 |
 |
11 May 2007
Cost is no longer the prohibitive factor holding back the adoption of deformable mirrors. Jacqueline Hewett speaks to Paul Bierden of pioneering US firm Boston Micromachines Corporation about the new and emerging applications of the technology.
Today’s new breed of deformable mirrors is finding uses ranging from planet detection to corrective eye surgery and scanning optical microscopy. Having almost exclusively gone hand in hand with astronomy for many years, falling costs and off-the-shelf availability are seeing the devices being adopted by a number of applications.
Having realized the commercial potential of deformable mirrors in the late 1990s, it’s a day that US company Boston Micromachines Corporation (BMC) has been waiting for. “Our products are made on a wafer and batch-fabricated so that costs can be driven down quickly as volumes increase,” Paul Bierden, BMC’s president and CEO, told OLE. “The technology has been proven now. It’s just a case of showing that it is commercially viable.”
What is a deformable mirror?
A deformable mirror can be thought of as a rubber mirror whose surface is made up of an array of tiny elements. Stretching, pulling and tilting each of these elements shapes light in a way that corrects for blurry images. Blurring is caused as the light passes through the medium between the source and the imaging system. For example, in astronomical applications, deformable mirrors correct for aberrations introduced as light passes through the atmosphere.
The key factors that govern the performance are the number of elements (known as actuators), the speed at which the actuators move and the stroke (the maximum actuator displacement). “We’re pushing the state of the art in MEMS manufacturing,” said Bierden. “The fields that we are supplying all want something a bit different. Through different designs or novel manufacturing, we are trying to meet those demands.”
Today, the company has a range of products starting with the Mini-DM that features an array of 32 actuators, a 1.5 or 2 mm aperture and a stroke ranging between 1.5 and 3.5 µm. The active mirror area can reach frame rates of up to 1 kHz, making it ideal for proof-of-concept studies.
Top of the range for high spatial resolution wavefront control is the Kilo-DM boasting 1024 actuators, a 10.5 mm aperture and 1.5 µm of stroke. Bierden says that the device suits fields such as astronomy and was originally developed for a DARPA coherent laser communication project.
Emerging applications
Keen to push the technology further and to work with customers on specific requirements, BMC is currently developing a 4000 actuator device (a 64 x 64 array) that will be the largest-ever MEMS deformable mirror.
“The device will be deployed on the Gemini Planet Imager – a ground-based telescope trying to find planets that are not in our solar system,” explained Bierden. “The reason they need 4000 actuators is to perform high-contrast imaging. The more actuators, the more complex a shape you can have on the mirror’s surface. In this application, you need to have precise control over the wavefront to generate a high-contrast image.” To put this into perspective, Bierden equates the challenge to viewing a speck of dust next to a light bulb across a room. “You need contrast ratios of 1,000,000:1 or even 10,000,000:1 just to pick up the small amount of light that is coming from the planet – or the speck of dust,” he said. “This is a high-spec mirror and will have to run at 5000 frames per second.”
While deformable mirrors continue to make in-roads into astronomical applications, one commercial, high-volume market that BMC is trying to gain traction in is refractive LASIK eye surgery. The idea is to use a deformable mirror to get subjective feedback from the patient. In other words, the patient customizes the new shape of their eye before surgery begins.
Current systems use wavefront sensors to measure aberrations and generate the best perceived new shape for the eye before surgery begins. However, when you get new spectacles, the optician tries various lenses and gets the patient’s feedback to find the best combination.
“Using a mirror like ours, you can measure the wavefront, make an adjustment and ask the patient if their vision is better, or the same, before surgery commences,” explained Bierden. “Higher-order aberrations are a real challenge. To include these higher orders, you need something that has 100 or more actuators. This is a huge market and we hope that our mirrors will be adopted into commercial systems.”
One commercial system that has already deployed BMC’s mirrors is a scanning optical microscope (SOM) from Thorlabs. “This SOM uses a deformable mirror in place of a complex optical set-up to compensate for off-axis aberrations,” said Bierden. “There is traditionally a trade off between wide field of view and resolution, but Thorlabs quotes a resolution of 1 µm over a 40 mm area. Our mirror reduces the cost of the overall system.”
Another application that could soon be reaping the rewards of using BMC’s technology is two-photon microscopy. According to Bierden, Jerome Mertz from Boston University’s Department of Biomedical Engineering has recently shown that deformable mirrors can improve image contrast and resolution.
“Mertz took a normal image with his two-photon system that had one of our mirrors in the optical path,” said Bierden. “He then put a pseudo-random shape on the mirror’s surface to disrupt the focus of his imaging system and took another picture. The key point is that the fluorescence signal coming from just above the plane of interest is the same for both images and can be removed by subtraction. This approach could work for any sub-surface imaging system.”
Because cost is no longer a limiting factor, the technology is rapidly becoming a viable alternative. And as researchers begin to see what can be achieved, a boom in the deformable mirror and adaptive optics market could be just around the corner.
• This article originally appeared in the April 2007 issue of Optics & Laser Europe magazine.
 |  |  |  |  |  |  |
| © 2024 SPIE Europe |
|
