This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 192.167.164.69

This content was downloaded on 01/12/2014 at 15:10

Please note that terms and conditions apply.

Graphene oxide intercalation into self-assembled porphyrin J-aggregates

View the table of contents for this issue, or go to the journal homepage for more

2014 Mater. Res. Express 1 045038

(http://iopscience.iop.org/2053-1591/1/4/045038)

Home Search Collections Journals About Contact us My IOPscience

Graphene oxide intercalation into self-assembledporphyrin J-aggregates

Harsha Vardhan R Maraka1, Rusul M Al-Shammari1, Nebras Al-Attar1,Sergio G Lopez2, Tia E Keyes2 and James H Rice11NanoPhotonics Research Group, University College Dublin, Dublin, Ireland2 School of Chemical Science, Dublin City University, Dublin, IrelandE-mail: james.rice@ucd.ie

Received 21 August 2014, revised 15 October 2014Accepted for publication 24 October 2014Published 24 November 2014

Materials Research Express 1 (2014) 045038

doi:10.1088/2053-1591/1/4/045038

AbstractStudies are undertaken to examine graphene oxide intercalation into self-assembled J-aggregate porphyrin structures. Fluorescence lifetime and fluores-cence anisotropy imaging were applied along with scanning electron microscopyto study the structure and optical properties of a graphene oxide/TMPyP hybridcomposite material. It was seen that the presence of graphene oxide alters themacroscale and nanoscale self-assembled structures of TMPyP in addition gra-phene oxide also alters the optical activity reducing the emission intensity andexciton recombination lifetime. Evidence exists to support a model where pla-ner-symmetric graphene oxide and TMPyP co-operate in the formation of self-assembled macro and nanostructures forming a composite with strong grapheneoxide/TMPyP interaction.

Keywords: graphene oxide, J-aggragate, fluorescence

Introduction

A considerable amount of research has been undertaken to advance the understanding of opticalprocesses in low dimensional nanomaterials such as quantum dots, nanodisks, fullerenes andcarbon nanotubes [1–10]. These materials exhibit a range of interesting physical properties, thestudy of which potentially advances fundamental science and also areas of applied researchsuch as photonics and sensor development where nanomaterials show great promise. This hasbeen reinforced by advances in optical spectroscopy and imaging methods that has supportedthe study of nanosystems [11–14]. Graphene is one type of nanomaterial that has recentlygained considerable interest. Graphene is a two-dimensional nanomaterial consisting of sp2-hybridized carbon atoms forming a one-atom thick honeycomb lattice. Graphene oxide sheetsare a form of graphene that is solution processable prepared by oxidizing graphite [15, 16]. A

Materials Research Express 1 (2014) 0450382053-1591/14/045038+14$33.00 © 2014 IOP Publishing Ltd

second type of nanomaterial that possess interesting properties are self-assembled nano andmicro-structured aggregates of π-conjugated molecules. These materials have attracted interestdue to their structural similarities to photosynthetic antenna systems of green bacteria [17–20].In the chlorosomes of green bacteria, bacteriochlorophylls (a porphyrin derivative) assemble toform nanorods, and these chromosomal rods are the most efficient harvesters of light known[21]. Nano and micro-structured porphyrin aggregates are readily formed by self-assembly. Theprocess of self-assembly is used as a facile method to create a range of structures [22–26]. Arange of porphyrins have been found to self-assemble into nanorods-like structures; for exampleReddy et al [27] demonstrated that H2TPP in acidic environments forms well-defined nanorodsassigned to occur due to J-aggregation. In porphyrins the formation of J aggregates are reportedto result in a red-shift in absorption bands. Kasha’s analysis of energy relaxation from excitedstates of molecular aggregates predicts that a stacked face-to-face H-aggregate leads to aspectral blue shift relative to the monomer excited-state level, whereas the a tilted ‘deck ofcards’ (J-aggregate) aggregation leads to a red shift in the Soret band [23, 30–32]. Khairutdinovet al noted that the extent of this shift been proportional to the degree of aggregation, theinterplanar separation distance, mutual orientation of the monomers in the aggregate, electronictransition probabilities, and the specific sensitivities of the chromophore [30].

Graphene materials such as GO are predicted to enable a range of devices via theirelectronic properties [28, 29]. Combining GO with organic aggregates such as porphyrin J-aggregates have the potential to create organic composite materials with optoelectronicproperties applicable for devices such as solar cells or sensors. For example, studies of blends offunctionalized graphene with TiO2 when applied to photovoltaics resulted in advancement inpower conversion efficiency of about five times [28]. In order to exploit the potential ofcombining the electrical and mechanical properties of GO with the optical properties ofporphyrins to create materials for device based applications, a better understanding of how GOinteracts with J-aggregate porphyrins is required in the solid state. Studies investigating thisquestion have been performed. Wojcik et al [29], studied the optical properties of 5,10,15,20-tetrakis(1-methyl-4-pyridinio)porphyrin (TMPyP) with chemically reduced graphene oxide(RGO). The authors reported the formation of a ground-state TMPyP/RGO complex in solutionvia absorption spectroscopy, monitoring of the porphyrins B band showed a red-shift withincreasing concentration of RGO along with a reducing of the fluorescence lifetime of theporphyrin from 5 to 1 ns demonstrating potentially excited-state interaction between singletexcited porphyrin and RGO. Other authors [33–39] have reported that GO interacts withporphyrins creating a complex which exhibits a reduced fluorescence quantum yield, which wasassigned to charge transfer from the porphyrin S1 to GO and/or molecular flatting of theprophyrin following interaction with the GO.

Here we investigate a solid state organic thin film composite of J-aggregated TMPyP andGO at a fixed concentration. We demonstrate graphene oxide intercalation into self-assembledJ-aggregate porphyrin structures. Graphene oxide–porphyrin (TMPyP) interaction is observedeven when conditions to form TMPyP J-aggregates are present. Fluorescence lifetime andfluorescence anisotropy imaging is applied in combination with electron microscopy to assessthe structure and optical properties of the resulting porphyrin/GO hybrid composite. We findevidence for strong porphyrin–GO interactions with the resulting self-assembled macro andnanostructures relatively unaffected by the uptake of GO. TMPyP J-aggregates are similar toTMPyP J-aggregates +GO hybrid composite in both optical properties and nano and macro

2

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

structured self-assembled formations. We note that the presence of GO reduces the lifetime ofthe porphyrin emission significantly in line with literature studies for GO-fluorphores.

Experimental

The porphyrins meso-tetra (n-methyl-4-pyridyl) porphyrine tetra chloride (TMPyP) (purchasedfrom Frontier scientific) were prepared as thin films with and without graphene oxide (GO)(purchased from Sigma Aldrich). Two sets of samples were prepared. The first prepared bydeposition of porphyrins directly onto a glass slide under neutral conditions, these samples arehereby referred as sample set n i.e. n-TMPyP. The second set of samples were prepared bydepositing the porphyrins onto a slide in the presence of acid (pH= 0.5), a sample set herebyreferred to as set a, i.e., a-TMPyP. In all cases the samples were dried in air. The structure ofGO was determined by SEM, the average length (long axis). The samples were prepared using1000 μl of porphyrin solution (0.2 × 10−3 molar concentration stock solution) with 50 μl of acid(concentrated HCl) with 50 μl of GO suspension, (used as received). The mixed solution wassonicated for 15min and then left standing 30min then 20 μl drop deposited onto a substrate.

Optical UV–vis absorption spectra were recorded using a Jasco UV/Vis instrument.Fluorescence microscopy studies were performed using a Zeiss inverted microscope systemcoupled with a 532 nm excitation wavelength, CCD camera and 650 nm band pass filter. Theresulting images were fitted with scale bars that indicate length. A Picoquant Microtime 200fluorescence lifetime imaging microscopy (FLIM) system was used for lifetime imaging. Boththe fluorescence anisotropy and the fluorescence lifetime images were recorded on a Microtime200 microscope system (PicoQuant GmbH, Berlin, Germany). The measurements wereperformed using the time-tagged time-resolved methodology. The excitation source was alinearly-polarized 532 nm PicoTA pulsed diode laser (PicoQuant GmbH, Berlin, Germany)working at a 10MHz repetition rate. The excitation light was directed towards a z532/635rpcdichroic mirror which focused the laser beam on a water immersion objective (NA 1.2UPlanSApo 60 × 1.2 water/CC1.48, Olympus). The fluorescence arising from the sample wascollected by the foregoing objective and passed through an HQ550lp AHF/Chroma filter, theaforementioned dichroic mirror, and a 50 μm pinhole. In the case of the fluorescence lifetimeimaging microscopy (FLIM) measurements, the fluorescence was then directly focused on aMPD SPAD detector (PicoQuant GmbH, Berlin, Germany), whereas in the case of thefluorescence anisotropy measurements, the parallel and perpendicular fluorescence componentswere separated using a polarization beamsplitter (PicoQuant GmbH, Berlin, Germany) andfocused on separate MPD SPAD detector (PicoQuant GmbH, Berlin, Germany). The analysis ofthe data was carried out using the SymPhotime software (PicoQuant GmbH, Berlin, Germany).This software calculates the steady-state anisotropy value r( ) of each pixel as

=−

− + −⊥

⊥[ ] [ ]r

I I

L I L I2 3 1 3(1)II

II1 2

= −⊥ ⊥ ⊥I I B (2)

= −[ ]I G I B (3)II II II

where III is the parallel fluorescence intensity, ⊥I is the perpendicular fluorescence intensity, BII

background signal in the parallel detection channel, ⊥B background signal in the perpendicular

3

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

detection channel, G is a correction factor equal to the ratio between the perpendicular andparallel fluorescence intensities, and L1 and L2 are correction factors related to thecharacteristics of the objective. In this work the values of L1 and L2 were assumed to bezero. The G value =G( 0.59) was obtained using a room-temperature solution of Rhodamine6G in water. The anisotropy [40] analysis described above (equations (1)–(3)) is based on thework of Schaffer et al [41]

Results

Electronic ground state porphyrin-porphyrin interactions with and without GO were probedusing optical absorption spectroscopy. The resulting optical absorption spectra are shown infigure 1 and are outlined in table 1. Optical absorption spectrum of a film of n-TMPyP filmshows clearly defined bands at 427, 470, 532, 602 and 658 nm. In comparison, the absorptionspectrum for the n-TMPyP +GO film shows bands at 428, 529, 593 and 646 nm. The opticalabsorption spectra of n-TMPyP ±GO shows not only differences in band positions but also inrelative band intensities. The presence of GO alters the position of the Q bands, blue shiftingthem from 3 to 12 nm. In addition, the relative intensity of the B band has increased while the Qband intensities decrease when GO is added. In n-TMPyP, relatively strong TMPyP–TMPyPinteractions occur, similarly to H/J aggregates which possess red-shifted absorption bands [27].When GO is added, the degree or magnitude of the red shift is reduced. One explanation for thismay be that GO is reducing TMPyP-TMPyP interactions. This may occur potentially by GOintercalating between the TMPyP–TMPyP formations which are held by weakly binding forcese.g. π–π or London force interactions. TMPyP possesses n-methylated pyridine substituentswhich are about perpendicular to the plane of the porphyrin [29]. While it is believed that GO

Figure 1. (a) Absorption spectra for all four TMPyP/GO complexes together, (b)normalised in order to show relative band intensities. Insert shows the absorption andemission spectra for the a-TMPyP/GO complex.

4

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

sheets when deposited exist as multi-layers [36]. The porphyrin can intercalate between the GOmulti-layers in solution enabling a molecular interaction between the two molecules toeffectively occur. The samples prepared here consist of materials first assembled together insolution before a thin film is prepared potentially providing an environment where GOintercalates with TMPyP (as well as TMPyP interacting with other TMPyP molecules to form aJ-aggregate) resulting in an effectively mixed TMPyP/GO film being prepared.

When acid is used in the preparation of a TMPyP film, the acidic environment is reportedto enhance the potential for porphyrin–porphyrin interaction, e.g. aggregation [27, 30]. Opticalabsorption spectra of a TMPyP film prepared at pH= 0.5, (referred to as a-TMPyP) wasrecorded and is shown in figure 1. The absorption spectra for a-TMPyP film possess a distinctspectrum with a broad B band (with a second peak at about 470 nm) and relatively intense Qbands with bands at about 375, 428, 470, 565, 615 and 668 nm. These band positions are redshifted compared to the (neutral) n-TMPyP film (as well as TMPyP in solution) and possessband intensities (i.e. B versus Q band intensities, together with the red-shift) consistent with J-aggregate formation [23, 27]. When GO is added to a-TMPyP, the absorption spectrum showsbands at 430, 528, 595 and 651 nm, very similar to both band position and relative (B versus Q)band intensities for n-TMPyP+GO, i.e. the presence of GO blue shifts the Q band positionsrelative to a-TMPyP alone. The optical absorption spectra of a-TMPyP ±GO shows not onlydifferences in band positions but also in relative band intensities, with the presence of GOincreasing the relative intensity of the B band and reducing the Q band intensities, mirroring thespectral features for n-TMPyP +GO.

Inspection of the B and Q bands from the absorption spectra for all four complexes showthat on the basis of the degree of spectral changes that a-TMPyP> n-TMPyP > a-TMPyP +GO≈ n-TMPyP +GO. The optical absorption spectral features are in line with the assertion thatGO reduces the TMPyP–TMPyP interaction potential, the mechanism for this potentiallyarising from GO intercalating between the TMPyP–TMPyP interacting layers, reducing theporphyrin–porphyrin interaction. It is seen that even when the TMPyP–TMPyP interactions areamplified via the formation of J-aggregation the presence of GO reduces significantly theprophyrin–porphyrin J-aggregation formations which are believed to be a stacking ofporphyrins one on top of the other [27, 30] such that about the same spectral features for a-TMPyP +GO occur as for n-TMPyP +GO.

Table 1. Band positions from optical absorption spectra. * denotes bands assigned toaggregation species. ** no band position was observed due to high background. ***optical absorption spectrum for GO showed no bands in the window 400 to 700 nm(noting bands are present < 400 nm which are outside of our window of interest), thusthis material did not contribute to peak assignments in this region.

Absorption band positions (nm)

Soret Band Q1 Q2 Q3 Q4

Porphyrin pH B band Qy(0−1) Qy(0−0) Qx(0−1) Qx(0−0)

TMPyP 7 427 532 ** 602 658TMPyP+GO 7 428 529 563 593 646TMPyP 0.5 428 522 565 615 668TMPyP 0.5 430 528 567 595 651GO*** 7 <300

5

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

Micro and macro-scale studies of the structures in the films for the porphyrin/grapheneoxide complexes were undertaken using optical microscopy. Fluorescence imaging (withoptical transmission contrast imaging) was applied to all samples; the resulting images areshown in figure 2. The fluorescence of TMPyP (in line with other free-base porphyrin moieties)consist of two emission peaks (figure 1) [27]. These peaks are assigned to Q(0,0) and Q(0,1)(mirror images of their respective absorption bands i.e. Q(0,0) and Q(1,0)). TMPyP, in commonwith other porphyrins, possesses a relatively large energy gap between the singlet states S1 (Qband) and S2 (B band). The B band corresponds to a strongly allowed transition, resulting in afast S2/S0 radiative transition. TMPyP fluoresces from the S1 and S2 states giving rise in turn totwo bands; this enables fluorescence imaging to occur.

The films show well defined structure resulting from self-assembly based processes, thesestructures possessing features on the tens of micron length scale and upwards. Figure 2 showsmacroscale structures via optical transmission and fluorescence imaging. Imaging of n-TMPyPshows well-defined self-assembled macroscale structures with rod- or needle-like formationsclearly visible in the image. Fluorescence imaging showing the same features further confirmingn-TMPyP as a film that self-assembles into macroscale ordered structures. For a-TMPyP whichpossesses a stronger J-aggregate formation there is a clear difference in the resulting self-assembled structures with macroscale square-like structures resulting. When GO is introducedfor both n and a-TMPyP materials there is a resulting destabilisation of the self-assembledstructures. In n and a-TMPyP, the structures are less well-defined, more amorphous andclustered.

Figure 2. Optical transmission (left hand side) and fluorescence images (right hand sidein red) from all four porphyrin samples.

6

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

Fluorescence lifetime imaging (FLIM) was applied and the resulting images are shown infigure 3. FLIM images for n-TMPyP show a large rod-like shape running across the image. Fora-TMPyP, FLIM images show the presence of square shapes. The FLIM image for n-TMPyP +GO shows rod-like shapes which are more broken compared to n-TMPyP. The FLIMimage for a-TMPyP +GO shows square-like shapes which are less defined compared to a-TMPyP alone. The FLIM images show features that are consistent with the features seen in theoptical transmission and florescence images shown in figure 2.

In order to compared the data from the FLIM images, plots of fluorescence intensity andlifetime versus its occurrence are extracted and plotted. Figure 4(a) shows a plot of fluorescenceintensity. The fluorescence intensity is highest for n-TMPyP (with an intensity of 1.6 × 107

+/−1 × 107 au), with the fluorescence reduced for a-TMPyP (to 6.2 × 106 +/−1.6 × 106 au). WhenGO is introduced additional reduction in fluorescence intensity results, e.g. n-TMPyP+GO(3 × 106 +/−2 × 106) and for a-TMPyP+GO (3.65 × 106 +/−2 × 106). Interestingly, a relativelysmall reduction in the fluorescence intensity for a-TMPyP occurs when GO is added (about atwofold drop) compared to n-TMPyP where about a fivefold drop is seen.

Figure 4(a) shows (as an insert) the average decay lifetime. The average lifetime for bothn- and a-TMPyP is reduced when GO is present. It is noted that inspection of the FLIM image(figure 3) shows a degree of uniformity in regard to the lifetimes across the sample, i.e., aconsistent green colouring, indicating a significant degree of homogeneity in the material.However, there is the presence of additional colours with the green in the FLIM imageindicating that a more complex situation is present. A plot of lifetime versus occurrence is

Figure 3. FLIM images from all four porphyrin samples.

7

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

shown in figure 4(b). This plot shows more clearly the different lifetimes present in the imageand how often they occur, which in turn enables better understanding of the dynamics occurringin the sample. Figure 4(b) shows that for n-TMPyP there is a distribution of lifetimes. Thisdistribution is approximately Gaussian with a decay time value of 3.4 ns at the centre of theGaussian profile. When GO is introduced (i.e., n-TMPyP +GO) the lifetime is reduced to 1.1 ns(as shown by a Gaussian centred at 1.1 ns). It is noted that there is a second smaller Gaussianpresent with a lifetime centred at 3.4 ns, which corresponds to the original lifetime of theporphyrins in the absence of GO. It is noted that the occurrence of lifetimes at 1.1 ns occurs to afar higher extent than for lifetime at 3.4 ns for TMPyP +GO. From this we infer that GO iseffective in altering the majority of porphyrin molecules’ fluorescence decay lifetimes. For a-TMPyP the decay lifetime is 1.3 ns (from the centre of a Gaussian fit) significantly reducedcompared to n-TMPyP. Noting that some residual lifetime contributions at around 3.4 ns can beseen which can in turn be assigned to either H aggregates or TMPyP molecules that arefractured and are in weaker aggregate formations, i.e., less close-packed (e.g. as in n-TMPyP).For a-TMPyP with GO the lifetime is shortened to 0.6 ns, i.e., on the order of the instrumentresponse time. The presence of GO clearly reduces the lifetime for the resulting porphyrincomplex with the overwhelming majority of the porphyrins exhibiting a reduced decay lifetimewhen GO is present.

Fluorescence anisotropy imaging was performed. The expression of polarization is directlyrelated to the molecular ordering and the observation of strong polarization effects indicates thatthe molecules are ordered [40, 41]. This enables a method to probe the internal structure of thecomplex, i.e., how the emitting molecule (e.g. TMPyP) is arranged. Fluorescence anisotropyimaging of n-TMPyP (see figure 5) shows a significant degree of ordering with the porphyrinaligned in rod-like sheets. The measured anisotropy depends on the direction that the rods arepointing (e.g. blue or green coloured features in the image). Looking at a single rod-like feature

Figure 4. (a) plot of fluorescence intensity for each complex, inset shows the averagedecay lifetime, (b) plot of time versus decay lifetime occurrence as extracted from theFLIM images shown in figure 3.

8

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

it is noted that the anisotropy is approximately uniform, i.e., there is a significant degree ofuniformity in the sample’s anisotropy with the average anisotropy being 0.19. To calculate theaverage anisotropy, fluorescence anisotropy histograms computed from polarization resolvedimages in figure 6 using equations outlined above were used. The data show that for the n-TMPyP ±GO anisotropy is the same, i.e., about 0.15, while for a-TMPyP ±GO anisotropy isalso the same, about +/−0.5, specifically the calculated average anisotropy values been: n-

Figure 5. Fluorescence anisotropy images for all four porphyrin samples.

Figure 6. Plots for all four porphyrin samples. Fluorescence anisotropy histogramscomputed from polarization resolved images from figure 5.

9

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

TMPyP (0.19 ± 0.1), n-TMPyP+GO (0.16 ± 0.04), a-TMPyP (0.05 ± 0.01) and a-TMPyP +GO(0.08 ± 0.02).

When GO is added to n-TMPy (see image in figure 5), while there is an alteration in theresulting macroscale structures, a uniform anisotropy is present with an average value about thesame as for n-TMPyP only (within error). This indicates a significant degree of ordering ispresent, and since the fluorescence originates from the prophyrins, this in turn indicates thatTMPyPs ordering is not affected by the introduction of GO. The inclusion of GO does notdisturb the ordering of the molecules on the macroscale.

In contrast to n-TMPyP, a-TMPyP shows a significant change in the anisotropy with amuch lower average anisotropy seen. Examining a single square self-assembled structure theanisotropy image shows an average anisotropy of 0.05, i.e.,≈ 0. However it is noted thatinspection of the image shows that there is a mixture of green and blue within the structurewhich indicates that the pophyrins are potentially orientated but only in domains smaller thanthe image resolution, i.e., on the nanoscale (see SEM imaging, details below). Examination ofthe edges of the square image for a-TMPyP shows that areas of red and blue can be seen whichpossess inverse anisotropy, showing that domains (or orientation) may exist within thestructure. For a-TMPyP +GO inspection of the image shows that an approximately squareshape is present, predominantly green in colour with an a slightly higher average anisotropy of0.08, indicating that a larger ordering is present when GO is introduced. It is noted that thisaverage ordering is still relatively low overall but higher than a-TMPyP. This may be due toremoval of strong nanoscale ordering replacing it with macroscale ordering processes.

Nanoscale imaging of the TMPyP±GO complexes was undertaken using scanning electronmicroscopy (SEM). The resulting images are shown in figure 7. The n-TMPyP SEM image

Figure 7. SEM images of all four complexes. Insets are SEM images with wider viewsof the samples.

10

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

shows micron-sized rod-like features indicating strong alignment in agreement with anisotropydata. For a-TMPyP large square-like features are observed (see inset in figure 7(b)) which whenresolved on the nanoscale show quasi-aligned nanorods packed within the squares, indicatingnanoscale sub-structure. Figure 7(b) 1 shows in more detail the nanorod structures. Nanoscaledomains of differently aligned nanorods may be present resulting in reduction in anisotropy onthe macroscale. When GO is present in the sample, changes in the resulting images can be seenwith n-TMPyP +GO exhibiting fragmentation of the rod-like structures, however, these rod-likestructures still hold their form on the macroscale, notwithstanding that the presence of GOseems to cause an increase in fragmentation. This fragmentation may be responsible for the lesswell-defined structures on the macroscale. For a-TMPyP +GO it is seen that nanostructures arestill formed but are less defined in their shape, being more square with an increased variation inshapes compared to the nanorods seen in a-TMPyP. The SEM image of a-TMPyP +GO showsthat the surface of the structure is relatively smooth.

DiscussionFluorescence and optical transmission images show that self-assembled porphyrins

structures are affected, (albeit to a relatively small extent) by the presence of GO. In addition,the spectroscopy data (fluorescence and absorption) indicate also that the average opticalproperties are affected when GO is added. FLIM data shows that GO is intercalated within theTMPyP self-assembled structures resulting in the average lifetime shifting downward.Significantly, analysis of FLIM data shows that the lifetimes are changed for the majority ofthe porphyrins when GO is added. This indicates that the porphyrins and GO are well mixed.We note that the self-assembled structures of the porphyrins are approximately maintained,which in turn indicates that the porphyrin and GO are interacting in a manner maintaining ingeneral the self-assembled mechanism, ex J-aggregate π–π stacking. GO+TMPyP may form amacroscale complex through aromatic π–π stacking whereby the porphyrin is lying flat onto theGO sheet, indicating that the other side of the porphyrin can also interact with a GO sheet orinteract with a second porphyrin or x number of porphyrins creating a GO-aggregate- likecomplex. If the aggregate complex consists of a small number of porphrins with a single GOsheet, this substructure potentially may form a similar structure to when GO is absent, i.e., GObehaves as a porphyrin molecule (in that both are planer in symmetry enabling potentiallysimilar stacking structures such as J-aggregation to occur). The presence of an acid (e.g.pH= 0.5 as used here) is reported to enhance porphyrin–porphyrin interactions, i.e., J-aggregation. Even with strong J-aggregation, evidence suggests that GO is present within theself-assembled porphyrin structures, resulting in the porphyrins’ optical properties and also self-assembled structure being effected. Mechanical force studies (using AFM) showed that whenGO is present, the resulting materials (i.e., for both n- and a-TMPyP +GO) wasabout >magnitude stronger.

The optical properties of the materials may be interpreted in terms of excitons. Excitedstate dynamics of molecular aggregates such as for TMPyP can be explained via molecularexciton coupling using Frenkel exciton theory [42]. These Frankel excitons exhibit a collectivephenomenon due to intermolecular coupling, linear exciton–phonon coupling and disorder. Thebalance between the delocalizing of resonant intermolecular coupling and the localizing ofstatic and dynamic disorder determines the coherence range of the exciton. In GO they may bepotentially two types of excitons: Frenkel-like exciton originated from the sp2 carbon clusterand charge-transfer excitons formed by localized states involving both oxygen and carbonatoms. Potentially these materials (i.e., TMPyP and GO) could form a complex where there is

11

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

strong molecular and also exciton interaction, noting that the degree of static and dynamicdisorder would be high, resulting in a low coherence length for the exciton. The quenching ofthe porphyrin fluorescence is indicative of full charge transfer, i.e., the transfer of porphyirnFrankel excitons to GO. This potentially supports an assertion that strong TMPyP–GOinteractions occur.

Conclusion

In conclusion, we demonstrate graphene oxide intercalation into self-assembled (J-aggregate)porphyrin structures. GO/TMPyP interaction is observed even when conditions to form TMPyPJ-aggregates are present. Fluorescence lifetime imaging (FLIM) and fluorescence anisotropyimaging were applied along with scanning electron microscopy to study the structure andoptical properties from the resulting GO/TMPyP composites. It was seen that the presence ofGO alters the macroscale and nanoscale self-assembled structures of TMPyP. In addition, GOsalso alters the optical activity, reducing the emission intensity and exciton recombinationlifetime. Evidence exists to support a model where planer-symmetric GO and TMPyP co-operate in the formation of self-assembled macro- and nanostructures forming a composite withstrong GO–TMPyP interaction.

Acknowledgments

This publication has emanated from research conducted with the financial support of SFI PIgrant and the DGPP which is funded under the Programme for Research in Third LevelInstitutions Higher Education Authority PRTLI Cycle5 and co-funded by the EuropeanRegional Development Fund. The Nanophotonics and Nanoscopy Research Group is supportedby SFI grants 11/RFP.1/MTR/3151, 12/IP/1556.

References

[1] Zhang J Z 2009 Optical Properties and Spectroscopy of Nanomaterials (Hackensack, NJ: World Scientific)[2] Al-Attar N, Kopf I, Kennedy E, Flavin K, Giordani S and Rice J H 2012 Surface-enhanced Raman scattering

from small numbers of purified and oxidised single-walled carbon nanotubes Chem. Phys. Lett. 535146–51

[3] Lordan F, Rice J H, Jose B, Forster R J and Keyes T E 2012 Effect of cavity architecture on the surface-enhanced emission from site-selective nanostructured cavity arrays J. Phys. Chem. C 116 1784–8

[4] Damm S et al 2013 Surface enhanced luminescence and Raman scattering from ferroelectrically defined Agnanopatterned arrays Appl. Phys. Lett. 103 083105

[5] Fanti M, Zerbetto Z, Galaup J P, Rice J, Birkett P, Wachter N and Taylor R J 2002 A computational and solidsolution spectroscopic study J. Chem. Phys. 116 7621

[6] Reddy M H V et al 2014 Micro-or nanorod and nanosphere structures derived from a series of phenyl-porphyrins Phys. Chem. Chem. Phys. 16 4386–93

[7] Damm S, Carville N C, Rodriguez B J, Manzo M, Gallo K and Rice J H 2012 Plasmon enhanced Ramanfrom Ag nanopatterns made using periodically poled lithium niobate and periodically proton exchangedtemplate methods J. Phys. Chem. C 116 26543–50

[8] Taylor R A, Robinson J W, Rice J H, Jarjour A, Smith J D, Oliver R A, Briggs G A D, Kappers M J,Humphreys C J and Arakawa Y 2004 Dynamics of single InGaN quantum dots Physica E 21 285–9

12

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

[9] Rice J H, Oliver R A, Robinson J W, Smith J D, Taylor R A, Briggs G A D, Kappers M J,Humphreys C J and Yasin S 2004 InGaN quantum dots grown by MOVPE via a droplet epitaxy routePhysica E 21 546–50

[10] Damm S et al 2014 Application of AAO matrix in aligned gold nanorod array substrates for surface-Enhanced fluorescence and Raman scattering Plasmonics 9 1371–6

[11] Kennedy E, Al-Majmaie R, Al-Rubeai M, Zerulla D and Rice J H 2013 Nanoscale infrared absorptionimaging permits non-destructive intracellular photosensitizer localization for subcellular uptake analysisRSC Adv. 3 13789–95

[12] Gould T J, Gunewardene M S, Gudheti M V, Verkhusha V V, Yin S R, Gosse J A and Hess S T 2008Nanoscale imaging of molecular positions and anisotropies Nat. Methods 5 1027–30

[13] Yarrow F, Kennedy E, Salaun F and Rice J H 2011 Sub-wavelength infrared imaging of lipids Biomed. Opt.Express 2 37–43

[14] Kennedy E et al 2013 Quantifying nanoscale biochemical heterogeneity in human epithelial cancer cellsusing combined AFM and PTIR absorption nanoimaging J. Biophoton. 1–9

[15] Hummers W S and Offeman R E 1958 Preparation of graphitic oxide J. Amer. Chem. Soc. 80 1339[16] Li S, Aphale A N, Macwan I G, Patra P K, Gonzalez W G, Miksovska J and Leblanc R M 2012 Graphene

oxide as a quencher for fluorescent assay of amino acids, peptides, and proteins ACS Appl. Mater.Interfaces 4 7069–75

[17] Wang Z, Medforth C J and Shelnutt J A 2004 Porphyrin nanotubes by ionic self-assembly J. Am. Chem. Soc.126 15954–5

[18] Liu Q, Zhu J, Sun T, Zhou H, Shao Q, Li G and Yin Y 2013 Porphyrin nanotubes composed of highlyordered molecular arrays prepared by anodic aluminum template method RSC Adv. 3 2765–9

[19] White N T, Beddard G S, Thorne J R, Feehan T M, Keyes T E and Heathcote P 1996 Primary chargeseparation and energy transfer in the photosystem I reaction center of higher plants J. Phys. Chem. 10012086–99

[20] Scholes G D, Fleming G R, Olaya-Castro A and van Grondelle R 2011 Lessons from nature about solar lightharvesting Nat. Chem. 3 763–74

[21] Sengupta S and Würthner F 2012 Covalently stabilized self-assembled chlorophyll nanorods by olefinmetathesis Chem. Commun. 48 5730–2

[22] Ariga K, Hill J P, Lee M V, Vinu A, Charvet R and Acharya S 2008 Challenges and breakthroughs in recentresearch on self-assembly Sci. Technol. Adv. Mater. 9 014109

[23] Reddy M, Al-Shammari R M, Al-Attar N, Rogers L, Lopez S, Forster R J, Senge M O, Keyes T E andRice J H 2014 Fractal structures in n-phenyl-porphyrin J-aggregate films Mater. Chem. Phys. 143 963–8

[24] Kennedy E, Yarrow F and Rice J H 2011 Nanoscale spectroscopy and imaging of hemoglobin J. Biophoton.4 588–91

[25] Yamamoto Y, Fukushima T, Suna Y, Ishii N, Saeki A, Seki S, Tagawa S, Taniguchi M, Kawai T and Aida T2006 Photoconductive coaxial nanotubes of molecularly connected electron donor and acceptor layersScience 314 1761–4

[26] Röger C, Müller M G, Lysetska M, Miloslavina Y, Holzwarth A R and Würthner F 2006 Efficient energytransfer from peripheral chromophores to the self-assembled zinc chlorin rod antenna: a bioinspired light-harvesting system to bridge the ‘green gap’ J. Am. Chem. Soc. 128 6542–3

[27] Reddy M H V, Al-Shammari R M, Al-Attar N, Rogers L, Lopez S, Forster R J, Senge M O, Keyes T E andRice J H 2014 Micro-or nanorod and nanosphere structures derived from a series of phenyl-porphyrinsPhys. Chem. Chem. Phys. 16 4386–93

[28] Geim A K 2009 Graphene: status and prospects Science 324 1530–4[29] Wojcik A and Kamat P V 2010 Reduced graphene oxide and porphyrin. An interactive affair in 2D ACS

Nano 4 6697–706[30] Khairutdinov R F and Serpone N 1999 Photoluminescence and transient spectroscopy of free base porphyrin

aggregates J. Phys. Chem. B 103 761–9

13

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al

[31] Udal’tsov A V, Kazarin L A, Sinani V A and Sweshnikov A A 2002 Water–porphyrin interactions and theirinfluence on self-assembly of large scale porphyrin aggregates J. Photochem. Photobiol. A 151 105–19

[32] Verma S and Ghosh H N 2012 Exciton energy and charge transfer in porphyrin aggregate/semiconductor(TiO2) composites J. Phys. Chem. Lett. 3 1877–84

[33] Xu Y, Zhao L, Bai H, Hong W, Li C and Shi G 2009 Chemically converted graphene induced molecularflattening of 5, 10, 15, 20-tetrakis (1-methyl-4-pyridinio) porphyrin and its application for optical detectionof cadmium (II) ions J. Am. Chem. Soc. 131 13490–7

[34] Liu J D, Zhao H X and Huan C Z 2012 Obstruction of photoinduced electron transfer from excited porphyrinto graphene oxide: a fluorescence turn-on sensing platform for iron (III) ions PloS One 7 e50367

[35] Yea T X, Yea S L, Chena D M, Chenc Q, Qiua B and Chena X 2012 Spectroscopic characterization oftetracationic porphyrins and their noncovalent functionalization with grapheme Spectrochim. Acta A 86467–71

[36] Amir Kaplan A, Korin E and Bettelheim A 2014 Structures self-assembled from anionic graphene andcationic manganese porphyrin: characterization and application in artificial photosynthesis Eur. J. Inorg.Chem. 2014 2288–95

[37] Kiessling D et al 2013 Novel nanographene/porphyrin hybrids—preparation, characterization, andapplication in solar energy conversion schemes Chem. Sci. 4 3085

[38] Guo P, Penglei C and Liu M 2013 One-dimensional porphyrin nano assemblies assisted via graphene oxide:sheetlike functional surfactant and enhanced photocatalytic behaviors ACS Appl. Mater. Interfaces 55336−45

[39] Malig J, Romero‐Ni.e.to C, Jux N and Guldi D M 2012 Integrating water‐soluble graphene into porphyrinnanohybrids Adv. Mater. 24 800–5

[40] Weber G 1953 Rotational Brownian motion and polarization of the fluorescence of solutions Adv. ProteinChem. 8 415–59

[41] Schaffer J, Volkmer A, Eggeling C, Subramaniam V, Striker G and Seidel C A M 1999 Identification ofsingle molecules in aqueous solution by time-resolved fluorescence anisotropy J. Phys. Chem. A 103331–6

[42] Lim S H, Bjorklund T G, Spano F C and Bardeen C J 2004 Exciton delocalization and superradiance intetracene thin films and nanoaggregates Phys. Rev. Lett. 92 107402

14

Mater. Res. Express 1 (2014) 045038 H V R Maraka et al