GLOBAL

INNOVATION
REPORT

Social Contribution by Photonics

Integration Circuit Technology
Latest Technology Trends and VLC Photonics S.L. (Spain) Initiatives
Iñigo Artundo David Doménech José Galán
Chief Executive Officer, Chief Technology Officer, Sales and Marketing Manager,
VLC Photonics S.L. VLC Photonics S.L. VLC Photonics S.L.

instruments, healthcare wearables and diagnostic

1 | Introduction lab-on-a-chip devices (see Figure 2).
These PICs use silicon as substrate material as
Electronic integrated circuits have revolutionized the well, although they can use other material platforms*1,
world by embedding miniature components in hun- each having its own advantages and disadvantages.
dreds of millions of devices at different markets since All the technology and processes developed over
the 1960’s, when basic components like transistors, the years for the micro-electronics industry can now
resistors or capacitors could be integrated into a be reused for PICs, saving costs and accelerating
monolithic silicon wafer die. These tiny chips could time-to-market. This is making that the adoption for
be mass produced by the millions using lithography them is growing at an amazing pace in some mar-
at a very low cost, enabling to embed multiple func- kets, and investment on the technology is flourish-
tionalities and complex architectures into very robust ing globally.
and reliable microelectronic components. Many semiconductor business models are being
In a similar fashion, but a few decades later, a adopted in photonics as well, like the pure-play found-
similar revolution is coming with the possibility of ries that fabricate for fabless companies, so foundries
miniaturizing optical components into photonic inte- amortize faster on their investment by loading their
grated circuits (PICs) at low cost and high volumes. lines as much as they can with many customers,
Photonic components like light emitters (lasers), light while these don’t need to invest huge amounts in
detectors (photodiodes) and others (modulators, fil- setting up a fab in a clean-room. Other models are
ters, etc.) can now be integrated as well into mono- that of design houses, providing outsourced com-
lithic PICs (see Figure 1). Such possibility enables ponent design and circuit layout, or the outsourced
all the advantages of integration for applications assembly and test companies that take care of the
where such optical components are used, for exam- very challenging back-end, much more complex than
ple fiber optic communications (both long-distance in electronics due to the 100x size reduction between
telecom networks and short distance datacenter electrical and optical input/output ports.
links), laser-based radar [light detection and ranging There are many good reasons why PICs are being
(LiDAR)] mainly for autonomous vehicles, quantum adopted by many industries:
computing and cryptography, and light-based medical *1 E.g. indium phosphide (InP), silicon nitride (SiN), silicon dioxide (SiO2), etc.

Hitachi Review Vol. 71, No. 4 438–439 29.

 GLOBAL INNOVATION REPORT

Figure 1 | Example PIC with Different Integrated Optical Elements (Center), Optical Waveguides (Left),
and Dimensions of a Typical Chip (Right)

Wavelength filters Photodetector

100 nm Electro-optic modulators

Vertical
couplers to
optical fiber

Example of one PIC 5 to 10 mm

PIC: photonic integrated circuit

(1) At large data centers, like those of the Big Five Figure 2 | PIC Applications
tech companies, PIC-based transceivers establish
optical communications among switches and com-
puting racks, contributing to increase data transmis-
Information Telecom
sion speeds and save power consumption to reduce technology
carbon dioxide (CO2) emissions.
(2) For autonomous vehicles like cars, trucks or
drones, and also robots, PICs enable miniaturized
Optical sensing,
laser ranging systems, gyroscopes and laser radar lighting, and Photonic integration Healthcare and
energy displays technologies life science
(also known as LiDAR). These allow to reach the
compact and light form-factor required, while being
very robust and adequate for harsh environments.
They will be at the front seat of the upcoming self- Security and
Manufacturing
defence
driving revolution, allowing for a safer society with
more efficient transportation and thus with lower
CO2 emissions.
(3) Quantum computing and machine learning for
artificial intelligence (AI) are new computing para- based on PICs. Simple diagnostic devices for dis-
digms that vastly outperform classical computers eases and other bio-markers, or healthcare moni-
and algorithms for some specific tasks key to our tors for e.g. blood pressure or glucose, can be
society, like cryptography for security and banking, augmented with photonic technology to evaluate
chemical drugs research, big data analysis, image and more than one parameter and in a more precise way.
language processing, etc. Again, PICs can provide a Deoxyribonucleic acid (DNA) sequencing can also be
faster and more power efficient way of implementing made faster and more affordable by making use of
some algorithms, and in some cases, the only way to PICs for light processing to the samples.
implement certain quantum processes not available
in electronics. 2 | Applications
(4) Bulky and expensive medical instrumentation, like
optical coherence tomography (OCT), can now be The next sections describe more in detail some of the
miniaturized into hand-held affordable instruments example applications that have been presented above.

30.
2.1 Optical Communications control electronics, and overall circuit modelling,
Silicon PICs and other variants like SiN or III-V mate- which demand concerted efforts to be addressed.
rials (mostly InP), have been attracting considerable Today, there is a strong movement towards co-pack-
interest as they encourage the fabrication of very aging and on-board optics. These technologies need
complex photonic chips for high-performance optical to be able to support faster photonic devices such as
communications. PICs have the potential for enabling >80 GHz silicon modulators that can operate at very
high density, high data rates and flexible solutions high speeds*2. There are other modulator technolo-
not only in optical telecommunication networks, but gies that are being developed with both high speed
also in datacenter communications. From intra-chip and low power potential that include: polymer, slot,
to chip-to-chip and chip-to-board, PIC technology can plasmonic, thin-film lithium niobate, or barium tita-
provide significant advantages. nate among others.
It is expected that PICs must achieve not only Overall, data centers need faster optical devices,
increased performance, but effective reliability and higher integration levels, higher reliability, lower
economic scalability to become competitive in those power consumption, a higher degree of scalable
markets. PICs, as opposed to discrete optical com- economics that address $/Gbps metrics, and smaller
ponents like individual lasers or photodiodes, com- more miniaturized foot-print platforms. Co-packaged
bine more than one optical function on a single chip. solutions with hybrid integration between different
The functions can be active functions such as optical technologies are required. The integration of electron-
signal generation, optical modulation, or photode- ics, photonics, and packaging is an important need
tection, as well as passive functions such as opti- that the industry is addressing today and will explore
cal signal routing or filtering. PICs provide a means this in more intensity over the next decade.
for significantly reducing system size, weight, and
power, improving reliability, and increasing system 2.2 LiDAR for Autonomous Driving
performance and functionality. Traditionally, optical LiDAR, short for light detection and ranging or “light
communication systems like transceivers are built radar,” is used to measure distances with high reso-
from discrete components and the components are lution and precision by illuminating the object with
connected together using fiber connections. This is a scanning laser beam, and consequently measur-
cumbersome in that each time light is coupled from a ing the reflections. A LiDAR basically consists of a
device to an optical fiber, or from an optical fiber to a laser source, a laser beam scanner, and the detec-
device, signal power is reduced. PICs can overcome tion optics. Various techniques are used to combine
these issues by connecting components on the chip these elements into a functional system, such as
using optical waveguides. They can realize novel sys- pulsed, flash and frequency-modulated LiDAR. As an
tem architectures employing advanced modulation example, many companies are working on bringing
formats where systems built from discrete compo- a PIC-based frequency modulated continuous-wave
nents might be too complex and negatively impact (FMCW) LiDAR to the market, that can detect not only
size, weight, and power requirements. the distance, but also the speed of target objects.
However, like any other emerging technologies, The opportunities for using LiDAR are plenty, most
PIC design comes with several specific challenges notably in automotive, where it is used in advanced
such as thermal management, photonic packaging, driver-assistance systems (ADAS) and in autonomous
radio frequency interface management for high-speed *2 100 Gbps NRZ or 200 Gbaud PAM-4.

Hitachi Review Vol. 71, No. 4 440–441 31.

 GLOBAL INNOVATION REPORT

driving, providing a better resolution than radar. Over powerful supercomputers cannot solve. Rather than
the last decade, pushed by the automotive market, store information using bits, quantum computers use
the price of LiDAR systems has come down by more quantum bits, or qubits, to encode information as 0s,
than an order of magnitude. 1s, or both at the same time. This superposition of
PICs offer new opportunities for cost-efficient states enables quantum computers to manipulate
LiDAR, as they can allow for on-chip high density enormous combinations of states at once.
integration of the optical components required and Quantum computers are exceedingly difficult to
can be massively manufactured in volumes. PICs engineer, build and program. As a result, they are
can be used as the LiDAR laser source: when com- crippled by errors in the form of noise, faults, and
bined with the on-chip components typically used in loss of quantum coherence, which is crucial to their
communications technology, pulsed lasers and fre- operation. This loss of coherence (called decoher-
quency-modulated lasers can be realized. An added ence), caused by vibrations, temperature fluctuations,
advantage is that PICs are typically designed for electromagnetic waves, ultimately destroys the quan-
operation around the telecom wavelengths of 1,310 tum properties of the computer.
and 1,550 nm, which is a sweet spot for LiDAR due PICs can overcome many of these limitations
to eye-safety requirements. This allows the range to by implementing the core functionalities of future
be extended significantly, to over 200 m. quantum computers. A sampling algorithm, a proce-
PICs can also be used to replace the beam-steering dure that takes a random sample of data, at a speed
part of the LiDAR, with optical phased arrays. Such beyond the reach of classical computers, has been
optical phased arrays can shape the laser beam and implemented in a PIC on a silicon nitride chip that is
steer it fast for video-rate three-dimensional imag- compatible with the fabrication processes used by
ing. Laboratory based implementations have already the semiconductor industry(3).
shown the feasibility, and the technology is now mov-
ing to the market. Sources and detectors can be 2.4 Security
integrated on the beam-steering PIC as well, with In the wake of the fourth industrial revolution, when
the potential of realizing a fully integrated, single-chip data is collected, transferred, and stored in networks
LiDAR, and allowing unprecedented high volumes at at a global scale, cybersecurity and cryptography
low cost. Moreover, this technology allows for close are of utmost importance. Technologies such as the
integration with electronics. Internet of Things (IoT), AI or the blockchain increase
As an example, a chip-scale FMCW LiDAR system the data traffic across the networks from their gen-
that incorporates wavelength tuning and phased array eration locations to cloud-based data repositories.
operation enables two-dimensional non-mechanical The daily operation of businesses, administrations,
beam steering(1). and individuals are increasingly confronted with the
transmission of sensitive—if not critical—data that
2.3 Computing require protection, notably against threats to its long-
Recent developments in chip-based photonic quan- term confidentiality.
tum circuits have radically impacted quantum infor- Modern cryptography provides fundamental tools
mation processing. One of the most promising to enable digital security in our society. Invented 40
(2)
applications is quantum computing . Quantum com- years ago, public-key cryptography enables secure
puters will solve complex problems that today’s most communication over open networks, i.e., the Internet.

32.
Figure 3 | Cleanroom Personnel Supporting VLC Photonics PIC Testing Services (Left)
and Technician Adjusting the Automated PIC Testing Tool (Right)

Computational cryptography however faces chal- and Multimedia Applications Institute (iTEAM) at
lenges: widely deployed public-key techniques are Universitat Politècnica de València (UPV), Spain.
vulnerable to the advent of a quantum computer. HVLC is a resident company in the UPV fab clean-
Cryptographic protocols relying on classical tech- room, the microfabrication research and develop-
niques are vulnerable to undetected eavesdropping ment and pilot line at UPV (see Figure 3). At first
*3
and to “store now-attack later ” attacks. By combin- the company started its business as a PIC design
ing physical layer technologies and quantum informa- company and it continued evolving to respond to
tion, quantum cryptography offers unique tools to the growing needs of PIC adoption. Now the com-
address these challenges. pany offers development services for customers in
The most prominent quantum cryptographic pro- many application markets, including the ones men-
(4)
tocol is quantum key distribution (QKD) , which tioned above on photonic communications (mainly
allows two remote network nodes to generate an fiber optics), autonomous driving, sensing, signal
encryption key with information theoretic security processing, biomedical, and quantum. HVLC offers
(ITS). Tremendous technological progress has been nowadays the following PIC development services:
achieved in the last two decades in QKD research, consultancy, design, fabrication, characterization/test,
culminating in the deployment of QKD over optical and packaging/assembly (see Figure 4)(6). The com-
fiber communication networks. QKD systems need pany also collaborates with Fortune 500 companies
to evolve technologically towards co-existence with and startups, and some world-famous researchers.
standard coherent optical communication systems Welcoming HVLC to the family, Hitachi High Tech
and towards miniaturization to achieve cost-effec- Group will further enhance its engineering services
tiveness and scalability. PIC components such as and will contribute to customers with its solutions.
(5)
quantum random number generators , QKD emitters
and receivers will thus make a leap towards QKD on 4 | Conclusions
a chip, which would lead to 100 times reduction in
price and 10 to 50 times reduction in system volume. Photonic integration technology is positioned to be
a key enabling technology for many high-tech appli-
3 | Hitachi High-Tech Services cations over the next years: from optical communi-
cation transceivers to LiDAR or gyroscope sensors
VLC Photonics S.L. (HVLC), a Hitachi High-tech for autonomous cars, cryptographic applications for
Group company since November 2020, was estab- banking and secure commerce and communications,
lished by researchers involved for over 20 years in
*3 An attack method that involves stealing and storing data that cannot
PIC technologies and photonic communication sys-
be decrypted at the present time, and then using advanced computers
tems, and spin off from the Telecommunications in the future to decrypt all data collected in the past.

Hitachi Review Vol. 71, No. 4 442–443 33.

 GLOBAL INNOVATION REPORT

Figure 4 | Cycle of the Service Provided by VLC Photonics

1 Consultancy 2 Design 3 Fabrication 4 Characterization and testing

• Analysis of PIC requirements • Custom PIC and building • PIC manufacturing via world • Test plan development
• Advise on PIC technology and blocks design leading foundry partners • Die and wafer level char-
platform • Circuit layout • Multi-project, engineering/ acterization and testing:
• Foundry and packager analy- • Design support pilot, and volume production Optical and electrical (DC
sis and selection • Process design kit (PDK) runs and RF) probing; Vertical and
• PIC design review towards development • All main photonic integrated edge PIC-to-fiber coupling;
testing and packaging platforms: III-V materials (InP, Automated data collection and
GaAs); Silicon photonics (SOI processing
• PIC development and produc-
tion plan estimates and SiNx); Silica/PLC • Root cause analysis (RCA)

5 Test assemblies 6 Packaging t e r iz a ti o n , t e s t

a ra c ing
• Standard low-cost test • Packaging design and devel- Ch 4
assemblies for easy functional opment via third parties
characterization, includ- • Packaging production ramp-
ing: Chip bonding into PCB up via high-volume partners

Fabrication
A ss e m bly
or ceramic carriers for DC
and RF operation; Electrical 5 3
wire bonding or flip-chip;
Connectorization; Thermal
management; Fiber array
pigtailing

InP: indium phosphide GaAs: gallium arsenide 1 2 6

SOI: silicon-on-insulator SiNx: silicon nitride
PLC: planar lightwave circuit PCB: printed circuit board Consultancy Design Packaging
DC: direct current RF: radio frequency

or medical instrumentation and healthcare wearable Iñigo Artundo, Ph.D.

VLC Photonics S.L.
devices. While there are many challenges to mature Worked as a reviewer for several scientific
and scale the technology to the level that electronic journals and national funding agencies.
Joined VLC Photonics in 2011, and is
integration has reached over the last decades, the currently the CEO from 2013. A member
of the high-level steering committee set
potential for growth is phenomenal, and there will up by the European Commission for the
establishment of the Quantum Technologies
be many other applications that benefit from this
Flagship program.
technology in the future too.
David Doménech, Ph.D.
VLC Photonics S.L.
Working on the design of integrated optic
circuits in InP/SiN/SOI technologies within
References
several European and national research
1) P. J. M. Suni et al., “Photonic Integrated Circuit projects. Co-founded VLC Photonics in 2011
FMCW Lidar On A Chip,” 19th Coherent Laser Radar and since then has worked as the CTO. In
Conference (Jun. 2018). 2012, awarded with the Intel PhD Honor
Program award.
2) IBM, “What is quantum computing?,” https://www.ibm.
com/quantum-computing/what-is-quantum-computing/
José Galán, Ph.D.
3) J. M. Arrazola et al, “Quantum circuits with many VLC Photonics S.L.
photons on a programmable nanophotonic chip,” More than 14 years of experience in
Nature 591, pp. 54–60 (Mar. 2021). technological projects management.
Extensive professional experience as head
4) I. Khan et al, “Towards continuous-variable quantum of the presales and business application
key distribution at GHz rates,” 5th International engineering team. Joined VLC Photonics
Conference on Quantum Cryptography, Tokyo (Sept. in 2011 and appointed Sales and Marketing
2015). Manager.

5) C. Abellan et al., “Quantum entropy source on an

InP photonic integrated circuit for random number
generation,” Optica, 3, pp. 989-994 (Sept. 2016).
6) VLC Photonics, https://www.vlcphotonics.com

34.