1. Introduction

Simple sub-micron and nano- patterning techniques are gaining interest for applications that require on-site fabrication of 1D nanopatterns (nano-grating and nano-waveguides). Over the years several techniques have been proposed that are complex, bulky and require sophisticated facilities. Existing lithography techniques can be broadly categorized as, mask-based (photolithography and its variants) and mask-less (direct laser writing, laser interference lithography, STED lithography, electron beam lithography) [1,2,3]. A major constraint with the existing light based nanolithography technique is the feature size that is limited by the Abbe diffraction limit. However, STED based techniques have been shown to achieve a resolution below the diffraction limit for nanolithography [4,5]. Other key issue associated with nanolithography is its inability for axial z-plane selectivity and lack of control over the patterning area. Existing techniques does not allow patterning on a selective plane in a 3D volume. Selective nature ensures that specific plane can be patterned without affecting nearby planes. The ability to vary patterning area for specific applications need control and brings flexibility. Light-based lithography techniques such as laser interference lithography (LIL) has the ability to address these issues. In LIL, intensity of the interference pattern produced by a small number of coherent beams is recorded on the photoresist and the pattern is subsequently developed by well known chemical protocols [6,7]. The separation between two successive features is approximately, $\lambda /2\eta sin\theta$; where, $\theta$, $\lambda$ and $\eta$ are respectively the inclination angle, wavelength of light and refractive index of the medium [8,9]. Few recent techniques that are capable of large area patterning are, single refracting prism holographic lithography and phase spatial light modulators interference lithography [10,11,12,13]. Prism holographic lithography has shown large area two-and three-dimensional submicron fabrication of polymer photonic crystals by manipulating the polarizations of the interfering beams using a prism. Whereas, phase spatial light modulator (SLM)-assisted interference lithography have demonstrated lsub-micrometer photonic structures on photoresist. It may however be noted that, proposed iCLASS technique is an alternate fabrication technique. In general, LIL has the advantage of generating periodic structures that makes it useful for a wide range of applications in photonic crystals [14], energy devices [15], magnetic storage devices [16], health-care [17,18], broad band reflectors [19], diffraction gratings [20], solar cells [21] and plasmonic biosensors [22,23].

2. iCLASS system design and theory

Light-sheet based technique has been proposed for fabricating simple micro/nano-structures [24–26]. However the experimental validation and the fabrication of nano-structure was not shown. Here we report the fabrication of nano-structures using light-sheet technique for the first time. This technique is well-suited due to its specificity, plane-selectivity and positionable patterning area. It uses coherently-illuminated counter-propagating interfering light-sheets for fabricating nanostructures. A distinct feature of this technique is the $2\pi$-geometry that allows single-shot fabrication of an entire specimen plane. The schematic diagram of light-sheet based nanolithography system is shown in Fig. 1. A coherent laser beam (Laserglow LSS-052, 532 nm, coherence length, $l_{coh} >280~m$) is appropriately expanded using a beam-expander (BE), linearly-polarized using LCR and directed towards the beam-splitter (BS). The coherent light is split into two parts: the first part is directed to the mirror M3 while the second part pass through a movable retro-reflector (RR-tS) to reach the mirror M4. RR-tS is placed on a linear translator so as to introduce necessary change in path-difference for interference to occur. The beams are then reflected by mirrors, M3 and M4 to the cylindrical lenses, C1 and C2 resulting in a pair of counter-propagating light-sheets. Note that path of beam in the illumination arm A2 is altered using RR-tS to enable interference of light sheets at the common geometrical focus of C1 and C2 (see, Fig. 1). The specimen to be patterned is placed at the focus and subsequently standard chemical protocols are followed to develop the pattern.

 

Fig. 1. Schematic diagram of the proposed light-sheet based iCLASS nanopatterning system. The technique involves splitting the light using $50:50$ beam-splitter BS followed by phase-change using translational retro-reflector RR-tS. The beams along optical arms A1 and A2 produce counter-propagating light-sheets after passing through a pair of cylindrical lenses, C1 and C2. These light-sheets interfere to produce the resultant pattern at the common geometrical focus.

Download Full Size | PDF

The scalar theory behind the proposed iCLASS system is essentially based on the interference of counter-propagating light waves. However, the patterning area is defined by the lateral-size of overlapping lightsheets. In general, the interference of two counter-propagating waves ($e^{ikx}$ and $e^{-ikx}$) give rise to an intensity distribution at the common geometrical focus, $I=2I_0[1+\cos (\Delta \phi )]$, where, $\Delta \phi = (\phi _1 -\phi _2)=2kx$ is the phase-difference, $\Delta x =2x$ is the path-dfference and $k$ is the wave-number of light. The maximum (of the intensity distribution) occurs when, $\Delta \phi =2kx=2n\pi ~\Rightarrow x=n\lambda /2$, where, $n$ is an integer. So the period of the resulting pattern is, $\lambda /2$, further suggesting that a selective plane can be patterned with a period of $\lambda /2$. However in practice, the photochemical nature of photopolymer film may alter the periodicity. A much more precise and explicit expression for computing the field at the focus can be found using vectorial theory of light-sheet [27]. In addition, $iCLASS$ offers user-defined patterning area (decided by the dimension of the interfering light-sheets assuming complete overlap) with plane selectivity.

3. Results and discussion

The goal is to exploit selective plane illumination property of interfering light-sheets for transferring the nano-pattern on a photopolymer system. A commercially available positive photopolymer (Michrochem-S1813) is exposed to the interfering pattern. Specific protocol for the development of photopolymer film is followed as shown in Fig. 2. Note that, the exposure takes place between soft-bake and development. Corresponding photopolymer sensitivity graph is also shown. UV-Vis absorption spectra is recorded using a UV-Vis spectrometer ( SHIMADZU, UV- 2600). The absorption spectrum was recorded by diluting the photopolymer mixture in MF319 ($1 ~\mu l$ in 1 ml of MF319) and by keeping MF319 as a Ref. [28,29]. S1813 has relatively weak absorbance for visible region of the electromagnetic spectrum. This ensures depth-penetration over the thickness of photoresist film. We have used relatively higher laser power to account for the decrease in photosensitivity. For the sample preparation, a coverglass (of size, $1~cm \times 1~cm$) is immersed in Chromic acid cleaning solution at $70^o \pm 2^o C$ for 15 min to remove organic and inorganic contaminants. Subsequently, the substrates were coated with Michrochem-S1813 photoresist by drop-casting method. Samples were soft-baked at $115^o \pm 2^o C$ for 30 min and developed using MF319 buffer (see, Fig. 2). $10 ~\mu l$ of S1813 on a $1~cm \times 1~cm$ coverglass has resulted in a film thickness of $\approx 300 ~\mu m$. Thickness measurements were carried out on a dektak surface profiler. Exposed samples are developed for 1 min in MF319 buffer solution. After development, the substrates were rinsed in ample of distilled water for 10 min and left for drying. Exposure of the photoresist is carried out in dark environment. The protocol as mentioned in Refs.[30,31,32,33] is closely followed. S1813 is tuned for g-line exposure (436 nm). Hence determining a optimum exposure dose for 532 nm illumination is necessary for achieving patterns on the photoresist.

 

Fig. 2. Step-by-step protocol for exposure and development of S1813 photoresist film. The UV-Vis absorption spectra of Michrochem S1813 (for a photoresist film thickness of about $300 ~\mu m$) is also shown.

Download Full Size | PDF

As a first step towards nanopatterning, it is imperative to determine the optimum light-dose interaction-time, hence experiments were carried out for different values of exposure time by keeping the illumination intensity and the developing time constant. In the present experimental study, we used a light of wavelength, $\lambda =532~nm$ operating at a laser power (W) of 18 mW. Since light- sheets are used for generating the interference structure, the irradiated area is about, $8000 ~\mu m^2$. The samples were mounted perpendicular to the direction of illumination as shown in Fig. 2. The exposure time is varied from 5 s to 20 s in steps of 5 s. Exposed samples were developed for 1 min in MF319 buffer solution. Developing mainly involves the diffusion of the solvated polymer chains into the developing solution. Developing time generally depends on the thickness of the photoresist film, exposure dose, temperature of the developing solution and also on the geometry of the patterns. The substrates were subsequently rinsed in ample distilled water for about 10 minutes and left in air. All the procedures were performed at a temperature of $18^o \pm 2^o C$ under low light conditions. The results obtained are as shown in Fig. 3. The stripe corresponds to the patterned region and it varies with change in exposure time. An exposure time of $\tau _{exp}=15 ~s$ is found to be optimal for fabrication. Exposure time of less and more than $15 ~s$ show aberration (also known as dark-erosion effect) driven by the chemical nature of photopolymer [34]. The combination of low exposure dose and developing time show dark-erosion effect in the developed samples. Figure 3 shows that the dark-erosion effect is strong for 5 s exposure (with a 1 min developing time) and a relatively weak effect can be seen for an exposure of 10 s. The effect is found to be minimum for a 15 s exposure time. Results suggest that optimal exposure time is critical and an effective balance between exposure time and development time is essential for quality fabrication.

 

Fig. 3. Effect of light-exposure time on photoresist film as captured by optical microscope (Lateral XY-view). An exposure of $\tau _{exp}=15~s$ combined with a development time of $\tau _{dev}=60~s$ is found to be optimal for nano-fabrication.

Download Full Size | PDF

Figure 4 shows AFM images of the photosensitive specimen exposed for 15 s (exposure time) and variable development time. All AFM images are recorded from Bruker AFM Probes operated in non-contact mode. The probe material is a n-type semiconductor with a tip height of $12-18 ~\mu m$ and the radius of the uncoated tip is $8 ~nm$. A full top angle of $40^o$ is used in the present measurement. The images are recorded at the periphery (edge) of the photoresist. AFM images clearly shows that the contrast is low for $\tau _{dev}=30~s$ when compared to images for 45 s and 60 s. This is purely due to insufficient developing time. In addition, we noticed isolated dots in Fig. 4. These dots correspond to dust particles that may have deposited during AFM measurement process. The intensity profile plots across the 1D pattern are also shown. The average peak-width for each case is calculated from the profile plots. The experimental results show slightly large periodicity of 560 nm, 460 nm and 366 nm for a light-dose interaction-time of (15 s, 30 s), (15 s, 45 s) and (15 s, 60 s) respectively. The average peak-width for the optimal light-dose interaction time $(\tau _{exp}, \tau _{dev})= (15~s, 60~s)$ is found to be approximately, 366 nm. We attribute the variation in periodicity primarily due to experimental errors, light-dose interaction time and sensitivity of the photoresist that is determined by its composition [35]. We conclude that peak-width and periodicity depend heavily on the photochemical nature of the polymer film. Fig. 4 also shows the intensity profile plots taken across the channels. The corresponding parameters such as, average peak-width ($W_c$), peak-height ($H_c$) and periodicity ($P_c$) are subsequently calculated for each case. The Aspect-Ratio ($AR =W_c : H_c$) for varying $(\tau _{exp}, \tau _{dev})$ is tabulated in Table. 1. This indicates a healthy aspect-ratio for nano-patterning applications.

 

Fig. 4. Visualization of the developed 1D pattern using atomic force microscope (AFM) at an exposure time($\tau _{exp}$) of 15 s and varying development time ($\tau _{dev}$) of 30 s, 45 s and 60 s . The corresponding intensity line plots are taken across the pattern that shows variable channel-width thereby signifying the importance of interaction time ($\tau _{exp}, ~\tau _{dev}$) in nanofabrication. Bright dot-like pattern correspond to dust particles.

Download Full Size | PDF

Table 1. Aspect ratio and periodicity calculations.

View Table

4. Conclusion

Counter-propagating light-sheet based iCLASS patterning technique is an alternate route to fabricate simple 1D sub-micron structures. At the heart of the system, we have $2\pi$-optical configuration that enables sheet-based 1D interference pattern. Proposed technique enables plane-selective 1D sub-micron structure patterning. Experimental results predict nano-patterns of the size few hundred nanometers. A strong dependence of light-dose interaction with the photoresist system is observed which ultimately determines the periodicity and quality of the fabricated nano-pattern. The novelty of the proposed technique lies in its ability to generate nano-pattern on a photoresist using interfering coherent light-sheets on a pre-defined patterning area. Light sheet based lithography has the distinct advantage of selective plane patterning in a 3D substrate volume. In addition, proposed light sheet based fabrication technique enables patterning on well-defined area with high precision. This is essential for applications (such as, nano-plasmonics and nano-optical devices) that require flexibility and control over the patterning area (determined by the lateral dimension of light sheet). These advantages make light sheet as a potential candidate for designing nanofluidic chips and sub-micron sized optical components (nano-grating and nano-waveguide). However, the limitations of iCLASS technique could be depth-penetration and sensitivity. In addition, complex structures and 3D patterning could be challanging. We envision that the proposed technique may enable fabrication of nano-components such as, nano-channels, nano-gratings and nano-waveguides [36,37,38]. Since the technique is miniaturizable, it may be a potential candidate for onsite fabrication. The applications are in nanobiotechnology, nano-optics and health-care.