la104499bThickness, Surface Morphology, and Optical Properties of Porphyrin

ARTICLE

https://www.wendangku.net/doc/b112810145.html,/Langmuir Thickness,Surface Morphology,and Optical Properties of Porphyrin Multilayer Thin Films Assembled on Si(100)Using Copper(I)-Catalyzed Azide-Alkyne Cycloaddition

Peter K.B.Palomaki,Alexandra Krawicz,and Peter H.Dinolfo*

Department of Chemistry and Chemical Biology and The Baruch’60Center for Biochemical Solar Energy Research, Rensselaer Polytechnic Institute,1108th Street,Troy,New York12180,United States

b Supporting Information

ABSTRACT:We report the structure,optical properties and

surface morphology of Si(100)supported molecular multi-

layers resulting from a layer-by-layer(LbL)fabrication method

utilizing copper(I)-catalyzed azide-alkyne cycloaddition

(CuAAC),also known as“click”chemistry.Molecular based

multilayer?lms comprised of5,10,15,20-tetra(4-ethynyl-

phenyl)porphyrinzinc(II)(1)and either1,3,5-tris(azidomethyl)-

benzene(2)or4,40-diazido-2,20-stilbenedisulfonic acid disodium

salt(3)as a linker layer,displayed linear growth properties up to

19bilayers.With a high degree of linearity,specular X-ray re?ectivity(XRR)measurements yield an average thickness of1.87nm/bilayer for multilayers of1and2and2.41nm/bilayer for multilayers of1and3.Surface roughnesses as determined by XRR data?tting were found to increase with the number of layers and generally were around12%of the?lm thickness.Tapping mode AFM measurements con?rm the continuous nature of the thin?lms with roughness values slightly larger than those determined from XRR.Spectroscopic ellipsometry measurements utilizing a Cauchy model mirror the XRR data for multilayer growth but with a slightly higher thickness per bilayer.Modeling of the ellipsometric data over the full visible region using an oscillator model produces an absorption pro?le closely resembling that of a multilayer grown on silica https://www.wendangku.net/doc/b112810145.html,paring intramolecular distances from DFT modeling with experimental?lm thicknesses,the average molecular growth angles were estimated between40°and70°with respect to the substrate surface depending on the bonding con?guration.

’INTRODUCTION

The fabrication of thin?lms utilizing layer-by-layer(LbL) assembly techniques provides a straightforward and?exible method for the creation of nanostructured materials with excellent precision.The LbL approach allows for facile manip-ulation of physical,electronic,photophysical,and chemical properties of the thin?lms.Films with well-de?ned structures have successfully been made utilizing the LbL technique and their properties are generally dictated by the polymeric or molecular components used in the assembly process.This ?exibility in thin?lm design has signi?cant implications in the development of advanced materials for nonlinear optics, photovoltaics,memory devices,sensors,molecular electronics, etc.1-3

LbL based thin?lm structures are unique among self-assem-bled systems in that they provide control over the nanoscopic ordering of the materials with respect to a macroscopic orienta-tion,i.e.,the axis normal to substrate surface.The most common variety of LbL assemblies are those based on the sequential deposition of polyanions and polycations on a charged substrate surface.These materials,initially described by Decher,4-6rely on the electrostatic interactions between the opposite charges on the di?erent polyelectrolytes and have resulted in the construc-tion of a wide range of nanostructured thin?lm architectures.1-3,7

Numerous variations of the LbL assembly scheme have also been developed utilizing discrete molecular species as the layer components.Molecular based LbL methodologies,also known as molecular layer deposition(MLD),typically employ self-limiting reactions between symmetric multifunctional molecules and linkers that yield single monolayers deposited at each step. Sagiv was the?rst to describe molecular-based LbL assemblies using siloxane couplings between layers of alkane chains.8,9 Following these reports,Mallouk described the use of zirconium phosphonates as an e?cient coupling group for the creation of mixed inorganic-organic multilayers with e?cient growth over several multilayers.10-14Since these?rst reports of molecular based LbL assemblies,others have expanded on the siloxane15-21 and zirconium(or hafnium)phosphonates,22-26as well as using additional organic1,27-31and inorganic32-35coupling chemis-tries as a means of multilayer formation.Generally,these types of reaction sequences have been shown to produce well ordered, multilayer thin?lms that are amenable to various molecular

components.

Received:November11,2010

Revised:February14,2011

Published:March16,2011

In an e ?ort to provide even greater ?exibility in terms of the molecular species that can be incorporated into the thin ?lm materials,we recently developed a new LbL methodology that employs copper(I)-catalyzed azide -alkyne cycloaddition (CuAAC)reactivity,commonly referred to as “click ”chemistry.36Following its discovery by Sharpless 37and Meldal 38in 2001,CuAAC has gained signi ?cant popularity in recent years for a variety of synthetic and materials applications due to its simplicity,versatility,and high yield for the isomerically pure 1,4-triazole.39,40The use of CuAAC to covalently modify surfaces was ?rst described by Collman and Chidsey,who employed azido terminated alkane thiols and ethynlyl substituted redox probes to functionalize Au-(111)electrode surfaces.41-44These Au(111)electrodes coated with mixed azido-alkane self-assembled monolayers provide a highly tunable substrate to which a wide variety of ethynlyl substituted molecules could be attached.Subsequent studies have extended this surface modi ?cation technique to include oxide surfaces through the use of azido-alkylsiloxane SAMs.45-48

Our extension of this surface modi ?cation technique to molec-ular based LbL assemblies is outlined in Figure 1.36While CuAAC has been used to build triazole-linked polymer 49-51and dendrimer 52based multilayers,it has not,to our knowledge,been used to assemble molecular multilayers prior to our report.36The initial step involves the modi ?cation of an oxide substrate (ITO,silica glass,etc.)with an alkyl siloxane self-assembled monolayer (SAM)to provide the initial attachment point.Following this step,a multiethynyl functionalized mole-cule,such as porphyrin 1,is reacted with the surface using CuAAC chemistry to generate a molecular monolayer covalently attached through 1,4-triazole linkages.The resulting ethynyl terminated surface is then reacted with a multiazide molecule (termed a “linker ”from here on),such as 2or 3,to regenerate an azide-rich surface.These two self-limiting reactions are then repeated sequentially to generate the desired number of molec-ular bilayers.The choice of 1as our primary molecular building block was initially made based on its relative ease of synthesis,multiple ethynyl functionalities,and high extinction coe ?cient that allows for the straightforward monitoring of multilayer growth through absorption measurements.The highly tunable

optical and electrical properties of porphyrins have led them to be of signi ?cant importance in a range of materials chemistry applications.53The inclusion of porphyrin based molecular building blocks,such as 1,into multilayer ?lms could lead to a wide range of applications including arti ?cial photosynthetic processes,54-56semiconductor sensitization,57-59sensors,54-56and catalysts.63-65

Molecular multilayer thin ?lms grown on silica glass and ITO using 1and 2or 3showed highly linear absorbance growth trends through 10s of bilayers and alternating surface properties as probed by water contact angle.Additionally,polarized absorbance measurements indicated the presence of some molecular ordering of the porphyrins with respect to surface normal.Preliminary specular X-ray re ?ectivity (XRR)results suggest a growth of 2.47nm/bilayer for multilayers containing 1and 3on a glass substrate.36Herein we expand on these results by examining the thickness,surface morphology,and optical properties of thin ?lms made by a CuAAC LbL technique on Si(100)through the use of specular XRR,tapping mode atomic force microscopy (AFM),spectroscopic ellipsometry,and UV -visible specular re ?ectance spectroscopy.

’RESULTS

Multilayer Growth and UV -Visible Specular Reflectance Spectroscopy.Molecular multilayer growth on Si(100)sub-

strates was achieved in an analogous method as described previously for silica and indium tin oxide (ITO).36The method requires an azide-terminated SAM attached to the substrate surface as a starting point.Here,we use a 11-azidoundecyltri-methoxysilane hydrolyzed on the native oxide of Si(100).Multi-layer formation starts with the attachment of one layer of ethynyl porphyrin (1)via CuAAC chemistry (Figure 1,step 1).Under the reaction conditions described previously and herein,a dense monolayer of 1is covalently attached to the azido SAM,yielding a surface that is now terminated with ethynyl groups.This is then followed with a multiazido linker layer (2or 3)again using CuAAC chemistry (Figure 1,step 2)to create one molecular bilayer and regenerating an azido terminated surface.Steps 1and 2are repeated sequentially to create the desired number of molecular

bilayers.

Figure 1.Schematic representation of the CuAAC based LbL formation of molecular multilayer thin ?lms on Si(100)substrates and the molecular components used throughout this work.

The high molar absorptivity of tetraphenyl porphyrins allows us to track multilayer growth with UV-visible absorption spectroscopy.To monitor the absorbance pro?le of these?lms on nontransparent substrates(i.e.,silicon),we have employed specular re?ectance spectroscopy techniques at a near normal incidence angle.The spectra resulting from the Fresnel re?ectiv-ity of1through19bilayers of1and3is shown in the bottom panel of Figure2.Instead of the normal absorption pro?le for tetraphenylporphyrins,the re?ectivity spectrum of these samples show a?rst-derivative like line shape,with increased re?ectance on the high energy side of the absorption features resulting from the classic Kramers-Kronig e?ect.66The top panel of Figure2 shows the absorption pro?le(transmission mode)from a multi-layer grown with the same molecular components on silica glass.36 Despite the convolution due to the Kramers-Kronig e?ect on the re?ectance spectra,two Q-band features around560and605nm as well as the large Soret peak at440nm are clearly observed and they increase in intensity(decreasing re?ectivity)with an increase in the number of bilayers.The spectra consistently decrease in re?ectivity as each bilayer is added to the silicon substrate.Similar trends in re?ectivity changes were observed for multilayers formed with1 and2(see Supporting Information,Figure S1).

X-ray Specular Reflectivity.X-ray specular reflectivity(XRR) has been used previously to probe the height,roughness,and density of multilayer thin-film materials grown using LbL assembly techniques.16,23,32,67,68X-rays impinging on a thin-film will reflect off of the upper air-film and lower film-substrate inter-faces resulting in an interference pattern as the angle of incidence is changed.The dependence of the position of the interference patterns or Kiessig fringes69on thickness is given by the following Bragg equation modified for index of refraction23,67

nλ?2deθn2-θc2T1=2e1TIn eq1,d is the?lm thickness,θn is the position of the n th fringe,andθc is the critical angle for the?lm.The thickness of a thin?lm can be found using this equation by plottingθn2vs n2 where the slope will equalλ2/4d2.A more comprehensive analysis involves?tting of the XRR pro?les to a model that is based on the electron density pro?le of the thin?lm perpendi-cular to the substrate surface.This is done via eq2where R(q z)is the Fresnel re?ectivity of the thin?lm sample,R f(q z)is the ideal Fresnel re?ectivity of the Silicon substrate,F Si is the electron density of silicon,F(z)is the electron density pro?le of the thin-?lm and q z is the wave transfer vector(q z=(4π/λ)sinθ).70,71

Req zTR Feq zT?

1

F Si

Z

DFezT

D z e

-izq z d z

2

e2T

Figure3a shows the XRR curves resulting from multilayers of1 and3grown on a Si(100)substrate(open circles).The inter-ference patterns increase in frequency and shift to lower q z(and lowerθ)as the number of bilayers and thickness of the?lms increase.The attenuation of the fringe patterns at higher q z(or θ)is a result of the roughness of the?lms along with the limitations of the instrument.Fitting of the XRR curves was performed using the LEPTOS software suite from Bruker AXS where a layered sample model was constructed to generate a simulated re?ectivity curve(solid lines)based on eq2.The model consisted of four simulated layers:(1)bulk Si(100),(2)a thin?lm of SiO2to represent the native oxide surface,(3)an alkane self-assembled monolayer,and(4)the porphyrin based multilayer structure.We found it necessary to include an oxide layer in the model to achieve a good correlation between the raw data and?ts.The solid lines in Figure3a represent the best?ts to the X-ray re?ectivity data using this model.

Figure3b shows the electron density pro?les of the multilayer structures normalized to Si as a function of the distance(z)from the substrate surface.The electron density drops sharply from that of Si to a local minimum as a result of the lower SAM density and then rises to that of the multilayer?lm which remains constant until the?lm/air interface,before dropping to zero. Throughout the XRR data?tting process,it was observed that the total organic?lm thickness(SAM and multilayer)was a robust variable and remained nearly constant when either the SAM or multilayer thicknesses were altered.XRR analysis of the 11-azidoundecyltrimethoxysilane SAM on Si(100)yielded1.8nm for the thickness and0.85g/cm3for the density(see the Sup-porting Information).These values resulted in the best data?ts for the multilayer structures and were held constant for the remaining samples.The results of the XRR data?tting for all four sample sets are shown in Table S1.It is important to note that thicknesses of the multilayer?lms calculated manually using only the observed Kiessig fringe maxima via eq1were nearly identical to those obtained using the LEPTOS software.

Figure4shows the XRR derived multilayer thickness as a function of the number of bilayers for all samples examined.The thickness results obtained by XRR con?rm the expectation that multilayers grown with di?erent azido linker units(2and3) result in di?erent thicknesses.The average thickness of multi-layers grown using porphyrin1and linker2was found to be 1.87nm/bilayer,whereas multilayers grown with1and3had an average thickness of2.41nm/bilayer(R2=0.988).It is

important Figure2.Top:UV-visible absorption pro?le(transmission mode)for 10bilayers of1and3assembled on glass modi?ed with11-azidoun-decyltrimethoxysilane(data adapted from a previous report).36Bottom: UV-visible specular re?ectivity pro?les of1through19bilayers of1and 3on Si(100)modi?ed with11-azidoundecyltrimethoxysilane.

to note that for one sample set,multiple slides were grown in parallel using the same solutions and reaction times.Up to 7separate samples were made in parallel and still resulted in a linear thickness trend:R 2=0.989for 1and 2and R 2=0.988for 1and 3.This shows that CuAAC is a reproducible and robust method to build multilayer thin ?lms.

As can be seen in Figure 3B,the interfacial roughness gives rise to the gradual changes in the electron density pro ?le of the ?lms.The roughness of the multilayer ?lms generally increase with greater number of molecular bilayers.This is apparent in the gradual change from a near vertical electron density pro ?le at the ?lm/air interface (3bilayers,black)to a more gradual sloping pro ?le for the ?nal 19bilayer sample (blue).The roughness at the multilayer/air interface as determined by XRR ?tting is shown versus the number of bilayers in Figure 5for all samples used in this study.There is a clear trend of increasing surface roughness with an increase in the number of bilayers,most likely due to the dendritic nature of the porphyrin used in this study (vide infra).Further work is underway to investigate the structure and morphology of ?lms grown using di ?erent molecular components.Spectroscopic Ellipsometry.Spectroscopic ellipsometry was employed as a secondary method to compare thicknesses and to determine the optical constants of the multilayer films.Ellipso-metry has often been used in conjunction with XRR to determine thickness of nanometer scale thin films 72-80and LbL

assemblies 19,32,81-84on a variety of surfaces.Fewer examples employ spectroscopic ellipsometry to determine the optical constants,n and k ,of the thin film materials over a wide spectral range.81,85To determine the multilayer thicknesses,ellipso-metric incidence data,Δ(λ)and Ψ(λ),were collected at three angles of incidence (60°,65°,and 75°)and analyzed over the transparent region of the samples (where the extinction coeffi-cient k (λ)=0,from 673.9to 741.5nm,10data points)and fit using the Cauchy dispersion model that describes the refractive index (n )as a function of the wavelength (λ)via eq 3.86

n eλT?A -B λ2-C λ4

e3T

Spectroscopic ellipsometry was performed on all multilayer samples following analysis by XRR.The data were ?t using a three layer model consisting of (1)silicon substrate,(2)a thin layer of SiO 2,and (3)the multilayer ?lm including the 11-azidoundecyltrimethoxysiloxane SAM.The SiO 2thickness was determined from an unfunctionalized Si(100)substrate cleaned in the same manner as the multilayer functionalized samples.Thickness values for the SAM-multilayer ?lms are included in Table S1.Incorporation of a fourth layer to explicitly model the SAM did not signi ?cantly improve the quality of the ?ts.Typical values for A and B,determined by ?tting the Cauchy dispersion model (eq 3)were found to be 1.45-1.91and 0.01-0.14for multilayers of 1and 2and 1.54-1.90and 0.01-0.10

for

Figure 4.Multilayer ?lm thicknesses of all samples determined by ?tting of the XRR curves.Blue circles are multilayers constructed with porphyrin 1and mesitylene linker 2,and red triangles are multilayers built with porphyrin 1and stilbene linker 3.For both linkers two sample sets are di ?erentiated by ?lled and open data

symbols.

Figure 3.Panel A:Specular XRR of multilayers of 1and 3on Si(100)for the stilbene A sample set.Data are shown as open circles and the best ?t models as solid lines (black,red,green,brown,and blue for 3,7,11,15,and 19bilayers,respectively).The data and ?ts are o ?set for clarity.Panel B:Electron density pro ?les normalized to Si as a function of distance away from the substrate surface for 3,7,11,15,and 19bilayers of 1and 3.These pro ?les correspond to the best ?ts of the XRR data shown in panel

A.

Figure 5.XRR determined roughness as a function of the number of bilayers (all data sets included).Samples grown with azido linker 2are shown in blue circles and 3in red triangles.

multilayers of 1and 3,respectively.The C term was found to have little in ?uence on the data ?tting and was set to 0for all samples.

Figure 6shows a comparison of the total SAM and multilayer ?lm thickness for the mesitylene A and stilbene B samples sets as determined by XRR and spectroscopic ellipsometry.The ellipso-metric ?lm thicknesses mirror those determined by XRR,showing a linear growth trend over 19bilayers,but are consistently higher by approximately 15%.Similar trends have been observed in other thin ?lms where the ellipsometrically determined thicknesses are larger than thicknesses determined by XRR.19,72,76,79,80Notably,the variations in data with respect to the trend lines in Figure 6are mirrored for almost all samples,suggesting a high degree of precision for the individual techni-ques but also the presence of systematic error of one or both techniques.Several people have pointed out that an accurate knowledge of n is required for correct thickness determination by ellipsometry,especially for ultrathin ?lms.72,73,76,77,80Although we do not know the exact reason for the discrepancy between ellipsometrically and XRR determined thicknesses,we believe it may be in part due to inaccurate values of n for the multilayer ?lms.The spectroscopic analysis of these porphyrin based ?lms at wavelengths shorter than 673nm requires the use of a parame-trized oscillator model to account for the absorption pro ?le of the ?lms and its e ?ect on the complex index of refraction.The complex index of refraction (~n )is composed of the real refractive index (n )and the imaginary part (ik )according to the Kramers -Kronig relationship:85,86

~n eλT?n eλT-ik eλT

e4T

In this equation,k is extinction coe ?cient relating to the

absorption pro ?le of the multilayer ?lms.The thicker samples were modeled using four Lorentzian oscillators to reconstruct the overall absorption pro ?le of the porphyrin based multilayers (see Figure 2).A spectral deconvolution was performed on trans-mission absorption data for the analogous multilayers synthe-sized on silica glass (top spectra of Figure 2)and the resulting Lorentzian oscillator parameters were used as a starting point for spectroscopic ellipsometry data ?tting where the ?lm thickness was held constant at the values derived from the Cauchy layer modeling.Figure 7shows the ?ts to the Δ(λ)and Ψ(λ)data in the range of 410.0to 741.5nm (44data points)and the resulting n and k pro ?les.The k pro ?les closely match that of the absor-

ption pro ?le for the porphyrin multilayers and the n spectra matches that of the specular re ?ectivity seen in Figure 2.

Atomic Force Microscopy.Several of the multilayer samples were also characterized by tapping mode atomic force micro-scopy (TM-AFM).For each of the multilayer sample sets,three representative samples were chosen:thinnest,middle,and thickest.Figure 8shows 1μm ?1μm TM-AFM images collected for the Mesitylene A sample set.These are representative

of

Figure https://www.wendangku.net/doc/b112810145.html,parison of ?lm thicknesses as determined by XRR and ellipsometry for the mesitylene A and stilbene B sample sets.Notice the consistent di ?erence between the two methods.1.8nm has been added to the XRR determined thickness values to account for the 11-azidoundecyltrimethox-ysiloxane SAM in order to directly compare it to the ellipsometry

data.

Figure 7.Spectroscopic ellipsometry data (410.0to 741.5nm,44data points)for 16bilayers of 1and 2on Si(100)from the mesitylene A sample set.The top and middle panels show the Δand Ψdata (open symbols)and ?ts (solid lines)collected at an incident angle of 65,70,and 75°(red circles,green squares,and orange triangles,respec-tively).The bottom panel shows the resulting optical constants for the multilayer ?lm as determined by the oscillator model.The index of refraction (n )is shown as the dashed red line and the extinction coe ?-cient (k )as the blue line.

multilayer thin films for all samples shown in Table S1.Topo-graphy was analyzed focusing on the roughness trends that occur with increasing the number of deposited layers and the use of different linkers.Root-mean squared roughness (R q )of a cleaned Si(100)substrate was found to be on the order of 0.35nm.The values of R q for the images in Figure 8,for 4,10,and 19bilayers of 1and 2on Si(100)were 3.4,3.7,and 4.1nm,respectively.While these values are slightly higher than those determined from XRR data fitting,they are consistent across all samples (see values in Table S1).Larger area scans up to 10μm ?10μm reveal continuous multilayer films and similar roughness values.

XPS Analysis of Cu Content of Multilayer Films.X-ray photoelectron spectroscopy (XPS)was used to determine the concentration of copper catalyst remaining in the films after the multilayer fabrication process.Since no effort was made to remove the copper catalyst from the films other than simple solvent rinsing,we expected our films to contain measurable amounts of copper.Integration of the signals resulting from Cu 2p 3/2and Zn 2p 3/2allow us to quantify the amount of total copper with respect to zinc present in the samples (see the Supporting Information).87,88Films of 1and 2have a Cu:Zn ratio of 1.36:1,whereas films of 1and 3have a much lower ratio,0.16:1.The surprisingly high amount of copper in the films made with stilbene linker 1may be a result of the sulfonate groups present on linker 1.We suspect that the sulfonate moieties sequester the copper catalyst causing it to remain in the film at high concentrations.The lack of a sodium peak in the XPS supports the conclusion that the Cu ions exchange with the sodium ions originally present on linker 1.Azido linker 3does not contain any chemical groups with a strong affinity for copper which may be the reason films grown using linker 3have much lower levels of copper.Multilayer films made using CuAAC chemistry are likely to contain some amount of copper due to the ability of the basic nitrogen on the triazole ring to chelate copper.The ability of CuAAC modified surfaces to chelate copper ions has been suggested before.89We suspect that the triazole functionality causes the fixation of low levels of copper in films grown using linker 3while the sulfonates are the major cause of the high levels of copper in films grown using linker 2.It has been shown that EDTA is an effective Cu scavenger that can be used to remove copper from CuAAC functionalized films.89Further studies are underway to develop a method to remove the majority of copper from the multilayer structures during LbL formation.

’DISCUSSION

The thickness results obtained by XRR and spectroscopic ellipsometry highlight the linearity and reproducibility in growth of multilayers up to 19bilayers and across the 23independent samples (see Figures 3and 5).As determined by XRR,the thick-ness of multilayers grown with 1and 2was 1.87nm/bilayer,and with 1and 3was 2.41nm/bilayer.The thicknesses determined for multilayers of 1and 3was only slightly smaller than the value of 2.47nm/bilayer for the same multilayers grown on silica glass as determined in our previous study.36As described above,the XRR results also show a steady increase in surface roughness with increasing number of bilayers.The XRR determined roughness values average approximately 12%of the ?lm thickness for multi-layers grown using both linkers (2and 3).Similar thickness and roughness trends,deduced from a combination of XRR,AFM,and ellipsometry,have been observed for other porphyrin or phthalocyanine based multilayer ?lms grown in a LbL fashion.Polyelectrolyte multilayer ?lms consisting of nickel phthalocya-nine tetrasulfonate poly(diallydimethylammonium)chloride yielded thin ?lm structures with linear growth trends up to a 27bilayers and an increase XRR determined surfaces roughness that was approximately 5%of the ?lm thickness.AFM analysis of the same ?lms yielded roughness values that were approximately a factor of 2larger,the di ?erence being attributed to the di ?erent probing techniques.81AFM analysis of polyimide linked multi-porphyrin architectures grown on Si(100)surprising showed that thinner ?lms were rougher than thicker,but still averaged between 10and 20%of the overal multilayer thickness.30AFM determined surface morphology of uniform porphyrin based Zr-bisphosphonate multilayers ?lms grown on Si and Au surfaces displayed substrate dependent roughnesses that ranged between 1and 4nm rms for three to ?ve bilayers.55

Considering the size,number of azide groups,and di ?erences in the rotational ?exibility between the multiazido linkers (2and 3)in our ?lms,it is expected that the corresponding multilayer thicknesses should vary as well.In an e ?ort to better understand the molecular orientation within the multilayer thin ?lms we employed density functional theory to model the basic molecular building blocks to estimate the length of the repeat unit.The left side of Figure 9shows the DFT-B3LYP optimized structures of 1“clicked ”with four pendant azido linker molecules (1-24and 1-34).Due to the tetra-ethynyl functionality of 1and triazido functionality of 2,it is expected that several di ?erent bonding motifs could be present within these multilayer ?lms.The right side of Figure 9shows a schematic representation of the

two

Figure 8.Representative AFM images from the mesitylene A sample set obtained in tapping mode with a 1μm ?1μm scan area.The left,middle and right images correspond to 4,10,and 19bilayers of 1and 2grown on Si(100),with calculated root mean squared roughness values of 3.4,3.7,and 4.1nm,respectively.

di ?erent possible modes of growth of the porphyrin multilayers (trans and cis)representing the extremes of possible bonding con ?gurations.The trans con ?guration assumes that each por-phyrin utilizes only two of the four available ethynyl groups for multilayer growth,one on the top and one on the bottom,yielding the longest possible repeat unit.The cis con ?guration uses two ethynyl groups on each side of the porphyrin,resulting in the shortest possible repeat unit.Included with the optimized structures are relevant distances corresponding to the length of the repeat unit when considering trans (long dashed line)and cis (dotted line)bonding conformations within the multilayer structure.These distances are measured center-to-center be-tween the azido linker units,corresponding to one repeat unit.Assuming that these two binding motifs represent the extremes of longest and shortest possible repeat units,we can calculate a range of average molecular orientations for the individual components.These values are show in Table 1.Multilayers of 1and 2growing at 1.87nm/bilayer give a range of molecular growth angles from 40.6°(trans)to 65.7°(cis)whereas ?lms with 1and 3growing at 1.41nm/bilayer are slightly higher at 46.7°(trans)to 70.2°(cis)with respect to the substrate surface.The real growth angles are probably within these ranges as the bonding motifs within the multilayer ?lms are likely a mixture of both trans and cis.

The molecular building blocks used in this study were chosen to provide the best possible opportunity for linear LbL growth by providing multiple functional groups with CuAAC reactivity.The dendritic nature of 1and 2helps ensure that a maximal amount of terminal ethynyl and azide groups,respectively,are present at the surface of ?lm following each molecular layer addition.This dendritic functionality may also lead to the increased roughness of the ?lms as additional bilayers are added (see Figure 5).Attempts are currently underway in our laboratory to examine purely trans

disubstituted components in an e ?ort to create multilayer ?lms with lower roughness values.

’CONCLUSION

The structure,optical properties and surface morphology of two series of Si(100)supported molecular multilayers thin ?lms assembled in a LbL fashion utilizing copper(I)-catalyzed azide -alkyne cycloaddition (CuAAC)reactivity has been reported.Molecular multilayers constructed using Zn(II)5,10,15,20-tetra-(4-ethynylphenyl)porphyrin (1)and two azido-linker molecules (2and 3)displayed linear growth trends up to 19bilayers as determined by visible specular re ?ectance spectroscopy,XRR and spectroscopic ellipsometry.The di ?erence in average bilayer thickness of the multilayers corresponds with variations in the intramolecular distances of the two azido linkers and triazole linked porphyrins as determined by molecular modeling.Due to the presence of multiple ethynyl and azido functional groups on 1and 2,there are several potential bonding con ?gurations that may be present in the multilayer ?lm leading to a range of possible growth angles of the molecular https://www.wendangku.net/doc/b112810145.html,paring intramolecular distances from DFT modeling with experimental ?lm thicknesses,the average molecular growth angles were estimated between 40°and 70°with respect to the substrate surface depending on the speci ?c bonding con ?guration.Tap-ping mode AFM measurements con ?rm the continuous nature of the thin ?lms with roughness values slightly larger than those determined from XRR and generally increase with the increasing number of bilayers in the ?lm.Spectroscopic ellipsometry measurements utilizing a Cauchy model mirrors the XRR data for multilayer growth,but with a slightly higher thickness per bilayer.Modeling of the spectroscopic ellipsometry data over the full visible region using a Kramers -Kronig consistent oscillator model produces an absorption pro ?le that closely resembles that of a multilayer grown on silica glass.

The CuAAC based LbL assembly technique provides a straightforward and ?exible method for the construction of molecular multilayer thin ?lms.The results described herein illustrate the reproducibility of this method in growing ordered and uniform thin ?lms.This technique has the potential to provide a truly ?exible method for the construction of nanos-tructured thin ?lms with a variety of molecular components.While additional experiments are required to determine the scope and extent of this method,it is reasonable to assume that

the

Figure 9.Left:DFT-B3LYP optimized structures of the individual repeat units for multilayer growth.Included with the optimized structures are relevant distances corresponding to the length of the repeat unit when considering trans (long dashed line)and cis (dotted line)bonding conformations within the multilayer structure.Right:Schematic representation of the two di ?erent possible modes of growth of the porphyrin multilayers (trans and cis).

Table 1.Calculated Values of Molecular Repeat Units Length and Corresponding Growth Angles

repeat length (nm)

growth angle (deg)

molecular components bilayer thickness (nm)trans cis trans cis 1and 2 1.87 2.87 2.0540.765.81and 3

2.41

3.31

2.56

46.7

70.3

?exibility of the CuAAC reaction would allow the use of a wide variety of azido-or ethynyl-functionalized molecules to be used in the fabrication of other multilayer assemblies.We are currently examining other molecular building blocks that may lead to smoother?lms with a higher degree of order.

’EXPERIMENTAL SECTION

Materials.Solvents,ACS reagent grade or better,were purchased from Sigma Aldrich or Fisher Scientific and used as received.Dry toluene was purged with nitrogen and stored over molecular sieves before use. Sodium ascorbate(Aldrich)and4,40-diazido-2,20-stilbenedisulfonic acid disodium salt tetrahydrate(3)(Fluka)were used as received.11-Azidoundecyltrimethoxysilane,90Zn(II)5,10,15,20-tetra(4-ethynylphe-nyl)porphyrin(1),91tris-(benzyltriazolylmethyl)amine(TBTA),92and 1,3,5-tris(azidomethyl)benzene(2)93were synthesized according to literature methods.Si(100)wafers(p-type,1-100ohm/cm)were obtained from University Wafer.

Azido-SAM Formation on Silicon.Prior to use,silicon wafers were washed with solvents,last DI water and then cleaned in a piranha solution for at about30min(piranha=3:1v/v sulfuric acid to30% hydrogen peroxide,CAUTION!Reacts violently with organics!).The silicon wafers were then rinsed with copious amounts of DI water,dried under a stream of nitrogen,and placed in a Schlenk?ask at a pressure of 10-4Torr to remove residual water.The slides were then submerged in a solution of approximately1mM11-azidoundecyltrimethoxysilane in anhydrous toluene.The reaction vessel was then heated at60-70°C overnight.After cooling to room temperature,the slides were removed and sonicated in toluene for5min,after which they were washed with acetone,DCM,ethanol,and water,and then dried in a stream of nitrogen.Slides were then placed in an oven at75°C for at least4h.

Multilayer Fabrication.Ethynyl-porphyrin layers:A solution of DMSO,containing<2%water,consisting of1.3mM1,0.325mM CuSO4,0.358mM TBTA,and0.48mM sodium ascorbate was placed in contact with one side of a SAM functionalized silicon sample.After5min the slide was washed with acetone,dichloromethane,ethanol,and water. Azidolinker layer:DMSO solution,containing<8%water,was used as described above consisting of2.2mM of the selected azide functiona-lized linker molecule(2or3),4.4mM CuSO4,4.8mM TBTA,and 8.9mM sodium ascorbate.

Electronic Specular Reflectance Spectroscopy.Electronic reflectance spectra were taken on a Perkin-Elmer Lamda950spectro-meter using a near normal specular reflectance accessory(6°from surface normal).A background spectrum of an11-azidoundecyltrimeth-oxysilane SAM functionalized Si(100)substrate was subtracted from each spectrum.

X-ray Specular Reflectivity.X-ray specular reflectivity profiles were obtained on a Bruker D8Discover with a2-circle Theta/2Theta goniometer and a centric Eulerian cradle.The sealed tube copper X-ray source(Cu K Rλ=1.54?)was operated at40kV followed by a40mm Gobel collimating mirror and a4position rotary absorber for incident beam with attenuation of approximately1:1,1:10,1:100,and1:1000. Slits of1.0mm and0.2mm were used before and after the rotary absorber respectively.Samples were mounted on a5in.vacuum chuck.A knife edge was used for some samples but was found not to affect the resulting reflectivity spectrum beyond the critical angle.Step sizes and dwell times varied from sample to sample depending on the fringe patterns and desired signal-to-noise.

Simulated XRR curves were generated using the LEPTOS software from Bruker with a three-layer model consisting of a layer of SiO2,SAM (organic),and porphyrin multilayer(organic)on top of a Si(100) substrate.A genetic algorithm was used to create?t curves and repetitive ?ts were performed until the?t curve reached a constant value for each parameter.Good?tting results were obtained with the density of the SiO2set to2.20g/cm3and the SAM layer density at0.85g/cm3.67,94,95 In general the variables that were?t were constrained to values that were within reason,and the?tting was performed on the re?ectivity curve from approximately2θ=0.4°(just below the critical angle)to a region where the fringe patterns could no longer be discerned.For the?nal?t curves(reported)the following parameters were?t:SiO2roughness, SAM roughness,and multilayer thickness,roughness,and density.The Si roughness,SiO2thickness,and SAM thickness were assumed to be constant across a sample batch and were?xed to values deemed appropriate based on initial?tting results.In general the roughness of silicon was about0.3nm,the thickness of SiO2was1-2nm,and the value used for the SAM thickness was kept at1.8nm for all samples.The value of1.8nm used for the SAM thickness is roughly in agreement with reported values for similar length SAMs on silicon(without terminal azides),94as well as the value obtained from XRR?tting of a sample functionalized with a SAM only(see the Supporting Information).It is important to note that when the SAM thicknesses were manually adjusted(higher or lower)then the resulting multilayer thickness (after?tting)would adjust accordingly such that the total?lm thickness (SAMtmultilayer)remained nearly identical.Mass densities deter-mined by LEPTOS were given for a pure carbon?lm and converted to electron densities by the following eq567

F m?

F el A

N A Z

e5T

The reported densities were converted back to mass densities using eq696

F m?

F el∑A j c j

N A∑Z j c j

e6T

This method accounts for the fact that the electron density for our ?lms is not necessarily the same as carbon.Formulas for oligomers of porphyrin and the two di?erent linkers were used to calculate the reported mass density.

Spectroscopic Elipsometry.Spectroscopic ellipsometry mea-surements were performed on the same samples following analysis by XRR using a J.A.Woollam VASE Ellipsometer,model M-44.ΔandΨdata were acquired over the full spectroscopic range of the ellipsometer (410to741.5nm,44data points)at incident angles of65°,70°,and 75°.The data were fit using the WVASE software package from J.A. Woollam using a three layer model consisting of the Si(100)substrate layer,a thin SiO2layer,and the Cauchy layer that describes the multilayer film including the11-azidoundecyltrimethoxysiloxane SAM.The SiO2thickness for each sample set was determined from an unfunctionalized Si(100)substrate cleaned in the same manner as the multilayer functionalized samples.To determine the multilayer film thicknesses,the Cauchy dispersion model was employed over the nonabsorbing range of the porphyrin(where the extinction coefficient k(λ)=0,673.9to741.5nm,10data points)to determine n(λ)via eq3.Optical modeling of the real and complex dielectric functions of the thicker multilayer samples films were accomplished by generating several Lorentz oscillator to reproduce the absorption profile of the film.This method imposes Kramers-Kronig consistency(via eq4) between n(λ)and k(λ)across the wavelength range being analyzed.66 An initial guess of k(λ)for the multilayer films was generated from a point-by-point fit of data following a determination of the film thickness by the Cauchy model.During the optical modeling of the spectroscopic ellipsometry data,the film thickness was held constant and only n(λ)and k(λ)for the multilayer film were fitted.

Atomic Force Microscopy.Surface topology was analyzed and imaged using atomic force microscopy in tapping mode.Samples were pretreated by washing with MQ water and drying with a stream of nitrogen,and then keeping the sample under vacuum for a few hours.

Atomic force microscopy(AFM)images were recorded using a Digital Instruments Multimode IIIa operated in tapping mode.The cantilevers employed herein were ultrasharp silicon probes(MikroMasch)with a tip radius of～10nm and a resonant frequency of～160kHz.Roughness values were obtained using n-Surf image processing software.

X-ray Photoelectron Spectroscopy.XPS measurements were made on a Phi5000Versa Probe from Physical Electronics with an Al K R X-ray source and a hemispherical analyzer with the analytical chamber at a vacuum level e5?10-7Pa.Spectra were collected with an angle of incidence of45°.All spectra were referenced to the C1s peak at284.5eV.Survey scans were acquired at1eV data intervals with a band-pass energy of117.4eV,while the high resolution scans for Cu and Zn were taken at0.1eV intervals and a band-pass energy of23.5eV.All high resolution spectra were fit using the Peak Fit software suite with a GaussiantLorentzian peak shape after a linear background subtraction. Peak areas for the Cu2p3/2and Zn2p3/2peaks were used to determine the percent copper in the films with the following atomic sensitivity factors:5.321for Cu2p3/2and3.726for Zn2p3/2.87,88

Computational Details.Geometry optimizations for the por-phyin-azido-linker constructs were carried out using DFT as imple-mented in Gaussian03,revision E.01.97Becke’s three-parameter hybrid functional98-101with the LYP correlation functional102 (B3LYP)was used with the6-31ttg**basis set for S and O and 3-21g for C,N,H,and Zn.Both porphyrin models were optimized under the S4point group.

’ASSOCIATED CONTENT

b Supporting Information.Table summarizing the physi-cal properties of all multilayer samples used in this study,UV-visible specular re?ectance spectra for the multilayers of1and2 grown on Si(100),XRR scans for the stilbene B and mesitylene A and B samples sets,spectroscopi

c ellipsometry from the stilbene A sample set,XPS survey an

d high resolution scans of multilayers of1and2and1and3grown on Si(100),and th

e complete ref97. This material is available free o

f charge via the Internet at http:// https://www.wendangku.net/doc/b112810145.html,.

’AUTHOR INFORMATION

Corresponding Author

*E-mail:dinolp@https://www.wendangku.net/doc/b112810145.html,.

’ACKNOWLEDGMENT

We gratefully acknowledge Prof.Gow-Ching Wang and Joseph Palazzo for assistance in collecting and analyzing the XPS spectra of our multilayer?lms.P.K.B.P acknowledges a Wiseman Family Fellowship from Rensselaer Polytechnic In-stitute.This work was supported by Rensselaer Polytechnic Institute through new faculty start-up funds.

’REFERENCES

(1)George,S.M.;Yoon,B.;Dameron,A.A.Acc.Chem.Res.2009, 42,498–508.

(2)Xi,Z.;Jiacong,S.Adv.Mater.1999,11,1139–1143.

(3)Ariga,K.;Hill,J.P.;Ji,Q.Phys.Chem.Chem.Phys.2007,2319–2340.

(4)Decher,G.;Hong,J.D.Makromol.Chem.Macromol.Symp.1991, 46,321–327.

(5)Decher,G.;Hong,J.D.Ber.Bunsen-Ges.Phys.Chem.1991, 95,1430–1434.

(6)Decher,G.;Hong,J.D.;Schmitt,J.Thin Solid Films1992, 210,831–835.

(7)Decher,G.Science1997,277,1232–1237.

(8)Netzer,L.;Sagiv,J.J.Am.Chem.Soc.1983,105,674–676.

(9)Pomerantz,M.;Segm€u ller,A.;Netzer,L.;Sagiv,J.Thin Solid Films1985,132,153–162.

(10)Lee,H.;Kepley,L.J.;Hong,H.G.;Akhter,S.;Mallouk,T.E. J.Phys.Chem.1988,92,2597–2601.

(11)Lee,H.;Kepley,L.J.;Hong,H.G.;Mallouk,T.E.J.Am.Chem. Soc.1988,110,618–620.

(12)Cao,G.;Hong,H.G.;Mallouk,T.E.Acc.Chem.Res.2002, 25,420–427.

(13)Yang,H.C.;Aoki,K.;Hong,H.G.;Sackett,D.D.;Arendt, M.F.;Yau,S.L.;Bell,C.M.;Mallouk,T.E.J.Am.Chem.Soc.2002, 115,11855–11862.

(14)Lee,H.;Kepley,L.J.;Hong,H.G.;Akhter,S.;Mallouk,T.E. J.Phys.Chem.2002,92,2597–2601.

(15)Tillman,N.;Ulman,A.;Penner,https://www.wendangku.net/doc/b112810145.html,ngmuir1989,5,101–111.

(16)Evmenenko,G.;Boom,M.E.v.d.;Kmetko,J.;Dugan,S.W.; Marks,T.J.;Dutta,P.J.Chem.Phys.2001,115,6722–6727.

(17)Facchetti,A.;Abbotto,A.;Beverina,L.;van der Boom,M.E.; Dutta,P.;Evmenenko,G.;Pagani,G.A.;Marks,T.J.Chem.Mater.2003, 15,1064–1072.

(18)Li,D.;Ratner,M.A.;Marks,T.J.;Zhang,C.;Yang,J.;Wong,

G.K.J.Am.Chem.Soc.1990,112,7389–7390.

(19)Lin,W.;Lin,W.;Wong,G.K.;Marks,T.J.J.Am.Chem.Soc. 1996,118,8034–8042.

(20)van der Boom,M.E.;Richter,A.G.;Malinsky,J.E.;Lee,P.A.; Armstrong,N.R.;Dutta,P.;Marks,T.J.Chem.Mater.2000,13,15–17.

(21)van der Boom,M.E.;Zhu,P.;Evmenenko,G.;Malinsky,J.E.; Lin,W.;Dutta,P.;Marks,https://www.wendangku.net/doc/b112810145.html,ngmuir2002,18,3704–3707. (22)Ne?,G.A.;Helfrich,M.R.;Clifton,M.C.;Page,C.J.Chem. Mater.2000,12,2363–2371.

(23)Zeppenfeld,A.C.;Fiddler,S.L.;Ham,W.K.;Klopfenstein,

B.J.;Page,

C.J.J.Am.Chem.Soc.1994,116,9158–9165.

(24)Katz,H.E.;Scheller,G.;Putvinski,T.M.;Schilling,M.L.; Wilson,W.L.;Chidsey,C.E.D.Science1991,254,1485–1487.

(25)Splan,K.E.;Hupp,https://www.wendangku.net/doc/b112810145.html,ngmuir2004,20,10560–10566.

(26)Massari,A.M.;Gurney,R.W.;Wightman,M.D.;Huang,

C.H.K.;Nguyen,S.B.T.;Hupp,J.T.Polyhedron2003,22,3065–3072.

(27)Kohli,P.;Blanchard,https://www.wendangku.net/doc/b112810145.html,ngmuir2000,16,4655–4661.

(28)Li,Y.-h.;Wang,D.;Buriak,https://www.wendangku.net/doc/b112810145.html,ngmuir2009.

(29)Schmidt,I.;Jiao,J.;Thamyongkit,P.;Sharada,D.S.;Bocian,

D.F.;Lindsey,https://www.wendangku.net/doc/b112810145.html,.Chem.2006,71,3033–3050.

(30)Jiao,J.;Anariba,F.;Tiznado,H.;Schmidt,I.;Lindsey,J.S.; Zaera,F.;Bocian,D.F.J.Am.Chem.Soc.2006,128,6965–6974. (31)Li,D.;Buscher,C.T.;Swanson,B.I.Chem.Mater.1994, 6,803–810.

(32)Altman,M.;Shukla,A.D.;Zubkov,T.;Evmenenko,G.;Dutta, P.;van der Boom,M.E.J.Am.Chem.Soc.2006,128,7374–7382. (33)Altman,M.;Zenkina,O.;Evmenenko,G.;Dutta,P.;van der Boom,M.E.J.Am.Chem.Soc.2008,130,5040–5041.

(34)Motiei,L.;Altman,M.;Gupta,T.;Lupo,F.;Gulino,A.; Evmenenko,G.;Dutta,P.;van der Boom,M.E.J.Am.Chem.Soc. 2008,130,8913–8915.

(35)Altman,M.;Zenkina,O.V.;Ichiki,T.;Iron,M.A.;Evmenenko,

G.;Dutta,P.;van der Boom,M.E.Chem.Mater.2009,21,4676–4684.

(36)Palomaki,P.K.B.;Dinolfo,https://www.wendangku.net/doc/b112810145.html,ngmuir2010,26,9677–9685.

(37)Rostovtsev,V.V.;Green,L.G.;Fokin,V.V.;Sharpless,K.B. Angew.Chem.,Int.Ed.2002,41,2596–2599.

(38)Tornoe,C.W.;Christensen,C.;Meldal,https://www.wendangku.net/doc/b112810145.html,.Chem.2002, 67,3057–3064.

(39)Sumerlin,B.S.;Vogt,A.P.Macromolecules2010,43,1–13.

(40)Meldal,M.;Torn?e,C.W.Chem.Rev.2008,108,2952–3015.

(41)Collman,J.P.;Devaraj,N.K.;Chidsey,https://www.wendangku.net/doc/b112810145.html,ngmuir2004, 20,1051–1053.

(42)Collman,J.P.;Devaraj,N.K.;Eberspacher,T.P.A.;Chidsey,

C.E.

https://www.wendangku.net/doc/b112810145.html,ngmuir2006,22,2457–2464.

(43)Devaraj,N.K.;Decreau,R.A.;Ebina,W.;Collman,J.P.; Chidsey,C.E.D.J.Phys.Chem.B2006,110,15955–15962.

(44)Devaraj,N.K.;Dinolfo,P.H.;Chidsey,C.E.D.;Collman,J.P. J.Am.Chem.Soc.2006,128,1794–1795.

(45)Devaraj,N.K.;Miller,G.P.;Ebina,W.;Kakaradov,B.;Coll-man,J.P.;Kool,E.T.;Chidsey,C.E.D.J.Am.Chem.Soc.2005, 127,8600–8601.

(46)Prakash,S.;Long,T.M.;Selby,J.C.;Moore,J.S.;Shannon, M.A.Anal.Chem.2006,79,1661–1667.

(47)Lummerstorfer,T.;Ho?mann,H.J.Phys.Chem.B2004, 108,3963–3966.

(48)Ku,S.-Y.;Wong,K.-T.;Bard,A.J.J.Am.Chem.Soc.2008, 130,2392–2393.

(49)Such,G.K.;Quinn,J.F.;Quinn,A.;Tjipto,E.;Caruso,F.J.Am. Chem.Soc.2006,128,9318–9319.

(50)Bergbreiter, D. E.;Chance, B.S.Macromolecules2007, 40,5337–5343.

(51)Rydzek,G.;Thomann,J.-S.;Ben Ameur,N.;Jierry,L.;Me sini, P.;Ponche,A.;Contal,C.;El Haitami,A.E.;Voegel,J.-C.;Senger,B.; Schaaf,P.;Frisch,B.t.;Boulmedais,https://www.wendangku.net/doc/b112810145.html,ngmuir2009,26,2816–2824.

(52)Vestberg,R.;Malkoch,M.;Kade,M.;Wu,P.;Fokin,V.V.; Sharpless,K.B.;Drockenmuller,E.;Hawker,C.J.J.Polym.Sci.,Part A: Polym.Chem.2007,45,2835–2846.

(53)Kadish,K.;Smith,K.M.;Guilard,R.The Porphyrin Handbook; Academic Press:New York,1999.

(54)Wasielewski,M.R.Chem.Rev.1992,92,435–461.

(55)Abdelrazzaq,F.B.;Kwong,R.C.;Thompson,M.E.J.Am. Chem.Soc.2002,124,4796–4803.

(56)Yamada,H.;Imahori,H.;Nishimura,Y.;Yamazaki,I.;Ahn, T.K.;Kim,S.K.;Kim,D.;Fukuzumi,S.J.Am.Chem.Soc.2003, 125,9129–9139.

(57)Tachibana,Y.;Haque,S.A.;Mercer,I.P.;Durrant,J.R.;Klug,

D.R.J.Phys.Chem.B2000,104,1198–1205.

(58)Cherian,S.;Wamser,C.C.J.Phys.Chem.B2000,104 3624–3629.

(59)Odobel,F.;Blart,E.;Lagree,M.;Villieras,M.;Boujtita,H.;El Murr,N.;Caramori,S.;Alberto Bignozzi,C.J.Mater.Chem.2003, 13,502–510.

(60)Lee,W.W.-S.;Wong,K.-Y.;Li,X.-M.;Leung,Y.-B.;Chan,C.-S.;Chan,K.S.J.Mater.Chem.1993,3,1031–1035.

(61)Rakow,N.A.;Suslick,K.S.Nature2000,406,710–713.

(62)Smith,V.C.;Batty,S.V.;Richardson,T.;Foster,K.A.; Johnstone,R.A.W.;Sobral,A.J.F.N.;Gonsalves,A.M.D.A.R.Thin Solid Films1996,284-285,911–914.

(63)Gonsalves,A.M.d.R.;Pereira,M.M.J.Mol.Catal.A:Chem. 1996,113,209–221.

(64)Merlau,M.L.;del Pilar Mejia,M.;Nguyen,S.T.;Hupp,J.T. Angew.Chem.,Int.Ed.2001,40,4239–4242.

(65)Collman,J.P.;Devaraj,N.K.;Decreau,R.A.;Yang,Y.;Yan, Y.L.;Ebina,W.;Eberspacher,T.A.;Chidsey,C.E.D.Science2007, 315,1565–1568.

(66)Stern,F.;Frederick,S.;David,T.Solid State Physics;Academic Press:New York,1963;Vol.15,pp299-408.

(67)Holy,V.;Pietsch,U.;Baumbach,T.High-Resolution X-Ray Scattering from Thin Films and Multilayers;Springer:New York,1999; Vol.149.

(68)Chason,E.;Mayer,T.M.Crit.Rev.Solid State Mater.Sci.1997, 22,1–67.

(69)Kiessig,H.Ann.Phys.1931,10,769–788.

(70)Parratt,L.G.Phys.Rev.1954,95,359.

(71)Als-Nielsen,J.Physica A(Amsterdam)1986,140A,376–89.

(72)Wasserman,S.R.;Whitesides,G.M.;Tidswell,I.M.;Ocko,

B.M.;Pershan,P.S.;Axe,J.D.J.Am.Chem.Soc.1989,111,5852–5861.

(73)Geer,R.E.;Stenger,D.A.;Chen,M.S.;Calvert,J.M.; Shashidhar,R.;Jeong,Y.H.;Pershan,https://www.wendangku.net/doc/b112810145.html,ngmuir1994, 10,1171–1176.

(74)Geo?rey,D.H.;David,L.A.J.Polym.Sci.,Part B:Polym.Phys. 1998,36,1247–1260.

(75)Toney,M.F.;Mate,C.M.;Pocker,D.IEEE Trans.Magnet. 1998,34,1774–1776.

(76)See,Y.-K.;Cha,J.;Chang,T.;Ree,https://www.wendangku.net/doc/b112810145.html,ngmuir2000,16 2351–2355.

(77)Richter,A.;Guico,R.;Wang,J.Rev.Sci.Instrum.2001,72,3004.

(78)Rozlosnik,N.;Gerstenberg,M.C.;Larsen,https://www.wendangku.net/doc/b112810145.html,ngmuir 2003,19,1182–1188.

(79)Kohli,S.;Rithner,C.D.;Dorhout,P.K.;Dummer,A.M.; Menoni,C.S.Rev.Sci.Instrum.2005,76,023906.

(80)Ishizaki,T.;Saito,N.;SunHyung,L.;Ishida,K.;Takai,O. Langmuir2006,22,9962–9966.

(81)Li,D.;L€u tt,M.;Fitzsimmons,M.R.;Synowicki,R.;Hawley, M.E.;Brown,G.W.J.Am.Chem.Soc.1998,120,8797–8804. (82)Kaschak,D.M.;Lean,J.T.;Waraksa,C.C.;Saupe,G.B.; Usami,H.;Mallouk,T.E.J.Am.Chem.Soc.1999,121,3435–3445. (83)Vuillaume,P.Y.;Glinel,K.;Jonas,A.M.;Laschewsky,A.Chem. Mater.2003,15,3625–3631.

(84)Erokhina,S.;Berzina,T.;Cristofolini,L.;Erokhin,V.;Folli,C.; Konovalov,O.;Marino,I.-G.;Fontana,https://www.wendangku.net/doc/b112810145.html,ngmuir2008, 24,12093–12096.

(85)Panteli c,N.;Wansapura,C.M.;Heineman,W.R.;Seliskar,C.J. J.Phys.Chem.B2005,109,13971–13979.

(86)Azzam,R.M.A.;Bashara,N.M.Ellipsometry and Polarized Light;North-Holland Publishing Co.:New York,1977.

(87)Meda,L.;Ranghino,G.;Moretti,G.;Cerofolini,G.F.Surf. Interface Anal.2002,33,516–521.

(88)Moulder,J.F.;Stickle,W.F.;Sobol,P.E.;Bomben,K.D. Handbook of X-Ray Photoelectron Spectroscopy;Physical Electronics: Japan,1995.

(89)Ciampi,S.;B€o cking,T.;Kilian,K.A.;James,M.;Harper,J.B.; Gooding,https://www.wendangku.net/doc/b112810145.html,ngmuir2007,23,9320–9329.

(90)Fu,Y.-S.;Yu,S.J.Angew.Chem.,Int.Ed.2001,40,437–440.

(91)Onitsuka,K.;Kitajima,H.;Fujimoto,M.;Iuchi,A.;Takei,F.; Takahashi,https://www.wendangku.net/doc/b112810145.html,mun.2002,2576–2577.

(92)Chan,T.R.;Hilgraf,R.;Sharpless,K.B.;Fokin,https://www.wendangku.net/doc/b112810145.html,.Lett. 2004,6,2853–2855.

(93)Song,Y.;Kohlmeir,E.K.;Meade,T.J.J.Am.Chem.Soc.2008, 130,6662–6663.

(94)Tidswell,I.M.;Ocko,B.M.;Pershan,P.S.;Wasserman,S.R.; Whitesides,G.M.;Axe,J.D.Phys.Rev.B1990,41,1111.

(95)del Campo,V.;Cisternas,E.;Taub,H.;Vergara,I.;Corrales,T.; Soza,P.;Colkmann,U.G.;Bai,M.;Wang,S.;Hansen,F.Y.;Mo,H.; Ehrlich,https://www.wendangku.net/doc/b112810145.html,ngmuir2009,25,12962–12967.

(96)van der Lee,A.;Hamon,H.;Holl,Y.;Grohens,https://www.wendangku.net/doc/b112810145.html,ngmuir 2001,17,7664–7669.

(97)Frisch,M.J.;et al.Gaussian03,revision E.01;Gaussian,Inc.: Wallingford,CT,2004.See the Supporting Information for the com-plete reference.

(98)Becke,A.D.Phys.Rev.A1988,38,3098–3100.

(99)Becke,A.D.J.Chem.Phys.1993,98,1372–1377.

(100)Becke,A.D.J.Chem.Phys.1993,98,5648–5652. (101)Stephens,P.J.;Devlin,F.J.;Chabalowski,C.F.;Frisch,M.J. J.Phys.Chem.1994,98,11623–11627.

(102)Lee,C.;Yang,W.;Parr,R.G.Phys.Rev.B1988,37,785.