PART I MICROLENS ARRAY
Chapter 2 Using Hydrophilic Effect to Fabricate Self-Assembled Microlens Array
2.3 Experimental Results and Discussions
As mentioned above, the contact angle became smaller with longer UV/ozone treatment time. Figure 2-2 shows the experimental result of contact angles of diluted SU-8 photoresist droplet (∼10 μL) with zero, one, two, three, and four minutes UV/ozone treatment time on an SU-8 PR base layer. The measurement system was based on the Sessile Drop Method [12], which was set up by a backlight, a camera, and software programmed by LabVIEW®. The program detected the two endpoints of the footprint and the spherical centroid of the droplet’s profile, where the profile was assumed to be spherical because the drop was small so that the gravity can be neglected.
Thereafter, the contact angle was derived by the three points. In Figure 2-2, a droplet (∼10 μL) was deposited on the sample surface. At least three samples were measured for each treatment time to obtain meaningful data. The contact angles reached its minimum value when UV/ozone treatment time was longer than 3 minute. We used the atomic force microscope (OBJ-204C, ITRI, Taiwan) to measure the surface roughness of the SU-8 base layer before and after four minutes UV/ozone treatment. The average height decreased by 40 nm, and the average surface roughness were both approximately 2 nm as shown in Figure 2-3. It reveals that the UV/ozone treatment did not increase the surface roughness effectively after four minutes UV/ozone treatment.
MLAs were fabricated by use of the 50 μm, 100 μm, and 200 μm openings of the shadow masks under one, two, three, and four minutes UV/ozone treatment time. The gaps between microlenses were 50 μm. The MLAs were finally cured by 2.4 J/cm2 of
angles of fabricated microlenses with different opening diameters of shadow mask under different UV/ozone treatment time. It was calculated from the measurement using surface profiler (Alpha Step 500, TENCOR). The longer treatment time resulted in a smaller contact angle. The contact angle of MLA was somehow smaller than those shown at Figure 2-2. This was mainly due to the evaporation of the solvent in the diluted SU-8 photoresist solution after UV curing. From Figure 2-4, the larger opening of the shadow mask also made smaller contact angles, which means the UV/ozone treatment was more effective. UV/ozone treatment longer than three minutes did not make any significant change. But, openings larger than 100
m did not make any
distinguishable difference with the same treatment time. According to the surface profiler measurement, the footprint of the fabricated microlens was smaller than the shadow mask opening. The sizes of MLA were closer to shadow mask openings under longer UV/ozone treatment time. The measured diameters of MLAs which were corresponded to the shadow masks of 50, 100, and 200 μm with one, two, three, and four minutes UV/ozone treatment are listed in Table 2-1. We could see that microlenses are closer to mask size under longer UV/ozone treatment. Focal length measurement of 100-μm diameter microlenses for one, two, three, and four minutes UV/ozone treatment is shown in Figure 2-5. Among the fabricated MLAs, the NAs approximately varied from 0.06 to 0.19 and the focal lengths were from 0.06 mm to 2.78 mm. We verified that longer UV/ozone treatment time results in longer focal length. This agrees well with contact angle and footprint diameter measurement results. In other words, longer UV/ozone treatment results in less curved microlens.Figure 2-3 Atomic force microscope (AFM) (OBJ-204C, ITRI, Taiwan) images of the SU-8 base layer (a) before and (b) after four minutes UV/ozone treatment. All images were taken at a scan size of 1x1 μm2 with 128 x 128 pixel2.
Figure 2-2 The experimental results of contact angles between a non-UV-exposed diluted SU-8 photoresist droplet (∼10 μL) and an SU-8 photoresist base layer.
Figure 2-5 Focal length versus UV/ozone treatment time.
Figure 2-4The contact angles for different UV/ozone treatment time and different shadow mask openings. The shadow mask openings were 50 μm, 100 μm, and 200 μm in diameters.
We took the pictures of the MLAs fabricated using 100-μm diameter shadow mask and measured the focus beam spot sizes by optical microscopy in Figure 2-6. It shows the uniform distribution of the intensity profile. Among the fabricated MLAs the average beam spot sizes were from 2.6 μm to 33.8 μm with variation less than 12%. The yield rate was higher than 96 %. We believe this can be further optimized in future. Figure 6 shows the scanning electronic microscopy (SEM) photograph of the microlenses using 100-μm diameter shadow mask under one minute UV/ozone treatment time. The cross section profile in Figure 2-7(e) indicates three layers, which are a glass substrate, a 5-μm thick SU-8 photoresist base layer, and an SU-8 photoresist microlens on the treated area. Two dimensional profiles of MLAs measured by surface profiler showed that the lenses had good surface smoothness. The surface roughness was less than 0.07 μm. This is a typical value for microlenses without going through etching process. We made the curve fitting of the microlens profile by using the SAG equation,
Table 2-1 The footprint diameters of MLAs.
constant. From the fitting result, the conic constant was between -0.005 and 0.00036, which means the surface curvature of MLA fabricated by this method was spherical. We measured the wavefront information by using the commercial Shack-Hartmann wavefront sensor (UI2210-m, UEye, NL) to evaluate the optical quality. The corresponding peak-to-valley (PV) and root-mean-square (RMS) values were lower than 0.949 λ and 0.143 λ as shown in Figure 2-8. The experimental results were summarized in the Table 2-2 below. The yield rates were obtained by average of each 5 samples using the beam spot images. We believe this can be further optimized in future.
Figure 2-6 The MLAs fabricated using 100-μm diameter shadow mask. (a) Optical microscopy and (b) the corresponding focus points with (c) intensity profile for four minutes UV/ozone treatment time. (d) Zoomed image of the optical intensity profile.
Figure 2-8 The interferogram of the wavefront surface measurement. (a) 50 μm (b) 100 μm (c) 200 μm.
Figure 2-7 The SEM of (a) microlenses of 100 μm diameter under one minute UV/ozone treatment time. (b) the cross section profile of one minute treated microlens.
Table 2-2 The summarized results of fabricated MLAs.