Chip-Scale Frequency Combs For Data Communications in Computing System
Chip-Scale Frequency Combs For Data Communications in Computing System
Received 1 April 2022; revised 25 June 2023; accepted 27 June 2023; published 25 July 2023
Recent developments in chip-based frequency-comb technology demonstrate that comb devices can be implemented in
applications where photonic integration and power efficiency are required. The large number of equally spaced comb
lines that are generated make combs ideal for use in communication systems, where each line can serve as an optical car-
rier to allow for massively parallel wavelength-division multiplexing (WDM) transmission. In this review, we summarize
the developments in integrated frequency-comb technology for use as a WDM source for communication systems in
data centers and high-performance computing systems. We highlight the following three approaches for chip-scale
comb generation: semiconductor modelocked lasers, electro-optic combs, and Kerr frequency combs. © 2023 Optica
Publishing Group under the terms of the Optica Open Access Publishing Agreement
https://doi.org/10.1364/OPTICA.460175
Fig. 1. Multiple wavelength sources for WDM communication. Conceptual schematic of (a) array of discrete wavelength lasers and three types of inte-
grated frequency combs: (b) extended cavity modelocked laser, (c) electro-optic frequency comb, and (d) microresonator-based Kerr frequency comb.
Table 1. Performance Metrics of Multiple Wavelength instabilities [29–31]. Furthermore, research efforts on monolith-
Sources in Communication Systems ically integrated MLLs have allowed for further miniaturization,
which is essential for applications including communications and
Power per channel 0.5–50 mW
Number of channels 4–160
microwave generation. These lasers are electrically pumped and
Channel spacing >50 GHz allow for picosecond and subpicosecond pulses with repetition
Linewidth 1 kHz–10 MHz rates from gigahertz to terahertz.
Relative intensity noise (RIN) <140 dB/Hz Modelocking in semiconductor lasers can be implemented
Side-mode suppression ratio (SMSR) <30 dB through active, passive, or hybrid approaches. To achieve active
modelocking, the optical gain is modulated by sending an electrical
the overall system performance has been extensively investigated in signal at a frequency near the cavity repetition rate or its subhar-
prior work [7,24–27]. monic [32]. Similarly, an electro-absorption modulator can be
In this review, we provide an overview of the recent develop- implemented to spur pulse formation [33]. These approaches are
ments in chip-based frequency-comb technology towards the largely limited by the availability of current sources at high RF
potential implementation of a multiple wavelength source for frequencies and typically result in longer pulses in the picosecond
WDM communication systems relevant for short-reach applica- range. Alternatively, a saturable absorber can be inserted within
tions (i.e., <100 m), including high-performance computing and the cavity to allow for passive modelocking. The nonsaturable
data centers. The three different approaches highlighted in this losses, recovery time, and saturation fluence of a saturable absorber
review are semiconductor modelocked lasers (MLLs), electro-optic strongly determine the properties of short pulse generation.
combs, and Kerr frequency combs, where in each case a distinct Furthermore, to reduce the phase noise and timing jitter of a MLL,
optical nonlinearity is responsible for generation of the OFC. hybrid approaches have been utilized where the saturable absorber
is modulated with an external RF source [34].
2. INTEGRATED SEMICONDUCTOR
MODELOCKED LASERS B. Integrated Solutions
A. Overview Since the demonstration of light emission from a gallium arsenide
Modelocked solid-state and fiber lasers are mature technologies (GaAs) diode laser in the 1960s [35], semiconductor diode lasers
that have been used to realize fully stabilized OFCs via super- have become a widely utilized technology for cw lasers. III-V tech-
continuum generation in nonlinear waveguides [21,28]. The nology has been integrated with silicon photonics technology via
development of fiber lasers has allowed for the realization of a wafer bonding, epitaxial growth, or hybrid approaches [36–38].
portable compact platform for use beyond the lab environment The active gain region can be electrically pumped (forward bias) to
[28]. However, most of these MLL systems operate with repetition realize light emission. One design that is used for semiconductor
rates up to a few hundred megahertz, which is not suitable for com- laser diodes is a ridge waveguide, where the ridge section along with
munication applications. Over the past two decades, there has been the cladding allows for the index contrast required for confinement
significant development of modelocked semiconductor lasers, of the optical mode. Recently, MLLs have been realized by modi-
which enables high pulse repetition rates (>GHz), due to the fying this waveguide geometry and introducing a reverse-biased
large gain cross section, which assists in preventing Q-switching saturable absorber section [39]. This two-section approach serves
Review Vol. 10, No. 8 / August 2023 / Optica 979
as the basic structure for many waveguide-based MLLs and we 2. Extended Cavity MLLs
review recent advances in integrated semiconductor MLLs. Key desirable characteristics for MLLs are low phase noise and
timing jitter. One approach that has been used to realize low phase
1. Quantum-Well, Quantum-Dot, and Quantum-Dash MLLs noise MLLs is to use an extended cavity that incorporates a passive,
low-loss waveguide. While the repetition rate is reduced, the noise
Semiconductor quantum structures can be used to provide both properties are also improved due to the increase in the photon life-
gain and SA action in integrated MLLs. The three different struc- time of the cavity [50,51]. One approach utilizes a 4-mm-long ring
tures that have been used for MLLs are quantum wells, quantum cavity, where the saturable absorber, SOA, and passive waveguide
dots, and quantum dashes. A quantum-well system consists of an are all embedded [52], and such a laser has been shown to generate
embedded thin layer of semiconductor material between other 900-fs pulses (11.5-nm 3-dB bandwidth) with a repetition rate
layers that have wider bandgaps, creating a potential well with of 20 GHz, a repetition-rate linewidth of 800 kHz (3-dB), and
discrete energy levels. There have been many demonstrations of an optical linewidth of 800 MHz near the central optical mode.
quantum-well-based MLLs where a reduction in cavity losses and Another approach implements an intracavity gain-flattening filter
an increase in gain saturation energy allows for reduction in the based on a Mach–Zehnder interferometer to realize a 900 fs pulse
pulse-timing jitter. Research efforts have focused on the laser struc- train (15 nm 10-dB bandwidth) with a 300-GHz repetition rate,
ture such as the cladding composition and the number of quantum repetition-rate linewidth of 500 kHz, and an optical linewidth of
wells to improve these metrics [40], and modelocking is achieved 29 MHz [53]. A narrower repetition-rate linewidth of 6.1 kHz and
with a two-section approach and 1-ps pulses with a repetition rate an optical bandwidth of 3 nm were demonstrated using a longer
of 40.77 GHz are generated with a timing jitter of 1.2 ps. Another (33-mm-long) InP-based ring cavity (Fig. 2) [54]. However, as a
approach is to use what is known as colliding-pulse modelocking, consequence of the longer cavity, the repetition rate was reduced to
in which the saturable absorber is placed at a location where pulses 2.5 GHz.
from each direction overlap, allowing for faster saturation and A second approach uses a Fabry–Perot (FP) cavity with long
shorter pulses [41]. In Lo et al. [42], two semiconductor optical passive waveguides typically in a spiral geometry to extend the
amplifiers (SOAs) are placed symmetrically with respect to the overall cavity length. Similar to the ring laser, the gain and SA
saturable absorber, multimode interference reflectors (MMRs) are regions are within the extended cavity. The high reflectors at the
used as end mirrors, and phase modulators are integrated to allow ends of the cavity are realized using MMRs [55]. Guzmán et al. uses
for bandwidth control. Using this approach, they demonstrate a 41-mm-long monolithic InP extended cavity to realize a 1-GHz
a 1.8-THz modelocked spectrum (350-fs pulse duration) with a repetition-rate MLL with a repetition-rate linewidth of 398 kHz,
21.5-GHz repetition rate and a 450 kHz repetition-rate linewidth. which is largely limited by the optical loss of the spiral waveguide
The two-section approach allows for high repetition rates with [56]. To achieve higher repetition rates using an extended cavity
demonstrations up to 100 GHz [43]. configuration, harmonic modelocking has been achieved using
Modelocking has also been demonstrated using a single-section three SA sections that divide the cavity into four gain segments
indium arsenide/indium phosphide (InAs/InP) quantum-dot with a total length of 1.66 mm. Such a system produces 500-fs
waveguide [44]. Quantum dots are nanometer-scale semicon- pulse trains with 100-GHz repetition rates [57]. Alternatively, a
ductor particles where the electrons and holes are constrained, hybrid approach has been implemented, where a III-V reflective
resulting in discrete energy levels. In contrast to the reverse-biased SOA section, consisting of gain and SA regions, is coupled to a
saturable absorber used in the two-stage structures, modelocking 30-mm-long silicon nitride (Si3 N4 , SiN) based extended cavity
here is achieved via self- and cross-phase modulation, four-wave region that consists of a long spiral and a Sagnac-based reflector
mixing (FWM), and the Kerr-lensing effect, which leads to the [51], which produces a bandwidth of 4 nm (6.31 ps pulses) at a
generation of 312-fs pulses with a repetition rate of 92 GHz and a repetition rate of 2.18 GHz. The long cavity allows for narrowing
repetition-rate linewidth <20 kHz. of the repetition-rate linewidths down to as low as 31 Hz. One
The quantum dash (QDash) consists of an elongated semicon- drawback of the InP extended cavity is the relatively high propa-
ductor structure that allows for strong carrier confinement. The gation losses (3 dB/cm) and the nonlinear loss due to two-photon
absorption [51].
broad gain bandwidth and low coupling of amplified spontaneous
emission (ASE) make this material ideal for on-chip modelocking
[45]. A key characteristic of QDash MLLs is that pulse formation 3. III-V on Silicon MLLs
can occur without a saturable absorber [46] and modelocking in There has been significant development towards heterogeneous
this system is attributed to FWM in the gain region [47]. Gosset integration of lasers with silicon PICs. Silicon PICs utilize mature
et al. [47] have shown 800-fs pulse duration, 42-GHz QDash complementary metal oxide semiconductor technology and read-
MLLs with narrow repetition-rate linewidths of 50 kHz, offer- ily offer numerous photonic components including modulators,
ing promise for timing jitter as low as 200 fs. Further studies by filters, detectors, splitters, and multiplexers with capabilities of
Merghem et al. have shown reduction of timing jitter down to wafer-scale fabrication for high-yield, mass production [58]. A
130 fs for 1.3 ps pulses at a repetition rate of 47.54 GHz [45]. The key development has been the heterogeneous integration of an
single stage structure is ideal for achieving high repetition rates and electrically pumped optical source, where the III-V gain mate-
pulse trains with subpicosecond pulses, with demonstrations of rial is placed on the silicon PIC. Approaches for III-V on silicon
245 and 346 GHz operation [46,48] and subpicosecond timing jit- integration include bonding [59–62], direct epitaxial growth via
ter [49]. Like the quantum-dot laser, the QDash laser also exhibits intermediate buffer layers [63,64], and micro-transfer printing
higher performance than the two-section quantum-well laser. [38]. Recent developments of a characterization method involving
Review Vol. 10, No. 8 / August 2023 / Optica 980
Fig. 2. (a) Microscope image of the ring-cavity MLL: semiconductor optical amplifier (SOA), saturable absorber (SA), electrical isolation (ISO), elec-
tro refractive modulator (ERM), multimode interference coupler (MMI), and passive waveguides (PWG). Measured (b) output spectrum and (c) repetition
linewidth of ring-cavity MLL. Figure adapted from Ref. [54].
a stepped-heterodyne approach has allowed for phase and ampli- For applications in WDM communications the channel spac-
tude measurements of the laser output indicating that transform ing (i.e., the repetition rate) needs to be larger. As seen in both
limited pulses can be achieved in these systems [65]. InP and III-V on silicon extended cavities, to achieve the narrow
For implementation of a MLL on silicon, oxide plasma repetition-rate linewidth, long cavities are required, reducing
assisted wafer bonding was used to transfer the III-V epitaxial the pulse repetition rate. However, the low propagation losses in
layer structure, which consists of a gain and SA region, to the silicon allow for additional elements to be incorporated into the
silicon-on-insulator wafer, where the silicon waveguides are ini- cavity, which can permit higher repetition rates while maintaining
tially fabricated to interact with the III-V structure via evanescent narrow linewidths. Srinivasan et al. utilized a 2-GHz cavity with an
coupling [66], and the laser cavity is formed by polishing the facets. intracavity filter that has a 20-GHz spacing, allowing for low noise
Using such a system, both passive and hybrid modelocking was harmonic modelocking at a 20-GHz repetition rate with a 52-kHz
demonstrated with repetition rates as high as 40 GHz. The system linewidth and an optical linewidth of 1 MHz, which is 10× lower
was further improved to allow for integration with other photonic than a 20 GHz cavity without a ring filter [74]. Alternatively,
components by using a racetrack geometry, eliminating the end Liu et al. demonstrated a low timing jitter (82.7 fs) and narrow
facets [67], and produced 7-ps pulses at a 30-GHz repetition rate. repetition-rate linewidth (1.8 kHz) quantum-dot MLL on silicon
The main advantage that silicon offers is the low propagation (Fig. 3) [75]. The gain region implements a chirped quantum-dot
loss in the passive silicon waveguide (0.7 dB/cm), allowing for a design to allow for broader bandwidths (6.2 nm), and the narrow
reduction in the repetition-rate linewidths [68]. Similar to the linewidth performance is attributed to the low ASE noise and low
extended cavity approaches demonstrated in InP, Keyvaninia et al. confinement factor in the quantum-dot material. The bandwidth
demonstrated an extended cavity MLL using the III-V on silicon of the generated laser spectrum spans 6.2 nm with a repetition rate
platform with both a FP and ring-cavity geometry [68]. The III-V of 20 GHz. The average optical linewidth of the modes is measured
region is bonded using divinylsiloxane-bis-benzocyclobutene to be 10.6 MHz. We have highlighted the performance of the
[60], allowing for evanescent coupling to the silicon waveguide. various semiconductor MLLs in a table at the end of the section
The FP cavity and ring cavity had a repetition rate of 4.7 GHz (Table 2).
and both showed a reduction in repetition-rate linewidth (12 kHz
and 16 kHz, respectively) with optical bandwidths up to 9 nm. In C. Semiconductor MLLs for Communication
addition, Keyvaninia et al. presented an anti-colliding-pulse-type MLL sources have applications in various WDM and coherent
MLL which produced a 3-ps pulse train with a repetition rate of optical communication systems [76]. One key issue for semicon-
4.83 GHz and a repetition-rate linewidth of 1.7 kHz [69]. The end ductor MLLs particularly for applications in communications is
mirrors are implemented using high reflectivity distributed Bragg the high relative intensity noise (RIN), which is largely attributed
reflectors to allow for further integration with other components in to the laser linewidth, which is dependent on the cavity Q, and
the silicon layer. Wang et al. demonstrated a 1 GHz repetition-rate the mode partition noise [77,78]. The linewidth requirement is
MLL using an extended silicon waveguide design, producing a generally set by the data rate of the system along with the modula-
10-nm optical bandwidth, a 450 Hz repetition-rate linewidth, and tion format that is employed [79,80], thus making a system level
a 250 kHz optical linewidth [70]. Further repetition-rate linewidth performance study important. Kurczveil et al. investigated the per-
reduction has been achieved by using a III-V on SiN platform formance of a quantum-dot laser bonded on silicon, where three
[71,72]. SiN has even lower loss than silicon (5 dB/m) [72], comb lines of their 38-GHz output are modulated with a non-
enabling low-noise modelocking. Here, the SiN spiral waveguides return to zero (NRZ) signal [81]. The measured signal-to-noise
(two 20-cm-long waveguides) act as the extended cavity and are ratio is >11.5 dB, and the bit-error rate (BER) is less than 10−12
coupled to a InP/InAlGaAs-based multiple-quantum-well ampli- for all channels. In addition, Moscoso-Mártir et al. implemented
fier waveguide through the silicon layer. The generated spectrum an eight-channel WDM transceiver based on a semiconductor
spans 3.27 nm with a repetition rate of 775 MHz, a repetition-rate MLL and studied the performance in the context of data-center
linewidth of 1 Hz, and an optical linewidth <200 kHz, which is interconnects (Fig. 4) [82]. The transmitter consisted of an exter-
the lowest for an on-chip MLL to date. The low-loss nature of SiN nal quantum-dash semiconductor MLL, a silicon chip consisting
waveguides offers promise for incorporating rare earth doped laser of filters to remove the unused lines and resonant ring modulators
gain media, with recent work demonstrating a MLL operating at for on–off-key modulation, and an SOA for optical amplification.
1.9 µm with a repetition rate of 1.2 GHz and an optical bandwidth The MLL has a 100-GHz repetition rate, and each of the eight
of 17 nm [73]. channels has power >500 µW. Analysis of the transmitter link has
Review Vol. 10, No. 8 / August 2023 / Optica 981
Fig. 3. (a) Schematic of the quantum-dot MLL. (b) Output spectrum (blue) and optical linewidth of the comb modes (red). (c) BER measurement and
(d) eye diagrams for comb lines modulated using a PAM-4 signal. Figure adapted from Ref. [75].
Fig. 4. (a) Block diagram of WDM transmitter using QDash MLL as the WDM source. (b) Image of transceiver with I/O. (c) MLL spectrum (blue) and
transmitter response (black) showing the eight wavelength channels. Figure adapted from Ref. [82].
shown that the system operates up to 14 Gbps with hard-decision Alternatively, Arsenijević et al. demonstrated 160-Gbps data trans-
forward error correction (FEC). The QDash MLL was also used mission using differential quadrature phase-shift keying (QPSK)
to demonstrate a 10.68 Tbps aggregate net data rate but using and time division multiplexing based on a 40-GHz repetition-rate
a 38-channel 16QAM dual-polarization WDM system [83]. quantum-dot MLL [84]. Furthermore, using 64 comb lines from
Review Vol. 10, No. 8 / August 2023 / Optica 982
a quantum-dot laser on silicon, Liu et al. characterized the data hybrid modulators. The performance of the various platforms is
transmission performance using a 32-GBd PAM-4 signal offering summarized in a table at the end of the section (Table 3).
promise towards 4.1 Tbps of transmission capacity [75]. Pan et al.
used a 100-GHz repetition-rate quantum-dot MLL to demon-
1. Thin-Film Lithium Niobate Modulators
strate 64-GBd PAM-4 data transmission using seven wavelength
channels [43]. Another issue with semiconductor MLL is the rela- Electro-optic modulators based on lithium niobate (LiNbO3 ,
tively low power per comb line. This can be mitigated with external LN) are well established as a key module in fiber-optics technol-
SOAs. However, the SOA current must be optimized as the ASE ogy [105]. These modulators rely on the Pockels effect and often
from the SOA can be coupled to the MLL cavity affecting the MLL use the relatively large EO tensor component (r 33 = 31 pm/V)
cavity and can affect the modelocking process [85]. [106]. These widely available commercial devices are fabricated
via ion-indiffusion or annealed proton exchange techniques
using bulk crystal wafers, resulting in a low index contrast and
3. ELECTRO-OPTIC COMBS necessitating a large optical mode [107]. Over the past several
A. Overview years, advancements in thin-film LN have led to the realization
of low-loss (<0.3 dB/cm), high-confinement waveguides with
Frequency combs can be generated via the electro-optic (EO)
cross sections on the wavelength scale. This allows for higher
effect where the refractive index of a material is changed in the
presence of an applied, near DC electric field. This allows for the electro-optic efficiency due to the electrodes being closer to the
generation of sidebands separated by the modulation frequency waveguide to allow for stronger interactions between the optical
of the driving field. For the simplest case where a cw laser with and electric fields [107]. Using the lithium-niobate-on-insulator
frequency ω0 is sent into a phase modulator driven with an electric (LNOI) platform, modulators with a 3-dB bandwidth of 100 GHz
field V (t) = V0 sin(ωm t), where ωm is the modulation frequency, have been demonstrated [88], and high-speed data transmission at
the output in the frequency domain can be expressed as 120 Gbps (NRZ) and 220 Gbps (PAM-4) has been shown [108].
Luke et al. have shown that wafer-scale fabrication is possible using
∞ deep ultraviolet lithography, offering promise for a scalable and
X V0
Ã(ω) = A 0 Jn π δ(ω − nωm − ω0 ) (1) cost-effective LN PIC [109].
n=−∞
Vπ
One approach for comb generation involves placing the EO
with Vπ as the half-wave voltage [86]. This expression shows that modulator inside a FP cavity to allow for resonant enhancement
a discrete spectrum is generated with the frequency of the nth of the optical field [110–112]. More recently, the development of
comb line given by ωn = nωm + ω0 . In order to shape the spectral high-Q resonators has enabled resonant EO combs where both
bandwidth of the generated comb output, various configurations the optical carrier and the microwave frequencies are resonant
have been considered including cascaded intensity and phase [89,113–115]. Zhang et al. demonstrated broadband resonant
modulators, dual-drive Mach–Zehnder modulators (MZMs), EO comb generation using a phase modulator embedded in a
dual-parallel MZMs, and EO cavities [87]. low-loss, dispersion-engineered LN microresonator based on the
LNOI platform (Fig. 5) [89]. Here, the comb spans >80 nm such
that 900 comb lines were generated with a 10-GHz spacing. The
B. Photonic Platforms microwave efficiency can be improved by embedding the optical
Although a key application of EO modulators is their use in cavity within the microwave cavity [114], and a recent demon-
imprinting data from the electrical to the optical domain, the past stration shows that high pump-to-comb conversion efficiencies
decade has also witnessed tremendous development of integrated of 30% can be achieved by using a coupled-resonator geometry
modulators for frequency-comb generation. Research efforts have to overcome the change in coupling conditions induced by the
focused on exploring materials with high electro-optic coefficients microwave drive signal used for comb generation, resulting in a
or low Vπ , compact footprint, and low propagation losses. An 132-nm bandwidth EO comb with 30.9 GHz spacing [116].
important efficiency metric for modulators is the voltage-length Another approach for EO comb generation involves the use
product Vπ L. Here, we review recent developments in inte- of cascaded intensity and phase modulators. To achieve pulse
grated EO-comb platforms, including thin-film lithium niobate compression and consequently spectral broadening, such a sys-
modulators, silicon modulators, plasmonic modulators, and tem uses a time lens [117,118], which takes advantage of the
Fig. 5. (a) Microscope image of lithium niobate resonator for EO comb generation. Electrodes (gold) are used for microwave modulation of the res-
onator. (b) Generated EO comb spectrum. Figure adapted by permission from Macmillan Publishers Ltd.: [89].
space–time duality of paraxial diffraction in the spatial domain a 4.5-mm-long traveling-wave phase modulator [94]. The wave-
and dispersive propagation in the time domain. This is a well- guide geometry allows for continuous tuning of the repetition
established technology using fiber-optic components [118–124], rate, and tuning between 7.5 and 12.5 GHz was demonstrated. In
with demonstrations of octave-spanning bandwidths [123,124]. order to further reduce the footprint and RF power, a relatively low
The low-loss LNOI platform allows for full integration of the cas- Q (3000–10,000) ring resonator modulator was used for comb
caded modulators and a dispersive element, offering potential for generation [136]. The low Q-factor is accompanied by a photon
a fully integrated comb source with spacings >30 GHz [90]. Since lifetime <8 ps such that the resonator free-spectral range (FSR)
the phase noise of the nth comb line due to the microwave signal does not dictate the optical bandwidth and is only dependent on
scales as n 2 , the noise becomes a significant issue as the number of the modulation frequency. In order to generate a frequency comb
comb lines increases [123]. This issue can be mitigated by using a
with a 10-GHz spacing, the modulator is driven with two different
cavity for phase noise suppression with suppression of up to 40 dB
RF frequencies, a 10-GHz signal with 1.5 Vpp and a 20-GHz signal
demonstrated experimentally [125].
with 2.9 Vpp . By controlling the bias voltage, seven comb lines are
generated within a 20-dB bandwidth. The generated bandwidth
2. Silicon Modulators can be further increased through such cascading of modulators
Silicon is a material platform that is used extensively for PICs [137,138].
and has become a mature technology with hybrid photonic and
electronic integration done via interposers or through-silicon vias.
3. Plasmonic Modulators
Frequency-comb generation has been demonstrated using phase
modulators based on free-carrier injection. These modulators Plasmonic structures allow for confinement of the optical mode
utilize the index change that results from the free-carrier plasma to subwavelength waveguide geometries allowing for compact
dispersion effect, which is dependent on the number of carriers photonic devices [139]. In these structures, the refractive index is
and the effective conductivity masses of the electrons and holes altered with the presence of an applied field based on the Pockels
[126,127]. Carriers can be injected by sending an electric field effect. Assisted by a surface-confined plasmon-polariton mode,
across a PIN junction, which consists of heavily doped p-type the strong localization of the optical and electrical fields in the slot
and n-type semiconductors that surround an undoped intrin- structure allows for large mode overlap between the two fields,
sic semiconductor. PIN modulators have enabled modulation resulting in high modulation indices and low power consump-
speeds beyond 100 GHz [91,92], and have been implemented
tion with electro-optic coefficients as high as r 33 = 390 pm · V−1
for electro-optic conversion and high data rates in integrated
demonstrated [140], and Vπ L = 0.06 V · mm has been demon-
platforms [128–132]. Jacques et al. demonstrated 240 Gbps with
strated [95]. Moreover, plasmonic modulators based on organic
PAM-8 modulation using a segmented-electrode MZM [132], and
EO materials exhibit flat frequency responses up to 360 GHz [96].
Sepehrian et al. showed 232 Gbps with 16-QAM using a silicon
IQ modulator [131]. Alternatively, resonator-based modulators Such modulators have been used to demonstrate data transmission
have been shown to improve the modulation efficiency in a com- up to 120 Gbps (NRZ) [141] and 120 Gbps (PAM-4) (Fig. 6)
pact footprint [93,133–135]. Baba et al. achieved 50-Gbps NRZ [142]. While current demonstrations have been limited to the
operation with Vπ L = 2.8 V · mm [93]. generation of a pair of sidebands [96,142], the platform could be
This technology has been used for frequency-comb generation, ideal for high efficiency EO comb generation with wider comb
and eight comb lines over a 20-dB bandwidth were produced in spacings.
Review Vol. 10, No. 8 / August 2023 / Optica 984
Fig. 6. (a) Microscope image of silicon Mach–Zehnder interferometer with a plasmonic phase modulator in each arm. (b) Measured eye diagram for
three different modulation formats. Figure adapted from Ref. [142].
4. Indium Phosphide Modulators (Vπ L = 30 V · mm, 8 GHz 3-dB bandwidth) [101], BaTiO3
III-V materials offer the potential for full integration of both on silicon (Vπ L = 2 V · mm, 2 GHz 3-dB bandwidth) [102],
active and passive elements of a transceiver, including the pump LiNbO3 on silicon (Vπ L = 94 V · mm, 40 GHz 3-dB bandwidth)
laser, optical amplifier, and modulator. There has been much [103], and InP on silicon (Vπ L = 12.5 V · mm, 1.5-GHz 3-dB
work done on modulators based on InP. The EO effect in III-V bandwidth) [104]. More recently, Churaev et al. have shown wafer-
materials can arise from several different processes, including scale bonding of thin-film LN to a SiN PIC [148]. While there
the Pockels effect (r 41 = −1.4 pm/V) [143], the Franz–Keldysh exists an increase in the Vπ L by a factor of 2 as compared to LN
effect (bulk), quantum confined Stark effect (quantum well), and PICs, the platform offers promise as a hybrid low-loss PIC combin-
free-carrier effects [104,144]. Demonstrations of comb generation ing χ (2) and χ (3) nonlinearities. For data transmission, 112 Gbps
using InP modulators have implemented the waveguide geometry data transmission for an NRZ signal has been demonstrated using
where a single modulator or several modulators (both intensity a LN-on-silicon hybrid modulator [149].
and phase) are connected in series. For the single-modulator
approach, a traveling-wave-electrode dual-drive MZM [145] and
4. KERR FREQUENCY COMBS
a dual-electrode MZM [146] have been used to generate as many
as 29 comb lines [146]. The number of comb lines was further A. Overview
increased by using a cascaded modulator approach on a single chip. A particularly promising approach for producing on-chip
Bontempi et al. use a dual-drive MZM and two phase modulators frequency combs for data communications utilizes the χ (3)
in series to generate 55 comb lines with a 1-GHz spacing [97]. nonlinearity within microresonators. This process known as Kerr
Furthermore, a similar device has been demonstrated that inte-
frequency-comb generation is based on a nonlinear optical process
grates the cascaded modulators with a distributed Bragg reflector
of FWM in the presence of group-velocity dispersion (GVD),
laser and an SOA to boost the output [147]. The system offers the
which can lead to parametric oscillation in a high-Q microres-
potential for supporting higher data rates with Ogiso et al. previ-
onator pumped with a single-frequency laser source [20,21].
ously reporting InP modulators with a bandwidth of 50 GHz [98].
Previously, FWM had been used in a single-pass fiber geometry to
The high modulation frequency allows for wider comb spacing,
generate multiple sidebands using a two-tone pump [150–154], in
minimizing crosstalk between channels.
which the comb spacing is dictated by the frequency spacing of the
pump fields, and bandwidths >900 nm have been demonstrated
5. Hybrid Modulators [154]. In a microresonator geometry, the pump power builds up
Hybrid modulators consisting of a silicon or SiN waveguide through cavity enhancement and leads to parametric gain for other
combined with an EO material to realize a modulator for comb cavity modes. Once the gain exceeds the propagation and coupling
generation. One approach uses an organic EO material (chromo- losses, parametric oscillation occurs and under suitable conditions
phore DLD164) on a silicon slot waveguide, which has a Vπ L leads to highly cascaded oscillations and phase locking of these
of 2 V · mm [99]. Here, flat-top spectra with seven comb lines modes. Kerr combs are characterized by a rich set of nonlinear
within a 2-dB bandwidth are demonstrated with a spacing of dynamics and can be generated in both the anomalous and normal-
40 GHz. The organic EO material also offers potential for high- GVD regimes. Early demonstrations of Kerr comb generation were
bandwidth operation up to 100 GHz [100]. Other platforms performed in whispering gallery mode (WGM) resonators com-
that show a low Vπ L include a SiN ridge waveguide with LN posed of silica or fluorides [155,156]. Levy et al. first demonstrated
Review Vol. 10, No. 8 / August 2023 / Optica 985
Fig. 8. (a) Generated single-soliton DKS spectrum. (b) SiN microresonator and cross-section images. (c) Bit-error ratio comparison between DKS comb
line and external-cavity laser using 16QAM signal at 40 GBd. Figure adapted by permission from Macmillan Publishers Ltd.: [184].
A key figure of merit (FOM) for Kerr combs is the pump- thermorefractive noise can be reduced by laser cooling with an
to-comb conversion efficiency, which is critical for applications auxiliary laser [200] or via self-cooling by pumping an auxiliary
such as data communications that require sufficient power per cavity mode [201,202].
line for data transmission across the full link, and high wall-plug There have been investigations for implementing Kerr combs
efficiency is needed. Extensive theoretical and experimental studies in other aspects of communications. Previously, Marin-Palamo
of conversion efficiency have been performed [176,189–191], et al. [184] used two separate DKS combs, one for the transmitter
and efficiencies beyond 20% can be achieved only with high FSR and the other as the LO in the receiver. An alternative approach
(>1 THz) and over-coupled resonators [191]. Alternatively, a for generating the LO uses a comb line from the master DKS
feedback approach using a hybrid Mach–Zehnder and ring geom- comb for generating the LO comb in the receiver link for coherent
etry has allowed for the generation of soliton crystal states with detection [203]. The coherence between the master and secondary
pump-to-comb conversion efficiencies of 55% [192]. In addition, combs can allow for phase recovery in the receiver. Alternatively,
conversion efficiencies beyond 50% have been demonstrated by the narrow linewidth comb lines in DKS combs have been used as
using an auxiliary resonator as a selective frequency shifter [193]. coherent pumps for wavelength multicasting, enabling superior
Low comb mode noise is also critical for communication appli- error-free performance as compared to free-running pump lasers
cations. Previous studies have shown that the phase noise of a [204,205].
generated comb line increases quadratically for increasing mode
numbers with respect to the pump mode [194], which is largely
C. Normal-GVD Combs
attributed to the thermorefractive noise in the microresonator
[195,196]. The temperature fluctuation lead to fluctuations in Kerr combs can also be generated and modelocked in the normal-
the repetition rate of the generated soliton. Researchers have GVD regime. While such combs have often been described as
revealed a quiet point for Kerr comb operation [197–199]. This “dark soliton” combs [206,207], bifurcation analysis has revealed
refers to the pump-cavity detuning operating point that balances that the dark structures formed in the normal-GVD regime corre-
the nonlinear effects in the cavity including spectral recoil due spond to interlocked wavefronts, or switching waves [158,161].
to dispersive wave formation and minimizes the coupling of the Switching waves are structures of light that connect the upper and
fluctuations in detuning frequency to the soliton repetition rate. lower homogeneous state solutions of the bistable cavity response.
Yang et al. demonstrated repetition-rate phase noise reduction Two switching waves of opposite polarity that lock to one another
of 36 dB [198]. Furthermore, recent demonstrations show that form the pulsed structure corresponding to a stable, modelocked
Review Vol. 10, No. 8 / August 2023 / Optica 987
normal-GVD comb [158,206,208–212]. Key benefits of normal- of normal-GVD Kerr combs as WDM sources. Xue et al. demon-
GVD Kerr combs include the relative ease of accessing high strated the low RF noise of a normal-GVD comb [217], and Kim
pump-to-comb conversion efficiencies (>30%) even for relatively et al. showed the high coherence of individual comb lines (Fig. 9)
small FSRs (e.g., 100 GHz) and slower power falloffs within the [213]. Jin et al. showed that the comb lines of a normal-GVD comb
spectral region of interest [207,213], which are more ideal for data can reach a fundamental linewidth on the order of a hertz, and
communications. Rizzo et al. showed open eye diagrams at 16 Gbps with some lines
achieving natively error-free BERs (<10−12 ) [225].
1. Background and Comb Properties
The generation of a normal-GVD Kerr comb requires a hard 2. Normal-GVD Combs for Communication
excitation to enable the formation of switching waves. This initial
Several demonstrations of WDM communications have utilized
perturbation can be achieved through pump modulation at the res-
normal-GVD Kerr combs. In one experiment by Fulop et al., 20
onator FSR [211,214], injection locking [215], or local changes in
lines of a 230-GHz comb were split into even and odd channels
the dispersion to allow for MI [207,210,212,213,216–219]. The
pump modulation demonstration showed not only that harmonic where the data in each arm was modulated at 20-GBd using PM-
driving can generate normal-GVD combs but also that desynchro- 64-QAM. At the receiver side, after 80 km of transmission fiber, a
nization of the pump repetition rate from the cavity FSR can shift local oscillator was tuned to each data channel to measure the BER
the comb spectrum [211]. The injection locking scheme showed of each channel. This coherent communications scheme resulted
that a normal-GVD comb maintains its benefits even under such in a final post-FEC BER below 10−15 with an aggregate data rate of
conditions [215]. 4.4 Tbps [219]. A more recent demonstration by Rizzo et al. used
The most prevalent method to generate a normal-GVD comb is a 200-GHz normal-GVD comb for data transmission on a single
through the modification of the effective dispersion of a microres- 32-channel integrated photonic link architecture (Fig. 10) [225].
onator through mode splittings [207–210,212,213,216–219]. This method utilized simple intensity modulation and direct
A mode splitting is formed when modes of different mode fam- detection of each comb line at modest channel speeds to reduce
ilies spectrally overlap and couple. Common mechanisms of the latency and energy per bit and eliminate the requirement of
mode splittings include coupling to different polarization modes FEC and a local oscillator used in coherent communications. The
[220], spatial modes [221], and auxiliary resonator modes [222]. photonic link measured a direct BER better than 10−8 for all data
Simulations of such Kerr combs can be achieved via various meth- channels at 16 Gbps, yielding an aggregate data rate of 512 Gbps
ods. One approach is to approximate the avoided mode crossing [225].
as a simplified two-parameter model [223] that modifies the dis- In addition, the high conversion efficiency and relatively flat
persion operator in the LLE. A more accurate approach simulates spectral profile of a normal-GVD comb make it a strong candidate
the avoided mode crossings, which is important when the splitting for direct modulation of the generated lines without an additional
strength is strong, through a unitary coupling matrix and a modi- amplifier. Even so, the powers generated by a Kerr comb are limited
fied NLSE to account for the FSR difference of the mode families due to the high field confinement and effective nonlinearity. Such
[212]. This method can be extended to different types of splitting a limitation can be overcome by using coherent beam combining,
schemes and accounts for the periodicity of the mode crossings
which can be achieved with high efficiency via the mechanism of
and the power exchange between the modes. Alternatively, the
synchronization [212]. By passively coupling individual combs
dispersion can be locally modified via a frequency-domain point
together, synchronization allows for the repetition rates of indi-
defect in a photonic crystal resonator. By controlling the strength
of this defect along with the pump-cavity detuning, both dark and vidual combs to become equal, and if the pump frequencies are
bright pulses can be induced, allowing for control of the bandwidth identical, the comb lines will overlap spectrally to allow for coher-
and efficiency of the generated comb [224]. Table 5 summarizes ent combining [212,226]. Recently, Kim et al. showed that not
results for several different methods of generating normal-GVD only can normal-GVD Kerr combs be synchronized like their DKS
combs. counterparts [226,227], but they also can be coherently combined
Although the noise properties of normal-GVD combs have via on-chip synchronization for a nearly 2× increase in the comb
not been studied as extensively as those of DKS combs, several power to extend the power range of a natively spaced comb with
demonstrations have shown key properties that show the usability high efficiency [212].
Fig. 9. (a) Microscope image of SiN coupled ring resonator with integrated resistive heaters. (b) Simulation of resonances using experimental parameters
from [213]. Periodic effects of mode crossings can be observed. (c) Normal-GVD comb spectrum with 201.6 GHz comb spacing and 41% pump-to-comb
conversion efficiency. Power levels of 50 µW, 100 µW, and 1 mW are indicated with red, green, and black dashed lines, respectively. Figure adapted from
Ref. [213].
Fig. 10. (a) Microscope image of fabricated 32-channel transmitter chip and zoom-in of all active and passive devices on the chip. (b) Measured BER
comparing comb performance to cw laser at 1559.8 nm. (c) Measured BER for multiple comb channels in the C- and L-bands. Figure adapted from Ref.
[225].
3. Fully Integrated Kerr Comb Source comb [230]. Shen et al. fully packaged a DFB laser chip with a SiN
While the microresonator was fabricated on a compact pho- chip to show injection locking and DKS generation [Figs. 11(d)
tonic platform, most previous demonstrations of Kerr comb and 11(e)] [232]. Furthermore, work by Jin et al. has shown that
generation required pumping with a CW laser amplified with an injection locking can also be used for generating a normal-GVD
erbium-doped fiber amplifier, resulting in a table-top demonstra- comb [215]. These approaches show a path towards realizing a
tion. Recently, significant progress has been made towards full
fully integrated Kerr comb source that is critical for realizing a
integration of Kerr combs. These advancements used a hybrid inte-
gration approach with SOA gain chips [228], FP laser diode chips compact photonic integrated WDM source that could be not only
[229,230], and distributed feedback (DFB) laser chips [215,231– implemented in data-center transceivers but also copackaged with
233]. Early work by Liang et al. showed that the DFB laser can be electronic chips.
injection locked to a cavity mode of a magnesium fluoride WGM To overcome power limitations of the comb source, on-chip
resonator and allow for subsequent comb generation [231]. Stern amplifiers can be implemented to boost the power of the generated
et al. demonstrated hybrid integration of a reflective SOA gain
comb lines. One approach is to utilize a III-V gain medium with
chip and SiN chips and demonstrated DKS generation using two
different mechanisms. One approach is to use the microresonator, epitaxial growth on silicon [234,235] or with bonding to silicon
not only the comb generator but also the end reflection facet of [236–238]. Another approach is to use rare earth doping in wave-
the entire integrated laser structure. The other method used a guides with demonstrations in aluminum nitride, silicon nitride,
Sagnac loop as the end mirror before the microresonator, which silicon dioxide, and lithium niobate [239–244]. Alternatively, cw
separates the pump laser portion from the nonlinear element parametric gain has been utilized for integrated amplifiers with
[Figs. 11(a)–11(c)] [228]. Subsequently, Pavlov et al. showed that a
demonstrations in waveguides [245,246] and resonators [247].
Fabry–Perot laser diode chip can be coupled to a WGM microres-
onator to produce a DKS [229]. Raja et al. demonstrated that Similar to MLLs, the amplifier properties may require optimiza-
the backscattering from a SiN chip to a Fabry–Perot laser diode tion to minimize the ASE being coupled back into the cavity that
chip allowed for injection locking and the generation of a DKS could cause destabilization of the Kerr comb.
Review Vol. 10, No. 8 / August 2023 / Optica 989
Fig. 11. (a) Microscope image of integrated Kerr comb source with the reflective SOA waveguide and the nonlinear microresonator. (b) Photograph of
integrated comb source. The RSOA is edge-coupled to the SiN chip. (c) Measured single-soliton Kerr comb spectrum. (d) Image of fully packaged DFB
laser and SiN chip in a butterfly package. (e) Optical spectrum of 40 and 15 GHz Kerr comb states. (a)–(c) adapted from Ref. [228]; (d) and (e) adapted by
permission from Macmillan Publishers Ltd.: [232].
5. CONCLUSIONS AND OUTLOOK Data availability. No data were generated or analyzed in the presented
research.
In order to overcome the energy-bandwidth limitations of
electronic interconnects, more optical technologies are being
deployed in data centers. To meet the demands towards ultra-high- REFERENCES
bandwidth links in the Tbps regime, much tighter optoelectronic 1. M. A. Taubenblatt, “Optical interconnects for high-performance
integration is required for communication at the board and chip computing,” J. Lightwave Technol. 30, 448–457 (2012).
level. The recent advances in chip-scale frequency-comb devices 2. T. Liljeberg, “Silicon photonics and the future of optical connectivity in
offer promise as multiwavelength optical carriers for WDM trans- the data center,” in IEEE Optical Interconnects Conference (OI) (2017),
pp. 1–2.
ceivers Integrated comb sources with 64 and 128 channels in 3. X. Zhou, H. Liu, and R. Urata, “Datacenter optics: requirements, tech-
conjunction with higher-order modulation formats such as PAM4 nologies, and trends (Invited Paper),” Chin. Opt. Lett. 15, 120008
at 20 Gbps can allow for reaching aggregate data rates of 5 Tbps (2017).
and 10 Tbps, respectively, with a distinct path towards further 4. S. Rumley, M. Bahadori, R. Polster, S. D. Hammond, D. M. Calhoun,
K. Wen, A. Rodrigues, and K. Bergman, “Optical interconnects for
data rate scaling. In this review, we presented an overview of the extreme scale computing systems,” Parallel Comput. 64, 65–80
chip-scale OFC technology that offers promise for realizing ultra- (2017).
high-bandwidth integrated photonic transceivers for deployment 5. Z. Zhou, R. Chen, X. Li, and T. Li, “Development trends in silicon pho-
in future high-bandwidth-density data-center architecture. Key tonics for data centers,” Opt. Fiber Technol. 44, 13–23 (2018).
6. Q. Cheng, M. Bahadori, M. Glick, S. Rumley, and K. Bergman, “Recent
challenges for next-generation chip-scale comb technology include
advances in optical technologies for data centers: a review,” Optica 5,
increased power per comb line and comb-line power equalization 1354–1370 (2018).
while achieving a high overall wall-plug efficiency. We envision 7. P. Marin-Palomo, J. N. Kemal, T. J. Kippenberg, W. Freude, S. Randel,
a multilayer photonic platform that allows for the integration of and C. Koos, “Performance of chip-scale optical frequency comb gen-
passive and active photonic elements and for the realization of erators in coherent WDM communications,” Opt. Express 28, 12897–
12910 (2020).
a photonic transceiver that enables the integration of photonic 8. H. Hu and L. K. Oxenløwe, “Chip-based optical frequency combs
interconnects in data centers for high-bandwidth density, low-cost, for high-capacity optical communications,” Nanophotonics 10,
energy-efficient communication networks. 1367–1385 (2021).
9. A. Ghiasi, “Large data centers interconnect bottlenecks,” Opt. Express
Funding. Defense Advanced Research Projects Agency (HR0011-19-2-0014); 23, 2085–2090 (2015).
Advanced Research Projects Agency - Energy (DE-AR0000843); Air Force Office 10. L. Carroll, J.-S. Lee, C. Scarcella, K. Gradkowski, M. Duperron, H. Lu,
of Scientific Research (FA9550-15-1-0303). Y. Zhao, C. Eason, P. Morrissey, M. Rensing, S. Collins, H. Y. Hwang,
and P. O’Brien, “Photonic packaging: Transforming silicon photonic
Acknowledgment. The authors thank Dr. J. K. Jang and Y. Zhao for criti- integrated circuits into photonic devices,” Appl. Sci. 6, 426 (2016).
cally reading the manuscript and helpful discussions. We gratefully acknowledge 11. N. Kucharewski, C. Minkenberg, and G. Rodriguez, “Network architec-
E. Sahagún (Scixel) for illustrations in Fig. 1. ture in the era of integrated optics,” in Optical Fiber Communication
Conference (Optical Society of America, 2018), paper M3J.7.
Disclosures. YO and ALG: Xscape Photonics (I,E,P), ML: Xscape Photonics 12. S. Pitris, M. Moralis-Pegios, T. Alexoudi, Y. Ban, P. D. Heyn, J. V.
(I,P). Campenhout, J. Lambrecht, H. Ramon, X. Yin, J. Bauwelinck, and
Review Vol. 10, No. 8 / August 2023 / Optica 990
N. Pleros, “O-band silicon photonic transmitters for datacom and 33. K. Sato, I. Kotaka, Y. Kondo, and M. Yamamoro, “Actively mode-locked
computercom interconnects,” J. Lightwave Technol. 37, 5140–5148 strained-InGaAsP multiquantum-well lasers integrated with electroab-
(2019). sorption modulators and distributed Bragg reflectors,” IEEE J. Sel. Top.
13. B. Buscaino, B. D. Taylor, and J. M. Kahn, “Multi-Tb/s-per-fiber Quantum Electron. 2, 557–565 (1996).
coherent co-packaged optical interfaces for data center switches,” 34. R. Arkhipov, A. Pimenov, M. Radziunas, D. Rachinskii, A. G. Vladimirov,
J. Lightwave Technol. 37, 3401–3412 (2019). D. Arsenijević, H. Schmeckebier, and D. Bimberg, “Hybrid mode lock-
14. S. Fathololoumi, K. Nguyen, H. Mahalingam, et al., “1.6Tbps silicon ing in semiconductor lasers: simulations, analysis, and experiments,”
photonics integrated circuit for co-packaged optical-IO switch appli- IEEE J. Sel. Top. Quantum Electron. 19, 1100208 (2013).
cations,” in Optical Fiber Communication Conference (OFC) (Optical 35. R. N. Hall, G. E. Fenner, J. D. Kingsley, T. J. Soltys, and R. O. Carlson,
Society of America, 2020), paper T3H.1. “Coherent light emission from GaAs junctions,” Phys. Rev. Lett. 9, 366–
15. D. F. Logan, S. Gebrewold, K. Murray, A. Dewanjee, E. Huante-Ceron, 368 (1962).
D. Kim, A. Baker, M. Kukiela, F. Znidarsic, M. Koehler, J. Whiteaway, R. 36. D. Liang, G. Roelkens, R. Baets, and J. E. Bowers, “Hybrid integrated
Chen, C. Dorschky, and G. Roell, “800 Gb/s silicon photonic transmit- platforms for silicon photonics,” Materials 3, 1782–1802 (2010).
ter for copackaged optics,” in IEEE Photonics Conference (IPC) (2020), 37. M. J. R. Heck, J. F. Bauters, M. L. Davenport, J. K. Doylend, S. Jain,
pp. 1–2. G. Kurczveil, S. Srinivasan, Y. Tang, and J. E. Bowers, “Hybrid silicon
photonic integrated circuit technology,” IEEE J. Sel. Top. Quantum
16. S. Fathololoumi, D. Hui, S. Jadhav, et al., “1.6 Tbps silicon pho-
Electron. 19, 6100117 (2013).
tonics integrated circuit and 800 Gbps photonic engine for switch
38. J. Zhang, G. Muliuk, J. Juvert, et al., “III-V-on-Si photonic integrated cir-
co-packaging demonstration,” J. Lightwave Technol. 39, 1155–1161
cuits realized using micro-transfer-printing,” APL Photonics 4, 110803
(2021).
(2019).
17. C. Minkenberg, R. Krishnaswamy, A. Zilkie, and D. Nelson, “Co-
39. E. U. Rafailov, M. A. Cataluna, W. Sibbett, N. D. Il’inskaya, Y. M.
packaged datacenter optics: opportunities and challenges,” IET
Zadiranov, A. E. Zhukov, V. M. Ustinov, D. A. Livshits, A. R. Kovsh,
Optoelectron. 15, 77–91 (2021).
and N. N. Ledentsov, “High-power picosecond and femtosecond
18. B. Buscaino, E. Chen, J. W. Stewart, T. Pham, and J. M. Kahn, “External pulse generation from a two-section mode-locked quantum-dot laser,”
vs. integrated light sources for intra-data center co-packaged optical Appl. Phys. Lett. 87, 081107 (2005).
interfaces,” J. Lightwave Technol. 39, 1984–1996 (2021). 40. L. Hou, M. Haji, J. Akbar, B. Qiu, and A. C. Bryce, “Low divergence
19. S. T. Cundiff and J. Ye, “Colloquium: femtosecond optical frequency angle and low jitter 40 GHz AlGaInAs/InP 1.55 µm mode-locked
combs,” Rev. Mod. Phys. 75, 325–342 (2003). lasers,” Opt. Lett. 36, 966–968 (2011).
20. T. J. Kippenberg, A. L. Gaeta, M. Lipson, and M. L. Gorodetsky, 41. H. Fan, C. Wu, M. El-Aasser, N. Dutta, U. Koren, and A. Piccirilli,
“Dissipative Kerr solitons in optical microresonators,” Science 361, “Colliding pulse mode locked laser,” in Optical Fiber Communication
eaan8083 (2018). Conference. Technical Digest Postconference Edition. Trends in
21. A. L. Gaeta, M. Lipson, and T. J. Kippenberg, “Photonic-chip-based Optics and Photonics Vol.37 (IEEE Cat. No. 00CH37079) (2000), Vol. 1,
frequency combs,” Nat. Photonics 13, 158–169 (2019). pp. 260–262.
22. S. A. Diddams, K. Vahala, and T. Udem, “Optical frequency combs: 42. M.-C. Lo, R. Guzmán, M. Ali, R. Santos, L. Augustin, and G. Carpintero,
coherently uniting the electromagnetic spectrum,” Science 369, “1.8-THz-wide optical frequency comb emitted from monolithic pas-
eaay3676 (2020). sively mode-locked semiconductor quantum-well laser,” Opt. Lett. 42,
23. T. Fortier and E. Baumann, “20 years of developments in optical fre- 3872–3875 (2017).
quency comb technology and applications,” Commun. Phys. 2, 1–16 43. S. Pan, H. Zhang, Z. Liu, M. Liao, M. Tang, D. Wu, X. Hu, J. Yan, L.
(2019). Wang, M. Guo, Z. Wang, T. Wang, P. M. Smowton, A. Seeds, H. Liu,
24. T. Pfau, S. Hoffmann, and R. Noé, “Hardware-efficient coherent digital X. Xiao, and S. Chen, “Multi-wavelength 128 Gbit/s/λ PAM4 optical
receiver concept with feedforward carrier recovery for M-QAM constel- transmission enabled by a 100 GHz quantum dot mode-locked optical
lations,” J. Lightwave Technol. 27, 989–999 (2009). frequency comb,” J. Phys. D 55, 144001 (2021).
25. J. N. Kemal, J. Pfeifle, P. Marin-Palomo, M. D. G. Pascual, S. Wolf, F. 44. Z. G. Lu, J. R. Liu, S. Raymond, P. J. Poole, P. J. Barrios, and D. Poitras,
Smyth, W. Freude, and C. Koos, “Multi-wavelength coherent trans- “312-fs pulse generation from a passive C-band InAs/InP quantum dot
mission using an optical frequency comb as a local oscillator,” Opt. mode-locked laser,” Opt. Express 16, 10835–10840 (2008).
Express 24, 25432–25445 (2016). 45. K. Merghem, C. Calò, R. Rosales, X. Lafosse, G. Aubin, A. Martinez, F.
26. A. Kakkar, J. Rodrigo Navarro, R. Schatz, X. Pang, O. Ozolins, A. Lelarge, and A. Ramdane, “Stability of optical frequency comb gen-
Udalcovs, H. Louchet, S. Popov, and G. Jacobsen, “Laser frequency erated with InAs/InP quantum-dash-based passive mode-locked
noise in coherent optical systems: spectral regimes and impairments,” lasers,” IEEE J. Quantum Electron. 50, 275–280 (2014).
Sci. Rep. 7, 844 (2017). 46. G.-H. Duan, A. Shen, A. Akrout, F. V. Dijk, F. Lelarge, F. Pommereau,
27. V. Torres-Company, J. Schröder, A. Fülöp, M. Mazur, L. Lundberg, O. B. O. LeGouezigou, J.-G. Provost, H. Gariah, F. Blache, F. Mallecot, K.
Merghem, A. Martinez, and A. Ramdane, “High performance InP-
Helgason, M. Karlsson, and P. A. Andrekson, “Laser frequency combs
based quantum dash semiconductor mode-locked lasers for optical
for coherent optical communications,” J. Lightwave Technol. 37, 1663–
communications,” Bell Labs Tech. J. 14, 63–84 (2009).
1670 (2019).
47. C. Gosset, K. Merghem, A. Martinez, G. Moreau, G. Patriarche, G.
28. T. Sylvestre, E. Genier, A. N. Ghosh, P. Bowen, G. Genty, J. Troles,
Aubin, A. Ramdane, J. Landreau, and F. Lelarge, “Subpicosecond
A. Mussot, A. C. Peacock, M. Klimczak, A. M. Heidt, J. C. Travers,
pulse generation at 134 GHz using a quantum-dash-based Fabry-
O. Bang, and J. M. Dudley, “Recent advances in supercontinuum
Perot laser emitting at 1.56 µm,” Appl. Phys. Lett. 88, 241105
generation in specialty optical fibers [invited],” J. Opt. Soc. Am. B 38,
(2006).
F90–F103 (2021). 48. K. Merghem, A. Akrout, A. Martinez, G. Aubin, A. Ramdane, F. Lelarge,
29. C. Hönninger, R. Paschotta, F. Morier-Genoud, M. Moser, and U. and G.-H. Duan, “Pulse generation at 346 GHz using a passively mode
Keller, “Q-switching stability limits of continuous-wave passive mode locked quantum-dash-based laser at 1.55 µm,” Appl. Phys. Lett. 94,
locking,” J. Opt. Soc. Am. B 16, 46–56 (1999). 021107 (2009).
30. C. G. E. Alfieri, D. Waldburger, S. M. Link, E. Gini, M. Golling, G. 49. R. Rosales, K. Merghem, A. Martinez, A. Akrout, J.-P. Tourrenc, A.
Eisenstein, and U. Keller, “Optical efficiency and gain dynamics of Accard, F. Lelarge, and A. Ramdane, “InAs/InP quantum-dot pas-
modelocked semiconductor disk lasers,” Opt. Express 25, 6402–6420 sively mode-locked lasers for 1.55-µm applications,” IEEE J. Sel. Top.
(2017). Quantum Electron. 17, 1292–1301 (2011).
31. D. Waldburger, A. S. Mayer, C. G. E. Alfieri, J. Nürnberg, A. R. Johnson, 50. K. Van Gasse, S. Uvin, V. Moskalenko, S. Latkowski, G. Roelkens,
X. Ji, A. Klenner, Y. Okawachi, M. Lipson, A. L. Gaeta, and U. Keller, E. Bente, and B. Kuyken, “Recent advances in the photonic integra-
“Tightly locked optical frequency comb from a semiconductor disk tion of mode-locked laser diodes,” IEEE Photonics Technol. Lett. 31,
laser,” Opt. Express 27, 1786–1797 (2019). 1870–1873 (2019).
32. J. Bowers, P. Morton, A. Mar, and S. Corzine, “Actively mode-locked 51. E. Vissers, S. Poelman, C. O. D. Beeck, K. V. Gasse, and B. Kuyken,
semiconductor lasers,” IEEE J. Quantum Electron. 25, 1426–1439 “Hybrid integrated mode-locked laser diodes with a silicon nitride
(1989). extended cavity,” Opt. Express 29, 15013–15022 (2021).
Review Vol. 10, No. 8 / August 2023 / Optica 991
52. V. Moskalenko, S. Latkowski, S. Tahvili, T. D. Vries, M. Smit, and “High-pulse-energy III-V-on-silicon-nitride mode-locked laser,” APL
E. Bente, “Record bandwidth and sub-picosecond pulses from a Photonics 6, 096102 (2021).
monolithically integrated mode-locked quantum well ring laser,” Opt. 72. S. Cuyvers, B. Haq, C. O. D. Beeck, S. Poelman, A. Hermans, Z. Wang,
Express 22, 28865–28874 (2014). A. Gocalinska, E. Pelucchi, B. Corbett, G. Roelkens, K. V. Gasse,
53. J. S. Parker, A. Bhardwaj, P. R. A. Binetti, Y.-J. Hung, and L. A. Coldren, and B. Kuyken, “Low noise heterogeneous III-V-on-silicon-nitride
“Monolithically integrated gain-flattened ring mode-locked laser for mode-locked comb laser,” Laser Photonics Rev. 15, 2000485 (2021).
comb-line generation,” IEEE Photonics Technol. Lett. 24, 131–133 73. K. Shtyrkova, P. T. Callahan, N. Li, E. S. Magden, A. Ruocco, D.
(2012). Vermeulen, F. X. Kärtner, M. R. Watts, and E. P. Ippen, “Integrated
54. S. Latkowski, V. Moskalenko, S. Tahvili, L. Augustin, M. Smit, K. CMOS-compatible Q-switched mode-locked lasers at 1900nm with
Williams, and E. Bente, “Monolithically integrated 2.5 GHz extended an on-chip artificial saturable absorber,” Opt. Express 27, 3542–3556
cavity mode-locked ring laser with intracavity phase modulators,” Opt. (2019).
Lett. 40, 77–80 (2015). 74. S. Srinivasan, A. Arrighi, M. J. R. Heck, J. Hutchinson, E. Norberg, G.
55. E. Kleijn, M. K. Smit, and X. J. M. Leijtens, “Multimode interference Fish, and J. E. Bowers, “Harmonically mode-locked hybrid silicon laser
reflectors: A new class of components for photonic integrated circuits,” with intra-cavity filter to suppress supermode noise,” IEEE J. Sel. Top.
J. Lightwave Technol. 31, 3055–3063 (2013). Quantum Electron. 20, 8–15 (2014).
56. R. Guzmán, C. Gordon, L. Orbe, and G. Carpintero, “1 GHz InP on-chip 75. S. Liu, X. Wu, D. Jung, J. C. Norman, M. J. Kennedy, H. K. Tsang, A.
monolithic extended cavity colliding-pulse mode-locked laser,” Opt. C. Gossard, and J. E. Bowers, “High-channel-count 20 GHz passively
Lett. 42, 2318–2321 (2017). mode-locked quantum dot laser directly grown on Si with 4.1 Tbit/s
57. M.-C. Lo, R. Guzmán, and G. Carpintero, “Femtosecond pulse and transmission capacity,” Optica 6, 128–134 (2019).
terahertz two-tone generation from facet-free multi-segment laser 76. P. Delfyett, S. Gee, M.-T. Choi, H. Izadpanah, W. Lee, S. Ozharar, F.
diode in InP-based generic foundry platform,” Opt. Express 26, Quinlan, and T. Yilmaz, “Optical frequency combs from semiconduc-
18386–18398 (2018). tor lasers and applications in ultrawideband signal processing and
58. T. Komljenovic, M. Davenport, J. Hulme, A. Y. Liu, C. T. Santis, A. communications,” J. Lightwave Technol. 24, 2701–2719 (2006).
Spott, S. Srinivasan, E. J. Stanton, C. Zhang, and J. E. Bowers, 77. K. Ogawa, “Analysis of mode partition noise in laser transmission sys-
“Heterogeneous silicon photonic integrated circuits,” J. Lightwave tems,” IEEE J. Quantum Electron. 18, 849–855 (1982).
Technol. 34, 20–35 (2016). 78. K. Haneda, M. Yoshida, M. Nakazawa, H. Yokoyama, and Y. Ogawa,
59. A. Fang, H. Park, R. Jones, O. Cohen, M. Paniccia, and J. Bowers, “Linewidth and relative intensity noise measurements of longitudinal
“A continuous-wave hybrid AlGaInAs-silicon evanescent laser,” IEEE modes in ultrahigh-speed mode-locked laser diodes,” Opt. Lett. 30,
Photonics Technol. Lett. 18, 1143–1145 (2006). 1000–1002 (2005).
60. S. Keyvaninia, M. Muneeb, S. Stanković, P. J. V. Veldhoven, D. V. 79. I. Henning, L. Westbrook, A. W. Nelson, and P. J. Fiddyment,
Thourhout, and G. Roelkens, “Ultra-thin DVS-BCB adhesive bonding “Measurements of the linewidth of ridge-guide DFB lasers,” Electron.
of III-V wafers, dies and multiple dies to a patterned silicon-on-insulator Lett. 20, 885–887 (1984).
substrate,” Opt. Mater. Express 3, 35–46 (2013). 80. L. G. Kazovsky, “Impact of laser phase noise on optical heterodyne
61. J. E. Bowers, J. T. Bovington, A. Y. Liu, and A. C. Gossard, “A path communication systems,” J. Opt. Commun. 7, 66–78 (1986).
to 300 mm hybrid silicon photonic integrated circuits,” in Optical 81. G. Kurczveil, M. A. Seyedi, D. Liang, M. Fiorentino, and R. G.
Fiber Communication Conference (Optical Society of America, 2014), Beausoleil, “Error-free operation in a hybrid-silicon quantum dot comb
paper Th1C.1. laser,” IEEE Photonics Technol. Lett. 30, 71–74 (2018).
62. T. Komljenovic, D. Huang, P. Pintus, M. A. Tran, M. L. Davenport, 82. A. Moscoso-Mártir, A. Tabatabaei-Mashayekh, J. Müller, J. Nojić, R.
and J. E. Bowers, “Photonic integrated circuits using heterogeneous Setter, M. Nielsen, A. Sandomirsky, S. Rockman, E. Mentovich, F.
integration on silicon,” Proc. IEEE 106, 2246–2257 (2018). Merget, A. Garreau, F. Lelarge, and J. Witzens, “8-channel WDM silicon
63. A. Lee, Q. Jiang, M. Tang, A. Seeds, and H. Liu, “Continuous-wave photonics transceiver with SOA and semiconductor mode-locked
InAs/GaAs quantum-dot laser diodes monolithically grown on Si laser,” Opt. Express 26, 25446–25459 (2018).
substrate with low threshold current densities,” Opt. Express 20, 83. P. Marin-Palomo, J. N. Kemal, P. Trocha, S. Wolf, K. Merghem,
22181–22187 (2012). F. Lelarge, A. Ramdane, W. Freude, S. Randel, and C. Koos,
64. A. Y. Liu, C. Zhang, J. Norman, A. Snyder, D. Lubyshev, J. M. Fastenau, “Comb-based WDM transmission at 10 Tbit/s using a DC-driven
A. W. K. Liu, A. C. Gossard, and J. E. Bowers, “High performance con- quantum-dash mode-locked laser diode,” Opt. Express 27,
tinuous wave 1.3 µm quantum dot lasers on silicon,” Appl. Phys. Lett. 31110–31129 (2019).
104, 041104 (2014). 84. D. Arsenijević, H. Schmeckebier, M. Kleinert, E. Rouvalis, R. Ziegler,
65. A. Verschelde, K. V. Gasse, B. Kuyken, M. Giudici, G. Huyet, and M. A. G. Steffan, and D. Bimberg, “Quantum-dot mode-locked lasers for
Marconi, “Analysis of the phase-locking dynamics of a III-V-on-silicon microwave-signal generation and 160 Gb/s optical communication,” in
frequency comb laser,” OSA Contin. 4, 129–136 (2021). IEEE Photonics Conference (IPC) (2015), pp. 635–636.
66. B. R. Koch, A. W. Fang, O. Cohen, and J. E. Bowers, “Mode-locked sili- 85. B. R. Koch, J. S. Barton, M. Masanovic, Z. Hu, J. E. Bowers, and D. J.
con evanescent lasers,” Opt. Express 15, 11225–11233 (2007). Blumenthal, “Monolithic mode-locked laser and optical amplifier for
67. A. W. Fang, B. R. Koch, K.-G. Gan, H. Park, R. Jones, O. Cohen, regenerative pulsed optical clock recovery,” IEEE Photonics Technol.
M. J. Paniccia, D. J. Blumenthal, and J. E. Bowers, “A racetrack Lett. 19, 641–643 (2007).
mode-locked silicon evanescent laser,” Opt. Express 16, 1393–1398 86. A. Parriaux, K. Hammani, and G. Millot, “Electro-optic frequency
(2008). combs,” Adv. Opt. Photonics 12, 223–287 (2020).
68. S. Keyvaninia, S. Uvin, M. Tassaert, X. Fu, S. Latkowski, J. Mariën, 87. M. Imran, P. M. Anandarajah, A. Kaszubowska-Anandarajah, N.
L. Thomassen, F. Lelarge, G. Duan, P. Verheyen, G. Lepage, J. V. Sambo, and L. Potí, “A survey of optical carrier generation techniques
Campenhout, E. Bente, and G. Roelkens, “Narrow-linewidth short- for terabit capacity elastic optical networks,” IEEE Commun. Surv.
pulse III-V-on-silicon mode-locked lasers based on a linear and ring Tuts. 20, 211–263 (2018).
cavity geometry,” Opt. Express 23, 3221–3229 (2015). 88. C. Wang, M. Zhang, X. Chen, M. Bertrand, A. Shams-Ansari, S.
69. S. Keyvaninia, S. Uvin, M. Tassaert, Z. Wang, X. Fu, S. Latkowski, J. Chandrasekhar, P. Winzer, and M. Lončar, “Integrated lithium niobate
Marien, L. Thomassen, F. Lelarge, G. Duan, G. Lepage, P. Verheyen, electro-optic modulators operating at CMOS-compatible voltages,”
J. V. Campenhout, E. Bente, and G. Roelkens, “III–V-on-silicon anti- Nature 562, 101–104 (2018).
colliding pulse-type mode-locked laser,” Opt. Lett. 40, 3057–3060 89. M. Zhang, B. Buscaino, C. Wang, A. Shams-Ansari, C. Reimer, R. Zhu,
(2015). J. M. Kahn, and M. Lončar, “Broadband electro-optic frequency comb
70. Z. Wang, K. Van Gasse, V. Moskalenko, S. Latkowski, E. Bente, B. generation in a lithium niobate microring resonator,” Nature 568, 373–
Kuyken, and G. Roelkens, “A III-V-on-Si ultra-dense comb laser,” Light 377 (2019).
Sci. Appl. 6, e16260 (2017). 90. M. Yu, D. Barton, III, R. Cheng, C. Reimer, P. Kharel, L. He, L. Shao, D.
71. A. Hermans, K. Van Gasse, J. O. Kjellman, C. Caër, T. Nakamura, Y. Zhu, Y. Hu, H. R. Grant, L. Johansson, Y. Okawachi, A. L. Gaeta, M.
Inada, K. Hisada, T. Hirasawa, S. Cuyvers, S. Kumari, A. Marinins, Zhang, and M. Lončar, “Integrated femtosecond pulse generator on
R. Jansen, G. Roelkens, P. Soussan, X. Rottenberg, and B. Kuyken, thin-film lithium niobate,” Nature 612, 252–258 (2022).
Review Vol. 10, No. 8 / August 2023 / Optica 992
91. K. Li, S. Liu, D. J. Thomson, W. Zhang, X. Yan, F. Meng, C. G. Littlejohns, lithium niobate on insulator modulator,” in 45th European Conference
H. Du, M. Banakar, M. Ebert, W. Cao, D. Tran, B. Chen, A. Shakoor, P. on Optical Communication (ECOC) (2019), pp. 1–4.
Petropoulos, and G. T. Reed, “Electronic–photonic convergence for sil- 109. K. Luke, P. Kharel, C. Reimer, L. He, M. Loncar, and M. Zhang, “Wafer-
icon photonics transmitters beyond 100 Gbps on–off keying,” Optica 7, scale low-loss lithium niobate photonic integrated circuits,” Opt.
1514–1516 (2020). Express 28, 24452–24458 (2020).
92. H. Zhang, M. Li, Y. Zhang, D. Zhang, Q. Liao, J. He, S. Hu, B. Zhang, 110. M. Kourogi, K. Nakagawa, and M. Ohtsu, “Wide-span optical fre-
L. Wang, X. Xiao, N. Qi, and S. Yu, “800 Gbit/s transmission over 1 km quency comb generator for accurate optical frequency difference
single-mode fiber using a four-channel silicon photonic transmitter,” measurement,” IEEE J. Quantum Electron. 29, 2693–2701 (1993).
Photonics Res. 8, 1776–1782 (2020). 111. T. Kobayashi, T. Sueta, Y. Cho, and Y. Matsuo, “High-repetition-rate
93. T. Baba, S. Akiyama, M. Imai, N. Hirayama, H. Takahashi, Y. Noguchi, T. optical pulse generator using a Fabry-Perot electro-optic modulator,”
Horikawa, and T. Usuki, “50-Gb/s ring-resonator-based silicon modu- Appl. Phys. Lett. 21, 341 (2003).
lator,” Opt. Express 21, 11869–11876 (2013). 112. M. Kato, K. Fujiura, and T. Kurihara, “Generation of super-stable
94. K. P. Nagarjun, V. Jeyaselvan, S. K. Selvaraja, and V. R. Supradeepa, 40 GHz pulses from Fabry-Perot resonator integrated with
“Generation of tunable, high repetition rate optical frequency combs electro-optic phase modulator,” Electron. Lett. 40, 299–301 (2004).
using on-chip silicon modulators,” Opt. Express 26, 10744–10753 113. V. S. Ilchenko, A. A. Savchenkov, A. B. Matsko, and L. Maleki,
(2018). “Whispering-gallery-mode electro-optic modulator and photonic
95. C. Haffner, W. Heni, Y. Fedoryshyn, J. Niegemann, A. Melikyan, D. L. microwave receiver,” J. Opt. Soc. Am. B 20, 333–342 (2003).
Elder, B. Baeuerle, Y. Salamin, A. Josten, U. Koch, C. Hoessbacher, 114. A. Rueda, F. Sedlmeir, M. Kumari, G. Leuchs, and H. G. L. Schwefel,
F. Ducry, L. Juchli, A. Emboras, D. Hillerkuss, M. Kohl, L. R. Dalton, “Resonant electro-optic frequency comb,” Nature 568, 378–381
C. Hafner, and J. Leuthold, “All-plasmonic Mach-Zehnder modulator (2019).
enabling optical high-speed communication at the microscale,” Nat. 115. B. Buscaino, M. Zhang, M. Lončar, and J. M. Kahn, “Design of efficient
Photonics 9, 525–528 (2015). resonator-enhanced electro-optic frequency comb generators,” J.
96. S. Ummethala, T. Harter, K. Koehnle, Z. Li, S. Muehlbrandt, Y. Lightwave Technol. 38, 1400–1413 (2020).
Kutuvantavida, J. Kemal, P. Marin-Palomo, J. Schaefer, A. Tessmann, 116. Y. Hu, M. Yu, B. Buscaino, N. Sinclair, D. Zhu, R. Cheng, A. Shams-
S. K. Garlapati, A. Bacher, L. Hahn, M. Walther, T. Zwick, S. Randel, Ansari, L. Shao, M. Zhang, J. M. Kahn, and M. Lončar, “High-efficiency
W. Freude, and C. Koos, “THz-to-optical conversion in wireless com- and broadband electro-optic frequency combs enabled by coupled
munications using an ultra-broadband plasmonic modulator,” Nat. micro-resonators,” Nat. Photonics 16, 679–685 (2021).
Photonics 13, 519–524 (2019). 117. B. H. Kolner and M. Nazarathy, “Temporal imaging with a time lens,”
97. F. Bontempi, N. Andriolli, F. Scotti, M. Chiesa, and G. Contestabile, Opt. Lett. 14, 630–632 (1989).
118. J. van Howe and C. Xu, “Ultrafast optical signal processing based upon
“Comb line multiplication in an InP integrated photonic circuit based
space-time dualities,” J. Lightwave Technol. 24, 2649–2662 (2006).
on cascaded modulators,” IEEE J. Sel. Top. Quantum Electron. 25,
119. R. Wu, V. R. Supradeepa, C. M. Long, D. E. Leaird, and A. M. Weiner,
3500107 (2019).
“Generation of very flat optical frequency combs from continuous-
98. Y. Ogiso, J. Ozaki, Y. Ueda, N. Kashio, N. Kikuchi, E. Yamada, H.
wave lasers using cascaded intensity and phase modulators driven by
Tanobe, S. Kanazawa, H. Yamazaki, Y. Ohiso, T. Fujii, and M. Kohtoku,
tailored radio frequency waveforms,” Opt. Lett. 35, 3234–3236 (2010).
“Over 67 GHz bandwidth and 1.5 V Vπ InP-based optical IQ modulator
120. A. Ishizawa, T. Nishikawa, A. Mizutori, H. Takara, A. Takada, T. Sogawa,
with n-i-p-n heterostructure,” J. Lightwave Technol. 35, 1450–1455
and M. Koga, “Phase-noise characteristics of a 25-GHz-spaced opti-
(2017).
cal frequency comb based on a phase- and intensity-modulated laser,”
99. C. Weimann, P. C. Schindler, R. Palmer, S. Wolf, D. Bekele, D. Korn,
Opt. Express 21, 29186–29194 (2013).
J. Pfeifle, S. Koeber, R. Schmogrow, L. Alloatti, D. Elder, H. Yu, W.
121. A. J. Metcalf, V. Torres-Company, D. E. Leaird, and A. M. Weiner, “High-
Bogaerts, L. R. Dalton, W. Freude, J. Leuthold, and C. Koos, “Silicon-
power broadly tunable electrooptic frequency comb generator,” IEEE J.
organic hybrid (SOH) frequency comb sources for terabit/s data
Sel. Top. Quantum Electron. 19, 231–236 (2013).
transmission,” Opt. Express 22, 3629–3637 (2014).
122. A. J. Metcalf, F. Quinlan, T. M. Fortier, S. A. Diddams, and A. M. Weiner,
100. L. Alloatti, R. Palmer, S. Diebold, K. P. Pahl, B. Chen, R. Dinu, M.
“Broadly tunable, low timing jitter, high repetition rate optoelectronic
Fournier, J.-M. Fedeli, T. Zwick, W. Freude, C. Koos, and J. Leuthold,
comb generator,” Electron. Lett. 51, 1596–1598 (2015).
“100 GHz silicon–organic hybrid modulator,” Light Sci. Appl. 3, e173 123. K. Beha, D. C. Cole, P. Del’Haye, A. Coillet, S. A. Diddams, and S. B.
(2014). Papp, “Electronic synthesis of light,” Optica 4, 406–411 (2017).
101. S. Jin, L. Xu, H. Zhang, and Y. Li, “LiNbO3 thin-film modulators using sil- 124. D. R. Carlson, D. D. Hickstein, W. Zhang, A. J. Metcalf, F. Quinlan, S. A.
icon nitride surface ridge waveguides,” IEEE Photonics Technol. Lett. Diddams, and S. B. Papp, “Ultrafast electro-optic light with subcycle
28, 736–739 (2016). control,” Science 361, 1358–1363 (2018).
102. F. Eltes, C. Mai, D. Caimi, M. Kroh, Y. Popoff, G. Winzer, D. Petousi, S. 125. J. Kim, D. J. Richardson, and R. Slavík, “Cavity-induced phase noise
Lischke, J. E. Ortmann, L. Czornomaz, L. Zimmermann, J. Fompeyrine, suppression in a Fabry–Perot modulator-based optical frequency
and S. Abel, “A BaTiO3 -based electro-optic Pockels modulator mono- comb,” Opt. Lett. 42, 1536–1539 (2017).
lithically integrated on an advanced silicon photonics platform,” J. 126. R. Soref and B. Bennett, “Electrooptical effects in silicon,” IEEE J.
Lightwave Technol. 37, 1456–1462 (2019). Quantum Electron. 23, 123–129 (1987).
103. A. J. Mercante, P. Yao, S. Shi, G. Schneider, J. Murakowski, and D. W. 127. V. R. Almeida, C. A. Barrios, R. R. Panepucci, and M. Lipson, “All-
Prather, “110 GHz CMOS compatible thin film LiNbO3 modulator on optical control of light on a silicon chip,” Nature 431, 1081–1084
silicon,” Opt. Express 24, 15590–15595 (2016). (2004).
104. P. Pintus, Z. Zhang, S. Pinna, M. A. Tran, A. Jain, M. Kennedy, L. 128. P. Dong, L. Chen, C. Xie, L. L. Buhl, and Y.-K. Chen, “50-Gb/s sil-
Ranzani, M. Soltani, and J. E. Bowers, “Characterization of hetero- icon quadrature phase-shift keying modulator,” Opt. Express 20,
geneous InP-on-Si optical modulators operating between 77 K and 21181–21186 (2012).
room temperature,” APL Photonics 4, 100805 (2019). 129. P. Dong, C. Xie, L. Chen, L. L. Buhl, and Y.-K. Chen, “112-Gb/s mono-
105. E. Wooten, K. Kissa, A. Yi-Yan, E. Murphy, D. Lafaw, P. Hallemeier, D. lithic PDM-QPSK modulator in silicon,” Opt. Express 20, B624–B629
Maack, D. Attanasio, D. Fritz, G. McBrien, and D. Bossi, “A review of (2012).
lithium niobate modulators for fiber-optic communications systems,” 130. Q. Li, R. Ding, Y. Liu, T. Baehr-Jones, M. Hochberg, and K. Bergman,
IEEE J. Sel. Top. Quantum Electron. 6, 69–82 (2000). “High-Speed BPSK modulation in silicon,” IEEE Photonics Technol.
106. M. Zhang, C. Wang, P. Kharel, D. Zhu, and M. Lončar, “Integrated Lett. 27, 1329–1332 (2015).
lithium niobate electro-optic modulators: when performance meets 131. H. Sepehrian, J. Lin, L. A. Rusch, and W. Shi, “Silicon photonic IQ
scalability,” Optica 8, 652–667 (2021). modulators for 400 Gb/s and beyond,” J. Lightwave Technol. 37,
107. C. Wang, M. Zhang, B. Stern, M. Lipson, and M. Lončar, 3078–3086 (2019).
“Nanophotonic lithium niobate electro-optic modulators,” Opt. 132. M. Jacques, Z. Xing, A. Samani, E. El-Fiky, X. Li, M. Xiang, S. Lessard,
Express 26, 1547–1555 (2018). and D. V. Plant, “240 Gbit/s silicon photonic Mach-Zehnder modulator
108. Y. Zhang, M. Xu, H. Zhang, M. Li, J. Jian, M. He, L. Chen, L. Wang, X. enabled by two 2.3-Vpp drivers,” J. Lightwave Technol. 38, 2877–2885
Cai, X. Xiao, and S. Yu, “220 Gbit/s optical PAM4 modulation based on (2020).
Review Vol. 10, No. 8 / August 2023 / Optica 993
133. Q. Xu, B. Schmidt, S. Pradhan, and M. Lipson, “Micrometre-scale sili- 153. B. P.-P. Kuo, E. Myslivets, V. Ataie, E. G. Temprana, N. Alic, and S.
con electro-optic modulator,” Nature 435, 325–327 (2005). Radic, “Wideband parametric frequency comb as coherent optical
134. B. G. Lee, B. A. Small, K. Bergman, Q. Xu, and M. Lipson, carrier,” J. Lightwave Technol. 31, 3414–3419 (2013).
“Transmission of high-data-rate optical signals through a 154. J. Chavez Boggio, S. Moro, N. Alic, M. Karlsson, J. Bland-Hawthorn,
micrometer-scale silicon ring resonator,” Opt. Lett. 31, 2701–2703 and S. Radic, “Nearly octave-spanning cascaded four-wave-mixing
(2006). generation in low dispersion highly nonlinear fiber,” in 35th European
135. X. Xiao, X. Li, H. Xu, Y. Hu, K. Xiong, Z. Li, T. Chu, J. Yu, and Y. Yu, “44- Conference on Optical Communication (2009), pp. 1–2.
Gb/s silicon microring modulators based on zigzag PN junctions,” IEEE 155. A. B. Matsko, A. A. Savchenkov, D. Strekalov, V. S. Ilchenko, and L.
Photonics Technol. Lett. 24, 1712–1714 (2012). Maleki, “Optical hyperparametric oscillations in a whispering-gallery-
136. I. Demirtzioglou, C. Lacava, K. R. H. Bottrill, D. J. Thomson, G. T. Reed, mode resonator: Threshold and phase diffusion,” Phys. Rev. A 71,
D. J. Richardson, and P. Petropoulos, “Frequency comb generation in a 033804 (2005).
silicon ring resonator modulator,” Opt. Express 26, 790–796 (2018). 156. P. Del’Haye, A. Schliesser, O. Arcizet, T. Wilken, R. Holzwarth, and T.
137. Z. Wang, M. Ma, H. Sun, M. Khalil, R. Adams, K. Yim, X. Jin, and L. R. J. Kippenberg, “Optical frequency comb generation from a monolithic
Chen, “Optical frequency comb generation using CMOS compatible microresonator,” Nature 450, 1214–1217 (2007).
cascaded Mach–Zehnder modulators,” IEEE J. Quantum Electron. 55, 157. J. S. Levy, A. Gondarenko, M. A. Foster, A. C. Turner-Foster, A. L.
8400206 (2019). Gaeta, and M. Lipson, “CMOS-compatible multiple-wavelength oscil-
138. S. Liu, K. Wu, L. Zhou, L. Lu, B. Zhang, G. Zhou, and J. Chen, “Optical lator for on-chip optical interconnects,” Nat. Photonics 4, 37–40
frequency comb and Nyquist pulse generation with integrated silicon (2010).
modulators,” IEEE J. Sel. Top. Quantum Electron. 26, 8300208 (2020). 158. P. Parra-Rivas, D. Gomila, E. Knobloch, S. Coen, and L. Gelens,
139. M. Burla, C. Hoessbacher, W. Heni, C. Haffner, Y. Fedoryshyn, D. “Origin and stability of dark pulse Kerr combs in normal dispersion
Werner, T. Watanabe, H. Massler, D. L. Elder, L. R. Dalton, and J. resonators,” Opt. Lett. 41, 2402–2405 (2016).
Leuthold, “500 GHz plasmonic Mach-Zehnder modulator enabling 159. C. Godey, I. V. Balakireva, A. Coillet, and Y. K. Chembo, “Stability
sub-THz microwave photonics,” APL Photonics 4, 056106 (2019). analysis of the spatiotemporal Lugiato-Lefever model for Kerr optical
140. C. Haffner, W. Heni, D. L. Elder, Y. Fedoryshyn, N. Ðordevi ¯ ć, D. frequency combs in the anomalous and normal dispersion regimes,”
Chelladurai, U. Koch, K. Portner, M. Burla, B. Robinson, L. R. Dalton, Phys. Rev. A 89, 063814 (2014).
and J. Leuthold, “Harnessing nonlinearities near material absorption 160. J. Gärtner, P. Trocha, R. Mandel, C. Koos, T. Jahnke, and W. Reichel,
resonances for reducing losses in plasmonic modulators,” Opt. Mater. “Bandwidth and conversion efficiency analysis of dissipative Kerr soli-
Express 7, 2168–2181 (2017). ton frequency combs based on bifurcation theory,” Phys. Rev. A 100,
141. B. Baeuerle, W. Heni, C. Hoessbacher, Y. Fedoryshyn, U. Koch, A. 033819 (2019).
Josten, T. Watanabe, C. Uhl, H. Hettrich, D. L. Elder, L. R. Dalton,
161. P. Parra-Rivas, E. Knobloch, L. Gelens, and D. Gomila, “Origin, bifurca-
M. Möller, and J. Leuthold, “120 GBd plasmonic Mach-Zehnder
tion structure and stability of localized states in Kerr dispersive optical
modulator with a novel differential electrode design operated at a
cavities,” IMA J. Appl. Math. 86, 856–895 (2021).
peak-to-peak drive voltage of 178 mV,” Opt. Express 27, 16823–16832
162. L. A. Lugiato and R. Lefever, “Spatial dissipative structures in passive
(2019).
optical system,” Phys. Rev. Lett. 58, 2209–2211 (1987).
142. C. Hoessbacher, A. Josten, B. Baeuerle, Y. Fedoryshyn, H. Hettrich,
163. D. Hillerkuss, R. Schmogrow, M. Meyer, et al., “Single-laser 32.5 Tbit/s
Y. Salamin, W. Heni, C. Haffner, C. Kaiser, R. Schmid, D. L. Elder, D.
Nyquist WDM transmission,” J. Opt. Commun. Netw. 4, 715–723
Hillerkuss, M. Möller, L. R. Dalton, and J. Leuthold, “Plasmonic modu-
(2012).
lator with >170 GHz bandwidth demonstrated at 100 GBd NRZ,” Opt.
164. J. S. Levy, K. Saha, Y. Okawachi, M. A. Foster, A. L. Gaeta, and M.
Express 25, 1762–1768 (2017).
Lipson, “High-performance silicon-nitride-based multiple-wavelength
143. H. G. Bach, J. Krauser, H. P. Nolting, R. A. Logan, and F. K.
source,” IEEE Photonics Technol. Lett. 24, 1375–1377 (2012).
Reinhart, “Electro-optical light modulation in InGaAsP/InP double
165. P.-H. Wang, F. Ferdous, H. Miao, J. Wang, D. E. Leaird, K. Srinivasan,
heterostructure diodes,” Appl. Phys. Lett. 42, 692–694 (1983).
L. Chen, V. Aksyuk, and A. M. Weiner, “Observation of correlation
144. G. Sinatkas, T. Christopoulos, O. Tsilipakos, and E. E. Kriezis, “Electro-
between route to formation, coherence, noise, and communication
optic modulation in integrated photonics,” J. Appl. Phys. 130, 010901
performance of Kerr combs,” Opt. Express 20, 29284–29295 (2012).
(2021).
145. N. Yokota and H. Yasaka, “Operation strategy of InP Mach–Zehnder 166. M. Karpov, M. H. P. Pfeiffer, H. Guo, W. Weng, J. Liu, and T. J.
modulators for flat optical frequency comb generation,” IEEE J. Kippenberg, “Dynamics of soliton crystals in optical microresonators,”
Quantum Electron. 52, 5200207 (2016). Nat. Phys. 15, 1071–1077 (2019).
146. R. Slavík, S. G. Farwell, M. J. Wale, and D. J. Richardson, “Compact 167. Y. Okawachi, M. R. E. Lamont, K. Luke, D. O. Carvalho, M. Yu, M.
optical comb generator using InP tunable laser and push-pull Lipson, and A. L. Gaeta, “Bandwidth shaping of microresonator-
modulator,” IEEE Photonics Technol. Lett. 27, 217–220 (2015). based frequency combs via dispersion engineering,” Opt. Lett. 39,
147. N. Andriolli, T. Cassese, M. Chiesa, C. de Dios, and G. Contestabile, 3535–3538 (2014).
“Photonic integrated fully tunable comb generator cascading optical 168. M. H. P. Pfeiffer, C. Herkommer, J. Liu, H. Guo, M. Karpov, E. Lucas,
modulators,” J. Lightwave Technol. 36, 5685–5689 (2018). M. Zervas, and T. J. Kippenberg, “Octave-spanning dissipative Kerr
148. M. Churaev, A. Riedhauser, R. N. Wang, C. Möhl, T. Blésin, M. soliton frequency combs in Si3 N4 microresonators,” Optica 4, 684–691
A. Anderson, V. Snigirev, A. Siddharth, Y. Popoff, D. Caimi, S. (2017).
Hönl, J. Riemensberger, J. Liu, P. Seidler, and T. J. Kippenberg, 169. V. Brasch, M. Geiselmann, M. H. P. Pfeiffer, and T. J. Kippenberg,
“Heterogeneously integrated lithium niobate photonics,” arXiv, “Bringing short-lived dissipative Kerr soliton states in microresonators
arXiv:2112.02018 (2021). into a steady state,” Opt. Express 24, 29312–29320 (2016).
149. M. He, M. Xu, Y. Ren, J. Jian, Z. Ruan, Y. Xu, S. Gao, S. Sun, X. Wen, 170. Z. Gong, A. Bruch, M. Shen, X. Guo, H. Jung, L. Fan, X. Liu, L. Zhang, J.
L. Zhou, L. Liu, C. Guo, H. Chen, S. Yu, L. Liu, and X. Cai, “High- Wang, J. Li, J. Yan, and H. X. Tang, “High-fidelity cavity soliton genera-
performance hybrid silicon and lithium niobate Mach–Zehnder tion in crystalline AlN micro-ring resonators,” Opt. Lett. 43, 4366–4369
modulators for 100 Gbit s−1 and beyond,” Nat. Photonics 13, 359–364 (2018).
(2019). 171. C. Joshi, J. K. Jang, K. Luke, X. Ji, S. A. Miller, A. Klenner, Y. Okawachi,
150. G. Sefler and K.-I. Kitayama, “Frequency comb generation by four- M. Lipson, and A. L. Gaeta, “Thermally controlled comb generation
wave mixing and the role of fiber dispersion,” J. Lightwave Technol. 16, and soliton modelocking in microresonators,” Opt. Lett. 41, 2565–2568
1596–1605 (1998). (2016).
151. A. C. S. Jr, J. M. C. Boggio, A. A. Rieznik, H. E. Hernandez-Figueroa, 172. H. Guo, M. Karpov, E. Lucas, A. Kordts, M. H. P. Pfeiffer, V. Brasch,
H. L. Fragnito, and J. C. Knight, “Highly efficient generation of broad- G. Lihachev, V. E. Lobanov, M. L. Gorodetsky, and T. J. Kippenberg,
band cascaded four-wave mixing products,” Opt. Express 16, “Universal dynamics and deterministic switching of dissipative Kerr
2816–2828 (2008). solitons in optical microresonators,” Nat. Phys. 13, 94–102 (2017).
152. E. Myslivets, B. P. P. Kuo, N. Alic, and S. Radic, “Generation of wide- 173. E. Obrzud, S. Lecomte, and T. Herr, “Temporal solitons in microres-
band frequency combs by continuous-wave seeding of multistage mix- onators driven by optical pulses,” Nat. Photonics 11, 600–607
ers with synthesized dispersion,” Opt. Express 20, 3331–3344 (2012). (2017).
Review Vol. 10, No. 8 / August 2023 / Optica 994
174. Z. Kang, F. Li, J. Yuan, K. Nakkeeran, J. N. Kutz, Q. Wu, C. Yu, and P. K. 193. Ó. B. Helgason, M. Girardi, Z. Ye, F. Lei, J. Schröder, and V. T. Company,
A. Wai, “Deterministic generation of single soliton Kerr frequency comb “Power-efficient soliton microcombs,” arXiv, arXiv:2202.09410 (2022).
in microresonators by a single shot pulsed trigger,” Opt. Express 26, 194. K. Nishimoto, K. Minoshima, K. Minoshima, T. Yasui, T. Yasui, N. Kuse,
18563–18577 (2018). and N. Kuse, “Investigation of the phase noise of a microresonator soli-
175. D. C. Cole, J. R. Stone, M. Erkintalo, K. Y. Yang, X. Yi, K. J. Vahala, and ton comb,” Opt. Express 28, 19295–19303 (2020).
S. B. Papp, “Kerr-microresonator solitons from a chirped background,” 195. N. L. Thomas, A. Dhakal, A. Raza, F. Peyskens, and R. Baets, “Impact of
Optica 5, 1304–1310 (2018). fundamental thermodynamic fluctuations on light propagating in pho-
176. X. Yi, Q.-F. Yang, K. Y. Yang, M.-G. Suh, and K. Vahala, “Soliton fre- tonic waveguides made of amorphous materials,” Optica 5, 328–336
quency comb at microwave rates in a high-Q silica microresonator,” (2018).
Optica 2, 1078–1085 (2015). 196. G. Huang, E. Lucas, J. Liu, A. S. Raja, G. Lihachev, M. L. Gorodetsky,
177. X. Yi, Q.-F. Yang, K. Y. Yang, and K. Vahala, “Active capture and stabi- N. J. Engelsen, and T. J. Kippenberg, “Thermorefractive noise in
lization of temporal solitons in microresonators,” Opt. Lett. 41, 2037– silicon-nitride microresonators,” Phys. Rev. A 99, 061801 (2019).
2040 (2016). 197. X. Yi, Q.-F. Yang, X. Zhang, K. Y. Yang, X. Li, and K. Vahala, “Single-
178. C. Bao, Y. Xuan, D. E. Leaird, S. Wabnitz, M. Qi, and A. M. Weiner, mode dispersive waves and soliton microcomb dynamics,” Nat.
“Spatial mode-interaction induced single soliton generation in Commun. 8, 14869 (2017).
microresonators,” Optica 4, 1011–1015 (2017). 198. Q.-F. Yang, Q.-X. Ji, L. Wu, B. Shen, H. Wang, C. Bao, Z. Yuan, and
179. Y. He, Q.-F. Yang, J. Ling, R. Luo, H. Liang, M. Li, B. Shen, H. Wang, K. Vahala, “Dispersive-wave induced noise limits in miniature soliton
K. Vahala, and Q. Lin, “Self-starting bi-chromatic LiNbO3 soliton microwave sources,” Nat. Commun. 12, 1442 (2021).
microcomb,” Optica 6, 1138–1144 (2019). 199. T. Tetsumoto, J. Jiang, M. E. Fermann, G. Navickaite, M. Geiselmann,
180. Z. Lin, Z. Kang, Z. Kang, Z. Kang, P. Xu, Y. Tian, S. He, S. He, and S. and A. Rolland, “Effects of a quiet point on a kerr microresonator fre-
He, “Turnkey generation of Kerr soliton microcombs on thin-film lithium quency comb,” OSA Contin. 4, 1348–1357 (2021).
niobate on insulator microresonators powered by the photorefractive 200. T. E. Drake, J. R. Stone, T. C. Briles, and S. B. Papp, “Thermal decoher-
effect,” Opt. Express 29, 42932–42944 (2021). ence and laser cooling of Kerr microresonator solitons,” Nat. Photonics
181. H. Zhou, Y. Geng, W. Cui, S.-W. Huang, Q. Zhou, K. Qiu, and C. Wei 14, 480–485 (2020).
Wong, “Soliton bursts and deterministic dissipative Kerr soliton gen- 201. F. Lei, Z. Ye, and V. Torres-Company, “Thermal noise reduction in soli-
eration in auxiliary-assisted microcavities,” Light Sci. Appl. 8, 50 ton microcombs via laser self-cooling,” Opt. Lett. 47, 513–516 (2022).
(2019). 202. K. Nishimoto, K. Minoshima, T. Yasui, and N. Kuse, “Thermal control of
182. S. Fujii, S. Tanaka, T. Ohtsuka, S. Kogure, K. Wada, H. Kumazaki, S. Kerr microresonator soliton comb via an optical sideband,” Opt. Lett.
Tasaka, Y. Hashimoto, Y. Kobayashi, T. Araki, K. Furusawa, N. Sekine,
47, 281–284 (2022).
S. Kawanishi, and T. Tanabe, “Dissipative Kerr soliton microcombs for
203. P. Liao, C. Bao, A. Almaiman, A. Kordts, M. Karpov, M. H. P. Pfeiffer, L.
FEC-free optical communications over 100 channels,” Opt. Express
Zhang, F. Alishahi, Y. Cao, K. Zou, A. Fallahpour, A. N. Willner, M. Tur,
30, 1351–1364 (2021).
T. J. Kippenberg, and A. E. Willner, “Demonstration of multiple Kerr-
183. J. Pfeifle, V. Brasch, M. Lauermann, Y. Yu, D. Wegner, T. Herr, K.
frequency-comb generation using different lines from another Kerr
Hartinger, P. Schindler, J. Li, D. Hillerkuss, R. Schmogrow, C. Weimann,
comb located up to 50 km away,” J. Lightwave Technol. 37, 579–584
R. Holzwarth, W. Freude, J. Leuthold, T. J. Kippenberg, and C. Koos,
(2019).
“Coherent terabit communications with microresonator Kerr frequency
204. C. Bao, P. Liao, A. Kordts, M. Karpov, M. H. P. Pfeiffer, L. Zhang, Y. Yan,
combs,” Nat. Photonics 8, 375–380 (2014).
G. Xie, Y. Cao, A. Almaiman, M. Ziyadi, L. Li, Z. Zhao, A. Mohajerin-
184. P. Marin-Palomo, J. N. Kemal, M. Karpov, A. Kordts, J. Pfeifle,
Ariaei, S. R. Wilkinson, M. Tur, M. M. Fejer, T. J. Kippenberg, and A. E.
M. H. P. Pfeiffer, P. Trocha, S. Wolf, V. Brasch, M. H. Anderson, R.
Willner, “Demonstration of optical multicasting using Kerr frequency
Rosenberger, K. Vijayan, W. Freude, T. J. Kippenberg, and C. Koos,
comb lines,” Opt. Lett. 41, 3876–3879 (2016).
“Microresonator-based solitons for massively parallel coherent optical
205. P. Liao, C. Bao, A. Kordts, M. Karpov, M. H. P. Pfeiffer, L. Zhang, Y. Cao,
communications,” Nature 546, 274–279 (2017).
A. Almaiman, A. Mohajerin-Ariaei, M. Tur, M. M. Fejer, T. J. Kippenberg,
185. B. Corcoran, M. Tan, X. Xu, A. Boes, J. Wu, T. G. Nguyen, S. T. Chu, B.
and A. E. Willner, “Pump-linewidth-tolerant wavelength multicasting
E. Little, R. Morandotti, A. Mitchell, and D. J. Moss, “Ultra-dense opti-
using soliton Kerr frequency combs,” Opt. Lett. 42, 3177–3180 (2017).
cal data transmission over standard fibre with a single chip source,”
Nat. Commun. 11, 2568 (2020). 206. V. E. Lobanov, G. Lihachev, T. J. Kippenberg, and M. L. Gorodetsky,
186. C. Bao, P. Liao, A. Kordts, M. Karpov, M. H. P. Pfeiffer, L. Zhang, Y. Cao, “Frequency combs and platicons in optical microresonators with
G. Xie, C. Liu, Y. Yan, A. Almaiman, A. Mohajerin-Ariaei, A. Fallahpour, normal GVD,” Opt. Express 23, 7713–7721 (2015).
M. Tur, T. J. Kippenberg, and A. E. Willner, “Tunable insertion of multi- 207. X. Xue, P.-H. Wang, Y. Xuan, M. Qi, and A. M. Weiner, “Microresonator
ple lines into a Kerr frequency comb using electro-optical modulators,” Kerr frequency combs with high conversion efficiency,” Laser
Opt. Lett. 42, 3765–3768 (2017). Photonics Rev. 11, 1600276 (2017).
187. Y. Geng, X. Huang, W. Cui, Y. Ling, B. Xu, J. Zhang, X. Yi, B. Wu, S.-W. 208. A. A. Savchenkov, A. B. Matsko, W. Liang, V. S. Ilchenko, D. Seidel, and
Huang, K. Qiu, C. W. Wong, and H. Zhou, “Terabit optical OFDM super- L. Maleki, “Kerr frequency comb generation in overmoded resonators,”
channel transmission via coherent carriers of a hybrid chip-scale soli- Opt. Express 20, 27290–27298 (2012).
ton frequency comb,” Opt. Lett. 43, 2406–2409 (2018). 209. W. Liang, A. A. Savchenkov, V. S. Ilchenko, D. Eliyahu, D. Seidel, A. B.
188. M. Mazur, M.-G. Suh, A. Fülöp, J. Schröder, V. Torres-Company, M. Matsko, and L. Maleki, “Generation of a coherent near-infrared Kerr
Karlsson, K. Vahala, and P. Andrekson, “High spectral efficiency coher- frequency comb in a monolithic microresonator with normal GVD,”
ent superchannel transmission with soliton microcombs,” J. Lightwave Opt. Lett. 39, 2920–2923 (2014).
Technol. 39, 4367–4373 (2021). 210. X. Xue, Y. Xuan, Y. Liu, P.-H. Wang, S. Chen, J. Wang, D. E. Leaird, M.
189. C. Bao, L. Zhang, A. Matsko, Y. Yan, Z. Zhao, G. Xie, A. M. Agarwal, L. Qi, and A. M. Weiner, “Mode-locked dark pulse Kerr combs in normal-
C. Kimerling, J. Michel, L. Maleki, and A. E. Willner, “Nonlinear con- dispersion microresonators,” Nat. Photonics 9, 594–600 (2015).
version efficiency in Kerr frequency comb generation,” Opt. Lett. 39, 211. Y. Xu, A. Sharples, J. Fatome, S. Coen, M. Erkintalo, and S. G.
6126–6129 (2014). Murdoch, “Frequency comb generation in a pulse-pumped normal
190. X. Yi, Q.-F. Yang, K. Y. Yang, and K. Vahala, “Theory and measure- dispersion Kerr mini-resonator,” Opt. Lett. 46, 512–515 (2021).
ment of the soliton self-frequency shift and efficiency in optical 212. B. Y. Kim, J. K. Jang, Y. Okawachi, X. Ji, M. Lipson, and A. L. Gaeta,
microcavities,” Opt. Lett. 41, 3419–3422 (2016). “Synchronization of nonsolitonic Kerr combs,” Sci. Adv. 7, eabi4362
191. J. K. Jang, Y. Okawachi, Y. Zhao, X. Ji, C. Joshi, C. Joshi, M. Lipson, M. (2021).
Lipson, A. L. Gaeta, and A. L. Gaeta, “Conversion efficiency of soliton 213. B. Y. Kim, Y. Okawachi, J. K. Jang, M. Yu, X. Ji, Y. Zhao, C. Joshi, M.
Kerr combs,” Opt. Lett. 46, 3657–3660 (2021). Lipson, and A. L. Gaeta, “Turn-key, high-efficiency Kerr comb source,”
192. J. M. C. Boggio, D. Bodenmüller, S. Ahmed, S. Wabnitz, D. Modotto, Opt. Lett. 44, 4475–4478 (2019).
and T. Hansson, “Efficient Kerr soliton comb generation in micro- 214. V. E. Lobanov, G. Lihachev, and M. L. Gorodetsky, “Generation of plati-
resonator with interferometric back-coupling,” Nat. Commun. 13, 1292 cons and frequency combs in optical microresonators with normal GVD
(2022). by modulated pump,” Europhys. Lett. 112, 54008 (2015).
Review Vol. 10, No. 8 / August 2023 / Optica 995
215. W. Jin, Q.-F. Yang, L. Chang, B. Shen, H. Wang, M. A. Leal, L. Wu, M. 231. W. Liang, D. Eliyahu, V. S. Ilchenko, A. A. Savchenkov, A. B. Matsko, D.
Gao, A. Feshali, M. Paniccia, K. J. Vahala, and J. E. Bowers, “Hertz- Seidel, and L. Maleki, “High spectral purity Kerr frequency comb radio
linewidth semiconductor lasers using CMOS-ready ultra-high-Q frequency photonic oscillator,” Nat. Commun. 6, 7957 (2015).
microresonators,” Nat. Photonics 15, 346–353 (2021). 232. B. Shen, L. Chang, J. Liu, H. Wang, Q.-F. Yang, C. Xiang, R. N.
216. Y. Liu, Y. Xuan, X. Xue, P.-H. Wang, S. Chen, A. J. Metcalf, J. Wang, D. Wang, J. He, T. Liu, W. Xie, J. Guo, D. Kinghorn, L. Wu, Q.-X. Ji, T. J.
E. Leaird, M. Qi, and A. M. Weiner, “Investigation of mode coupling in Kippenberg, K. J. Vahala, and J. E. Bowers, “Integrated turnkey soliton
normal-dispersion silicon nitride microresonators for Kerr frequency microcombs,” Nature 582, 365–369 (2020).
comb generation,” Optica 1, 137–144 (2014). 233. A. S. Voloshin, N. M. Kondratiev, G. V. Lihachev, J. Liu, V. E. Lobanov,
217. X. Xue, Y. Xuan, P.-H. Wang, Y. Liu, D. E. Leaird, M. Qi, and A. M. N. Y. Dmitriev, W. Weng, T. J. Kippenberg, and I. A. Bilenko, “Dynamics
Weiner, “Normal-dispersion microcombs enabled by controllable of soliton self-injection locking in optical microresonators,” Nat.
mode interactions,” Laser Photonics Rev. 9, L23–L28 (2015). Commun. 12, 235 (2021).
218. J. K. Jang, Y. Okawachi, M. Yu, K. Luke, X. Ji, M. Lipson, and A. L. 234. S. Liu, J. Norman, M. Dumont, et al., “High-performance O-band
Gaeta, “Dynamics of mode-coupling-induced microresonator fre- quantum-dot semiconductor optical amplifiers directly grown on a
quency combs in normal dispersion,” Opt. Express 24, 28794–28803 CMOS compatible silicon substrate,” ACS Photonics 6, 2523–2529
(2016). (2019).
219. A. Fülöp, M. Mazur, A. Lorences-Riesgo, Ó. B. Helgason, P.-H. Wang, Y. 235. Y. Wan, J. Norman, S. Liu, A. Liu, and J. E. Bowers, “Quantum
Xuan, D. E. Leaird, M. Qi, P. A. Andrekson, A. M. Weiner, and V. Torres- dot lasers and amplifiers on silicon: Recent advances and future
Company, “High-order coherent communications using mode-locked developments,” IEEE Nanotechnol. Mag. 15(2), 8–22 (2021).
dark-pulse Kerr combs from microresonators,” Nat. Commun. 9, 1598 236. G. Roelkens, L. Liu, D. Liang, R. Jones, A. Fang, B. Koch, and J.
(2018). Bowers, “III-V/silicon photonics for on-chip and intra-chip optical
220. S. Ramelow, A. Farsi, S. Clemmen, J. S. Levy, A. R. Johnson, Y. interconnects,” Laser Photonics Rev. 4, 751–779 (2010).
Okawachi, M. R. E. Lamont, M. Lipson, and A. L. Gaeta, “Strong polari- 237. S. Cheung, Y. Kawakita, K. Shang, and S. J. B. Yoo, “Highly efficient
zation mode coupling in microresonators,” Opt. Lett. 39, 5134–5137 chip-scale III-V/silicon hybrid optical amplifiers,” Opt. Express 23,
(2014). 22431–22443 (2015).
221. T. Carmon, H. G. L. Schwefel, L. Yang, M. Oxborrow, A. D. Stone, and 238. K. V. Gasse, R. Wang, and G. Roelkens, “27 db gain III-V-on-silicon
K. J. Vahala, “Static envelope patterns in composite resonances gener- semiconductor optical amplifier with >17 dbm output power,” Opt.
ated by level crossing in optical toroidal microcavities,” Phys. Rev. Lett. Express 27, 293–302 (2019).
100, 103905 (2008). 239. G. C. Righini, M. Brenci, G. N. Conti, S. Pelli, M. Ferrari, M. Bettinelli,
222. S. A. Miller, Y. Okawachi, S. Ramelow, K. Luke, A. Dutt, A. Farsi, A. L. A. Speghini, and B. Chen, “Integrated optical amplifiers based on rare-
Gaeta, and M. Lipson, “Tunable frequency combs based on dual micro- earth-doped oxide glasses,” Proc. SPIE 5061, 34–40 (2003).
ring resonators,” Opt. Express 23, 21527–21540 (2015). 240. H. C. Frankis, H. M. Mbonde, D. B. Bonneville, C. Zhang, R. Mateman,
223. T. Herr, V. Brasch, J. D. Jost, I. Mirgorodskiy, G. Lihachev, M. L. A. Leinse, and J. D. B. Bradley, “Erbium-doped TeO2 -coated Si3 N4
Gorodetsky, and T. J. Kippenberg, “Mode spectrum and temporal waveguide amplifiers with 5 dB net gain,” Photonics Res. 8, 127–134
soliton formation in optical microresonators,” Phys. Rev. Lett. 113, (2020).
123901 (2014). 241. W. A. P. M. Hendriks, L. Chang, C. I. van Emmerik, J. Mu, M. de Goede,
224. S.-P. Yu, E. Lucas, J. Zang, and S. B. Papp, “A continuum of bright and M. Dijkstra, and S. M. Garcia-Blanco, “Rare-earth ion doped Al2 O3 for
dark-pulse states in a photonic-crystal resonator,” Nat. Commun. 13, active integrated photonics,” Adv. Phys. X 6, 1833753 (2021).
3134 (2022). 242. Z. Chen, Q. Xu, K. Zhang, W.-H. Wong, D.-L. Zhang, E. Y.-B. Pun, and
225. A. Rizzo, A. Novick, V. Gopal, B. Y. Kim, X. Ji, S. Daudlin, Y. Okawachi, C. Wang, “Efficient erbium-doped thin-film lithium niobate waveguide
Q. Cheng, M. Lipson, A. L. Gaeta, and K. Bergman, “Integrated amplifiers,” Opt. Lett. 46, 1161–1164 (2021).
Kerr frequency comb-driven silicon photonic transmitter,” arXiv, 243. X. Yan, Y. Liu, J. Wu, Y. Chen, and X. Chen, “Integrated spiral wave-
arXiv:2109.10297 (2021). guide amplifiers on erbium-doped thin-film lithium niobate,” arXiv,
226. J. K. Jang, A. Klenner, X. Ji, Y. Okawachi, M. Lipson, and A. L. Gaeta, arXiv:2105.00214 (2021).
“Synchronization of coupled optical microresonators,” Nat. Photonics 244. Y. Liu, Z. Qiu, X. Ji, A. Lukashchuk, J. He, J. Riemensberger, M.
12, 688–693 (2018). Hafermann, R. N. Wang, J. Liu, C. Ronning, and T. J. Kippenberg,
227. J. K. Jang, X. Ji, C. Joshi, Y. Okawachi, M. Lipson, and A. L. Gaeta, “A photonic integrated circuit-based erbium-doped amplifier,” Science
“Observation of Arnold tongues in coupled soliton Kerr frequency 376, 1309–1313 (2022).
combs,” Phys. Rev. Lett. 123, 153901 (2019). 245. Z. Ye, P. Zhao, K. Twayana, M. Karlsson, V. Torres-Company, and P. A.
228. B. Stern, X. Ji, Y. Okawachi, A. L. Gaeta, and M. Lipson, “Battery- Andrekson, “Overcoming the quantum limit of optical amplification in
operated integrated frequency comb generator,” Nature 562, 401–405 monolithic waveguides,” Sci. Adv. 7, eabi8150 (2021).
(2018). 246. J. Riemensberger, J. Liu, N. Kuznetsov, J. He, R. N. Wang, and T.
229. N. G. Pavlov, S. Koptyaev, G. V. Lihachev, A. S. Voloshin, A. S. J. Kippenberg, “Photonic chip-based continuous-travelling-wave
Gorodnitskiy, M. V. Ryabko, S. V. Polonsky, and M. L. Gorodetsky, parametric amplifier,” arXiv, arXiv:2107.09582 (2021).
“Narrow-linewidth lasing and soliton Kerr microcombs with ordinary 247. Y. Zhao, J. K. Jang, X. Ji, Y. Okawachi, M. Lipson, and A. L. Gaeta,
laser diodes,” Nat. Photonics 12, 694–698 (2018). “Large microresonator-assisted optical-parametric amplification at
230. A. S. Raja, A. S. Voloshin, H. Guo, S. E. Agafonova, J. Liu, A. S. ultra-low pump powers,” arXiv, arXiv:2209.05589 (2022).
Gorodnitskiy, M. Karpov, N. G. Pavlov, E. Lucas, R. R. Galiev, A.
E. Shitikov, J. D. Jost, M. L. Gorodetsky, and T. J. Kippenberg,
“Electrically pumped photonic integrated soliton microcomb,” Nat.
Commun. 10, 680 (2019).