Content uploaded by Cosimo Lacava
Author content
All content in this area was uploaded by Cosimo Lacava on Jan 31, 2019
Content may be subject to copyright.
Coupling strategies for silicon photonics
integrated chips [Invited]
RICCARDO MARCHETTI,1COSIMO LACAVA,2,*LEE CARROLL,3KAMIL GRADKOWSKI,3AND PAOLO MINZIONI1
1Electrical, Computer and Biomedical Engineering Department, University of Pavia, Pavia 27100, Italy
2Optoelectronics Research Centre, Highfield Campus, University of Southampton, Southampton SO17 1BJ, UK
3Photonic Packaging Group, Tyndall National Institute, Lee Maltings Complex, Cork T12R5CP, Ireland
*Corresponding author: C.Lacava@soton.ac.uk
Received 6 July 2018; revised 4 December 2018; accepted 6 December 2018; posted 13 December 2018 (Doc. ID 338163);
published 31 January 2019
Over the last 20 years, silicon photonics has revolutionized the field of integrated optics, providing a novel and
powerful platform to build mass-producible optical circuits. One of the most attractive aspects of silicon pho-
tonics is its ability to provide extremely small optical components, whose typical dimensions are an order of
magnitude smaller than those of optical fiber devices. This dimension difference makes the design of fiber-
to-chip interfaces challenging and, over the years, has stimulated considerable technical and research efforts
in the field. Fiber-to-silicon photonic chip interfaces can be broadly divided into two principle categories:
in-plane and out-of-plane couplers. Devices falling into the first category typically offer relatively high coupling
efficiency, broad coupling bandwidth (in wavelength), and low polarization dependence but require relatively
complex fabrication and assembly procedures that are not directly compatible with wafer-scale testing.
Conversely, out-of-plane coupling devices offer lower efficiency, narrower bandwidth, and are usually polariza-
tion dependent. However, they are often more compatible with high-volume fabrication and packaging processes
and allow for on-wafer access to any part of the optical circuit. In this paper, we review the current state-of-the-art
of optical couplers for photonic integrated circuits, aiming to give to the reader a comprehensive and broad view
of the field, identifying advantages and disadvantages of each solution. As fiber-to-chip couplers are inherently
related to packaging technologies and the co-design of optical packages has become essential, we also review the
main solutions currently used to package and assemble optical fibers with silicon-photonic integrated circuits.
Published by Chinese Laser Press under the terms of the Creative Commons Attribution 4.0 License. Further distribution of this work
must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI.
https://doi.org/10.1364/PRJ.7.000201
1. INTRODUCTION: SILICON PHOTONICS
OVERVIEW AND OPTICAL-COUPLING
“OPPORTUNITIES TREE”
Modern telecommunications require significant technological
advancements to cope with the tremendous growth of data ex-
changed over networks, which is mainly driven by mobile ap-
plications, video streaming, and cloud services. Optical
technologies have already revolutionized the communications
field, allowing for modern high-bandwidth transoceanic trans-
mission through optical fibers. Over the last decade, silicon
photonics has established itself as a platform for the realization
of optical transceivers and optical processors, aiming to provide
low-cost and high-performance components for telecom and
datacom applications [1]. Using silicon (Si) waveguides as a
basic element [2–4], a variety of optical components can be
implemented, such as directional couplers [5], Y-branches
[6], distributed waveguide Bragg gratings [7], and arrayed
waveguide gratings (AWGs) [8]. Interferometric structures such
as Mach–Zehnder interferometers [9] and ring resonators
[10,11] have been demonstrated, and a variety of high-speed
optical modulators have also been implemented [12,13].
Although silicon photonics can now be considered as a ma-
ture technological platform, its compatibility with optical fiber
components is still relatively limited, mainly due to the large
size mismatch between the optical fibers and silicon photonic
waveguide modal distributions. Because of this, coupling light
to and from silicon photonic components with large efficiencies
is still a relevant challenge. To overcome this issue, two main
solutions are usually adopted.
•Edge (also indicated as “in-plane,”“end-fire,”or “butt”)
coupling
In this case, the light beam is coupled in/out from the wave-
guide from lateral sides, thus always propagating in the same
plane. This technique usually requires the realization of
Review Vol. 7, No. 2 / February 2019 / Photonics Research 201
2327-9125/19/020201-39 Journal © 2019 Chinese Laser Press
optical-quality facets on the chip sides, in order to allow for
high coupling efficiencies (typically greater than 80%), with
negligible polarization dependence.
•Vertical coupling
When this technique is adopted, the light beam is incident
from the top surface of the silicon chip (or the bottom one if
required), and a suitably designed coupling structure modifies
the k-vector direction of the incident light beam, allowing
coupling of light into the integrated waveguide. The most
widely adopted vertical coupling solution is based on diffractive
gratings; it is characterized by relatively relaxed positioning tol-
erances and ease of lithographic fabrication and allows multi-
point wafer characterizations. On the other hand, the simplest
forms of these structures are intrinsically polarization and
wavelength sensitive, and careful design and optimization are
required to reduce the impact of these limitations.
In this paper, we review both the aforementioned ap-
proaches, showing the advantages and disadvantages of solu-
tions proposed by different research groups, giving a short
introduction to explain the physical working principle of the
analyzed structures. We note that coupling techniques and
technology have a strong impact on chip-packaging solutions;
therefore, we also review the most relevant packaging tech-
niques and trends.
The paper is organized as follows (see Fig. 1). Section 2gives
an overview of the different fiber types commonly used as the
interface of photonic integrated circuits. We then focus atten-
tion on edge-coupling strategies (Section 3), considering in-
verted tapers (Sections 3.A and 3.B), metamaterials-based
structures (Section 3.C), and vertical coupling to bent wave-
guides (Section 3.D). Section 4reports on the principle, design,
and optimization of grating coupler (GC) structures for single-
polarization beams, considering different materials, fabrication
techniques, approaches, and specific applications. We then dis-
cuss polarization-insensitive GCs in Section 5, where we analyze
1D and 2D photonic structures (Sections 5.A and 5.B,respec-
tively). Finally, Section 6is devoted to packaging techniques and
reports on the opportunities and limitations connected to the
use of fiber arrays, microlenses, and photonic-wire bonds. As a
conclusion, in Section 7, we offer a comprehensive table, sum-
marizing in a compact form the main data discussed in the re-
view, which can also be used by the reader as a reference while
reading. A schematic representation of the conceptual organiza-
tion of the topics discussed in the text is given in Fig. 1.
The basic structure of a silicon-on-insulator (SOI) wafer in-
cludes the presence of a thick Si substrate, a bottom oxide layer
(BOX), a thin Si layer, and eventually of a top oxide layer (TOX).
In the following, we will indicate the thickness of the BOX,
Si thin layer, and TOX with B,S,andT, respectively.
2. STANDARD AND COUPLING-DEDICATED
OPTICAL FIBERS
The interface between a silicon photonics circuit and any optical
fiber component typically includes a section of fiber that is either
a standard straight-polished fiber or a specialized device that, on
some occasions, makes it possible to improve the overall coupler
efficiency. In this section, we briefly review the fiber solutions
currently being considered by researchers and market specialists.
The standard fiber for telecom and datacom photonic
applications is single-mode, generally indicated as single-mode
fiber (SMF) 28 [shown in Fig. 2(a)], which shows an attenu-
ation coefficient lower than 0.18 and 0.32 dB/km at
Fig. 1. Conceptual organization of the different structures proposed for optical coupling and discussed in the present text.
202 Vol. 7, No. 2 / February 2019 / Photonics Research Review
wavelengths of 1.55 and 1.31 μm, respectively, and correspond-
ing mode field diameters (MFDs) of 10.4 and 9.2 μm[14]. The
SMF28 fiber, whose structure is schematically illustrated in
Fig. 2(a), consists of a 125 μm cladding layer (nclad 1.45
at λ1.55 μm), surrounding an 8.2 μm core, which has a
refractive index (ncore) that is just 0.3% higher than nclad .
Polarization-maintaining fiber variants of SMF28 are available
(typically referred to as P-SMF or PMF28), and they use in-
ternal stressor rods, in the fiber-cladding region, to induce bi-
refringence so that the mode degeneracy due to the cylindrical
symmetry of the fiber core is broken. This slight asymmetry is
sufficient to suppress the random polarization hopping, which
occurs in standard SMF28 because of external mechanical
vibrations or temperature shifts [15]. In general, PMF28 tends
to be used in high-value, low-volume applications only because
its cost-per-length is one order of magnitude higher than that of
standard SMF28, and it also exhibits higher propagation losses
than SMF28.
Visible-light analogues of SMF28 are also commercially
available, with reduced MFDs of the order of 4–7μm, which
scales with the typical target wavelength range of 0.60–0.85 μm
[16]. There is no single dominant standard for single-mode vis-
ible-light fibers because they tend to be used in highly appli-
cation-specific devices, such as a biosensor, where there is no
need for easy integration into a fiber-network. As a result, while
SMF28 dominates the telecom and datacom markets, there is a
rich ecosystem of visible-light SMFs with different diameters,
core sizes, and numerical apertures.
When the SMF28 facet is planar-polished, as shown in
Fig. 2(b), the fiber mode diverges into the air with a nearly
perfect symmetric Gaussian 2D profile and a numerical
aperture of 0.12, which gives a Rayleigh length of the order
of 50 μm[17]. Alternatively, the SMF28 facet can be angle-
polished (usually to 40°) [see Fig. 2(c)] to facilitate fiber-
to-photonic integrated circuit (PIC) grating-coupling. In this
geometry, the fiber mode undergoes total internal reflection
(TIR) when it is incident on the angled facet and exits the fiber
nearly orthogonal to the direction of the fiber core.
The SMF28 facet can also be polished or laser-ablated into a
hemispherical (or conical) tip, which acts to focus the 10 μm
MFD of the fiber core to a 2–4μm diameter “hot spot”located
20–40 μm from the end of the fiber, as illustrated in Fig. 2(d).
At the focal point, the resulting focused fiber mode offers
improved modal overlap with 3μm×3μmsilicon-nitride
(Si3N4) spot-size converters (SSCs) generally used in an edge
coupler-based silicon-on-insulator (SOI) photonic platform
[18]. This approach significantly improves the efficiency of fi-
ber-to-PIC edge coupling, though the smaller spot sizes result
in proportionally tighter alignment tolerances. A lensed SMF28
is well suited to single-fiber edge coupling but does not scale
efficiently to multifiber coupling (see Section 6) because the
mutual alignment of multiple-lensed fibers into an array is
practically challenging.
An alternative approach involves using ultrahigh numerical
aperture (UHNA) fibers, which use a Δnbetween the core and
cladding refractive indices much higher than that of an SMF28
in order to provide a fiber mode with MFD 3–4μmin the
O and C bands [19]. A planar-polished UHNA fiber offers
good modal overlap with the 3μm×3μmSSCs on standard
SOI edge couplers and can be spliced to an SMF28 fiber with
very low losses (≈0.2dB) [see Fig. 2(e)], thus guaranteeing easy
compatibility with standard fiber networks [20]. Additionally,
multiple UHNA fibers can be aligned into a single V-groove
array and then planar-polished to create a common facet for
the resulting fiber array (FA). This assembly offers improved
fabrication tolerances compared with arrays of lensed fibers
and is well-suited to multifiber edge coupling of PICs.
3. EDGE COUPLING
This section is dedicated to the analysis of coupling solutions
exploiting end-fire coupling schemes. Typically, when one of
these schemes is adopted, the incoming and outcoming optical
fiber needs to be tailored according to the coupling scheme;
therefore, it is crucial to carefully select the proper fiber geom-
etry. For a list of fiber solutions, we refer the reader to Section 2.
A. Inverted Taper-Based Solutions
A basic edge coupler (EC) consists of an inverse taper section,
i.e., the waveguide width is gradually reduced along the direction
of light propagation, down to a small value at the end tip. As the
waveguide dimensions decrease, the guided mode becomes less
confined, and its effective cross-section increases, while its effec-
tive index decreases. This allows for coupling to a (usually lensed)
fiber with a coupling loss determined by effects such as reflection
at the chip facet, fiber-to-tip mode mismatch, and mode-
conversion loss along the tapered SOI waveguide. Typically,
Fig. 2. (a) Cross-section schematic of an SMF28 fiber, showing the
8.2 μm fiber core centered in the cladding layer, wave-guiding the
10.4 μm MFD 1.55 μm mode. Side view schematics of (b) planar
polished, (c) angle polished, and (d) lensed SMF28. (e) Schematic
of UHNA-to-SMF28 splicing, showing the thermally expanded adia-
batic taper. The (b), (d), and (e) geometries are commonly used for
edge coupling, while the (c) geometry is preferred for grating coupling.
Review Vol. 7, No. 2 / February 2019 / Photonics Research 203
tapers longer than 100 μm are required for adiabatic conversion
of the optical mode in the waveguide [21], which is contrary to
the drive to minimize the dimensions of photonic components,
in order to achieve dense on-chip integration.
In the effort to reduce the EC dimensions, Almeida et al.
[22] demonstrated a compact tapered coupler, in which a single-
mode SOI waveguide, having a cross-section equal to 470 nm ×
270 nm,wastapereddowntoa100nmwidetipoveralength
of only 40 μm, using a parabolic-shaped transition. Numerical
simulations carried out using a 2D finite-difference time-domain
(FDTD) approach predicted a taper-induced loss of only
0.25 dB (assuming a minimum taper length of 40 μmanda
minimum tip width of 120 nm). The experimental coupling
efficiency (CE) between the taper and a 5 μmMFDfiberwas
found to be ≈46.8% (−3.3dB) for transverse magnetic (TM)
light polarization and ≈25.1% (−6.0dB) for transverse electric
(TE) polarization, at λ1.55 μm[22]. The 1-dB bandwidth
with respect to fiber misalignment was assessed to be 1.2μm.
The reason for the relatively low CE can be explained by the
fact that the mode mismatch between the tip and the fiber is
minimized for two different values of the tip width, when con-
sidering TE and TM polarization, respectively (the optimum tip
width value was found to be 50 nm for TM polarization and
120 nm for TE polarization). Also, the guided mode is enlarged
only in the in-plane direction, and so a high mode mismatch
is still present between the fiber and the tip. To achieve better
efficiency, the wave-guided mode must undergo an expansion
not only in the in-plane but also in the out-of-plane direction.
This can be achieved by fabricating a 3D waveguide taper, where
the waveguide height (and not only the width) is gradually varied
along the direction of light propagation, as reported in [23,24].
The fabrication processes required for these types of tapers in-
volve gray-scale masks and ultraviolet (UV) gray-scale lithogra-
phy, which are rather complex and not immediately compatible
with the two-tone lithography in standard CMOS process flows.
Another possibility, which is more viable from a mass fab-
rication perspective, is encapsulating a lateral Si inverse taper
structure inside an additional overlay with a lower refractive
index material, in order to implement a 2D spot size converter
(SSC). This approach is schematically shown in Fig. 3and
allows for subdecibel CEs, for both TE and TM polarized light,
over a large spectral bandwidth (1-dB bandwidth ≥100 nm).
The SSC can be implemented using polymer materials
[18,25,26], silicon oxynitride (SiOxNyor SiON) [27],
silicon-rich oxide (SiOx)[28]orSi3N4. Considering the
SOI platform, the standard SSC is realized in Si3N4because
it can be grown on the Si layer of the SOI wafer using
CMOS compatible processes. The refractive index of Si3N4
is approximately 2.0 in the O and C bands, which lies almost
exactly between the effective index of an SMF28 fiber and the
waveguides on the SOI platform, allowing for nearly optimum
index-matching conditions. A standard Si3N4SSC has a cross-
section on the order of 3μm×3μm. Growing thicker layers
of Si3N4, to implement SSCs for beams exhibiting a larger
MFD, is quite challenging because the stress added to the
SOI wafer during the Si3N4growth process could lead to
wafer-bowing and even fracturing. An increase in the MFD
of an SOI EC, to better match the 10 μm MFD of
SMF28, can be achieved by using a silicon oxynitride SSC.
By properly tuning the ratio between N2and O2gases used
during the SSC-layer deposition, a continuous variation of re-
fractive index from 2.0 (pure Si3N4) to 1.5 (pure SiO2) can be
achieved. The resulting graded-index SiON SSC can provide an
MFD of 6–10 μm, which offers good modal overlap to planar-
polished SMF28 as well as improved index-matching. These
recent developments have been well received by the photonic
foundries, allowing them to relax fiber-to-PIC alignment tol-
erances and reducing the need for expensive, specialized fibers,
such as lensed SMF28 and UHNA fibers.
To provide close access to optical waveguides and ECs, the
edges of the PIC die are typically polished to achieve optical-
quality facets. Although this process is well established and
works well at small volumes, its scale up to large volumes is
not practical. Therefore, other solutions for creating optical
facets on PICs are being investigated. For instance, the optical
facet can be realized using a two-step reactive ion-etch (RIE)
process [29], where (i) a first etching step is used to fabricate
high optical quality trenches through the SSCs and into the first
few μm of the SOI surface and (ii) a coarser deep-etching step
is used to create an approximately 80–100 μm deep trench into
the SOI substrate to facilitate fiber access and positioning, as
shown in Fig. 4.
For example, in [18], an inverse taper was designed to re-
duce the width of a 300 nm ×300 nm Si wire waveguide down
to 60 nm over a 200 μm length. Both the waveguide and the
taper were fabricated by e-beam lithography and electron cyclo-
tron resonance (ECR) plasma ion etching and then encapsu-
lated in a 3μm×3μmpolymer waveguide. Experimental
measurements showed a CE of 83.2% (−0.8dB) and a
polarization-dependent loss (PDL, defined as the CE variation
when changing the polarization of the input fiber optical signal
from TE to TM) of about 0.5 dB, for an optical fiber with an
MFD of 4.3 μm.
A similar inverse taper structure was demonstrated in [27],
regarding 400 nm ×200 nm Si wire waveguides and SiON-
SSCs, deposited by means of plasma-enhanced chemical vapor
deposition (PECVD). The use of SiON allowed for better
durability and resistance to humidity (with respect to polymer
Fig. 3. Schematic of a standard SOI EC for coupling light between
an SOI waveguide and a tapered single-mode fiber. The waveguide
(WG) is tapered down to a small tip to allow mode expansion in
the horizontal direction, whereas an overlay of polymer, Si3N4,
SiON, or SiOx, is deposited over the taper, to allow mode expansion
in the vertical direction.
204 Vol. 7, No . 2 / February 2019 / Photonics Research Review
materials) as well as lower absorption in the telecom C band.
This approach resulted in a measured CE of 89.1% (−0.5dB)
(for a 4.3 μm fiber MFD) and a 350 nm increase in band-
width, with respect to [18], due to reduced absorption by
the CH bond in the SiON material. Measurements carried out
using larger MFD fibers (9 μm) showed a maximum CE of
56.2% (−2.5dB).
Another high-efficiency SOI EC (optimized for TE light
polarization) was demonstrated by McNab et al. [26], with
a Si wire waveguide having a 450 nm ×220 nm cross-section.
The authors took into consideration the polymer SSC inverse
taper structure reported in [18] and optimized it in order to
improve the CE for TE light polarization at the expense of
the PDL. This approach allowed shortening the taper down
to 150 μm, with respect to 200 μm, as used in [18], by realizing
a polymer waveguide with a cross-section of 2μm×2μm,
while increasing the tip width to 75 nm (thus relaxing the fab-
rication tolerances). As a consequence, the improved design
allowed for a measured CE for TE polarization of 89.1%
(−0.5dB)atλ1.55 μm, with a nearly flat spectral response
over a 300 nm bandwidth. The CE for TM polarization was
instead found to be about 79.4% (−1dB). The tolerance to
lensed fiber misalignments was also experimentally character-
ized, recording a 3 dB efficiency drop for a 1 μm displacement
from the optimal position, thus proving how critical the fiber
alignment is for achieving high coupling efficiency with edge
couplers. All of the inverse tapers reported in [18,26,27] were
fabricated with e-beam lithography, in order to realize suffi-
ciently small tip widths. In contrast, the work described in
[25], used an inverse taper with a 175 nm tip width [thus com-
patible with 248 nm deep UV (DUV) lithography], which was
designed for a Si wire waveguide having a 590 nm ×220 nm
cross-section. It was theoretically shown that a CE of about
87.1% (−0.6dB) could be achieved, without excessively in-
creasing the taper length, by placing a benzocyclobutene (BCB)
spacing layer (with thickness of 200 nm) between the taper and
the polymer waveguide and by reducing the polymer waveguide
height from 3 μm (the value reported in [18]) to 1.3 μm. The
experimental characterization carried out over fabricated sam-
ples with optimum design reported, however, a fiber-taper CE
of only 64.6% (−1.9dB).
In [28] instead, a Si strip waveguide with a thickness of
220 nm (S) and initial width of 500 nm, was linearly tapered
down to 80 nm over a length of 300 μm, by means of 193 nm
DUV lithography and RIE techniques and then encapsulated
in a SiOxSSC ridge waveguide. This allowed experimental
demonstration of CE as high as 94.4% (−0.25 dB) at a wave-
length of 1.55 μm for both TE and TM polarizations, by using
a lensed fiber with MFD 3μmfor coupling.
A number of different types of SOI ECs were reported in the
scientific literature, such as those employing multilevel Si in-
verse tapers [30,31] or using a SiO2waveguide as an SSC
[31,32]. In [30], for example, it was theoretically shown that
the CE between a lensed fiber and a 500 nm high Si wire wave-
guide could be increased up to 90% (−0.46 dB) by using a dou-
ble-layer inverse taper, in which the top layer is composed by a
parabolic tapered section, followed by a sine tapered section.
Two overlapped Si tapers were also employed in [31], with
the double taper located inside a suspended SiO2waveguide
(6μm×6μm), used for mode size conversion (see Fig. 5).
The device was fabricated starting from an SOI wafer with
S220 nm and B2000 nm; in addition, 248 nm DUV
lithography was used to pattern both the wire waveguides
(width 500 nm) and the tapers (both taper layers have a
110 nm wide tip), which were then defined by two different
etching processes, one partial 140 nm deep etch and a full etch.
Fig. 5. Schematic of the SOI edge-coupling structure proposed in
[31], based on the use of a double-layer Si inverse taper and a SiO2
waveguide. Reproduced from [31].
Fig. 4. (a) SEM image of the optical facet and edge-coupler region
on an SOI-PIC, showing the mirror-finish optical facet, deeper RIE
trench for fiber access, and the diced edge of the PIC for singulation
from the rest of the wafer. (b) Schematic of a multichannel fiber array,
showing the 250 μm pitched array of fibers sandwiched between a
V-groove array and a contact plate.
Review Vol. 7, No. 2 / February 2019 / Photonics Research 205
The encapsulating SiO2waveguide was then obtained by grow-
ing a 4 μm thick SiO2layer (by means of PECVD) over the
BOX and by using an octofluorocyclobutane (C4F8) and sulfur
fluoride (SF6) based etch to define the suspended structure,
which is essential to prevent light dispersion in the BOX
and in the Si substrate. The experimental CE for the fabricated
device, when using a lensed fiber with 5 μm MFD, was be-
tween 67.6% (−1.7dB) and 63.1% (−2.0dB) for TE polari-
zation and between 63.1% (−2.0dB) and 57.5% (−2.4dB) for
TM polarization, in the wavelength range of 1.52–1.60 μm.
The alignment tolerance was also characterized, with a 1 dB
efficiency drop for a 1.7 μm displacement from the optimal
coupling position. When coupling light from a cleaved fiber
with 9.2 μm MFD, the CE was reduced to 39.8% (−4.0dB)
for both TE and TM polarizations, but with 1 dB misalign-
ment tolerance increased to 2.5μm.In[32], a Si inverse
taper was designed in an SOI platform with S205 nm and
B3000 nm, exploiting the thicker BOX to implement a
SiO2waveguide underneath the taper, which acts as a SSC. The
Si wire waveguide (width 500 nm) was tapered to a 200 nm
width over a length of 400 μm, whereas the cross-section of the
SiO2waveguide was set to 8μm×3μm; the Si substrate was
also etched in order to create a V-groove structure for allowing
fiber passive auto-alignment techniques (see Fig. 6). When con-
sidering a fiber MFD of 8 μm, CEs of 44.7% (−3.5dB) and
42.7% (−3.7dB) were theoretically demonstrated for the TE
and TM polarizations, respectively, at λ1.55 μm.
B. Trident Structures
Recently, Hatori et al. [33] proposed an alternative method
to design silicon photonic inverted tapers, without the need
for 3D structures or overlay waveguides. The authors’coupler
is composed of a three-waveguide structure, implemented in
an SOI platform (S220 nm) and having an overall length
of 150 μm, as shown in Fig. 7. The topology includes a main
tapered Si wire waveguide in the center, and two secondary
tapered waveguides placed symmetrically to the sides of the
main taper.
The trident design allows greater flexibility during the de-
sign phase. In fact, while in standard tapers the mode size at the
chip facet is only determined by the tip width (such as in [22]),
the resulting mode size from the trident design can be tuned by
adjusting the distance between the two lateral waveguides and
the width of their tips. In order to obtain a 3 μm guided mode
spot size at the chip facet, the distances between the secondary
waveguides and the width of their taper tips were set to 1 μm
and 150 nm, respectively. This acts to prevent the excitation of
high-order guided modes. The width of the side waveguides
was set to 300 nm, while the widths of both the lateral ending
tip and central waveguide tip, were set as 100 nm: these values
allowed the modal mismatch between the two discontinuities
inside the coupler to be minimized. Test structures with a
different number of trident elements were designed and fabri-
cated, to experimentally assess the fiber-waveguide CE and the
excess loss. At the central wavelength of λ1.55 μm, the CE
for TE and TM input polarization was equal to 80.9%
(−0.92 dB) and 80.5% (−0.94 dB), respectively. The simulated
CE values were 92.5% (−0.34 dB) and 86.7% (−0.62 dB),
respectively, giving an excess loss of 0.55 and 0.24 dB,
respectively.
C. Metamaterial-Based Edge Couplers
An alternative solution to implement highly efficient planar
ECs, without the need of an overlay SSC, relies on the use
Fig. 6. Schematic of the SOI edge-coupling structure proposed in
[32], based on the use of a Si inverse taper and a SiO2waveguide
implemented in the BOX. A V-groove is etched in the Si substrate
to allow for fiber auto-alignment. Reproduced from [32].
Fig. 7. (top) Schematic of the SOI trident EC structure proposed in
[33]. Reproduced from [34]. (bottom) Top view of the trident SOI
EC structure proposed in [33].
206 Vol. 7, No . 2 / February 2019 / Photonics Research Review
of subwavelength gratings (SWGs) [35–37]. As shown in
Fig. 8, an SWG consists of a Si wire waveguide in which fully
etched trenches are periodically formed along the direction of
light propagation. If the period of the grating is smaller than the
optical wavelength λand is also small enough to avoid Bragg
diffraction, then the SWG can be treated as a metamaterial,
according to effective medium theory [38]. The effective refrac-
tive index of this structure can be varied between that of Si and
of the cladding material, by changing the grating duty cycle,
i.e., the ratio between the length of the unetched grating
portion and the grating period Λ.
Implementing a Si inverse-taper-based EC with the aid of an
SWG structure offers a fundamental advantage with respect to
standard Si inverse tapers. Narrow waveguide tips are often re-
quired for high-efficiency taper-based EC designs, and these
require e-beam fabrication. Additionally, given that the optimal
tip size for TE and TM polarization is generally different, these
designs result in couplers with high PDL [22]. In contrast,
SWG-based structures offer an extra degree of freedom that
is useful at the design stage because the spot size at the coupler
tip can be controlled by both the tip width and the waveguide
effective refractive index, which can be tuned by modifying the
SWG structure.
Cheben et al. [35] carried out the first analytical study of
SWG-based ECs for coupling light into an SOI strip wave-
guide. The authors focused on an SOI platform with S
300 nm and a SiO2cladding (T6000 nm). An SWG struc-
ture with a constant period Λwas implemented, with a duty
cycle linearly chirped from 0.1 (at the tip of the coupler) to 1
(at the conjunction with the strip waveguide), in order to allow
the SWG effective refractive index to evolve along the direction
of light propagation. Different coupler lengths (from 10 to
60 μm) and profiles, including lateral tapering (by one- or
two-step linear width tapering) and vertical tapering (by RIE
lag effect), were analyzed. The performance of the different
SWG-based EC configurations was evaluated by means of
2D-FDTD numerical simulations.
The best results were obtained when numerically simulating
an SWG coupler with a length of 50 μm and two-step linear
width tapering, where the SWG width was initially varied from
30 nm (at the tip of the SWG) to 150 nm along the first
two-thirds of the coupler length and then from 150 to
300 nm, without any tapering of the SWG height. This con-
figuration showed a theoretical CE of 76.1% (−1.2dB)at
λ1.55 μm, when coupling light from a standard SMF hav-
ing an MFD of 10.4 μm and a CE of 81.4% (−0.89 dB) when
using a high numerical aperature (NA) fiber with a reduced
MFD of 5.9 μm (see Section 2). Despite the good coupling
performance, it should be noted that no fabrication-related
constraints were considered during the theoretical optimiza-
tion. As a result, the optimal tip width is 30 nm, which is
far below the resolution of the lithographic process currently
employed by silicon photonics foundries.
The first experimental demonstration of such an SWG-
based EC was reported by the same research group in [36],
using an SOI platform, with S260 nm,B2000 nm,
and a 2 μm thick SU-8 (n1.58) polymer layer as the
top-coating material. The SWG period Λwas chirped from
400 nm (at the coupler edge) to 200 nm (at the junction with
the strip waveguide) while simultaneously tapering the width
from 350 to 450 nm, so as to match the width of the strip
WG. The taper itself was composed of two different sections:
in the first low-confinement section, close to the chip-edge, the
grating gaps were fully etched, with their length linearly
decreasing from 200 to 170 nm, whereas, in the second section,
the gaps had a constant length of 100 nm and were partially
filled by Si bridging elements, having a width linearly increasing
from 100 to 450 nm, so as to obtain a smooth transition from
the SWG to the strip waveguide. The average values of the ex-
perimentally measured CE, at λ1.55 μmand using a lensed
SMF with 2 μm waist, were assessed to be 81.3% (−0.9dB)
and 75.9% (−1.2dB) for TE and TM light polarization, re-
spectively. As the SWG is based on a nonresonant phenome-
non, the transmission band was found to be almost flat for
more than 100 nm. The system also showed good tolerance
to fabrication-induced feature-size variations, as a change of
tip width from 350 to 300 nm led to an increase of the
insertion loss by less than 0.1 dB for both polarizations.
Starting from the SWG-based EC configuration described
in [36], a refined design was proposed in [37], which showed
improved performances in terms of CE and PDL. A schematic
representation and a scanning electron microscope (SEM) im-
age of the improved SWG edge coupler are shown in Fig. 9.
This coupler was implemented in a standard SOI plat-
form (S220 nm,B3000 nm) using SiO2as cladding
material. Comparing this structure with the one previously
Fig. 8. SEM image of an SWG waveguide. Inset shows the
dispersion diagrams (for TE polarization) of an SWG waveguide (blue
curve) and of a standard strip waveguide (red curve) having an effective
refractive index of 2.65. The two curves show a good match when
away from the bandgap resonance. Reproduced from [36].
Fig. 9. (a) Schematic representation of the SWG-based EC de-
scribed in [37]. (b) SEM image of the fabricated SWG-based EC.
Reproduced from [37].
Review Vol. 7, No. 2 / February 2019 / Photonics Research 207
reported, the thicker BOX (3000 nm instead of 2000 nm)
allowed for reduction in parasitic coupling to the substrate, thus
improving the overall performance. As shown in Fig. 9, the cou-
pler is still composed of two different sections, similar to the one
reported in [36]. The design consists of an initial low-confine-
ment section with fully etched gaps, followed by a second tran-
sition section with the gaps partially filled with Si bridging
elements. The initial section was designed to have a tip width
of 220 nm, and both the grating period and duty cycle were
engineered to obtain an optical MFD at the chip edge equal
to 3.2 μm. The use of a square-shaped tip allows the PDL
to be essentially eliminated (the residual PDL is 0.01 dB); this
is ultimately limited by the residual vertical asymmetry of the
structure, due to the presence of a Si substrate under the BOX,
while air is present above the SiO2cladding. It is important to
note that it is not currently easy to realize polarization-insensi-
tive inverse tapers using a standard SOI configuration. This is
due to the fact that the optimal tip width for TE and TM polari-
zation is lower than the standard Si layer thickness of commer-
cially available SOI wafers [22]; thus, solutions to reduce the Si
layer height should be included in the process [4].
Another aspect that was improved with respect to the initial
design in [36] was the transition between the first and the
second section of the SWG coupler, where an effective index
mismatch at the junction led to an extra loss of 0.3 dB. This
loss originated from the difference between the volume fraction
of silicon (the high index material of the grating) at the junction
between the fully etched and partially etched sections [see gra-
ting elements labeled A and B in Fig. 9(a)]. In order to mitigate
this issue, thus reducing the excess loss, the width of the first
element (A) of the fully etched section was slightly increased
with respect to the corresponding element in the partially filled
section (B), approximately by the size of the first Si bridging
element (C). Some test structures of the nominally optimized
design were fabricated using e-beam lithography to pattern the
SWG and showed an experimental CE of 89.1% (−0.5dB)
with an extremely low PDL (<0.05 dB). Even better CEs were
obtained by two additional designs with a small bias on the tip
width (−20 nm or 30 nm): the first one resulted in a CE of
91.2% (−0.4dB) with negligible PDL, whereas the second one
resulted in a CE of 92.9% (−0.32 dB) (for TE polarization)
with a PDL of 0.5 dB.
Many of the previously discussed edge coupling solutions
made it possible to to obtain CEs better than −1dB, although
they required the use of relatively expensive tapered or lensed
fibers for mode-matching. In contrast, Papes et al. [39] recently
theoretically demonstrated the possibility of achieving efficient
edge coupling using standard cleaved SMF-28 fibers with a
10.4 μm MFD, or high numerical aperture (HNA) fibers with
a6μm MFD, which represent a good trade-off between ta-
pered/lensed fibers and standard SMF-28 fibers. This was fa-
cilitated by the use of a standard linear Si taper (with a tip width
of 150 nm), in a 220 nm Si thick platform having a 3 μmBOX
layer and with an increased upper cladding refractive index near
the chip facet, achieved via the deposition of a stack of Si3N4
layers, as shown in Fig. 10.
This technique allowed the optical mode dimensions at the
chip edge to be increased, while simultaneously pulling it
toward the upper cladding and therefore helped to reduce op-
tical leakage to the Si substrate, as shown in Fig. 11 (bottom).
At the same time, the Si3N4layers (20 nm thick and 30 nm
thick, respectively, for the 6 μm MFD and the 10.4 μmMFD
fibers) were patterned as SWGs with varying duty cycle to
gradually reduce the effective refractive index of the layers from
that of pure Si3N4(nSi3N42.016) in correspondence with
the coupler tip, to that of pure SiO2(nSiO21.444) at the
end of the layers.
The performances of the different edge coupler configura-
tions were numerically evaluated using a mixed 3D-FDTD and
3D eigenmode expansion (EME) approach for the 6 μmMFD
configuration and a pure 3D-EME approach for the 10.4 μm
MFD configuration, resulting in a CE of 90.8% (−0.42 dB)
and 84.1% (−0.75 dB), respectively, at λ1.55 μmfor TE
mode. The wavelength dependence of the coupling efficiency
for both configurations was evaluated by means of 3D-EME
simulations, showing variations of less than 0.2 dB over the
1.45–1.60 μm wavelength window. The 1 dB fiber misalign-
ment tolerance was also assessed to be 1.3 μm for a 6 μmMFD
and 2.2 μm for a 10.4 μm MFD, which is larger than that
reported for standard inverse-taper ECs.
Fig. 10. Schematic representation of the EC structure optimized
for a 10.4 μm MFD, presented in [39]. A ridge waveguide is formed
in the upper cladding over the Si inverse taper, and Si3N4layers are
deposited in order to increase the effective refractive index of the upper
cladding. The dimensions are not to scale.
Fig. 11. Fundamental TE mode distribution at the coupler tip, of
the structure optimized for a 10.4 μm MFD. The mode is pulled to-
ward the upper cladding, therefore overcoming optical leakage to the
Si substrate. Reproduced from [39].
208 Vol. 7, No . 2 / February 2019 / Photonics Research Review
D. Vertical End-Fire Couplers
One of the main limitations for edge coupling solutions is related
to their incompatibility with wafer-level testing of the photonic
circuits and devices. A possible solution to this problem has been
provided by Yoshida et al. [40–43] who proposed a novel vertical
end-fire coupler scheme, also known as an “elephant”coupler.
This coupler used a vertically swept SOI strip waveguide, realized
by means of an ion implantation process to achieve a radius of
curvature in the order of a few μm. An SEM image of a vertically
curved coupler is shown in Fig. 12.
The vertical couplers described in [41] were implemented
in a standard SOI platform having S220 nm and
B2000 nm. The first fabrication step involved the realiza-
tion of Si wire waveguides with a width of 430 nm and nano-
tapers having a 20 μm length and 190 nm tip width, which
marks the end of the waveguides. As a second step, the SiO2
underneath the Si taper was partially removed by wet etching,
in order to obtain cantilever structures of different lengths
(ranging from 5 to 40 μm). A 0.7 μm thick SiO2protective
layer was then deposited, to protect the waveguide region out-
side of the cantilevers, and Si ions were implanted perpendicu-
larly to the chip surface. Si ion implantation (with a penetration
depth controlled by the acceleration energy, 80 keV in [41])
increases the concentration of lattice defects near the top sur-
face of the cantilever and produces a spatially confined amorph-
ization of the crystalline silicon. The change of the material
structure affects the internal stress of the cantilever, allowing
for the engineering of the cantilever bending. As a final step,
the bent Si cantilevers were encapsulated in a 2 μm thick SiO2
cladding, deposited by means of PECVD at 350°C. The lon-
gest couplers (having length between 10 and 40 μm) were fur-
ther reinforced with a spin-cast epoxy resin having a refractive
index equal to that of SiO2. The optimal ion dose required to
obtain a 90° bend varied according to the cantilever length
(from 5.5×1015 cm−2for the 40 μm long cantilevers, up to
1.7×1016 cm−2for the 5 μm long cantilevers), as higher inter-
nal stress is required to achieve shorter radii of curvature. The
main drawback of this technique is due to the additional wave-
guide propagation loss introduced by the increased concentration
of lattice defects. The propagation loss increases from 0.05 dB/μm
at 2×1015 ions∕cm2to 0.1 dB/μmat1.7×1016 ions∕cm2,
showing saturation for higher levels of ions doses [42].
It is worth noting that the overall coupling loss of the fab-
ricated elephant couplers (including the scattering/absorption
loss due to ion implantation, the radiation loss due to bending,
and the tip-fiber coupling loss) decreases when relatively short
cantilevers are considered. This suggests that a trade-off exists
between relatively long (low implantation doses, long propaga-
tion distances) and relatively short tapers (high implantation
doses, short propagation distances). The measured CE at λ
1.55 μmfor 5 μm long vertical couplers, when coupling to a
2μm spot tip-lensed single-mode fiber, was found to be 60.3%
(−2.2dB) for TE polarization and 43.7% (−3.6dB) for TM
polarization, with very low wavelength dependence. It was also
shown that this CE could be further improved by performing a
600°C annealing process after coating the cantilevers with
SiO2. This process makes it possible to recrystallize the previ-
ously amorphized silicon, theoretically making it possible to
reduce the ion implantation by more than one order of mag-
nitude, without deforming the bended coupler. However, such
an annealing process would compromise the fabrication
CMOS compatibility, as other devices would be permanently
damaged by the high annealing temperatures.
Recently, the same authors also demonstrated the possibility
to obtain light coupling between silicon vertical couplers and
high numerical aperture fibers (5 μmMFD)[44,45]. Unlike
their previous works, this time the Si nano-tapers were designed
to show an exponential tapering profile, with the waveguide
width decreasing from 430 to 50 nm over a 6 μm length, while
the 90° waveguide bending was obtained by means of argon ion
(Ar) implantation (7×1015 ions∕cm2). A 2.7 μm thick SiO2
cladding was then deposited over the curved Si nano-taper by
means of tetraethyl orthosilicate (TEOS) PECVD: as a result of
the isotropic nature of the deposition process, a dome-like SiO2
structure with 5.4 μm diameter was formed at the coupler cap.
This structure basically acted as a collimation lens, making it
possible to reduce the high divergence angle of the beam emit-
ted from the Si taper tip (where the high divergence is due to
the extremely short taper length), so as to obtain a 5 μmMFD
output beam with a nearly flat phase plane. This structure
showed a theoretical CE of 83.2% (−0.8dB) with a 0.5 dB
loss bandwidth of 420 nm for TE polarization, whereas mea-
surements carried out on experimental samples reported a CE
of only 38% (−4.2dB) with a reduced 0.5 dB loss bandwidth
of about 150 nm; the discrepancy between the simulated and
the experimental efficiency values can be explained by a re-
duced radius of curvature of the curved waveguides in the
fabricated devices with respect to the nominal value [45].
4. SINGLE-POLARIZATION GRATING
COUPLERS
One of the most popular solutions to implement fiber-to-
waveguide optical couplers is represented by vertically coupled
diffractive grating structures. The wide adoption of these devi-
ces in the field of silicon photonics is due to several factors:
(i) they provide access to any point on the PIC, thus facilitating
wafer-level testing; (ii) they do not require intensive post-fab-
rication processing, such as cleaving and facet-polishing; and
(iii) they typically provide relatively relaxed fiber-positioning
tolerances, usually higher than that of typical ECs (see
Section 3). A diffractive GC is realized by varying the wave-
guide refractive index profile according to a periodic pattern
defined along one or more dimensions. This allows for
phase-matching between the (near vertical) optical mode
Fig. 12. SEM image of a vertically bent optical coupler, obtained by
ion implantation in a silicon waveguide. The bent waveguide is 5 μm
long, and the curvature radius is approximately equal to 3 μm.
Reproduced from [43].
Review Vol. 7, No. 2 / February 2019 / Photonics Research 209
incident on the grating structure and the Si waveguides in the
horizontal plane of the PIC (see Fig. 13).
If the refractive index distribution is engineered only along
the direction of light propagation, then a 1D-GC is obtained.
If the refractive index is also varied along the width of the
waveguide, then a 2D-GC is realized. SOI-based integrated
waveguides usually exhibit a high birefringence, thus making
it challenging to achieve efficient coupling into a standard
1D-GC for both the orthogonal (and degenerate) light polari-
zation states propagating in an SMF. As a consequence, 1D-
GCs are usually optimized for a single state of polarization,
whereas more complex structures are required to implement
polarization-insensitive couplers. This section is focused on
single-polarization 1D-GCs. We first describe the working
principle (in Section 4.A); subsequently, we review some of
the most relevant 1D-GC structures proposed in the scientific
literature (Sections 4.B and 4.C); then, we will also report
about GCs suitable for vortex modes and higher-order modes
(Sections 4.D and 4.E). Polarization-insensitive GCs instead
are discussed in Section 5.
A. Introduction to 1D Grating Couplers
1. Structure and Working Principle
In a GC structure, the refractive index variation can either be
periodic (uniform GC) or nonperiodic. Nonperiodic grating
structures are usually referred to as “apodized”or “chirped”gra-
tings. Here, we will take into account a uniform 1D-GC imple-
mented in a standard SOI platform, as shown in Fig. 14, where
the kvector of the waveguide mode and the waveguide effective
index variation are assumed to develop along the zaxis.
The periodic index variation is created by partially etching
the Si waveguide with an etch depth e, thus defining etched
trenches with length LEand original thickness (i.e., unetched)
teeth with length LO. The period Λis defined as the length
of each scattering unit, thus being
ΛLELO:(1)
The grating fill factor (FF) can be defined as the ratio between
the unetched tooth LOand the grating period Λ[see Eq. (2)],
even if some authors [46] define the fill factor FF as the ratio
between LEand Λ:
FF LO
ΛLO
LOLE
:(2)
As a first-order approximation, the effective refractive index
(neff ) of the waveguide grating section can be defined as a func-
tion of the effective index of the unetched tooth and of the
etched trench (labeled as nOand nE, respectively), as shown
in Eq. (3). The resulting neff of the grating is therefore influ-
enced both by the choice of the FF and by the chosen value of e,
as increasing the etch depth causes a reduction of nE:
neff FF ·nO1−FF·nE:(3)
The grating structure can be exposed to air (as shown in
Fig. 14), or it can be coated with a top oxide layer.
In the following, we describe the basic working principle of a
1D-GC. For the sake of simplicity, we consider its use as an
outcoupling device (i.e., used to couple light from the inte-
grated chip to an optical fiber), but an analogous description
holds when the grating is used as an incoupling element.
The physical behavior of a diffractive GC can be described
in terms of the Bragg condition, which defines the relation be-
tween the wave vector of the incident optical beam and the
wave vectors of the diffracted beams. With reference to
Fig. 14 (where the index variation is introduced along the z
axis), the Bragg equation can be expressed as
km,zβ−mK,(4)
where mis the diffraction order, km,zis the zaxis component of
the wave vector of the mth diffracted order km,βis the propa-
gation constant of the waveguide guided mode, and Kis a vec-
tor lying in the zdirection, which describes the grating
according to the following expression:
jKj2π
Λ:(5)
If we take n1as the refractive index of the top medium and n2as
the refractive index of the bottom medium (which is the BOX
in this case), the wave vector of the mth diffracted order kmcan
be expressed according to Eqs. (6) and (7), whereas the wave-
guide propagation constant can be expressed according to
Eq. (8), neff being the effective refractive index of the waveguide
grating [see Eq. (3)]:
Fig. 13. Schematic representation of a 1D-GC used as an outcou-
pling device. Si layer thickness, BOX layer, etched depth, and fiber
core are shown with the right relative proportions.
Fig. 14. Cross-sectional schematic of a uniform GC implemented
in SOI technology.
210 Vol. 7, No . 2 / February 2019 / Photonics Research Review
jkmj2π
λn1in top medium,(6)
jkmj2π
λn2in bottom medium,(7)
β2π
λneff :(8)
The Bragg condition [Eq. (4)] describes phase-matching along
the zaxis for the modes diffracted by the grating.
A pictorial representation of the Bragg condition, expressed
by Eq. (4), can be obtained by means of wave-vector diagrams
(e.g., Fig. 15). Ideally, the grating is located in the center of
the diagram, and concentric semicircles are drawn around it,
with a radius directly proportional to the magnitude of the
optical wave vector kin the corresponding medium. Using
these types of diagrams, the mth order diffracted beam can be
graphically constructed by adding mtimes the grating vector K
to the zaxis component of the wave-guided beam (blue vector)
and then by tracing vertical lines from the end point of the
resulting vector, perpendicularly to the interface of the two
circles. If the vertical line crosses the wave-vector circle, then
the mth order diffracted beam is permitted, with its vector
starting at the center of the circles and ending at the intersec-
tion point. Conversely, if there is no intersection between the
semicircle and the vertical line, then the diffracted mode has
no physical meaning. Two scenarios are of particular interest:
in Fig. 15 we show the situation occurring when Λλ∕neff
(i.e., the optical wavelength inside the grating). In this con-
figuration the 1st order diffracted mode is vertically emitted
from the grating and can therefore be coupled to an optical
fiber, but the 2nd order diffracted mode is reflected back to
the waveguide. This configuration, also known as resonant
configuration, is usually not employed in practice, unless a
specific strategy to suppress the reflected contribution is
adopted. This can be realized, for example, by implementing
a partially reflecting mirror before the grating, as reported in
[47]. On the other hand, if we consider the case Λ>λ∕neff
(see Fig. 16), the 1st order diffracted mode is emitted with
its direction slightly detuned from the vertical axis, while
the 2nd order in-waveguide reflection contribution is sup-
pressed. Given the considerable advantages of this setting, this
configuration is the most widely adopted in designing 1D-GCs.
By considering the 1st order diffraction angle (θ1) in the de-
tuned configuration, the Bragg condition of Eq. (4) can be re-
formulated as
neff −n1·sin θ1λ
Λ:(9)
This equation shows that, when an optical beam having a
central wavelength λinteracts with a uniform GC, it is dif-
fracted at an angle θ1, which can be tailored by appropriately
adjusting its geometrical parameters (Λ,e, FF). Both in the res-
onant and in the detuned configurations, two 1st order dif-
fracted contributions exist: one is emitted upward (i.e.,
toward the collection fiber), while the second one downward
toward the substrate. This practically implies that part of the
optical power incident on the grating will be directed toward
the substrate, representing a power loss. It is also important to
note that the Bragg condition is strictly valid only for infinite
grating structures and for small index variations. For a finite
structure, diffraction does not occur at a single discrete k-vector
only but for a range of k-vectors around that predicted by the
Bragg condition. As a result, real GCs show a nonzero spectral
bandwidth for waveguide-to-fiber coupling.
SOI-based GCs can be implemented by etching vertical
trenches in the Si slab, resulting in square-shaped devices, as
shown in Fig. 17(a). In this case, the width of the grating (along
the xdirection, with reference to Fig. 17) is usually chosen to be
about 12 μm[48], in order to properly accommodate the
Gaussian beam emitted by an SMF fiber typically exhibiting
a 10.4 μm MFD. In this case, a linear taper whose length is
in the order of hundreds of μm (400 μm, for example, in
[48]) is required to connect the wide grating section to the sub-
micrometer-sized Si waveguide, thus allowing for low-loss
mode conversion. An alternative solution is to use a focusing
configuration [49], as shown in Fig. 17(b), which allows one to
considerably reduce the coupler footprint space. In this case,
the grating trenches are no longer straight lines but describe
sections of different ellipses, all having a common focal point
located between the integrated waveguide and the coupler sec-
tion. If the grating surface lies in a plane of coordinates xand z
(according to Fig. 17), where zis the direction of light propa-
gation and the origin is set in the desired focal point, the focus-
ing grating can be defined by curving the grating trenches [50],
so that
Fig. 15. Wave-vector diagrams of waveguide GCs in resonant
configuration.
Fig. 16. Wave-vector diagrams of waveguide GCs in detuned
configuration.
Review Vol. 7, No. 2 / February 2019 / Photonics Research 211
q·λ0neff ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
x2z2
p−z·nt·cos θc,(10)
where qis an integer number for each grating trench, θcis the
angle between the fiber and the chip surface, ntis the external
refractive index, λ0is the optical wavelength in vacuum, and
neff is the effective refractive index experienced by the cylindri-
cal wave in the grating section. If a straight-trench grating and a
focusing grating are characterized by the same etch depth,
period, and fill factor, their CEs are almost exactly the same;
when designing a focusing grating, it is therefore convenient
to initially assume a straight-trench configuration and, once
the optimal geometrical parameters (period, fill-factor, etch
depth) have been identified, use them to curve the grating
according to Eq. (10).
The performance of a generic GC is typically determined by
three parameters.
•Directionality: the ratio between the optical power
diffracted upward (Pup) toward the fiber and the optical power
propagating along the integrated waveguide before the grating
(Pwg). (Pup and Pwg are defined with reference to Fig. 14.)
•Reflectivity: the ratio between the optical power reflected
back to the waveguide by the GC and the optical power im-
pinging from the waveguide on the GC (Pwg). A nonzero
reflectivity is always present, even in detuned grating configu-
rations, because of the refractive index contrast between the Si
waveguide and the grating section; it is important to minimize
this effect, as it can cause parasitic Fabry–Perot oscillations in
the PIC.
•Overall coupling efficiency: the ratio between the opti-
cal power coupled to the fundamental mode of the optical fiber
and the optical power propagating along the waveguide (Pwg).
Thanks to the reciprocity theorem, which can be applied to
linear open systems as silicon gratings, if we consider single-
mode waveguides and single-mode fibers, the CE in the output
configuration (i.e., light travels from the PIC waveguide to the
fiber) and in the input configuration (i.e., light couples from
the fiber to the PIC waveguide) is exactly the same [51].
In order to provide the reader with reference values,
which are useful to better understand the advantages and
disadvantages of structures proposed in Sections 4.B and 4.C,
we describe in the following the typical performance offered by
a 1D-GC with standard design and fabrication parameters.
2. Performance of Uniform Grating Couplers
For a uniform GC (UGC) based on a standard SOI platform,
having S220 nm and B2000 nm, the maximum CE is
usually achieved for shallow etching levels (e≈70 nm) and FF
values close to 0.5, and it is limited to values <60%. This can
be seen in Fig. 18, where the CE spectrum calculated as a func-
tion of the wavelength is reported for different values of eand
FF, assuming TE light polarization (along the xdirection, with
reference to Fig. 14). In these simulations, the grating period Λ
was set to 634 nm, so as to obtain a diffraction angle θequal to
14.5° for a central wavelength of 1.55 μm. As previously men-
tioned, a 1D-GC optimized for TE light polarization will sup-
press light in the opposite polarization, as shown in Fig. 19.
The low CE of standard UGCs can be explained in terms of
poor directionality and poor mode-matching between the radi-
ated field distribution and the Gaussian field distribution of the
fundamental mode of an SMF.
Fig. 17. (a) Schematic top view of a GC with straight trenches.
(b) Schematic top view of a focusing GC.
Fig. 18. (Top) CE for a UGC implemented in a standard SOI plat-
form (S220 nm and B2000 nm), with a grating period Λ
634 nm and FF 0.5. CE is reported as a function of optical wave-
length λfor different values of etching depth e, ranging from 60 to
80 nm. (Bottom) CE for a UGC implemented in a standard SOI plat-
form (S220 nm and B2000 nm), with a grating period Λ
634 nm and e70 nm. CE is reported as a function of optical wave-
length λfor different values of FF, ranging from 0.4 to 0.6.
212 Vol. 7, No . 2 / February 2019 / Photonics Research Review
Concerning directionality, a considerable part of the optical
power incident on the GC can be diffracted downward, toward
the substrate, unless a proper structure is designed. In a per-
fectly symmetrical situation, assuming infinitely extending
BOX and TOX and full etching of the Si layer, the 1st order
upward and downward diffracted modes would have the same
intensity, thus clipping the maximum achievable CE to 50%.
In practical SOI implementations, part of the power diffracted
downward is reflected at the interface between the BOX and
the Si substrate, thus making it possible to increase direction-
ality and, consequently, the maximum achievable CE. The GC
directionality shows a sine-type dependence on the BOX layer
thickness: the maximum directionality values are achieved
when the reflected contribution is in phase with the upward
diffracted field. However, it is important to note that this
parameter generally cannot be optimized in the grating design
stage, i.e., it directly depends on the structure of the SOI wafer
chosen for fabrication, and other components may define the
required BOX thickness. Nevertheless, it is interesting to note
that SOI wafers with different geometrical specifications are
currently commercially available [52]. Some of the techniques,
which have been proposed and demonstrated in the scientific
literature to control and boost the GC directionality, will be
described in Section 4.C.1.
In order to better understand the mode-matching issue, it is
useful to analyze the normalized power density profile radiated
from a UGC, as shown in Fig. 20. The specific data refer to a
uniform 1D-GC realized in an SOI wafer with S220 nm
and e70 nm, but similar results are obtained even varying
these parameters; thus, the conclusions that can be derived have
general validity. As shown in Fig. 20, in UGCs most of the
optical power Pis radiated in the initial section of the grating
and then an exponentially decaying trend is observed, which
can be expressed as
PzP0e−2αz:(11)
In the above equation, αis the grating coupling strength
(α≈0.146 μm−1for the example of Fig. 20), whose value
varies as a function of eand FF, as shown, for example, in
Fig. 21, where the normalized power density profiles of the dif-
fracted mode are reported, for GC realized in SOI wafers with
S220 nm, having e70 nm and FF ranging from 0.5
to 0.8.
Starting from the coupling strength α, the grating coupling
length Lc(the length of the region inside the grating where
spatial transients develop) can be defined as
Lc1
2α:(12)
As the fundamental mode of SMF is characterized by a
Gaussian distribution, with an MFD of 10.4 μm, the gra-
ting-fiber mode overlapping is, however, quite poor (usually
<80%), thus affecting the resulting CE.
B. Grating Couplers for Large-MFD Optical Beams
In this section and in the next, we review the most relevant GC
structures present in the literature to date. We start our discus-
sion from the oldest Si-GCs, designed to work at λ1.3μm
using external collimating optics to in/outcouple light to/from
the device. GCs specifically designed for fiber coupling will be
extensively discussed in the next section.
The first generation of SOI GCs was developed between the
end of the 1990s and the early 2000s. Twenty years after the
seminal work published by Neviere [53], Pascal et al. carried
out the first analytical study of the interaction of an SOI wave-
guide grating with an incident Gaussian beam, under the para-
xial approximation hypothesis [54]. In this work, the authors
found that, in order to maximize the grating CE, the waist of
Fig. 19. CE of a TE-optimized 1D-GC as a function of optical
wavelength for TE light polarization (blue trace) and for TM light
polarization (red trace).
Fig. 20. Normalized power density profile of the diffracted mode
from a UGC, implemented in a 220 nm Si thick SOI platform with
a 70 nm deep etch.
Fig. 21. Normalized power density profile of the diffracted mode
from UGCs, implemented in 220 nm Si thick SOI platform with
a 70 nm etch depth and FF ranging from 0.5 to 0.8.
Review Vol. 7, No. 2 / February 2019 / Photonics Research 213
the input Gaussian beam, which was focused on the grating
using a discrete-elements optical system to adjust the waist
radius, had to be equal to 1.37Lc. The theoretical results were
confirmed by the experimental characterization of a UGC fab-
ricated using a SIMOX SOI platform (S200 nm and
e140 nm), designed for λ1.3μmand showing a maxi-
mum CE of about 30% (−5.2dB) for an input waist radius
of 300 μm. A few years later, Ang et al. demonstrated the first
SOI GC based on the use of UNIBOND SOI wafers with rel-
atively thick silicon layers (≈1μm)[55,56]. Uniform GCs
were patterned using e-beam lithography, with a period Λof
400 nm, so as to couple light at a wavelength λ1.3μm, with
a diffraction angle θ1<12°. The maximum value of the exper-
imentally observed directionality (≈70%) was found when a
GC with a 140 nm etch depth was considered. Conversely,
the theoretical results (obtained using a perturbation theory ap-
proach) suggested that increasing the etch depth to 200 nm
could improve the GC directionality up to 85%. It is, however,
important to note that the relatively small refractive index con-
trast between the etched and unetched portions of the grating
caused in those systems a low coupling strength and a long Lc,
thus implying the need for long (e.g., 1 mm) structures [55].
As previously mentioned, one of the main limiting factors
toward the realization of highly efficient GC is the limited di-
rectionality provided by the device. To overcome this limita-
tion, two different approaches were proposed. The first
approach targeted to “break”the GC symmetry, properly de-
signing its structure to minimize the diffracted contribution in
a specific direction. The second approach was based on the
use of bottom mirrors, making it possible to recover the beam
diffracted in the unwanted direction. Both of these techniques
have also been later applied to GCs specifically designed for
fiber-coupling. An example of the first approach can be found
in [57], where Ang et al. proposed and experimentally charac-
terized a blazed grating structure using the same UNIBOND
SOI wafers reported in [56]. A schematic representation of the
blazed GC is shown in Fig. 22.
These structures were fabricated using e-beam lithography
and Ar-ion etching with a precisely controlled angle of inci-
dence. This technique made it possible to produce a tilting angle
of the grating teeth equal to 70° with a grating periodicity Λ
383 nm (slightly less than the 400 nm selected during the design
phase) so as to couple laser light at λ1.3μmwith an MFD
of approximately 300 μm (adjusted by means of a lens system).
The orientation of the grating teeth was chosen in order to min-
imize the power diffracted toward the substrate and to maximize
it in the upward direction, where a Ge-photodetector was used
to collect the outcoupled optical power, demonstrating a CE
of 84% (−0.76 dB). The second approach, based on the use
of mirrors to improve the grating performances, was first re-
ported in [58]. In this case, UGCs with period Λ380 nm
and FF 0.35 were implemented in a SIMOX SOI platform
with S200 nm and B400 nm. The gratings were coated
by a SiO2passivation layer, and an aluminum layer was deposited
on top of the input grating. Light was sent through the substrate
(where an antireflection coating was deposited) to the 1st (incou-
pling) grating, propagated through the Si waveguide and then
reached the 2nd (outcoupling) grating, which diffracted the
light through the SiO2passivation layer, as shown in Fig. 23.
Using this configuration (input beam waist of 24 μm), a maxi-
mum CE of 57% (−2.44 dB) was experimentally demonstrated.
C. Direct Fiber-to-Grating Couplers
A common feature of all the previously discussed UGC struc-
tures is that they were optimized considering free-space optical
beams with an MFD much larger than 10.4 μm, which is the
characteristic of optical beams propagating in an SMF (see
Section 2). As a consequence, all those solutions require the
use of collimating and focusing lenses in order to adjust the
beam properties of the incoming beam. Additionally, it must
be highlighted that the direct measurement of the outcoupled
radiation by means of a Ge photodetector does not allow one to
properly consider the mode-mismatch loss due to the different
spatial profile between the exponential decay shape of the GC
diffracted beam and the Gaussian profile of the fiber mode.
The first SOI GC specifically designed to be in/outcoupled
from/to SMFs was proposed only by Taillaert et al. [59], based
on an SOI wafer with a 220 nm thick Si layer, a 925 nm thick
BOX (value obtained by an optimization procedure), and a
1μm thick SiO2TOX. In this paper, the authors showed that
the use of a nonuniform (apodized) GC can strongly enhance
the overall CE by reshaping the spatial profile of the optical
beam outcoupled by the GC.
Taking into account the coupling strength αand indicating
with zthe grating-coupler direction (as shown, for example, in
Fig. 17), it is possible to write the equation giving the change of
the optical power propagating along the GC, as a function of
the position:
dP
dz−2αz·P: (13)
As a consequence, when a GC with a uniform αis realized, the
diffracted optical power decreases exponentially while moving
along the grating direction, according to Eq. (11), while a
Fig. 22. Schematic of the SOI-blazed GC proposed in [57].
Fig. 23. Schematic of the 1D-GC structure proposed in [58]. Good
CE is obtained thanks to the presence of an aluminum layer deposited
in the region above the GC.
214 Vol. 7, No . 2 / February 2019 / Photonics Research Review
Gaussian intensity distribution can be obtained if αis properly
varied along the structure. To obtain a Gaussian output beam,
the αzprofile should be tuned by controlling the evolution
of ezand FFzalong the GC so that the condition reported
in Eq. (14) is satisfied, where Gzrepresents a normalized
Gaussian distribution with a standard deviation corresponding
to the MFD of the considered fiber (i.e., 10.4 μm for SMF):
2αz G2z
1−Rz
0G2tdt:(14)
In the following, we briefly report on the design procedure used
by Taillert et al. to maximize the CE of the designed struc-
ture [59].
•A UGC was first designed. Λwas selected to obtain
coupling at λ1.55 μmfor a specific diffraction angle (θ),
and the etch depth (e) was defined so as to achieve the maxi-
mum value of the coupling-strength coefficient (α) required
by Eq. (14).
•αFFcalculation. While keeping Λand efixed, the value
of the FF parameter was scanned from 0 to 1, and a curve of
αFFwas then constructed.
•Apodized grating buildup. Starting from the results of
Eq. (14) and those of αFF, the FF was adjusted in each period
of the grating, so that each unit provides the required coupling-
strength coefficient, as reported in Fig. 24.
It is important to note that the numerical simulation results
(by 2D approximation of the eigenmode expansion method),
achieved considering the apodized grating coupler [(AGC) in
the following] designed according to the above strategy, showed
only a 75% (instead of the expected 100%) mode overlap be-
tween the grating diffracted mode and the fiber Gaussian mode.
This is due to the combination of three different effects, which
are not fully considered by the previously described optimiza-
tion procedure.
•Equation (14) is strictly accurate only for an infinite gra-
ting structure and for small values of coupling strength αz.
•By modifying the FF in each section, the neff value of each
scattering unit is varied [see Eq. (3)]. As the GC period Λis
kept constant along the whole grating structure, the neff varia-
tion implies that the Bragg condition [Eq. (9)] is satisfied only
at a specific GC position and not along the whole structure,
thus causing a nonzero phase mismatch between the radiation
components scattered by two consecutive scattering units.
•The nonuniform grating is based on the assumption that
the coupling strength (α) of each single scattering unit, with
given eand FF, perfectly matches that calculated by considering
a uniform and infinite GC, but this assumption is not gener-
ally true.
In order to further increase the modal overlap, the AGC,
designed by the previously described methodology, was used
as the starting point for a numerical optimization, which
allowed the distance between the grating teeth to vary by
5 nm steps. As a result of this optimization, based on a genetic
algorithm (GA), the modal overlap increased from 75% to
97%, thus making it possible to achieve a theoretical CE of
61% (−2.15 dB). The optimal grating configuration was also
simulated with the addition of a backreflector inside the BOX,
implemented as a two-pair distributed Bragg reflector (DBR),
which boosted the theoretical maximum CE up to 92%
(−0.36 dB), with a 1-dB bandwidth of 35 nm, as shown
in Fig. 25.
Despite the high CE obtained by numerical simulations, it
must be highlighted that introducing a DBR structure strongly
increases the complexity of the GC fabrication and that the
optimum GC structure reported in [59] requires trenches as
narrow as 30 nm, which are hardly compatible with the reso-
lution allowed by common UV and even e-beam lithography
processes. Nevertheless, the work by Taillaert et al. marked a
turning point in GC design, and many other researchers started
to investigate different optimization strategies, as described in
the following section.
1. Directionality Improving Techniques
A first option to improve directionality is the adoption of mir-
rors, placed beneath the GCs, thus acting as backreflectors.
They make it possible to recover part of the light that would
travel in the substrate, thus improving the GC efficiency.
Mirrors can be realized by either using DBR structures or metal
Fig. 24. Gaussian output beam and corresponding α(z) calculated
according to Eq. (14). Dotted curve is the resulting output from the
simulation after numerical optimization. Reproduced from [59].
Fig. 25. CE of the optimized nonuniform SOI GC proposed in
[59], with (continuous line) and without (dashed line) of a two-pair
DBR. Reproduced from [59].
Review Vol. 7, No. 2 / February 2019 / Photonics Research 215
layers. In [60], the authors reported on the fabrication and
experimental characterization of a UGC with an embedded
backreflector realized as a DBR, by using a stack of amorphous
Si∕SiO2layers with λ∕4thickness and positioned at a dis-
tance of 1.48 μm from the Si waveguide layer, to guarantee
the constructive interference of the reflected beam with the up-
ward diffracted beam. The GC was patterned by means of
193 nm deep-UV lithography and RIE, with e70 nm and
Λ630 nm, so as to diffract the optical beam with λ
1.55 μmat an angle θ10° (see schematic in Fig. 26). The
measured CE was found to be 69.5% (−1.58 dB), with a 1-dB
bandwidth of 36 nm [60]. We note that, although SOI wafers
with embedded DBR structures are not currently available as
standard products from semiconductor manufacturing compa-
nies, some authors have shown that their fabrication could be
made commercially reproducible by employing a simple modi-
fied version of the ion-cut technique used to fabricate standard
SOI wafers [61].
Metallic mirrors have also been widely considered by the
research community: a gold backreflector was employed in
[62] to improve the efficiency of a UGC defined in a 220 nm
Si thick SOI platform with e50 nm and Λ610 nm (see
Fig. 27). After patterning, the GC was coated with a benzocy-
clobutene (BCB, n1.54 at λ1.55 μm) buffer layer,
whose thickness was set to 840 nm to satisfy the constructive
interference condition for the reflected wave. The gold mirror
was deposited and, subsequently, the SOI structure was bonded
to a Pyrex host substrate by a second BCB layer. Finally, the
Si substrate of the SOI wafer was removed, leaving the
BOX layer (1 μm thick) on top of the final device; light was
then injected into the grating through the BOX, making it pos-
sible to achieve an experimental CE of 69% (−1.61 dB), almost
identical to that reported in [60], where a DBR structure was
alternatively considered.
Aluminum can also be used as the constituent of a backre-
flector section, as shown, for example, in [63,64], where the
GC was implemented starting from SOI wafers with a native
Si thickness of 250 nm and BOX thickness of 3 μm. In [63],
the aluminum backreflector was deposited in membrane win-
dows obtained by etching the backside of the wafer underneath
the grating regions, whereas in [64] a flip-chip bonding similar
to the one described in [62] was used, employing SiO2as a
buffer layer.
When comparing the flip-chip bonding with the backside
etch techniques, it must be stressed that the latter approach
should be preferred when active photonic components, such
as modulators, photodetectors, or heaters, need to be integrated
on the PIC, thus needing direct access to the electrodes, placed
on the chip surface. However, it also has to be remarked that
the use of a metallic backreflector may involve the use of
non-CMOS compatible materials, thus making their adoption
quite challenging in a real CMOS photonic foundry scenario.
Another possible approach to increase the GC directionality
is based on the deposition of a polycrystalline silicon (p-Si) layer
over the grating structure [65,66], which makes it possible to
increase the refractive index contrast between the teeth and the
trenches on top of the grating. In order to maximize the direc-
tionality of this type of structure, the thickness of the overlayer
and the etching depth emust be optimized in such a way to
impose a πshift between the phase of the optical beam con-
tributions diffracted by each consecutive trench and tooth dur-
ing the propagation toward the upper cladding. It was
theoretically shown that, considering a grating with Λ
610 nm (λ1.55 μm,θ10°) implemented in an SOI wa-
fer with S220 nm and B2μm(see Fig. 28), a direction-
ality of 85% can be achieved by setting e220 nm and using
an overlayer thickness of 150 nm [65]. As a reference value, the
maximum directionality of a GC without the use of any over-
layer in the same conditions is limited to 55%. The overall CE
was found to be 66% (−1.81 dB) for a UGC, while it increased
to 78% (−1.08 dB) after grating profile apodization (by GA
optimization) while satisfying the requirements for deep-UV
lithography-based fabrication (i.e., 200 nm minimum feature).
The use of an overlayered structure also increases the cou-
pling spectral bandwidth. The simulated 3-dB bandwidth was,
in fact, found to be 85 nm, with a 25 nm increase with respect
to a simple grating structure without the use of any overlay
[48]. In [66], the optimal overlayered grating was fabricated
via adopting a uniform configuration with FF 0.35 and us-
ing poly-Si for the overlayer, demonstrating an experimental
CE of 69% (−1.6dB)atλ1.53 μm, with a 1-dB and 3-dB
bandwidth of, respectively, 44 and 80 nm. In [67] instead,
the overlayer was obtained by epitaxial silicon growth in a re-
duced-pressure chemical vapor deposition tool by using SiH4at
700°C. In this case, an experimental CE of 55% (−2.6dB) was
achieved. The reason for this discrepancy with respect to [66]
Fig. 26. Schematic structure of the DBR-assisted UGC reported
in [60].
Fig. 27. Schematic structure of the Au-mirror-assisted UGC re-
ported in [62].
Fig. 28. Schematic structure of the poly-Si overlayer UGC reported
in [65].
216 Vol. 7, No . 2 / February 2019 / Photonics Research Review
can be explained by the growth of an excessively thick Si over-
layer of 180 nm with respect to the optimum value of 150 nm.
A different solution was proposed by Saha and Zhou [68].
In this case, the overlayer can be obtained as a single crystalline
silicon nano-membrane (Si-nmb), which can be patterned on a
separate SOI wafer by a complete etch-through process, re-
moved from the initial SOI carrier by immersion in an aqueous
diluted HF solution and finally transferred on the target optical
waveguide implemented in a second SOI wafer (see Fig. 29).
This approach makes the Si-nmb overlay grating be indepen-
dent on etching depth errors, making the overall process ideally
robust; it is worth noting that the ultimate performance of this
class of grating can be reached only when a 150-nm-thick
silicon target wafer is used [66]. Conversely, when a standard
220-nm-thick target wafer is employed in conjunction with a
240-nm-thick Si-nmb (as in [68]), numerical simulations
showed a directionality of 81%, leading to a CE of 64%
(−1.94 dB), therefore slightly lower than in [65].
The possibility to implement the overlay using other high-
index and CMOS-compatible materials was also demonstrated.
Yang et al. [69] demonstrated the possibility to deposit a
230 nm thick germanium overlayer on a 220 nm Si-thick
SOI grating, already patterned with a 60 nm deep etch
[69]. The high refractive index of germanium (nGe 4.28
at λ1.55 μm) allowed a significant directionality increase
(92%), while the layer was kept sufficiently thin to avoid
the introduction of a large absorption loss (which was calcu-
lated to be 0.2 dB). Combining this system with a numerically
optimized nonuniform grating profile, a theoretical CE of 76%
(−1.2dB)atλ1.55 μmwas demonstrated.
A different solution is represented by the adoption of slanted
GCs, as proposed in [70,71]. Similar to the principle exploited
in blazed GCs [57], the slanted GC is composed by tilted gra-
ting sections (see Fig. 30), aiming to reduce the power leaked to
the substrate and, simultaneously, increasing the optical power
being sent toward the optical fiber. In [70], a slanted GC with
vertical emission was designed in a 240 nm Si thick SOI plat-
form (Λ649.5nm,FF 0.328,δ59.71°, λ1.55μm),
thus demonstrating a theoretical CE of 69.8% (−1.56 dB). An
increased CE of 75.8% (−1.2dB) was reached when employing
a nonuniform grating structure. Another demonstration was
provided in [71]. In this case, the emission was not vertical
(θ10°, λ1.55 μm), and the silicon thickness was
220 nm, demonstrating a theoretical directionality of 83%
and a CE of 64% (−1.9dB). It is important to note that
the actual efficiency of these devices strictly depends on the
ability to fabricate precisely tuned angled grating sections,
which may require relatively complex fabrication proce-
dures [71].
Despite the good theoretical results, which are definitely
comparable with the overlayer gratings reported in [65], the
fabrication of slanted gratings can be, however, quite tricky,
as it requires the use of direct etching techniques such as a
focused ion beam (FIB), in which the etch depth and the width
of the slanted slits are difficult to control precisely. In [71], for
example, the optimized grating configuration was fabricated us-
ing an FIB at an angle of 58°, employing Al2O3as a hard mask
and I2as a selective etching agent; the fabricated experimental
samples showed, however, a CE of only 46% (−3.3dB).
2. Apodized Grating Couplers
The grating CE is strictly related to the modal overlap between
the fiber mode and the optical mode of the grating structure.
An apodization or, equivalently, chirping technique represents a
powerful tool to increase the modal overlap, thus leading to
enhanced GC performance. Three main approaches to GC
apodization are commonly used and sometimes combined.
•Mode-targeting apodization. The apodization profile is
defined so as to match a specific intensity-distribution profile
of the diffracted beam.
•Numerical-optimization. Computationally intensive
simulations, often exploiting GAs are used to define the apod-
ization profile maximizing a specific performance figure
(generally CE).
•Linear-chirping. One of the GC’s profile parameters
(e.g., FF or Λ) is linearly chirped, while the other parameters
are kept constant, or modified accordingly, depending on the
specific embodiment.
A first approach is related to the apodization of the grating
fill-factor, as proposed in [59]. This allowed one to obtain the
“ideal”coupling strength distribution, i.e., αz, thus produc-
ing a Gaussian intensity distribution of the scattered beam. A
similar approach was employed in [46] by Chen et al. consid-
ering an SOI platform with a thicker Si layer (340 nm). In this
case, the authors first carried out 2D-FDTD simulations of a
uniform grating configuration in order to identify the value of e
(the etch depth) yielding the highest directionality. Once the
optimal ewas identified (200 nm), the design process was al-
most identical to that previously done by Taillaert et al. [59],
i.e., the αFFcurve was numerically calculated (FF
varied from 0.6 to 0.92), and then the grating structure was
composed by defining two sections. In the initial section, 11
units long, the FF was varied so as to satisfy Eq. (14), while
in the final section a uniform GC with Λ610 nm and FF
0.6was realized. In the initial section (with apodized FF), the
period Λwas also tuned to satisfy the phase matching condition
Fig. 29. Schematic structure of the Si-nmb overlayer UGC reported
in [68].
Fig. 30. Schematic representation of a slanted SOI GC. The period
of the uniform grating is equal to Λand the etched slits angle is equal
to δ.
Review Vol. 7, No. 2 / February 2019 / Photonics Research 217
[expressed by Eq. (9)], so that no further numerical optimiza-
tion was required (contrarily to [59]), and a final CE (theoreti-
cal) of 84% (−0.76 dB) was obtained when considering a
coupling angle θ10°. It is interesting to note that the apo-
dized structure also showed great reduction of the in-waveguide
reflectivity compared with the case of a UGC with the same
etch depth e. The experimentally measured CE of fabricated
samples was, however, found to be 75.8% (−1.2dB)at
λ1.53 μmshowing a 3-dB bandwidth of about 45 nm.
An alternative technique relies on the use of subsequent
numerical optimization (for example, those based on GAs)
to obtain the desired structure, showing improved modal over-
lap and enhanced CE. A GA-based apodization, combined with
the use of an embedded backreflector, was used to optimize the
performances of a GC implemented in a 250 nm Si thick SOI
platform with a 3 μm thick BOX and made it possible to ex-
perimentally demonstrate a CE of 87% (−0.62 dB), which, to
the best of our knowledge, constitutes the current record value
for SOI 1D-GCs [63].
A third strategy relies on the linear chirp of the FF [72–76].
In this case, the FF of each scattering unit is linearly decreased,
as a function of the distance of the considered scattering unit
from the GC starting point, according to Eq. (15), where FF0is
the initial fill factor of the first radiative unit, Ris the linear
apodization factor (expressed in μm−1), and zis the distance
of each radiative unit from the starting point of the grating.
A cross-sectional schematic representation of a linearly
apodized GC is shown in Fig. 31:
FF FF0−R·z: (15)
He et al. designed and realized a linearly apodized shallow-
etched GC in a 220 nm Si thick SOI platform (B2μm,
e60 nm)[72]. The grating was divided into two sections.
The first 13 grating elements were apodized according to
Eq. (15), with FF00.719,R0.03 μm−1, and
Λ640 nm, followed by a UGC section with the same Λ
and constant FF 0.469. The initial fill factor (FF0) of the
apodized section was selected so as to obtain a minimum trench
length LEof 180 nm, thus allowing one to fabricate the optimal
grating configuration using standard deep-UV (248 nm) lithog-
raphy. The fabricated samples showed an average CE of 49%
(−3.1dB) and a 1-dB bandwidth of ≈41 nm, with a best per-
forming device exhibiting a CE of 54% (−2.7dB). A similar
design approach was also applied to a 250 nm Si thick SOI
platform with B3μmand e70 nm [73]. The optimal
design was found for FF00.9,R0.06 μm−1, and a
constant period Λof 590 nm, demonstrating a maximum
theoretical CE of 59% (−2.3dB) and an experimental CE
of 54% (−2.7dB), with a 1-dB bandwidth of ≈30 nm. In this
case, the initial fill factor corresponds to a minimum trench
feature of 59 nm, which is compatible with the ultimate
resolution limitation of an e-beam lithography process.
Following a modified approach, Chen et al. demonstrated a
linear apodization of the grating period Λ, which was used to
reduce the reflectivity and enhance the CE of a GC designed for
perfect vertical coupling [74]. The structure, divided into two
sections, was implemented in a 220 nm Si thick SOI platform
(B2μm), with a shallow etch depth e70 nm. In the first
apodized section, the GC had a constant FF 0.47, and its
period Λvaried according to Eq. (16), where Λ1is the period
of the first grating element, pis the number of the pth scattering
element, pTOT is the overall number of elements composing the
apodized section, and Δis the maximum grating period
deviation:
ΛpΛ1p
pTOT −1Δ:(16)
The second part of the grating was instead designed as a uni-
form section, with its parameters (Λ590 nm,FF 0.34)
chosen so as to maximize the reflectivity into the waveguide.
In vertically emitting GCs, in fact, a high-reflectivity grating
rear section makes it possible to improve the overall CE of
the device [74]. The CE was predicted to be 42% (−3.77 dB)
by numerical simulations of an optimized structure with Δ
−120 nm and pTOT 9, while the experimental characteriza-
tion of fabricated samples showed a CE of 34% (−4.69 dB). A
3-dB bandwidth of 45 nm was experimentally measured, which
matched quite well the theoretically expected value of 48 nm.
One common feature of the previously described linear
apodization techniques [72–74] is that the etching depth e
is chosen before the apodization profile optimization, i.e., ac-
cording to the value, which maximizes the directionality of
UGCs implemented in the same SOI platform. Moreover,
the linear chirp is applied to the FF while keeping the period
Λconstant [72,73] or, similarly, to Λwhile having constant FF
[74]. This constraint hinders the possibility of satisfying the
Bragg condition [Eq. (9)] along the whole grating length for
given λand θ, thus affecting the overall GC performance [76].
To overcome this limitation, a different approach was re-
cently proposed in [76], based on a simultaneous FF linear
apodization [according to Eq. (15)] and period (Λ) variation
for each scattering unit in order to fulfill the Bragg condition.
In this way, the length LE,iand LO,iof the, respectively, etched
and unetched portions of the ith scattering section can be ex-
pressed as a function of the fill factor of the ith unit (FFi)by
Eqs. (17) and (18), where θair is the diffraction angle in air:
LE,iλ1−FFi
nEFFinO−nE−sin θair
,(17)
LO,iλ·FFi
nEFFinO−nE−sin θair
:(18)
In Eqs. (17) and (18), ziis the distance of the ith grating
scattering unit from the origin of the grating, which can be
written as
Fig. 31. Cross-sectional schematic of a nonuniform GC imple-
mented in an SOI wafer, based on a linear apodization of the
grating FF.
218 Vol. 7, No . 2 / February 2019 / Photonics Research Review
ziX
i−1
j0
LO,jLE,j:(19)
2D-FDTD simulations of the grating structures designed using
Eqs. (17)–(19) were carried out exploring different combina-
tions of the grating apodization factor Rand etch depth e
and taking into account SOI platforms with Si layer thickness,
respectively, equal to 220 and 260 nm (both having a BOX
thickness of 2 μm), as shown in Fig. 32.
Regarding the grating implemented in the 220 nm Si thick
SOI platform, it was shown that a CE of 70% (−1.55 dB)at
λ1.55 μmcould be achieved when performing a deep etch
of 110 nm, with R0.0275 μm−1, while for the 260 nm Si
thick SOI platform, a CE of 83% (−0.81 dB)atλ1.55 μm
was theoretically demonstrated for e160 nm and
R0.025 μm−1. The latter result is close to the efficiency re-
ported in [46], although making use of an SOI platform with a
thinner silicon layer. Because of the fulfillment of the phase-
matching condition, the obtained coupling efficiencies are
also greater than the ones reported in [72,73], where similar
linear FF apodization (but with constant Λ) was employed.
Comparing the GC directionality as a function of efor both
the apodized and UGC implemented in a 260 nm Si thick plat-
form, it was shown that the apodized configuration makes it
possible to increase the GC directionality by ≈10% and, even
more interestingly, that the etch depth eyielding the maximum
directionality is significantly different between the apodized
and uniform structures, as shown in Fig. 33 in [76]. This
proved that the chosen apodization profile could influence
the resulting directionality and also that the etching depth e
cannot directly be derived from a uniform GC analysis but in-
stead has to be optimized together with the apodization profile
in order to obtain the maximum CE.
An experimental characterization of apodized grating sam-
ples fabricated in a 260 nm thick SOI platform, according to
the optimum design described in [76], resulted in an average
CE of 77.6% (−1.1dB), with the best performing device show-
ing a CE of 81.3% (−0.9dB), which currently represents the
best experimental result achieved in an SOI platform without
the use of any backreflector or overlayer.
In another recent work, Bozzola et al. [75] performed a cam-
paign of 2D-FDTD simulations, analyzing the performance of
GCs, implemented in different SOI platforms, apodized by
using a mixed approach, based on an FF linear chirp and
numerical optimization by a GA [75]. As a first step, the
authors took into consideration a 220 nm Si thick SOI plat-
form and derived the maximum CE of GCs implemented
with a linear FF chirp (and constant Λ), as a function of
the etch depth e: the best result was found to be CE
61.6% (−2.1dB) for e100 nm, assuming no constraints on
the GC minimum feature, and CE 59.4% (−2.26 dB) when
assuming a minimum feature compatible with deep-UV lithog-
raphy (100 nm), corresponding to e80 nm. Starting from
the best-performing linearly chirped GC configurations at
each elevel, a GA was subsequently applied to further improve
the achievable CE. For the unconstrained case, this led to
CE 64.7% (−1.9dB) when e100 nm, while obtaining
CE 61.9% (−2.08 dB) for e220 nm (full etch) in the
constrained situation. As a second step, the same mixed opti-
mization technique was applied considering the SOI Si thick-
ness as a variable parameter. The simulation results, obtained
after the GA optimization and without minimum
feature constraints, are shown as the red curve in Fig. 34.
The maximum CE was found to be 88.2% (−0.55 dB) for a
Si thickness of 340 nm (and a 1950 nm thick BOX) and
e200 nm. The authors then reperformed the GC optimi-
zation in a different fashion. The GCs were initially apodized
(for different values of the SOI Si thickness) using an ap-
proach similar to the one described in [46], and afterward
the obtained AGC configurations were further optimized by
a GA without feature constraints. The simulation results are
shown as the blue curve in Fig. 34, with a maximum CE of
89.3% (−0.49 dB) obtained for a Si thickness of 340 nm
(B1950 nm) and e200 nm. When the 100 nm mini-
mum feature constraint was considered, the overall CE slightly
decreased to 84.8% (−0.72 dB), as shown by the green square
in Fig. 34.
3. Multiple Etch-Depth Grating Couplers
A common characteristic of the previously described GC struc-
tures is that the etch depth (e) is the same along the whole gra-
ting; thus, a single etch step is sufficient to define all the grating
trenches in the Si waveguide. An alternative approach to enhance
Fig. 32. Maximum CE at λ1.55 μmfor the linear AGC based
on 220 nm SOI (red plot) and 260 nm SOI (blue plot). Reproduced
from [76].
Fig. 33. GC directionality, at λ1.55 μm, as a function of efor
the linearly apodized grating (green curve) and for a uniform grating
configuration (purple curve). Reproduced from [76].
Review Vol. 7, No. 2 / February 2019 / Photonics Research 219
the CE of the gratings is to design more complex grating struc-
tures, including multiple etch depths in a single grating [77–81].
In [77], for example, a uniform SOI grating implemented with a
double-level etch was proposed and analyzed from an antenna
theory point of view. A schematic cross-sectional view of the pro-
posed grating is shown in Fig. 35. By considering each grating
trench as an individual point-scatterer and defining the horizon-
tal and vertical phase delay between two consecutive trenches
as ϕhand ϕv, respectively, a constructive interference for the
upward radiated beam (and destructive interference for the
downward radiated beam) occurs if ϕhϕvπ∕2. Further,
2D-FDTD simulations showed that, if the grating is designed
satisfying this phase relation, a directionality of 97.2% and a
CE of 74% (−1.3dB)atλ1.55 μm, with a negative radiation
angle of −30°, are obtained.
A similar approach was used in [78,79] to optimize a GC
implemented in an SOI platform with S220 nm, employ-
ing a shallow etch e170 nm and a full etch e2220 nm.
By optimizing the distance between the shallow and deep
trenches, a theoretical directionality exceeding 95% was dem-
onstrated [78], and an experimental CE of 74.1% (−1.3dB)
with a 3-dB bandwidth of 52 nm was reported at λ1.55 μm
[79]. A double-level etch strategy with an “L-shape”configu-
ration was also applied in [80] to GCs implemented in an SOI
platform with S300 nm: grating samples designed with
e1150 nm and e2300 nm and fabricated using 193 nm
DUV lithography, made it possible to demonstrate an exper-
imental CE of 53.7% (−2.7dB) with a 3-dB bandwidth of
62 nm at λ1.55 μm.
Another possibility is to take advantage of the lag effect
present in the ICP-RIE process, which results in a shallower
etch for narrower trenches, to design a nonuniform multi-etch
depth GC [81]. A cross-sectional schematic of the grating struc-
ture proposed in [81], which was implemented in a SIMOX
SOI platform with Si thickness of 250 nm and BOX thickness
of 3 μm, is shown in Fig. 36.
The possibility of varying the etched trenches in two dimen-
sions made it possible to enhance the variation range of the
grating coupling strength α; the etch-depth profile was there-
fore chosen in order to match the ideal αdistribution function
reported in [59] [see Eq. (14)], whereas the distance between
each grating trench was set in order to satisfy the Bragg
condition [see Eq. (9)]. With the optimized multi-etch depth
GC configuration, a theoretical CE of 74% (−1.3dB)atλ
1.52 μmwas demonstrated, whereas experimental results
showed a CE of 64% (−1.9dB) with a 1-dB bandwidth of
43 nm. Despite the good efficiency results reported in [77–81],
it must be considered that using multiple etching steps, or
relying on the etching lag effect, can significantly complicate
the device fabrication process.
4. Metamaterial-Based Grating Couplers
As discussed in previous sections, the GC-CE is sensitive to any
etch-depth fabrication error. Moreover, if the other optical
components to be integrated on the PIC require the use of
Si channel waveguides, two different etching processes are re-
quired: one to define the grating trenches and one to define the
waveguides. A recently proposed solution is based on the use of
single-etched gratings based on the use of a subwavelength
structure or photonic crystals. An example of such a structure,
made of nano-holes and implemented in an SOI platform with
S220 nm and B2μm, was reported in [82], and its
structure is shown in Fig. 37.
A subwavelength grating can be seen as a standard grating,
where the etched trenches are replaced by a metamaterial of
refractive index nL(refer to Fig. 38), while the unetched teeth
are still characterized by a refractive index nH(refractive index
of a silicon slab, unetched). The metamaterial refractive index
nLis influenced by the fill factor FFyin the ydirection and by
the refractive index (nhole) of the hole filling material (SiO2in
the case of [82]). According to first-order effective medium
theory [38], if the period Λyin the ydirection is much smaller
than the optical wavelength λ,nLcan be expressed, in the case
of TE or TM polarized input light, by the following equations:
Fig. 34. Efficiency comparison between optimized uniform grating
designs (black line and squares) and optimized apodized grating de-
signs (red and blue lines and symbols) as functions of the SOI thick-
ness. The red curve with dots refers to apodized gratings obtained from
an FF linear chirp and genetic algorithm optimization, while the blue
curve with triangles refers to those obtained by optimizing the struc-
ture reported in [46]. The theoretical CE of the record-efficiency de-
sign for a 1D-GC with an Al backreflector ([63]) is denoted with a
dashed horizontal black line. The CE of the apodized design with
deep-UV lithographic constraints (minimum feature =100 nm) is
denoted with an open green square. Adapted from [75].
Fig. 35. Schematic cross-section of a UGC implemented with the
double-etch technique reported in [77]. ϕhand ϕv, respectively, re-
present the horizontal and vertical phase shift between two consecutive
grating trenches.
Fig. 36. Schematic representation of the multi-etch GC proposed
in [81], where the lag effect of ICP-RIE is exploited.
220 Vol. 7, No . 2 / February 2019 / Photonics Research Review
nTM
LFFyn2
hole 1−FFyn2
Si1∕2,(20)
1
nTE
L
FFy
n2
hole
1−FFy
n2
Si 1∕2
:(21)
Once the metamaterial refractive index nLhas been set by
choosing a proper value of FFy, the effective refractive index
along the xdirection can still be calculated using Eq. (3)
(replacing nEwith nL), and the grating diffraction properties
can still be evaluated using the Bragg equation [Eq. (9) where
the period Λxalong the xdirection is considered]. In [82]a
UGC (Λx610 nm) was implemented in a 220 nm Si thick
SOI and optimized for TE light polarization: for a nano-hole
diameter of 200 nm and Λy450 nm, an experimental CE of
34% (−4.69 dB) was achieved, with a 3-dB bandwidth of
40 nm at λ1.46 μm. However, a strong Fabry–Perot
(FP) parasitic oscillation was observed on the output spectrum,
indicating a large (≈9%) residual in-waveguide reflectivity. In
order to reduce the in-waveguide reflectivity, two approaches
have been proposed: (i) employing longer Λyor (ii) reducing
the nano-hole diameters. Fully etched subwavelength gratings
with 143 nm (diameter) nano-holes, fabricated by e-beam
lithography in an SOI platform with S250 nm and
B1μm, made it possible to experimentally demonstrate a
CE of 42% (−3.77 dB)atλ1.55 μm, with a 1-dB
bandwidth of 37 nm, while simultaneously reducing the
in-waveguide reflectivity to 0.9% [83].
On the other hand, a different type of uniform subwave-
length structure was reported in [84], aiming at a large 1-dB
coupling bandwidth. In general, a reduction of the grating ef-
fective refractive index neff is required in order to increase the
bandwidth, but this is difficult to achieve in standard GC, as nO
[with reference to Eq. (3)] is set by the native SOI Si thickness
(this aspect will also be treated in detail in Section 4.C.5). To
overcome this issue, a GC based on the use of Si nano-pillars
(instead of Si nano-holes) was designed: nL(with reference to
Fig. 38) was fixed to be equal to the refractive index of the
cladding (SiO2), whereas nHcould be easily reduced by con-
trolling Λy. The grating was implemented in a 340 nm Si thick
SOI platform (B2μm), and the experimentally assessed CE
was found to be 27.5% (−5.6dB)atλ1.583 μm, with a 1-
dB bandwidth of 73 nm. By means of numerical simulations,
the authors also showed that the CE could be increased up to
46% (−3.4dB), with a 1 dB bandwidth of 86 nm, using an
optimized BOX thickness of 1.64 μm.
In order to increase the CE of a subwavelength grating, the
grating structure can be apodized similarly to shallow etch stan-
dard GCs, thus changing the effective refractive index neff of
each scattering cell (composed by a metamaterial “trench”
and a Si tooth) along the direction of light propagation inside
the grating. In [85–87], where rectangular nano-holes were
considered, this was achieved by fixing Λyand varying FFy
along the xdirection (according to the Cartesian reference sys-
tem reported in Fig. 38), thus varying neff in a controlled man-
ner. At the same time, the length Λxof each scattering unit was
varied along xin order to satisfy the Bragg condition in each
section of the grating. In [86], this method was applied to a
260 nm Si thick SOI platform (B2μm), optimizing the
device for TM light coupling and varying FFyin order to
achieve a linear variation of neff (from 3.22 to 2.16) along
the grating. The minimum feature size of the grating was set
to 100 nm, to guarantee compatibility with 193 nm deep-UV
lithography. The theoretical CE of the subwavelength grating
was found to be 52.5% (−2.8dB), with a 1-dB bandwidth of
35 nm, whereas measurements on fabricated samples showed a
43% (−3.7dB) CE and a 1-dB bandwidth of 40 nm [86]. In
[87] instead, the square hole grating was implemented in a
220 nm SOI platform (B3μm) and optimized for coupling
TE light polarization. A linear apodization of the hole size was
applied, and an Al backreflector was deposited (by Si substrate
back-etch, using sulfur hexafluoride and electron beam evapo-
ration) under the BOX layer. This configuration made it pos-
sible to demonstrate a theoretical CE of 85.7% (−0.67 dB)at
λ1.55 μm(for a diffraction angle θair of 27°), and the ex-
perimental results showed a CE of 85.3% (−0.69 dB) with a 3-
dB bandwidth of 60 nm that, to the best of our knowledge,
represents the highest CE ever obtained in a 220 nm Si thick
SOI platform-based device.
A different design approach was employed in [64,88]to
apodize the neff of a grating based on the use of a triangular
hole lattice and implemented in a 250 nm Si thick SOI plat-
form (with B1μm). In this case, the grating hole dimen-
sions were not varied linearly, but they were engineered in order
to shape the grating coupling strength αaccording to the ideal
distribution reported in [59], suitable to produce a Gaussian
Fig. 38. (Left) Top view of the grating based on a nano-hole array.
(Right) 2D model of the waveguide grating with a nano-hole array
based on a slab structure.
Fig. 37. Schematic representation of a nano-hole GC designed for a
coupling angle θof 8°.
Review Vol. 7, No. 2 / February 2019 / Photonics Research 221
profile of the diffracted beam. A CE of 67% (−1.74 dB) was
experimentally demonstrated, with a 3-dB bandwidth of 60 nm
[88]. The grating performance was then increased by including
an Al backreflector, making it possible to achieve a high CE of
87.5% (−0.58 dB) and a 3-dB bandwidth of 71 nm [64].
5. Large-Bandwidth Grating Couplers
Besides CE, another important parameter, which defines the
performances of SOI gratings, is the coupling bandwidth,
which is usually characterized in terms of the 1-dB or the
3-dB points. The 1-dB bandwidth of the coupling efficiency
between an SOI grating and an SMF with NA ≪1can be
expressed by the following equation [89]:
Δλ1dB ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2ln10
5
rNA λ0cos θ1
neff −n1sin θ1
,(22)
where λ0is the central coupling wavelength, neff is the grating
effective refractive index, n1is the cladding refractive index, and
θ1is the diffraction angle in the cladding. It must be noted that
Eq. (22) returns a slightly overestimated value of Δλ1dB,asitis
derived assuming the diffracted and the SMF mode to have a
Gaussian distribution with the same width. In real cases, the
diffracted mode profile is an exponentially decaying distribu-
tion, if a UGC is designed, or a general approximation of
a Gaussian distribution of an apodization profile is used.
However, although a small bandwidth improvement can be ob-
tained by increasing the diffraction angle θ1, the only way to
significantly increase the 1 dB bandwidth is by reducing the
grating effective refractive index neff . This can be achieved us-
ing a dielectric material having a refractive index lower than
that of Si to implement the grating. A possible material choice
is represented by silicon nitride (Si3N4), which has nSi3N4
2.0and is CMOS-compatible. The effective refractive index
reduction also makes it possible to achieve the optimal grating
coupling strength when full-etched trenches are considered,
thus greatly simplifying the fabrication process. In [89], for
example, a uniform full-etched GC was demonstrated in a
400 nm thick Si3N4platform, reporting a CE of 38%
(−4.2dB)atλ1.57 μmand a 1-dB bandwidth of 67 nm.
Similar to the case of Si gratings, the directionality and hence
the CE of a Si3N4based GC can also be improved by employ-
ing an embedded backreflector. In [90], for example, a CE of
55% (−2.6dB) with a 1-dB bandwidth of 53 nm was demon-
strated realizing a UGC in a 400 nm thick Si3N4platform with
an amorphous-silicon/silicon-dioxide DBR in the BOX.
It is interesting to note that silicon nitride is also nowadays
emerging as a promising photonic platform for nonlinear ap-
plications [91,92], thanks to the possibility of implementing
low-propagation loss and TPA (two photon absorption) free
waveguides. In this case, thicker Si3N4layers are usually re-
quired in order to achieve broadband zero-dispersion operation,
with typical layer thickness in the order of 700 nm [93] or even
thicker [94]. Implementing efficient GCs in a thick waveguide
layer can be quite challenging, as the excitation of higher-order
vertical Bloch modes inside the grating usually leads to a drastic
deterioration of the CE. This problem was mitigated in [95],
where a UGC with a peak CE of 43% (−3.7dB) and a 1-dB
bandwidth of 54 nm was implemented in a 700 nm thick
silicon nitride-on-insulator (SNOI) platform. In that case, to
prevent the excitation of high-order Bloch modes, the access
waveguide was etched down to the same level as the grating
trenches, and an inverse taper was used to connect the
Si3N4strip waveguide and the grating section, as shown in
Fig. 39. Moreover, to reduce the taper-induced mode-conver-
sion loss while at the same time keeping a compact footprint for
the device, a focusing configuration was employed, so as to
taper the access waveguide to a smaller lateral size at the focal
position.
In the case of Si3N4-based GCs, one of the drawbacks re-
lated to the relatively low refractive index contrast (Δn≈0.5)is
the difficulty to obtain high-directionality devices. This issue
was recently faced in [96], where the authors proposed the
use of a two-level etched structure, therefore resulting in a
“blazing effect,”allowing them to demonstrate an experimental
CE of ≈71% (−1.5dB) and a 3-dB bandwidth of 60 nm,
which constitutes a record performance for structures with only
Si3N4deposited on the oxide layer.
Another interesting strategy to simultaneously increase both
the GC-CE and bandwidth of a Si3N4structure was proposed
in [97]. Here, as shown in Fig. 40, a dual-level grating was
implemented, with a set of 400 nm thick fully etched Si3N4
grating teeth placed above an aligned set of thin partially etched
Si teeth. Due to the close proximity of the two different sets,
which are spaced by a 135 nm SiO2layer, the grating behaves
as a collection of composite Si3N4∕Si teeth, breaking the
natural vertical symmetry of the structure and making it
possible to achieve constructive interference for the upward dif-
fracted radiation and destructive interference for the downward
diffracted radiation. The performance of the device was then
finally optimized by apodizing the GC structures by means
of 2D-FDTD simulations; as a result, an experimental CE
as high as 74% (−1.29 dB) was demonstrated, with a record-
high 1-dB bandwidth of 80 nm.
A dual-level Si3N4-on-SOI GC was also theoretically pro-
posed in [98], and its structure is shown in Fig. 41. Compared
with [97], a thicker SiO2separation layer (H1.68 μm) was
placed in between the 400 nm thick Si3N4layer, where the
actual fully etched apodized grating was implemented, and
the 220 nm thick Si layer, where a fully etched grating was
designed to act as a bottom reflector. By careful optimization
Fig. 39. (Top) Cross-sectional schematic of the Si3N4GC proposed
in [95]. The access waveguide thickness (t) is equal to thickness of the
native Si3N4layer (tSN ) minus the etch depth (tg). (Bottom)
Schematic top view of the focusing GC structure proposed in [95],
where an inverse taper was used to connect the grating section and
the Si3N4strip waveguides. The optimized geometrical parameters
are Wt150 nm,We4μm,Wg900 nm, and Lt20 μm.
222 Vol. 7, No . 2 / February 2019 / Photonics Research Review
of the Si-grating period and fill factor (WGR 470 nm and
PGR 870 nm, with reference to Fig. 41) a reflectivity higher
than 92% was achieved (therefore comparable with that of a
DBR with two pairs of Si∕SiO2layers), resulting in an overall
theoretical CE of 81.7% (−0.88 dB)atλ1.55 μm, with
a 1-dB bandwidth of about 60 nm.
Concerning GCs implemented in standard 220 nm Si thick
SOI platforms, a recent theoretical work showed the possibility
of obtaining larger coupling bandwidth by considering fibers
with an MFD different from the typical 10.4 μm, which is
characteristic of SMFs (see also Section 2)[99]. The authors
demonstrated that, when considering uniform gratings with
a large footprint (therefore designed for high MFD beams),
the spectral behavior exhibits a Lorentzian lineshape, originat-
ing from the quasi-guided photonic modes of the correspond-
ing photonic crystal slab, while for small-footprint gratings, a
Gaussian lineshape is observed, originating from the k-vector
spreading of the incoming Gaussian beam. This means that the
coupling bandwidth increases when reducing the input beam
MDF. The performances of different AGCs, optimized by
means of a particle-swarm algorithm, were then explored con-
sidering different values of the input beam MFD (10.4, 8, 6,
and 4 μm). This comparison allowed the authors to observe
three main effects.
•Impact on CE. The maximum CE slightly increases when
the MFD is increased, and it is equal to ≈64% (−1.94 dB),
≈61% (−2.15 dB), ≈60% (−2.22 dB), and ≈57% (−2.44 dB)
for an MFD of 10.4, 8, 6, and 4 μm, respectively.
•Impact on the coupling bandwidth (BW). The BW ob-
tained considering the maximum CE configuration increases
when the MFD is decreased (≈40, 55, 65, and 85 nm for
an MFD of 10.4, 8, 6, and 4 μm, respectively).
•BW-CE trade-off. Thanks to the reported data, once the
application requirement on the BW (or CE) is known, the MFD
maximizing the CE (or BW) parameter can be evaluated, thus
allowing for proper system design. In particular, it is shown that a
relatively small decrease of the required CE [e.g., from 64%
(−1.94 dB) to 54% (−2.68 dB)] allows one to broaden the
AGC BW by more than a factor of 2 (from 40 to 100 nm).
D. Grating Couplers for Vortex Modes
Optical vortex beams deliver information through optical an-
gular momentum (OAM). This has attracted considerable at-
tention in various fields, including optical and quantum
communications [100–102], thus inducing scientists to study
efficient ways to couple vortex beams to standard silicon wave-
guides (and vice versa). The grating groove design (size, posi-
tion of each single grating tooth) for a specific vortex
topological charge (χ) case, considering a focusing geometry,
can be calculated by equating the phase of the focused mode
with the phase of the inward optical vortex beam [103], using
the following phase-matching equation:
Zr
0
βeff u,ϕdu2πmχarctanycos θ
xky sin θ,(23)
where kis the cladding wavenumber and mis a specific grating
tooth, while rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
x2yR2
pand ϕarctanx∕yR
represent the polar coordinates from the origin to the feed focus.
It is interesting to note that, in the case of χ0, Eq. (23) leads
to the design of a focused UGC, while if χ≠0the same equation
produces elliptically distorted patterns, as shown in Fig. 42.
To date, the only available demonstration of such a device
has been proposed, fabricated, and measured by Nadovich et al.
[103]. The authors presented the design of forked GC in SOI
technology, for a vortex topological case up to χ2.
Although the focus of this work was on the design strategy
rather than the performance of the actual device, the authors
estimated a CE of 31.6% (−5dB), anticipating that most of the
losses come from substrate leakage and optical mode matching.
E. Grating Couplers for Higher-Order Modes
In the last decade, several discussions on how to increase the
current network capacity have been started. To tackle this issue,
researchers have proposed the use of the so-called spatial
division multiplexing (SDM) technology, in which data are
allocated on different fiber spatial modes, thus increasing the
Fig. 40. Perspective representation of the Si3N4-on-SOI dual-level
GC reported in [97]. Reproduced from [97].
Fig. 41. Cross-sectional schematic of the Si3N4-on-SOI dual-level
GC reported in [98]. Reproduced from [98].
Fig. 42. GC forked design for vortex beam optical mode.
Reproduced from [103].
Review Vol. 7, No. 2 / February 2019 / Photonics Research 223
information throughput of a single optical fiber [104].
Recently, practical SDM implementation on silicon photonics
has been proposed [105,106], thus revealing the need for
efficient fiber-to-chip interfaces for higher-order modes.
Specifically, the scientific community has revealed the necessity
to perform two different operations involving higher-order
modes: (i) to interface a higher-order fiber mode to a higher-
order waveguide mode and vice versa and (ii) to interface a
higher-order fiber mode to the fundamental waveguide mode
and vice versa. Although SDM has been discussed for several
years, the topic discussed in this section is relatively new, and
only a few implementations of GCs for higher-order modes
have been proposed to date.
1. Higher-Order Fiber Mode to Higher-Order Waveguide
Mode Coupling Devices
The functionality of devices falling in this category is to
efficiently couple a fiber optical mode to a silicon photonic
waveguide, irrespectively from the spatial mode occupied by
the light beam. A first demonstration of such a device was
proposed in [107].
The authors of this paper showed that higher-order wave-
guide modes can be directly coupled to their respective optical
fiber counterparts, using a standard GC (i.e., designed for single
mode coupling). In Fig. 43(b), the authors show the effective
index of three different waveguide modes (TE00 ,TE10, and
TE20) as a function of the waveguide width. It is clear that,
when the waveguide width is greater than 10 μm, no significant
differences can be appreciated in terms of neff . By exploiting
this, the authors anticipated that GCs already designed for sin-
gle-mode operation can be used for higher-order modes, given
that the width of the grating section is greater than 10 μm [see
Fig. 43(b)]. However, no experimental demonstration of such a
device has been reported so far; thus, additional data and real
implementations of them are highly desirable.
A different implementation scheme has recently been pro-
posed in [108]. The proposed structure is shown in Fig. 44.
Note that this device cannot be operated to couple multiple
spatial modes at the same time, while it offers the functionality
of coupling the LP11 fiber mode to the TE10 waveguide mode.
As depicted in Fig. 44, the two lobes of the LP11b are coupled
to two, separate, single-mode silicon waveguides, which are
subsequently coupled to a multimode silicon waveguide by
means of a Y-junction coupler. Because the two optical paths
experienced by the two lobes of the original LP11b mode are the
same, the phase relationship between them is maintained, mak-
ing it possible to excite the TE10 mode in the multimode silicon
waveguide. A maximum theoretical CE of 50% (−3dB) has
been reported, exhibiting a 67 nm 3-dB wavelength band-
width. Fabricated samples showed, instead, a CE of 37.9%
(−4.21 dB) with a 3-dB bandwidth of 62 nm. In [109] instead,
the same authors demonstrated an improved and more compact
version of the higher-order mode GC, employing a curved
Y-junction and two short tapers having lengths equal to 20 μm.
This structure made it possible to experimentally demonstrate
an LP11b-to-TE10 CE of 42.9% (−3.68 dB) with a 1-dB band-
width of 35 nm. In this version, the GC structure was also com-
plemented with an on-chip mode converter (TE10 to TE00),
able to separate TE10 and TE00 input modes on two different
output ports. This allows the GC to simultaneously couple
LP01 and LP11 fiber modes (with a crosstalk below −15 dB),
effectively increasing its compatibility with SDM systems.
2. Higher-Order Fiber Mode to Fundamental Waveguide
Mode Coupling Devices
The purpose of devices falling in this category is the ability to
excite higher-order fiber modes, starting from a waveguide fun-
damental mode and vice versa. This problem has only recently
attracted the attention of the scientific community; therefore,
only a few studies have been published by researchers. In a re-
cently published paper [110], the authors showed four different
GC design strategies, allowing the excitation of four different
fiber modes (LP01,LP11a ,LP11b , and LP21b) from the funda-
mental mode of a standard silicon photonic waveguide. The
fundamental idea is to change the grating structures with
the aim of shaping the phase of the required fiber mode at
the output. An example of this functionality is reported in
Fig. 45. In order to break the fundamental waveguide mode
into two different lobes (as required to shape the LP11b mode),
the authors proposed to split the common grating structure
into two sections, shifted by a constant factor D, in such a
way that a phase difference of πis created between the upper
and the lower lobe. Similar structures have also been shown in
[110] for the excitation of LP11a and LP21b fiber modes. The
Fig. 43. (a) Schematic of a GC excited either with a TE00 (blue) or TE10 (red) waveguide mode. (Inset) Field plot of a scattered TE10 mode.
(b) Effective refractive indices of the first three guided modes in an SOI nanowire of height 220 nm at λ1.55 μm. (c) Cross-section of the
scattered electric field profiles of a standard GC excited from both ends with Δφ0° (blue) and Δφ180° (red). Reproduced from [107]. This
device can be used to simultaneously couple different fiber spatial modes to silicon photonic multimode waveguides.
224 Vol. 7, No . 2 / February 2019 / Photonics Research Review
theoretical CEs for LP11a,LP11b , and LP21b fiber modes were
evaluated by means of 3D-FDTD simulations and were, re-
spectively, found to be equal to 55% (−2.6dB), 51.5%
(2.9 dB), and 50% (−3dB). However, to the best of our knowl-
edge, no experimental results have been reported yet on
this topic.
5. POLARIZATION-INSENSITIVE GRATING
COUPLERS
A. 1D Grating Couplers (Standard and Metamaterial-
Based)
Silicon photonic waveguides typically exhibit strong birefrin-
gence, which does not allow simultaneous optimization of
the CE of a simple 1D-GC for the orthogonal polarization
states of incoming light. In fact, as nTE
eff ≠nTM
eff , two different
grating periods Λwould be required in order to diffract the
polarization states at the same central wavelength and diffrac-
tion angle, indicated as λand θ1in Eq. (9). A number of
solutions have been proposed by the scientific community.
In Ref. [111], two nonuniform GC structures were proposed:
the first one was obtained by the geometrical intersection of
two UGC structures, whose periods Λwere chosen to couple
the TE and TM light polarization, respectively, while the
second one was obtained by the geometrical union of the same
TE and TM UGC designs. A schematic of the design strategy
reported in Ref. [111] is shown in Fig. 46.
The design technique proposed in [111] was implemented
in a standard SOI platform (S220 nm and B2μm) and
made it possible to demonstrate a theoretical CE of 20.4%
(−6.9dB) for the intersection grating and a CE of 29%
(−5.4dB) for the union grating, when using an etch depth
of 60 and 100 nm, respectively. The theoretical PDL, which
is defined as the difference CETE −CETM, was theoretically as-
sessed to be 0.57 dB for the intersection grating and 0.5 dB for
the union grating. A proof-of-concept sample of the intersec-
tion grating was also fabricated, employing 248 nm DUV
lithography to define the grating pattern. As some of the teeth
of the ideal intersection grating were smaller than the minimum
feature size allowed by this type of lithography (180 nm), they
were simply removed without being replaced from the mask
design, even if this resulted in a 1 dB reduction of the TE peak
CE with respect to the ideal design. Experimentally, maximum
values CETE 16.6% (−7.8dB) and CETM 15.9%
(−8.0dB) at the central wavelength λc1.55 μmwere dem-
onstrated, with a PDL lower than 0.8 dB within a 20 nm range
centered around λc.
A better performance in terms of average CE was instead
reported in [112], where a uniform 1D-GC was designed so
as to couple TE and TM light polarizations in opposite direc-
tions into the same SOI waveguide, as shown in Fig. 47.
The GC proposed in [112] is based on an SOI platform hav-
ing a 250 nm thick Si layer, a 3 μm thick BOX, and an Al back-
reflector deposited underneath the BOX layer, as in [63]. The
proposed GC theoretically allows for achieving CETE 85.1%
(−0.7dB) and CETM 77.6% (−1.1dB); however, the maxi-
mum CETE and CETM values are achieved considering two dif-
ferent positions of the optical fiber with respect to the grating.
When the position of the fiber is fixed, as it practically happens
when packaging a PIC, a compromise position between the TE
and TM hot spots must be chosen, and the theoretical CE drops
to 49% (−3.1dB)atλ1.55 μmfor both polarizations.
Measurements carried out on fabricated samples showed, in-
stead, a maximum CETE of 63% (−2dB)atλ1.559 μm,
with a 1-dB bandwidth of 29 nm, and a maximum CETM of
58.9% (−2.3dB)atλ1.551 μm, with a 1-dB bandwidth
of 23 nm. When fixing the fiber position, however, a lower
CE of 41.7% (−3.8dB) was found for both polarizations.
Another possibility for implementing a polarization-
insensitive 1D-GC, is to use an SOI platform with a thick Si
layer, which allows the propagation of TE and TM 1st order
Fig. 44. LP11 GC structure proposed by [108]. Reproduced from
[108].
Fig. 45. Structure proposed by [110] to excite the LP11b mode from
a silicon photonic single-mode waveguide.
Fig. 46. Schematic of the designed strategy proposed in [111]to
obtain a polarization insensitive 1D-GC. Grating (a) is obtained by
the geometrical intersection of two different UGCs, having a grating
period optimized for TE and TM light polarization, respectively.
Grating (b) is obtained by the union of the same two UGCs.
Reproduced from [111].
Review Vol. 7, No. 2 / February 2019 / Photonics Research 225
modes. In [113], for example, a 400 nm thick Si layer was used,
and, thanks to a proper choice of the grating period Λ,aUGC
was demonstrated, which was able to simultaneously couple the
two orthogonal SMF light polarization states to the second dif-
fraction order (m2)ofTE00 mode and to the first diffraction
order (m1)ofTM10 mode, respectively, with a CE higher
than 50% (−3dB) for both polarization states. However, it is
important to underline that the adoption of this technique re-
quires all the other integrated optical components (such as
waveguides, filters, and modulators) to be optimized for multi-
modal operation, which is not trivial and is impractical in many
scenarios.
Finally, another possibility for designing a polarization-insen-
sitive 1D-GC is based on the use of subwavelength gratings
[114]. The structure of the proposed GC is similar to the
one previously shown in Fig. 37 and by properly exploiting
1st order [see Eqs. (20)and(21)] or 2nd order effective medium
theory equations. It is possible to identify a range of FFyallowing
one to obtain nL,TM >n
L,TE (see Section 4.C.4). As this condi-
tion is symmetric with respect to what happens in the Si tooth
section (where nH,TM <n
H,TE), for any value of FFybelonging
to the previously identified range, a corresponding value of FFx
can be identified, which makes the effective refractive index of
the GC identical for both polarizations (nTE
eff nTM
eff ). The exact
values of FFxand FFyare selected as a trade-off between the
coupling strength αfor the TE and TM modes. Thanks to this
design strategy, the subwavelength grating structure proposed in
[114] was implemented in an SOI platform (S340 nm and
B2μm), setting FFy0.65,FFx0.39,andΛx750 nm.
Considering a central coupling wavelength λof 1.55 μmand
a diffraction angle of 15°, a theoretical CE of ≈40% (−4dB)
for both the TE and TM modes was obtained by numerical sim-
ulations, which could be improved up to 64% (−1.94 dB)by
including a DBR reflector under the GC.
B. 2D Grating Couplers
As discussed in Section 4.A, 1D-GCs generally exhibit strong
polarization sensitivity and offer optimum CE for only one
polarization state. While certain 1D-GC designs, employing
subwavelength features, can overcome such polarization-
dependence issues, 2D-GCs offer an alternative solution for
fiber-to-PIC coupling, where the fiber mode has an unknown
or unstable polarization state. In its simplest form, a 2D-GC
can be viewed as the superimposition of two orthogonally ori-
entated 1D-GCs, each of which couples light from the fiber
mode into a TE-polarized waveguide mode on the surface
of the SOI-PIC, as shown in Fig. 48(a) [115,116]. In this
scheme, any arbitrary polarization state of the input fiber mode
can be projected onto the 2D-GC as a pair of orthogonal
modes, each of which is TE-polarized with respect to one of
the 1D-GCs. As the polarization state varies, the fraction of
power coupled by an individual 1D-GC changes, but the over-
all power coupled to both 1D-GCs is almost constant, as illus-
trated in Fig. 48(b).
In practice, the superposition of two 1D-GCs to create a
2D-GC results in a structure that is highly analogous to a pho-
tonic crystal array, with cylindrical features partially or fully
etched into the SOI layer of the PIC [118]. The 2D-GC offers
high CE, when the band structure of the photonic crystal array
(PCA) intersects the “light line”of the fiber mode at the target
wavelength [117]. For an arbitrary polarization state, some frac-
tion of the power from the input fiber mode will be coupled
into both “arms”of the 2D-GC (see Fig. 49). This figure also
shows that, while the PCA structure is made up of a strictly
Fig. 47. Cross-sectional schematic of the GC proposed in [112].
Fig. 48. (a) Schematic of a 2D grating-coupler (2D-GC) on the
SOI photonic platform, showing the angle-of-incidence (θ) and the
polarization angle (φ) of the incident fiber mode. Inset shows the def-
inition of the pitch (P) and radius (R) of the partially etched cylinders
making up the 2D-GC. (b) The coupling efficiency into the two
orthogonal arms (in the xand ydirections) of the 2D-GC as a function
of the polarization angle. Adapted from [117].
226 Vol. 7, No . 2 / February 2019 / Photonics Research Review
orthogonal array of etched cylinders, the coupled modes in the
SOI layer propagate at a small angle (3.3°) with respect to the
2D-GC symmetry axes. This angular offset is due to a refrac-
tion effect from the tilt angle of the fiber mode with respect to
the 2D-GC and has practical consequences for the design of
focusing couplers and efficient adiabatic tapers to guide the
light into 450 nm wide standard SOI waveguides [119].
Ideally, the total coupling efficiency (CETCExCEy)
of a 2D-GC would be entirely independent of the polarization
state of the input fiber mode polarization. However, the same
tilt angle of the fiber mode that gives rise to the slight offset in
coupled-mode propagation also creates an asymmetry in the
E-field projection on the 2D-GC. Specifically, the tilt angle cre-
ates a condition wherein a vertical (i.e., along the zdirection)
projection of the incident E-field exists for certain polarization
states and not others. This introduces a slight amplitude varia-
tion, termed a PDL, around the mean value of the CETof a
simple uniform 2D-GC, as shown in Fig. 50. In the specific
case of a 2D-GC designed for operation at 1.55 μm and realized
in an SOI wafer with a 220 nm thick Si layer, one finds an
average value (with respect to the different polarization states)
of CET47.9% (−3.2dB) and PDL 0.3dB(see Fig. 50).
In addition to this slight variation in coupling efficiency, the
same effect causes a slight “wobble”of 1nmin the central
wavelength of the coupling spectrum [117].
1. 2D-GCs with Optimized Etch Depth, Duty Cycle, and
Period
While 1D-GC designs can be quickly simulated and optimized
using 2D-FDTD simulations, more computationally intensive
3D-FDTD simulations are needed to design and optimize 2D-
GCs [117,120]. For many years, this led to a “performance
gap”between the CE of 1D-GCs and 2D-GCs, but full de-
vice-scale simulation of a 2D-GCs is now practical with desk-
top computers, and several optimized designs have been
identified and described. For simple uniform 2D-GCs realized
by partially etching cylindrical features into the SOI layer, the
key design parameters are the Si layer thickness (S), etch depth
(E), hole radius (R), and grating pitch (P). It is often useful to
parameterize the 2D-GC in terms of R∕Pbecause this is analo-
gous to the duty cycle in 1D-GCs, and it is often constrained by
the lithographic tolerances of the SOI-PIC fabrication. Once a
set of the three main grating parameters (S,E, and R∕P) has
been selected for evaluation, the grating pitch is tuned until the
peak of the simulated coupling spectrum aligns with the target
wavelength, e.g., 1.55 μm. Repeating this process for different
values of Eand R∕Pallows a contour plot of the 1.55 μm cou-
pling efficiency to be built up (see Fig. 51) and the optimum
design parameters to be identified. This contour plot for an
S220 nm 2D-GC has its peak at E120 nm and R∕P
0.3(R185 nm,P635 nm), which offers a 48%
(−3.2dB)CET[117]. It is interesting to note that the optimum
etch depth for a 220 nm 2D-GC is almost double that of a
1D-GC with the same Si layer thickness (70 nm).
Measurements carried out on 2D-GCs designed with the same
optimized parameters have shown an experimental CE of 42%
(−3.75 dB) with a 3-dB bandwidth of 43 nm [121].
It is possible to further improve the 2D-GC coupling effi-
ciency by increasing the thickness of the Si layer (S). In practice,
this is usually achieved by depositing a layer of p-Si on top of
the 220 nm layer (which represents a de facto standard for
Fig. 49. (a) Schematic of a 2D grating-coupler (2D-GC) with long
adiabatic taper waveguides to match the SMF28 coupled mode to the
dimensions of the SOI waveguide (i.e., 450 nm ×220 nm). (b) A large
3D-FDTD simulation (75 μm×25 μm) is used to illustrate the slight
angular offset between the direction of coupled-mode propagation and
the symmetry axis of the 2D-GC. Adapted from [117].
Fig. 50. Dependence of CEx,CEy, and CETon the input-beam
polarization angle. Adapted from [117].
Review Vol. 7, No. 2 / February 2019 / Photonics Research 227
silicon photonics companies) and then using modified selec-
tive-etch and full-etch recipes to realize the periodic features
in the 2D-GC and the taper-and-waveguide structures, respec-
tively. An optimal 2D-GC considering an SOI wafer with S
400 nm (i.e., with 180 nm p-Si on top of a 220 nm Si layer)
makes it possible to achieve CET65% (−1.9dB)[120]
when E291 nm,R167 nm, and P584 nm. This
“thick”400 nm Si-layer design offered a 1-dB bandwidth of
38 nm and a PDL 0.3dB, similar to that of a standard
2D-GC design realized using S220 nm.
2. 2D-GC Designs with Fully Etched SOI
There have been proposals to simplify the 2D-GC fabrication
process by eliminating the need for the partial etch used to real-
ize the periodic features of the coupler, and particular attention
has been paid to the possibility of exploiting the “lag effect”of
lithographic dry-etching processes. As the etching rate depends
on the feature size, the “full etch”used to define the strip wave-
guide structures on the SOI-PIC does act as a partial etch on
very small (i.e., subwavelength) features patterned onto the
2D-GC, as shown in Fig. 52(a).
While an individual subwavelength feature is too small to
provide useful coupling, clusters of these features can form
an analogue of the etched cylinders of the simple 2D-GCs de-
scribed in Section 5.B.1. Tuning the radius (r) and spacing (d)
of the “intra-cluster”subwavelength features allows their effec-
tive etch depth (from the “full etch”process) to be controlled
and the effective index of the cluster to be engineered, as shown
in Fig. 52(b). Tuning the pitch of the “inter-cluster”spacing
[see Fig. 52(c)] then allows the coupling spectrum to be cen-
tered at the target frequency. For the 2D-GC design in Fig. 52,
clusters of five identical cylinders are arranged in an “X”con-
figuration, aligned with respect to the symmetry axes of the 2D-
GC [122]. A parameter sweep made it possible to identify the
optimum design for a 2D-GC operating at 1.55 μm (consid-
ering an SOI wafer with S220 nm)asr75 nm,
d230 nm, and P650 nm, where the effective etch
depth of the subwavelength features is E150 nm. This
2D-GC design provided a CE 26.3% (−5.8dB), and it
was relatively robust to realistic fabrication tolerances (1nm
change in hole radius gives 1nmin wavelength).
3. 2D-GC Designs for Reduced Polarization-Dependent
Loss
Clusters of subwavelength features can also be used to reduce
the PDL of 2D-GCs, as shown in Fig. 53(a).Zouet al. dem-
onstrated that asymmetric cluster configurations allow for a
first-order compensation of the previously cited asymmetry
in the E-field projection on the 2D-GC [123]. As shown in
Figs. 53(b) and 53(c), a cluster of five subwavelength cylinders,
arranged in a diamond shape can significantly reduce the PDL
exhibited by a 2D-GC realized in an SOI platform with
S220 nm. Thanks to a sweep of the intracluster and inter-
cluster dimensions, it was possible to tune the 2D-GC peak
to 1.55 μm while obtaining a CET32% (−5dB) and
PDL 0.25 dB over the entire C band [123].
2D-GCs based on the symmetric patterning of subwave-
length features are particularly well suited to single-ended pho-
tonic applications, i.e., where the fiber mode is only incoupled
to the SOI-PIC but not outcoupled. In applications that use
end-to-end coupling (i.e., a pair of identical 2D-GCs on the
PIC for in- and outcoupling to a fiber), an alternative ap-
proach can be used [124]. Here, a πphase shifter was intro-
duced onto one of the two arms of the input 2D-GC (see
Fig. 54), which made the polarization state of incoupled and
outcoupled modes be orthogonal, regardless of the initial
polarization state.
Therefore, the end-to-end coupling efficiency (i.e., the
product of the in- and outcoupling) always included a balanced
ratio of the unequal s- and p-polarization projections of
the fiber mode onto the 2D-GC, due to the small fiber tilt an-
gle. For perfectly fabricated 2D-GCs and perfectly aligned in-
put and output fibers, the πphase shifter could provide perfect
cancellation of the PDL. If the fiber-to-PIC system has residual
asymmetry from manufacture and assembly tolerances, then
thermal tuning of the phase shifter can help recover a low PDL
condition, though this can be impractical for real devices
Fig. 51. CE contour plot at λ1.55 μmas a function of Eand
R∕Pfor a 2D-GC realized on the SOI platform with S220 nm.
The optimum performance corresponds to E120 nm and
R∕P0.3. Reproduced from [117].
Fig. 52. (a) Schematic showing that a nominally “full”etch provides
only partial etching for small sub-λfeatures. (b) Layout of the sub-λ
cluster that acts as a unified scattering site for the 2D-GC. (c) The
periodic layout of the sub-λclusters to create the 2D-GC.
228 Vol. 7, No . 2 / February 2019 / Photonics Research Review
because it transforms what should be a passive element into an
active element requiring a power source for proper operation.
The principal drawback of the πshift approach is that it re-
quires the introduction of a 100 μm long phase-shifter on the
PIC, and it is only useful for applications that use symmetric
incoupling and outcoupling.
4. 2D-GC Designs with Metal Backside
As already shown when discussing 1D-GC (see Section 4.C.1),
adding a bottom reflector to a 2D-GC can substantially in-
crease its CE. Here, a reflecting element should be deposited
on the bottom surface of the BOX layer, using a process that
is CMOS-compatible. The following is one such simplified
process flow, as graphically illustrated in Fig. 55(a):
(i) the GC is etched in the SOI layer using standard lithog-
raphy techniques,
(ii) a TOX layer is added for handling and process flow,
(iii) a dry/wet selective etch of the Si substrate is used to open
a high-aspect-ratio trench to the underside of the BOX layer,
(iv) a 20 μm×20 μmpatch of aluminum is deposited onto
the BOX.
Alternative approaches, substituting the aluminum for mul-
tilayer dielectric thin films (as discussed for 1D-GC previously),
to create a DBR, are also possible. The same contour plot, as
described in Section 5.B.1, can be used to identify the opti-
mum design parameters of an SOI 2D-GC with a bottom
reflector, as shown in Fig. 55. In these systems, clearly the
BOX thickness is an important design variable because it
controls whether the reflected mode will constructively (or
destructively) interfere with the input mode in the Si layer
of the coupler. By considering an optimum BOX thickness
(B2175 nm) an 80% CE (−0.95 dB) at 1.55 μm was re-
cently reported [120], by setting E80 nm,S160 nm,
R209 nm, and P696 nm.In[125], instead an experi-
mental CE of 66.1% (−1.8dB) with a 1-dB bandwidth of
32 nm was reported, for a 2D-GC with a bonded gold mirror.
6. PACKAGING TECHNIQUES
This section describes and analyses different photonic packag-
ing techniques, which are fundamental to realizing commer-
cially scalable photonic products. Key challenges include
identifying fast, reliable, high-yield, and low-cost packaging
techniques that can deliver stable and highly efficient optical
connections between the photonic chip and external environ-
ment. Single-mode fiber-to-chip coupling is a fundamental
component of photonic packaging (see Section 2). Many ap-
plications require multichannel fiber-to-chip coupling, and this
is most easily achieved using an FA, in which several fibers are
precisely aligned in V-grooves etched into a glass or silicon
block, with a pitch of 127 or 250 μm, dictated by the fiber
cladding diameter and polished to a common optical facet.
This approach has been used to package an opto-MEMS chip
Fig. 53. (a) Periodic layout of the asymmetric clusters used to realize
a 2D-GC with reduced polarization dependent loss (PDL). (b) and
(c) Detail of the tuned clusters, showing the asymmetric dSand dP
spacing of the different subcylinders etched into the SOI layer. The
optimum design parameters are dP250 nm,dS360 nm,
E70 nm,R200 nm, and P612 nm.
Fig. 54. Schematic of an end-to-end photonic-circuit with an iden-
tical 2D-GC for both the input and output optical interconnects.
(a) In the initial scheme, the input mode and the output mode both
have the same polarization angle, i.e., φin φout, so the end-to-end
transmission exhibits twice the polarization dependence of a single
2D-GC, i.e., PDLφin·PDLφout PDL2φin. (b) In the πshift
scheme, the polarization states of the input and output modes are
rotated such that the maximum and minimum of PDLφinfrom
the input 2D-GC are anticorrelated with PDLφout, so that
PDLφin·PDLφout 0.
Review Vol. 7, No. 2 / February 2019 / Photonics Research 229
with 136 optical channels [126]. In order to reduce the FA
footprint on the photonic chip and improve the form factor
of the packaged device, it is possible to use total internal reflec-
tion from a suitably polished fiber surface to direct the beam
into a GC at the required angle.
Several factors determine the CE between the PIC and fiber
in a packaged photonic device:
•modal overlap,
•presence of reflections and interferences,
•accuracy of the positioning of the elements and ability to
preserve the alignment during packaging as well as during
normal working conditions.
The first two points can be addressed by designing appro-
priate couplers and using suitable coatings and index-matching
epoxies. The third point is a critical step in the packaging of the
device and is a major contributor to the cost and yield of pho-
tonic devices [127]. Figure 56 compares measured alignment
tolerances of a GC (designed for 10.4 μm MFD) and an
EC (with 3.5 μm MFD) made with appropriate fiber types,
i.e., standard SMF and UHNA, respectively.
For multichannel coupling between an FA and array of on
PIC coupling structures, the accurate alignment of the “roll”
angle, i.e., the orientation in the plane of the facet, is critical
to ensuring high coupling efficiency and low channel-to-chan-
nel variation. The following equation estimates the roll-related
displacement between the different fiber channels, assuming
that the first channel is perfectly aligned:
dN−1Ptan θ,(24)
where Nis the number of channels in the FA, Pis the pitch,
and θis the relative angle between the PIC and FA facet.
In practical situations, most of the displacement occurs along
the yaxis, with an xaxis displacement accountable for <0.1%
of the total. Naturally, a pitch reduction always benefits the role
tolerance, so state-of-the-art waveguide arrays to fiber trans-
poser (WAFT) arrays, with a channel pitch of 15 μm, offer
particularly relaxed angular tolerances. The manufacturing
tolerances of the FAs must also be taken into consideration,
as they introduce an additional penalty to both the CE and
channel-to-channel variation. For a two-channel array, where
the first core defines the origin of the array, the second core
has an in-plane alignment tolerance of 1.4 μm, as indicated
by the red shaded area in Fig. 56. The additional losses arising
from this alignment tolerance are 0.6 dB for SMF and 2.1 dB
for UHNA fibers, respectively. The channel-to-channel varia-
tion, alignment tolerances, and differences between coupling
array designs are the subject of continuous further research and
optimizations.
A. Alignment Procedures
Photonic packages must provide a suitable, stable mechanical
support for the secure optical connection as well as DC and
high-speed electronic connectivity and, in many cases, a thermal
stabilization component, such as a thermo-electric cooler
(TEC). Due to their tight spatial and angular tolerances, the
fiber-to-PIC coupling step should be the last assembly step in
practical photonic packages [128]. Before a systematic optimi-
zation of the fiber-to-PIC coupling is carried out, it is necessary
Fig. 55. (a) Schematic of the wafer-level postprocessing steps used
to deposit a metal (or DBR) bottom reflector beneath a 2D-GC to
enhance the fiber-to-PIC CE. (b) Contour plot of 1.55 μmCEas
a function of etch-to-thickness ratio (E/S) and radius-to-pitch ratio
(R/P) for a 160 nm thick SOI-PIC with a BOX layer thickness of
2175 nm. Adapted from [120].
Fig. 56. Alignment tolerances for (a) 10.4 μm MFD GC and
(b) 3.5 μm MFD EC.
230 Vol. 7, No . 2 / February 2019 / Photonics Research Review
to find a “first light”condition, often with the FA >100 μm
away from the PIC, to exploit the larger spot sizes that result
from the divergence of the fiber mode. Figure 57 shows a
color map of the power collected by the output fiber during
an x–yscan in a 30 μm×30 μmarea. In Fig. 57(a), the fiber
is far away from the PIC, making the signal weak and broad.
It is, however, visible enough above the background that it
can be used to refine the position, as shown in Fig. 57(b).
Epoxy plays a critical role in photonic packaging, not only
for mechanical stability, but also as an element of the optical
interface, where it can act as an index-matching layer and/or
interacts with antireflection (AR) coatings applied to the
PIC. In general, as epoxies cure, they undergo “shrinkage,”with
most UV curable optical epoxies exhibiting a shrinkage be-
tween 0.4% and 1.5% and purely “mechanical”epoxies exhib-
iting shrinkages of ≈0.08%.
Figure 58 schematically shows the effect of shrinkage on
four types of coupling, where the red arrow shows the direction
of the force during curing. Multichannel GC schemes [see
Fig. 58(a)] are weakly affected by the shrinkage because it pulls
the FA toward the PIC surface, improving the coupling and
providing a large surface area for mechanical stability. Single-
fiber ECs [Fig. 58(b)] may not have a large contact area with
the PIC; however, the fiber itself can be placed on a submount,
which can be glued to the base providing stress relief. The flex-
ibility of the fiber is sufficient, so the curing process will not
alter the coupling efficiency. The situation changes significantly
when multichannel edge coupling is considered, due to the ri-
gidity of the FA. If we utilize the previous method and secure
the interface, followed by underfilling the FA to mechanically
secure it to the base, the force exerted during curing can be large
enough to pull it out of alignment [see Fig. 58(c)]. Current
state-of-the-art packaging techniques involve placing the FA
on a transparent (glass) submount and putting low-shrinkage
epoxy between it and the PICs’own submount to secure the
alignment, as shown in Fig. 58(d). In this way, the contact area
is enlarged for a stable connection and the shrinkage is in the
direction toward the coupler, which can only improve the effi-
ciency. Figures 58(e) and 58(f) present images of the packages
realized using methods in Figs. 58(b) and 58(d), respectively.
B. Use of Microlenses
Figure 59 shows Zemax ray-tracing models of photonic pack-
ages that use micro-optic elements. Figure 59(a) shows how a
single 6 mm ball lens is used to collimate coupling modes from
six GCs, which were used in a device for noncontact pulse-wave
velocity measurement from skin above the carotid artery [129].
In this case, the micro-optics helps mediate between the
micrometer-scale of the PIC and the macroscale of human skin.
Figure 59(b) describes a method for hybrid integration of a
laser onto PICs in the form of a micro-optical bench (MOB)
[130]. A divergent beam from a laser diode (on the left, not
shown) is collimated using a 500 μm ball lens, while a second
lens with the same diameter is used to focus the beam on a
standard GC (15 μm×12 μm), allowing one to obtain a suit-
able spot size. In this arrangement, the beam is redirected at a
correct angle toward the GC by a microprism, and an isolator
Fig. 58. Schematic effects of epoxy shrinkage on coupling interface. Black elements represent the PIC, the submount is indicated in gray, the fiber
(or FA) is reported in blue, yellow is the mechanical epoxy, and green is the optical epoxy. Red arrows show the direction of the force during
shrinkage. (a) GC, notice that optical epoxy also plays a mechanical role. (b) Single fiber. (c) FA attached directly to PIC submount. (d) FA attached
to PIC submount through its own. Panels (e) and (f) show a practical realization of package designs in (b) and (d), respectively.
Fig. 57. Finding first light during alignment. (a) FA is far away from
the PIC leading to weak albeit wide signal. (b) Optimized distance
between FA and PIC leads to a strong, narrow Gaussian beam shape.
(c) Using red light to align the fiber to the EC. Scattering is observed as
the waveguide turns 90°.
Review Vol. 7, No. 2 / February 2019 / Photonics Research 231
between the ball lenses limits the effects of backreflection on
high-frequency stability of the laser diode. The submount of
the MOB offers high thermal conductivity for efficient heat
extraction from the laser diode and can be machined with suf-
ficiently high precision to ensure alignment tolerances are sat-
isfied during packaging. In [130] the simulated CE to a
standard GC was >75% and the reported −3dBalignment
tolerances were 11 μmfor the xaxis and 8μmfor the
yaxis. Comparing those values to those shown in Fig. 56(a),
we can observe that MOBs relax alignment tolerances by a
factor of ≈2.
Finally, microlenses (μ-lenses) can be used to create free-
space pluggable optical interconnects [131] (see also Fig. 60).
Microlens arrays (μLAs) can be mounted on both sides of the
connection, i.e., on the FA and on the array of GCs on the
surface of the PIC. Considering that the alignment tolerances
for this operation are as shown in Fig. 56(a), and taking into
account that, thanks to the μLA, the beam is now collimated
(and it’s waist is increased by a factor of 10), the tolerance of
the relative positioning of those elements is now relaxed by the
same factor, up to 30 μm. This means that the required tol-
erance is now sufficiently large enough to allow using standard
manufacturing techniques and also molded plastic systems.
For currently available SOI ECs (see Section 3), the typical
fiber mode area is of the order of 3μm×3μm, which leads to
submicrometer 1 dB fiber-to-PIC alignment tolerances, when
coupled to a matching UHNA fiber.
It is challenging to guarantee the required fiber-to-PIC tol-
erances for a packaged device, and practical issues, such as
shocks and vibrations or thermal expansion, can be sufficient
to have a significant impact on the fiber-to-PIC alignment.
One approach to relax these tolerances is to allow the mode
from these edge couplers to expand through natural diffraction
and divergence effects before collimating the expanded mode
using a microlens directly printed on the edge of the PIC
[132,133] (see Fig. 61). In the first configuration, a relatively
small μ-lens, with high NA, is used to generate a collimated
10 μm MFD mode from the 3μm×3μmEC, which can then
be directly coupled into a planar-polished SMF28 fiber, with
near perfect modal overlap. In the second configuration, a
larger μ-lens allows the natural divergence of the mode from
the edge coupler to expand such that MFD 25–50 μm,
before collimation. This large mode is then free-space coupled
to a second μ-lens grown on the facet of a planar-polished
SMF28 fiber, which refocuses the collimated beam onto the
fiber core with an NA that is compatible with that of SMF
and guarantees a good modal overlap.
C. Photonic Wire Bonds for Edge Coupling
Recent works, principally led by researchers at the Karlsruher
Institute of Technology, have shown that a single-mode optical
analogue of electrical wire bonds can be realized [134,135].
These photonic wire bonds (PWBs) can be used to couple light
from an SMF to an inverted taper on the Si layer of a PIC. This
fiber–PWB–PIC approach can offer high coupling-efficiency
and large optical bandwidth. The PWB is fabricated through
local two-photon polymerization of a negative monomer resist,
where a high-NA lens focuses the pulse train from a fs laser. The
two-photon lithographic process allows the PWBs to be defined
with a positional and cross-section precision, which is better
than the diffraction limit of the wavelength used for the laser
writing. The PWB is laser written in the monomer, through
precision control and 3D translation of the focusing lens, trac-
ing the laser focal-point along the calculated trajectory. Because
the PWB writing process allows for dynamic routing between
the start and end points, only a rough fiber-to-PIC prealign-
ment (on the order of 50 μm) is needed to ensure a
high CE of the overall fiber-PWB-PIC system. PWBs are an
attractive technology for photonic packaging because they have
low material costs and do not need active fiber-to-PIC align-
ment; for these reasons, considerable research effort is currently
being devoted to developing suitable processes for rapid
writing of high-performance PWBs for mass-manufacturing
applications.
7. CONCLUSIONS AND PERSPECTIVES
Silicon photonics has undoubtedly changed the paradigm of
integrated optics. Novel, compact, and low power devices
are now available to designers, and the need to efficiently
interface and package silicon photonics PICs to fiber optics
Fig. 59. Zemax Gaussian beam propagation simulations of pack-
ages utilizing micro-optics. (a) Collimation of six beams from GCs.
(b) Micro-optical bench. (c) Pluggable free-space coupler.
Fig. 60. Picture of a package exploiting the μ-lens-assisted pluggable
connector described in [131]. A pair of lenses (highlighted) are at-
tached to the FA and the PIC. Adapted from [131].
232 Vol. 7, No . 2 / February 2019 / Photonics Research Review
is now a fundamental necessity. In the past years, researchers
have faced this theme with enthusiasm and excitement, produc-
ing an impressively large volume of research results and solu-
tions. Three main aspects have driven their research efforts:
(i) the ability to efficiently couple light beams in/out to/from
a silicon photonics PIC, (ii) the bandwidth and polarization
dependent losses, and (iii) the possibility to implement any pro-
posed device in a high-density CMOS-compatible fabrication
environment and package it with practical and realistic tech-
niques. Naturally, the first approach that has historically been
studied, is the edge-coupling technique, where light is squeezed
into the silicon photonics device, without changing its traveling
direction. Introducing a less-intuitive approach, GCs have re-
vealed themselves as an extremely powerful tool to couple light
to silicon photonics PICs from an out-of-plane direction, thus
facilitating the wafer-scale testing and circuit diagnostics.
Undeniably (see Table 1), edge-coupling strategies offer better
performance (higher coupling efficiency, almost flat bandwidth,
and no polarization-dependent loss) than those typically achiev-
able using GC devices. However, performance has to be sup-
ported by other aspects, such as practicality and compatibility
with large-volume fabrication techniques. For the latter, GCs
possess significant advantages; therefore, it is clear that the se-
lection of the coupling technique has to be tailored to the spe-
cific needs of the application under consideration. In particular,
while edge couplers still offer better performance than those
achievable by grating-based configurations, the coupling
Fig. 61. (a) Schematic of free-space fiber-to-PIC coupling using a
single μ-lens. Here, the mode emitted by the edge coupler is weakly
focused by a μ-lens to give a 10 μm MFD mode size and NA that
matches the SMF28 fiber. The weakly focused mode is then directly
free-space coupled to the core of the SMF28 fiber. (b) Schematic of
free-space fiber-to-PIC coupling using a pair of μ-lenses. Here, the first
μ-lens collimates the emitted mode when it has diverged to an
MFD 25–50 μm. This large collimated mode is incident on the
second μ-lens, which refocuses it onto the core of the SMF28 fiber
with the required MFD and NA.
Table 1. This Table Allows Rapid Comparison Between the Performance of Different Structures Described in the Texta
Structure Description and Geometrical Dimensions Reference Section
CE
[dB]
PDL
[dB]
BW
[nm]
TOL
[μm]
MFD
[μm] Notes
EC: Parabolic-shape inverted taper, wt100 nm,lt40 μm[22]3.A −6.0 // // 1.2 5 E-TE
−3.3 1.2 5 E-TM
EC: 3μm×3μmpolymer SSC, wt60 nm,lt200 μm[18]3.A −0.8 0.5 >100 // 4.3 E-TE
EC: 3μm×3μmSiON SSC, wt80 nm,lt300 μm[27]3.A −0.5 // >100 // 4.3 E-TE
−2.5 9 E-TE
EC: 2μm×2μmpolymer SSC, wt75 nm,lt150 μm[26]3.A −0.5 // ≈340Δ1†4.2 E-TE
−1//<300Δ// 4.2 E-TM
EC: 3μm×1.3μmpolymer SSC, BCB spacing layer, wt175 nm,
lt175 μm
[25]3.A −0.6 // // // 2.5 S-TE
−1.9 2.5 E-TE
EC: SiOxridge WG SSC, hWG 3.5μm,hridge 1.5μm,
wt80 nm,lt300 μm
[28]3.A −0.25 // ≈100 // 3 E-TE
−0.25 ≈100 3 E-TM
EC: 6μm×6μm SiO2suspended SSC 2overlapped Si tapers,
wt110 nm
[31]3.A −1.4 // // // 6 S-TE
−1.8 >100 1.7 5 E-TE
−2.2 >100 1.7 5 E-TM
−3.8 // 2.5 9.2 E-TE
−4.0 2.5 9.2 E-TM
EC: 8μm×3μm SiO2(BOX) SSC, V-groove struct.,
wt200 nm,lt400 μm
[32]3.A −3.5 // >100 // 8 S-TE
−3.7 >100 8 S-TM
EC: Trident struct., wt100 nm,ltotal 150 μm,dlateral 1μm[33]3.B −0.34 // // 0.8 3 S-TE
−0.62 0.8 3 S-TM
−0.92 0.85/0.9 3 E-TE
−0.94 0.85/0.9 3 E-TM
EC: SWG struct. (S300 nm), two-step linear taper FF linear chirp,
wt30 nm,lt50 μm
[35]3.C −0.89 // // // 5.9 S-TE
−1.19 >210.4 S-TE
EC: SWG struct. (S260 nm,B2μm), two-step linear taper,
wt350 nm
[36]3.C −0.9 // >100 // 4 E-TE
−1.2 >100 4 E-TM
EC: SWG struct. (S220 nm B3μm), two-step linear taper,
wt220 nm
[37]3.C −0.5 <0.05 >100 // 3.2 E-TE
(Table continued)
Review Vol. 7, No. 2 / February 2019 / Photonics Research 233
Structure Description and Geometrical Dimensions Reference Section
CE
[dB]
PDL
[dB]
BW
[nm]
TOL
[μm]
MFD
[μm] Notes
EC: inv.-taper, Si3N4SWGs in SiO2ridge WG SSC
(S220 nm,B3μm), wt150 nm
[39]3.C −0.42 // >100 1.3 6 S-TE
−0.75 >100 2.2 10.4 S-TE
EC: Vertical struct. (S220 nm,B2μm),
wt190 nm,lt20 μm
[42]3.D −2.2 // >100 // 2 E-TE
−3.6 <100 2 E-TM
EC: Vertical struct. (S220 nm,B2μm)
wt50 nm,lt6μm
[44]3.D −0.8 // 420◯0.8⋄5 S-TE
−2.4 // // 5 S-TM
[45]−4.2 150◯5 E-TE
1D-UGC (S220 nm,B1μm,E70 nm, w/o TOX) [48]4.A.1 −4.32 // 40 ≈2.510.4 S-TE
−5.1 // 40 ≈2.510.4 E-TE
1D-UGC (S220 nm,B1μm,E70 nm, TOX) −3.57 // 10.4 S-TE
−4.69 // 10.4 E-TE
1D-UGC (S220 nm,B900 nm,E70 nm)−2.76 // 10.4 S-TE
1D-UGC (S220 nm,B900 nm,E70 nm), DBR −1.02 // 10.4 S-TE
1D-AGC, GA (S220 nm,B925 nm,E70 nm)[59]4.C −2.15 // // // 10.4 S-TE
1D-AGC, GA, DBR −0.36 21 35 10.4 S-TE
1D-UGC, DBR (S220 nm,E70 nm)[60]4.C.1 −1.19 // // // 10.4 S-TE
−1.58 // 36 // 10.4 E-TE
1D-UGC, BR(Au) (S220 nm,B1μm,E50 nm)[62]4.C.1 −1.43 // ≈46 // 10.4 S-TE
−1.61 // ≈35 // 10.4 E-TE
1D-AGC, GA BR(Al) (S250 nm,B3μm,E70 nm)[63]4.C.1 −0.33 // 43 // 10.4 S-TE
4.C.2 −0.62 // 40 // 10.4 E-TE
1D-UGC, p-Si (S220 nm,O150 nm,E220 nm)[65]4.C.1 −1.81 // // // 10.4 S-TE
1D-AGC, p-Si, GA −1.08 // 85Δ1.5 10.4 S-TE
1D-UGC, p-Si (S220 nm,O150 nm,E220 nm)[66]4.C.1 −1.6 // 44 // 10.4 E-TE
1D-UGC, e-Si (S220 nm,O180 nm,E250 nm)[67]4.C.1 −2.29 // 55 // 10.4 S-TE
−2.60 // 50 10.4 E-TE
1D-UGC, Si-nmb (S150 nm,O240 nm)[68]4.C.1 −1.94 // // // 10.4 S-TE
1D-AGC, Ge (S220 nm,O230 nm,E290 nm)[69]4.C.1 −1.2 // 40 // 10.4 S-TE
1D-UGC, slanted (S240 nm) (vertical emiss.) [70]4.C.1 −1.56 // // // 4.4 S-TE
1D-AGC, slanted −1.20 // // // 4.4 S-TE
1D-UGC, slanted struct. (S220 nm,B2μm)[71]4.C.1 −1.94 // 100Δ// 10.4 S-TE
−3.32 // 80Δ// 10.4 E-TE
1D-AGC (S340 nm,B2μm,E200 nm)[46]4.C.2 −0.8 // // // 10.4 S-TE
−1.2 // 45Δ// 10.4 E-TE
1D-AGC, FF-chirp (S220 nm,B2μm,E60 nm)[72]4.C.2 −2.6 // // // 10.4 S-TE
−2.7 // 41 // 10.4 E-TE
1D-AGC, FF-chirp (S250 nm,B3μm,E70 nm)[73]4.C.2 −2.3 // 37 // 10.4 S-TE
−2.7 // 29.9 // 10.4 E-TE
1D-AGC UGC,Λ-chirp (S220 nm,
B2μm,E70 nm) (vertical emiss.)
[74]4.C.2 −3.77 // 48Δ// 10.4 S-TE
−4.69 // 45Δ// 10.4 E-TE
1D-AGC, GA (S220 nm,E100 nm)[75]4.C.2 −1.9 // // // 10.4 S-TE
1D-AGC, GA (S220 nm,E220 nm,B2μm)−2.08 // // // 10.4 S-TE
1D-AGC, GA (S340 nm,E200 nm)−0.49 // 33 // 10.4 S-TE
1D-AGC, GA (S340 nm,dmin 100 nm)−0.72 // // // 10.4 S-TE
1D-AGC, FF-chirp, Λvaried (S220 nm,E110 nm)[76]4.C.2 −1.55 // // // 10.4 S-TE
1D-AGC, FF-chirp, Λvaried (S260 nm,E160 nm)−0.81 // 32.8 // 10.4 S-TE
−0.9 // 38.8 // 10.4 E-TE
1D-GC, double-etch depth [77]4.C.3 −1.3 // 60Δ// 10.4 S-TE
1D-GC, double-etch depth (S220 nm)[78]4.C.3 −1.05 // 30 // 10.4 S-TE
1D-GC, double-etch depth (S220 nm)[79]4.C.3 −1.3 // 52Δ// 10.4 E-TE
1D-GC, double-etch depth (S220 nm)[80]4.C.3 −2.2 // // // 10.4 S-TE
−2.7 // 62Δ// 10.4 E-TE
1D-AGC, etch lag effect (S250 nm,B3μm)[81]4.C.3 −1.3 // // // 10.4 S-TE
−1.9 // 43 // 10.4 E-TE
SWG-UGC (S220 nm,dhole 200 nm)[82]4.C.4 −4.69 // 40Δ// 10.4 E-TE
SWG-UGC (S250 nm,B1μm,dhole 143 nm)[83]4.C.4 −3.77 // 37 // 10.4 E-TE
SWG-UGC, nano-pillars (S340 nm,B2μm)
(optimized B1.64 μm)
[84]4.C.4 −5.6 // 73 // 10.4 E-TE
−3.4 // 86 // 10.4 S-TE
SWG-AGC (S260 nm,B2μm)[85]4.C.4 −3 // // // 10.4 S-TM
SWG-AGC (S260 nm,B2μm,dmin 100 nm)[86]4.C.4 −2.8 // 35 // 10.4 S-TM
−3.7 // 40 // 10.4 E-TM
SWG-AGC BRAl(S220 nm,B3μm,dmin 100 nm)[87]4.C.4 −0.67 // 35 // 10.4 S-TE
−0.69 // 60Δ// 10.4 E-TE
(Table continued)
234 Vol. 7, No. 2 / February 2019 / Photonics Research Review
Structure Description and Geometrical Dimensions Reference Section
CE
[dB]
PDL
[dB]
BW
[nm]
TOL
[μm]
MFD
[μm] Notes
SWG-AGC (S250 nm,B1μm,dmin 60 nm)[88]4.C.4 −1.8 // // // 10.4 S-TE
−1.74 // 60Δ// 10.4 E-TE
SWG-AGC BR(Al) (S250 nm,B3μm,dmin 70 nm)[64]4.C.4 −0.43 // 76Δ// 10.4 S-TE
−0.58 // 71Δ// 10.4 E-TE
Si3N4-UGC (N400 nm)[89]4.C.5 −3.9 // 67 >1μm10.4 S-TE
−4.2 // 67 // 10.4 E-TE
<−18 // // // 10.4 E-TM
Si3N4-UGC DBR (N400 nm)−1.9 // 90 // 10.4 S-TE
Si3N4-UGC (N400 nm,B2.6μm)[90]4.C.5 −2.6 // // // 10.4 S-TE
−5.2 // // // 10.4 E-TE
Si3N4-UGC DBR (N400 nm,B2.6μm)−1.2 // // // 10.4 S-TE
−2.6 // 53 // 10.4 E-TE
Si3N4-UGC (N700 nm,B3μm)[95]4.C.5 −3.7 // 54 // 10.4 E-TE
Si3N4-AGC (N600 nm,E1300 nm,E2600 nm)[96]4.C.5 −1.5 // 60Δ// 10.4 S-TE
Dual-level Si3N4∕Si-AGC (N400 nm,B2μm)[97]4.C.5 −1.0 // 82 2 10.4 S-TE
–1.3 // 80 // 10.4 E-TE
Dual-level AGC (N400 nm,S220 nm,B2μm)[98]4.C.5 −0.88 // ≈60 // 10.4 S-TE
FGC (S220 nm,B2μm,E70 nm)[103]4.D −5 // // // 10 E-O.V.M./
TE
Double UGC (S220 nm,E70 nm)[108]4.E.1 −3//67Δ// 14 S-LP11 ∕TE
−4.21 // 62Δ// 14 E-LP11 ∕TE
Double UGC (S220 nm,E70 nm)[109]4.E.1 −3.61 // 70Δ// 14 S-LP11 ∕TE
−3.68 // 35 // 14 E-LP11∕TE
Two-section AGC (S220 nm,B2μm,E100 nm)[110]4.E.2 −2.19 // 50Δ// 18 S-LP01 ∕TE
−2.60 // 30Δ// 18 S-LP11a ∕TE
−2.88 // 55Δ// 18 S-LP11b ∕TE
−3.0 // 33Δ// 18 S-LP21b ∕TE
Non-UGC (S220 nm,B2μm,E60 nm)[111]5.A −6.9 0.57 40 // 10.4 S-TE
−7.8 <0.8// // 10.4 E-TE
Non-UGC (S220 nm,B2μm,E100 nm)−5.4 0.5 40 // 10.4 S-TE
1D UGC, BR(Al) (S250 nm,B3μm,E70 nm)[112]5.A −0.7 // // // 10.4 S-TE
−1.1 // // // 10.4 S-TM
−2 // 29 // 10.4 E-TE
−2.3 // 23 // 10.4 E-TM
1D UGC (S400 nm,E120 nm)[113]5.A −2.57 <0.5 30 // 10.4 S-TE
−2.73 // 35 // 10.4 S-TM
SWG-UGC (S340 nm,B2μm)[114]5.A −4 // // // 10.4 S-TE
−4 // // // 10.4 S-TM
SWG-UGC DBR (S340 nm,B2μm)−1.94 // 65Δ// 10.4 S-TE
−1.94 // 65Δ// 10.4 S-TM
2D-GC (S220 nm,B1μm,E90 nm)[115]5.B −7 // // // 10.4 E
Focusing 2D-GC (S220 nm,B2μm,E70 nm)[116]5.B −5.7 0.4 // // 10.4 E
2D-GC (S220 nm,B2μm,E70 nm)[118]5.B −6.7 0.66 60Δ// 10.4 E
Focusing 2D-GC (S220 nm,B2μm,E70 nm)[119]5.B −4.1 // // // 10.4 E
2D-GC (S220 nm,B2μm,E120 nm)[117]5.B.1 −3.2 0.6 40 // 10.4 S
2D-GC (S220 nm,B2μm,E120 nm)[121]5.B.1 −3.75 // 43Δ// 10.4 E
2D-GC, p-Si (S220 nm,O180 nm,E291 nm)[120]5.B.1 −1.9 0.3 38 // 10.4 S
2D-GC, BR(Al) (S160 nm,B2175 nm,E80 nm)5.B.4 −0.95 0.3 42 // 10.4 S
SWG-2D-GC, etch lag effect (S220 nm,
B3μm,E150 nm)
[122]5.B.2 −5.6 0.2 // // 10.4 S
−5.8 0.2 // // 10.4 E
SWG-2D-GC (S220 nm,B2μm,E70 nm)[123]5.B.3 −4.4 0.25 // // 10.4 S
−5.0 0.25 // // 10.4 E
2D-GC, BR(Au) (S220 nm,E150 nm)[125]5.B.4 −1.37 // // // 10.4 S
−1.8 1 32 // 10.4 E
aIn the structure description table, EC, 1D-GC, and 2D-GC indicate edge coupler, 1D-GC, and 2D-GC, respectively. wtand ltindicate tip width and taper length.
The values between parenthesis (S, B, T, O, N) indicate the thickness of the silicon layer, BOX, TOX, overlay, and silicon nitride layers, respectively. The bandwidth
(BW) column reports the 1 dB bandwidth of the described coupling solution. If the 3 dB bandwidth is given, a Δapex is added to the given value. If the 0.5 dB
bandwidth is given, a ◯apex is added to the given value. The tolerance (TOL) column reports the alignment deviation causing an extra-loss of 1 dB, unless a symbol
appears as an apex. In the case of the †symbol, the position offset corresponds to a 3 dB extra loss, while in the case of the ⋄symbol, the position offset corresponds to a
0.5 dB extra loss. If two values are given, they correspond to the horizontal and vertical tolerance, respectively. In the notes column, we report if the given figures were
obtained by experiments or by numerical simulations (E/S) and the considered mode (TE/TM).
Review Vol. 7, No. 2 / February 2019 / Photonics Research 235
strategy to be implemented in a specific PIC must be selected
considering many different factors depending on the specific
application (e.g., bandwidth requirements and acceptable
polarization dependence), also on the acceptable packaging
cost and on the expected production volume. As a conse-
quence, recently, significant attention is being paid to the iden-
tification of design and processes, thus allowing one to design
high-efficiency couplers, which can be fabricated using stan-
dard materials and processes commonly exploited in CMOS
production lines. A common trend now emerging is that of
realizing large SSCs on SOI edge couplers, so as to match
the 10 μm MFD of common fibers. This approach offers a
1 dB tolerance of 2.5μm, which is nowadays compatible
with flip-chip and pick-and-place technology. Such an ap-
proach may yield assembly times shorter than 1 min, provided
that two technical challenges will be positively solved: (i) stress
and torque accumulated on the FA during placement must be
properly relieved and (ii) the drift occurring during epoxy cur-
ing must be avoided or at least corrected by a proper process
control. On the other hand, from the industrial point of view,
other considerations, e.g., material costs, assembly time, chip-
to-chip variation, lifetime, and resilience of the optical connec-
tion play a fundamental role in influencing the coupling and
packaging strategy to be applied. In the field of active cables, Si-
photonics-based solutions have already reached commercial
maturity and different products with stable and reliable cou-
plers are available on the market. For this kind of sector, sat-
isfied by relatively small production volumes requiring higher
performance, fast, serial, machine-vision assembly of FAs to
edge couplers may provide a good trade-off between required
optical performance and packaging costs. Nevertheless, it is evi-
dent that, in order to access mass markets (with a production
volume in the order of millions of devices per year), it will be
necessary to develop strategies for wafer-level testing and pack-
aging (as happens for electronic components); in this direction,
the combination of GCs and μ-lenses currently constitutes the
most promising solution.
Funding. Engineering and Physical Sciences Research
Council (EPSRC) (EP/L00044X/1).
REFERENCES
1. B. Jalali and S. Fathpour, “Silicon photonics,”J. Lightwave Technol.
24, 4600–4615 (2006).
2. Y. A. Vlasov and S. J. McNab, “Losses in single-mode silicon-on-
insulator strip waveguides and bends,”Opt. Express 12, 1622–1631
(2004).
3. P. Dumon, W. Bogaerts, V. Wiaux, J. Wouters, S. Beckx, J. Van
Campenhout, D. Taillaert, B. Luyssaert, P. Bienstman, D. Van
Thourhout, and R. Baets, “Low-loss SOI photonic wires and ring
resonators fabricated with deep UV lithography,”IEEE Photon.
Technol. Lett. 16, 1328–1330 (2004).
4. R. Marchetti, V. Vitali, C. Lacava, I. Cristiani, B. Charbonnier, V.
Muffato, M. Fournier, and P. Minzioni, “Group-velocity dispersion
in SOI-based channel waveguides with reduced-height,”Opt.
Express 25, 9761–9767 (2017).
5. G. F. R. Chen, J. R. Ong, T. Y. L. Ang, S. T. Lim, C. E. Png, and D. T.
H. Tan, “Broadband silicon-on-insulator directional couplers using a
combination of straight and curved waveguide sections,”Sci. Rep. 7,
7246 (2017).
6. Y. Zhang, S. Yang, A. E.-J. Lim, G.-Q. Lo, C. Galland, T.
Baehr-Jones, and M. Hochberg, “A compact and low loss Y-junction
for submicron silicon waveguide,”Opt. Express 21, 1310–1316
(2013).
7. X. Wang, W. Shi, H. Yun, S. Grist, N. A. F. Jaeger, and L.
Chrostowski, “Narrow-band waveguide Bragg gratings on SOI
wafers with CMOS-compatible fabrication process,”Opt. Express
20, 15547–15558 (2012).
8. W. Bogaerts, P. Dumon, D. Van Thourhout, D. Taillaert, P. Jaenen,
J. Wouters, S. Beckx, V. Wiaux, and R. G. Baets, “Compact
wavelength-selective functions in silicon-on-insulator photonic
wires,”IEEE J. Sel. Top. Quantum Electron. 12, 1394–1401
(2006).
9. C. Lacava, M. J. Strain, P. Minzioni, I. Cristiani, and M. Sorel,
“Integrated nonlinear Mach Zehnder for 40 Gbit/s all-optical switch-
ing,”Opt. Express 21, 21587–21595 (2013).
10. W. Bogaerts, P. de Heyn, T. van Vaerenbergh, K. de Vos, S. Kumar
Selvaraja, T. Claes, P. Dumon, P. Bienstman, D. van Thourhout, and
R. Baets, “Silicon microring resonators,”Laser Photon. Rev. 6,47–
73 (2012).
11. R. Marchetti, V. Vitali, C. Lacava, I. Cristiani, G. Giuliani, V. Muffato,
M. Fournier, S. Abrate, R. Gaudino, E. Temporiti, L. Carroll, and
P. Minzioni, “Low-loss micro-resonator filters fabricated in silicon
by CMOS-compatible lithographic techniques: design and charac-
terization,”Appl. Sci. 7, 174 (2017).
12. G. T. Reed, G. Mashanovich, F. Y. Gardes, and D. J. Thomson,
“Silicon optical modulators,”Nat. Photonics 4, 518–526 (2010).
13. D. J. Thomson, F. Y. Gardes, J.-M. Fedeli, S. Zlatanovic, Y. Hu, B. P.
P. Kuo, E. Myslivets, N. Alic, S. Radic, G. Z. Mashanovich, and G. T.
Reed, “50-Gb/s silicon optical modulator,”IEEE Photon. Technol.
Lett. 24, 234–236 (2012).
14. “Corning®SMF-28®ultra optical fiber,”https://www.corning.com/
media/worldwide/coc/documents/Fiber/PI1424_11-14.pdf.
15. “PANDA PM specialty optical fibers,”https://www.corning.com/
media/worldwide/csm/documents/PANDA%20PM%20and%20RC
%20PANDA%20Specialty%20Fiber.pdf.
16. “Corning®RGB 400 specialty optical fibers,”https://www.corning.
com/media/worldwide/csm/documents/Corning%20RGB%20400%
20Specialty%20Fiber.pdf.
17. A. M. Kowalevicz, Jr. and F. Bucholtz, “Beam divergence from an
SMF-28 optical fiber,”Tech. Rep. NRL/MR/5650-06-8996 (Naval
Research Lab, 2006).
18. T. Shoji, T. Tsuchizawa, T. Watanabe, K. Yamada, and H. Morita,
“Low loss mode size converter from 0.3 μm square Si wire wave-
guides to singlemode fibres,”Electron. Lett. 38, 1669–1670 (2002).
19. “Ultra-high NA single-mode fibers,”http://www.nufern.com/pam/
optical_fibers/spec/id/988/.
20. “UHNA fiber: efficient coupling to silicon waveguides,”https://www.
nufern.com/library/getpdf/id/485/.
21. K. Kasaya, O. Mitomi, M. Naganuma, Y. Kondo, and Y. Noguchi, “A
simple laterally tapered waveguide for low-loss coupling to single-
mode fibers,”IEEE Photon. Technol. Lett. 5, 345–347 (1993).
22. V. R. Almeida, R. R. Panepucci, and M. Lipson, “Nanotaper for
compact mode conversion,”Opt. Lett. 28, 1302–1304 (2003).
23. L. Vivien, S. Laval, E. Cassan, X. L. Roux, and D. Pascal, “2-D taper
for low-loss coupling between polarization-insensitive microwave-
guides and single-mode optical fibers,”J. Lightwave Technol. 21,
2429–2433 (2003).
24. A. Sure, T. Dillon, J. Murakowski, C. Lin, D. Pustai, and D. Prather,
“Fabrication and characterization of three-dimensional silicon
tapers,”Opt. Express 11, 3555–3561 (2003).
25. G. Roelkens, P. Dumon, W. Bogaerts, D. van Thourhout, and R.
Baets, “Efficient silicon-on-insulator fiber coupler fabricated using
248-nm-deep UV lithography,”IEEE Photon. Technol. Lett. 17,
2613–2615 (2005).
26. S. McNab, N. Moll, and Y. Vlasov, “Ultra-low loss photonic integrated
circuit with membrane-type photonic crystal waveguides,”Opt.
Express 11, 2927–2939 (2003).
27. T. Tsuchizawa, K. Yamada, H. Fukuda, T. Watanabe, J. I.
Takahashi, M. Takahashi, T. Shoji, E. Tamechika, S. I. Itabashi,
and H. Morita, “Microphotonics devices based on silicon
236 Vol. 7, No. 2 / February 2019 / Photonics Research Review
microfabrication technology,”IEEE J. Sel. Top. Quantum Electron.
11, 232–240 (2005).
28. B. Ben Bakir, A. V. De Gyves, R. Orobtchouk, P. Lyan, C. Porzier,
A. Roman, and J. M. Fedeli, “Low-Loss (< 1 dB) and polarization-
insensitive edge fiber couplers fabricated on 200-mm silicon-on-
insulator wafers,”IEEE Photon. Technol. Lett. 22, 739–741 (2010).
29. M. Schnarrenberger, L. Zimmermann, T. Mitze, J. Bruns, and K.
Petermann, “Facet preparation of SOI waveguides by etching and
cleaving compared to dicing and polishing,”in IEEE International
Conference on Group IV Photonics (IEEE, 2004), pp. 72–74.
30. D. Dai, S. He, and H. K. Tsang, “Bilevel mode converter between a
silicon nanowire waveguide and a larger waveguide,”J. Lightwave
Technol. 24, 2428–2433 (2006).
31. Q. Fang, T.-Y. Liow, J. F. Song, C. W. Tan, M. B. Yu, G. Q. Lo, and
D.-L. Kwong, “Suspended optical fiber-to-waveguide mode size con-
verter for silicon photonics,”Opt. Express 18, 7763–7769 (2010).
32. J. V. Galán, P. Sanchis, G. Sánchez, and J. Martí, “Polarization in-
sensitive low-loss coupling technique between SOI waveguides and
high mode field diameter single-mode fibers,”Opt. Express 15,
7058–7065 (2007).
33. N. Hatori, T. Shimizu, M. Okano, M. Ishizaka, T. Yamamoto, Y. Urino,
M. Mori, T. Nakamura, and Y. Arakawa, “A hybrid integrated light
source on a silicon platform using a trident spot-size converter,”
J. Lightwave Technol. 32, 1329–1336 (2014).
34. Y. Urino, T. Usuki, J. Fujikata, M. Ishizaka, K. Yamada, T. Horikawa,
T. Nakamura, and Y. Arakawa, “High-density and wide-bandwidth
optical interconnects with silicon optical interposers,”Photon. Res.
2,A1–A7 (2014).
35. P. Cheben, D.-X. Xu, S. Janz, and A. Densmore, “Subwavelength
waveguide grating for mode conversion and light coupling in inte-
grated optics,”Opt. Express 14, 4695–4702 (2006).
36. P. Cheben, P. J. Bock, J. H. Schmid, J. Lapointe, S. Janz, D.-X. Xu,
A. Densmore, A. Delâge, B. Lamontagne, and T. J. Hall, “Refractive
index engineering with subwavelength gratings for efficient micro-
photonic couplers and planar waveguide multiplexers,”Opt. Lett.
35, 2526–2528 (2010).
37. P. Cheben, J. H. Schmid, S. Wang, D.-X. Xu, M. Vachon, S. Janz, J.
Lapointe, Y. Painchaud, and M.-J. Picard, “Broadband polarization
independent nanophotonic coupler for silicon waveguides with
ultra-high efficiency,”Opt. Express 23, 22553–22563 (2015).
38. S. M. Rytov, “Electromagnetic properties of a finely stratified
medium,”Sov. Phys. JETP 2, 446–475 (1956).
39. M. Papes, P. Cheben, D. Benedikovic, J. H. Schmid, J. Pond, R.
Halir, A. Ortega-Moñux, G. Wangüemert-Pérez, W. N. Ye, D.-X.
Xu, S. Janz, M. Dado, and V. Vašinek, “Fiber-chip edge coupler with
large mode size for silicon photonic wire waveguides,”Opt. Express
24, 5026–5038 (2016).
40. T. Yoshida, T. Nishi, S. Tajima, R. Takei, M. Suzuki, E. Omoda, M.
Nagao, N. Miura, M. Mori, and Y. Sakakibara, “Vertically-curved
silicon waveguide fabricated by ion-induced bending method for ver-
tical light coupling,”in IEEE International Conference on Group IV
Photonics (IEEE, 2013), pp. 89–90.
41. T. Yoshida, S. Tajima, R. Takei, M. Mori, N. Miura, and Y.
Sakakibara, “Vertical silicon waveguide coupler bent by ion implan-
tation,”Opt. Express 23, 29449–29456 (2015).
42. T. Yoshida, E. Omoda, Y. Atsumi, T. Nishi, S. Tajima, N. Miura, M.
Mori, and Y. Sakakibara, “Vertically curved Si waveguide coupler
with low loss and flat wavelength window,”J. Lightwave Technol.
34, 1567–1571 (2016).
43. T. Yoshida, E. Omoda, Y. Atsumi, T. Nishi, S. Tajima, R. Takei, N.
Miura, M. Mori, and Y. Sakakibara, “CMOS-compatible vertical Si-
waveguide coupler fabricated by ion implantation,”in Integrated
Photonics Research, Silicon and Nanophotonics (Optical Society
of America, 2016), paper ITu2B-1.
44. Y. Atsumi, T. Yoshida, E. Omoda, and Y. Sakakibara, “Design of
compact surface optical coupler based on vertically curved silicon
waveguide for high-numerical-aperture single-mode optical fiber,”
Jpn. J. Appl. Phys. 56, 090307 (2017).
45. Y. Atsumi, T. Yoshida, E. Omoda, and Y. Sakakibara, “Broad-band
surface optical coupler based on a SiO2-capped vertically curved
silicon waveguide,”Opt. Express 26, 10400–10407 (2018).
46. X. Chen, C. Li, C. K. Y. Fung, S. M. G. Lo, and H. K. Tsang,
“Apodized waveguide grating couplers for efficient coupling to optical
fibers,”IEEE Photon. Technol. Lett. 22, 1156–1158 (2010).
47. H.-L. Tseng, E. Chen, H. Rong, and N. Na, “High-performance
silicon-on-insulator grating coupler with completely vertical emis-
sion,”Opt. Express 23, 24433–24439 (2015).
48. D. Taillaert, F. Van Laere, M. Ayre, W. Bogaerts, D. Van Thourhout,
P. Bienstman, and R. Baets, “Grating couplers for coupling between
optical fibers and nanophotonic waveguides,”Jpn. J. Appl. Phys. 45,
6071–6077 (2006).
49. F. Van Laere, T. Claes, J. Schrauwen, S. Scheerlinck, W. Bogaerts,
D. Taillaert, L. O’Faolain, D. Van Thourhout, and R. Baets, “Compact
focusing grating couplers for silicon-on-insulator integrated circuits,”
IEEE Photon. Technol. Lett. 19, 1919–1921 (2007).
50. R. Waldhäusl, B. Schnabel, P. Dannberg, E.-B. Kley, A. Bräuer, and
W. Karthe, “Efficient coupling into polymer waveguides by gratings,”
Appl. Opt. 36, 9383–9390 (1997).
51. F. Olyslager, Electromagnetic Waveguides and Transmission Lines
Monographs in Electrical and Electronic Engineering (Clarendon,
1999).
52. “Photonics SOI, for high-speed optical transceivers in data centers,”
https://www.soitec.com/en/products/photonics-soi.
53. M. Neviere, D. Maystre, and P. Vincent, “Determination of the leaky
modes of a corrugated waveguide: application to the study of anoma-
lies of dielectric coated gratings,”J. Opt. 8, 231–242 (1977).
54. D. Pascal, R. Orobtchouk, A. Layadi, A. Koster, and S. Laval,
“Optimized coupling of a Gaussian beam into an optical waveguide
with a grating coupler: comparison of experimental and theoretical
results,”Appl. Opt. 36, 2443–2447 (1997).
55. T. W. Ang, G. T. Reed, A. Vonsovici, A. G. R. Evans, P. R. Routley,
T. Blackburn, and M. R. Josey, “Grating couplers using silicon on
insulator,”Proc. SPIE 3620,79–86 (1999).
56. T. W. Ang, G. T. Reed, A. Vonsovici, A. G. R. Evans, P. R. Routley,
and M. R. Josey, “Effects of grating heights on highly efficient uni-
bond SOI waveguide grating couplers,”IEEE Photon. Technol.
Lett. 12,59–61 (2000).
57. T. W. Ang, G. T. Reed, A. Vonsovici, A. G. Evans, P. R. Routley, and
M. R. Josey, “Highly efficient unibond silicon-on-insulator blazed gra-
ting couplers,”Appl. Phys. Lett. 77, 4214–4216 (2000).
58. R. Orobtchouk, A. Layadi, H. Gualous, D. Pascal, A. Koster, and S.
Laval, “High-efficiency light coupling in a submicrometric silicon-on-
insulator waveguide,”Appl. Opt. 39, 5773–5777 (2000).
59. D. Taillaert, P. Bienstman, and R. Baets, “Compact efficient broad-
band grating coupler for silicon-on-insulator waveguides,”Opt. Lett.
29, 2749–2751 (2004).
60. S. Selvaraja, D. Vermeulen, M. Schaekers, E. Sleeckx, W. Bogaerts,
G. Roelkens, P. Dumon, D. V. Thourhout, and R. Baets, “Highly
efficient grating coupler between optical fiber and silicon photonic
circuit,”in Conference on Lasers and Electro-Optics (Optical
Society of America, 2009), paper CTuC6.
61. M. K. Emsley, O. Dosunmu, and M. S. Ünlü, “Silicon substrates with
buried distributed Bragg reflectors for resonant cavity-enhanced opto-
electronics,”IEEE J. Sel. Top. Quantum Electron. 8, 948–955 (2002).
62. F. Van Laere, G. Roelkens, M. Ayre, J. Schrauwen, D. Taillaert, D.
Van Thourhout, T. F. Krauss, and R. Baets, “Compact and highly
efficient grating couplers between optical fiber and nanophotonic
waveguides,”J. Lightwave Technol. 25, 151–156 (2007).
63. W. S. Zaoui, A. Kunze, W. Vogel, M. Berroth, J. Butschke, F.
Letzkus, and J. Burghartz, “Bridging the gap between optical fibers
and silicon photonic integrated circuits,”Opt. Express 22, 1277–1286
(2014).
64. Y. Ding, C. Peucheret, H. Ou, and K. Yvind, “Fully etched apodized
grating coupler on the SOI platform with −0.58 dB coupling effi-
ciency,”Opt. Lett. 39, 5348–5350 (2014).
65. G. Roelkens, D. Van Thourhout, and R. Baets, “High efficiency
silicon-on-insulator grating coupler based on a poly-silicon overlay,”
Opt. Express 14, 11622–11630 (2006).
66. D. Vermeulen, S. Selvaraja, P. Verheyen, G. Lepage, W. Bogaerts, P.
Absil, D. Van Thourhout, and G. Roelkens, “High-efficiency fiber-to-
chip grating couplers realized using an advanced CMOS-compatible
silicon-on-insulator platform,”Opt. Express 18, 18278–18283 (2010).
Review Vol. 7, No. 2 / February 2019 / Photonics Research 237
67. G. Roelkens, D. Vermeulen, D. Van Thourhout, R. Baets, S. Brision,
P. Lyan, P. Gautier, and J. M. Fédéli, “High efficiency diffractive gra-
ting couplers for interfacing a single mode optical fiber with a nano-
photonic silicon-on-insulator waveguide circuit,”Appl. Phys. Lett. 92,
131101 (2008).
68. T. K. Saha and W. Zhou, “High efficiency diffractive grating coupler
based on transferred silicon nanomembrane overlay on photonic
waveguide,”J. Phys. D 42, 085115 (2009).
69. S. Yang, Y. Zhang, T. Baehr-Jones, and M. Hochberg, “High effi-
ciency germanium-assisted grating coupler,”Opt. Express 22,
30607–30612 (2014).
70. B. Wang, J. Jiang, and G. P. Nordin, “Embedded slanted grating
for vertical coupling between fibers and silicon-on-insulator planar
waveguides,”IEEE Photon. Technol. Lett. 17, 1884–1886 (2005).
71. J. Schrauwen, F. Van Laere, D. Van Thourhout, and R. Baets,
“Focused-ion-beam fabrication of slanted grating couplers in silicon-
on-insulator waveguides,”IEEE Photon. Technol. Lett. 19, 816–818
(2007).
72. L. He, Y. Liu, C. Galland, A. E. J. Lim, G. Q. Lo, T. Baehr-Jones, and
M. Hochberg, “A high-efficiency nonuniform grating coupler realized
with 248-nm optical lithography,”IEEE Photon. Technol. Lett. 25,
1358–1361 (2013).
73. M. H. Lee, J. Y. Jo, D. W. Kim, Y. Kim, and K. H. Kim, “Comparative
study of uniform and nonuniform grating couplers for optimized fiber
coupling to silicon waveguides,”J. Opt. Soc. Korea 20, 291–299
(2016).
74. X. Chen, C. Li, and H. K. Tsang, “Fabrication-tolerant waveguide
chirped grating coupler for coupling to a perfectly vertical optical
fiber,”IEEE Photon. Technol. Lett. 20, 1914–1916 (2008).
75. A. Bozzola, L. Carroll, D. Gerace, I. Cristiani, and L. C. Andreani,
“Optimising apodized grating couplers in a pure SOI platform to –
0.5 dB coupling efficiency,”Opt. Express 23, 16289–16304 (2015).
76. R. Marchetti, C. Lacava, A. Khokhar, X. Chen, I. Cristiani, D. J.
Richardson, G. T. Reed, P. Petropoulos, and P. Minzioni, “High-
efficiency grating-couplers: demonstration of a new design strategy,”
Sci. Rep. 7, 16670 (2017).
77. M. Fan, M. A. Popović, and F. X. Kärtner, “High directivity, vertical
fiber-to-chip coupler with anisotropically radiating grating teeth,”in
Conference on Lasers and Electro-Optics (CLEO) (Optical Society
of America, 2007), paper CTuDD3.
78. C. Alonso-Ramos, P. Cheben, A. Ortega-Moñux, J. H. Schmid, D.-X.
Xu, and I. Molina-Fernández, “Fiber-chip grating coupler based on
interleaved trenches with directionality exceeding 95%,”Opt. Lett.
39, 5351–5354 (2014).
79. D. Benedikovic, C. Alonso-Ramos, P. Cheben, J. H. Schmid, S.
Wang, D.-X. Xu, J. Lapointe, S. Janz, R. Halir, A. Ortega-Moñux,
J. G. Wangüemert-Pérez, I. Molina-Fernández, J.-M. Fédéli, L.
Vivien, and M. Dado, “High-directionality fiber-chip grating coupler
with interleaved trenches and subwavelength index-matching struc-
ture,”Opt. Lett. 40, 4190–4193 (2015).
80. D. Benedikovic, C. Alonso-Ramos, D. Pérez-Galacho, S. Guerber,
V. Vakarin, G. Marcaud, X. Le Roux, E. Cassan, D. Marris-Morini,
P. Cheben, F. Boeuf, C. Baudot, and L. Vivien, “L-shaped fiber-chip
grating couplers with high directionality and low reflectivity fabricated
with deep-UV lithography,”Opt. Lett. 42, 3439–3443 (2017).
81. Y. Tang, Z. Wang, L. Wosinski, U. Westergren, and S. He, “Highly
efficient nonuniform grating coupler for silicon-on-insulator nanopho-
tonic circuits,”Opt. Lett. 35, 1290–1292 (2010).
82. X. Chen and H. K. Tsang, “Nanoholes grating couplers for coupling
between silicon-on-insulator waveguides and optical fibers,”IEEE
Photon. J. 1, 184–190 (2009).
83. L. Liu, M. Pu, K. Yvind, and J. M. Hvam, “High-efficiency, large-
bandwidth silicon-on-insulator grating coupler based ona fully-etched
photonic crystal structure,”Appl. Phys. Lett. 96, 051126 (2010).
84. X. Chen, K. Xu, Z. Cheng, C. K. Y. Fung, and H. K. Tsang,
“Wideband subwavelength gratings for coupling between silicon-
on-insulator waveguides and optical fibers,”Opt. Lett. 37,
3483–3485 (2012).
85. R. Halir, P. Cheben, S. Janz, D.-X. Xu, Í. Molina-Fernández, and
J. G. Wangüemert-Pérez, “Waveguide grating coupler with sub-
wavelength microstructures,”Opt. Lett. 34, 1408–1410 (2009).
86. R. Halir, P. Cheben, J. H. Schmid, R. Ma, D. Bedard, S. Janz,
D.-X. Xu, A. Densmore, J. Lapointe, and Í. Molina-Fernández,
“Continuously apodized fiber-to-chip surface grating coupler with
refractive index engineered subwavelength structure,”Opt. Lett.
35, 3243–3245 (2010).
87. D. Benedikovic, P. Cheben, J. H. Schmid, D.-X. Xu, B. Lamontagne,
S. Wang, J. Lapointe, R. Halir, A. Ortega-Moñux, S. Janz, and M.
Dado, “Subwavelength index engineered surface grating coupler
with sub-decibel efficiency for 220-nm silicon-on-insulator wave-
guides,”Opt. Express 23, 22628–22635 (2015).
88. Y. Ding, C. Peucheret, and H. Ou, “Ultra-high-efficiency apodized
grating coupler using a fully etched photonic crystal,”Opt. Lett.
38, 2732–2734 (2013).
89. C. R. Doerr, L. Chen, Y. Chen, and L. L. Buhl, “Wide bandwidth
silicon nitride grating coupler,”IEEE Photon. Technol. Lett. 22,
1461–1463 (2010).
90. H. Zhang, C. Li, X. Tu, H. Zhou, X. Luo, M. Yu, and P. G.-Q. Lo, “High
efficiency silicon nitride grating coupler with DBR,”in Optical Fiber
Communications Conference and Exhibition (OFC) (IEEE, 2014),
paper Th1A.4.
91. D. J. Moss, R. Morandotti, A. L. Gaeta, and M. Lipson, “New CMOS-
compatible platforms based on silicon nitride and Hydex for nonlinear
optics,”Nat. Photonics 7, 597–607 (2013).
92. C. Lacava, S. Stankovic, A. Z. Khokhar, T. D. Bucio, F. Y. Gardes,
G. T. Reed, D. J. Richardson, and P. Petropoulos, “Si-rich silicon
nitride for nonlinear signal processing applications,”Sci. Rep. 7,
22 (2017).
93. Y. Okawachi, K. Saha, J. S. Levy, Y. H. Wen, M. Lipson, and A. L.
Gaeta, “Octave-spanning frequency comb generation in a silicon
nitride chip,”Opt. Lett. 36, 3398–3400 (2011).
94. K. Luke, Y. Okawachi, M. R. E. Lamont, A. L. Gaeta, and M. Lipson,
“Broadband mid-infrared frequency comb generation in a Si3N4
microresonator,”Opt. Lett. 40, 4823–4826 (2015).
95. X. Zhao, D. Li, C. Zeng, G. Gao, Z. Huang, Q. Huang, Y. Wang, and
J. Xia, “Compact grating coupler for 700-nm silicon nitride strip
waveguides,”J. Lightwave Technol. 34, 1322–1327 (2016).
96. Y. Chen, T. D. Bucio, A. Z. Khokhar, M. Banakar, K. Grabska, F. Y.
Gardes, R. Halir, Í. Molina-Fernández, P. Cheben, and J.-J. He,
“Experimental demonstration of an apodized-imaging chip-fiber
grating coupler for Si3N4waveguides,”Opt. Lett. 42, 3566–3569
(2017).
97. W. D. Sacher, Y. Huang, L. Ding, B. J. F. Taylor, H. Jayatilleka, G.-Q.
Lo, and J. K. S. Poon, “Wide bandwidth and high coupling efficiency
Si3N4-on-SOI dual-level grating coupler,”Opt. Express 22, 10938–
10947 (2014).
98. J. Zou, Y. Yu, M. Ye, L. Liu, S. Deng, and X. Zhang, “Ultra efficient
silicon nitride grating coupler with bottom grating reflector,”Opt.
Express 23, 26305–26312 (2015).
99. M. Passoni, D. Gerace, L. Carroll, and L. C. Andreani, “Grating cou-
plers in silicon-on-insulator: the role of photonic guided resonances
on lineshape and bandwidth,”Appl. Phys. Lett. 110, 041107 (2017).
100. J. Romero, D. Giovannini, S. Franke-Arnold, S. M. Barnett, and M. J.
Padgett, “Increasing the dimension in high-dimensional two-photon
orbital angular momentum entanglement,”Phys. Rev. A 86, 012334
(2012).
101. N. Bozinovic, Y. Yue, Y. Ren, M. Tur, P. Kristensen, H. Huang,
A. E. Willner, and S. Ramachandran, “Terabit-scale orbital angular
momentum mode division multiplexing in fibers,”Science 340,
1545–1548 (2013).
102. G. Gibson, J. Courtial, M. J. Padgett, S. M. Barnett, and S. Franke-
Arnold, “Free-space information transfer using light beams carry-
ing orbital angular momentum,”Opt. Express 12, 5448–5456
(2004).
103. C. T. Nadovich, W. D. Jemison, D. J. Kosciolek, and D. T. Crouse,
“Focused apodized forked grating coupler,”Opt. Express 25, 26861–
26874 (2017).
104. D. J. Richardson, J. M. Fini, and L. E. Nelson, “Space-division
multiplexing in optical fibres,”Nat. Photonics 7, 354–362 (2013).
105. L. W. Luo, N. Ophir, C. P. Chen, L. H. Gabrielli, C. B. Poitras, K.
Bergmen, and M. Lipson, “WDM-compatible mode-division multi-
plexing on a silicon chip,”Nat. Commun. 5, 3096 (2014).
238 Vol. 7, No. 2 / February 2019 / Photonics Research Review
106. Y. Ding, V. Kamchevska, K. Dalgaard, F. Ye, R. Asif, S. Gross, M. J.
Withford, M. Galili, T. Morioka, and L. K. Oxenløwe, “Reconfigurable
SDM switching using novel silicon photonic integrated circuit,”Sci.
Rep. 6, 39058 (2016).
107. B. Wohlfeil, C. Stamatiadis, L. Zimmermann, and K. Petermann,
“Compact fiber grating coupler on SOI for coupling of higher order
fiber modes,”in Optical Fiber Communication Conference and
Exposition and the National Fiber Optic Engineers Conference
(OFC/NFOEC) (Optical Society of America, 2013), paper OTh1B.2.
108. Y. Lai, Y. Yu, S. Fu, J. Xu, P. P. Shum, and X. Zhang, “Compact
grating coupler for higher-order mode coupling,”in Optical Fiber
Communication Conference (OFC) (Optical Society of America,
2018), paper Th2A.6.
109. Y. Lai, Y. Yu, S. Fu, J. Xu, P. P. Shum, and X. Zhang, “Compact
double-part grating coupler for higher-order mode coupling,”Opt.
Lett. 43, 3172–3175 (2018).
110. M. Zhang, H. Liu, G. Li, and L. Zhang, “Efficient grating couplers for
space division multiplexing applications,”IEEE J. Sel. Top. Quantum
Electron. 24, 8200605 (2018).
111. J. H. Song, F. E. Doany, A. K. Medhin, N. Dupuis, B. G. Lee, and
F. R. Libsch, “Polarization-independent nonuniform grating couplers
on silicon-on-insulator,”Opt. Lett. 40, 3941–3944 (2015).
112. W. S. Zaoui, A. Kunze, W. Vogel, and M. Berroth, “CMOS-compatible
polarization splitting grating couplers with a backside metal mirror,”
IEEE Photon. Technol. Lett. 25, 1395–1397 (2013).
113. J. Zhang, J. Yang, H. Lu, W. Wu, J. Huang, and S. Chang,
“Subwavelength TE/TM grating coupler based on silicon-on-insulator,”
Infrared Phys. Technol. 71, 542–546 (2015).
114. X. Chen and H. K. Tsang, “Polarization-independent grating cou-
plers for silicon-on-insulator nanophotonic waveguides,”Opt. Lett.
36, 796–798 (2011).
115. D. Taillaert, H. Chong, P. I. Borel, L. H. Frandsen, R. M. De La Rue,
and R. Baets, “A compact two-dimensional grating coupler used as
a polarization splitter,”IEEE Photon. Technol. Lett. 15, 1249–1251
(2003).
116. F. Van Laere, W. Bogaerts, P. Dumon, G. Roelkens, D. Van
Thourhout, and R. Baets, “Focusing polarization diversity grating
couplers in silicon-on-insulator,”J. Lightwave Technol. 27, 612–
618 (2009).
117. L. Carroll, D. Gerace, I. Cristiani, S. Menezo, and L. C. Andreani,
“Broad parameter optimization of polarization-diversity 2D grating cou-
plers for silicon photonics,”Opt. Express 21, 21556–21568 (2013).
118. W. Bogaerts, D. Taillaert, P. Dumon, D. Van Thourhout, R. Baets,
and E. Pluk, “A polarization-diversity wavelength duplexer circuit in
silicon-on-insulator photonic wires,”Opt. Express 15, 1567–1578
(2007).
119. S. Pathak, M. Vanslembrouck, P. Dumon, D. Van Thourhout, and
W. Bogaerts, “Compact SOI-based polarization diversity wavelength
de-multiplexer circuit using two symmetric AWGs,”Opt. Express
20, B493–B500 (2012).
120. L. Carroll, D. Gerace, I. Cristiani, and L. C. Andreani, “Optimizing
polarization-diversity couplers for Si-photonics: reaching the −1dB
coupling efficiency threshold,”Opt. Express 22, 14769–14781
(2014).
121. C. Lacava, L. Carrol, A. Bozzola, R. Marchetti, P. Minzioni, I.
Cristiani, M. Fournier, S. Bernabe, D. Gerace, and L. C. Andreani,
“Design and characterization of low-loss 2D grating couplers for sil-
icon photonics integrated circuits,”Proc. SPIE 9752, 97520V (2016).
122. J. Zou, Y. Yu, and X. Zhang, “Single step etched two dimensional
grating coupler based on the SOI platform,”Opt. Express 23,
32490–32495 (2015).
123. J. Zou, Y. Yu, and X. Zhang, “Two-dimensional grating coupler with a
low polarization dependent loss of 0.25 dB covering the C-band,”
Opt. Lett. 41, 4206–4209 (2016).
124. R. Halir, D. Vermeulen, and G. Roelkens, “Reducing polarization-
dependent loss of silicon-on-insulator fiber to chip grating couplers,”
IEEE Photon. Technol. Lett. 22, 389–391 (2010).
125. Y. Luo, Z. Nong, S. Gao, H. Huang, Y. Zhu, L. Liu, L. Zhou, J. Xu,
L. Liu, S. Yu, and X. Cai, “Low-loss two-dimensional silicon pho-
tonic grating coupler with a backside metal mirror,”Opt. Lett. 43,
474–477 (2018).
126. H. Y. Hwang, P. Morrissey, J. S. Lee, P. O’Brien, J. Henriksson, M. C.
Wu, and T. J. Seok, “128 × 128 silicon photonic MEMS switch package
using glass interposer and pitch reducing fibre array,”in IEEE 19th
Electronic Packaging Technology Conference (IEEE, 2017), pp. 1–4.
127. E. R. Fuchs, E. J. Bruce, R. J. Ram, and R. E. Kirchain, “Process-
based cost modeling of photonics manufacture: the cost competi-
tiveness of monolithic integration of a 1550-nm DFB laser and
an electroabsorptive modulator on an InP platform,”J. Lightwave
Technol. 24, 3175–3186 (2006).
128. L. Carroll, J.-S. Lee, C. Scarcella, K. Gradkowski, M. Duperron, H.
Lu, Y. Zhao, C. Eason, P. Morrissey, M. Rensing, S. Collins, H.
Hwang, and P. O’Brien, “Photonic packaging: transforming silicon
photonic integrated circuits into photonic devices,”Appl. Sci. 6,
426 (2016).
129. “Cardis H2020 project,”http://www.cardis-h2020.eu.
130. M. Duperron, L. Carroll, M. Rensing, S. Collins, Y. Zhao, Y. Li, R.
Baets, and P. O’Brien, “Hybrid integration of laser source on silicon
photonic integrated circuit for low-cost interferometry medical de-
vice,”Proc. SPIE 10109, 1010915 (2017).
131. C. Scarcella, K. Gradkowski, L. Carroll, J. S. Lee, M. Duperron, D.
Fowler, and P. O’Brien, “Pluggable single-mode fiber-array-to-PIC
coupling using micro-lenses,”IEEE Photon. Technol. Lett. 29,
1943–1946 (2017).
132. P.-I. C. Dietrich, I. Reuter, M. Blaicher, S. Schneider, M. R. Billah, T.
Hoose, A. Hofmann, C. Caer, R. Dangel, B. Offrein, M. Möhrle, U.
Troppenz, M. Zander, W. Freude, and C. Koos, “Lenses for low-loss
chip-to-fiber and fiber-to-fiber coupling fabricated by 3D direct-write
lithography,”in CLEO: Science and Innovations (Optical Society of
America, 2016), paper SM1G-4.
133. P. I. Dietrich, M. Blaicher, I. Reuter, M. Billah, T. Hoose, A. Hofmann,
C. Caer, R. Dangel, B. Offrein, U. Troppenz, M. Moehrle, W. Freude,
and C. Koos, “In situ 3D nanoprinting of free-form coupling elements
for hybrid photonic integration,”Nat. Photonics 12, 241–247 (2018).
134. N. Lindenmann, G. Balthasar, D. Hillerkuss, R. Schmogrow, M.
Jordan, J. Leuthold, W. Freude, and C. Koos, “Photonic wire bond-
ing: a novel concept for chip-scale interconnects,”Opt. Express 20,
17667–17677 (2012).
135. N. Lindenmann, S. Dottermusch, M. L. Goedecke, T. Hoose, M. R.
Billah, T. P. Onanuga, A. Hofmann, W. Freude, and C. Koos,
“Connecting silicon photonic circuits to multicore fibers by photonic
wire bonding,”J. Lightwave Technol. 33, 755–760 (2015).
Review Vol. 7, No. 2 / February 2019 / Photonics Research 239
