Open Access
Add to BibSonomyAbstract
We report a new simple and inexpensive sub-micrometer two dimensional patterning technique. This technique combines a use of a photomask featured with self-organized particles in the micro- to nano-meter size range and a photoresist-covered substrate. The photomask was prepared by depositing monodispersed silicon dioxide (SiO2)- or polystyrene- spheres on a quartz substrate to form a close-packed pattern. The patterning technique can be realized in two configurations: a hard-contact mode or a soft-contact mode. In the first configuration, each sphere acts as a micro ball-lens that focuses light and exposes the photoresist underneath the sphere. The developed pattern therefore reproduces exactly the same spatial arrangement as the close-packed spheres but with a feature size of developed hole smaller than the diameter of the sphere. In the soft-contact mode, an air gap of few micrometers thick is introduced between the 2D array of self-organized spheres and the photoresist-covered substrate. In this case, a phase mask behavior is obtained which results in an exposure area with a lattice period being half of the sphere diameter. A 2D lattice structure with period and feature size of a developed hole as small as 750 nm and 420 nm, respectively, was realized in this configuration. We further applied this technique to host the deposition of organic films into the 2D nanostructure and demonstrated the realization of green and red nano-structured OLEDs.
© 2014 Optical Society of America
1. Introduction
In the past decades, organic light emitting diodes (OLEDs) have been widely studied for various applications including lighting and displays [1]. OLEDs with micro- or nano-scale feature size can have wide variety of applications. Examples ranging from bio-sensing [2], to emission sources for quantum encryption [3] and scanning near field optical microscopy [4] have been reported. Moreover, sub-micron patterns with lattice spacing in a 300 nm to 400 nm range are potentially interesting for two-dimensional distributed feedback (2D DFB) organic lasers operated in the visible spectral regime [5].
However, lithographic patterning of OLEDs extending large area of micro- or nano-scale feature size remains a challenge in the context of mass production. The well-known electron beam (e-beam) lithography technique offers a possibility of fabricating nano-OLEDs in a very controllable manner but only on a limited area, and thus can hardly meet the requirement for mass production [6]. With this regard, development of alternative nano-patterning processes has drawn attention. In particular, a method that uses a photoresist layer covered with self-organized micro-spheres and subject to UV exposure has been proposed to realize nano-arrays of holes and pillars [7]. In this mask-less process the micro-spheres are removed afterward in a wet process and therefore are not reusable. Nevertheless, a reusable mask featured with 8 µm-size SiO2 microspheres has been reported by Claire et al. and used for laser ablation with an excimer laser [8]. Moreover, Wu et al presented a technique using a reusable planar array of 1.5 µm micro-spheres embedded in a PDMS layer [9]. Such design acts as an array of ball-lenses to project the image of an illuminated pattern onto a photoresist layer.
It is worth mentioning that none of the above mentioned techniques were applied to the fabrication of organic light emitting diodes. To date, the main-stream techniques used for patterning the OLEDs are the e-beam lithography [6,10], the imprint techniques [11,12] and the soft lithography [1,13]. For the latter two approaches, a mould or a stamp prepared by the e-beam lithography and/or a dry etching (RIE) process are required. Laser-interfering lithography is another alternative [14]. In comparison, self-organized nano-particles based methods remain a simple, inexpensive, and high-throughput process. For example, Veinot et al. applied the nanosphere lithography process to realize sub-100 nm feature-size OLEDs [15]. Such nano-OLEDs were deposited inside the voids between the self-organized microspheres. However this technique requires repetitive deposition of the self-organized microspheres for each sample preparation. On the contrary, Bolognesi et al. used self-organized honeycomb-shape polymer films to realize a matrix for the imprint lithography [16]. In that case the resultant matrix of 1 µm to 1.5 µm lattice-space is re-useable.
In this paper we present an alternative method for introducing two-dimensional (2D) micro- and nano- patterning of OLEDs with a feature size in the 700 nm to 2.4 µm range. In our case, the patterning method is based on the conventional photolithography techniques but with a mask made of micro- and sub-micro sized SiO2 or polystyrene mono-dispersed spherical particles deposited in a self-organized manner on a quartz substrate. The advantages of this technique are that (i) the photomask is reusable, and (ii) the diameter of the micro-nanoparticles can be adapted to a desired lattice spacing in the 700 nm to 2.34 µm range. Our nano-structured mask can further be complied with the commonly used photolithography processes in terms of hard and soft-mode contact operation using a G-line optical light source of 405nm wavelength, photoresist and mask aligner. Compared with the aforementioned nano-patterning techniques, our approach is more simple, adaptable, and time/cost-effective. It can generate 2D hole patterns in a hard-contact mode with feature diameter down to 420 nm which is close to the diffraction limit with a photolithography light source at λ = 405 nm. This technique therefore proves a simple way to produce sub-micrometer patterns. In addition, under a soft-contact mode we notice that the period of the 2D lattices can be reduced by a factor of two compared with the diameter of the self-organized microspheres. By providing numerical analysis, we show that our nano-structured mask can have a function of phase-mask and thus provide 2D patterning with a feature size range from 800 nm to 2.34 µm as proved by the experimental work.
This paper is organized as follows: In section 2 we present a theoretical and numerical study of the proposed nano-structured lithography, whereas in the section 3 we provide the experimental procedures leading to the fabrication of 2D patterned OLEDs. Finally, in section 4, the experimental study of 2D patterns as a function of the particles size is presented.
2. Theoretical and numerical studies
This section is dedicated to the theoretical and numerical studies of the 2D patterns generated by the self-organized nanoparticles which play a role as an array of micro ball-lenses. We consider two experimental configurations which can be distinguished as the hard-contact mode and the soft-contact mode. First, let's consider the hard-contact mode. The behavior of a micro-ball-lens array in that configuration is illustrated in Fig. 1(a).It shows the intensity distribution of an optical field as calculated using 3D FDTD simulation with the particles in close contact with the photoresist.
Fig. 1 FDTD simulation illustrating the principle of microsphere based photolithography: NP: nanoparticles, SP: Spacing, PR: photoresist, SB: Substrate. The light is focused by the microspheres (a) Electric field distribution in the hard-contact mode. The green line indicates where the intensity profile is calculated. (b) Intensity profile in the hard contact mode. (c) Electric field distribution in the soft-contact mode. The green line indicates where the intensity profile is calculated; (d) Intensity profile in the soft-contact mode.
In this configuration, the incident beam at wavelength λ = 405 nm eminent from the Hg-light source of a mask aligner was assumed to travel in a downward direction. It successively passes through the air, the micro nano-particles (NP) (with diameter D and refractive index nbl), the photoresist (PR) (thickness e with refractive index nPR) and finally through the glass substrate (SB) with refractive index nSB. The optical field intensity was calculated at the middle of the photoresist layer with D = 1.68 nm, nbl = 1.47, e = 200 nm, nPR = 1.69, nSB = 1.5. From Fig. 1(b), an optical focusing effect due the micro-spheres can be observed as judged by the field intensity distribution with a period matching to the lattice constant D of the microspheres array. With this regard, each micro-sphere can be modeled as a micro ball-lens for light focusing. Another aspect is to consider the micro-sphere photomask in the soft-contact mode which can act as a phase-mask as illustrated in Fig. 1(c). There, a spacing (h = 1.5 µm) (SP on Fig. 1) has been introduced between the micro-sphere mask and the photoresist layer. In the latter case the optical fields arising from light passing through the adjacent micro-spheres can interfere with each other. This would lead to a phase-mask effect and introduce an optical field intensity modulation at a period to be half of the microsphere diameter D/2. Indeed, the intensity profile exhibits a local maximum at the plumb of each microsphere and another local maximum in the plumb of the contact points of the microspheres. The intensity profile in Fig. 1(d) exhibits a quasi period that corresponds to the half of the microsphere diameter D/2. The above statement can be understood by modeling the microsphere array of diameter D as a grating with a pitch D. The effect of phase-mask occurs as the optically transmitted beams from the grating order + 1 and −1 interfere constructively. In that case, the period of the fringes is exactly one half of the pitch of the phase mask. Therefore, in the soft-contact mode, a nanostructuration with a period D/2 at the half of the micro-sphere diameter D is expected.
At this step, we would like to specify the conditions under which the phase-mask effect can be obtained. The geometry of the system is presented in Fig. 2(a). Unlike the case of blazed gratings, and because of the symmetry of the structure, both of the transmitted beams for grating order + 1 and −1 can exist under the normal incidence condition if satisfying the condition shown in Eq. (1). The angle of transmission θ1 of the order m = 1 is calculated using the grating fundamental law:
Where nm is the refractive index of the medium underneath the grating. Depending on the distance of the photoresist to the microsphere, nm is either the refractive index of the photoresist nPR (hard-contact mode) or that of the air nar (soft-contact mode). Under normal incidence θi = 0, the orders ± 1 exist if the angle θ1 associated to the order 1 is smaller than the maximum angle θmax transmitted by the microspheres and related to their numerical aperture.Fig. 2 (a) Schematic showing the optical path and the interference conditions. MS1, MS2: Microspheres (center O1 and O2), D: diameter and lattice, BFL: back focal length, h: distance from the plane of focus to the plan of interference. (b) The aperture Angles θmax and the phase-mask angle of the orders ± θ1 as a function of the diameter D of the microspheres. (c) Effective distance S for the existence of the interference between the orders + 1 and −1 calculated from Eq. (8).
The back focal length (BFL), the effective focal length (EFL) and the numerical aperture NA of a ball-lens can be written, respectively, as follows [17]:
Where a is the aperture, which in our case of normal incidence, is the micro particles diameter d = D. The NA gives the maximum angle θmax for the transmitted optical field according to Eq. (4):The condition for coexistence of the transmitted grating orders ± 1 is:Figure 2(b) shows the dependence of θmax and θ1, respectively, as a function of the diameter D (red dashed line for θmax and blue solid line for θ1). Under our assumptions θmax is shown independent of D and it appears as a horizontal line in Fig. 2(b). Data shown in Fig. 2(b) further illustrates the fact that for D larger than a critical value D0, (i.e., 624 nm) the angle of the transmitted grating orders ± 1 can be smaller than the aperture angle. It follows that interference between these two grating orders may occur. The critical diameter Dθ for the self-organized micro-spheres to make a phase-mask is written as:Under this assumption and with the current parameters, Dθ is 624 nm. This imposes a lower limit on the diameter D of the microspheres for making such a phase mask. This limiting case of Dθ is indicated with a green solid vertical line in Figs. 2(b) and 2(c). Above Dθ , we have an angular condition of θ1 > θmax which leads to the coexistence of transmitted grating orders + 1 and −1. It imposes a theoretical lower limit Dθ = 624 nm for the case of forming phase-mask and thus causes a reduced lattice period of Dθ /2 which remains larger than the Abbe limit. Another aspect of wave interference can be referred Fig. 2(a) when the optical wave emitted from the microsphere 1 (MS1 center O1) with angle θMax interferes constructively at point A with wave emitted from the microsphere 2 (MS2 center O2).Knowing the distance D between the microspheres, it is straight-forward from the relation in the triangle HBA to calculate the distance h from the plane of focus to the plan of interference.
By substituting the angular function of θmax from Eq. (5) to Eq. (7) and note a linear relation between the distance S and h as follows:With further mathematical manipulation, we retain an S-D relation according to Eq. (9):Equation (9) is plotted on Fig. 2(c). This curve shows a minimum distance h = 0.7 µm to h = 2.8µm, according to a variation of micro-sphere diameter from 600 to 2400 nm, is required to achieve a phase mask behavior with a period of the fringes equal to the half of the microsphere diameter.3. Experimental procedures
This section is dedicated to the experimental demonstration of the 2D patterning applied to the fabrication of nano-structured OLEDs. The experiments are conducted in four steps: the synthesis of the nanoparticles, the fabrication of the photomask, the processing of the patterns, and the deposition of the OLED.
The experiments were started with the synthesis of different mono-dispersed nanoparticles made from SiO2 and polystyrene (PS). Nano-particles of different diameters were used in the experiments; 800 nm, 1 μm, 1.25 μm, 1.53 μm, 2 μm, and 2.34 μm for SiO2 and 1.68 μm for PS. The SiO2 microspheres were prepared by the modified Stober and sol–gel processes in an alcohol-rich phase [18]. Fresh alcohol, purified water, and ammonium hydroxide were firstly added into a sealed bottle and well mixed by a magnetic stirrer. Tetraethylorthosilicate (TEOS) was then introduced into the system. This would change the transparent solution into a milk-like solution. After stirring for 2 h, the milk like solution was subject to a centrifugal force to collect the SiO2 microspheres. The latter were washed with fresh alcohol three times before the SiO2 micro-nanoparticles are dispersed into an alcohol phase. The solid content of the dispersed SiO2 spheres was controlled to be about 10 wt% [19].
Our fabrication of the nano-structured photomask is based on a dip-coating method. The SiO2 micro-nanoparticles were dip-coated on both sides of a double-stacked quartz slides. The slides were immersed into the solution of micro nanoparticles and vertically withdrawn at a speed of 5 mm/s. The pair of coated quartz slide samples were separated and heated at 100°C in the air for 10 min to remove the residue of solution.
The mono-dispersed PS micro-particles were synthesized via emulsion polymerization at 70°C. After heating at 70°C for 24 hours, the latex was centrifuged. The PS spheres were collected and washed with methyl alcohol six times to remove the remaining impurities. They were then dispersed into a solvent. The solid content was controlled to about 10 wt %. After fabricating the PS spheres, they were deposited in a highly ordered self-organized structure on a quartz substrate by a dip coating method as in our previous study [20,21].
Figure 3 illustrates the proposed fabrication process of micro nano-OLEDs. The patterning process started with the cleaning of ITO coated glass used as a substrate for the OLEDs. AZ-1505 (MicroChemicals) was used as a positive photoresist and diluted with AZ-EBR solvent (MicroChemicals) with a volume ratio = 2:1 (2 parts for 1 part of AZ-1505). The photoresist was then spin coated on the ITO coated substrate at 6000 rpm for 60 seconds. The thickness of the deposited photoresist was 190 nm as measured by profilometer (Alpha-Step IQ Surface Profiler). The photoresist at the edge of the substrate was thicker as more resist is accumulated during the spin coating process. This edge bead was removed by AZ-EBR solvent to improve the flatness of the samples and ultimately to improve the contact between the surface of the photoresist and the microsphere mask.
Fig. 3 Fabrication process of micro-nanoOLEDs. (a) schematic of the contour mask (left) and of the photomask made of self-organized micro-nanoparticles (right). (b) the different steps of the process.
The sample was soft-baked for 120 seconds at 100°C before UV exposure. A printed plastic mask (see Fig. 3(a)) with 4˟5 mm2 rectangular transparent region surrounded by UV blocking region was used as a contour mask to select the region where the pattern is to be transferred onto the substrate. This ensures that the OLED area is defined by the region defined by the transparent part of the plastic mask. The photomask was prepared by forming a monolayer of self-organized silica micro-spheres on a pre-cleaned quartz substrate as mentioned above. As shown in Fig. 3(b), the part of photomask containing the single layer of micro- and nano-particles was brought into contact with the photoresist in either of the hard-contact or the soft-contact mode in compliance with operation of the mask aligner. The sample was then subject to UV light exposure (12.5 mW/cm2) through the monolayer of micro-spheres mask for 0.9 second. The exposed resist was then removed by AZ-726-MIF (MicroChemicals) developer for 9 seconds. The sample was then rinsed in deionized water to remove the remaining developer and blow-dried with nitrogen.
As an example, we have fabricated two types of OLED devices with red and green emission despite blue emission was also demonstrated in the lab although not shown here. The OLED layers were deposited onto the opening holes of the patterned photoresist, by evaporation at a vacuum pressure of 2.10−7 Torr. The energy diagrams of the green and red organic hetero-structures are shown on Figs. 4(a) and 4(b), respectively. The green heterostructure was composed of a 40 nm thick 4,4’,4”,tris-(3-methylphenylphenylamino) triphenylamine (m-MTDATA) layer as hole injection layer (HIL), a 15 nm of N,N'-diphenyl-N,N'-bis(1-naphthyl)-1,1'-biphenyl-4,4-diami (NPD) as a hole transport layer, a 50 nm of Tris(8-hydroxyquinolinato) aluminum (AlQ3) host material as the emissive layer (EL) with an intrinsic peak emission centered at 520 nm. The organic layers were followed by a 1 nm of lithium fluoride (LiF) and a 150 nm thick aluminum (Al) layer as a cathode that was vapor deposited at the same background pressure.
Fig. 4 Band diagram of the organic heterostructures (a) organic heterostructure for green emission (b) organic heterostructure for red emission.
The red organic heterostructure was constructed as follows: a 40 nm thick m-MTDATA layer, a 15 nm thick NPD layer, a 30 nm of AlQ3 doped DCM2 (4-(dicyanomethylene) −2–methyl −6-(p-dimethylaminostyrl) −4H-pyran) as the red emitting layer. The emitting zone was embedded within the 30 nm thick AlQ3:DCM2 emitting layer. The heterostructure was completed with a 10 nm thick BCP (bathocuproine: 2,9-Dimethyl-4,7-diphenyl-1,10-phenanthroline)hole blocking layer, and a 10 nm thick Alq3 layer as an electron emitting layer. Finally we added a 1 nm thick lithium fluoride (LiF) and a 150 nm thick aluminum layer to form the cathode. In total, the OLED hetero-structure thickness without the aluminum layer was 105 nm with an active area of 4˟5 mm2.
4. Experimental results
The effects of the geometric parameters on the evolution of surface morphology for the 2D nano-structured can be verified with a Scanning Electron Microscope (SEM) and an Atomic Force Microscope (AFM) at each step of the process. Figure 5(a) presents a panoramic view of the photomask made with a single layer of self-organized polystyrene micro-particles of 1.68 µm in diameter on a 17˟25 mm2 quartz substrate. The monolayer of micro-particles covers the whole quartz substrate and is uniform without significantly aggregates at a glance. A Scanning Electron Microscope (SEM) image of a single layer made with 1.68 µm polystyrene micro-particles shown on Fig. 5(b) indicates the regularity of the self-organization. Possibly, some line defects may appear.
Fig. 5 (a) Optical microscope image of the 17x25mm photomask covered with 1.68 µm self-organized polystyrene micro-nanoparticles. (b) SEM image of the self-organized polystyrene micro-nanoparticles (1.68 µm-diameter) (c) SEM image of the patterned photoresist (1.68 µm lattice). (d) Atomic force microscope image of a patterned photoresist (1.68 µm lattice). (e) Unpatterned glass substrate with the ITO layer (f), (g) (h) different diffracted colors of the patterned substrate for different viewing angles.
A SEM image of the resultant patterned photo-resist is shown on Fig. 5(c) with a 1.68 µm lattice. Note that possibly the line defects of the self-organization are reproduced on the patterned photo-resist. This is a clear indication of a direct transfer mechanism of the micro-particle array onto the photoresist and suggests the micro ball-lens focusing effect and excludes the phase-mask behavior. The diameter of the resultant hole diameter is 1.20 ± 0.08 µm and is smaller than the micro-particles diameter composing the nano-structures upon the mask. These observations confirm the advantageous use of such ball-lens like micro-spheres on the photo-mask for fulfilling the focusing effects. To measure the depth of the holes and to make sure that the exposed photo-resist had been etched down to the ITO layer, the surface morphology of the patterned photo-resist was observed with an atomic force microscope in contact mode (AFM Nanosurf). Figure 5(d) presents the AFM images of the photo-resist pattern after the development procedure as shown in Fig. 3(b) with a photo-mask composed of arrays of 1.68 µm-dia. nano-particles. The depth of the holes measured by AFM microscope was comparable to the thickness of the photoresist measured by the profilometer, before the patterning was performed. This confirms that the ITO is locally left bare without being covered by the photoresist. The latter ensures that an electric contact path can exist between the ITO and the organic layers. Note that in contrary to the unpatterned substrates (Fig. 5(e)), the processed substrates exhibit different coloration due to iridescence from different angles (see Figs. 5(f)-5(h)). The latter is an evidence of supporting nano-structured patterns on the whole ITO covered glass.
Figures 6(a)-6(c) present two types of micro-OLEDs. A red micro-OLED array based on the organic heterostructure presented in Fig. 4(b), patterned with a photomask made with 1.68 µm polystyrene micro-particles and resultant hole diameters of 1.20 µm, is shown on Fig. 6(a). The peak emission wavelength of this device is at λ = 620 nm. The photography was taken under a 20X objective with the device biased at a turn on voltage of 10 V and a driving current of 30 mA. The second type of devices is an array of green micro OLEDs shown on Figs. 6(b) and 6(c) taken with a 50 X and a 20X objectives, respectively. It corresponds to the green organic hetero-structure shown on Fig. 3(a), structured with 1.68 µm micro-particles mask, and has a resultant hole diameter of 1.20 µm.
Fig. 6 Red and green Micro-OLEDs. (a) Optical microscope image of an array of red OLEDs (λ = 620 nm, diameter is 1.2 µm, size of PS micro-nanoparticles is 1.68 µm) (b) Optical microscope image of an array green OLEDs (λ = 520m OLED diameter is 1.2 µm, size of PS micro nanoparticle is 1.68 µm), (c) Optical microscope image of large area of a green OLED (OLED diameter is 1.2 µm, size of PS micro nanoparticles is 1.68 µm).
The current-versus-voltage (IV) characteristics and the luminance-versus-voltage (LV) characteristics of the green patterned and the unpatterned (reference) OLEDs are shown in Fig. 7.At 8.5 V, the patterned OLED was driven with 16 mA and shown under normal incidence a luminance of 2142 cd/m2 with 2.35 lm/W of power efficiency. In comparison the unpatterned green reference OLED biased at 8.5 V was driven with 30mA, and a luminance of 5355 cd/m2 with an efficiency of 3.28 lm/W. For the nano-structured patterned device, the active area represents 46% of the unpatterned OLED. Taking into account the uncertainties on the mono-disperssion of the microparticles (5%) and on the defects in the self-organization pattern, the reduction by 46% of the active area explains the 40% decrease in the luminance. This is a clear demonstration that this patterning method based on the nano-particles photolithography remains compatibles with the OLED deposition process. We speculate that the reduction in the active area is expected to decrease the luminance by the same amount. Since we observe a decrease of 60% in the luminance instead of 54%, it is very likely that the active area is in reality smaller than what the calculation predicts. This can be explained with a non-uniformity of −15% (0.46-0.4)/0.46). The negative sign indicates that the dark spots participating to the active area reduction suppress more light than what the bright spots add.
Fig. 7 IVL curves of the green patterned and unpatterned OLEDs. Current as function of the voltage (IV) (left). Luminance as a function of the voltage (LV)(Right).
In this section, we investigate the lattice parameters as a function of the nanoparticle size in the range 2.34 µm to 800 nm. The idea is to demonstrate the possibility of scaling-down the nano-micro-structures using this patterning technique. Moreover, we would like to investigate the phase-mask behavior of micro-nanoparticle arrays and demonstrate the possibility to obtain pattern periods equal to half of the particle diameter D.
For this reason, and as an intermediate step, the photolithography process was repeated with 1.68 µm polystyrene micro-nanoparticles on an irregular photoresist layer. This induces variations of the spacing between the photoresist and the photomask. As a result, unreduced (D) and reduced (D/2) lattice periods clearly appear on the same sample (see Fig. 8).Despite a lack of control in the positioning of the different patterns, we present the Fig. 8 for the sake of the demonstration. This is indeed a clear experimental demonstration that two patterning configurations exist and may coexist with the same photolithography mask made with the same diameter D of self-organized micro nanoparticles: On the one hand the focusing effect of the ball-lens like microspheres that results in a pattern with a period D, and the other hand a phase-mask behavior that produces nanostructures with period D/2.
Fig. 8 Optical microscope image showing a patterned photoresist with an irregular thickness. The pattern exhibits two lattice periods. D = 1.68 µm (unreduced) and D/2 = 840 nm (reduced) periods created on the same sample with the same photomask. (size of spheres is 1.68 µm).
In order to investigate the diameter dependence of the phase mask properties, the process was systematically repeated in both the hard-contact and soft-contact modes for the different sizes of the SiO2 and Polystyrene micro-nanoparticles measured by SEM (800 nm, 1 μm, 1.25 μm, 1.53 μm, 1.68 µm, 1.8 µm, 2 μm, and 2.34 μm). For the different particle sizes, the hole diameters and the lattice periods were measured in both the hard-contact and soft-contact modes. The results are summarized in the Fig. 9.The blue dots corresponding to the period of the nanostructures obtained in the hard-contact mode, are perfectly aligned and exhibits a slope of 1 ± 0.1 which confirms that the self-organized micro-nanoparticles lattice is exactly reproduced in the photoresists with a period equal to the particle diameter D. The hole diameters (empty circle) of this unreduced patterns exhibit a lower alignment with an averaged slope obtained by a linear regression of 0.69 ± 0.1 indicating the proportion of the hole diameters to the particles size.
Fig. 9 The dependence of the lattice and of the size of holes as a function of the micro nanoparticle diameter: ● The period of the pattern in the hard-contact mode (unreduced period), ○ Diameter of the holes in the hard-contact mode, ■ Reduced period of the pattern in the soft-contact mode, □ The diameter of holes in the soft-contact mode.
We emphasize that the same self-organized micro-nanoparticles based photolithography masks were used both in the-soft contact and hard-contact modes. Clearly, the same mask allows an unreduced period pattern (see sample S3, S5) in hard-contact mode and a reduced period pattern (S4, S6) in the soft-contact mode as already demonstrated on Fig. 8. The black squared line on Fig. 9 clearly indicates a two-fold reduction of the pattern periods compared to the particles diameter used for the photomask. Indeed, a linear regression gives a slope of 0.51 ± 0.1 indicating that the reduced pattern period is D/2 as expected from the interference between the orders + 1 and −1 in a phase-mask with a pitch D.
The hole diameters obtained in the soft-contact mode (empty squares in Fig. 9) exhibit a linear regression with the diameter D giving a slope of 0.35 ± 0.1 which corresponds to the half of the hole diameters obtained in the hard-contact mode. Note that the smaller hole diameter obtained with 1.25 µm-microparticles (sample S4) was 420 nm which is close to the diffraction limit with a light source at λ = 405 nm. In addition to that, the fabrication of nanostructures with a reduced period D/2 were not observed with particle diameter smaller than 1.25 µm, whereas the theory predicted a limit on the particle size at D0 = 624 nm which was not achieved here. We believe that the difference is due to the fact that the soft-contact mode introduces a too large gap G between the photomask and the PR layer compared to the spacing S required for particles size below 1.25 µm taking into account a tolerance T. More precisely, we assume the gap distance in the soft-contact mode to be about 2 µm < G <5 µm. Based on Fig. 2(c), for D <1 µm dia. the distance S <1.5 µm. Based on Fig. 1(c), the tolerance estimated from the z range of existence of the interferences is approximately T = 0.5 µm. We speculate that the reason why the reduced pattern is not obtained when D < 1.25 µm is that S + T < G.
Finally, we fabricated two types of 2D reduced arrays of green OLEDs based on 1.68 µm PS micro-nanoparticles and 1.25 µm SiO2 micro-particles in the soft-contact modes resulting in 910nm and 750 nm lattices with 650 nm and 420 nm hole diameters, respectively.
5. Conclusion
As a conclusion, we numerically and experimentally studied the 2D nano-structuration of a photoresist deposited on the ITO anode of an OLED. The nano-structuration process was based on photolithography and a reusable mask made with a self-organized monolayer of SiO2 and polystyrene micro nanoparticles. This work included a study of the lattice parameters (period and hole diameter) of the patterned photoresist as a function of the diameter D of the micro nanoparticles ranging from 800 nm to 2.4 µm. Depending on the spacing between the photomask and the photoresist, two different configurations can be distinguished: the hard-contact and the soft-contact modes.
In the hard-contact mode (i.e. the mask was in close contact with the photoresist) each micro nanoparticle behaved like a micro-ball-lens that focuses the UV light on the photoresist. This creates a periodic variation of the energy absorbed by the photoresist resulting in a periodic arrangement of holes. In this case the period of the two-dimensional pattern is equal to the particles diameter D.
In the soft-contact mode, the distance between the photomask and the photoresist was of few micrometers. This allowed interferences to occur in a similar manner as in a 2D phase-mask. The interferences of the orders + 1 and −1 produced a pattern with a lattice period equal to the half of the micro nanoparticles diameter (D/2). Period as low as 700 nm with holes diameters down to 420 nm were achieved with a simple photolithography process at the wavelength of 405 nm using a mask made of self-organized 1.25 µm micro-nanoparticles.
Finally, two different organic hetero-structures were then evaporated on the patterned photoresist deposited on an ITO anode in order to fabricate red and green micro-nano OLEDs. We believe that this process can still be applicable for smaller size of micro nanoparticles (range 400 nm – 800 nm). As a perspective, we will extend this process with shorter wavelength light sources (365 nm, 193 nm) to produce 2D-DFB laser resonators compatible with the visible emission of organic hetero-structures.
Acknowledgment
This work was supported by the Agence National de la Recherche (OLD-TEA project), the RTB +, the Région Ile de France (C'nano IdF), and the Fonds Européens de Développement Economique Régional (FEDER) as well as the Labex SEAM. We also thank the Bureau de la Représentation de Taipei en France (BRTF) for its support in the framework of the Taïwan Fellowship 2013.
References and links
1. S. P. Price, J. Henzie, and T. W. Odom, “Addressable, large-area nanoscale organic light-emitting diodes,” Small 3(3), 372–374 (2007). [CrossRef] [PubMed]
2. M. Vasilopoulou, D. G. Georgiadou, L. C. Palilis, A. Botsialas, P. S. Petrou, S. E. Kakabakos, and P. Argitis, “Photopatterned PLED arrays for biosensing applications,” Microelectron. Eng. 86(4-6), 1511–1514 (2009). [CrossRef]
3. H. Yamamoto, J. Wilkinson, J. P. Long, K. Bussman, J. A. Christodoulides, and Z. H. Kafafi, “Nanoscale organic light-emitting diodes,” Nano Lett. 5(12), 2485–2488 (2005). [CrossRef] [PubMed]
4. J. Lu, H. W. Lin, H. Y. Chen, Y. C. Yang, and S. Gwo, “Single InGaN nanodisk light emitting diodes as full-color subwavelength light sources,” Appl. Phys. Lett. 98(23), 233101 (2011). [CrossRef]
5. S. Riechel, C. Kallinger, U. Lemmer, J. Feldmann, A. Gombert, V. Wittwer, and U. Scherf, “A nearly diffraction limited surface emitting conjugated polymer laser utilizing a two-dimensional photonic band structure,” Appl. Phys. Lett. 77(15), 2310–2312 (2000). [CrossRef]
6. F. A. Boroumand, P. W. Fry, and D. G. Lidzey, “Nanoscale conjugated-polymer light-emitting diodes,” Nano Lett. 5(1), 67–71 (2005). [CrossRef] [PubMed]
7. W. Wu, D. Dey, O. G. Memis, A. Katsnelson, and H. Mohseni, “A novel self-aligned and maskless process for formation of highly uniform arrays of nanoholes and nanopillars,”',” Nanoscale Res. Lett. 3(3), 123–127 (2008). [CrossRef]
8. C. O’Connell, R. Sherlock, and T. J. Glynn, “Fabrication of a reusable microlens array for laser-based structuring,” Opt. Eng. 49(1), 014201 (2010). [CrossRef]
9. M. H. Wu and G. M. Whitesides, “Fabrication of arrays of two-dimensional micropatterns using microspheres as lenses for projection photolithography,” Appl. Phys. Lett. 78(16), 2273 (2001). [CrossRef]
10. H. Yamamoto, J. Wilkinson, J. P. Long, K. Bussman, J. A. Christodoulides, and Z. H. Kafafi, “Nanoscale organic light-emitting diodes,” Nano Lett. 5(12), 2485–2488 (2005). [CrossRef] [PubMed]
11. K. Trivedi, U. S. Bhansali, B. Gnade, and W. Hu, “The fabrication of high density nanochannel organic light emitting diodes with reduced charge spreading,” Nanotechnology 20(40), 405204 (2009). [CrossRef] [PubMed]
12. X. Cheng, Y. Hong, J. Kanicki, and L. J. Guo, “High-resolution organic polymer light-emitting pixels fabricated by imprinting technique,” J. Vac. Sci. Technol. B 20(6), 2877 (2002). [CrossRef]
13. C. Piliego, M. Mazzeo, B. Cortese, R. Cingolani, and G. Gigli, “Organic light emitting diodes with highly conductive micropatterned polymer anodes,” Org. Electron. 9(3), 401–406 (2008). [CrossRef]
14. Y.-C. Kim and Y. R. Do, “Nanohole-templated organic light-emitting diodes fabricated using laser-interfering lithography: moth-eye lighting,” Opt. Express 13(5), 1598–1603 (2005). [CrossRef] [PubMed]
15. J. G. C. Veinot, H. Yan, S. M. Smith, J. Cui, Q. Huang, and T. J. Marks, “Fabrication and properties of organic light-emitting nanodiode arrays,” Nano Lett. 2(4), 333–335 (2002). [CrossRef]
16. A. Bolognesi, C. Botta, and S. Yunus, “Micro-patterning of organic light emitting diodes using self-organised honeycomb ordered polymer films,” Thin Solid Films 492(1-2), 307–312 (2005). [CrossRef]
17. H. Zappe, Fundamentals of Micro-Optics, (Cambridge University, 2010).
18. W. Stöber, A. Fink, and E. J. Bohn, “Controlled growth of monodisperse silica spheres in the micron size range,” J. Colloid Interface Sci. 26(1), 62–69 (1968). [CrossRef]
19. C.-H. Chan, A. Fischer, A. Martinez-Gil, P. Taillepierre, C.-C. Lee, S.-L. Yang, C.-H. Hou, H.-T. Chien, D.-P. Cai, K.-C. Hsu, and C.-C. Chen, “Anti-reflection layer formed by monolayer of microspheres,” Appl. Phys. B 100(3), 547–551 (2010). [CrossRef]
20. C.-H. Chan, C.-H. Hou, C.-K. Huang, T.-J. Chen, S.-Z. Tseng, H.-T. Chien, C.-H. Kuo, K.-H. Hsieh, Y.-L. Tsai, K.-C. Hsu, and C.-C. Chen, “Patterning periodical motif on substrates by monolayer of microspheres: Application on GaN light-emitting diodes,” Jpn. J. Appl. Phys. 48, 020212 (2009). [CrossRef]
21. C.-H. Hou, S.-Z. Tseng, C.-H. Chan, T.-J. Chen, H.-T. Chien, F.-L. Hsiao, H.-K. Chiu, C.-C. Lee, Y.-L. Tsai, and C.-C. Chen, “Output power enhancement of light-emitting diodes via two-dimensional hole arrays generated by a monolayer of microspheres,” Appl. Phys. Lett. 95(13), 133105 (2009). [CrossRef]
