Using AFM-IR for the Nanoscale Chemical Characterization of Polymeric and Thin Film Samples

In government, academic, and industrial research and development laboratories, infrared (IR) spectroscopy is regarded as one of the most established analytical measurement methods for characterizing polymeric materials.

Generally, Abbe diffraction laws restrict the spatial resolution of traditional bulk IR spectroscopy to between 3 and 10 μm, based on the type of technique used. Atomic force microscopy, or AFM, is a popular nanoscale imaging method that offers users a high spatial resolution topographic map of the surface of a sample. However, AFM cannot be used for the chemical characterization of materials underneath the tip and this was considered a major disadvantage until now.

When integrated with an IR source, the ensuing AFM-IR method breaks the diffraction limit of traditional IR spectroscopy by orders of magnitude and, at the same time, provides the high-resolution imaging capabilities of the AFM technique.¹ This article discusses the application of AFM-IR to tackle nanoscale chemical characterization on an array of thin film and polymeric samples.

Resonance-Enhanced AFM-IR

The latest combination of AFM and pulsed tunable IR laser sources has made it possible to collect IR spectra at spatial resolutions less than 100 × 100 nm.^2,3 The sharp AFM tip serves as a local detector of IR absorbance at the sample surface it contacts. The IR radiation is absorbed when the laser source’s wavenumber is in resonance with a molecular vibrational frequency, and the sample will expand once the molecules return to their ground vibrational state after exchanging energy with the sample matrix.

This results in thermal expansion of the sample over an area relative to the focused IR laser spot. In addition, the local thermal expansion of the material close to the apex of the AFM probe causes the AFM cantilever to deflect, offering a considerably higher spatial resolution that is no longer restricted by the diffraction limit of the IR wavelength. In the preliminary configuration of this method, the optical parametric oscillator, or OPO, tunable laser source had a pulse length of ~10 ns and a repetition rate of 1 kHz, which would cause the sample to expand rapidly and induce an impulse in the cantilever.

This, in turn, would cause the oscillation of the cantilever to ring down at its natural resonance frequencies following each laser pulse. This article demonstrates how the signal enhancement of the AFM-IR signal by two orders of magnitude is achieved by substituting the OPO tunable laser source with a variable repetition rate quantum cascade laser, or QCL. This signal enhancement is achieved by tuning the repetition rate of the QCL to correspond with the contact resonant frequency mode of the AFM cantilever.^2,3

Then, at the contact resonance, the cantilever’s oscillation amplitude is considerably increased in relation to off-resonance frequencies. The use of a gold-coated AFM tip results in additional enhancement of the AFM-IR signal, creating a “lightning rod” effect that localizes or improves the electric field at the tip apex. The combination of using a gold-coated probe and matching the contact resonance of the AFM cantilever to the repetition rate of the laser makes it possible to collect IR spectra of samples on arbitray substrates down to thicknesses of approximately ~10 nm. Upon depositing the thin film sample onto a gold substrate, an additional increase in the local improvement of the electric field enables measurements as small as 1 nm. This allows the AFM-IR method to detect monolayer coverage of a material on the surfaces of metals at lateral spatial resolutions as small as 25 × 25 nm.

Characterization of Multilayer Films

Multilayer films are extremely significant in a number of products, particularly packaging materials. As multilayer films get thinner and the number of separate layers increases, these structures have to be characterized.⁴ FTIR spectroscopy is a method that is extensively used for defining cross-sections of multilayer films, offering data regarding the chemical nature of each polymer layer, provided they have a thickness of at least a few micrometers. Earlier, films had to be delaminated to examine individual layers that are smaller compared to the diffraction limit of traditional FTIR. AFM-IR has the potential to overcome these limitations and thus can offer true chemical characterization at the nanoscale.

In order to reverse engineer multilayer films, a microtome is used to cross-section the samples and these are subsequently placed on an IR transparent substrate (ZnS) for further analysis, as illustrated in Figure 1a.

Figure 1. AFM height image (a) showing cross-sectioned multilayer film and AFM-IR spectra at 2800–3600 cm⁻¹ (b) and 1200–1800 cm⁻¹ (c). The spectra collected from layers C, E, and G are consistent with a polyamide. The IR spectrum recorded from layer D, which can be considered as the barrier layer in the film, is consistent with polyethylene-co-(vinyl alcohol) (EVOH).

The respectively colored markers denote the subsequent AFM-IR spectra. In this example, the film’s contents were effectively detected as polyethylene and polyamide, respectively (see Figure 1b and 1c).

Within multilayer films, each layer can have a particular function, for example, oxygen or moisture barrier layers. These materials are generally incompatible with each other and, therefore, “tie” layers are required to bind these layers together during the formation of the film. Due to their thickness (<500 nm), the tie layers could not be characterized by FTIR spectroscopy until now.

Such an example can be seen in the association between polyamide (PA) and polyethylene (PE). These polymers are incompatible within films, and therefore, a tie layer is usually needed. An AFM height image and AFM-IR spectra from the boundary region between the PA and PE layer of a cling film cross-section is shown in Figure 2. On the AFM image, the nine colored marker locations match with the AFM-IR spectra of the same color (a spacing of 100 nm).

Figure 2. (a) AFM height image and (b) AFM-IR spectra across PE/PA layer interface.

Tapping AFM-IR of a PEMA/PMMA

Often, polymer blend systems demonstrate proof of phase separation at spatial scales. These are submicron domains and are usually obvious in AFM images of the sample topography and phase. Conversely, AFM alone cannot ultimately establish the chemical composition of the domains seen. Figure 3 illustrates tapping AFM-IR spectra as well as an image that shows the phase separation between the poly (methyl methacrylate) (PMMA) and the poly(ethyl methacrylate) (PEMA) matrix domains based on the typical IR absorption band of PEMA at 1026 cm⁻¹ (ambient condition).

Figure 3. (a) Tapping AFM-IR spectra of PMMA and PEMA; (b) AFM height image showing the locations where the correspondingly colored tapping AFM-IR spectra were collected; (c) tapping AFM-IR image collected with the QCL tuned to 1026 cm⁻¹. Sample courtesy of the University of Minnesota.

Measuring Monolayers

A monolayer island sample of poly(ethylene glycol) methyl ether thiol (PEG) and self-assembled monolayers, or SAMs, of 4-nitrothiophenol (NTP) were deposited on template-stripped gold substrates. Next, AFM topography measurements were used to confirm that the thickness of the NTP monolayer film was below 1 nm, as indicated in Figure 4a.

Figure 4. (a) An AFM deflection image showing SAMs of NTP on a gold substrate; (b) comparison of AFM-IR spectrum (blue) and ATR spectrum (red) of NTP SAMs; (c) array of AFM-IR spectra collected across the red line shown in (a), demonstrating spatial resolutions of ~20 nm.

Figure 4 shows AFM-IR spectra and molecular structures of NTP SAMs on gold (in blue). Each AFM-IR spectrum comes from a sample surface area of 25 × 25 nm, restricted only by the contact area of the AFM probe with the sample.³ For comparison purposes, corresponding IR reflection absorption spectra recorded over a considerably larger area of NTP SAMs are shown in red (see Figure 4c). A robust NTP absorption peak around 1339 cm⁻¹ is associated with the symmetric NO₂- stretching mode, whereas the weaker absorption band around 1175 cm⁻¹ is correlated to an aromatic CH-bending mode. Finally, an array of AFM-IR spectra was acquired across a gap in the monolayer (see Figure 3c); ~20 nm is the spatial resolution obtained for these measurements.

In Figure 5, the AFM topography image (top left) is shown along with an IR absorption image with the IR laser adjusted to the fixed wavenumber of 1340 cm⁻¹ (top right) of a PEG monolayer island film on gold.

Figure 5. (a) AFM topography image; (b) IR absorption image at 1340 cm⁻¹ of a monolayer island film of PEG on gold; and (c) AFM-IR spectrum of one of the PEG islands.

The PEG islands have an approximate thickness of 4 nm, as indicated by the AFM image. At 1340 cm⁻¹, the IR absorption band is assigned to a CH₂-wagging mode and the location of the PEG island regions is confirmed by the image. PEG monolayer island regions down to 25 × 25 nm are effortlessly resolved in the IR absorption image. The broad IR band that is centered at 1102 cm⁻¹ is assigned to the C-O-C antisymmetric stretching mode. Although a strong CH₂-scissoring band would be expected at 1460 cm⁻¹, it is not evident in this AFM-IR spectrum. The absence of the CH₂-scissoring band indicates that the ethoxylate chains extend in a vertical fashion from the gold surface.

Given that the incident is polarized normal to the surface in this experiment, the incident IR radiation will be absorbed only by vibrations with electric dipole-transition moments oriented parallel to this direction. On the other hand, the C-O-C antisymmetric stretching and the CH₂-wagging mode have electric dipole-transition moments that are oriented along the surface normal, and this is the reason for their extremely strong IR absorbances.

Quantifying Chemical Content in Nanoscale Polymer Domains

One of the most important and extensively used polymers is polypropylene (PP). This is because of its processability, tensile strength, heat resistance, and low cost. Yet, the use of PP is mostly restricted by its poor impact resistance, particularly at low temperatures. This consequently led to the development of a novel copolymerization process that makes it possible to blend other polymers with PP in order to enhance the overall performance, while boosting high-impact resistance.⁵

Three separate regions in a high-impact polypropylene (HIPP) material are shown in Figure 6 — the core, the matrix, and the intermediate layer. To improve HIPP’s performance, it is important to have the ability to analyze the chemical composition of each region.

Figure 6: (a) Schematic diagram and (b) AFM phase image showing different regions within high-impact polypropylene sample.

Generally, IR spectroscopy is a robust tool for assessing chemical composition but domains within HIPP are so small that they cannot be examined by traditional FTIR microspectroscopy. Based on the powerful relationship between traditional AFM-IR and FTIR, FTIR was used for creating a calibration from standards with known material composition to associate the copolymers’ ethylene content with the peak area ratio of the CH₂- and CH₃- bending bands at ∼1456 cm⁻¹ and ∼1378 cm⁻¹, respectively.

A close-up AFM image of the nanoscale domains within HIPP is shown in Figure 7. Markers on the image match with the location where AFM-IR spectra were acquired from the matrix (black), the intermediate layer (blue), and the core (red).

Figure 7. (a) AFM height image; (b) AFM-IR map of the methyl symmetric CH-bending at 1378 cm⁻¹; and (c) AFM-IR spectra taken at the locations marked in (a) and (b), normalized to the 1378 cm⁻¹ band, indicative of different ethylene contents as shown by the intensity of the 1456 cm⁻¹ band.

To compare the AFM-IR peak ratio data with the FTIR calibration curve (see Figure 7c), the collected spectra were normalized to 1378 cm⁻¹. The chemical content of each region can be determined by comparing the peak ratios with the calibration curve generated by the FTIR spectra (see Table 1).

Table 1. Average PE content for each domain within HIPP

Conclusions

AFM-IR is considered a robust tool for defining a wide range of polymer materials. Through the combination of AFM and IR spectroscopic capabilities, Bruker’s Anasys nanoIR3 system can acquire topographical images of the surface of a sample and define these materials’ chemical composition with a spatial resolution of <20 nm. This article has clearly demonstrated that AFM-IR can be used for reversing or characterizing engineer polymer multilayer films, determining the molecular orientation in self-assembled monolayer island films, chemically detecting separate components in a polymer blend, and measuring the chemical content of polymer crystallinity with excellent spatial resolution.

References

1. Dazzi, A.; Prazeres, R.; Glotin, F.; and Ortega, J.M.; Optical Letters 30, 2388-90 (2005).
2. Lu, F.; and Belkin, M.A.; Optics Express 29, 19942-47 (2011).
3. Lu, F.; Jin, M.; Belkin, M.A.; Nature Photonics 8, 307-12 (2014).
4. Kelchtermans, M.; Lo, M.; Dillon, E.; Kjoller, K.; Marcott, C.; Vibrational Spectroscopy 82, 10-15 (2016).
5. Tang, F.; Bao, P.; Su, Z., Analytical Chemistry 88 (9), 4926-30 (2016).

This information has been sourced, reviewed and adapted from materials provided by Bruker Nano Surfaces, who has acquired Anasys Instruments. For more information on this source, please visit Bruker Nano Surfaces.