Technology

Nematic liquid crystal devices & controllers

Nematic liquid crystal (LC) electro-optical transmissive devices are characterized by low power consumption, high light throughput at 450 - 1800 nm, large clear aperture, small weight and size, and insensitivity to vibrations. They allow for full-frame image acquisition or manipulation with no moving parts.

The traditional shortcomings of LC devices are their relatively slow response time (tens of milliseconds), temperature sensitivity, and limited field of view. The time response can be improved by parallel alignment of the LC layers (pi-cell), heating the cell, transient nematic effect, and voltage biasing. More sophisticated techniques to improve the time response involve the application of transverse voltage and dual-frequency LC materials. The problem of temperature sensitivity can be solved by an appropriate housing and temperature control servo system. The field of view can be greatly improved by stacking properly oriented cells or the use of pi-cells.

Our liquid-crystal controllers are capable of driving continuously-variable LC devices in either AC (carrier + envelope) or DC (carrier-suppressed, envelope only) modes. AC programs (typically with a 2 kHz carrier) are used in research and development applications that require holding voltages on LC cells for an indefinite period of time. DC programs are often used with shutters or polarization switches in stereoscopic projection displays; for this application, the DC programs are synchronized to the frame rate.

Controllers need to maintain overall DC balance to prevent slow ion migration to the electrodes, which can damage the LC device over time. For AC modes, our controllers maintain DC balance by carrier-cycle accurate programs; for DC programs, our controllers invert the envelope sign after every program cycle to maintain DC balance.

To change the electric field interior to the cell, the charge on the transparent, conductive indium tin oxide (ITO) layers must be redistributed. For large-area, highly capacitive cells, this charge redistribution can require a large (~amps) inrush current that must be supplied by the LC controller. Our proprietary analog output drive stage provides enough inrush current (>3.5 A) to change states rapidly for large-area cells while retaining full short-circuit protection on the LC outputs. Once a short is removed, the output is automatically re-established.

Our µ-micro series controllers use 16-bit & 32-bit flash microcontroller technology to run a proprietary real-time pre-emptive kernel. This permits dynamic LC program editing from a host computer via a full-speed USB 2.0 compliant interface while insuring glitch-free program execution throughout the editing process. Once edited, the LC programs are stored on the controller to permit true embedded operation without a host computer present.

Clients around the globe are utilizing our LC controllers in a variety of impressive custom applications. For example:

• Our controllers run the projector polarization switches that create the 3-D effect in technologically sophisticated amusement park rides in the US and abroad.
• Our µLC400-30's operate left-eye/right-eye projector shutters in world-class virtual reality environments and visualization facilities.
• Our high-voltage LCC1-100 shutters telescope light in a system that tracks potentially hazardous "space junk" in low-earth orbit.
• A customized version of our LCC2-12 operates a LC polarization rotator as part of the optical flight package aboard a high-altitude airborne science aircraft for atmospheric aerosol measurements.

Phase-sensitive detection & precision signal mixing

Demodulator or lock-in amplifier techniques are frequently encountered in situations with inherently low signal-to-noise, but can be applied in any situation with a periodic signal source. What is desired is to extract the small-signal bandwidth (~100 Hz or less) from a mixture of noise or other undesired components. The source must generate a reference or sync signal phase-locked to the excitation; this reference signal is provided to the lock-in, along with the signal to be measured.

There are two complementary ways of looking at the technique. In the frequency domain view, it is the process of mixing the bandwidth of an amplitude-modulated signal down to DC by convolving it with a delta function at the reference frequency. The resulting DC signal bandwidth is then recovered by multiplying it with a rectangle function (= ideal low-pass filter), which removes the undesired band at twice the reference frequency.

In the time domain view, it is the equivalent of multiplying the input (= desired signal at some frequency, plus undesired noise at various other frequencies) with a sinusoidal reference at the same frequency as the signal, and then time-integrating the product. The reference has a fixed, definite phase relationship that is locked to the signal. For a reference at zero degrees (perfectly in phase with the signal), the time-averaged result is a positive DC amplitude directly proportional to the signal amplitude.

For a reference locked 180 degrees out of phase with the signal, the time-averaged result is a negative DC amplitude, again directly proportional to the signal amplitude.

For a reference at 90 degrees with respect to the signal, the time-averaged result is zero. A phase slip is associated with all other frequencies other than the signal frequency. This phase slip means that noise components multiplied with the reference and time-averaged will rapidly flutuate between positive and negative contributions, time-averaging to zero. In this sense, the integral of the signal with the reference picks out just the signal frequency and discriminates against all other frequency components.

For either view, a reference frequency- and phase-locked to the periodic signal must either be externally provided or internally synthesized. With an appropriate reference, demodulators can "lock in" on small signals (~µvolts) in the presence of substantial background, plucking them out of the noise, amplifying them, and accurately measuring their RMS amplitudes.

Application examples

• Detection of weak signals in spectroscopy: lock-in detection can be used in conjunction with existing spectrometry systems to detect weak absorption, fluorescence, or Raman signals in the presence of substantial background noise. Depending on the application, the technique could work with a modulated source, a mechanical chopper, an electro-optic, acousto-optic or photoelastic modulator. For example, a fiber-coupled spark-plug laser sensor can be used in situ with lock-in detection to measure combustion-chamber conditions using wavelength-modulation spectroscopy. This technique requires recovery at both 1F and 2F.
• Measurement of optical activity: circular dichroism (CD) and optical rotation (OR) measurements can benefit from lock-in techniques. CD is the difference in absorbance between right and left circularly polarized light in a sample. Since the magnitude of the CD signal is typically ~10⁴ times smaller than that of the isotropic absorbance, lock-in detection and light modulators are typically employed in CD measurements. In OR measurements, the polarization of linearly polarized light is rotated upon passing through the optically active sample.
• Phase-sensitive fluorescence spectroscopy: this technique is of particular interest to biochemical applications, as it can be used to resolve fluorescence interferences of intrinsic fluorophores such as tyrosine and tryptophan derivatives. The technique requires modulation of a light source at a frequency comparable to the inverse of the fluorescence lifetimes (10-100 ns), which can be done acousto- or electro-optically.
• Feedback and feed-forward controlled instruments: lock-ins can be applied in numerous control applications, including laser power, frequency and polarization stabilization, and optimization techniques. For example, lock-in amplifiers are used in the piezo & tuning-fork probe feedback loop of near-field scanning optical microscopes to control sample-to-probe distances and obtain surface topographic data.

Our µLIA-320 ARM-based lock-in amplifier can be used as a standalone general-purpose instrument. This device delivers a complete, compact, cost-effective USB-enabled lock-in solution suitable for a variety of applications. The µLIA-320 can be economically integrated into OEM applications and higher-level instruments. We also offer custom solutions for situations that require an application-specific lock-in.

Direct digital synthesis for frequency & phase control

Direct digital synthesis (DDS) is a numerically-controlled oscillator technique that can deliver extraordinary levels of frequency & phase control. In the DDS technique, a clock source is used to drive both additions to a digital phase accumulator register and the output latch of a digital-to-analog converter. In state-of-the-art implementations, the clock frequency can be as high as 1 GHz.

At each clock, the contents of a frequency tuning word register are added to the phase accumulator. The bit width of these registers can range from 28 to 48 bits and determines the frequency resolution of the DDS scheme. The phase accumulator is truncated, combined with a phase tuning word register, and the result is used as the angle to look up a digital amplitude value in a cosine (or other periodic function) look-up table. This digital amplitude drives the bit switches of the output DAC. The result is a periodic analog signal whose frequency and phase are under complete digital control. DDS can be used in any situation that requires tuning with sub-Hz settability and repeatability.

Because the output is a sampled signal, the maximum frequency is subject to the Nyquist criteria--it can be no greater than half the clock frequency for a sinusoidal output. Additionally, the sampling process results in an output with a |sinc(f_out / f_clock)| amplitude profile, and with image responses about integer multiples of f_clock. For clean DDS clock generation, these images must be suppressed by an appropriately designed anti-aliasing filter. There are also digital artifacts (spurs) in the filter's passband due to the finite digital-to-analog converter resolution and truncation of the phase accumulator.

DDS is a key enabling technology for our µLIA-320 ARM-based dual-phase lock-in amplifier, which has three channels of common-clocked 32-bit DDS: two channels for the (x, y) signal paths, and one for the reference channel input/output path. The common clocking permits the channels to have an exact integer relationship between their frequencies while remaining perfectly phase-locked. This permits conventional 1F quadrature signal recovery, as well as measurement schemes that require recovery at both the fundamental frequency (1F) and higher harmonic frequencies (e.g. 2F).

Mueller & Stokes polarimetry

In Stokes polarimetry, the state of polarization of light is analyzed. The measured light may be transmitted, reflected, scattered or radiated by the sample. Full state-of-polarization measurements are most useful in situations where the sample exhibits a combination of polarization-related phenomena--for example, linear birefringence, depolarization, and optical rotation. The analyzed light is then expressed in terms of four Stokes parameters. These four parameters form a vector (Stokes vector) that provides a complete description of the state of polarized light, including partial depolarization.

Mueller matrix polarimetry provides a complete description of the polarization properties of an optical sample, expressed in terms of 16 elements of a 4 x 4 real matrix. The Mueller matrix of a sample determines how an input polarization state (input Stokes vector) is transformed by an optical medium into an output polarization state (output Stokes vector). Mueller matrix imaging reveals information that is not available from conventional, non-polarized imaging. For example, many biological tissues and materials possess internal structure that can be revealed by Mueller matrix imaging.

In spectropolarimetry, the state of polarization and/or Mueller matrix are measured at different wavelengths, to provide additional information in the form of a spectral signature of the sample.

Application examples

• Determination of optical properties of biological materials and turbid suspensions through analysis of backscattered light.
• Surface and sub-surface imaging of skin abnormalities and carcinomas, including early detection of melanoma.
• Glucose sensing in presence of retardation and scattering in blood and aqueous humors.
• Detection, diagnostics and monitoring of structural changes in biological tissues.
• Characterization and testing of optical components, materials and liquid crystal displays.
• Object detection and contrast enhancement for feature mapping (e.g., landmine detection).

Our MSP-meter can measure either the full Mueller matrix of optical samples, or the Stokes vector of the incoming light. Polar decomposition of the measured Mueller matrix allows extraction of the sample's optical properties, such as diattenuation, linear retardance, optical rotation and depolarization. The key optical components of the MSP-meter are two integrated modules: the polarization generator and the polarization analyzer. In its basic configuration, each module contains two liquid crystal elements, and can be used as either a polarization analyzer (for a Stokes polarimeter), or as a polarization generator, in conjunction with an analyzer (for a Mueller polarimeter). The other components of the MSP-meter are the LC controller, and proprietary algorithms incorporated into the application software. Owing to its modular design, the optimal combination of the light source, polarizers and detector can be selected to match the user's application. Since the modules can be calibrated for any wavelength in the 450 nm-1800 nm spectral region, the polarimeter can be easily converted into a spectropolarimeter.

The modules consist of two optically-contacted nematic liquid crystal variable retarders enclosed in a temperature-controlled aluminum housing. The external optical surfaces are broadband AR coated. The 1-inch clear aperture size makes the modules suitable for imaging Mueller matrix polarimetry applications. The housings are equipped with a resistive heater element and a temperature sensor for closed-loop control. A unique feature of the housing design enables the entire optic assembly to be rotated +/- 5 degrees about the transmission axis, for optimal system alignment. The modules can be used to generate or to analyze the polarization state of light at any wavelength band within its range.

The host application incorporate proprietary algorithms that allow for fast calibration (~30 s) and operation (~5 s per Mueller matrix measurement). A learning algorithm makes it unnecessary to calibrate the individual LC elements; one can also calibrate out the background polarization signature of the optics and sample holders. A front-panel GUI seamlessly ties together calibration, data acquisition, & data processing.

More details on the PGA-based MSP-meter are available in this SPIE Newsroom release.