XPS Depth-Profiling Studies of Chlorophyll Binding to Poly(cysteine methacrylate) Scaffolds in Pigment–Polymer Antenna Complexes Using a Gas Cluster Ion Source

X-ray photoelectron spectroscopy (XPS) depth-profiling with an argon gas cluster ion source (GCIS) was used to characterize the spatial distribution of chlorophyll a (Chl) within a poly(cysteine methacrylate) (PCysMA) brush grown by surface-initiated atom-transfer radical polymerization (ATRP) from a planar surface. The organization of Chl is controlled by adjusting the brush grafting density and polymerization time. For dense brushes, the C, N, S elemental composition remains constant throughout the 36 nm brush layer until the underlying gold substrate is approached. However, for either reduced density brushes (mean thickness ∼20 nm) or mushrooms grown with reduced grafting densities (mean thickness 6–9 nm), elemental intensities decrease continuously throughout the brush layer, because photoelectrons are less strongly attenuated for such systems. For all brushes, the fraction of positively charged nitrogen atoms (N+/N0) decreases with increasing depth. Chl binding causes a marked reduction in N+/N0 within the brushes and produces a new feature at 398.1 eV in the N1s core-line spectrum assigned to tetrapyrrole ring nitrogen atoms coordinated to Zn2+. For all grafting densities, the N/S atomic ratio remains approximately constant as a function of brush depth, which indicates a uniform distribution of Chl throughout the brush layer. However, a larger fraction of repeat units bound to Chl is observed at lower grafting densities, reflecting a progressive reduction in steric congestion that enables more uniform distribution of the bulky Chl units throughout the brush layer. In summary, XPS depth-profiling using a GCIS is a powerful tool for characterization of these complex materials.

In addition to the C-C/C-S peak at 285 eV, the C1s spectrum of a DTBU SAM exhibits peaks at 286.7 eV, corresponding to the C-O and C-Br environment, and at 289.1 eV, corresponding to the carboxylate carbon (O-C=O).The C1s spectrum of an MUL SAM exhibits a small feature at 286.9 eV, due to the carbon atom adjacent to the hydroxyl group, in addition to the where ꭓBr(sol) is the mole fraction of initiator sites in the solution, Aexp is the experimentally determined ratio of the fitted peak areas corresponding to the C1s carbonyl group and the C-C/C-S peak and Acalc is the calculated ratio of the functional groups in the SAM, assuming that the surface composition reflects that of the solution composition.
Tables S3 -S5 show the results of fitting of the C1s spectra of the samples shown in  For a gold surface covered with a polymeric overlayer, the Au4f signal intensity is expected to be attenuated progressively by inelastic scattering of the escaping photoelectrons as the polymer film thickness increases. 1The Au4f:C1s signal intensity ratio was obtained from XPS spectra of a series of surface-grafted polymers of varying thicknesses determined by spectroscopic ellipsometry.Data are shown in Figure S2 for fully-dense surface-grafted polymer films.It can be seen that the log of the peak area ratio declines linearly as the thickness of the polymer layer increases.Thus, it is possible to determine the thickness of a polymer film directly from the XPS Au4f and C1s spectra providing the grafting density is the same.

Figure S2.
Calibration profile for dry thickness determination of full density brush films using XPS analysis.The natural logarithm of the Au4f7/2/C1s peak area ratio, extracted from the high- Polymer brush thickness / nm resolution narrow spectra, is presented as a function of the polymer dry brush thickness determined by spectroscopic ellipsometry.
Table S6.Fitting details of the data presented in Figure S2.S6.

Determination of chlorophyll content within brushes using XPS analysis
Poly(cysteine methacrylate) (PCysMA) contains one sulfur and one nitrogen atom per repeat unit.After the attachment of a Chl molecule, the repeat unit gains an additional four nitrogen atoms.The Chl/CysMA molar ratios were calculated using the relative sensitivity factor corrected N/S peak area ratios.In samples where the S2p high resolution spectra contained peaks due to the Au-S bonding from the SAM, only the peaks corresponding to the monomer were used.

Tapping mode atomic force microscopy (AFM) of micropatterned surface-grafter pCysMA films
During the Chl coupling reaction the polymer chains become solvated, allowing the pigment to diffuse into the polymer network for binding.This leads to an increase in the thickness of the film.Atomic force microscopy (AFM) analysis of PCysMA grown from micropatterned surfaces shows the difference in height due to the Chl addition (Figure S4).Briefly, a SAM of 11-MUL was exposed to a Coherent Innova 300 C frequency-doubled argon ion laser (λ = 244 nm) through a copper mesh grid.The dose used was 62 J cm -2 .The samples were then rinsed with ethanol and placed in either a 2 mM solution of DTBU in ethanol or a mixed solution of DTBU and 11-MUL in ethanol for 2 h.The samples were removed from the solution, rinsed with ethanol and dried with nitrogen, after which the PCysMA brushes were grown from exposed regions of the samples for 30 min (square regionjs in Fig S4). 2,3The resulting pCysMA squares were analysed by tapping mode AFM before and after the Chl reaction.The fully dense brush film showed a thickness of ca.23 nm before the functionalisation.After the Chl binding, a 30% increase in height was observed, attributed to the covalent coupling of Chl to the polymer scaffold.When ꭓBr(Au) = 0.39 ± 0.07, the film grown for 30 min had a thickness of ca.1.5 nm.The relatively low thickness is likely due to the small elastic moduli of surface tethered polymers, which are easily compressed by the AFM tip during the imaging process.Attachment of the pigment, however, doubles the thickness to 3 nm, based on the cross-section analysis.The doubling of the reduced density polymer film thickness indicates a higher Chl:monomer ratio compared to the fully dense brush, which increases only by ca.1/3 of its original height.

Figure
Figure S1(c).Using equation S1, the values of with ꭓ Br(Au) were determined for each sample

Figure S4 .
Figure S4.High resolution S2p XPS spectra corresponding to the samples presented in

Monomer unit: 1 S + 1 Nthe
Monomer + Chl unit: 1 S + 5 N Since the sulfur concentration on the surface remains unchanged after the Chl attachment, the N/S peak area ratios before !Chl addition, determined from the high-resolution XPS spectra, can be used to account for the excess N within the brush after the coupling reaction.Note that as each Chl contains four nitrogens, the result must be divided by four.Therefore, we can use the following equation to estimate the Chl:

Table S1 .
XPS fitting details of a SAM of DTBU, presented in FigureS1(a).

Table S2 .
XPS fitting details of a SAM of MUL, presented in Figure S1(b).
Mixed SAMs were formed by the co-adsorption of MUL and DTBU.FigureS1(c) shows C1s spectra of three different mixed monolayers.The carbonyl functional group is present only in DTBU; thus the mole fraction of the adsorbed initiator site (ꭓBr(Au)) within the film can be derived from the C1s peak area ratios:

Table S3 .
XPS fitting details of a SAM of DTBU + MUL, with ꭓ Br(Au) = 0.52, presented in

Table S6 .
XPS fitting details of a drop-cast Chl on a gold slide, for the corresponding highresolution spectra shown in FigureS3.

Table S7 .
XPS fitting details of the high-resolution N1s and S2p spectra before and after the functionalisation of PCysMA brushes with Chl, as presented in Figure5and FigureS4.RSF refers to the relative sensitivity factor, which is used to scale the raw peak areas of different elements within the sample surface.For N1s, the RSF value of 1.68 was used, while for S2p the value was 1.79.T is the transmission and MFP is the mean-free path.The N1s/S2p area ratio was calculated from the sum of the RSF corrected peak areas of the N1s high resolution spectra and the S2p peak areas corresponding to the C-S-C peak of the CysMA monomer.