PCSELs are the future of laser production
PCSELs are a new class of semiconductor laser. They utilise a 2D grating structure which scatters light linearly and orthogonally. This makes PCSELs the only laser where feedback is in-plane and light emission is out of plane, emanating from the laser’s top surface.
Out of plane, orthogonal, surface emission offers a huge cost advantage for lasers, as it makes them easy to package and incorporate into PCBs and electronic assemblies. Out of plane emission is enabled and stabilised by our 2D grating structure (photonic crystal), creating feedback and single mode emission. The PCSEL structure provides advantages in data rate, wavelength and power performance when compared to equivalent EELs or VCSELs.
Vector Photonics’ PCSELs deliver low cost, high-volume, robust build, broad wavelength range and high power.
“Our PCSELs produce the speed performance of EELs and VCSELs, whilst their tested and packaged cost is 50% that of EELs and they deliver over 10x the power of VCSELs.”
Neil Martin, CEO, Vector Photonics
The laser technology evolution.
Fabry Perot lasers are the original, semiconductor laser technology. The laser feedback and emission are both in-plane, so light comes out of the end of the laser and the gain reflection is produced by facet mirrors. DFB lasers also have in-plane feedback and emission, but this time the gain reflection is produced with a grating structure.
VCSEL technology has out of plane gain and emission, where the light emits from the top surface of the laser. This makes both the test and packaging of the lasers much cheaper than EELs.
The PCSEL is the only laser using in-plane feedback and out of plane, surface emission. Test and packaging remain cheaper, like VCSELs, but the PCSEL structure provides advantages in data rate, wavelength and power performance when compared to equivalent sized EELs or VCSELs.
PCSELs – low cost with high speed and power – 3 in one!
PCSELs have it all! They deliver low cost with high speed and power. All other current laser technologies offer only two of these three, key characteristics in one device.
VCSELs (Vertical Cavity Surface Emitting Lasers) compromise power for low cost and high speed. EEL’s (Edge Emitting Lasers), including FP (Fabry-Perot) and DFB (Distributed Feedback) lasers offer speed and power, but are expensive to make.
PCSELs have other advantages too. They emit light from the top surface, like VCSELs, making them easy to package and incorporate into PCBs and electronic assemblies. They are also made in a similar way to EELs, so existing, experienced, supply chain capability and capacity can be utilised in their production. Vector Photonics’ PCSELs can be made at any wavelength, so can be used to address a wide breadth of applications.
cost and ease of assembly
EEL Lasers (Edge Emitting Lasers) have been in production for more than 40 years, where they have proven their reliability and longevity in telecoms and data systems.
FP (Fabry-Perot) and DFB (Distributed Feedback) lasers are both types of legacy, semiconductor EELs. EELs offer high levels of single-mode performance, both from optical spectrum range and power perspectives. However, EELs have two significant disadvantages. The first disadvantage is that they must be precisely aligned and handled to be integrated into systems. This is because single-mode light is emitted from the edge, not the front, meaning the lasers must be precisely aligned within subassemblies to re-direct the light into the correct direction for optical fibres or free space. The second disadvantage is that they require complex manufacturing and testing processes. The semiconductor wafers must be split into bars and finished on each side with reflective coatings. Each laser must be tested at bar level before being ‘singulated’ into individual laser devices for system integration. These multiple processing and testing steps increase cost and reduce yield.
wavelength range and power
VCSELs (Vertical Cavity Surface Emitting Lasers) were first produced in the early 1990’s. They have limited operational wavelengths due to the manufacturing challenges caused by the various material systems required for multi-wavelength operation.
A VCSEL is produced by having two Bragg stacks above and below the active region of the laser. The Bragg stack comprises layers of two materials of differing refractive index ‘grown’ on top of one another. The number of layers, and therefore periods and the refractive index contrast, gives rise to the reflectivity. A high number of periods gives high reflectivity, and high index contrast increases reflectivity per layer.
The VCSEL grating structure also has inherent limitations to the single mode, power levels that can be produced. So, although VCSELs can achieve high speeds and can be produced cost-effectively, their limited single mode performance makes them unsuitable for high-speed datacoms and long-distance telecoms. These limitations also restrict their use in sensing applications to relatively short distances.
Laser technology comparison chart
Why making a PCSEL is so much easier than a VCSEL.
The materials used in the Bragg stack of a VCSEL and their corresponding wavelengths determine the stacks dimensions and properties. The longer the wavelength, the thicker each period of the Bragg stack will be and the larger it becomes overall. If the two materials used have a small ∂n, then the number of periods, or stack layers, becomes higher, increasing resistance and limiting power.
For optimal performance, the two materials used in a Bragg stack should have over 99% reflectivity. VCSELs have limited operational wavelengths due to the manufacturing challenges caused by the various material systems required. In fact, in practise, no suitable pairs of lattice-matched compounds, with a high index contrast, are available for Indium Phosphide (InP); the material associated datacoms and telecoms wavelengths.
PCSELs have the following advantages over VCSELs:
- They do not use a Bragg stack structure, like a VCSEL (or indeed a Facet coating like DFBs), they instead have a 2D grating which is easy to scale.
- They have low contact resistance, ensuring the highest power output of any laser, size for size.
- They can be made in any material, so are suitable for a wide range of wavelengths and applications.
- Their power is single mode, which will scale up proportionally with area, easily enabled by their 2D construction and in-plane feedback.
- They have no Oxide aperture, as used by VCSELs, so do not require specialist oxidation furnaces in manufacture.
Why 1310nm and 1550nm wavelengths are critical in fibre optics.
Single-mode, fibre optic cables have two wavelength windows which offer high performance for different applications. These occur at 1310nm, the optimal datacoms transmission wavelength and at 1550nm, the optimal telecoms transmission wavelength.
At 1310nm, dispersion in a single-mode, fibre-optic cable is at its minimum. This means a pulse of light transmitted through the fibre will arrive at its destination at mostly the same time and relatively intact. This is important because this intrinsically low dispersion transmission puts less demand on a semiconductor laser for coherence. The downside is that there is greater attenuation, or power loss, in the cable at this wavelength, hence 1310nm is used in datacoms applications where distances are shorter.
At 1550nm, attenuation, or loss, in the fibre optic cable is at its minimum. This makes it the ideal wavelength for long-distance transmission, such as telecoms, where distances of 200km or more are achieved before reamplification. The downside is that there is greater dispersion, meaning a compound semiconductor laser with greater coherence is needed for good system receiver performance.
PCSELs offer a serious breakthrough for all communications applications as they can provide more power and coherence than any other existing laser devices. This means they can be used to increase existing transmission distances, which is especially important in datacoms at 1310nm, and they improve coherence, which is especially important for telecoms at 1550nm.
The performance benefits of PCSELs
The 2D grating structure used in PCSELs scatters light linearly, in plane, and orthogonally, out of plane. The out of plane, surface emission offers huge cost and performance advantages for lasers. It makes them easy to manufacture, test, package and incorporate into electronic assemblies. This gives rise to some critical performance features of PCSELs, which are:
- Coherence and coherent arrays
- Solid-state array, beam steering
- Wavelength flexibility
- High speed and data rates
Coherence and coherent arrays
The in-plane feedback in PCSELs is two-dimensional and facet free, enabling coherently coupled arrays. The laser elements of the array are joined by a coupler region. This means the in-plane light is linked between laser elements, creating coherence. Using lenses, coherent light can be focused down to a small spot with greater light density, which has advantages in cutting, welding, melting, engraving and drying applications.
Due to the laser elements of the array being coupled in plane, in an ‘n x n’ array, power scaling is also possible. High brightness and the unique geometry enable kilowatts of coherent power, which is simply not achievable with any of the other laser technology.
Solid-state array, beam steering
PCSELs can have an array structure where the coupler region can be controlled in what is called an ‘optical phased array’. By electronically tuning the phase of the coupler region, the laser beam can be steered, in real time, without moving parts. This makes PCSELs applicable to rapidly evolving LiDAR applications, where the beam is steered for imaging. High power, 3D printing in metal and plastic are other applications that could be revolutionised. As a solid-state solution, system size is reduced by 10x, whilst reliability is increased over existing mechanical solutions.
Wavelength flexibility and speed
Semiconductor materials inherently emit different wavelengths (colours) of the light. The structure of PCSELs allows them to be made readily in any gain material to produce any colour or wavelength, yet they still retain all the advantages explained in this article.
Other laser types must be made in specific materials. For example, VCSELs have been demonstrated in high volume in GaAs at 850nm but are difficult to make in InP at 1310nm and 1550nm, so cannot be used reliably for high-speed datacomms.
PCSELs permit multi-Gb modulation rates, with huge data speeds.
To achieve the high speed 1310nm datacoms and 1550nm telecoms wavelengths, the mode volume of a semiconductor device must be minimised, by reducing its size. EELs have a large mode length and a small mode height. VCSELs have a small mode length, but a comparatively large mode height, because the mode penetrates the Bragg stack. So, although VCSELs can achieve high speeds and can be produced cost-effectively, their limited single mode performance makes them unsuitable for high-speed datacoms and long-distance telecoms. These limitations also restrict their use in sensing applications to relatively short distances.
PCSELs have a mode width and length like a VCSEL, but the mode height of an EEL. This means that for the equivalent emission area of a VCSEL, the PCSEL can be up to 2.5 times faster than VCSELs and 3x faster than a high-speed EEL.