3.7 Polymer Materials

Polymer (organic) based technologies for optoelectronic and photonic applications have been developed by both academia and industry over the past 40+ years and are currently in the process of commercialization with the first announcements of active polymers occurring in 2023. 

Both active and passive polymers have been commercialized and made into products by several companies globally. The key applications for photonic based active electro-optic polymers are fiber optic communications, namely telecommunications, data communications, high performance computing, where fiber optic interconnects and links are utilized. The key applications for passive polymers such as polymer waveguides, plastic optical fiber, and two photon polymermization (TPP) are fiber optic communications, automotive, medical, display, and consumer. 

The whole industry today (material vendors, semiconductor foundries, various chip suppliers, standards organization, test equipment vendors, package houses) are growing their business for line rates of 100G, and gearing up for line rates of 200G.  These line rates are being used to generate products such as the optical pluggable transceiver for aggregate data rates of 400Gbps, 800Gbps, 1600Gbps (or 1.6Tbps), 3200Gbps, 6400Gbps and beyond.  While companies are looking to complete the designs for 200G line rates in 2023-2024 timeframe, they are also looking how to extend the data rates and line rates further.  Interest in line rates of 300G and 400G are still at an early stage, however, active electro-optic polymers offer modulator device demonstrations today that achieve these goals.  Passive polymers and POF will be able to support these extended line rates with waveguides and interface coupling components as part of a subsrate platform. Commercializaiton of these performance metrics by both active and passive polymers is expected over the next decade. 

The obvious, yet most difficult next move for device designers of modulators is to revisit increasing the optoelectronic device speed as semiconductor solutions are stuggling to achieve the bandwidth, low power and small size needed for the markets.  Active electro-optic polymers can exceed the semiconductor incumbent modualtors by nearly an order of magnitude on 3dB bandwidth as well as power consumption with sub 1V operation.  While incumbent semiconductor modulators are operating fiber optic communications traffic with optical 3dB bandwidths in the range of 30-40 GHz (to generated data speeds of 50Gbps NRZ and 100GBaud PAM4), active electro-optic polymer modulators have demonstrated optical 3dB bandwidths that are in excess of 250GHz using plasmonic slot designs, and silicon slot modulators similarly with over 100GHz. 

Further, speed is not the only issue for modulators: low power and size are also important.  The demands by datacenter designers to lower power consumption and maximize space efficiency are critical metrics for the architecture.  The good news is that active electro-optic polymer slot modulators either plasmonic or silicon slot are very small in footprint with less than one square millimeter. As part of a PIC platform, many modulators can be arrayed subject to the architecture design of for example: 4 lane, 8 lane, 16 lane PIC designs.  Active electro-optic polymer slots have been demonstrated to operate with very low voltage levels of less than 1V (which means that the modulators themselves not only save power but can be directly driven by normal electronics and thereby eliminiating the driver chips). The potential elimination of driver chips means that the architectures save more power than the modulator itself – which is a strong differential advantage for low power designs.   When these performance metrics of low power and small size footprint are added to the natural high optical bandwsith of modulators designed with active electro-optic polymers, commercialization opportunities look extremely promising over the next decade. 

Today’s roadmaps are demanding photonic components with increased performance.  With the latest results from seminconductor modulators stretching to operate at 50-70GHz, while suffering higher drive voltage, the need for 100GHz bandwidth solutions to achieve 100G, 200G line rates with cleaner optics is increasing. Already, companies are looking at lowering power consumption in datacenter switches and routers by eliminating the DSP chip and directly driving the optics from ASIC switch chips. While this movement is still at an early stage, the network system architects are looks creatively to save power consumption both from the electronics as well as the optics. Faster optics, i.e., cleaner signals for NRZ, PAM4 using higher bandwidth optical modulators, operating with sub 1V voltage levels for lower power consumption is quickly becoming key for next generation optical network systems. 

Unlike conventional modulator materials such as InP, Lithium Niobate, silicon photonics and GaAs, the polymer material system is naturally fast. There are now companies that are commercially designing commercial high-speed optical modulators made from electro-optic polymers that will be capable of line rates of 200G beyond using either discrete modulators or arrayed modulators in a PIC with 3dB bandwdiths that easily exceed 70GHz and can range up to 250GHz. 

Several commercial companies have been able to prove the maturity of planar polymer waveguides for passive light guiding. The polymers provide low loss and further good optical quality to enable fast optical signaling on-board and board-board via optical backplanes. The availability of single-mode polymer waveguides to support cost-effective pluggable optical transceiver packages as well as co-packaged platforms with and enables PIC platforms such as silicon photonics.

In polymer based optical fibers (known as polymer optical fiber (POF), optical attenuation is a key metric, and recent maturity in terms of dB/km is enabling short distance interconnects to be designed with low cost for volume consumer markets.  While POF is larger in diameter to glass fibers, it does not need to be cleaved, and can be installed quickly and maintained efficiently for small networking environments.

1. ^[1] ETH Zurich, Polariton, CA?U University Kiel, EU Horizon 2020, LWLG
2. ^[2] Polariton, ETH Zurich, LWLG
3. ^[3] Traditional polymer optical slot modulator application as designed for example from Polariton, LWLG

The polymer roadmap that displays the purple brick walls is shown below in Figure 1 and Figure 2.  The first figure shows the individual line purple brick walls for each technology segment. The following figure shows the purple brick wall areas that require focus to addesss the challenges in order to achieve the metrics laid out in the polymer roadmap. 

The main roadmap challenge for the communications industry is to plan future data line rates that exceed 200G and extend towards 300G and 400G.  These line rates are needed to achieve pluggable transceiver and/or co-packaged transceiver aggregate data rates of 400Gbps, 800Gbps, 1600Gbps (or 1.6Tbps), 3200Gbps, 6400Gbps and beyond.  While companies are looking to complete the designs for 200G line rates in 2023-2024 timeframe, they are also looking how to extend the data rates and line rates further.  Interest in line rates of 300G and 400G are still at an early stage, however, active electro-optic polymers offer modulator device demonstrations today that achieve these goals.

 Several vendors are looking at how to address 200G line rates, and are planning spatial multimplexing (adding fibers), wavelength division multiplexing, (adding wavelengths in a single fiber), encoding with more complex symbols per bit (PAM, QAM etc), designing optical and electrical devices for higher bandwidth, lower voltage (power), and smaller footprint (size) as part of a PIC platform.  These higher performance drivers are being accelerated by data hungry customers such as data centers, high performance computing, and shorter reach telecommuications.  Figure 3 below shows the trends in optical pluggable transceivers for the data communications market and range out to 3.2Tbps using 16 channels and 200G line rates.

A particularly difficult challenge is to increase an optical device’s analog bandwidth to that beyond 100GHz, while keeping the power consumption very low. This is further made complex as the modulation signal will need to be driven with similar data rates electronically from electronic drivers for the transmitter, and received by associated detection circuitry, transimpedance amplifiers etc., at the receive end of the optical interconnect. With the drive in 2023 to potentially eliminate the DSP chip, network archtiects are not only looking to polymers for the optical solution, but to combine cleaner, faster polymer optics with more power efficient electronic designs called Linear Drive Optics (LDO) that potentially could drive the optics directly from ASIC switch chips as opposed to DSP and driver chips.  Figure 4 below shows the perspective from a network systems company that is looking to implement LDO to save not only cost but power consumption in the network.

1. ^[1] Arista presentation from OFC 2023 for elimination of DSP using Linear Drive Optics (LDO)
2. ^[1] Arista presentation from OFC 2022 for OSFP pluggable optical transceivers

In the field of active electro-optic polymers, there has been significant technological and commercial progress.  There are now commercial companies that are supplying active electrop-optic polymer material as well as commercial companies supplying silicon slot and plasmonic slot modulator devices, and devices embedded into photonic integrated circuit (PIC) platforms. 

With the natural ability for active elctro-optic polymers to be operated at very fast optical switching and become part of the popular trend of design PIC based solutions, optical devices such as modulators have the potential to drive this market opportunity quickly over the next decade.

The performance of electro-optic active polymers in the last few years has been incredible.  While incumbent semiconductor modulators are operating fiber optic communications traffic with optical 3dB bandwidths in the range of 30-40 GHz (to generated data speeds of 50Gbps NRZ and 100GBaud PAM4), active electro-optic polymer modulators have exceeded these bandwidths by an order of magnitude.  Plasmonic slot modulators have demonstrated optical 3dB bandwidths that are in excess of 250GHz as seen in Figure 5 and plasmonic ring resonator slot modulators similarly with over 100GHz as seen in Figure 6 and the table in Figure 7. This approximately translates to line rates that not only exceed 200G PAM4 which is where the datacomunications and datacenter industry is today in 2023 but offers line rate potential of 300G and even 400G in the near future.  These line rates will enable pluggable optical transceivers to be designed with aggregate data rates of 800Gbps, 1600Gbps (or 1.6Tbps), 3200Gbps, 6400Gbps and beyond as can bee seen in Figure 8.

1. ^[1] LWLG at https://www.lightwavelogic.com/presentation/8th-photonic-integrated-circuits-pic-international-conference/
2. ^[2] Polariton (post deadline paper ECOC 2022), LWLG 
3. ^[3] ETH Zurich, Polarition, LWLG, https://doi.org.10.1038/s41566-023-01161-9

Unlike conventional modulator materials, the polymer material system for both active and passive solutions have natural abilities that is attracting attention from the fiber optic communication industry.  Several commercial companies are currently designing commercial high speed optical modulators made from electro-optic polymers that will be capable of 200G line rates using 3dB bandwidths more than 70GHz. Technical data showing 3dB bandwidths that exceed 250 GHz (corresponding to approximately 3-400G line rates) have already been demonstrated, which indicates even higher speeds should be possible in future. The polymer material system also has the potential of extremely low voltage levels that can reach below 1V. Furthermore, the polymer material system has the potential for small device size and in fact, small enough to fit into standard fiber optic transceiver form factors that are used for 100G today, and 200G communications that are emerging in the 2023 and 2024 timeframe.

The polymer material breaks through the technology barrier and opens up new possibilities. Two examples of solutions that could inhabit the new design space are shown in green below.  It is expected that the industry roadmap will be extended to >100GHz bandwidth devices in the next few years, and >200GHz over the next decade.

1. ^[1] Source: M. Lebby. [Online]. https://www.lightwavelogic.com/presentation/2023-annual-shareholder-meeting-management-presentation/