Shedding New Light on Materials Analysis: Tunable Mid-IR Laser Spectrometry

August 1, 2010

Special Issues

Volume 0, Issue 0

Infrared (IR) spectroscopy is one of the most versatile and powerful analytical tools that we have today for the characterization and identification of materials. Its strength lies in its ability to handle a broad range of material types, in any physical state, at a wide range of concentrations, and on many occasions, with direct methods of measurement. These strengths are about to be enhanced by the use of instrumentation that utilizes a choice of broadly tunable laser devices, covering the sweet spot of the mid-IR spectrum, the "fingerprint region." These systems currently cover the spectral range of 6–12 ?m (1665–830 cm-1), which provides spectroscopic access to almost all classes of chemical compounds. This article reviews the benefits offered by such a laser system for a wide range of new and challenging applications.

Infrared (IR) spectroscopy as an analytical tool has been around since the mid-1950s. In the early days, it was perceived as an important tool for material identification, but it also was perceived to be weak on performance, being described in many circles as an energy-limited technique. In fact, in 1970, a well-publicized editorial suggested that the "death knell" should be sounded for the technique. This was around the time that the new Fourier-transform IR (FT-IR)–based technology was emerging. A key, but not widely known company, Block Engineering, was at the forefront of the FT-IR technology at that time, and produced a range of commercial products that were later marketed by another vendor. In the ensuing years, FT-IR gradually took over the IR spectroscopy market, and by the late 1980s and early 1990s, it was being heralded as a critical technology for the analyst's tool box. This became IR spectroscopy, although for many years, and even today, some still call the spectroscopic technique FT-IR. What has been important about FT-IR is that in combination with a broad range of sample-handling techniques, virtually any type of sample can be addressed. This does not mean that there are not areas of difficulty. There are limits, and these often are related to the photon budget provided by a conventional thermal IR source.

It has long been recognized that the ideal light source for IR spectroscopy would be a tunable IR laser; such a device has been considered to be a panacea by the spectroscopist. Laser-based spectrometers based upon what can be defined loosely as tunable lasers have been around for at least two decades. The problem has been in the interpretation of the term tunable. Early lasers only provided a tuning range of fractions on a micrometer. In IR spectroscopy, such a tuning range can be considered to be extremely limited, if not unusable, for most practical IR spectroscopy. Yes, if one is looking at certain gas rotational lines of specific gases, the limited tunability range can be used beneficially for the monitoring of these gases. However, typically, multiple lasers are required for the measurement of different gases, and seldom does a single laser handle more than one gas or vapor. Such systems have been used in applications such as fence-line monitoring of gases and some vapors for some time. However, they have not been available as a general purpose analytical tool. The IR spectroscopist needs to have a usable range covering many micrometers (hundreds of wavenumbers [cm-1]) if a tunable laser is to become a practical tool. While it is early days, one is tempted to draw parallels back to the introduction of FT-IR.

The role of lasers in spectroscopy has already become established by techniques such as Raman spectroscopy. There are many parallels between Raman and IR spectroscopy, with the most important being the complementary nature of the information content of the two forms of spectra. In recent years, Raman has gained in popularity both in traditional, laboratory-based applications and for portable handheld measurements. There have been significant advances in the performance of Raman systems, in part based upon new laser technologies, including the use of both ultraviolet (UV) and near-IR (NIR) lasers. Issues still exist relative to eye safety from scattered radiation and sample fluorescence, which still can be a major spectral interference. Also, Raman scattering is weak phenomenon and is not readily applicable to trace-level measurements, such as for low concentrations of deposited materials on surfaces, low concentrations of solutes in solutions (typically requires surface-enhanced Raman spectroscopy [SERS]), or for most forms of gas measurements (normal and trace levels). IR spectroscopy is strong in all of these areas, and is applied widely to measurements of most materials in both bulk concentration levels and at trace-level measurements, and is particularly well applied to gases and vapors, even down to parts-per-billion levels, with the right form of sample handling. In the case of the use of mid-IR lasers, such as the broadly tunable quantum-cascade laser (QCL) devices, eye safety and sample fluorescence are not issues.

Tunable Laser Spectrometers

It does not take much imagination to define some critical application areas that can benefit from the increased photon power of a laser-based system. Applications in which IR spectroscopy can "run out of steam" are in measurements on highly absorbing materials, in microsampling, especially down at the diffraction limits, and in remote sample handling with either optical fibers or with a standoff configuration. The QCL systems provide the performance in the mid-IR region to over come these shortcomings. One such system (LaserScan, Block Engineering, Marlborough, Massachusetts) provides a nominal measurement range of 6–12 μm, with current versions offering 6–10 μm and 7–12 μm. The current specifications of the instrument are summarized in Table I.

Table I: Tunable QCL spectrometer specifications

One configuration for the instrument is for standoff measurements, where the analysis of materials on surfaces can be made from several feet away from the sample (typically 3–5 ft). A red (visible) sighting laser is included to help the user locate the correct sampling point, remembering that IR laser radiation is not visible to the human eye. Upon projection, the IR sample beam is currently set to 5 mm in diameter. An important point here is that the user does not have to worry about scattered radiation from the laser because the beam is eye-safe. The large IR collection optic provides efficient light collection from a remote sampling point. Note that the system is also offered with a laboratory-oriented configuration that enables the use of standard FT-IR accessories. In this mode of operation, standard IR cells, reflectance accessories, and even microscope accessories can be accommodated. Gas, vapor, and liquid analysis is accommodated by the use of appropriate accessories and attachments.

Figure 1: Tunable QCL principle of operation.

The principle of operation of this broadly tunable QCL is outlined in Figures 1–3. The laser features external cavity tuning by means of an optically matched grating; optimized for resolution and dispersion in the lasing region of the QC chip. The laser radiation is extracted from the laser assembly, as illustrated in Figure 1. A proprietary rapid scanning mechanism enables the wavelength to be tuned at high speed providing a full scan potentially in milliseconds. A schematic overview of the total system is provided in Figure 2, and the device represented in Figure 1 is designated as the compact tuner module. This is controlled by a dedicated high-speed processor, which synchronously acquires the laser pulse signals from the "sample detector." The laser pulse repetition rate is 200 kHz. The tuning and acquisition sequence is summarized in Figure 3, which illustrates the synchronization between the grating rotation and the QCL generated laser pulses, to generate the analytical spectrum as represented at the bottom of Figure 3. The angled flat reflector in the upper right hand side represents the reflective sample surface, where the system is operated in a stand-off configuration. For gas and vapor sensing, the beam passes through free space where it interacts with the "ambient air" and returns to the instrument via a remote mirror target.

Figure 2: Tunable QCL spectrometer system architecture.

In traditional terms, the QCL spectrometer operates very much in the same manner as a traditional spectrometer, such as an FT-IR. The main difference in operation is in the ability to acquire spectral data very rapidly, and under difficult or unusual sample handling conditions. An example of a nonstandard operation is the remote standoff (a completely noncontact noninvasive sampling procedure) mode of measurement. The system is designed for benchtop or tripod-mounted operation where the remote measurement is made from a fixed target. In this arrangement, the visible sighting laser is used to position the sampling point in the correct location relative to the spectrometer.

Figure 3: Laser tuning and predispersive spectroscopy.

The instrument is also available in a portable format, which allows handheld remote inspections of surfaces at varying distances. To enable this mode of operation a special colinear optical arrangement is used where the laser illumination path follows directionally the same effective path as the returning reflected beam. This allows the instrument to be used from distances of a few inches to several feet from the sample surface, making it ideal as an inspection tool or as a stealth or noncontact tool for the detection of materials on surfaces for law enforcement, military, or security applications. This mode of operation will be covered later with practical examples of remote, stand-off measurements.

Applications of the Tunable QCL Spectrometer

Fundamentally, a broadly tunable QCL spectrometer functions as any other IR spectrometer. It can be used in a laboratory environment for standard transmission measurements on solids, liquids, and gases. A spectrum of a standard, free-standing polystyrene film is shown in Figure 4, which illustrates the standard transmission mode of operation. The beam size can be as low as 5 mm in diameter at the sample, which corresponds to the average beam size at the sample focus of standard laboratory (FT-IR) instruments. Consequently, all standard methods of sample handling can be applied, including compressed halide disks and mulls, reflection measurements, including attenuated total reflectance (ATR), and standard methods of gas sampling with either short or long-pass cells. One of the benefits of the use of a laser is the inherent beam collimation, which usually simplifies the imaging requirements of the sample handling optics. The laser also inherently provides a high optical throughput, for trace monitoring applications, as well as for low-level transmission or low reflectivity measurements, as experienced with highly absorbing sample substrates, or with microsampling methods.

Figure 4: Polystyrene spectrum recorded in standard laboratory measurement mode.

The higher throughput is the consequence of a higher IR power density at the sample; estimated to be at least 400× per wavelength when compared to a traditional IR source as used in an FT-IR. As a result, it is possible to carry out measurements normally considered to be too difficult by traditional IR methods. This increased performance can be realized in terms of higher signal-to-noise ratios (S/N) for highly absorbing samples or as faster rates of spectral acquisition (hundreds of milliseconds to seconds). An added benefit is the directional characteristics of the IR beam that results from attributes of the laser source (beam coherence with a minimally divergent beam), and this makes it possible to study samples remotely, at a significant standoff distance. This includes the ability to perform ambient air open path gas or vapor monitoring applications without the need for the extra large source assemblies or for the telescope collection optics normally required by traditional remote monitoring IR instrumentation. This makes measurement scenarios such as cross-room work-place monitoring or cross duct emissions monitoring very practical and easy to implement.

Examples of applications that need higher throughput or improved (S/N) and increased data collection rates include IR microscopy and remote sample handling via fiber-optic cables. There are two practical issues for consideration here with regard to IR microscopy. One is the ability to pump more photons through the sample when imaging down to a small spot (aperture size) with the resultant improvement is spectrum quality. The other equates to both higher speed acquisition of high-quality spectra, and the ability to record spectra with higher-than-normal spectral resolution. It is common to record spectra at 8 cm-1 resolution (or lower) to help expedite the measurement with good S/N values within practical measurement times. With a laser-based system, spectral resolutions as high as 0.5 cm-1 can be considered. This is particularly important for situations in which fine spectral detail is required, such as in the analysis of crystalline modifications, as in the case of pharmaceutical polymorphs. The other area of improvement for microscopy that can be realized with the aid of the laser is for spectral (hyperspectral) imaging. In this case, the increased photon flux can enable faster spectral acquisition, with the possibility to recording data with high spatial resolution (using a smaller aperture) and higher spectral resolution as addressed previously.

The other important area of application for the laser-based system is for process or chemical reaction monitoring where there is a need to record spectral data from the sample in a remote location. This is a critically important requirement for IR spectroscopy, in which there is a need to enhance its process-monitoring capabilities, as compared to Raman or UV-visible-based methods. These latter methods have the benefit that they can utilize silica-based fibers that inherently provide a high optical throughput. Fibers that transmit in the mid-IR exhibit a high degree of optical attenuation, which leads to very large light loses as a function of fiber length. Typical fibers available for the mid-IR include chalcogenide fibers, or more recently, hollow-core fibers. Both types can be throughput-constrained in the mid-IR region. Most users need the ability to transmit the spectral information over long distances. The higher IR flux provided by the laser system will enable longer lengths of these fibers to be used, therefore enabling process–reaction monitoring measurements not normally accessible to FT-IR process analyzer users.

The main focus for applications of the QCL spectrometer in this article is in the area of remote or standoff measurements. This mode of measurement is unique for routine sampling and is usually impractical with a conventional FT-IR-based system. Commercial instruments for handheld measurements (FT-IR and Raman) are limited to close-up measurements, in which the spectrometer is used a few millimeters from the surface or for ATR-based measurements, in which actual surface contact is required. While both of these approaches work, they have limitations, both in terms of basic performance, and in the case of ATR, in which there is a high risk of contamination of the sampling head and the sample surface. A standoff measurement eliminates these problems and risks.

One example of the use of the laser described here for remote measurements is in the measurement of surface coatings, as illustrated in Figure 5, where the spectra of two different thin-film coatings are presented: one from an epoxy resin and the other from a clear spray lacquer, both deposited on a painted metal surface. These spectra exhibit unique anomalous dispersion effects (in this case Restrahlen bands) as typically observed in thin-film reflectance spectra from highly absorbing substrate surfaces. From the point of view of characterization, these materials can be differentiated, even though they can be convolved with some of the spectral characteristics of the background or substrate. From the point of view of characterization, this format of spectrum can be used directly with a library of similarly recorded data for known surface materials. If there is a requirement to compare the recorded spectrum with a standard library of spectra in transmittance or absorbance format, then the spectrum can be qualitatively compared to the library after a band inversion (for the transmittance format).

Figure 5: Stand-off for remote measurements of coatings on painted metal surfaces.

Materials deposited as thick films, designated "infinitely thick" relative to the IR absorption, will yield characteristically distorted mixed-mode spectra. Conversions such as the KK-transform (Kramers–Kroenig Transform) can be used to reformat the data into an absorption-style of format, by separating the refractive index components from the spectra. An example of such a measurement is the standoff recording of the spectrum from a pool of liquid, as shown in Figure 6. The spectrum can be characterized as that of a common phosphate ester (tricresyl phosphate). The generated absorbance form of the spectrum is suitable for direct comparison to a standard spectral library.

Figure 6: Stand-off for remote measurements of liquid films.

Standoff measurements are important for a wide range of practical applications, for example, surface quality or contamination measurements for engineering and production applications, security applications, in which dangerous materials such as chemical agents and explosives might be involved, and for law enforcement for illicit drug detection (both production and usage). Unlike other methods, this can be performed at a reasonable distance, and without contamination of either the surface or the instrument. As an example, Figure 7 illustrates the stand-off measurement of explosive residues on a painted surface at approximately 3 ft separation between the target–sample and the instrument. For comparison purposes, the FT-IR reference spectra, recorded with a standard 30° specular reflectance accessory, also provided. In spite of the large separation of the standoff measurement, the spectra of the explosive materials compare well to the reference spectra.

Figure 7: Stand-off measurements of explosives on painted surfaces: (a) QCL stand-off spectra and (b) FT-IR reference spectra.

Conclusion

In the future, it is expected that the overall spectral range covered will increase to around 5 μm (2000 cm-1 ), and the spectral measurement times will be shortened to milliseconds. The latter is important for accurate handheld measurements, in which shorter observation times are necessary.

John Coates is with Coates Consulting, Newton, Connecticut.