2.1.

Chlorophyll a Preparation

Chlorophyll a from anascystis nidulans algae was commercially obtained (Sigma Aldrich, C6144) and dissolved in a solution of 80% (v/v) acetone (Sigma Aldrich, 650501, Chromasolv plus ≥ 99.9%) with 20% de-ionized water (Millipore Direct 3Q water purification system). The initial concentration of chlorophyll a in the stock solution was 27.3 μM. To obtain solutions with a final concentration of 1.4 μM of chlorophyll a, the stock solution was diluted in 7 mM of sodium phosphate with 1 or 10 mM of sodium chloride (pH = 7.25).

2.2.

Optical Waveguide Spectroscopy of Chlorophyll a

The single-mode, integrated optical waveguide spectrometer (shown in Fig. 2) for the data acquisition reported in this study was developed in our laboratories and reported in detail elsewhere.^{31, 32, 35} The setup used is composed of a broadband light source, a single-mode planar optical waveguide with integrated grating couplers, a flowcell, a monochromator, and a charge coupled device (CCD) array detector. The light source used was a tungsten-halogen lamp (Philips, FocusLine) with 4.5 V and 2.55 A. An optical filter (BG 40 from Schott) was placed in front of the light source to limit the spectral content inside the probing region to wavelengths shorter than 650 nm and to prevent the near infrared wavelengths, which are not coupled to the waveguide, from getting inside the flowcell and generating heat and photobleaching of the solution dissolved species.³⁶ To couple broadband light into the single-mode planar integrated optical waveguide, two diffraction gratings (3.4 cm apart from each other) were micro-fabricated into the glass substrate. For this purpose, a photoresist film (Shipley 1805) was photo-patterned by a holographic technique.³⁷ After exposure and development of the holographic pattern, an ion milling dry etching process with argon and CF₄ as etchants was performed to transfer the periodic modulation present in the photoresist film onto the surface of the glass slide substrate. The periodicity of the fabricated grating couplers was 250 nm and the coupled bandwidth was from 400 nm (due to the spectral limitations on the light source) to 500 nm. The broad spectral bandwidth was achieved by a solid immersion lens device to create a highly anamorphic beam of large numerical aperture, as previously described by our group.³¹ A linear polarizer was placed in the optical path to allow for only the transverse electric (TE) light modes to couple into the waveguide. The outcoupled light beam was fiber-guided to a monochromator (SpectraPro 2300, Princeton Instruments) that dispersed the light beam to a charge coupled device (CCD Pixis 400, Princeton Instruments) with each column in the CCD array corresponding to a specific wavelength.

The glass substrates, each with a pair of integrated grating couplers, were coated with a thin-film of aluminum oxide using an atomic layer deposition tool (Beneq TFS200). The alumina thin-film was chosen as the waveguide material due to its high refractive index and low propagation loss (less than 1 dB/cm) over a broad spectral range, including the violet/blue region of the optical spectrum.³² The precursors used in the atomic layer deposition process for the alumina film were trimethylaluminum and water. These precursors were alternately injected into the deposition chamber to create a self-limiting atomic layer on the glass substrate surface for each complete cycle.^{32, 35} The deposition temperature in the reactor chamber was 200°C and the number of cycles to reach the thickness of 178 nm was 1670.

To functionalize the waveguide surface, an organic monolayer was coated over the aluminum oxide layer by molecular vapor deposition (MVD^TM, Applied MicroStructures Inc.).³⁸ The deposition was obtained by sequential vapor injection of the precursors to the reaction chamber under vacuum. To obtain a hydrophilic monolayer, the precursors used were methoxy(polyethyleneoxy)propyltrimethoxysilane (PEG silane, Gelest Inc.) and DI water with a reaction time of 15 min (Fig. 3). To obtain a hydrophobic monolayer, the precursors used were heptadecafluoro1,1,2,2-(tetrahydrodecyl)trichlorosilane (FDTS, Gelest Inc.) and DI water with a reaction time of 5 min (Fig. 4). In both cases, the temperature of the precursors was kept at 55°C and the reaction chamber temperature was 35°C. Before the deposition of the organic films, a 60-s oxygen-plasma cleaning was performed to remove any organic residue from the alumina surface. The presence of the aluminum oxide layer, compared to metal or polymeric surfaces, improves the stability and durability of the organic films deposited because of the higher density of the reactive hydroxyl groups on the surface of the oxide film.³⁹

The chlorophyll a solution was injected into the flowcell with a volume of approximately 2 ml covering the waveguide surface. All the data acquisition was performed at room temperature (25°C).

2.3.

Absorbance Spectra of Chlorophyll a in Solution and Surface

The concentration of chlorophyll a in solution was obtained by measuring the absorbance at the Q band (672 nm) in solution with a conventional spectrophotometer (Cary 300, Varian). The molar absorptivity of chlorophyll a in solution is well known (ɛ ₆₇₂ = 73,265 M⁻¹ cm⁻¹).⁴⁰ To ensure that the chlorophyll a concentration remained unaltered during our experiment, the sample to be measured was collected after the passage through the waveguide flowcell. Figure 5 shows the absorbance spectra obtained for the solutions with the same buffer pH and chlorophyll a concentration, but different concentrations of NaCl; as expected they exhibit a good agreement with each other.

Fig. 5

Absorbance spectra for 1.4 μM of chlorophyll in solutions of different ionic strength. The black curve is chlorophyll a dissolved in 7 mM sodium phosphate plus 10 mM of sodium chloride, pH 7.25. The gray curve is chlorophyll a dissolved in 7 mM sodium phosphate plus 1 mM of sodium chloride, pH 7.25. The different concentration of sodium chloride does not markedly affect the absorbance of chlorophyll a in solution.

To obtain the absorbance spectra of chlorophyll a that were surface-adsorbed onto the functionalized waveguide surface, the integration time for data acquisition in the CCD was fixed at 60 s. The sequence of data collection was the following: first, the buffer solution (without chlorophyll a) was injected into the flowcell and a reference signal was acquired; next, the chlorophyll a solution was introduced into the flowcell and data was sequentially acquired every 10 min until equilibrium in the amount of surface-adsorbed chlorophyll a was reached. A dark current signal was finally acquired by blocking the input light beam.

3.1.

Absorbance Measurements of Chlorophyll a Adsorbed to Chemically Modified Waveguide Surfaces

The absorbance data for surface-adsorbed chlorophyll a are presented in Fig. 6 for the aluminum oxide waveguide functionalized with a hydrophilic surface, and in Fig. 7 for the hydrophobic surface. In all cases, the solution concentration of chlorophyll a injected into the flowcell was held constant, but the sodium chloride concentration was varied at two different levels: 1 and 10 mM. The spectra presented in Figs. 6, 7 correspond to the situation where equilibrium had been reached for the surface-adsorbed species, i.e., enough time had elapsed that no further changes in the spectra were observed. From these data we qualitatively observed that chlorophyll a adsorbs to a higher degree to the hydrophilic surface. In order to quantitatively describe the surface density, the sensitivity of the waveguide spectrometer had to be considered.

Fig. 6

Absorbance spectra of chlorophyll a adsorbed to a hydrophilic surface, acquired after adsorption equilibrium was reached. In both spectra, the amount of chlorophyll a dissolved in solution was the same (1.4 μM), as was the pH of the 7 mM sodium phosphate buffer (7.25). Black shows that the concentration of NaCl was 1 mM; gray shows that the concentration of NaCl was 10 mM.

Fig. 7

Spectra of chlorophyll a adsorbed to a hydrophobic functionalized surface, acquired after the steady state was reached. In both spectra, the amount of chlorophyll a dissolved in solution was the same (1.4 μM), as was the pH of the 7 mM sodium phosphate buffer (7.25). Black show that the concentration of NaCl was 1 mM; gray shows that the concentration of NaCl was 10 mM.

3.2.

Sensitivity of the Waveguide Spectrometer

The sensitivity factor, S, which is defined as the ratio between the absorbance measured using a waveguide platform, A_WG, and the absorbance measured in direct transmission, A_TR, for an arbitrary layer of chromophores in both cases, is given by^{26, 30}

Fig. 8

Sensitivity values calculated for a single-mode waveguide with 178 nm of thickness and considering the following values for the constants: L = 3.4 cm, n_c = 1.33, with the approximation that n_l ≈ n_c, n_s = 1.51. The effective refractive index, N_TE, and the effective thickness, t_eff,TE, were calculated from the waveguide dispersion equation. The refractive index of the alumina film was calculated using the relation n_w (λ) = a + b/λ² + c/λ⁴, where a = 1.645 76, b = 42.898 98, and c = 308 958 233.142 from Ref. 32.

3.3.

Determination of Surface Coverage and Kinetics of Chlorophyll a Adsorbed to Different Functionalized Surfaces

In order to calculate the surface density of chlorophyll a, we solved Eq. 1 for Г

Fig. 9

Adsorption kinetics for chlorophyll a absorbed onto the hydrophilic surface. In both cases, the amount of chlorophyll a dissolved in solution was the same (1.4 μM), as was the pH of the 7 mM sodium phosphate buffer (7.25). Black squares show that the concentration of NaCl was 1 mM; gray circles show that the concentration of NaCl was 10 mM.

Fig. 10

Adsorption kinetics for chlorophyll a in contact to the hydrophobic surface. In both cases, the amount of chlorophyll a dissolved in solution was the same (1.4 μM), as was the pH of the 7 mM sodium phosphate buffer (7.25). Black squares show that the concentration of NaCl was 1 mM; gray circles show that the concentration of NaCl was 10 mM.

The experimental results of surface coverage for chlorophyll a adsorbed to different surfaces and dissolved in different buffer compositions are presented in Table 1; data of surface coverage correspond to values when the equilibrium has been reached. For a 1.4 μM solution of chlorophyll a dissolved in 7 mM of phosphate buffer, we obtained a surface density of 0.209 pmol/cm² for chlorophyll a adsorbed to the hydrophilic PEG surface when the sodium chloride concentration was 1 mM. When the sodium chloride concentration was raised to 10 mM, the surface density increased to 0.528 pmol/cm². For a surface functionalized with a relatively more hydrophobic organic monolayer, the surface density of chlorophyll a adsorbed was 0.125 pmol/cm² for a sodium chloride concentration of 1 mM. When the concentration of sodium chloride was raised to 10 mM, the surface density of chlorophyll a increased to 0.337 pmol/cm². Therefore, our data show that increasing the concentration of sodium chloride in the dissolving solution dictates a higher surface density of chlorophyll a, regardless of the hydrophobicity of the surface. In addition, our data also shows that a higher surface density is reached for a hydrophylic surface indicating that chlorophyll a has a higher affinity to hydrophilic surfaces.

Table 1

Surface density for chlorophyll a adsorbed to different surfaces that was equally dissolved in solutions of phosphate buffer with different concentrations of sodium chloride.

3.4.

Molar Absorptivity of Chlorophyll a Adsorbed to Different Functionalized Surfaces

The molar absorptivity is generally a characteristic spectral fingerprint of a particular material mainly determined by the molecular structure. However, the nano-environment surrounding a particular molecule may affect its structures with possible consequences on the spectral profile of the molar absorptivity. For example, if a molecule is adsorbed to a surface with a specific chemical composition and physical structure, the surface may perturb the spectroscopic properties of the pristine molecule. In addition, the composition of dissolved species in the solvent may also affect the optical properties of a particular analyte. To obtain the molar absorptivity for the adsorbed chlorophyll a across the entire spectral region available from our experimental data, we employed Eq. 1 and the results obtained in Secs. 3.2, 3.3 for the waveguide sensitivity and surface density for each experiment. The results are presented in Fig. 11 for the hydrophilic surface and in Fig. 12 for the hydrophobic surface. In both Figs. 11, 12 we have included the molar absorptivity of the dissolved species as obtained from data shown in Fig. 5. As previously mentioned, Fig. 5 clearly indicates that the increase of sodium chloride concentration in the solution from 1 to 10 mM does not affect the molar absorptivity of chlorophyll a in the dissolved phase. However, the results reported in Figs. 11, 12 clearly demonstrate the dependence of chlorophyll a upon adsorption on both types of surfaces studied here. For the adsorbed molecules it generated a red-shift (5 nm on the hydrophobic surface and 15 nm on the hydrophilic surface) of the Soret band located at approximately 435 nm. A possible explanation for the observed red-shift of the Soret band at higher ionic strength of the bulk solvent is an aggregation of the chlorophyll a molecules. Red-shifted Soret bands have previously been observed for chlorophyll a dissolved in benzene and distributed as a thin liquid film upon a buffered aqueous subphase.²¹ Highly similar shifts of the Soret band have also been observed for chlorophyll dissolved in solutions of water-miscible organic solvents upon the addition of water to the organic solution.^{3, 41} A wealth of experimental evidence supports aggregation of the pigment molecules in these solution studies.³ Aggregation in solutions have been attributed to a stronger hydrophobic effect between chlorophyll a molecules as the dielectric constant of the solvent is increased. Higher order aggregation of micellar chlorophyll a to polymeric forms has also been observed upon the addition of salt (KCl), accompanied by further red-shifts in the absorption peaks.³ If the organic layer on the functionalized waveguide surfaces used in this study are behaving as a thin-film of liquid organic solvent, effectively dissolving both some water and chlorophyll a (or at least the phytyl chain in the case of the hydrophobic surface), then an increase in the ionic strength of the buffer, which increases the polarity of the bulk solvent, might be expected to induce aggregation.

Fig. 11

Molar absorptivity of chlorophyll a adsorbed to hydrophilic surface. Black curve: surface-adsorbed chlorophyll a that was previously dissolved in 7 mM sodium phosphate buffer (pH 7.25), 1 mM sodium chloride; dark gray curve: surface-adsorbed chlorophyll a that was previously dissolved in 7 mM sodium phosphate buffer (pH 7.25), 10 mM sodium chloride; light gray curve: chlorophyll a in solution under the same conditions (obtained from a cuvette in a traditional spectrophotometer).

Fig. 12

Molar absorptivity of chlrophyll a adsorbed to hydrophobic surface. Black curve: surface-adsorbed chlorophyll a that was previously dissolved in 7 mM sodium phosphate buffer (pH 7.25), 1 mM sodium chloride; dark gray curve: surface-adsorbed chlorophyll a that was previously dissolved in 7 mM sodium phosphate buffer (pH 7.25), 10 mM sodium chloride; light gray curve: chlorophyll a in solution under the same conditions (obtained from a cuvette in a traditional spectrophotometer).