to E-enol can also occur.31 In the latter case, the E-enol population is still effectively depleted as

RESULTS AND DISCUSSION It is well-known that irradiation of alkyl phenyl ketones with UV light results in formation of intermediate photoenol species, a mechanism that has been extensively studied for lighttriggered photorelease or photoremovable protecting groups.21,22 As previously reported by our group, one-photon excitation at UV wavelengths can be replaced by two-photon excitation at 700 nm, opening up photoenol chemistry for 2D and 3D laser lithography.23−27 The proposed mechanism is depicted in Scheme 1. A methylbenzaldehyde species is excited by two-photon absorp6398

DOI: 10.1021/acsnano.7b02820 ACS Nano 2017, 11, 6396−6403

Article

ACS Nano the concentration of Z-enol can be assumed to be zero due to the rapid reketonization process and, thus, no considerable Eenol formation occurs by irradiation with the depletion light. In both cases the concept remains applicable for STED-inspired laser lithography. In order to investigate the proposed mechanism experimentally, the sample preparation procedure depicted in Figure 1 is used. First, the surface of a glass substrate is activated by an oxygen plasma and silanized with benzaldehyde-carrying silane molecules (Figure 1a). To relate to Scheme 1, the R group of the benzaldehyde derivative is chosen to be a triethoxysilane. Thus, the photoactive groups are covalently bound to the surface. This step eliminates diffusion, which might otherwise have a detrimental effect on line width and resolution. Second, the laser lithography step is performed in a custom-built setup that allows for focused irradiation with femtosecond pulses at 700 nm center wavelength and 440 nm continuous-wave laser light. A 1 mmol·L−1 solution of bromide-carrying maleimide in dimethylformamide (DMF) is added as a drop that is held in place by a polydimethylsiloxane ring on the substrate. Such, the R′ group depicted in Scheme 1 is chosen to be a species containing bromine. During lithography, the cycloaddition reaction takes place in the regions which have been exposed solely to the excitation light, resulting in bromine groups being covalently bound to the surface. In areas that are simultaneously irradiated by both excitation and depletion lasers, the cycloaddition reaction is suppressed by the above mechanism. As a result, the substrate surface is covalently functionalized with bromine groups with an area density that spatially varies following the lithographically defined patterns (Figure 1b). Third, in order to render the surface functionalization measurable, the sample is transferred to a reaction chamber and surface-initiated atom-transfer radical polymerization (SIATRP)32 is performed using the bromine groups as initiators. Hence, polymer brushes consisting of repeating units of poly(ethylene glycol) methyl ether methacrylate (PEGMEMA) are grown from the functionalized areas (Figure 1c). Fourth, the sample is characterized by atomic-force microscopy (AFM), to reveal the lithographically defined patterns (Figure 1d). In principle, many different functionalization procedures are feasible using our approach because the R′ group of the maleimide can be arbitrarily selected.33 As an example, we have alternatively used a biotin-carrying maleimide and subsequent immobilization of streptavidin dyes, followed by investigation with fluorescence microscopy (Figure S1). However, the procedure stated above proved to be most reliable and provided the most insight for further investigations toward diffraction-unlimited optical lithography. In particular, the height information gained in AFM measurements can be directly related to the functionalization density, as the height of the polymer brushes grown by SI-ATRP is reported to relate proportionally to the initiator density.34 The experimental setup employed for the lithography step was described in previous publications,16 yet here modified to include a 440 nm laser diode. The laser beams are focused by an objective lens with a numerical aperture of 1.4 and remain fixed, while the sample is moved by a 3D piezo stage. The laser foci were measured by scanning gold beads and detecting the light scattered thereof. Results are depicted in Figure 2. As illustrated, we have used two different spatial profiles for the depletion laser. The first is a regular Gaussian laser focus that is slightly widened to match the size of the diffraction-limited excitation laser focus. For the second, we introduce an edge

Figure 2. Point-spread functions of the laser foci used for laser lithography are measured by scanning gold beads and detecting the backscattered light. The results are shown in false-color plots and corresponding cuts through the data for the excitation laser at 700 nm wavelength and the depletion laser at 440 nm wavelength. The depletion laser can be modified by introducing an edge phase mask, adding a phase of π to one half-space, imposing a line of zero intensity onto the focus.

phase mask, i.e., a phase mask adding a phase of π to one halfspace in the conjugate plane. The phase mask results in a line of zero intensity imposed onto the Gaussian profile. In a first set of experiments, we use the Gaussian depletion focus without a phase mask and write a series of patterns as depicted in Figure 3. While the excitation laser is switched on

Figure 3. A test pattern is written using a Gaussian depletion focus as indicated in the scheme (left) and characterized by AFM measurements (right). In the central part, the depletion laser is switched on, completely suppressing the functionalization, while previously functionalized parts are not affected. By writing again through previously depleted areas, the reversibility of the depletion effect is demonstrated.

permanently, the depletion laser is only activated in the central part. It is clearly visible that the depletion beam completely suppresses the functionalization and also does not affect already written lines. This is an important finding for practical applications that involve more complex patterns than just isolated individual lines. Furthermore, by writing through the previously depleted area, it becomes evident that the depletion effect is predominantly reversible, i.e., that the depleted 6399

DOI: 10.1021/acsnano.7b02820 ACS Nano 2017, 11, 6396−6403

Article

ACS Nano

Figure 4. Polymer brush patterns for determination of the accessible line widths are written and characterized by AFM measurements. (a) Both excitation and depletion laser powers are varied to determine ideal parameters. (b) To demonstrate depletion effects, the depletion laser is switched on for the central part of each line. With increasing depletion laser power, the line width as well as the line height is reduced. (c) Cuts along the dashed lines reveal a minimum full-width-at-half-maximum (fwhm) of the lines of typically 60 nm. A model incorporating effects of an imperfect depletion focus and a nonlinear relation between functionalization density and polymer height reproduces the experimental data in excellent agreement (panels d and e).

after a shift of 25 μs, which fits to reported lifetimes of similar species.21 In order to investigate the possible resolution enhancement of the mechanism, we use the depletion laser focus with an edge phase mask, precisely overlaid onto the Gaussian writing focus. Line patterns are written for different ranges of excitation and depletion laser powers following the general scheme, where, while writing a line with the excitation laser, the depletion laser is solely switched on for the central part of the line. Typical results are shown in Figure 4. Line profiles are extracted from the AFM measurements at the indicated positions and averaged along the direction perpendicular to the lines in order to compensate the typical roughness caused by nonuniform polymer growth. Profile A shows a cut through an undepleted part of a line, which exhibits full-width-at-halfmaximum (fwhm) of 355 nm and a flattened top in the central part. Naively, one would expect the line shape to follow exactly the intensity distribution of the excitation laser focus squared, as excitation is a two-photon process. However, due to the finite silanization density, saturation of the functionalization density leads to the observed line shape deviation. Figure S5 depicts a set of lines written with increasing excitation laser power and the depletion laser switched off. Using a simple rate equation model, saturation effects can be reproduced as seen in the experimental data. In particular, it becomes obvious that the fwhm of the squared intensity of excitation laser focus is actually only a lower limit of the achievable line width in the case of depletion being switched off. The profiles B to E of Figure 4 correspond to depletion powers spanning almost 2 orders of magnitude. Clearly, a considerable line width narrowing can be achieved, down to a fwhm of 60 nm, i.e., a factor of 3.5 below the fwhm of the squared excitation intensity profile. Similar fwhm have been obtained for seven other data sets. The smallest fwhm found is 54 nm. For a resist system

photoenols transfer back to the benzaldehyde ground state, which is another necessary prerequisite for writing complex patterns. However, the line height decreases by 23% in this region. Moreover, this irreversible contribution becomes more pronounced for larger excitation and depletion powers (see Figure S2). The higher the excitation power, hence the higher the depletion power needed to achieve a considerable depletion effect, the larger the irreversible contribution becomes. At yet higher excitation and depletion powers, even previously written lines are damaged irreversibly. These observations obviously pose a limit to excitation and depletion laser powers that can be used. For the following experiments, however, we stay in the reversible regime. A further piece of evidence for our proposed mechanism is the observed influence of the atmospheric environment (see Figure S3). For the depicted sample, the lithography step is first conducted in a sealed chamber under nitrogen atmosphere. Subsequently, the chamber is opened to allow ambient oxygen to diffuse into the maleimide solution before writing the same lithographic patterns again. Depending on the writing power, the lines written under nitrogen atmosphere are up to 135% higher than the lines written under air. We attribute this effect to the intermediate biradical state (see Scheme 1) being partially scavenged by oxygen. As a result, the affected benzaldehydes are lost irreversibly, lowering the initiator density and thus reducing the polymer brush height. As mentioned above, the time scales involved in a depletion mechanism play an important role. We have assessed the lifetime of the intermediate E-enol by performing writing experiments, similar to a previously reported method.35 The results are depicted in Figure S4. We record the depletion effect with respect to a time shift between excitation and depletion exposure and find the depletion light to have no effect anymore 6400

DOI: 10.1021/acsnano.7b02820 ACS Nano 2017, 11, 6396−6403

Article

ACS Nano

Figure 5. (a) Grating polymer brush structures are fabricated for the determination of the lateral resolution by sequentially writing lines. The structures are characterized by AFM measurements. In the parts where the depletion laser is switched off, flat plateaus are formed. In the parts that have seen the depletion light, the grating corrugation can be retrieved down to lateral lattice constants of a = 100 nm. (b) Line cuts through the data reveal that the trenches of the gratings are partly filled up due to the lack of a sharp thresholding mechanism. The employed model reproduces the experimental data well.

In addition to the line width obtained for individual lines, an additional important measure is the minimum distance in which one can write two lines while they can still be separated from each other, i.e., the resolution. In order to investigate this aspect, we fabricate line gratings consisting of sequentially written individual lines for different grating periods a. Typical results are shown in Figure 5. Again, to demonstrate the depletion effect, the depletion laser is switched on only for the second half of the line. In the part that has not seen the depletion light, flat top plateaus are formed that show irregular corrugations stemming from the polymer growth process but no clear periodicity. With decreasing grating period, the height of the plateaus increases, indicating that at, e.g., at a = 200 nm, the functionalization is not yet fully saturated. Looking at the profiles B, D, and F of the regions that were irradiated by the depletion laser, periodic corrugations are visible down to a grating period of a = 100 nm, whereas at a = 75 nm, the periodicity cannot be retrieved unambiguously anymore presumably because of polymer roughness. Without depletion effects, for a two-photon process at 700 nm fundamental wavelength and for our focusing conditions, the resolution limit according to Sparrow’s criterion is 177 nm. Therefore, we have demonstrated subdiffraction resolution. In contrast to previously reported work in the field based on negative-tone acrylate resists, our system does not show a steep thresholding behavior.10 Hence, as the exposure dose of neighboring lines adds up, polymer brushes grow between the lines as well, lowering the measurable contrast of the gratings. As in the single-line experiments mentioned above, the height of the lines is lowered considerably by the depletion light. Using the same combined model as in Figure 4, the experimental data of the line profiles can be reproduced with good agreement.

without any threshold, this fwhm determines the line width. In this sense, we have demonstrated patterning with subdiffraction line width. Obviously, the height of the lines also decreases considerably with increasing depletion laser power, posing a limitation for further line-width reduction. Such a behavior can be explained by a combination of configurational properties of the grown polymer brushes and optical imperfections in the laser lithography process. First, it was reported in literature that the initiator density no longer translates linearly into the height of the grown polymer when going to low initiator densities or when the initiators are confined to small lithographically defined submicron-sized patterns.36 The reason is that brushes grown on a small densely functionalized area in close vicinity to nonfunctionalized areas, such as the spot in the minimum of the depletion focus, do not erect to full height but rather stay in a more compact configuration. In Figure S6, we model such effects by Gaussian blurring of the brush height and show that this in part reconstructs the measured data. Second, a residual intensity, i.e., a nonperfect zero, in the minimum of the depletion focus leads to lower initiator densities for depleted lines that translate into lower line heights, even when neglecting the aforementioned effect of initiator density on the brush height. The focus measurement in Figure 2 shows a residual intensity of 1.6% of the maximum. However, it remains unclear, whether this is reasonably close to the real value, as the typical measurement uncertainty is on the order of 1% and it is conceptually impossible to detect infinitesimally narrow features by scattering measurements with finite-sized gold beads. A model incorporating the measured residual intensity in the depletion focus partially reproduces the experimental data as well (Figure S7). This result agrees with findings in STED microscopy employing the same type of phase mask with comparable fractional residual intensity.37 We conclude that both effects play a role in our experiments and therefore use a combination of the two models. This combined model reproduces the experimental data very well (compare panels d and e of Figure 4).

CONCLUSION We introduce and experimentally demonstrate a photochemical molecular switch based on photoenol chemistry that allows STED-inspired laser lithography. A process based on cis−trans photoisomerization has been realized and is exploited to 6401

DOI: 10.1021/acsnano.7b02820 ACS Nano 2017, 11, 6396−6403

Article

ACS Nano

AUTHOR INFORMATION

overcome the diffraction limit in terms of the single line width as well as the lateral resolution. In order to decouple different influences, we have exploited our approach by subdiffraction surface functionalization followed by polymerization. However, the underlying mechanism is not necessarily limited to surface functionalization but can be exploited in various ways to achieve subdiffraction structuring by optical lithography. For example, the development of suitable negative-tone photoresists based on this concept could enable 3D structuring with resolution below the diffraction limit. Generally, along the lines of recent developments in super-resolution microscopy, the introduction of an isomeric switch for STED-inspired lithography constitutes a conceptually crucial step in order to enable higher precision in optical nanofabrication.

Corresponding Authors

*E-mail: [email protected], christopher. [email protected]. *E-mail: [email protected]. ORCID

Patrick Mueller: 0000-0001-6737-614X Gerd Ulrich Nienhaus: 0000-0002-5027-3192 Christopher Barner-Kowollik: 0000-0002-6745-0570 Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS All authors received funding and support from the Karlsruhe Institute of Technology and the Helmholtz-Gemeinschaft in the context of the Helmholtz program Science and Technology of Nanosystems (STN). M. Zieger received funding from Evangelisches Studienwerk Villigst. C. Barner-Kowollik received funding from Queensland University of Technology. We acknowledge support by the Karlsruhe School of Optics & Photonics (KSOP) and the KIT Nanostructure Service Laboratory (NSL).

METHODS Material Preparation. Benzaldehyde-functionalized silane38 and bromine-carrying maleimide32 were prepared according to previously reported procedures. Silanization. Glass substrate surfaces were thoroughly cleaned and activated by an oxygen plasma for 15 min, before being immersed in a 0.1 mmol·L−1 solution of benzaldehyde-functionalized silane in anhydrous toluene. After 4 h the substrates were rinsed with toluene, acetone, 2-propanol, and water, before being dried with nitrogen. Laser Lithography. Laser lithography is performed with a modified version of a previously described setup.16 As excitation laser source a Ti:sapphire oscillator (Chameleon Ultra II, Coherent) is used, tuned to emit femtosecond laser pulses at a central wavelength of 700 nm with a repetition rate of 80 MHz. As depletion laser source, a 440 nm continuous-wave laser diode (LDH-D-C-440, Picoquant) is added to the setup. A custom-built edge phase mask can be introduced into the depletion beam path (see above). Acousto-optic modulators (AA Opto-Electronic) allow for power control. Both laser beams are linearly polarized and overlaid using a dielectric beam splitter, before being focused by an oil-immersion objective lens with a numerical aperture of 1.4 (HCX PL APO 100× /1.4−0.7 Oil CS, Leica). The sample is moved with respect to the laser beam focus by a 3D piezo stage (Physik Instrumente). The samples are fabricated with a writing speed of 10 μm·s−1. After fabrication, samples are rinsed with DMF, acetone, 2-propanol, and water and dried with nitrogen. Surface-Initiated Atom-Transfer Radical Polymerization. SIATRP is performed according to a previously reported procedure.32 A solution of 2,2′-dipyridyl (435 mg), PEGMEMA (8.5 g, Mn = 300 g· mol−1, previously deinhibited by passing through a basic alumina column), CuBr2 (12.2 mg), CuCl (80 mg) in 20 mL water:methanol (1:1) is prepared and degassed by nitrogen bubbling for 1 h. Next, the solution is transferred into a reaction vial containing the samples in inert atmosphere. The polymerization is carried out for 45 min at 60 °C. Subsequently the samples are thoroughly rinsed with methanol, water, and DMF and dried with nitrogen. Atomic-Force Microscopy. Polymer brushes are characterized in dry state using a commercial AFM system (Easyscan 2, Nanosurf) in dynamic force mode with high-precision tips (PPP-NCLR, Nanosensors).

REFERENCES (1) Abbe, E. Beiträge zur Theorie des Mikroskops und der mikroskopischen Wahrnehmung. Arch. Für Mikrosk. Anat. 1873, 9, 413−418. (2) Hell, S. W.; Wichmann, J. Breaking the Diffraction Resolution Limit by Stimulated Emission: Stimulated-Emission-Depletion Fluorescence Microscopy. Opt. Lett. 1994, 19, 780−782. (3) Rust, M. J.; Bates, M.; Zhuang, X. Sub-Diffraction-Limit Imaging by Stochastic Optical Reconstruction Microscopy (STORM). Nat. Methods 2006, 3, 793−796. (4) Betzig, E.; Patterson, G. H.; Sougrat, R.; Lindwasser, O. W.; Olenych, S.; Bonifacino, J. S.; Davidson, M. W.; Lippincott-Schwartz, J.; Hess, H. F. Imaging Intracellular Fluorescent Proteins at Nanometer Resolution. Science 2006, 313, 1642−1645. (5) Hell, S. W. Microscopy and Its Focal Switch. Nat. Methods 2009, 6, 24−32. (6) Reuss, M.; Engelhardt, J.; Hell, S. W. Birefringent Device Converts a Standard Scanning Microscope into a STED Microscope That Also Maps Molecular Orientation. Opt. Express 2010, 18, 1049− 1058. (7) Heintzmann, R.; Jovin, T. M.; Cremer, C. Saturated Patterned Excitation Microscopya Concept for Optical Resolution Improvement. J. Opt. Soc. Am. A 2002, 19, 1599−1609. (8) Gustafsson, M. G. L. Nonlinear Structured-Illumination Microscopy: Wide-Field Fluorescence Imaging with Theoretically Unlimited Resolution. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 13081− 13086. (9) Klar, T. A.; Hell, S. W. Subdiffraction Resolution in Far-Field Fluorescence Microscopy. Opt. Lett. 1999, 24, 954−956. (10) Fischer, J.; Wegener, M. Three-Dimensional Optical Laser Lithography beyond the Diffraction Limit. Laser Photonic Rev. 2013, 7, 22−44. (11) Klar, T. A. STED Lithography and Protein Nanoanchors. In Optically Induced Nanostructures: Biomedical and Technical Applications; König, K., Ostendorf, A., Eds.; De Gruyter: Berlin, 2015; pp 303−324. (12) Three-Dimensional Microfabrication Using Two-Photon Polymerization: Fundamentals, Technology, and Applications; Baldacchini, T., Ed.; William Andrew Publishing: Oxford, 2016. (13) Multiphoton Lithography: Techniques, Materials, and Applications; Stampfl, J., Liska, R., Ovsianikov, A., Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, 2016.

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.7b02820. Details about alternative functionalization approach, reversibility test patterns, influence of the atmospheric environment, experiments for lifetime determination, saturation effects, rate models, model for nonlinear brush growth, and model for imperfect depletion beam intensity minimum, (PDF) 6402

DOI: 10.1021/acsnano.7b02820 ACS Nano 2017, 11, 6396−6403

Article

ACS Nano (14) Fischer, J.; von Freymann, G.; Wegener, M. The Materials Challenge in Diffraction-Unlimited Direct-Laser-Writing Optical Lithography. Adv. Mater. 2010, 22, 3578−3582. (15) Fischer, J.; Wegener, M. Three-Dimensional Direct Laser Writing Inspired by Stimulated-Emission-Depletion Microscopy (Invited). Opt. Mater. Express 2011, 1, 614−624. (16) Fischer, J.; Mueller, J. B.; Quick, A. S.; Kaschke, J.; BarnerKowollik, C.; Wegener, M. Exploring the Mechanisms in STEDEnhanced Direct Laser Writing. Adv. Opt. Mater. 2015, 3, 221−232. (17) Li, L.; Gattass, R. R.; Gershgoren, E.; Hwang, H.; Fourkas, J. T. Achieving λ/20 Resolution by One-Color Initiation and Deactivation of Polymerization. Science 2009, 324, 910−913. (18) Scott, T. F.; Kowalski, B. A.; Sullivan, A. C.; Bowman, C. N.; McLeod, R. R. Two-Color Single-Photon Photoinitiation and Photoinhibition for Subdiffraction Photolithography. Science 2009, 324, 913−917. (19) Cao, Y.; Gan, Z.; Jia, B.; Evans, R. A.; Gu, M. HighPhotosensitive Resin for Super-Resolution Direct-Laser-Writing Based on Photoinhibited Polymerization. Opt. Express 2011, 19, 19486. (20) Hell, S. W. Strategy for Far-Field Optical Imaging and Writing without Diffraction Limit. Phys. Lett. A 2004, 326, 140−145. (21) Sankaranarayanan, J.; Muthukrishnan, S.; Gudmundsdottir, A. D. Photoremovable Protecting Groups Based on Photoenolization. In Adv. Phys. Org. Chem.; Richard, J. P., Ed.; Academic Press, 2009; Vol. 43, pp 39−77. (22) Klán, P.; Šolomek, T.; Bochet, C. G.; Blanc, A.; Givens, R.; Rubina, M.; Popik, V.; Kostikov, A.; Wirz, J. Photoremovable Protecting Groups in Chemistry and Biology: Reaction Mechanisms and Efficacy. Chem. Rev. 2013, 113, 119−191. (23) Richter, B.; Pauloehrl, T.; Kaschke, J.; Fichtner, D.; Fischer, J.; Greiner, A. M.; Wedlich, D.; Wegener, M.; Delaittre, G.; BarnerKowollik, C.; Bastmeyer, M. Three-Dimensional Microscaffolds Exhibiting Spatially Resolved Surface Chemistry. Adv. Mater. 2013, 25, 6117−6122. (24) Quick, A. S.; Rothfuss, H.; Welle, A.; Richter, B.; Fischer, J.; Wegener, M.; Barner-Kowollik, C. Fabrication and Spatially Resolved Functionalization of 3D Microstructures via Multiphoton-Induced Diels−Alder Chemistry. Adv. Funct. Mater. 2014, 24, 3571−3580. (25) Kaupp, M.; Hiltebrandt, K.; Trouillet, V.; Mueller, P.; Quick, A. S.; Wegener, M.; Barner-Kowollik, C. Wavelength Selective Polymer Network Formation of End-Functional Star Polymers. Chem. Commun. 2016, 52, 1975−1978. (26) Blasco, E.; Wegener, M.; Barner-Kowollik, C. Photochemically Driven Polymeric Network Formation: Synthesis and Applications. Adv. Mater. 2017, 29, 1604005. (27) Fruk, L.; Kerbs, A.; Mueller, P.; Kaupp, M.; Ahmed, I.; Quick, A. S.; Abt, D.; Wegener, M.; Niemeyer, C. M.; Barner-Kowollik, C. Photo-Induced Click Chemistry for DNA Surface Structuring via Direct Laser Writing. Chem. - Eur. J. 2017, 23, 4990−4994. (28) Hofmann, M.; Eggeling, C.; Jakobs, S.; Hell, S. W. Breaking the Diffraction Barrier in Fluorescence Microscopy at Low Light Intensities by Using Reversibly Photoswitchable Proteins. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 17565−17569. (29) Bossi, M.; Fölling, J.; Dyba, M.; Westphal, V.; Hell, S. W. Breaking the Diffraction Resolution Barrier in Far-Field Microscopy by Molecular Optical Bistability. New J. Phys. 2006, 8, 275. (30) Netto-Ferreira, J. C.; Scaiano, J. C. A Laser Flash Photolysis Study of the Mechanism of the Photocyclization of α-Chloro-oMethylacetophenones. J. Am. Chem. Soc. 1991, 113, 5800−5803. (31) Pelliccioli, A. P.; Klán, P.; Zabadal, M.; Wirz, J. Photorelease of HCl from O-Methylphenacyl Chloride Proceeds through the ZXylylenol. J. Am. Chem. Soc. 2001, 123, 7931−7932. (32) Rodriguez-Emmenegger, C.; Preuss, C. M.; Yameen, B.; PopGeorgievski, O.; Bachmann, M.; Mueller, J. O.; Bruns, M.; Goldmann, A. S.; Bastmeyer, M.; Barner-Kowollik, C. Controlled Cell Adhesion on Poly(Dopamine) Interfaces Photopatterned with Non-Fouling Brushes. Adv. Mater. 2013, 25, 6123−6127.

(33) Delaittre, G.; Goldmann, A. S.; Mueller, J. O.; Barner-Kowollik, C. Efficient Photochemical Approaches for Spatially Resolved Surface Functionalization. Angew. Chem., Int. Ed. 2015, 54, 11388−11403. (34) Jones, D. M.; Brown, A. A.; Huck, W. T. S. Surface-Initiated Polymerizations in Aqueous Media: Effect of Initiator Density. Langmuir 2002, 18, 1265−1269. (35) Fischer, J.; Wegener, M. Ultrafast Polymerization Inhibition by Stimulated Emission Depletion for Three-Dimensional Nanolithography. Adv. Mater. 2012, 24, OP65−OP69. (36) Lee, W.-K.; Patra, M.; Linse, P.; Zauscher, S. Scaling Behavior of Nanopatterned Polymer Brushes. Small 2007, 3, 63−66. (37) Klar, T. A.; Engel, E.; Hell, S. W. Breaking Abbe’s Diffraction Resolution Limit in Fluorescence Microscopy with Stimulated Emission Depletion Beams of Various Shapes. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 2001, 64, 066613. (38) Pauloehrl, T.; Delaittre, G.; Winkler, V.; Welle, A.; Bruns, M.; Börner, H. G.; Greiner, A. M.; Bastmeyer, M.; Barner-Kowollik, C. Adding Spatial Control to Click Chemistry: Phototriggered Diels− Alder Surface (Bio)Functionalization at Ambient Temperature. Angew. Chem., Int. Ed. 2012, 51, 1071−1074.

6403

DOI: 10.1021/acsnano.7b02820 ACS Nano 2017, 11, 6396−6403