Raman spectroscopy and Monocrom products

Raman spectroscopy and Monocrom products

ABSTRACT

This application note gives a brief introduction into Raman spectroscopy and related applications. It gives an overview over the physics, the requirements on the laser source and the detector properties necessary to detect a proper spectrum. The most critical Figures Of Merit will be discussed and last but not least it follows a presentation of Monocrom products that can be used to serve Raman spectroscopy applications.

1. RAMAN SPECTROSCOPY – AN OVERVIEW

Raman spectroscopy is a characterization technique concerning the irradiation of a sample with a light source and the analysis of its scattered light, in particular those photons that are scattered in-elastically (elastic scattering is known as Rayleigh scattering). The photons that suffer inelastic scattering interact with the matter either by gaining or loosing energy in the form of phonons. The portion of photons with a resulting energy lower than the incident beam constitute the Stokes radiation, while those photons that gain energy are part of the Anti-Stokes radiation. The possible elastic and inelastic scattering transitions are illustrated in the upper part of Fig. 1.1. Both, Stokes and Anti-Stokes, transitions have different probabilities. The latter is quite raw, but the signal can be enhanced via coherent excitation.

In fact, it is the frequency shift (expressed in wavenumbers, generally in units of cm−1) what is measured in Raman spectroscopy, and its value is independent from the excitation wavelength λin. However, in practice there are several factors that make one or another excitation wavelength more or less appropriate, including the sample itself.

In principle, Stokes and Anti-Stokes radiation can be analyzed in order to measure the Raman shift. However, Stokes shows inherently a higher intensity since the relative thermal population of energy levels are defined by the Boltzmann factor:

Figure 1.1: Illustration of the energy shift suffered by photons inelastically scattered when the incident photon excites the material (a phonon is absorbed) or de-excites it (a phonon is released). [1]

In Eq. 1.1 the ni represents the population of a specific state i, kB is the Boltzmann constant, h is the Plank constant and together with the photon frequency Vi = cvacin it gives the energy of that state. The last variable missing is T, the temperature in Kelvin. Due to the intrinsically low quantum efficiency of this interaction(1 in 106 − 108 events), a high brightness light source is mandatory. Brightness Br can be expressed via:

Here Plas denotes the laser power and Mo2 describe the beam quality in both axes x and y, respectively. Cross sections for Raman scattering reach from 10−31cm−2 −10−25cm−2, which is still five orders of magnitude lower than Rayleigh scattering [3]. It is easy then to understand why although the Raman Effect was observed for the first time in 1928 [4, 5], the development of Raman spectroscopy was not possible in a reliable way until the 60’s, when the first lasers appeared. Additionally, the intensity of the Raman scattered light Iscat is proportional to the incident beam intensity I0 and to the fourth power of its wavelength λin [6]:

Figure 1.2: Relationship between the three main elements involved in Raman Spectroscopy. The analyte and the laser source are closely interrelated. Besides, the detector is mainly a consequence of the wavelength, although the range of the detector plays an important role on the wavelength choice. Detector and analyte do not condition to each other directly.

So, in principle lower wavelengths should be the best option. However, things are never that easy, so let’s consider this idea further in detail in the following paragraphs. It is important to have in mind the three main elements involved in Raman Spectroscopy:

• The analyte

• The laser source

• The detector

The relationship and mutual influence of these three elements are illustrated in Fig. 1.2.1.1 The analyte Raman Spectroscopy is a material characterization technique that uses the interaction between laser light and the rotational-vibrational energy-level structure of molecular compounds, as a way to provide fingerprints of it. Therefore, Raman spectroscopy essentially represents a tool to identify certain materials and some relevant aspects of its molecular and lattice structure, although it can be also used for quantification. A good example of the application of Raman Spectroscopy is the identification of solid carbon materials, which show distinctive spectroscopic features in Raman scattering depending on whether the atoms are arranged to form diamond, graphite, amorphous carbon or even fullerenes, carbon nanotubes or graphene. The material under characterization can be either in solid, liquid or gaseous state, although for obvious reasons, the gaseous samples are not easily detected with conventional techniques. However, there are several Raman-based techniques that are able to obtain an enhanced scattering response when applying complementary strategies. This is the case of Surface Enhanced Raman Spectroscopy (SERS), where the sample is placed over a novel metal–coated substrate (like silver or gold). This way, the signal detected can be amplified by several orders of magnitude, allowing even single-molecule detection if the metal layer is nanostructured [3, 6, 8]. Another well established technique consists of irradiating the sample at a wavelength close to one of its electronic transitions, leading to a truly photon absorption (in Raman scattering the incident photon is not really absorbed by the sample) and a subsequent highly enhanced Raman response. This is called Resonance Raman (RR) spectroscopy [6, 7]. Also worth to mentioned is Coherent Anti-Stokes Raman Spectroscopy (CARS), which is based on third-order susceptibilities. In order to get a significant signal, high laser intensities are required to ensure a two-photon absorption. In the literature can be found some more signal enhancement techniques like—among others—Coherent Stokes Raman Spectroscopy (CSRS), Photo-Acoustic Raman Spectroscopy (PARS) or Stimulated Raman Gain Spectroscopy (SRGS) [9].

1.2 The laser source

As explained above, due to the low quantum efficiency of Raman scattering, there are two conditions for a proper light source to promote observable Raman scattered photons. One is high brightness Br (see Eq.1.2)—so laser is obviously the best choice—, and the other is an as low as possible wavelength (see Eq. 1.3). UV and visible lasers in the NUV-blue-green region are a good option for inorganic compounds [7]. But when it comes to organic matter and biological samples, many compounds show a strong fluorescence in the visible region when irradiated with these wavelengths. Therefore, the Raman signal can be easily “buried” under an intense, broad fluorescent emission background (and noise). A quick path to avoid this is shifting the incident wavelength of choice out of the visible and NUV region towards the MUV or the NIR, but then we must consider other aspects like the influence of the detector. Usually, the range of Raman shift that goes from 100 cm−1 up to 4000 cm−1 covers almost the entire set of Raman-active species [7, 9]. However, when we translate this range into nanometers, the observable range turns to be only a 26nm wide spectral window if λin = 248 nm, which implies a big challenge from the grating resolution and detector perspectives (UV-enhanced silicon detectors are a solution, but sensitivity is still low). At the other extreme, if λin = 1064 nm, the window of observation is about 700 nm, which means that germanium or InGaAs detectors are necessary. The relation between the excitation wavelength and the width of the Raman spectrum is

Figure 1.3: Popular Raman excitation laser lines and their corresponding spectral observation window for the Raman scattering (Stokes radiation). The horizontal length of the side rectangles represent the spectral range in nanometers corresponding to a Raman shift going from 100 cm−1 to 4000 cm−1, while the height illustrates the relative scattering intensity according to the excitation wavelength (notice that the vertical left axis is in logarithmic scale). Additionally, relative spectral response of traditional detector technologies are superimposed to illustrate the link between laser source and detector.

shown in Fig. 1.3. Moreover, MUV wavelengths can induce unwanted changes in many samples (ionization, polymerization or bond-breaking transitions) [6].

1.3 The detector

CCD cameras with Si-based detectors shows an excellent sensitivity specially in the red-NIR region, so the optimum combination seems to be a laser source in the range 400 − 550nm with a regular room temperature CCD camera. But having fluorescence in mind, the use of MUV lasers presents to main drawbacks: there are not yet cheap and widely available options in the laser industry below 300 nm, and the sensitivity of CCD cameras in the NUV-violet is not that good. If we decide to move towards the NIR lasers, Raman scattered radiation will show lower and lower intensity as the irradiation wavelength increases. In addition, Si-based detectors are no longer a good option when the Raman scattered light is over 1 micron, so other type of detectors (like those based in InGaAs) must be used instead. So all in all, it is a question of identifying the best trade-off. In this sense, the progressive adoption of Raman spectroscopy by the industry has converged towards the “happy” combination of 785nm plus CCD cameras, being nowadays the gold standard. This has enabled, for example, the creation of affordable portable or even handheld Raman spectrometers, which contrast with the old, bulky and expensive laboratory equipment from the 80’s. Lasers at 785nm can be diode-based (so very efficient, compact, excellent emission characteristics, cheap and widely available), while silicon photo-sensitivity stays within acceptable levels (Raman shifts over 3000 cm−1 are still detectable under these conditions) and fluorescence is overcome in many cases. Nonetheless, many cases still require different wavelengths and different detectors.

CCD cameras with Si-based detectors shows an excellent sensitivity specially in the red-NIR region, so the optimum combination seems to be a laser source in the range 400 − 550nm with a regular room temperature CCD camera. But having fluorescence in mind, the use of MUV lasers presents to main drawbacks: there are not yet cheap and widely available options in the laser industry below 300 nm, and the sensitivity of CCD cameras in the NUV-violet is not that good. If we decide to move towards the NIR lasers, Raman scattered radiation will show lower and lower intensity as the irradiation wavelength increases. In addition, Si-based detectors are no longer a good option when the Raman scattered light is over 1 micron, so other type of detectors (like those based in InGaAs) must be used instead. So all in all, it is a question of identifying the best trade-off. In this sense, the progressive adoption of Raman spectroscopy by the industry has converged towards the “happy” combination of 785nm plus CCD cameras, being nowadays the gold standard. This has enabled, for example, the creation of affordable portable or even handheld Raman spectrometers, which contrast with the old, bulky and expensive laboratory equipment from the 80’s. Lasers at 785nm can be diode-based (so very efficient, compact, excellent emission characteristics, cheap and widely available), while silicon photo-sensitivity stays within acceptable levels (Raman shifts over 3000 cm−1 are still detectable under these conditions) and fluorescence is overcome in many cases. Nonetheless, many cases still require different wavelengths and different detectors.

1.4 Figures of merit of lasers concerning Raman spectroscopy

Apart from the wavelength of the laser source, there are other important figures to keep in mind from the joint perspective of the laser manufacturer and the Raman equipment integrator, which are listed below:

• Beam quality: For samples where composition or structure spatial distribution is analyzed, TEM00 beams maximize spatial resolution.

• Polarization: Laser beam must be linearly polarized in certain branches of Raman Spectroscopy where the polarization degree of the molecules is investigated.

• Spectral linewidth: around 10pm or less is required to guarantee an acceptable resolution of the Raman spectra (the smallest difference in cm−1 between Raman features that can be resolved)

• Spectral purity: Whichever side modes concerning the excitation wavelength need to be suppressed, so the main peak must prevail over them at a level > 60dB. This level of purity is usually acceptable at 1 − 2nm around the main peak, although this distance gets reduced as Raman shift goes into the sub − 100 cm−1 level.

• Frequency stability: Since the acquisition time is usually in the order of seconds or tens of seconds, it is important to keep the excitation wavelength still enough (< 10pm drift over time and operation temperature).

• Output power stability: Typical power range goes from 10 to 1000mW, depending on the analyte and the excitation wavelength as mentioned above. The power stability is important mainly for quantification purposes and it is linked to the integration time necessary to obtain a spectrum.

• Isolation against optical feed-back: In the particular case of confocal microscopy configurations (over-coupled excitation and backscattered beams), even a small portion of light backscattered into the laser source can cause power instabilitiesor even laser degradation. Optical isolators must be used.

1.5 Hot applications of Raman Spectroscopy

Raman spectroscopy is gaining presence in bio-science given its non-invasive nature. Lasers for Raman spectroscopy can range many wavelengths, but usually they are in the UV-VIS-NIR part of the electromagnetic spectrum. Also, remote identification of explosives can be carried out with Q-Switched lasers by using diode pumped solid state lasers 2nd, 3rd or 4th harmonic generation wavelengths. Raman spectroscopy can be found—for instance—in:

– Art & Archaeology

– Bio-science and Medical Diagnosis

– Polymers and Chemical Processes

– Semiconductor & Solar Industry

– Geology and Mineralogy

– Pharmaceutical Industry

– Environmental Science

– Raman Microscopy

– Forensic Analysis

– Gemology Teaching

– Quality Control as well as General Research

1.5.1 Take a closer look on food safety

Based on signal enhancement techniques like SERS it is possible to design on-line monitoring systems for food safety and food quality control. In the past decades, since the food industry is more profit driven and the globalization of the marked is progressing, the consumers are more and more concerned about the food quality in mass production [10]. These concerns are based on real evidences, which proof a link between harmful food and health diseases. Some of them are analyzed and named by [12, 13]. The connection between health and food quality works in both directions, harmful food can cause diseases and healthy food can improve mental and physical body strength [10, 11]. These examples can be taken as motivation to ensure high quality food products. A reliable quality control can—thanks to Raman spectroscopy (especially SERS) as well as other spectroscopy methods—easily be implemented into the production process. An overview concerning food safety and quality control can be found in [14, 16] and [15] describes, for instance, how pesticides on fruit surfaces can be detected.

Figure 2.1: Picture of the fiber coupled version of the S-series package. Also available as free-space version. Monocrom offers a broad variety of wavelength and output power combinations.

2. MONOCROM PRODUCTS THAT SERVE RAMAN SPECTROSCOPY APPLICATIONS
2.1 Low power, single frequency diode laser

Our S-series (depicted in Fig. 2.1) is ideal for the most common Raman spectroscopy set-ups. It comes with free-space or with SM-fiber output. Monocrom offers a broad spectrum of wavelength and output power combinations. Single frequency versions are available with a linewidth as low as a few tens of MHz and a high side mode suppression ratio (SMSR) of typically 50 dB. On the other hand side the SM-fiber coupling capabilities deliver superb beam quality, which is mandatory for a high brightness Br (see Eq. 1.2) and a high spatial resolution. The latter is important in Raman microscopy applications. Moreover, it opens the possibility to use PM-fibers for polarization depending Raman spectroscopy. Since most Raman spectroscopy applications need a stable output power Plas and a stablewavelength  λin over the integration time necessary to acquire a complete spectrum the S-series comes with athermoelectric cooler (TEC). The footprint of the standard package is 100×100mm2 but other packages can be manufactured on request.

2.2 High energy solid state laser with/without frequency conversion

Our high energy diode pumped solid state laser (HESSL), which can be seen in Fig. 2.2, open the possibility to step into all coherent Raman spectroscopy applications as well as remote Raman spectroscopy. The linewidth ( λin < 0.1nm) of the fundamental wavelength ( λin = 1064nm) transforms to < 1.77 cm−1— still narrow enough for the most requirements. The laser system is available in repetition rates 1 < Vrep < 500 Hz achieving up to Epulse = 1 J. Due to the high available output power the laser is suitable for remote Raman spectroscopy. Our high energy solid state laser  delivers a high power (< 2%@8 h) as well as pulse-to-pulse (< 1%rms) stability. Depending on the desired application the pulse width can be chosen between 4 ns < τpulse < 25 ns. It has to be kept in mind that τpulse influences Epulse and vice versa. Thanks to the high fundamental pulse energy, frequency conversion to the 2nd, 3rd or 4th harmonic are easily possible with high Epulse, harm. The footprint of the laser system is 900 × 500mm2 and can be adapted to customers needs.

Figure 2.2: Picture of the high energy solid state lasers

References

[1] J. Heath, N. Taylor: “Raman Microscopy”, John Wiley & Sons Ltd, 2017

[2] E Smith, G. Dent: “Modern Raman Spectroscopy – A practical Approach”, John Wiley & Sons Ltd, 2005

[3] X. Liu: “Organic Semiconductor Lasers and Tailored Nanostructures for Raman Spectroscopy”, Dissertation, Karlsruher Institute for Technology, 2015

[4] C. Raman, K. Krishnan: “A new type of secondary radiation”, Nature, Vol. 121, 501-502, 1928

[5] G. Landberg, L. Mandelstam: “Eine neue Erscheinung bei der Lichtzerstreuung in Kristallen”, Naturwissenschaften, Vol. 16, 557-558, 1928

[6] P. Vandenabeele: “Practical Raman Spectroscopy”, John Wiley & Sons Ltd, 2013

[7] E. Smith and G. Dent: “Modern Raman spectroscopy: a practical approach”, John Wiley & Sons Ltd, 2005.

[8] K. Kneipp: “Surface-enhanced Raman scattering”, Physics Today, Vol. 60, 40–46, 2007

[9] P. J. Larkin: “IR and Raman Spectroscopy – Principles and Spectral Interpretation”, Elsevier, 2011

[10] Z. You: “Application of Infrared Raman Spectroscopy in Analysis of Food Agricultural Products”, AIDIC Servizi S.r.l., Chemical Engineering Transactions, Vol. 59, 763-768, 2017 [11] J. Depciuch et al.: “Application of Raman spectroscopy and infrared spectroscopy in the identification of breast cancer”, Applied Spectroscopy, Vol. 70(2), 251-263, 2016

[12] A. A. Kadik et al.: “Application of IR and Raman spectroscopy for the determination of the role of oxygen fugacity in the formation of n–Ñ–Î–Í molecules and complexes in the iron-bearing silicate melts at high pressures”, Geochemistry International, Vol. 54(13), 1175-1186, 2016

[13] J. Yu et al.: “Recent applications of infrared and Raman spectroscopy in art forensics: a brief overview”, Applied Spectroscopy Reviews, Vol. 50(2), 152-157, 2015

[14] Z. Zhang: “Raman Spectroscopic Sensing in Food Safety and Quality Analysis”, University of Nebraska-Lincoln, 2017

[15] J. Chen, D. Dong, S. Ye: “Detection of pesticide residue distribution on fruit surfaces using surface-enhanced Raman spectroscopy imaging”, The Royal Society of Chemistry, Vol. 8, 4726-4730, 2018

[16] Y. S. Li, J. S. Church: “Raman spectroscopy in the analysis of food and pharmaceutical nanomaterials”, Elsevier, Journal of food and drug analyzes, Vol. 22, 29-48, 2014