Microresonator soliton dual-comb spectroscopy

See allHide authors and affiliations

Science  04 Nov 2016:
Vol. 354, Issue 6312, pp. 600-603
DOI: 10.1126/science.aah6516

Shrinking spectrometers

Dual-comb spectroscopy is a powerful technique that uses the interference of two closely related combs to map spectroscopic features directly into a frequency domain that can be read by electronics. Suh et al. developed a dual-comb spectroscopy approach using combs produced by silica microresonators fabricated on a silicon chip. Perhaps high-resolution spectroscopy will soon be shrunk to the chip scale, doing away with the need for bulky spectrometers.

Science, this issue p. 600


Measurement of optical and vibrational spectra with high resolution provides a way to identify chemical species in cluttered environments and is of general importance in many fields. Dual-comb spectroscopy has emerged as a powerful approach for acquiring nearly instantaneous Raman and optical spectra with unprecedented resolution. Spectra are generated directly in the electrical domain, without the need for bulky mechanical spectrometers. We demonstrate a miniature soliton-based dual-comb system that can potentially transfer the approach to a chip platform. These devices achieve high-coherence pulsed mode locking. They also feature broad, reproducible spectral envelopes, an essential feature for dual-comb spectroscopy. Our work shows the potential for integrated spectroscopy with high signal-to-noise ratios and fast acquisition rates.

Since their demonstration in the late 1990s (13), optical frequency combs have revolutionized precision measurements of time and frequency and have enabled new technologies such as optical clocks (3). One remarkable method they make possible is dual-comb spectroscopy, which leverages the coherence properties of combs for rapid broadband spectral analysis with high accuracy (410). Frequency comb systems exist across a broad spectral range spanning ultraviolet to infrared, making this method well suited for measurement of diverse molecular species (10).

In parallel with advancements in frequency comb applications, the past decade has witnessed the appearance of miniature optical frequency combs or microcombs (11, 12). These microcombs have been demonstrated using several dielectric materials across a range of emission bands (11, 1317). Under continuous-wave laser pumping, the combs are initiated by way of parametric oscillation (18, 19) and are broadened by cascaded four-wave mixing (11, 12) to spectral widths that can encompass an octave of spectrum (15). Four-wave mixing in the ultrafast intraband gain medium of quantum cascade lasers has also been shown to create frequency modulation (FM) combs (20). These FM systems have been applied to demonstrate dual-comb spectroscopy in the mid-infrared (21). Also, heterodyne of two conventional microcombs in the mid-infrared has been demonstrated, a key step toward dual-comb spectroscopy (22).

A major advancement in microcombs has been the realization of soliton mode locking (2327). Soliton microcombs feature dissipative Kerr solitons that leverage the Kerr nonlinearity to both compensate dispersion and overcome cavity loss by way of parametric gain (23). Unlike earlier microcombs, this new device provides phase-locked femtosecond pulses with well-defined, repeatable spectral envelopes, which is important for dual-comb spectroscopy. The pulse repetition rate of the device is detectable, and it has excellent phase noise characteristics (24). In this work, we demonstrate dual-comb spectroscopy using this new platform. The dual-comb source spans >30 nm and the interferogram spectra feature high signal-to-noise ratios (SNRs). Also, precise microfabrication control enables close matching of the repetition rates so that >4 THz of optical bandwidth is measured within 500 MHz of electrical bandwidth.

A schematic view of the dual-comb experimental setup (Fig. 1) shows two soliton trains having different repetition rates (Δfr = fr1fr2) generated from distinct microresonators and then combined using a directional coupler. One of the combined streams is coupled through a gas cell of molecules (test sample in the figure) whose absorption spectrum is to be measured. The other combined stream provides a reference. The slight difference in repetition rates of the soliton streams creates a periodically time-varying interferogram in the detected current with a period 1/Δfr. Fourier transform of this time-varying signal reveals the interfering soliton comb spectra, now shifted to radio-frequency rates. The signal spectrum containing the molecular absorption information is then normalized using the reference spectrum to reveal the spectral absorption of the gas cell. [See (28) for details of the experimental setup, methods, and equipment.]

Fig. 1 Microresonator-based dual-comb spectroscopy.

Two soliton pulse trains with slightly different repetition rates are generated by continuous-wave (CW) laser pumping of two microresonators. The pulse trains are combined in a fiber bidirectional coupler to produce a signal output path that passes through a test sample as well as a reference output path. The output of each path is detected (on Reference and Signal photo detectors) to generate an electrical interferogram of the two soliton pulse trains. The interferogram is Fourier-transformed to produce comb-like radio-frequency (RF) electrical spectra having spectral lines spaced by the repetition rate difference of the soliton pulse trains. The absorption features of the test sample can be extracted from this spectrum by normalizing the signal spectrum by the reference spectrum. Also shown are two silica disk resonators. The disks have a diameter of 3 mm and are fabricated on a silicon chip. The nth optical comb frequency (νn1 and νn2) for each soliton pulse train is given in terms of the respective repetition rate (fr1 and fr2) plus an offset frequency (fc1 andfc2). In the RF domain spectrum, the nth line occurs at a multiple of the difference in the repetition rates (Δfr = fr2fr1) plus an offset frequency (Δfc = fc2fc1).

Solitons are generated and stabilized in two microresonators by means of the active-capture/locking technique (fig. S1) (24, 28). The locking makes it possible for the two combs to remain stable indefinitely. Typical soliton optical spectra (Fig. 2, A and B) feature the characteristic sech2 envelope, observed in this case over a 60-nm wavelength span. The detected electrical spectrum for each soliton source was measured using a spectrum analyzer with a bandwidth of 26 GHz (Fig. 2, C and D). The narrow spectral lines (resolution bandwidth of 500 Hz) have SNRs greater than 75 dB, which shows that the corresponding repetition rates are extremely stable. The high-Q resonators used in this work are silica whispering-gallery devices fabricated on a silicon wafer by a combination of lithography and wet/dry etching (29). The unloaded quality factor of the microresonators is approximately 300 million, and the solitons have repetition rates of approximately 22 GHz that were determined primarily by the diameter of the devices (3 mm).

Fig. 2 Soliton comb spectral characterization.

(A and B) Optical spectra of the microresonator soliton pulse streams. (C and D) Electrical spectra showing the repetition rates of the soliton pulse streams. The rates and resolution bandwidth (RBW) are given within the panels.

The optical outputs from the stabilized soliton sources are combined and coupled into two paths (Fig. 1 and fig. S1). One path contains a H13CN gas cell (28), which functions as the test sample in the measurement. The other path is coupled directly to a photodetector and functions as the reference. The test sample path also includes an alternate path (shown in fig. S1) in which a WaveShaper is inserted (28). Temperature control of one of the microresonators is used to tune the relative optical frequency difference of the two soliton streams. In our measurements, this difference was held below 1 GHz, allowing the observation of the temporal interferogram on an oscilloscope (bandwidth 1 GHz).

The reference interferogram (Fig. 3A) has a period of 386 ns, corresponding to a soliton repetition rate difference of 2.6 MHz. This relatively small repetition rate difference was made possible by precise control of the resonator diameter (and hence the resonator free spectral range) using calibrated wet etching of the silica (29). It was possible to fabricate disks with even more closely matched repetition rates (<100 kHz). Figure 3B shows the calculated Fourier transform of the interferogram. The small repetition rate difference on the much larger 22-GHz soliton repetition rate makes it possible to compress an optical span of 4 THz (1535 to 1567 nm) into 500 MHz of electrical spectrum. The measured wavelength span is actually narrower than the observable wavelength span of the original soliton pulse streams and is limited by the photodetector noise. The interferogram spectrum has a SNR around 30 dB near the central lines. A zoom-in of the spectrum (multi- and single-line) is shown in Fig. 3C. The electrical comb lines are equidistantly separated by 2.6 MHz and have a linewidth (full width at half maximum) of <50 kHz, limited by the mutual coherence of the independent fiber pump lasers. The pump laser line in a dissipative Kerr soliton is also a comb tooth in the soliton optical spectrum. As a result, the frequency jitter in each pump is transferred as an overall shift on the resulting soliton comb. Externally locking the two combs should reduce the observed linewidth in the interferogram spectrum.

Fig. 3 Measured electrical interferogram and spectra.

(A) The detected interferogram of the reference soliton pulse train. (B) Typical electrical spectrum obtained by Fourier transform of the temporal interferogram in (A). Ten spectra each are recorded over a time of 20 μs and averaged to obtain the displayed spectra. (C) Resolved (multiple and individual) comb lines of the spectrum in (B) are equidistantly separated by 2.6 MHz, the difference in the soliton repetition rate of the two microresonators. The linewidth of each comb line is <50 kHz and is set by the mutual coherence of the pumping lasers. (D and E) Fourier transform (black) of the signal interferogram produced by coupling the dual-soliton pulse trains through the synthetic absorber (WaveShaper; see fig. S1) with programmed absorption functions (spectrally flat and sine-wave). The obtained dual-comb absorption spectra (red) are compared with the programmed functions (blue curves) from 1545 to 1565 nm.

The pump lines are placed toward the high-frequency side (near 550 MHz) of the spectral maximum in the interferogram spectrum (see Fig. 3B). In the optical spectra (Fig. 2, A and B), the pump is blue-detuned relative to the soliton spectral maximum [this occurs on account of the Raman self-frequency shift of the soliton (24)]. This spectral landmark shows that the relative spectral placement of the soliton combs is such that high optical frequencies are mapped to high interferogram frequencies. The way in which certain nonidealities in the soliton spectra map into the interferogram spectrum is also of interest. Specifically, there are avoided mode crossing–induced Fano-like spurs (24) in the soliton optical spectra (Fig. 2, A and B) occurring near 1535 nm, and this generates a corresponding feature at 750 MHz in Fig. 3B.

As an initial test of the dual-comb source, synthetic absorption spectra were programmed in a WaveShaper and then measured as dual-comb spectra (28). In Fig. 3, D and E, Fourier transforms of the signal interferograms produced using the synthetic absorber are shown. The two programmed functions are a spectrally flat 3-dB absorption and a sine-wave absorption having a 4-dB amplitude. The synthetic absorption spectra, obtained by normalizing the signal and reference electrical spectra, are compared with the programmed functions in Fig. 3, D and E. The ability to reconstruct these synthetic spectral profiles clearly demonstrates the reproducibility of the soliton spectral profile.

Finally, we studied the absorption spectrum of the H13CN 2ν3 band. In Fig. 4A, the measured dual-comb absorption spectrum from 1538 to 1562 nm is shown in red and compared with a directly measured absorption spectrum shown in blue. Both absorption spectra are normalized. A laser locked to a molecular absorption line enabled absolute frequency calibration of the spectra (28). Sampling-induced choppiness of the dual-comb spectrum is caused by the relatively coarse spectral resolution of the solitons in comparison to the spectral scale of the H13CN absorption lines. Nonetheless, the characteristic envelope of the H13CN 2ν3 band is clearly resolved. The residual difference between the two absorption spectra is shown in green in Fig. 4A; the calculated standard deviation is 0.0254. Furthermore, a line-by-line overlay of the measured optical and dual-comb spectra (Fig. 4B) confirms the wavelength precision and absorption intensity accuracy of the dual-comb source. The directly measured H13CN absorption spectrum is obtained by coupling an external cavity diode laser (ECDL) into the H13CN gas cell and then scanning the laser while monitoring the transmitted optical power. A separate signal is also tapped from the ECDL to function as a reference (28).

Fig. 4 Measured molecular absorption spectra.

(A) Absorption spectrum of 2ν3 band of H13CN measured by direct power transmission using a wavelength-calibrated scanning laser and comparison to the microresonator-based dual-comb spectrum. The residual difference between the two spectra is shown in green. (B) Overlay of the directly measured optical spectrum and the dual-comb spectrum, showing line-by-line matching. The vertical positions of the two spectra are adjusted to compensate for insertion loss.

In principle, a soliton source with finer comb spacing (i.e., lower repetition frequency) is possible. Nonsoliton microcombs with mode spacings as narrow as 2.4 GHz have been demonstrated using the silica resonator platform (16). The use of electro-optical modulators to modulate the microcombs by a fraction of the repetition frequency is another possible way to create a finer spectral comb grid. Also, tuning of the combs is, in principle, possible so as to create a nearly continuous high-resolution absorption measurement.

The dual-comb source is centered near 1550 nm (in the C-band) for this work; however, operation at other wavelengths within the transmission window of silica is also possible. Moreover, using fiber nonlinear broadening or internal (resonator) dispersive wave generation (25), it should be possible to greatly extend the comb spectral span. Resonator dispersion engineering (30) can also be used to extend comb bandwidth. More generally, a wide range of materials are available for microcombs enabling access to mid-infrared spectra. With further improvements, it should also be possible to realize chip-based dual-comb coherent anti-Stokes Raman spectroscopy (CARS). The potential for monolithic integration with other devices makes soliton-based microcombs well suited for possible realization of a dual-comb spectroscopic system-on-a-chip.

Supplementary Materials

Materials and Methods

Supplementary Text

Fig. S1

References (3134)

References and Notes

  1. See supplementary materials on Science Online.
Acknowledgments: We thank N. Newbury and G. Scalari for helpful comments on this manuscript. Supported by the Defense Advanced Research Projects Agency under the PULSE and SCOUT programs, NASA, and the Kavli Nanoscience Institute.
View Abstract

Navigate This Article