1. Introduction

In recent years there has been significant interest in developing materials that mimic naturally occurring systems in order to take advantage of certain useful behaviors, characteristics and properties that nature provides [1]. Such research includes the mimicking of hydrophobic lotus leaves [2, 3], firefly lanterns [4] and butterfly wings [5, 6], as well as mimicking the color-changing skins of chameleons [7, 8], the adhesive grip of geckos [9, 10] and rapid camouflage behavior of cephalopods [11, 12]. Through much research, it has been determined that many of these naturally occurring characteristics and abilities feature specifically patterned sub-micron structures that assist in imparting the exhibited, desirable properties.

The surfaces of moths’ eyes comprise another biological material that has received significant attention for its transmission and anti-glare capabilities [13–16]. Moths’ eyes, which have thousands of surface nanostructures, transmit a high percentage of incident light which allows the moth to see in low-light circumstances. Furthermore, the nanostructured surface of moths’ eyes provides anti-reflection (AR) character which prevents glare, thus allowing the moth to avoid visible detection from predators. These desirable properties have prompted much research, including reports of materials with compound surfaces [17], materials coated with nanoparticles [18], and glass fibers imprinted with moth eye-like patterns [19], many of which also exhibit high optical transmission and AR character [20]. Various materials, including vanadium dioxide [21], tin oxide [18], silica [14], and others [13, 22], have included some type of surface patterning in order to enhance the optical capabilities of the native substrate material.

We recently reported results of a surface modification on randomly nanopatterned fused silica substrates, resulting in the chemical attachment of a monolayer of a fluorinated compound, altering the surface wettability of the fused silica [23]. The surface modification was performed in order to impart superhydrophobicity to the previously hydrophilic substrates, ultimately resulting in surface behavior similar to that of the lotus leaf [3, 24]. In that report, it was determined that the surface modification did not negatively affect the high transmission and AR character of the substrates. It was also observed that the imparted superhydrophobic character was maintained in the presence of seawater. Also unique to that work was that the random nanopatterning of the fused silica substrate was directly done on the fused silica substrate itself [23], as opposed to being patterned with the aid of a coating, as is done in many other reports [18, 21, 25]. In the present report, we investigate the transmission behavior of various germanium substrates that have periodic, sub-micron structures patterned into their surfaces, and that were also subjected to surface modification. Germanium was chosen as the substrate material because it is a material that can transmit into the long wave-infrared (LWIR) region of the electromagnetic spectrum, which is a beneficial characteristic for applications such as thermal imaging and reconnaissance. Herein we outline the performance of surface-modified, periodically patterned germanium substrates.

2. Materials and methods

1H,1H,2H,2H-perfluorooctyl trichlorosilane (PFOTS) and n-hexanes were purchased from Sigma-Aldrich and used as received. Artificial seawater (meeting ASTM standard D1141-98) was purchased from Lake Products Company, LLC, and used as received. Germanium moth eye substrates were all patterned on only one side, and were purchased from TelAztec.

Germanium substrates were chemically modified using PFOTS as previously described [23]. The substrates were subjected to an O_2(g) atmosphere for 10 min in a March Plasma Reactive Ion Etcher under a pressure of 320 mTorr, and at a power of 200 W. The substrates were then removed, immersed and swirled for 30 s in hexanes that contained 0.5% PFOTS. The substrates were immediately rinsed with hexanes, blown dry with N_2(g) and placed in an oven to anneal for 15 min at 120 °C. Finally, the substrates were removed from the oven and allowed to cool.

FT-IR transmission data was obtained using an Analect Diamond-2O FT-IR Instrument. Scanning Electron Micrograph (SEM) images of the substrate surfaces were taken using a Carl Zeiss LEO Supra 55 scanning electron microscope.

Water contact angle measurements were taken as previously described using a Ramé-Hart instrument, model #290-U1 [23]. Deionized water droplets (10 µL) at room temperature were placed on the surface of the substrates. Measurements were taken in a 3 by 3 grid fashion, with 9 contact angle values recorded for each substrate. The best value and an average of all 9 values were determined for each substrate. Error is given as the standard deviation for the total series of measurements.

The seawater incubation process and data collection were performed as previously described [23]. The surface modified substrates were subjected to 20 h incubation in artificial seawater. Following incubation in the seawater, the substrates were dried and CA measurements were obtained as described before (using deionized water, not seawater).

3. Results and discussion

Recently we reported that nanopatterning IR transmitting materials can further enhance the transmission capabilities of the substrate material [23]. In that study, the substrate material studied was fused silica, and the substrate surfaces were randomly nanopatterned. In the present study, each of the substrates analyzed was made of germanium, and had a unique surface topography; however the surface patterns were not random, as shown in Fig. 1. Germanium is an element that is well known to transmit light into the LWIR. By periodically patterning the substrate surfaces, this transmission capability can be enhanced. The germanium substrate surface types studied were composed of pillars (Ge_P), truncated cones (Ge_C), and honeycomb (Ge_HC) patterns. For comparison purposes, a non-patterned, pristine (i.e. smooth) germanium substrate (Ge) was also analyzed.

 

Fig. 1 Figure showing the surface topography of the various germanium substrates. (a) Digital photograph images of various substrates where Ge represents non-patterned germanium, Ge_P represents germanium patterned with a surface pillar pattern, Ge_C represents germanium patterned with a surface cone pattern, and Ge_HC represents germanium patterned with a surface honeycomb pattern. (b) top down SEM images, (c) angled, zoomed in SEM images. Scale bars: (a) = 1 cm; (b) = 10 µm; (c) = 1 µm.

Download Full Size | PDF

The pristine Ge substrate (no patterning) was smooth, and its surface had a shiny, mirror-like appearance, as shown in Fig. 1. The Ge_P substrate had an array of cylindrical pillars that were ~0.75 µm in diameter, and were spaced ~1.5 µm apart, as shown in Fig. 1. Both on the plateau of the pillars, and in the gaps between pillars, much shorter, jagged nanostructures were seen in the SEM images obtained. The Ge_C substrate had an array of truncated cones that tapered from a 1 µm diameter base to a flat top that was slightly less than ~1 µm in diameter, as shown in Fig. 1. Unlike the Ge_P, which had jagged structures on the surfaces of its pillars, the surfaces of the truncated cones of Ge_C were smooth and devoid of such structures. That said, as with Ge_P, there were small, jagged nanostructures in the gaps between the bases of the cones of the Ge_C substrate. Finally, Fig. 1 shows that the Ge_HC substrate was unique in that instead of having discrete features, like pillars or cones, the raised portion of its surface was interconnected, but contained circular voids which formed a honeycomb-like pattern. The circular voids were ~1.75 µm in diameter, and had a uniform depth of ~0.75 µm.

Notable transmission profiles were obtained for each of the periodically patterned substrates before surface modification, and those profiles were compared to that of unmodified Ge, as shown in Fig. 2. The measured unmodified Ge sample showed a transmission of ~48% (note: this value is greater than the transmission of 41% that would be expected based on Fresnel reflection calculations alone, due to the etalon effect). For the mid-wave IR (MWIR) and LWIR (6–14 µm), there was an increase in measured transmittance for each of the patterned substrates in comparison to Ge. The transmittance for both Ge_P and Ge_C reached a peak of ~60% in the 8–12 µm range, and Ge_HC more gradually reached its peak transmission of nearly 65% at ~9 µm, before decreasing to 60% transmission at 14 µm, as shown in Fig. 2. Overall, these data show that each of the patterned substrate types analyzed significantly enhanced the transmission capabilities of germanium in the MWIR and LWIR regions.

 

Fig. 2 FT-IR plots for the various germanium substrate types prior to PFOTS modification. Ge represents non-patterned germanium, Ge(P) represents germanium patterned with surface pillar pattern, Ge(C) represents germanium patterned with surface cone pattern, and Ge(HC) represents germanium patterned with surface honeycomb pattern.

Download Full Size | PDF

Previously we showed that performing a surface modification on randomly nanopatterned fused silica to impart superhydrophobicity had little effect on the transmission profile of the substrate [23]. This modification entailed the attachment of 1H,1H,2H,2H-perfluorooctyl trichlorosilane (PFOTS) to the surface nanostructures of the substrate. In the present study, the germanium substrates were likewise subjected to PFOTS modification, and their post-modification transmission profiles compared to their pre-modification transmission profiles. As seen with fused silica in the previous study, the high transmission can also be maintained for periodically micropatterned germanium substrates following PFOTS modification in the present study, as shown in Fig. 3.

 

Fig. 3 Representative FT-IR plot of Ge(HC), before and after PFOTS surface modification, demonstrating the minimal effect of the PFOTS surface modification on transmission properties. Ge(S) represents non-patterned germanium, and Ge(HC) represents germanium patterned with a surface honeycomb pattern.

Download Full Size | PDF

With the exception of Ge_HC, all of the germanium substrates exhibited significant hydrophilic character prior to processing (Table 1), and upon surface contact, water would begin to spread outward from the center of the water droplet in a non-uniform fashion as time progressed. In the case of Ge_HC, however, the surface was only slightly hydrophilic, causing a water contact angle (CA) approaching 90°. Interestingly, small water droplets (10 µL) on the surface of Ge_HC did not roll off of the surface of the Ge_HC substrate, even when the substrate was completely inverted (see Visualization 1). Overall, these observations suggest that the surface structures of the substrates had a direct effect on the behavior of surface moisture when compared to the Ge control substrate surface.

Table 1. Water contact angle (CA) of germanium substrates at varying conditions.

View Table

In order to prepare the substrates for surface modification, each substrate was subjected to an oxygen plasma cleaning step. This process made the surfaces of each substrate even more hydrophilic, and thus more amenable to the PFOTS surface modification. Because the PFOTS molecule has hydrophilic chlorine atoms on one end and hydrophobic fluorine atoms on the other end, the surface modification works by attachment of the chlorines of the PFOTS molecule to the hydrophilic surface of the substrate, thus causing a layer of fluorine atoms to form at the interface of the air on the opposite end of the PFOTS molecule. Following the PFOTS modification, all of the substrates exhibited a dramatic increase in CA, with Ge_P and Ge_C exhibiting superhydrophobicity, having CA greater than 150°, as shown in Fig. 4(b). Not only did water droplets ‘bead up’ on the surfaces of Ge_P and Ge_C following the surface modification, water droplets readily rolled off of the surface at shallow slide angles (< 10°). The Ge_HC substrate also became more hydrophobic, with its CA increasing to 135° (Table 1). However instead of water droplets rolling off of its surface, water droplets tended to stick to its surface even after the PFOTS modification (see Visualization 2). Thus the PFOTS modified Ge_HC substrate falls under a class of surfaces that demonstrates hydrophobicity, while also having a high adhesive force [26, 27]. The behavior exhibited by the Ge_HC substrate is similar to that of certain rose petals, in which even water droplets with high CA’s can defy gravity and adhere to the surface in an inverted state [26, 27]. In contrast, the behavior of both Ge_P and Ge_C is reminiscent of lotus leaves, in which water droplets readily roll off the surface of the leaves at shallow slide angles [2, 3].

 

Fig. 4 (a) Digital image of water droplet on the various unmodified substrate surfaces before processing, (b) Water contact angle (CA) following PFOTS surface modification, (c) CA following seawater incubation, (d) CA following post-seawater incubation cleaning via sonication (Visualization 1 and Visualization 2).

Download Full Size | PDF

To determine whether the surface modification was viable in saltwater conditions, each substrate was incubated in seawater for 20 hours. After this, the CA’s were again measured as described above (Table 1). For each substrate type, there was a decrease in CA, with Ge_C and Ge_HC being the most affected by the seawater incubation. Interestingly, although Ge_P and Ge_C were the most alike in terms of CA immediately following the PFOTS modification, as well as transmission profile, they responded very differently to the incubation in seawater, with Ge_P being much less affected by the seawater incubation, as shown in Fig. 4(c). Despite a statistical decrease in CA, the Ge_P surface was still significantly hydrophobic following the seawater incubation while Ge_C and Ge_HC were not. In this regard the Ge_P substrate was the most similar to the previously studied, randomly nanopatterned fused silica substrates [23].

Following seawater incubation, each substrate was subjected to gentle sonication (5 mins) in deionized H₂O in order to determine if the pre-seawater incubation CAs could be restored by removal of the seawater salts. Unfortunately, SEM images showed that many of the pillars on the surface of the Ge_P substrate were damaged during the sonication process, as shown in Appendix Fig. 5. Despite the surface damage, with a CA of 125° following sonication, it was apparent that much of the hydrophobic character still remained on the Ge_P substrate surface. Although not as extensive, both Ge_C and Ge_HC also showed damage following sonication, but a large percentage of hydrophobicity was still recovered from both of those substrates after the sonication process. The maintenance (for Ge_P) and recovery (for Ge_C and Ge_HC) of enhanced hydrophobicity after the seawater salts were sonicated off of the substrate surfaces was likely due to the enduring presence of undamaged PFOTS-modified micro- and nanostructures that remained on each substrate surface. It is worth noting that Ge and Ge_HC both had sonication recovery CAs of 110°. This suggests that the primary damage to the Ge_HC substrate following sonication occurred on the surface of the raised honeycombs, thus the remaining hydrophobicity primarily happened on the smooth voided areas on the surface of the Ge_HC substrate. Essentially, damage to the Ge_HC substrate made it structurally more like the smooth Ge control substrate. Therefore, it is not surprising that the CAs for the two would be similar. Overall these data show that neither incubation in aqueous salts, nor surface damage, was fatal to the ability of the modified substrates to repel water.

 

Fig. 5 Top down SEM images showing damage done to the surface of Ge_P following sonication. (a) and (c) are before damage, (b) and (d) are after damage due to sonication. Scale bars for (a) and (b) = 2 µm, (c) and (d) = 10 µm

Download Full Size | PDF

4. Conclusions

The work presented shows not only a significant enhancement in the MWIR and LWIR transmission properties of germanium when germanium is periodically patterned with sub-micron features on its surface, but more importantly, it was also shown that the patterned germanium substrates could be chemically modified along the surface to impart superhydrophobicity without any change to the transmission. When incubated in seawater, there was a notable decrease in CA; however, much of the CA could be recovered upon brief sonication. Depending on the surface nanopatterning, upon PFOTS modification the substrates can exhibit both moth eye and lotus leaf characteristics (i.e. Ge_P, Ge_C), or can exhibit both moth eye and rose petal characteristics (i.e. Ge_HC).

The results are consistent with previous results obtained using randomly nanopatterned fused silica as the test substrate. In combination, this work and the prior results show that near-IR to mid-IR transmitting materials can undergo surface modification to become superhydrophobic while maintaining their desirable increase in transmission as compared to untreated surfaces. Thus, it is important to note that the key results of surface modification, superhydrophobicity, and maintenance of the transmission properties obtained for IR transmitting substrates are likely transferable to various other substrate materials with variously patterned surfaces.