Detecting a single species in a mixture of chemically similar molecules can be difficult. Several different techniques have been applied to this problem, as it concerns D2O in H2O mixtures. While spectroscopic (emission and absorption) techniques might seem the most logical, since they give the “chemical fingerprint” of the individual molecules in a mixture, this technique has not achieved low detection sensitivities. Alternate methods, such as protonic conductors and NMR have succeeded in detecting 5% and 10% D2O in H2O respectively. To date, the most sensitive technique is optothermal detection. While this is similar to infra-red detection, the signal can be further enhanced to give sensitivities of 30ppmv (parts per million per volume) of D2O in H2O.
Ultra-high-Q microcavities can offer an enhancement of this type of measurement. Changes in absorption modify the cavity intrinsic quality factor, and under constant loading conditions (provided by a coupling waveguide), these modifications to Q factor can be observed as changes in the on-resonant optical transmission through the coupling waveguide. In this work, we demonstrate this method for the first time, using the chemically similar species D2O and H2O. In a single pass measurement, which is typical of an optothermal detector or spectrophotometer, the light will pass through the sample, giving an absorption reading over a very small distance. However, the light orbiting inside an ultra-high-Q microcavity (Q > 10 million) interacts with the sample over a large effective length, and, as shown here, can increase sensitivity to 1ppmv or .0001% without increasing the testing footprint. To demonstrate this alternative to conventional chemical detection techniques, planar arrays of ultra-high-Q microtoroids were fabricated and the testing proceeded as follows: 1) immerse microtoroid in 100% D2O, 2) gradually increase concentration of H2O in D2O, until 100% H2O is reached, and 3) increase concentration of D2O to 100%. As shown previously, the difference between the quality factor in H2O and D2O is liquid-limited. Therefore, the quality factor can be described by: Qliq=2pn/la, where n=effective refractive index, l=wavelength, and a is the absorption rate introduced into the resonator whispering gallery due to the presence of the liquid. The refractive index of H2O and D2O is the same and the resonant wavelength is constant. Since the absorption of H2O is larger than D2O, the quality factor of the resonator in H2O is smaller than when the resonator is in D2O, at 1300nm.
In a first series of measurements, the solutions were prepared in 10% increments (10% H2O in D2O, 20% H2O in D2O, etc). The toroid was initially in 100% D2O. After the quality factor was determined, all of the D2O was removed, until the toroid was in air. The chamber was then flushed five times with the next concentration solution (in this case, the 10% H2O in D2O solution), and the quality factor was again determined. This flushing process was followed for all solutions to remove trace amounts of higher or lower concentration solutions. Initially, with the toroid immersed in 100% D2O, the quality factor was 1.55x107. As can be seen in Figure 2 and in table 1, when the concentration of D2O was reduced, the quality factor of the toroidal resonator began to decrease. The Q of the toroid in 100% H2O was 6.4x105. The theoretical values for each concentration of D2O in H2O were calculated and are also shown. These theoretical and experimental values are also listed in Table 1 for direct comparison. This Q decrease was reversible, and by increasing the D2O concentration, the quality factor is recovered. This cyclical refreshing process was repeated several times, demonstrating the reproducibility of this detector.
To determine the lower bound on the detection capabilities, larger dilutions of D2O in H2O were prepared, ranging from .01% to 1x10-9%. Starting at 100% H2O and slightly increasing the D2O concentration using the low concentration solutions, it was possible to set a lower limit on the detection. As can be seen in Figure 3, there is a strong signal at .001% D2O in H2O. However, a small, but detectable shift occurs with the .0001% D2O solution. These values are not believed to reflect a fundamental limit on the detection sensitivity of this device as no attempt to address operational sources of noise in the system has been attempted here.
The ultra-high-Q microcavity has demonstrated the ability to detect the difference between two chemically similar species, H2O and D2O, at low concentrations. This detection is based on the subtle difference in optical absorptions between D2O and H2O, which is then magnified by the quality factor of the resonator. Previous technologies have measured 30 ppmv (parts per million per volume) of D2O in H2O. This is the equivalent of .003% D2O in H2O. Using resonant cavities, it was possible to improve upon this detection sensitivity by over an order of magnitude. The ability to actively monitor the presence and simultaneously determine the quantity of D2O is very important, especially given its significance in current strategic locations.
More information can be found in the following papers:
A. M. Armani and K. J. Vahala
"Heavy water detection using ultra-high-Q microcavities"
Optics Letters, vol. 31, issue 12, June 2006.
A. M. Armani, D. K. Armani, B. Min and K. J. Vahala, S. M. Spillane
"Ultra-high-Q microcavity operation in H2O and D2O"
Applied Physics Letters, vol. 87, issue 15, October 2005.
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.