14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
The study of mixed self-assembled monolayers of perylenedii- mide and hexanethiol by cyclic voltammetry and absorption spectroelectrochemistry shows a non-linearity of absorbance maxima as a function of the surface coverage. This surprising result could be supported by an increase in the average tilt
angle at low coverage, suggesting that the dilution of redox species through an alkanethiol would induce a change of the orientation of immobilized chromophores and, therefore, a modification of the optical properties of the monolayer.
1. Introduction
Electrochemistry is known to be a powerful tool to probe redox-active self-assembled monolayers (SAMs)[1]and establish detailed structure-reactivity relationships for interfacial reac- tions, especially on mixed SAMs by taking advantage of its high sensitivity, specificity, accuracy and temporal resolution.[2–6]
Concerning absorption spectroelectrochemistry on SAMs, many works involving an electrochemical/spectroscopic cou- pling have been dedicated to this research field. Restricted to thiolate-on-gold, a few works have been devoted to UV-Vis spectroelectrochemistry and essentially performed in potentio- static conditions in order to visualize in situ or ex situ the optical bands of the electroactive species under different redox states or products arising from redox reactions.[7–11] Even rarer still is the real-time monitoring of a spectroscopic signature as a function of an electrical perturbation (potential step or linear scan) on SAMs. Two of our recent works have been dedicated to improve significantly resolution of absorption time-resolved spectroelectrochemistry (A-SEC) coupled to cyclic voltammetry on redox thiolate-on-gold. Drawing upon the benefits provided by the latest technological breakthroughs (i. e. high signal-to- noise ratio and high sampling rate of CCD detectors), we have demonstrated the possibility of monitoring very low absorb- ance variations of electroactive species (thiophene[12] and perylenediimide[13]) to extract voltabsorptograms of the corre- sponding SAMs.
The present work takes advantage of A-SEC for probing mixed SAMs derived from dialkyl disulfide tetrachloro-perylene- diimide (1) and alkanethiol (Schema 1) and seeks to establish
structure-property relationshipsviathe study of optical proper- ties (i. e. absorbance) versus the surface coverage.
2. Results and Discussion
As expected[13], SAM of1exhibits two successive fully reversible one-electron processes in negative direction, at 0.60 V and 0.75 V vs. AgNO3/Ag (Figure 1). The shape of voltammetric waves and the linear dependency between peak intensities and scan rates are characteristic of surface-confined redox species.
At 298 K, the full steady-state coverage (Gmax) was assessed to 1.50.1·1010mol cm2. This value fits with the one determined in 0.1 M Bu4NPF6/CH3CN and agrees with previous works.[14] Moreover, the surface coverages calculated from integrated charges of CVs corroborate those estimated by quartz crystal microbalance measurement during adsorption, and consequently indicate that all redox species are engaged in redox processes.
Estimated by direct analysis of experimental CVs and refined by curve fitting from two Generalized Lateral Interactions functions (GLI functions) of equal area on the basis of the procedure from reference,[15] the values of full width at half maximum (FWHM) deviate from the expected value (i. e.
~91 mV at 303 K) of an “ideal system”, based on a Langmuir model (i. e. all adsorption sites are equivalent and there are no interactions between immobilized electroactive centers).
This effect decreases on mixed SAMs with decreasing surface coverage (G), suggesting the presence of interactions between electroactive species only[16–18] (Figure 2). Analysis of the experimental data (i. e.EPeak andiPeak vs.G) by curve fitting leads to an estimation of the S and G interaction parameters involved during the first (S10 and G1=0.940.03) and second (S20 and G2=0.100.03) reduction steps (Fig- ure 2 A). These values are consistent with strong attractive interaction between 1 and 1C and a tiny lateral interaction [a] S. Bkhach, Dr. O. AlvÞque, Y. Morille, Dr. T. Breton, Prof. P. Hudhomme,
Dr. C. Gautier, Dr. E. Levillain Laboratoire MOLTECH-Anjou Universit d’Angers/CNRS UMR 6200
2 Boulevard Lavoisier, 49045 Angers Cedex (France) E-mail: eric.levillain@univ-angers.fr
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
during the second process. As previously demonstrated on TEMPO based SAMs, a linear dependence of FWHM vs. G traduces a random distribution and a deviation from linearity at low surface coverage suggests a phase segregation.[19]Because the small gap (150 mV) between the first and the second apparent redox potential leads to uncertainties in estimating the FWHM at low surface coverage, information about the redox sites distribution cannot be precisely assessed. Never- theless, the quasi linearity of FWHMvs.Gseems to indicate that the distribution of redox sites is rather random (Figure 2B).
Mixed SAMs of1have been studied by A-SEC between 350 and 950 nm from 5·1012to 1.5·1010mol cm2(Figure 3). Note that below 5·1012mol cm2, raw data are not fully exploitable because of the very low signal-to-noise (i. e. the detection limit of our bench is equal to an absorbance value of 104 in the worst case at 400 nm).
According to references,[13, 20]the absorption bands close to 540, 640 and 755 nm are assigned to 1, 12 and 1C species, respectively. The position and the shape of optical bands were found independent of the surface coverage, suggesting that the molecular structure of electroactive species does not change over the successive dilutions. It is noteworthy that no additional optical band was observed versus the surface cover- age, whichde factoexcludes a possible formation of dimers or multimers.
According to the book “Fundamentals of Analytical Chemistry”,[21] “Beer’s law describes the absorption behavior only of dilute solutions and in this sense is a limiting law. At concentrations exceeding about 0.01 M, the average distances between ions or molecules of the absorbing species are diminished to the point where each particle affects the charge distribution and thus the extent of absorption of its neighbors.
Because the extent of interaction depends on concentration, the occurrence of this phenomenon causes deviations from the linear relationship between absorbance and concentration”.
Nevertheless, despite these limitations, Beer’s law for sur- face-confined species is usually expressed as shown in Equa- tion (1)[22–25]:
Absorbance¼21000:e:G with
e;M1cm1 G;mol cm2 8<
: ð1Þ
Note here that the 2 factor is due to our spectroelectro- Scheme 1.Mixed SAMs are prepared by using a successive adsorption
protocol, immersing SAMs of1(dialkyl disulfide perylenediimide; see the Experimental Section) in a millimolar hexanethiol solution for the appropriate time to achieve the expected surface coverage (i. e., from 1 min to 12 h).
Figure 1.Cyclic voltammetry of mixed SAMs in 0.1 M Bu4NPF6/CH2Cl2at 100 mV s1and 303 K in a glovebox. The surface coverages are expressed as a percentage of the full steady-state coverage. Note that the reproducible shape of the CV at 100 % (dashed line) is different from those of mixed SAMs, as previously published in Ref. [13]. At this stage, this surprising behavior is uncommon and requires some additional experiments to provide a clear explanation. Also note that the same electrochemical behavior is observed with mixed SAMs prepared from successive dilutions with octa-, dodeca- and octadecanethiols. Insert: linear dependency between peak intensities of the first reduction step (i) and scan rates (v) for 81% of the surface coverage.
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
chemical setup which imposes an absorption measurement in reflection mode. The linear dependence of maximum absorb- ance vs. surface coverage is still valid if the chromophore remains in the same configuration and if the interactions between the molecules are negligible.
To confirm the equation 1, the linear dependence of maximum absorbance vs. surface coverage has been examined and, surprisingly, maximum absorbance values at 540, 670 and 755 nm are not linearly dependent on the surface coverage (Figure 4 A).
previous works have established that the transmission spectrum of a Langmuir-Blodgett film or organic thin films depends on the tilt angle (a) of the immobilized chromophores, defined with respect to the surface normal direction. In considering these findings and assuming that all PDI species are oriented in the same direction and randomly distributed in the SAM (i. e. a crude approximation), the anisotropic values of extinction coefficientej jande? (i. e. parallel and perpendicular projections to the surface normal direction, respectively) of 1 could be expressed[23]as Equation (2)
e?¼32eSAMsin2ð Þa ek¼3eSAMcos2ð Þa 8<
: witheSAM:extinction coefficient in SAM ð2Þ Note that the angle a represents the angle between the surface normal and the transition dipole moment and the orientation between this transition dipole and the molecule defines the final molecular orientation. In first approximation, we can consider that the transition dipole direction might coincide with the long molecular axis of the three PDI redox derivatives,[26–28] which can be supported by the trend of the normalized absorbance variations observed in Figure 4b. Even if each redox specie has its own tilt angle, we assume that these three angles are very close. At this stage, we can also postulate that the average tilt angle depends on the surface coverage because the degree of freedom of immobilized redox species increases at low surface coverage.[2, 19]
Accordingly, the surface coverage dependence of the extinction coefficient could be expressed through e? (i. e. due to the reflection mode at normal incidence), leading to ana- based formula of the Beer’s law [Eq. (3)]:
Absorbanceð Þ ¼G 3000:eSAM:sin2½a Gð Þ:G ð3Þ Because eSAM cannot be approximated by the molecular extinction coefficient of species in solution,[23] a normalization atG=Gmaxof equation 3 allows to extract an expression ofa independent ofeSAM[Eq. (4)]:
Figure 2.Fitted parameters extracted from CVs of mixed SAMs in 0.1 M Bu4
NPF6/CH2Cl2at 100 mV s1and 303 K). A) Peak intensity versus surface coverage. The solid lines correspond to a random distribution (Laviron’s modelipeakð Þ ¼G n2RTF2vA2 2GGG
Gmax
ð Þ, withA=area andv=scan rate).
G1=0.940.03 andG2=0.100.03, extracted by fitting of experimental data. B) FWHM versus surface coverage. The solid lines correspond to a random distribution (Laviron’s model) and the dashed lines to a phase segregation model, which have been simulated with the previous values of G1andG2.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
a Gð Þ ¼arcsin sin½a Gð maxÞ:
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Absorbanceð ÞG AbsorbanceðGmaxÞ:Gmax
G
" s #
ð4Þ
Note thata(G!0) depends on the initial value ofa(Gmax).
According to the previous works of Brochsztainet al.[14]and Saavedra et al.,[27] the average tilt angle between the PDI molecular axis and the ITO surface normal reaches 308 at the full surface coverage. On the assumption that a(Gmax) for 1- based SAM on Au substrate is of the same order, the average tilt angle vs G can be assessed from experimental data (Figure 5). The extrapolation of the tilt angle at the nil surface coverage reaches a value of 708leading to a 408angle shift at low surface coverage. Such angle changes were previously observed on mixed SAMs.[29–31]
3. Conclusions
This work shows that absorption spectroelectrochemical experi- ments on mixed SAMs afford the identification of electroactive species under different redox states or products arising from redox reactions with an absorbance detection limit close to 104.
A-SEC provides an opportunity to probe the structure of the monolayer in order to establish structure-properties relationships. Maximum absorbance of a dialkyl disulfide PDI and its reduced forms are not linearly dependent on the surface coverage, probably due to the change of the angle orientation of the electroactive species occurring during the dilution.
However, additional data are required to confirm this behavior and understand the impact of the distribution of immobilized chromophores on optical properties.
Experimental Section
Chemicals, Au Substrate, and SAM Preparation
The synthesis of the dialkyl disulfide perylenediimide (PDI) 1 (Scheme 1) was previously described.[13] We should take into account that the introduction of four chlorine atoms at the 1,6,7,12 positions of the bay region enforces a considerable twisting close to 35–378of the PDI skeleton as a result of electrostatic repulsion and steric effects among the substituents.[32–34]
Au substrates were prepared as previously described in[35]and were made immediately before use.
SAMs of 1 were prepared on fresh Au substrates (0.2 cm2) by immersion for 15 min in 0.5 mM solution of 1 (i. e. 1 mM per PDI moieties) in dichloromethane at 303 K. Mixed SAMs were prepared using a successive adsorption protocol, by immersing SAMs of 1 in a millimolar hexanethiol solution for the appropriate time to achieve the expected surface coverage (i. e. from 1 min to 12 h).
Spectroelectrochemical Experiments
Electrochemical measurements were carried out with a Biologic SP- 150 potentiostat driven by the EC-Lab software including ohmic drop compensation. Experiments were recorded in dry HPLC-grade dichloromethane with tetrabutylammonium hexafluorophosphate (nBu4NPF6, electrochemical grade, Fluka) as supporting electrolyte.
In order to use commercially available thermostated cell holders, the spectroelectrochemical cell is dimensionally close to the conventional quartz cuvette (outer dimensions=12.5 mm 12.5 mm 45 mm). The inner part of the cuvette (HellmaAnalytics) has been specially redesigned in order to insert, parallel to the quartz windows, a home-made interdigitated three Au electrodes with high precision.[4] Spectrophotometric measurements were carried out in direct reflecting mode on the working electrode with a home-made bench composed of Princeton Instruments modules (light sources, fibers, monochromators, spectroscopy camera and software). The connection between the light source, the cell and the spectrophotometer is ensured through a “Y-shaped” optical fiber bundle: 18 fibers guide the light to the cell, and 19 fibers collect the reflected light from the cell to visible (320-1080 nm/
Figure 3.Absorption spectroelectrochemical experiments on mixed SAMs in 0.1 M Bu4NPF6/CH2Cl2at 10 mV s1and 293 K. On theyaxis (i. e., time in s), 0, 6 and 12 s correspond to the potentials0.25 (first vertex),1.05 (potential of reverse scan) and0.25 V (second vertex) of the CVs in Figure 1 respectively.
Note that the absorbance at a given potential was determined by comparison with a reference spectrum recorded at the equilibrium potential.[13]
A)G=1.450.06 1010mol cm2, i. e., 96% of the full steady-state coverage.
B)G=5.60.2 1012mol cm2, i. e., 4% of the full steady-state coverage.
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
maximum acquisition frequency 2 MHz) and IR (900-1700 nm / maximum acquisition frequency 8 MHz) CCD detectors. The sensitivity of the spectroscopic measurement (<3 e at 100 kHz and <13 e at 2 MHz between 320 and 1080 nm; 400 e (high gain) and 5000 electrons (low gain) between 900 nm and 1700 nm) allows performing a spectroelectrochemistry experiment under the usual conditions of electrochemistry.
Acknowledgements
The authors gratefully acknowledge the MENRT for a PhD grant to S.B. The authors thank Hellma Analytics and Princeton Instruments companies for the quality of their products and
services. This work was supported by the Contrat Plan E´tat Re´gion 2007–2013 (Pays de la Loire - France).
Conflict of Interest
The authors declare no conflict of interest.
Keywords: Absorbance · perylenediimide · self-assembled monolayers · surface coverage · time-resolved spectroelectrochemistry
[1] A. L. Eckermann, D. J. Feld, J. A. Shaw, T. J. Meade,Coord. Chem. Rev.
2010,254, 1769–1802.
[2] J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo, G. M. Whitesides,Chem Rev2005,105, 1103–1170.
[3] S. Zhang, C. M. Cardona, L. Echegoyen, Chem Commun2006, 4461–
4473.
[4] P.-Y. Blanchard, S. Boisard, M. Dias, T. Breton, C. Gautier, E. Levillain, Langmuir2012,28, 12067–12070.
[5] O. Aleveque, T. Breton, E. Levillain,Soft Matter2012,8, 3875–3880.
[6] C. Orain, A. G. Porras-Gutirrez, F. Evoung Evoung, C. Charles, N.
Cosquer, A. Gomila, F. Conan, Y. Le Mest, N. Le Poul, Electrochem.
Commun.2013,34, 204–207.
[7] D. D. Schlereth, H.-L. Schmidt,J. Electroanal. Chem.1995,380, 117–125.
[8] T. Sagara, T. Midorikawa, D. A. Shultz, Q. Zhao, Langmuir 1998, 14, 3682–3690.
[9] J. Lukkari, K. Kleemola, M. Meretoja, T. Ollonqvist, J. Kankare,Langmuir 1998,14, 1705–1715.
[10] D. A. Brevnov, H. O. Finklea,J. Electrochem. Soc.2000,147, 3461–3466.
[11] T. Sagara, Y. Kubo, K. Hiraishi,J. Phys. Chem. B2006,110, 16550–16558.
[12] O. AlvÞque, E. Levillain, L. Sanguinet,Electrochem. Commun.2015,51, 108–112.
[13] S. Bkhach, Y. Le Duc, O. AlvÞque, C. Gautier, P. Hudhomme, E. Levillain, ChemElectroChem2016,3, 887–891.
[14] B. P. G. Silva, D. Z. de Florio, S. Brochsztain,J. Phys. Chem. C2014,118, 4103–4112.
[15] O. AlvÞque, E. Levillain,Electrochem. Commun.2016,67, 73–79.
Figure 4.Data processing from A-SEC. A) Maximum absorbances at 540, 670 and 755 nm extracted from absorption spectroelectrochemical experiments of mixed SAMs in 0.1 M Bu4NPF6/CH2Cl2at 10 mV s1and 303 K. The dashed lines are trend curves. B) Normalized absorbance maxima (rightyaxis) versus surface coverage. The absorbance values are normalized to 100% atG=Gmax. The dashed line is a trend curve.
Figure 5.Calculated average tilt angle (left y-axis) vs. surface coverage. The solid line visualizes the arbitrary value of tilt angle at the full steady-state coverage (i. e., 308). The dashed line is a trend curve.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
[16] E. Laviron,J. Electroanal. Chem.1979,101, 19–28.
[17] O. Aleveque, P. Y. Blanchard, T. Breton, M. Dias, C. Gautier, E. Levillain, F.
Seladji,Electrochem. Commun.2009,11, 1776–1780.
[18] O. Aleveque, P. Y. Blanchard, C. Gautier, M. Dias, T. Breton, E. Levillain, Electrochem Commun2010,12, 1462–1466.
[19] O. Aleveque, C. Gautier, M. Dias, T. Breton, E. Levillain,Phys Chem Chem Phys2010,12, 12584–12590.
[20] R. O. Marcon, S. Brochsztain,Langmuir2007,23, 11972–11976.
[21] D. A. Skoog, D. M. West, F. J. Holler, S. R. Crouch, Fundamentals of Analytical Chemistry 9th Edition,2014.
[22] T. Hasegawa, S. Takeda, A. Kawaguchi, J. Umemura,Langmuir1995,11, 1236–1243.
[23] T. Hasegawa, Y. Ushiroda, M. Kawaguchi, Y. Kitazawa, M. Nishiyama, A.
Hiraoka, J. Nishijo,Langmuir1996,12, 1566–1571.
[24] A. L. Bramblett, M. S. Boeckl, K. D. Hauch, B. D. Ratner, T. Sasaki, J. W.
Rogers,Surf. Interface Anal.2002,33, 506–515.
[25] T. Tang, J. Qu, K. Mllen, S. E. Webber,Langmuir2006,22, 26–28.
[26] P. A. J. De Witte, J. Hernando, E. E. Neuteboom, E. M. H. P. van Dijk, S. C. J. Meskers, R. A. J. Janssen, N. F. van Hulst, R. J. M. Nolte, M. F.
Garca-Paraj, A. E. Rowan,J. Phys. Chem. B2006,110, 7803–7812.
[27] Y. Zheng, A. J. Giordano, R. C. Shallcross, S. R. Fleming, S. Barlow, N. R.
Armstrong, S. R. Marder, S. S. Saavedra, J. Phys. Chem. C 2016, 120, 20040–20048.
[28] H. J. Park, M. C. So, D. Gosztola, G. P. Wiederrecht, J. D. Emery, A. B. F.
Martinson, S. Er, C. E. Wilmer, N. A. Vermeulen, A. Aspuru-Guzik, et al., ACS Appl. Mater. Interfaces2016,8, 24983–24988.
[29] F. Schreiber,Prog. Surf. Sci.2000,65, 151–257.
[30] M. Zwahlen, S. Tosatti, M. Textor, G. H hner,Langmuir2002,18, 3957–
3962.
[31] P. Gupta, A. Ulman, S. Fanfan, A. Korniakov, K. Loos,J. Am. Chem. Soc.
2005,127, 4–5.
[32] Z. Chen, M. G. Debije, T. Debaerdemaeker, P. Osswald, F. Wrthner, ChemPhysChem2004,5, 137–140.
[33] S. Leroy-Lhez, J. Baffreau, L. Perrin, E. Levillain, M. Allain, M.-J. Blesa, P.
Hudhomme,J. Org. Chem.2005,70, 6313–6320.
[34] F. Wrthner,Pure Appl. Chem.2006,78, 2341–2349.
[35] O. AlvÞque, P.-Y. Blanchard, T. Breton, M. Dias, C. Gautier, E. Levillain, Electrochem. Commun.2012,16, 6–9.
Manuscript received: November 29, 2016 Accepted Article published: December 20, 2016 Final Article published:&& &&, 0000
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
mide and hexanethiol by cyclic vol- tammetry and absorption spectroelec- trochemistry leads to a non-linearity of absorbance maxima as a function
change of the orientation of immobi- lized chromophores at low coverage, inducing a modification of the optical properties of the monolayer.
Coverage
