Combined Atomic Force Microscopy and Raman Spectroscopy – TERS and Co-Located AFM-Raman Systems

Atomic force microscopy and Raman spectroscopy are both methods that are used to obtain data about the surface properties of a sample, though their respective user base is generally quite different. The integration of these technologies ensures it can be used for a variety of applications. This application note looks both at the complementary information gained from both techniques and how a researcher who has access to a combined system can benefit from the additional information available.

Atomic Force Microscopy

In atomic force microscopy, a sharp probe is brought near to the sample and held at that distance using a force-based feedback loop. The force on which the primary feedback loop is based, the electrical current, surface potential and specific nanomechanical properties can be measured. By scanning the sample and tip relative to each other and measuring these quantities at discrete locations in a sequential manner, three-dimensional images of selected sample properties can be created. An atomic force microscope (AFM) setup is shown in Figure 1.

Figure 1. Shown are the most basic parts of an AFM system, a tip, raster scan mechanism, and data processing unit.

Raman Spectroscopy

The study of the interaction of electromagnetic radiation with matter is known as spectroscopy. The most common kinds are fluorescence, infrared, and Raman. In a Raman experiment, monochromatic light is focused on the sample and the inelastic scattered light is detected. A Raman spectrometer is often used along with an optical microscope to benefit from the high spatial resolution that a confocal optical setup can provide. The main components of a dispersive Raman setup are a laser illuminating the sample, optics to collect the backscattered radiation, a high-efficiency laser line rejection filter, and a spectrometer with an entrance slit, a diffraction grating and a CCD camera. A basic schematic of a simple Raman spectrometer is shown in Figure 2.

Figure 2. Schematic of a Raman spectrometer. The sample is illuminated by a monochromatic light source. After passing through a filter rejecting the laser light it is dispersed by a grating and imaged onto a CCD chip.

Figure 3 shows a Raman spectrum taken on one of the layers of a cross-sectioned piece of packing material for unfrosted animal crackers using Bruker’s SENTERRA Raman microscope.

Figure 3. Spectrum (black) of a layer of food packaging material and literature spectrum for Poly- propylene (red).

When the spectrum is compared with a literature database, it can be concluded that one of the layers is Polypropylene. It is not possible to determine the nanomechanical characteristics of the layer. But when the sample is transferred from the SENTERRA to the MultiMode 8 AFM having the Peak Force Tapping feature, quantification of parameters such as modulus and adhesion is possible. The resulting modulus data from the AFM are shown in Figure 4.

Figure 4. Modulus map of the cross-sectioned food packaging material on left. The outer layers are identified by the Raman data as polypropylene. The AFM data allow for quantification of the mechanical properties. The cross -sectional plot on the right highlights the drop in modulus from the outer to the inside layer and reveals some particles in the middle layer exhibiting higher modulus than the matrix.

The Next Step

It is possible to acquire an optical map along with topography and typical atomic force microscopy information by using a suitable AFM tip. A general setup for linear and non-linear tip-assisted spectroscopies is shown in Figure 5. Examples for suitable AFMs include Bruker’s Catalyst and Innova for transparent and non-transparent samples respectively.

Figure 5. General setup for linear and non-linear tip-assisted spectroscopies.

When a suitable combination of incident light and tip is chosen, a strong electromagnetic field is formed at the tip apex. Figure 6 shows that the field enhancement by the tip is two-fold. This field enhancement ensures that Raman spectroscopy is made feasible at the nanometer scale.

Figure 6. Field enhancement by the tip.

In the pharmaceutical industry, the occurrence of polymorphs, which is the same chemical compositions but in a different crystal lattice, can be critical for the property of a drug. Raman spectroscopy is used to study polymorphism, and co- localization with several AFM techniques that can increase the productivity of the research carried out.

Solutions for a Combined Approach

While using a combined instrument, it is important not to compromise the performance of either. The two typical factors to be considered include the following.

1. To keep the photodetector noise in an AFM low, typical beam-bounce systems in an AFM operate in the red with a power of about 1mW, which translates to 3x10¹⁸ photons/second. To allow parallel, simultaneous operation of the spectrometer and the AFM, the wavelength of the AFM beam-bounce system must therefore either be altered to the near-IR so as to not interfere with the optical measurements or a non-optical feedback system such as tuning forks must be employed.
2. Spectrometer systems often use several lasers that may be cooled by noisy external fans or water cooling systems or radiate quite an amount of heat in the proximity of the AFM. Both of these effects can negatively impact AFM performance. Noise from heating fans may couple into the AFM and result in instabilities in the feedback loop. Temperature changes will result in the AFM to drift and will make it very tough to keep the tip in the selected field-of-view.

The obstacles involved in combining a Raman microscope and AFM actually make another physical solution viable. That is a shuttle stage designed to transfer samples from one to the other instrument with a possibility to register them to a common coordinate system. A shuttle stage also has the potential benefit of allowing for increased productivity as both instruments can be used simultaneously.

A solution integrating Bruker’s industry-leading Dimension Icon with Horiba’s LabRam achieves the co-localization of data and ensures the high performance of both systems.

The Dimension Icon stage shuttles the sample between the AFM head at the left and the Raman objective at the right. The red spot emanating the objective is the Raman laser illuminating the sample during a Raman measurement. The two systems were mechanically coupled by using the accuracy of the Icon stage to shuttle the sample between the AFM and Raman microscope. Figure 7 shows the combined instrument with the sample in the position for (a) AFM imaging and (b) Raman imaging.

Figure 7. View of the Dimension Icon stage and optics arm of the Raman microscope.

Bruker has introduced the ScanAsyst, which almost completely automates AFM operation without scarifying performance. The following section discusses some results from co-localized measurements. Another solution from Bruker is the integrated design of the benchtop NEOS AFM and SENTERRA confocal Raman system shown in Figure 8, which allows for straightforward sample handling without the need to transfer the sample between methods for co-located measurements.

Figure 8. The Bruker NEOS SENTERRA AFM-Raman Spectroscopy System.

The NEOS AFM is highly compact housed in a microscope objective and integrated into an upright optical microscope, whereas the SENTERRA is a benchtop confocal Raman microscope integrated into an upright optical microscope. This enables the system to work in AFM and Raman modes and to utilize standard optical techniques, such as Nomarski differential interference contrast (DIC) to interrogate the sample.

Co-Localized AFM and Raman Measurements

Example One

The following section speaks about some results from co-localized measurements. The first shows an epoxy compound on a metal substrate. The analysis commences with the selection of an area using for instance, regular brightfield contrast. It is possible to obtain AFM and Raman in the order chosen by the user. Figure 9 shows such a sequence. The bright-field optical image is shown on the left and the sample topography as acquired by the AFM is shown on the right. Integration of a user selected spectrum area results in a Raman map in the middle of the sequence.

The two spectra in Figure 10 were taken at points of different heights of the sample. The highest intensity phenyl-ring vibration at 1004 cm^-1 can be found in the lower regions of the sample whereas intensities compared to the non-aromatic reference line at 1014 cm^-1 are lower in the thicker portions of the sample. The Raman map in the middle of Figure 9 shows this clearly. The steric orientation of the sample is known with Raman data. The molecular orientation is different in the thick and thinner areas of the sample.

Figure 9. AFM-Raman acquisition sequence with (left) bright-field optical image, (middle) Raman map, and (right) AFM sample topography image.

Figure 10. Two Raman spectra taken from sample in figure 9.

Example Two

Polymorphism is the ability of a material to exist in more than one crystal structure. The following example describes a study on Yttria-stabilized zirconia polycrystals (Y-TZP). Y-TZP is often used in dental implants for its biomechanical and esthetic characteristics. The material is sintered from a fine powder and can be crystallized in the tetragonal form. The two spectra shown in Figure 11 were obtained on different locations of a sample. The peaks on the spectra shown in red are attributed to Y-TZP. The blue spectrum clearly shows more peaks. After isolating the additional peaks and comparing them to literature data, they are assigned to the monoclinic phase of ZrO₂.

Figure 11. Raman spectra of ceramic sample. Red curve shows only Y-TZP while green curve suggests presence of additional phase

With the help of the built-in DIC optical contrast and the AFM objective, a region of the sample was identified that shows two patches of charcteristically different surface morphology. With the AFM it is possible to quantify the roughness further to just defining them as smooth or rough. The average roughness of the smooth region is 8.7 nm, whereas the rough region averages 15.7 nm. The automated analysis features of the NEOS AFM enable further analysis. Grain size may play a significant role in the change from tetragonal to monoclinic. Grain sizes can be extracted from the AFM data. Figure 12 shows an 83 x 83 um² DIC image of a smooth and rough patch and a 10x10um²AFM image of the smooth area of the sample highlighting the grains. The analysis yields an average grain size of 0.56 um².

Figure 12. DIC and grainsize data. The optical image (left) depicts a smooth and rough patch on the sample. AFM data (right) of the smooth region provides grain size detail after automated grain recognition and analysis.

With the mapping feature of the SENTERRA Raman microscope, it is possible to obtain a map of an area showing the previously topographically characterized smooth and rough areas. Once a map of Raman spectra is obtained, the SENTERRA software enables plotting the integrated intensities of a user-selected area. In this example, an area from 180-184 cm^-1 was selected, as this region highlights a peak only present for the monoclinic phase. By showing the intensities in an appropriate color scheme, a two-dimensional Raman map of the tetragonal and monoclinic occurrence is generated. The map and corresponding DIC image is shown in Figure 13. Using co-localized AFM, Raman, and DIC microscopy enabled the study of the process at the nanometer scale.

Figure 13. DIC and Raman map.

A Setup for TERS: Innova IRIS

In TERS, the tip needs to be as close to the sample as possible without affecting the integrity of the tip or the sample. In addition, a metallic tip is essential for the enhancement. STM provides an easy way to integrate these requirements and study the impact of several tip shapes, coupling mechanisms, and other variables. An excellent way to excite the AFM tip and collect the Raman signal is to place the Raman objective at a 60° angle with reference to the tip axis. The side illumination scheme has shown the highest enhancement factor for TERS in theoretical studies. A setup using this side-on geometry is realized in the Innova IRIS shown in Figure 14.

Figure 14. TERS-ready combination of the Bruker Innova Scanning Probe Microsope and the Renishaw inVia Raman microscope. The optical coupling is achieved via a trackball operated sampling arm.

Due to its openness, the Innova lends itself as a platform for TERS on opaque samples; it has a very stable and low-noise closed-loop feedback system, and a near-IR feedback diode. It can be operated in STM and a variety of AFM modes with convenient switching. The Innova is integrated with Renishaw’s inVia microscope to enable TERS, confocal Raman, and co- localized measurements.

The control of both the AFM and Raman microscope is done by a software package present in the AFM computer. An example of a TERS dataset achieved with such a setup is given in Figure 15. The sample used is Malachite Green, a dye for which literature data is available. Spectra like the one presented can be acquired in as little as 0.1s using just a few micro- watts of incident laser power.

Figure 15. TERS spectra of Malachite Green obtained using a gold tip illuminated by 633nm light at varying distances above the surface. Data acquired using the IRIS Innova-InVia combination. By comparing the peak intensities with the tip approached to the retracted spectra, one can clearly see the enhancement of Raman modes.

Conclusion

Since co-localized instrumentation is possible, researchers can study samples using optical spectroscopy techniques such as Raman and scanning probe techniques providing detailed information about nanoscale properties and composition. Bruker provides solutions for opaque samples and transparent samples with the Dimension Icon, BioScope Catalyst, and NEOS SENTERRA systems. TERS promises to push the resolution limits even further and enable the collection of chemical information on the nanometer scale. Bruker’s solutions for this advanced research include the Catalyst and Innova for transparent and opaque samples, respectively.

This information has been sourced, reviewed and adapted from materials provided by Bruker Nano Surfaces.

For more information on this source, please visit Bruker Nano Surfaces.