Management
President, CEO and Chairman is John Clark with 30 years in the industry. He was hired out of Berkeley labs by AMOCO and set up the Amoco Laser Company, which was involved in high-powered amplifiers. ATX was sold to Scientific Atlanta in 1996. John Clark joined iolon in November 2000.

VP of Operations is Tim Harris, a former Senior VP with disk drive manufacturer Seagate Technologies. He ran Seagate's Malaysian operations that produced 1 million thin film recording heads per day, as well as other worldwide operations.

Founder Dr John (Hal) Jerman has 25 years in the industry.

Director of Materials and also a founder is Pauline Church.

Funding
On February 26th 2001, iolon announced the completion of a $53 million strategic financing round led by Bowman Capital.

iolon's second round of funding included additional investments from first round investors Kleiner Perkins Caufield & Byers and Optical Capital Group, as well as new investments from Boston Millennia Partners, Corning Innovation Ventures, Doll Capital Management, Goldman Sachs Group, Integral Capital Partners, J. & W. Seligman, Banc of America Securities, CIBC World Markets, Credit Suisse First Boston, CrossBridge Venture Partners, Kalkhoven, Petit and Levin Ventures, SSB Page Mill Capital Partners, Techno-Venture and Wit SoundView.

Following the second round funding announcement, John Clark noted that iolon had raised more than $69 million in the previous seven months.

Technical resources
iolon was granted 8 patents following its spinout from Seagate and has 28 application filings. Seagate remains a major shareholder in iolon.

iolon works with two qualified, unnamed, MEMS suppliers and two qualified laser source manufacturers, again unnamed. The company also buys-in lenses and gratings.

iolon defines automated manufacturing as at least 60% use of its "pick and place" robot. iolon has equipped its 3,500 square feet, Class 100 clean room for production, and all builds are coming off the clean room production line. Robotic assembly will allow iolon to add capacity very easily instead of having to hire additional staff. The company expects to be producing "100s per month" by the end of 2001. The assembly line can handle 10,000 pieces per month.

iolon's facility is 56,000 square feet and employs approximately 100 staff, over 80% of who are engineers. iolon has been fortunate to attract great talent from many top companies and universities to form an elite team of professionals. Present facilities can handle a work force of up to 250 and 45 additional staff are currently being hired.

The customer requirement, for most applications, is a broadly tunable device with a minimum output power of several milliwatts. Although different applications have varying requirements, most customers would like to deploy one laser that will cover the entire C-Band, which is generally a range of between 30-35 nm. There are also requirements for L-Band devices. At this point in time, potential customers prefer to have distinct devices addressing the C- and L-bands separately.

For long-haul markets (defined as anything over 200 km) at rates of 2.5 Gbit/s or greater, the signal is typically externally modulated with a CW laser source, since the chirp and resulting nonlinear effects of direct modulation are unsuitable for transmission at these distances and bit rates. The preferred material for external modulation of broadly tunable lasers at 10 Gbit/s and higher is lithium niobate (LiNbO3) since its modulation characteristics are independent of wavelength. A number of established companies, such as Lucent and JDS Uniphase, as well as some promising startups like Codeon, are supplying LiNbO3 modulators.

Design advantages
iolon sees the inherent advantage of its product design as providing the required power, tuning range, and spectral fidelity in a device that is designed for highly automated volume production. When modulating with LiNbO3 a minimum of 7-10 mW of output power is necessary to overcome the losses of the modulator and still achieve the desired launch power of ~1mW. Therefore, a broadly tunable device that only delivers 2 mW or 4 mW, does not meet the specification, as at 10 Gbit/s these devices will be used exclusively with LiNbO3. With a target output power of 20mW and tuning range of 35 nm, iolon's Apollo offers both the required power and the desired tuning range to provide transmission flexibility across the C-Band.

Also the tunable laser must have better or equivalent spectral performance characteristics, compared to a fixed device, depending on bit rate and reach. The required characteristics include linewidth, wavelength stability, and power variation. iolon's external cavity laser has distinct advantages in spectral fidelity compared to other tunable laser technologies. For instance, where iolon's device has less than 0.25 dB end of life power variation between channels or across the range of channels, other technologies can exhibit a 2 dB or greater variation.

Another important factor is separation between channels. While 100 GHz filters can be interleaved to build a 50 GHz system, the laser still has to be specified at 50 GHz. Several equipment vendors now want to go down to 25 GHz spacing, mainly in order to squeeze more colours in the traditional erbium window.

According to iolon, it is still cost effective to put the wavelengths closer together in the simplest region of amplification. But to get down to that kind of spacing, very good wavelength stability is needed, and iolon is down to less than 20 picometres. This level of wavelength stability is a challenge for other fixed wavelength and broadly tunable laser technologies.

Spectral linewidth is also a tremendously important parameter. For ultra-long-haul, narrow linewidth is needed to minimize signal degradation. Narrow linewidth is difficult to achieve with both traditional fixed DFBs and other tuneable technologies, whereas the external cavity laser is well suited for producing narrow linewidth signals.

MEMS
In order to meet these tight performance characteristics, iolon set about applying its expertise in MEMS and micro-optics to the well-understood Littman-Metcalf external cavity configuration. In iolon's device, light from a FP laser chip is diffracted off of a grating and retroflected back into the laser cavity by a MEMS-controlled mirror. Rotating the mirror selects the desired wavelength.

Test equipment companies, like Agilent for example, have used the Littman-Metcalf configuration in test lasers for some time, exploiting the high power and spectral fidelity of this configuration to achieve test and measurement quality devices. However these lasers have not previously met the size, reliability, and cost requirements of network applications. iolon figured out how to miniaturize it by de-coupling the ECL components like the actuator, mirror, grating, lenses, and gain medium, and implementing miniaturization using MEMS and microptics. Servo control loops provide precise control of the mirror to enable broad tunability, accuracy, and stability over the wide operating temperature range.

The real bottleneck in the network today is in optical ADMs. According to iolon "OADMs have seen only limited deployment because each one has to be custom-configured, so that the wavelength to be added or dropped would be pre-defined." If the technology employed was all-tunable, however, then a low-cost optical ADM could be tuned to the required colour.

Customers who employ fixed wavelength CW devices have the additional cost of putting control circuitry on their cards. However, much of that control circuitry is included in iolon's tunable laser, and with iolon's device, customers will get an attenuation function built-in as well as the wavelength locker. This simplifies the customer's design, saves circuit board space and reduces cost.

iolon has been shipping sample units since late 2000, and predicts that towards the end of the year, system trials will begin with some customers, typically requiring 50 to 100 pieces each. Production shipments are expected to begin at the end of 2001 and into 2002. Others will evaluate the technology for a bit longer before going into trials.

Today's customers are deploying iolon's tunable lasers to replace current fixed wavelength designs and basing new system design around the laser sources being tunable - a key factor being whether the customer has an incumbent network management system. Once the device is installed, the next version of software can be designed to take advantage of tunability.

Carriers perceive much benefit in being able to provision services based on wavelength, an option that they are unable to achieve today based on fixed wavelength devices. Those equipment vendors developing new kit based at the outset on tunable devices will be able to strongly differentiate themselves at the service provisioning architecture level.

At the moment, customers are willing to pay a substantial premium for a tunable device over a fixed wavelength device. However, by the end of 2003, the tunable laser will likely approach price parity with fixed wavelength devices, perhaps with a 10-15% premium.

However, looking to the future, it is apparent that tunable is coming, and more aggressive companies will make the switch to tunable immediately in order to leverage the competitive advantages and flexibility that tunable offers. Furthermore, iolon's product will entice many customer to make the switch to tunable sooner because the optical performance and high output power will reduce costs in other equipment, plus add a whole new level of network capability.

From a strategic point of view, companies are often defined as much by what they succeed in doing well as by what they originally set out to be, and if iolon achieves the technical and commercial targets at which it is aiming in the laser market, that may well define its primary activities for the next two to three years.

iolon's entry into the tunable laser market appears to be well timed. Whilst the market is in the early stages of commercialisation, with quite a few suppliers, it is at that interesting point (which does not always occur with markets that take off much faster) where volumes being shipped are not large and no supplier can be said to be in anywhere near a dominant position. On the other hand, there is enough activity from first generation suppliers (as iolon points out in the case of Altitun) for customers to have a clearer idea about their priorities for the product. This gives second-generation suppliers something clearer to aim at and with decent volumes in the offing.

This latter point may be particularly important for iolon since the background of the company's management is at least partly in the field of volume production for high-tech systems. A core strategic assumption for the company would appear to be the belief that, provided it can get the product definition right, it can beat the socks off competitors in production, giving it the chance for commercial domination of any sector which it enters successfully from a technical standpoint. This may be the key to iolon's assumption that there will be close to price-equivalence between tunable lasers and fixed lasers by 2003. The assumption may be less a fact about the marketplace than a reflection of an instinctive sense by iolon that provided it can drive the cost of tunable lasers close to such parity, then it probably has a good chance of being able to penetrate the general laser market as well as its tunable superniche.

The risks of such an approach are well known. iolon might set out strategically to bring its price down this predefined slope (or possibly even get driven down it by customers already aware of its strategy and thus able to persistently bully it into price drops), then find either that it was not able to build capacity as fast as it had built demand, or alternatively, that despite a competitive market price and plenty of capacity, for various other reasons it was unable to penetrate the marketplace as deeply as it had hoped.