Tunable Ultra High-Q Microtoroid

The ability to tune resonant frequency in optical microcavities is an essential feature for many applications. It is equally important to have a reproducible fabrication technique. Until recently, the combination of these features has not been available in devices that operate in the ultra-high-Q regime where device quality factors (Q) can exceed 100 million. By introducing electrical control of the resonant frequency in an ultra-high-Q microtroroid, themooptic tuning is acheived. This is the first example of an UHQ planar microresonator with “integrated” electrical tuning.  The fabrication process for this device is shown in Fig 1.

The tuning range was determined by simultaneously scanning a single-frequency, external-cavity laser (coupled to a tapered fiber waveguide) across a frequency span of approximately 50 GHz and in the spectral vicinity of a high-Q resonance and incrementally applying voltage to the microtoroid resonator. Fig. 2 shows an example of a typical tuning curve for a microtoroid resonator with a resistance of 7 Ω and tuning rate of 85 GHz/V2. The tuning is plotted against V2 in order to stress the dependence of tuning on applied electrical power (V2/R).

The frequency response characteristics were measured by first tuning the laser near a resonance and simultaneously a function generator was used to apply a small-signal sinusoidal modulation voltage. A lock-in analyzer was set up to detect the modulation induced in the optical power transmission and was referenced to the modulation frequency of the function generator. Due to bandwidth limitations specific to the lock-in analyzer used in the experiment, the frequency response was measured up to 100 kHz. In Fig. 3 the frequency response of the tunable microtoroid resonators in both air and helium is plotted. The measured frequency response contains features consistent with the existence of a single, low-frequency pole. This corner frequency is believed to result from cooling to the ambient atmosphere. To substantiate our hypothesis, the corner frequency was measured in a helium ambient and was observed to double in value (see Fig. 3). This is due to the larger coefficient of thermal conduction of helium which is 5.62 times greater than that of air.

More information can be found in the following papers:

D. K. Armani, B. K. Min, A. L. Martin, and K. J. Vahala
“Electrical thermo-optic tuning of ultrahigh-Q microtoroid resonators”
Applied Physics Letters, Volume 85, No. 22, 5439-5441 November 29, 2004.

D. K.  Armani, T. Kippenberg, S. M.  Spillane and K. J. Vahala
"Ultra-high-Q toroid microcavity on a chip"
Nature, vol. 421, pp. 925-929, 27 February 2003.


tunable figure 1

A: Oxide is lithographically defined and etched leaving oxide disks with contact holes. B: Metal is evaporated on both sides of the wafer to create the electrical contacts. Ohmic contacts are formed by annealing in a nitrogen ambient at 500C in a tube furnace. C: The wafer is isotropically etched using xenon difluoride. D: The oxide disks are reflowed using a CO2 laser, creating the tunable UHQ microtoroids.

tunable figure 2

Resonant frequency shift versus voltage2. The quadratic coefficient is 85.56 Ghz/V2. Inset: A scanning electron micrograph of a tunable microtoroid resonator.

tunable figure 3

The frequency response of the tunable microtoroid resonators in both air and helium ambient atmospheres. Inset: A rendered depiction of the tunable microtoroid device coupled to a tapered optical fiber while being contacted by a metal probe.