and resonant filters. Previous work involving straight fiber-tapers required devices to be elevated by several microns above the chip surface to prevent parasitic coupling to the surrounding substrate. Curved fiber-taper probes [92, 98–101] have been demonstrated to reduce parasitic loss into the substrate. However, they tend to be less mechanically stable than their tensioned straight-taper counterparts and suffer from noise induced by fluctuations in the taper’s position. In this work we have developed a microscopic “dimpled”
fiber-taper probe which allows for low-noise local probing of individual devices on a wafer.
By increasing the tension in the taper, fluctuations in the taper-chip gap can be greatly reduced to the levels present in straight fiber-taper measurements. To demonstrate the utility of the dimpled taper optical probe, we describe the characterization of two types of devices on a SOI wafer platform: a dense two-dimensional array of high-Qsilicon microdisk resonators and, second, a planar microring resonator.
(c) (d)
Dimple Straight taper
Straight taper Pull Down Fiber mold
Hot zone around the mold
U-Mount with adjustable tension
High tension
Low tension
Dimple is shallower with less mechanical noise
Dimple is deeper with more mechanical noise Tapered fiber pulled
(e)
Heating torch -- On
Heating torch -- Off
Separate to increase
tension Heating torch
-- On Pull
Pull Hot zone at
edge of flame
(a) (b)
0 25 50 75 100 125 150
Time (s)
Norm. Trans.
1.00 0.90 0.80 0.70
Running Std. Dev. (%) 0 1 2 3 4 5 6
1060XP fiber SMF28e fiber
single mode
Figure 2.1: Process for producing dimpled fiber-taper probes. (a) Taper pulling with the fiber placed at the edge of the flame to produce a small hot zone. (b) Characteristicin situ down-sampled transmission traces for 1060-XP and SMF-28e fiber during a taper pull. The running standard deviation for these traces shows the onset of single-mode behavior. (c) Forming the dimple by heating taper while it is pressed against a mold fiber. (d) Detaching the probe from the mold. (e) Schematic of a U-mount with variable tension.
to trigger the end of the pull.1 To produce tapers with the desired length (∼1 cm for the region withd.10µm) and minimum diameter (∼1µm), typical pulls continue ∼10 s after the oscillations stop and take 120–150 s with the stages each moving at ∼1.5 cm/min, but there is significant variability depending on how the fiber is prepared. For instance, the single-mode jump in the running standard deviation will contain one or more discrete steps if the fiber is subjected to undue stress while the polymer coating is being removed.
After mounting the taper in a U-bracket [104], the narrowest part of the taper is pressed against a silica mold with the desired radius of curvature [Fig. 2.1(c)]; a bare optical fiber with a diameter of approximately 125µm is used as the mold in these experiments. The taper and mold are heated with a hydrogen torch and allowed to cool. To detach the taper from the mold, we simply pull the taper away slowly while moving it back and forth (i.e., we are inducing fatigue failure at the taper-mold joint). A thin layer of soot on the mold improves the detachment yield—an appropriate layer is usually deposited when the torch is sparked.2 After releasing the fiber from the mold, the taper retains an impression of the mold, Fig 2.2(b), which forms a global minimum with respect to the rest of the taper.
The dimpling process introduces negligible additional loss, and the total loss of the dimpled taper can be less than 0.5 dB relative to the unpulled optical fiber. However, tapers typically exhibit a total loss 1–2 dB. Using a specially designed U-mount with a set screw to control the tensioning, varying the taper’s tension changes the radius of curvature of the dimple.
Under high tension, the dimple becomes very shallow but never completely straightens.
After dimpling, the probe is mounted onto a three-axis 50-nm encoded stage and is fusion spliced into a versatile fiber-optic setup. During testing, devices are placed in the near field of the probe, as in Fig. 2.2(a,c); adjustments to a pair of goniometers ensure the straight run of the taper is parallel to the sample surface.
Measurement of the non-resonant insertion loss as the waveguide is moved relative to nearby semiconductor microstructures gives the effective interaction length and profile of the local probe. First, we record the loss as a 1.6-µm wide GaAs cantilever is scanned along the taper’s length (ˆx-direction) while holding the taper at a fixed height. At tensions used in standard testing, Fig. 2.3(a) shows only ∼20µm (full width at half maximum) of the taper
1This method was implemented by C. Chrystal. A similar procedure has been published by F. Orucevic et al. using a running Fourier transform [103].
2N.B. Do not spark H2 torches near the fiber either before or after tapering as it will significantly reduce the transmission of the final coupler.
5 μm
50 μm
20 μm
Figure 2.2: (a) Illustration of a “dimpled” taper coupled to an undercut microdisk. (b) Optical image of the taper probe. The taper diameter at the center of the dimple is∼1.2µm.
(c) At the center of a 5×5 array, the dimpled taper probe is critically coupled to the microdisk at the center of the array but not coupled to any of the neighboring disks.
at the bottom of the dimple is close enough to interact with the sample. Second, the loss is measured as a function of the probe’s height (ˆz-direction) above a 11.6-µm wide GaAs mesa. By assuming an exponential vertical dependence for the insertion lossL∝e−zt(x)/zo where zt(x) is the probe’s “near-field” profile and zo is the decay length from Fig. 2.3(b), we convert the axial dependence of the loss [Fig. 2.3(a)] into zt(x) [Fig. 2.3(c)]—i.e., the height of the taper relative to the lowest point of the dimple. Since only the lowest part of the dimple interacts with the sample, this method can only determine the taper’s profile within∼1.25µm of the surface. Fitting the profiles determines the effective probe radius to be 159, 228, and 498µm at low, medium, and high tension, respectively. These radii differ from the mold radius (∼62µm) due to tensioning of the taper and how the fiber detaches from the mold after heating.