Flexible and Micropatternable TripletâTriplet ... - ACS Publications

Subscriber access provided by UNIV OF DURHAM

Article

Flexible and Micropatternable Triplet–Triplet Annihilation Upconversion Thin Films for Photonic Device Integration and Anti-Counterfeiting Applications Anna L. Hagstrom, Hak-Lae Lee, Myung-Soo Lee, Hyun-Seok Choe, Joori Jung, Byung-Geon Park, Won-Sik Han, Jong Soo Ko, Jae-Hong Kim, and Jaehyuk Kim ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b17789 • Publication Date (Web): 14 Feb 2018 Downloaded from http://pubs.acs.org on February 15, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Flexible and Micropatternable Triplet–Triplet Annihilation Upconversion Thin Films for Photonic Device Integration and Anti-Counterfeiting Applications Anna L. Hagstrom,‡ Hak-Lae Lee,† Myung-Soo Lee,† Hyun-Seok Choe,† Joori Jung,§ ByungGeon Park,║ Won-Sik Han,§ Jong-Soo Ko,║ Jae-Hong Kim,‡ and Jae-Hyuk Kim*,† †

Department of Chemical and Environmental Engineering, Pusan National University, 46241 Busan, Korea ‡ Department of Chemical and Environmental Engineering, Yale University, New Haven, CT 06511, USA § Department of Chemistry, Seoul Women’s University, 01797 Seoul, Korea ║ Graduate School of Mechanical Engineering, Pusan National University, 48075 Busan, Korea KEYWORDS: Triplet–triplet annihilation, upconversion, thin films, soft lithography, anticounterfeiting ABSTRACT: Triplet–triplet annihilation upconversion (TTA-UC) has recently drawn widespread interest for its capacity to harvest low-energy photons and to broaden the absorption spectra of photonic devices such as solar cells. Although conceptually promising, effective integration of TTA-UC materials into practical devices has been difficult due to the diffusive and anoxic conditions required in TTA-UC host media. Of the solid-state host materials investigated, rubbery polymers facilitate the highest TTA-UC efficiency. To date, however, their need for long-term oxygen protection has limited rubbery polymers to rigid film architectures that forfeit their intrinsic flexibility. This study introduces a new multilayer thin film architecture in which scalable solution processing techniques are employed to fabricate flexible, photostable, and efficient TTA-UC thin films containing layers of oxygen barrier and host polymers. This breakthrough material design marks a crucial advance toward TTA-UC integration within rigid and flexible devices alike. Moreover, it introduces new opportunities in unexplored applications such as anti-counterfeiting. Soft lithography is incorporated into the film fabrication process to pattern TTA-UC host layers with a broad range of high-resolution microscale designs, and superimposing host layers with customized absorption, emission, and patterning ultimately produces proof-of-concept anti-counterfeiting labels with advanced excitation-dependent photoluminescent security features.

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 25

Introduction Recent advances in upconversion (UC) materials have drawn widespread attention to their potential for applications in photonic devices such as solar cells (SCs).1 Incorporation of UC materials that upconvert a low-energy portion of the solar spectrum that would otherwise be wasted (i.e., sub-bandgap photons) could theoretically allow SCs to reach power conversion efficiencies exceeding 40%.2-4 While earlier attempts to implement this approach using lanthanide-based UC materials were unable to achieve appreciable efficiency enhancements,5-6 triplet–triplet annihilation (TTA)-UC systems have emerged over the last decade as a promising alternative.7 TTA-UC functions through a series of energy transfers between sensitizer chromophores (e.g., metalloporphyrins) that absorb incident low-energy photons and acceptor chromophores (e.g., anthracene derivatives) that emit high-energy upconverted photons. Facile tuning of absorption and emission wavelengths through choice of sensitizer and acceptor, respectively, enables an ever-expanding range of NIR-to-visible, visible-to-visible, and visibleto-UV anti-Stokes shift schemes.8 Constraints intrinsic to the TTA-UC mechanism, however, complicate the search for effective chromophore host materials suited to practical applications. It is paramount that host materials provide robust protection from molecular oxygen (3O2),9-10 which quenches chromophores’ longlived triplet excited states, thereby producing reactive singlet oxygen (1O2) capable of chromophore degradation.11 Furthermore, they must facilitate collision between chromophores in order for the necessary Dexter-type energy transfers to occur.12 Although these conditions are most easily met in deoxygenated low-viscosity solvents contained in sealed vessels, device integration in practice demands robust solid-state host materials.13 Glassy host polymers with high glass transition temperatures (Tg) provide effective oxygen protection but prohibit


2

Page 3 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


chromophore diffusion,14 making organogels,15-18 glassy polymers with nanoscale fluidity,19-20 and rubbery (i.e., low-Tg) polymers more promising as efficient UC host materials.21-22 Of these options, rubbery polymers provide the best balance of mechanical flexibility and stability. Moreover, they have achieved UC with a quantum yield (QY) exceeding 30% (defining the QY as 100% when all absorbed photons undergo TTA-UC; Equation S5),23 which, to our knowledge, is the highest QY reported in a solid TTA-UC material. Though otherwise ideal as hosts, rubbery polymers require oxygen protection for long-term operation in ambient conditions. Their polymer matrices provide a degree of protection from ambient oxygen, but the same local polymer chain mobility that makes them effective TTA-UC hosts also allows oxygen to infiltrate over time. Thin films, the configuration most favorable for device integration, are particularly vulnerable. Encapsulating these films within oxygen barrier materials, an approach that could eventually be combined with emergent efforts to develop singlet-oxygen-scavenging chromophores and films,11, 24 is the most straightforward solution. Thus far, however, TTA-UC polymer films have only been encapsulated by curing chromophore-doped polymer precursor solutions between rigid barrier panels (e.g., glass).22, 25-27 While this approach has been shown to provide effective oxygen protection,25 it forfeits the flexibility of the resulting films and affords limited control over their thickness and uniformity. Here, we overcome these limitations by characterizing the first spin-coated polyurethane (PU) TTA-UC host and developing a new multilayer thin film architecture comprising layers of oxygen barrier polymer and PU host deposited by scalable solution processing. This enables us to fabricate the first flexible TTA-UC thin films that maintain high efficiency and photostability. While its prospects for SC device integration are obvious, this breakthrough multilayer UC film approach also allows us to explore the untapped promise of TTA-UC in anti-counterfeiting


3


Page 4 of 25

applications. The ease of mimicking the Stokes fluorescence of conventional UV-absorbing dyes, which are ubiquitous in the luminescent security features of valuable documents such as currency and identification cards, has inspired extensive research into advanced alternatives with photoluminescence more difficult to imitate.28 In recent years, lanthanide-based UC materials have drawn widespread attention as a promising alternative.29-32 Despite the considerable advantages of TTA-UC, which requires lower excitation intensities, achieves higher efficiencies, and enables a broader variety of anti-Stokes shifts, the lack of photostable TTA-UC host media suitable for patterning has previously precluded its use for anti-counterfeiting. Utilizing the moldable nature of rubbery polymers, we incorporate soft lithography into our fabrication process to produce films containing PU host layers patterned with a wide range of highresolution microscale designs.33 By superimposing patterned host layers containing a variety of chromophore combinations, we create, for the first time, anti-counterfeiting labels with photoluminescent security features harnessing TTA-UC.

Experimental Section Materials. Polyvinyl alcohol (PVA; Mowiol 4-88) and perylene were purchased from Aldrich, 4,7-di(thiophen-2-yl)benzo[c][1,2,5]thiadiazole

(DTPBTD)

was

purchased

from

Tokyo

Chemical Industry Co., Ltd. (TCI), and photocurable polyurethane acrylate (PUA; MINS-ERM) was purchased from Minuta Tech (Korea). All chemicals were used as received. Palladium(II) meso-tetraphenyltetrabenzoporphyrin (PdTPBP) was synthesized according to a procedure reported in the literature.34 Perylene, PdTPBP, and DTPBTD stock solutions were prepared in tetrahydrofuran (Aldrich). PU precursor was purchased from Polytek (Poly 74-30, US) in the form of a polyisocyanate component (A) and a polyol component (B). Chromophore-doped PU


4

Page 5 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


solutions for spin-coating were prepared by adding 2 mL of chloroform to 2 g of a mixture of PU precursors A and B (A:B = 1:1 by weight) and curing this mixture in a sealed glass vial at 70 °C for 2 h under constant magnetic stirring (800 rpm). Four 2-mL aliquots of chloroform were added during this curing period, typically at 40, 60, 80, and 120 min. Appropriate aliquots of chromophore stock solutions were added to the resulting viscous PU solution to achieve the desired chromophore concentrations. PVA solution (0.32 g mL-1) was prepared by dissolving PVA powder in deionized (DI) water at 90 °C under vigorous stirring and then cooling it to room temperature. Fabrication of Planar UC Films (UCF1 and UCF2). To prepare UCF1, aqueous PVA solution (2000 rpm/60 s) and PU solution (700 rpm/60 s) were sequentially spin-coated (ACE100, Korea) onto a polyethylene terephthalate (PET) film. After 2 h of heating at 55 °C to remove residual solvent, a second layer of PVA solution was spin-coated onto the PU (2000 rpm/60 s), and the resulting sample was left to cure at room temperature for >12 h. To prepare UCF2, ca. 60 µL of PUA solution was drop-casted onto UCF1 and covered with a glass plate, using Scotch tape around the edges of the film as a spacer ensuring a level surface. After >1 h of UV curing, the glass plate was removed. Fabrication of Micropatterned UC Films (P-UCF). Master Molds. Master molds were fabricated on silicon using a standard photolithography technique. Micropatterns were converted into AutoCAD files and transferred to chrome-coated quartz to prepare micropatterned photomasks. A photoresist (SU-8, MicroChem, USA) was spin-coated onto 6-inch silicon wafers at 500 rpm for 5 s followed by 4000 rpm for 30 s. The photoresist-coated wafers were soft-baked on a hot plate (MSH-10, WiseStir, Korea) at 95 °C for 180 s. Photomasks were then mounted onto the photoresist-coated wafers using a mask aligner (MA/BA6, SUSS MicroTec, Germany)


5


Page 6 of 25

and exposed to UV radiation for 4.5 s (exposure dose: 130 mJ cm-2). Post-exposure baking was performed on a hot plate at 95 °C for 240 s. The wafers were then immersed in SU-8 developer (MicroChem, USA) for 3 min, washed with DI water and isopropyl alcohol, and dried in a nitrogen-atmosphere glove box. The molds were finished by hard-baking on a hot plate at 95 °C for 4 h. Polydimethylsiloxane (PDMS) Replica Molds. Self-assembled monolayers (SAM) were formed on the master molds by depositing a few droplets of SAM solution (L-SAM, Minuta Tech, Korea) onto the molds and then placing them in a vacuum desiccator (10-3 torr) for 30 min. PDMS prepolymer, prepared by mixing PDMS (Sylgard 184, Dow Corning, USA) and its curing agent in a 10:1 ratio by weight, was poured onto the SAM-treated molds and left to cure for 2 days at room temperature. Once fully cured, the replica molds were peeled away from the master molds. P-UCF Fabrication by Soft Lithography. PDMS replica molds were each pressed into ca. 130 µL of PU solution drop-casted onto a PVA-coated PET film, left in place while the PU cured at room temperature for 3 h and at 55 °C for 3 h, and then carefully removed. PVA solution was spin-coated (2000 rpm/60 s) onto the micropatterned PU several times to ensure complete coverage, and PUA was coated on top as described above. Characterization. Scanning electron micrographs were taken using a field emission scanning electron microscope (Supra 25, Zeiss). Fluorescence micrographs were taken using an Axioskop 2 Plus microscope (Zeiss) equipped with various optical filters. Anti-Stokes emission spectra were collected using a custom laser setup in which UC film samples were excited at an angle of approximately 45° using a 635 nm commercial diode laser with a beam 6 mm in diameter. Film emission was modulated with an optical chopper (80 Hz) and directed to a monochromator (Oriel


6

Page 7 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Cornerstone, Newport) using a series of focusing lenses, and scattered laser light was removed using a 632.8 nm notch filter. The signal was then detected by an Oriel photomultiplier tube and processed by a lock-in amplifier (SRB10 DSP, Stanford Research Systems). Incident laser intensity was adjusted using a continuously variable neutral density filter and measured using a power meter (843-R, Newport). Static absorption and Stokes emission spectra and time-resolved emission spectra were collected and analyzed using a spectrofluorometer (FS5-MCS, Edinburgh Instruments). Time-resolved phosphorescence spectra were employed in lifetime analyses detailed in Text S1. The phosphorescence QY of UCF2 doped only with PdTPBP was measured using a specialized spectrophotometer equipped with an integrating hemisphere (QE-2100, Otsuka Electronics) and used to calculate the UC QY of UCF2 doped with both PdTPBP and perylene as detailed in Text S2.

Results and Discussion Planar UC Films. We first developed a sequential coating method for TTA-UC film fabrication in ambient conditions (i.e., without need for an anoxic atmosphere). We selected PdTPBP and perylene as sensitizer and acceptor, respectively, a benchmark chromophore pairing for red-to-blue UC (Figure 1a).35-36 As illustrated in Figure 1b, red excitation creates a 3PdTPBP* triplet excited state, which produces a 3perylene* triplet excited state through triplet–triplet energy transfer (TTET), and TTA between two of these 3perylene* states produces a 1perylene* singlet excited state that emits blue upconverted fluorescence.37 We dissolved these chromophores in the precursor of a new commercial PU and tuned the viscosity of the resulting solution through partial curing and stepwise dilution to prepare a UC-PU solution suitable for spin-coating. To fabricate our first-generation UC films, designated UCF1, we sequentially spin-


7


Page 8 of 25

coated layers of (1) PVA, a water-soluble polymer frequently employed as an oxygen barrier,38-41 (2) UC-PU (thickness ca. 15 µm; Figure S1), and (3) PVA (again) onto a PET substrate (Figure 2a). Concerned that the water solubility of PVA might compromise film stability in humid conditions, we also fabricated a second generation of UC films, UCF2, by drop-casting PUA onto the exposed PVA layer of UCF1 (Figure 2a, S1). Both UCF1 and UCF2 were flexible and highly transparent, with a slight green tint from PdTPBP (Figure 2b), and emitted strong blue fluorescence under excitation at 635 nm (Figure 2c).

Figure 1. a) Molecular structures and normalized absorption (solid) and emission (dashed) spectra of sensitizer (S) palladium (II) meso-tetraphenyltetrabenzoporphyrin (PdTPBP; red) and acceptor (A) perylene (blue) in oleic acid. b) Schematic illustrating the steps in the triplet–triplet annihilation (TTA) upconversion (UC) mechanism: (1) photon absorption, (2) intersystem


8

Page 9 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


crossing (ISC), (3) triplet–triplet energy transfer (TTET), (4) TTA, and (5) upconverted fluorescence. The relative TTA-UC performance of UCF1, UCF2, and control films containing bare UC-PU layers confirmed that PVA acted as an effective oxygen barrier for UC-PU and that reinforcement with PUA further improved its performance. While the unprotected control films rapidly lost their UC capabilities, UCF1 and UCF2 maintained strong UC performance after days of storage in ambient conditions. Their UC fluorescence emission remained steady under ca. 1 h of continuous laser excitation

at 48.9 mW cm-2 (Figure S2), exhibiting photostability

comparable to that recently reported in a singlet-oxygen-scavenging host polymer.24 Because the intensity of UCF2 fluorescence consistently proved ca. 30% higher than that of UCF1 fluorescence under identical excitation (Figure S2), we focused on UCF2 for further studies.

Figure 2. (a) Schematic illustrating the fabrication of solution-processed UC films UCF1 and UCF2, using polyethylene terephthalate (PET), polyvinyl alcohol (PVA), polyurethane (PU), and polyurethane acrylate (PUA). (b) Photographs of flat (left) and bent (right) UCF2 (charcoal traces superimposed for emphasis). (c) Photographs of UCF2 with and without 635 nm laser


9


Page 10 of 25

excitation, taken through a 632.8 nm notch filter to exclude excitation wavelengths. In the PU layer, [PdTPBP] = 0.52 µmol g-1 and [perylene] = 12.7 µmol g-1.

Analyzing the lifetime of PdTPBP phosphorescence in UCF2 as a function of perylene concentration revealed that the spin-coated PU host layer facilitated highly efficient TTET from 3

PdTPBP* to perylene. The results of this analysis are summarized in Figure 3a–b (details in Text

S1). In UCF2 containing only PdTPBP, its phosphorescence exhibited mono-exponential decay with a lifetime (τ0) of 242 µs (Figure 3a). This closely matched its lifetime in deoxygenated tetrahydrofuran (245 µs; Figure S3), confirming that UCF2 provided effective protection from oxygen. In UCF2 containing both PdTPBP and perylene, TTET to perylene competed with radiative decay of 3PdTPBP* states. As a result, PdTPBP phosphorescence exhibited biexponential decay with (1) a short lifetime component (τ1) reflecting the 3PdTPBP* states that underwent TTET to perylene and (2) a long lifetime component (τ2) reflecting those that did not (Figure 3a). Both the percentage of 3PdTPBP* states quenched by TTET and the overall TTET efficiency (Text S2) rose with increasing perylene concentrations, reaching 97% and 94%,42 respectively, in UCF2 whose PU layer contained 1.9 µmol g-1 perylene (Figure 3a–b, Table S1). Using τ0 and τ1 to perform a Stern–Volmer analysis of PdTPBP phosphorescence quenching by perylene, we calculated a bimolecular quenching constant (kq) of 8.94 × 107 M-1 s-1 (Figure S4), comparable to values reported for PdTPBP and perylene in deoxygenated ionic liquids with viscosities of 25–40 mPa s.36, 43


10

Page 11 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. (a) Decay of PdTPBP phosphorescence, detected at 800 nm, emitted by UCF2 following excitation at 635 nm ([PdTPBP] = 0.52 µmol g-1 and [perylene] = 0–1.9 µmol g-1 in PU). Mono-exponential and bi-exponential fits to the experimental data are shown (cyan and red traces, respectively), as are the resulting lifetime in the PdTPBP-only film (τ0) and lifetime components in the most concentrated UC film (τ1 and τ2, percentages denote their relative intensities). (b) TTET efficiency of UCF2, calculated through analysis of (a), as a function of [perylene]. (c) Integrated UC emission intensity and (d) UC quantum yield of UCF2 ([PdTPBP] = 0.52 µmol g-1 and [perylene] = 12.7 µmol g-1 in PU) as a function of the power density of laser excitation at 635 nm. Each point marks the average value obtained from excitation of three different spots on the film, and the error bars denote the standard deviation of these measurements. The dashed lines in (c) show the slope of linear fits to the data at the lowest and highest excitation intensities employed, and the gray arrow marks the threshold intensity (Ith) at which they meet. The inset in (d) shows the rising emission of UCF2 under increasingly intense excitation.

To maximize the efficiency of TTA between 3perylene* states, we thereafter increased the perylene concentration in the PU layers of UCF2 to 12.7 µmol g-1, near its solubility limit. Under red excitation, these UCF2 emitted strong UC fluorescence with minimal PdTPBP


11


Page 12 of 25

phosphorescence (Figure S2a), confirming that they retained the high TTET efficiency of their dilute counterparts. Time-resolved emission spectroscopy revealed their UC fluorescence lifetime to be ca. 200 µs (Figure S5), reflecting the time delay required for TTET to populate the 3

perylene* state and for subsequent TTA to populate the 1perylene* state. Under increasingly strong excitation, the intensity of their UC fluorescence gradually

transitioned from a quadratic dependence on excitation power (Figure 3c, slope = 2.0) to a nearly linear dependence (Figure 3c, slope = 1.1). This characteristic transition reflects an increase in TTA efficiency; by generating higher 3perylene* concentrations, higher excitation intensities favored TTA between neighboring 3perylene* states over 3perylene* decay through pseudo-firstorder processes.9, 44 The threshold excitation intensity (Ith) beyond which TTA was the dominant mechanism, a key figure of merit for TTA-UC systems, was ca. 70 mW cm-2 in UCF2 (Figure 3c). While this is higher than desirable, the Ith of a TTA-UC system is inversely proportional to its light harvesting abilities,45 so we expect that lower values could be achieved in UCF2 containing higher PdTPBP concentrations. As TTA approached maximum efficiency—i.e., as the slope in Figure 3c approached one—the overall QY of the TTA-UC process approached 7% (Figure 3c–d, Equation S5). To our knowledge, this is the first time the UC QY of PdTPBP and perylene has been measured in a solid material. Though this value is admittedly lower than those reported for the same chromophores in deoxygenated low-viscosity solvents,37 it is more than twice as high as those reported in micellar systems or viscous solvents such as oleic acid and serves as a useful benchmark for future studies.35, 46-47 The ability to fabricate efficient, flexible, and photostable TTA-UC films through a simple spin-coating procedure marks a crucial advance in ongoing SC integration efforts. While attempts to develop SCs specifically tailored to UC systems (e.g., intermediate band SCs with


12

Page 13 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


electrically integrated TTA-UC) show promise,48-50 efficient UC materials easily mounted to existing SC architectures could have far broader impacts. A number of studies have conclusively proven that applying TTA-UC layers to the back of a SC can extend its effective absorption range and thereby improve its performance.51-53 However, these all employed either glassencased UC films or sealed glass cuvettes containing deoxygenated UC solutions.51-53 UCF1 and UCF2, the first free-standing TTA-UC materials compatible with both rigid and flexible SCs, open the door to a host of new applications ranging from SCs in wearable devices to SCs installed on nonplanar surfaces (e.g., buildings and vehicles).25, 54-55 It is noteworthy that many flexible SCs are currently fabricated on PET substrates through spin-coating and compatible solution processing techniques.56-57 As a result, employing UC films as substrates for SC fabrication or incorporating a UC layer deposition step into the existing SC fabrication process seems highly feasible. Micropatterned UC Films. Recognizing that the success of UCF1 and UCF2 also introduced new opportunities for TTA-UC applications in unexplored fields such as anti-counterfeiting, we constructed a third generation of UC films, P-UCF, containing PU layers with intricate microscale patterns. Curing host polymers within macroscale molds offers an easy and versatile approach for creating TTA-UC materials with different shapes.22 To extend this approach to the microscale for the first time, which is impossible with benchmark solution-based TTA-UC systems, we micropatterned master molds by photolithography and used them as templates for PDMS replicas (Figure S6). Modifying our initial film fabrication procedure, we used these PDMS molds to micropattern the PU layers by soft lithography: after drop-casting aliquots of UC-PU solution onto PVA-coated PET, we covered the PU with PDMS molds and allowed it to


13


Page 14 of 25

cure before depositing PVA and PUA protection layers (Figure 4a). The resulting P-UCF, like their planar UCF2 analogues, were flexible and highly transparent (Figure S7).

Figure 4. (a) Schematic illustrating the fabrication of micropatterned UC films (P-UCF) by soft lithography. (b) Scanning electron and (c) fluorescence micrographs of P-UCF under laser excitation at 635 nm, taken through a 500 nm shortpass (SP) filter. (d) Fluorescence micrographs of P-UCF containing PdTPBP and/or perylene under xenon lamp excitation at 365 nm and laser excitation at 635 nm, taken through either a 420 nm longpass (LP) filter or a 500 nm SP filter. The PdTPBP and perylene concentrations employed in the PU layers were 0.52 µmol g-1 and 12.7 µmol g-1, respectively. All scale bars represent 100 µm.

The P-UCF fabrication process allowed myriad combinations of patterning and photoluminescence schemes. As Figure 4b–c demonstrates, any AutoCAD design composed of strokes at least 10 µm thick, ranging from simple stripes to intricate logos, could be replicated in


14

Page 15 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


P-UCF. Tuning the designs’ absorption and emission using different chromophore compositions enabled a wide range of Stokes and anti-Stokes photoluminescence; for example, depending on its combination of PdTPBP and perylene, a P-UCF pattern invisible under white light either (1) appeared blue only under UV excitation or (2) appeared blue under both red and UV excitation (Figure 4d). Expanding beyond this pair of chromophores provides countless additional photoluminescence possibilities. P-UCF pairing PdTPBP with DTPBTD instead of perylene, for instance, produced yellow-green patterns visible only under red or blue excitation (Figure S8). By superimposing PU layers containing different patterns and chromophores, we created prototype P-UCF security labels with unique photoluminescence features. This allowed us to incorporate multiple wavelengths of excitation and/or emission, thereby making the P-UCF more difficult to counterfeit and inherently providing far more rigorous security.58 As an initial proof of concept, we superimposed (1) a striped PU layer containing only perylene and (2) a UC-PU layer containing both PdTPBP and perylene, which we patterned with a QR code (i.e., a type of machine-readable barcode often used in covert security labels31, 59) storing the phrase “PNU Upconversion Lab” (Figure 5a). Under UV excitation of this P-UCF, which is typically employed to read conventional fluorescent security labels, the blue Stokes fluorescence of its stripes obscured that of its QR code (Figure 5b). Under red excitation, in contrast, its QR code emitted unobstructed blue anti-Stokes fluorescence (Figure 5b). As a result, simple processing of the Stokes fluorescence micrograph of this P-UCF yielded an illegible black-and-white image, whereas analogous processing of its anti-Stokes fluorescence micrograph yielded a black-andwhite QR code that could be read by a smart phone or designated barcode scanner (Figure 5b). Moving forward, we could easily improve on the data storage capacity of these initial labels by using more intricate patterns.


15


Page 16 of 25

Figure 5. (a) Schematic showing the layers in a P-UCF superimposing a striped PU layer doped with perylene and a UC-PU layer patterned with a QR code and doped with PdTPBP and perylene. (b) Fluorescence micrographs of this P-UCF under xenon lamp excitation at 365 nm and laser excitation at 635 nm, converted to black and white through black/white inversion and brightness cutoff extraction using the graphics editor Inkscape. Only the QR code in the bottom image can be read using a smart phone or QR code scanner. To exclude excitation wavelengths, fluorescence micrographs were taken through either a 500 nm SP filter or a 420 nm LP filter. (c) Schematic showing the layers in a P-UCF superimposing two UC-PU layers, one patterned with a logo and doped with PdTPBP and perylene and the other patterned with stripes and doped with PdTPBP and 4,7-di(thiophen-2-yl)benzo[c][1,2,5]thiadiazole (DTPBTD). (d) Stokes fluorescence micrographs of this P-UCF under xenon lamp excitation at 420 nm and 365 nm, taken through 510 nm and 420 nm LP filters, respectively, to exclude excitation wavelengths. (e) Anti-Stokes fluorescence micrographs of this P-UCF under laser excitation at 635 nm, taken through a 500 nm SP filter to isolate perylene emission, a bandpass filter (ca. 510–600 nm) to isolate DTPBTD emission, and a 632.8 nm notch filter to show the emission of both layers while excluding excitation wavelengths. The PdTPBP, perylene, and DTPBTD concentrations employed in the PU layers were 0.52 µmol g-1, 12.7 µmol g-1, and 27.2 µmol g-1, respectively. All scale bars represent 100 µm.


16

Page 17 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Finally, by incorporating additional chromophores, we developed P-UCF labels with more rigorous security features involving multiple excitation and emission wavelengths. As a preliminary demonstration of this strategy, we superimposed two UC-PU layers containing PdTPBP paired with perylene and DTPBTD, respectively (Figure 5c). Incorporation of a second acceptor provided additional excitation and emission options, thereby introducing a number of new possibilities for photoluminescence. For instance, using appropriate longpass filters, yellowgreen DTPBTD Stokes fluorescence could be either isolated or viewed together with blue perylene Stokes fluorescence under blue or UV excitation, respectively (Figure 5d). Moreover, under red excitation, strategic optical filter choice made it possible either to isolate anti-Stokes fluorescence from each layer individually or to view anti-Stokes fluorescence from both layers simultaneously (Figure 5e).

Conclusions In summary, we characterized a new spin-coated PU host and produced the first flexible and photostable TTA-UC films by developing a multilayer thin-film architecture fabricated through facile spin-coating and compatible solution processing techniques that can be readily scaled up. We expect this innovative material design to significantly advance ongoing research efforts targeting integration of TTA-UC into SC devices, an application with potential definitively established in prior studies. Pioneering an application previously unexplored, we incorporated soft lithography into our fabrication procedure in order to produce UC films containing multiple superimposed layers with high-resolution microscale patterns. By customizing these films’ patterns and chromophore compositions, we created proof-of-concept covert security labels storing encoded data legible only under specific light sources. Their ability to respond to low


17


Page 18 of 25

excitation intensities (

Flexible and Micropatternable TripletâTriplet ... - ACS Publications

Recommend Documents

Flexible and Micropatternable TripletâTriplet ... - ACS Publications