This journal is© the Owner Societies 2014 Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 | 18349

Cite this:Phys.Chem.Chem.Phys.,

2014, 16, 18349

Probing the aggregation behavior of4-aminophthalimide and 4-(N,N-dimethyl) amino-N-methylphthalimide: a combined photophysical,crystallographic, microscopic and theoretical(DFT) study†

Debashis Majhi, Sudhir Kumar Das, Prabhat Kumar Sahu, Saied Md Pratik,‡Arun Kumar and Moloy Sarkar*

Two fluorescent molecules, 4-aminophthalimide (AP) and 4-(N,N-dimethyl)amino-N-methylphthalimide

(DMP) have been used as the building blocks to fabricate fluorescent organic nano particles. DMP, the

analogue of AP, has been synthesized by substituting all the amine hydrogens of AP with methyl groups

to get an idea about the effect of intermolecular hydrogen bonding interactions (N–H� � �) on the aggre-

gation behavior of these molecules. All the systems have been characterized by field emission scanning

electron microscopy (FESEM). Photophysical behavior of these well characterized systems has been

investigated in molecular as well as aggregated forms. Interestingly, while the AP-aggregates exhibit a

blue-shifted absorption band (as compared to AP in its molecular form), DMP-aggregates exhibit a red-

shifted absorption band (as compared to DMP in its molecular form). These absorption data indicate the

formation of H and J aggregates for AP and DMP, respectively. The intermolecular interactions that are

responsible for the molecular self assembly of AP and DMP are studied by using X-ray crystallography.

X-ray analysis demonstrates the presence of strong intermolecular hydrogen bonding interactions in AP,

but only weak interactions (C–H� � �O, C–H� � �p) in the case of DMP. X-ray analysis also demonstrates

that varying the nature of intermolecular interactions leads to different modes of aggregation. Theoretical

studies (DFT and TD-DFT) have been carried out to investigate how different modes of aggregation lead

to changes in the optical (UV-VIS spectra) properties of these systems.

1 Introduction

Design and development of molecular materials form animportant area of contemporary research owing to the widerange of applications. This is mainly because of the fact thatphysicochemical properties are found to lie between the pro-perties of molecular and bulk forms.1 In this context, thedevelopment of organic nano/microparticles appears to be quiteattractive as they offer much more variability and flexibility inmaterial design and synthesis as compared to inorganic ones

that are based on atomic or ionic lattices.2–6 Nevertheless,organic nano/microparticles are promising molecular materialsfor optoelectronic devices,7,8 sensor applications,9 etc. Severalstudies have been carried out in recent years mainly demonstrat-ing the fabrication10–13 of organic nanomaterials and much lessemphasis has been given to understand how different modes ofaggregation lead to changes in the optical properties of thesemolecules.14–25

Organic nanocrystals are generally formed by the self-assembly of the individual monomer, which acts as the drivingforce for the synthesis of nanoparticles. Driving forces for self-assembly are based on different intermolecular forces like vander Waals interaction, hydrogen bonding, co-ordination bond-ing, and p–p stacking etc.26–28 The impact of various inter-molecular interactions can be realized as the crystal evolvesfrom a nanoscopic size domain to a microcrystalline and bulksolid.6,29 Moreover, intermolecular interactions also play animportant role in influencing the electronic properties of organicnano/micro particles.4 For example, aggregation-induced emission

School of Chemical Sciences, National Institute of Science Education and Research,

Bhubaneswar 751005, India. E-mail: moloysarkar@gmail.com;

Fax: +91-674-2302436; Tel: +91-674-2304037

† Electronic supplementary information (ESI) available: The crystal data of DMP,selected bond length and bond angles for DMP and distance from mean plane forAP and DMP. CCDC 993237. For ESI and crystallographic data in CIF or otherelectronic format see DOI: 10.1039/c4cp01912a‡ Present address: Department of Spectroscopy, Indian Association for theCultivation of Science, Kolkata – 700032, India.

Received 4th May 2014,Accepted 9th July 2014

DOI: 10.1039/c4cp01912a

www.rsc.org/pccp

PCCP

PAPER

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article OnlineView Journal | View Issue

18350 | Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 This journal is© the Owner Societies 2014

from organic nano/microcrystals are believed to be caused bychanges in the intermolecular vibronic interactions from molec-ular to aggregated states.30 Hence, it is extremely important tounderstand the correlation among intermolecular interactions,structure and optical properties of organic nano/micro crystalsso that organic molecular materials having desired electronicproperties are designed and developed for future applications.

Considering the above-mentioned facts, in the presentstudy, we have investigated the photophysical behaviors ofthe two fluorescent molecules 4-aminophthalimide (AP) and4-(N,N-dimethyl)amino-N-methylphthalimide (DMP) (Chart 1),in molecular as well as aggregated forms. As can be seen, inDMP, all hydrogen atoms of AP have been replaced by methylgroups so that the effect of hydrogen bonding interactions(N–H� � �) towards aggregation behavior is exclusively moni-tored. Among the two probes, 4-aminopththalimide (AP) is ahighly fluorescent probe that has been extensively applied inbiopolymers.31–34 It has also been used for studies on thesolvation and rotational dynamics in different solvents includ-ing ionic liquids.35–37Absorption, fluorescence emission,fluorescence lifetime and quantum yield of AP undergo sig-nificant changes with changing polarity of the solvent.38–41 Thesimplest higher analogue of AP, 4-(N,N-dimethylamino)-phthalimide (DAP) is also used in polarity sensing.42 ThoughAP and DAP are structurally similar, their photophysical pro-perties in polar medium are quite different.42 Low fluorescenceefficiency and lifetime of DAP in polar media as compared toAP have been ascribed to the low laying twisted intra molecularcharge transfer (TICT) state of DAP in polar solvent.42

All the aggregates are characterized via field-emission scan-ning electron microscope (FESEM) imaging techniques. Crys-tallographic analysis has been conducted to get further insightinto the various forces of interaction that may be responsiblefor the aggregation behavior of these two EDA systems.

Photophysical investigations on these well characterized sys-tems have been carried out. A red and blue shift in absorptionspectrum of the aggregates indicates the formation of H and Jaggregates for AP and DMP, respectively. Theoretical studies(DFT and TD-DFT) have been carried out to investigate howdifferent modes of aggregation lead to changes in the UV-VISspectra of these molecules.

2 Experimental sections2.1 Materials

4-Aminophthalimide and other starting materials for the synth-esis of DMP, such as 5-amino-2-methylisoindoline-1,3-dione,paraformaldehyde, sodium cyanoborohydride (NaBH3CN), andethyl acetate were obtained from Sigma-Aldrich and used asreceived without further purification. For photophysical studiesHPLC grade dimethyl sulfoxide (DMSO) and for NMR studiesDMSO-d6 were procured from Sigma-Aldrich and used withoutfurther purification.

2.2 Synthesis and characterization of 4-dimethylamino-N-methylphthalimide (DMP)

DMP has been synthesized by the reaction of 5-amino-2-methyliso-indoline-1,3-dione and paraformaldehyde in presence of sodiumcyanoborohydride (NaBH3CN) and ethyl acetate (25 mL) was usedas solvent in refluxing conditions under a N2 atmosphere(Scheme 1). After synthesis, DMP was characterized by 1H, 13C,DEPT NMR spectroscopic and Mass spectrometric analysis.

1H-NMR; (DMSO-d6): 3H (s, 2.98), 6H (s, 3.08), 1H (d, 6.90),1H (s, 7.00) 1H (d, 7.59) ESI-MS: 205 (M + H+). Yield 85%.

2.3 Preparation of AP and DMP solution

A dilute B10�6 (M) solution of AP and DMP in DMSO were usedfor the photophysical investigation of the molecular form.

2.4 Preparation and characterization of AP and DMPaggregate

The AP and DMP aggregate were prepared by the reprecipita-tion method. Typically, 30 mL of a DMSO solution of AP (0.01 M)was rapidly injected into 3 mL of Milli-Q-water with vigorousstirring. After addition, the stirring was continued for 15 minutes.To confirm aggregation, FESEM images of AP and DMP aggre-gates were captured. To prepare the FESEM sample, freshly

Chart 1 Molecular structures of AP and DMP.

Scheme 1 Synthetic route of DMP molecule.

Paper PCCP

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online

This journal is© the Owner Societies 2014 Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 | 18351

prepared aggregate solution was coated on cleaned alumina foiland kept for drying.

3 Instrumentation and methods

For 1H NMR spectra, a Bruker Avance 400 NMR spectrometerwas used with tetramethylsilane (TMS) as an internal standard.ESI-MS spectra were obtained with a Bruker micrOTOF-QII. Allsteady state absorption spectra were recorded on a PerkinElmer, Lambda-750 spectrophotometer and fluorescence spec-tra were recorded on a Perkin Elmer, LS 55 spectrofluorimeter.A time-correlated single-photon counting (TCSPC) spectrometer(Edinburgh, OB920) with laser diode source (lexc = 375 nm,FWHM = 95 ps) was used for all time-resolved measurements.A MCP photomultiplier (Hamamatsu R3809U-50) was used asthe detector (response time 40 ps) for all time resolved measure-ments. Dilute ludox solution in water was used as scatterer tomeasure the lamp profile. Decay curves were analyzed using thenonlinear least-squares iteration method using F900 decay ana-lysis software. The quality of the fit was judged by the chi square(w2) values, and weighted deviation was obtained by fitting. X-raydata were collected from a Bruker CCD X-ray diffraction systemwith a graphite-monochromatized Mo Ka radiation (l = 0.71073 Å).In order to understand the nature of interactions among the APand DMP molecules and their aggregation leading to changes inthe UV-VIS spectra, the ground state structures were optimizedusing the hybrid DFT functional M06-2X43–45 with 6-31+g(d,p)45

basis set as implemented in Gaussian 09.46 The hybrid metaexchange–correlation functional, M06-2X from Truhlar et al. hasbeen successful in accounting for medium and long-range electroncorrelation in charge-transfer systems and correctly incorporatesdispersion forces.47,48 The computed absorption spectra wereobtained through time-dependent DFT (TD-DFT) formalism.49

4 Results and discussion4.1 Optical studies of AP and DMP in molecular form

The absorption spectra of AP and DMP have been recorded inDMSO medium. The absorption spectra of both AP and DMP areprovided in Fig. 1. While the absorption maximum of AP appearsat 374 nm, DMP shows a peak maximum at 397 nm. The 23 nmred shift of absorption maximum for DMP can be attributed to thegreater inductive effect of the electron donating methyl group inDMP derivative. The red-shift of absorption is also supported bythe estimated ground state dipole moment values of AP (5.29 D)41

and the dimethyl analogue (N,N-dimethyl-4-aminophthalimide) ofAP (5.65 D). The molar extinction coefficient for AP and DMP areestimated to be 5 � 103 M�1 cm�1 and 6 � 103 M�1 cm�1 at theirrespective absorption maxima. These spectroscopic data indicatethat the absorption bands for AP and DMP are of intramolecularcharge transfer (ICT) in nature.

The steady state emission spectra of AP and DMP in DMSOare shown in Fig. 2. The emission maxima for AP and DMP arefound to be at 474 nm and 519 nm, respectively. The batho-chromic shift in emission maxima on going from AP to DMP is

found to be more (45 nm) as compared to the shift in absorp-tion spectra (Fig. 1). This observation indicates that the changein dipole moment from AP to DMP in the excited state is moreas compared to the same for them in the ground state. Inter-estingly, the emission intensity is observed to be significantlyless for DMP as compared to the same for AP. In fact, thequantum yield of DMP is found to be B10 fold lower than thatof AP. This observation can be rationalized by considering thepresence of non-fluorescent TICT state below the ICT state ofDMP in polar medium.41

Fluorescence decay times are also measured for AP and DMPin DMSO. A representative lifetime decay profile of AP and DMPare shown in Fig. 3. The lifetime of AP molecule (t = 18.5 ns) isobserved to be significantly higher than that of DMP molecule(t = 0.280 ns). The lower lifetime value in case of DMP also

Fig. 1 Absorption spectra of AP and DMP in molecular form in DMSO(5 � 10�6 M) solution at room temperature. Spectra are normalized withrespect to the lowest energy peak maxima.

Fig. 2 Emission spectra of AP and DMP in molecular form in DMSO(5 � 10�6 M) solution at room temperature. lexc = 375 nm.

PCCP Paper

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online

18352 | Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 This journal is© the Owner Societies 2014

indicates the presence of a non-radiative decay pathway occur-ring from the ICT - TICT energy level.41

4.2 Microscopic analysis of AP and DMP aggregates

To investigate the particle size of the aggregates of AP and DMP, theFESEM images have been captured at room temperature (Fig. 4). APparticles are found to be nearly spherical in shape, whereas DMP arenearly square. At room temperature, the most probable size of theaggregates from AP and DMP system are measured to be B220 nmand B250 nm, respectively. Dynamic light scattering measurementsalso provide similar size information about both aggregates (Fig. S2and S3, ESI†). The effect of both concentration and temperature onthe aggregation behavior of AP and DMP are also investigated. Forboth AP and DMP, sizes of the particles are observed to increasewith an increase in the monomer concentration. For example, whenmonomer concentration is doubled, sizes of AP and DMP particlesare found to be B500 nm and B400 nm, respectively. With anincrease in temperature, particle sizes of both aggregates are alsoincreased (B400 nm) (Fig. S1, ESI†).

4.3 Spectroscopic studies on AP and DMP aggregates

Photophysical properties of well-characterized aggregates havebeen studied and compared with those in molecular form.

Fig. 5 and 6 compare the absorption spectra of AP and DMPin their molecular and aggregated (colloidal) form, respectively.Table 1 collects the absorption spectral data for AP and DMP intheir molecular and aggregated states. As can be seen fromFig. 5, the absorption spectral profile of AP aggregates is blueshifted as compared to AP molecular form. However, interest-ingly, absorption spectra for DMP-aggregates are found to bered shifted as compared to DMP (Fig. 6). This blue shift in theabsorption for AP aggregates and red shift in the absorption forDMP aggregates as compared to their respective molecularforms indicates the formation of H and J aggregates for APand DMP, respectively.

Fig. 3 Lifetime decay profiles of AP and DMP in molecular form at roomtemperature. lexc = 375 nm. Solid black lines indicate the single exponen-tial fit to the experimental data points. Chi-square (w2) values for AP andDMP are 1.19 and 1.07, respectively.

Fig. 4 FESEM images of (a) AP and (b) DMP aggregates at roomtemperature.

Fig. 5 Absorption spectra of AP in molecular form in DMSO (5 � 10�6 M)solution and in aggregated form in 99% water–DMSO medium at roomtemperature. Spectra are normalized with respect to the second peakmaxima.

Fig. 6 Normalized absorption spectra of DMP in molecular form in DMSO(5 � 10�6 M) solution and in aggregated form in 99% water–DMSOmedium at room temperature. Spectra are normalized with respect tothe second peak maxima.

Paper PCCP

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online

This journal is© the Owner Societies 2014 Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 | 18353

H and J aggregate formation of the derivatives AP and DMP,respectively, in water–DMSO mixed solvent system is also truefor other solvent systems, such as water–dioxane, water–acetonitrile(Fig. S5, ESI†). For electron donor acceptor (EDA) systems like AP orDMP, a bathochromic shift of the absorption band is expectedto be observed upon increasing the polarity of the medium forthe charge-transfer transition. Thus, the present blue shift inthe absorption spectrum of AP aggregates clearly demonstratesthat this shift is due to the formation of H-type aggregates ofAP. Further, in order to confirm that the red shift in theaggregated state of DMP is not exclusively due to the changein the solvent polarity, the absorption spectrum of DMP hasbeen recorded in a polar (methanol + water) system. Thepolarity of this (methanol + water) system is maintained closeto that of the previous (DMSO + water) system. The absorptionmaxima of DMP in polar (methanol + water) is found to be404 nm, which is much lower than that observed for DMPaggregates (417 nm). The relatively larger red shift of theabsorption maximum of DMP aggregates confirms that thisshift is due to the aggregation of the particles and not due tothe change in solvent polarity. These shifts in the spectra ofaggregated forms of AP and DMP can be explained with thehelp of molecular exciton theory.39 According to this theory, amolecule is considered as a point dipole, and the energy levelcorresponding to the excitonic state of the aggregate splits intotwo levels through the interaction of transition dipoles.50 Thesplitting of the excitonic state leads to electronic transition tooccur either at lower or higher energy from the correspondingmonomeric form. These higher and lower energy transitionsrepresents H and J aggregates, respectively. The varying natureof dipole–dipole interaction between two molecules in the

aggregates are believed to be primarily responsible for splittingof energy states in the aggregates.50 A schematic diagram forexcitonic splitting for AP and DMP dimer is shown in Fig. 7.Because it is also known from molecular exciton theory50 that aface-to-face (parallel arrangements) of transition dipoles in caseof H-aggregates shifts absorption maximum to the blue andhead-to-tail (anti-parallel) arrangement for J-aggregates shiftsabsorption maximum to the red, H-type aggregates for AP andJ-type aggregates for DMP in the present case indicates that thearrangements of molecules and consequently the electroniccoupling between two molecules in the respective aggregatesare different. Crystal structure analysis and theoretical calcula-tion based on different modes of aggregates throws more lighton this aspect (vide infra).

The excited states of these aggregates are investigatedthrough steady state and time-resolved fluorescence studies.Fig. 8 shows the emission spectra of AP and DMP in theiraggregated forms. Data for emission maxima of AP and DMP inmolecular as well as aggregated forms are provided in Table 1.As can be seen from Fig. 8, the emission maxima for bothaggregates are red shifted in the respective molecular forms(homogeneous solution). Fig. 8 also shows that the emissionintensity of DMP aggregate is significantly less than that of APaggregate. Moreover, data that are collected in Table 1 alsodemonstrate large Stokes-shifted emission for both AP andDMP aggregates. This large Stokes shift perhaps indicates thatemission in these systems do not originate from the same stateto which they were vertically excited.29 Fluorescence decayprofiles of AP and DMP in aggregated forms at room tempera-ture are shown in Fig. 9. The average lifetime values corre-sponding to the aggregates of AP and DMP are estimated to bet = 1.19 ns and 0.213 ns, respectively. This lifetime data whencompared with the respective monomers of AP and DMP clearlyindicates that the average lifetime of the aggregated system hasdecreased due to the aggregation induced quenching of theexcited states.51 We would also like to note here that lifetimevalues are independent of different emission wavelengths(Fig. S7, ESI†). It may be mentioned in this context that veryrecently Durantini et al.52 demonstrated that when AP wasexcited around 300 nm in water, the emission maximum was

Table 1 Absorption labsmax (nm) and emission maxima lflu

max (nm) of AP andDMP in molecular forms in DMSO (5 � 10�6 M) solution and in aggregatedform in 99% water–DMSO (5 � 10�5 M) solution

Systems labsmax (nm) lflu

max (nm)

AP molecule 374 474AP aggregate 368 551DMP molecule 397 518DMP aggregate 417 595

Fig. 7 Schematic diagram illustrating allowed transitions for H- and J-aggregates. Here ‘m’ represents transition moment integral, which must begreater than zero for allowed transitions.

PCCP Paper

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online

18354 | Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 This journal is© the Owner Societies 2014

found to be at 562 nm, whereas the emission maximum wasfound to be at 546 nm on excitation at 370 nm. Moreover, thefluorescence decays of AP in water showed no emission wave-length dependence at lexc = 300 nm, whereas it was found to bedifferent when lexc = 370 nm was used. The excitation wave-length dependent emission behavior has been explained byconsidering the keto–enol tautomerism of the derivative AP inpure water.52 However, in the present study (99% water–DMSO)the excitation wavelength-dependent emission behavior of APhas not been observed (Fig. S4, ESI†). This behavior is notsurprising in a sense that in 99% water–DMSO binary mixture,AP forms aggregate (supramolecular assembly) through theinvolvement of all of its hydrogen atoms (vide crystal structuresof AP). In this context, we would also like to mention that keto–enol tautomerism was observed by Durantini et al. at much

lower solute concentration. In the present case, equilibriumshifts to keto form, which aggregates to the supramolecularassembly, and hence, behavior observed by Durantini et al. isnot observed. Further, we have studied the aggregation beha-vior of 5-amino-2-methylisoindoline-1,3-dione (starting mate-rial of DMP) in 99% water–DMSO binary mixture. As can beseen from Fig. S6 (ESI†), the absorption spectral profile of theaggregated form of this molecule is blue shifted as compared toits molecular form. While the absorption maximum of thissystem in its molecular form appears at 382 nm, its aggregatedform shows a peak maximum at 376 nm. This blue shift (6 nm)of the absorption maximum for methylisoindoline-1,3-dioneaggregates indicates the formation of H aggregates. The emis-sion maximum of this molecule in its molecular and aggre-gated forms appears at 486 nm and 560 nm, respectively(Fig. S7, ESI†). These data corresponding to the molecular andaggregated forms of the three systems, AP, methylisoindoline-1,3-dione and DMP, clearly indicate that hydrogen bondinginteraction can play an important role towards the aggregationbehavior of organic molecules.

We have also studied the absorption behavior of AP–DMPheteroaggregates. The hetero aggregates are formed by takingAP and DMP in 1 : 1 mole ratio. Interestingly, the observedheteroaggregates exhibits the absorption maxima at 395 nm,which is very close to the mean value of absorption maxima368þ 417

2¼ 392:5

� �obtained for individual aggregates from

AP and DMP, respectively. Fig. 10 represents the absorptionmaximum for heteroaggregates along with absorption maximafor individual AP and DMP aggregates.

4.4 Crystal structures of AP and DMP

To understand the role of intermolecular interactions duringthe self-assembly of molecules the crystal structures of both APand DMP has been analyzed. It may be mentioned that inter-molecular effects arising from interactions between moleculesdepend on the geometry of the packing structure.53 Moreover,the various intermolecular interactions also play crucial role ingoverning the photophysical properties of aggregates.22 Hence,the study of intermolecular interaction through the analysis of

Fig. 8 Emission spectra of AP and DMP in aggregated form in 99% water–DMSO medium at room temperature. lexc = 375 nm.

Fig. 9 Fluorescence decay profiles of AP and DMP in aggregated forms atroom temperature. lexc = 375 nm. Solid black lines indicate the singleexponential fit to the experimental data points. Chi square (w2) values forAP and DMP aggregates are 1.23 and 1.12, respectively.

Fig. 10 The absorption maxima of 1 : 1 AP–DMP aggregate along with APand DMP aggregate.

Paper PCCP

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online

This journal is© the Owner Societies 2014 Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 | 18355

the packing geometry of molecules will not only be helpful tounderstand the molecular aggregation pattern but also theirinfluence in controlling photophysical properties.

Crystals of DMP, suitable for X-ray diffraction, were obtainedby recrystallization of the crude product from methanol. Thecompound crystallizes in orthorhombic crystal fashion withspace group P21/n bearing four molecules per unit cell, whereasAP crystallizes in same manner with space group Pna21.54 Themolecular structure of DMP and AP are illustrated in Fig. 11. Thecrystal structure refinement data and selected structural para-meters are collected in Tables S1 and S2 (ESI†), respectively.

Both the molecules exhibit nearly planar geometry. The dis-placement of atoms from the mean plane through non-hydrogenatoms are given in Table S3 (ESI†). The fact that amine groups(NH2 for AP and NMe2 for DMP) is planarized is evident from thesum of their bond angles AP (HNH), 358.851; DMP (CNC), 359.721;and torsion angles (each being close to 01 or 1801). Furthermore,the dihedral angle of amine group and the mean planes of theadjacent phenylene ring AP, 5.20(8)1; DMP, 3.35(71)1 (Tables S4and S5, ESI†) also supports the planarization of the NH2 andNMe2 groups attached to the aryl ring.55

We have also carefully looked at the packing structures ofboth systems. As can be seen from Fig. 12, for AP, all threehydrogens from amido and amine N atoms forms N–H� � �Ohydrogen bonding interactions, where H atom of one APmolecule is found to be associated with the O acceptor from

the neighboring molecules. Interestingly, the packing diagramreveals that alternate molecules are parallel to each other. Themolecules in the lattice are interlinking in a head-to-headfashion (Fig. 12). The parallel aryl rings of the molecules areclose enough [centroid to centroid 3.7216(5) Å] to form chan-nelized p� � �p stacking (Fig. 13). The parallel geometry of twoadjacent molecules in the packing diagram of AP supports theformation of the H-type aggregates for AP as parallel alignmentof transition dipole moments is required to form H aggregates.

Quite interestingly, DMP, where all the hydrogen atoms arereplaced by methyl group, exhibits very different crystal packing ascompared to AP. DMP molecules are found to be packed throughtwo kinds of C–H� � �O and a weak C–H� � �p interaction (Table 2).The CH3 protons adjacent to N atom in the amine fragment aresufficiently acidic to form hydrogen bonds with the oxygenacceptor from the neighboring molecules. In DMP, the C–H� � �Ointeractions appears to be crystal director force, which build 2Dsupramolecular array of antiparallel chains by head to tail inter-linking of molecules in the lattice (Fig. 14). The weak C–H� � �pinteraction organizes the 2D sheets into 3D supramolecularassembly. The topological properties of weak C–H� � �p(C7) inter-actions as indicated by the 2.806 Å distance (ring-to-ring distanceis 3.846 Å with an angle 66.21) is apparent as its reciprocal natureresults in the formation of a zero dimensional centrosymmetricdimeric unit (Fig. 15), which are a consequence of 3D weaving ofsheet arrays in the crystal packing of DMP. Anti-parallel (head-to-tail) arrangements of molecules in the packing diagram of DMPmolecules support the formation of J-type aggregates. Presentanalysis on crystal structure data have clearly demonstrated thatthe nature of intermolecular interactions plays an important rolein the formation of aggregated states through the self-assembly ofmolecules, and the fine tuning of intermolecular interactions maylead to different mode of aggregations.

4.5 Theoretical calculation of the experimental results

Theoretically, we have studied the nature of aggregation of APand DMP molecules and investigated how different modes of

Fig. 11 ORTEP diagram (50% displacement thermal ellipsoids) withatomic numbering scheme of (a) AP and (b) DMP.

Fig. 12 Interplay of N–H� � �O hydrogen bonding interactions involving thebifurcated oxygen acceptor O1 in building 3D crystal lattice of AP.

Fig. 13 The parallel aryl rings of AP molecules form channelized p� � �pstacking.

PCCP Paper

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online

18356 | Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 This journal is© the Owner Societies 2014

aggregation lead to changes in the UV-VIS spectra of thesemolecules. The ground states of monomers AP, DMP, and theheterodimers (AP–DMP) were fully optimized, while only H-atoms were optimized in the case of self-assembled aggregateddimers AP–AP and DMP–DMP using the hybrid DFT functionalM06-2X56,57 with 6-31+g(d,p) basis set as implemented inGaussian 09.39,40 The hybrid meta exchange–correlation func-tional, M06-2X from Truhlar et al. has been successful in account-ing for medium-range electron correlation in charge-transfer

systems and correctly incorporates dispersion forces.58–60 Addi-tional frequency calculations were also performed for the fullyoptimized monomers and heterodimers for ensuring real vibra-tional modes for the minimum ground state structures. Thecomputed absorption spectra were obtained through time-dependent DFT (TD-DFT) formalism.61–65 All the calculationsare performed in gas phase. In our simplified models, we haveconsidered both linear and p-stack arrangements independently,for the self-assembled homodimers of AP and DMP taken fromthe respective crystal structures (Fig. 16). The center-to-centerdistance between two phenyl rings in AP and DMP are 6.63 Åand 8.49 Å respectively, while the p-stacking interaction betweenthe two molecules in AP and DMP self-assembled aggregateddimmers are 3.72 Å and 3.84 Å, respectively. The value is quitesimilar to their single crystal X-ray structure analysis. Thisindicates moderately strong p-stacking interactions between twochromophoric units of AP and DMP.63,64 Since the molecules arecloser in the p-stacked configuration of the monomers, excitonsplitting and hence shift in lmax should arise due to p-stackingrather than linear arrangements. We have also considered twohetero-type p-stacked dimers formed between AP and DMP withthe parallel and antiparallel interaction, where the two chromo-phores slipped parallel to each other with the stabilizationenergies of �12.0 kcal mol�1 and �15.8 kcal mol�1, respectively(Fig. 16).

The computed lmax (Fig. 17) for AP, DMP, and their homo-and heteroaggregates at TD-B3LYP/6-31+G(d,p) levels of theoryare summarized (Table 3). Here, optical transitions refer to theS0 - S1 transition. Hence, the spectroscopic studies indicatethat the AP dimer shows a blue shift of 7.4 nm as a conse-quence of the parallel p-stacked arrangement, while for theantiparallel arrangement of the DMP molecules leads to ared shift 9.7 nm on forming dimer, which is expected for aJ-aggregate like state of association based on a simple dipole–dipole interaction-induced exciton splitting.66–68 The lineararrangement of AP indicates red shift which is opposed bythe experimental observation. Hence, the blue-shifted spectraof AP self-assembled dimer arise due to the p-stacked arrange-ment. The UV-visible spectrum of linearly arranged DMP dimershows a small change from its monomer (3.7 nm red shifted).Therefore, our calculation suggests that for both AP and DMP,the blue and red shift in the spectra observed in our experimentcould be due to the results of p-stacking rather than a linear

Table 2 Hydrogen bond parameters obtained from crystal structures of AP and DMP

Systems D–H� � �A D–H (Å) H/p� � �A (Å) D–A (Å) D–H� � �A (1)

DMP molecule C10–H10B� � �O1i 0.96 2.618 (14) 3.529 (25) 158.710 (124)C10–H10A� � �O2ii 0.96 2.418 (14) 3.400 (26) 160.251 (126)C10–H10C� � �p(C7) 0.96 2.8854 (15) 3.6153 (26) 133.622 (133)

AP molecule N1–H1� � �O1iii 0.88 2.059 (420) 2.922 (41) 168.255 (3853)N2–H2A� � �O1iv 0.99 2.128 (602) 3.119 (54) 172.596 (5048)N2–H2B� � �O2v 0.91 2.106 (536) 2.992 (48) 162.945 (4816)p� � �p (centroid defined by C2–C8) 3.7216 (5)

Symmetry codes: (i) 1/2 + x, 3/2 � y, 1/2 + z; (ii) �1 + x, y, 1 + z; (iii) �1 + x, y, 1 + z; (iv) 1 � x, �y, 3/2 + z; (v) 3/2 � x, �1/2 + y, 1/2 + z; (vi) 1 � x, 1 � y,1/2 + z.

Fig. 14 Role of C–H� � �O interactions in head-to-tail interlinking ofmolecules in repeating antiparallel chains and 2D supramolecular arrayof DMP.

Fig. 15 Formation of centrosymmetric dimeric units via C–H� � �p andp� � �p interactions of DMP molecules.

Paper PCCP

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online

This journal is© the Owner Societies 2014 Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 | 18357

arrangement. Our calculation shows that for the antiparallelp-stacked arrangement, the computed lmax is 30.7 nm blueshifted with respect to AP monomer and 1.7 nm red shifted

with respect to DMP monomer. This is in agreement with theexperimental results (Table 1). Since from the binding energypoint of view, the antiparallel p-stacked AP–DMP heterodimeris more stable than parallel p-stacking arrangement by anamount of �3.8 kcal mol�1, we believe that the experimentallyobserved spectra is arising due to the antiparallel p-stacking ofthe AP–DMP heterodimer.

5 Conclusions

In this work, the aggregation behavior of two fluorescentmolecules, 4-aminophthalimide (AP) and 4-(N,N-dimethyl)-amino-N-methylphthalimide (DMP) have been investigated byphotophysical, crystallographic, microscopic and theoretical(DFT) studies. Out of the two fluoroprobes, DMP has beensynthesized by substituting all the amine hydrogen of AP bymethyl groups. These two systems are selected for the presentstudy to basically get an idea about the effect of intermolecularhydrogen bonding interactions (N–H� � �) on the aggregationbehavior of these molecular materials. The aggregates of bothAP and DMP are prepared based on the standard reprecipitationmethod without using any external stabilizer. Photophysicalbehavior on these well characterized systems has been carriedout in molecular as well as aggregated forms. Interestingly,AP-aggregate exhibits a blue-shifted absorption band as com-pared to AP in its molecular form. On the contrary, DMP-aggregate exhibits a red-shifted absorption band as comparedto its molecular form. These spectral data indicate the formationof H and J aggregates for AP and DMP, respectively. The crystalpacking of AP and DMP are analyzed by X-ray crystallographicdata to have a better idea about the intermolecular interactionsthat are responsible for the molecular self-assembly of AP andDMP. X-ray analysis demonstrates the different modes of aggre-gation for AP and DMP arises due the varying nature of inter-molecular interaction for AP and DMP. Density functional

Fig. 16 Ground state optimized structures [at M06-2X/6-31+G(d,p) level in gas phase] for (a) AP monomer (b) AP linear dimer (c) AP p-stacked dimer(d) DMP monomer (e) DMP linear dimer (f) DMP p-stacked dimer (g) AP–DMP parallel (h) AP–DMP antiparallel.

Fig. 17 The computed lmax for AP and DMP and their homo and hetero-aggregates at TD-B3LYP/6-31+G(d,p) level of theory.

Table 3 The computed lmax for AP and DMP, in monomer, homo- andheterodimers forms

Systemslmax

(nm)f (Oscillatorstrength) Transitions

AP monomer 339.2 0.07 H–L; 100%AP linear dimer 368.7 0.09 H–L + 1; 100%AP stacked dimer 331.8 0.06 H � 1 � L; 94.9%

H–L + 1; 5.1%DMP monomer 371.6 0.08 H–L; 100%DMP-linear 375.3 0.12 H–L + 1; 100%DMP-stacked 381.3 0.15 H� 1 � L; 54.6%

H–L + 1; 45.4%AP–DMP (parallel) 375.5 0.06 H–L + 1; 100%AP–DMP (antiparallel) 369.9 0.06 H � 1 � L; 78.5%

H � 1 � L + 1; 4.9%H–L + 1; 16.6%

PCCP Paper

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online

18358 | Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 This journal is© the Owner Societies 2014

calculations have also been carried out to investigate howdifferent modes of aggregation lead to changes in the UV-VISspectra of these systems.

Acknowledgements

This work was supported by the Council of Scientific andIndustrial Research (CSIR), New Delhi. S. K. D. thanks theCouncil of Scientific and Industrial Research (CSIR), New Delhifor awarding a fellowship. D. M. and P. K. S. thank NationalInstitute of Science Education and Research (NISER), Bhubaneswarfor a fellowship.

References

1 D. Horn and J. Rieger, Angew. Chem., Int. Ed., 2001, 40,4330–4361, and references therein.

2 P. A. Alivistos, Science, 1996, 271, 933–937.3 X. Peng, C. S. Michael, V. K. Andreas and A. P. Alivisatos,

J. Am. Chem. Soc., 1997, 119, 7019–7029.4 E. A. Silinsh, Organic Molecular Crystals: Their Electronic

States, Springer-Verlag, Berlin, 1980.5 D. Li and J. L. Guo, J. Phys. D: Appl. Phys., 2008, 41, 105115.6 S. Varghese and S. Das, J. Phys. Chem. Lett., 2011, 2, 863–873.7 S. S. Babu, S. Prasanthkumar and A. Ajayaghosh, Angew.

Chem., Int. Ed., 2012, 51, 1766–1776.8 H. Wang, X. Xu, A. Kojtari and H.-F. Ji, J. Phys. Chem. C,

2011, 115, 20091–20096.9 A. Ibanez, S. Maximov, A. Guiu, C. Chaillout and P. L.

Baldeck, Adv. Mater., 1998, 10, 1540–1543.10 F. Balzer and H. Rubahn, Nano Lett., 2002, 2, 747–750.11 T. Kietzke, D. Neher, K. Landfester, R. Montenegro,

R. Guntner and U. Scherf, Nat. Mater., 2003, 2, 408–412.12 L. Zhao, W. Yang, Y. Luo, T. Zhai, G. Zhang and J. Yao,

Chem. – Eur. J., 2005, 11, 3773–3777.13 M. Mas-Torrent and P. Hadley, Small, 2005, 1, 806–809.14 T. Onodera, H. Kasai, S. Okada, H. Oikawa, K. Mizuno,

M. Fujitsuka and O. Ito, Opt. Mater., 2002, 21, 595–598.15 H. Oikawa, T. Mitsui, T. Onodera, H. Kasai, H. Nakanishi

and T. Sekiguchi, Jpn. J. Appl. Phys., 2003, 42, L111–L113.16 S. Takahashi, H. Miura, H. Kasai, S. Okada, H. Oikawa and

H. Nakanishi, J. Am. Chem. Soc., 2002, 124, 10944–10945.17 H. B. Fu and J. N. Yao, J. Am. Chem. Soc., 2001, 123,

1434–1439.18 H. Fu, B. H. Loo, D. Xiao, R. Xie, X. Ji, J. Yao, B. Zhang and

L. Zhang, Angew. Chem., Int. Ed., 2002, 41, 962–965.19 D. Xiao, L. Xi, W. Yang, H. Fu, Z. Shuai, Y. Fang and J. Yao,

J. Am. Chem. Soc., 2003, 125, 6740–6745.20 B.-K. An, S.-K. Kwon, S.-D. Jung and S. Y. Park, J. Am. Chem.

Soc., 2002, 124, 14410–14415.21 S. Li, L. He, F. Xiong, Y. Li and G. Yang, J. Phys. Chem. B,

2004, 108, 10887–10892.22 A. Patra, N. Venkatram, D. N. Rao and T. P. Radhakrishnan,

J. Phys. Chem. C, 2008, 112, 16269–16274.

23 A. Patra, N. Hebalkar, B. Sreedhar and T. P. Radhakrishnan,J. Phys. Chem. C, 2007, 111, 16184.

24 V. Karunakaran, D. D. Prabhu and S. Das, J. Phys. Chem. C,2013, 117, 9404–9415.

25 P. Mazumdar, D. Das, G. P. Sahoo, G. Salgado-Moran andA. Misra, Phys. Chem. Chem. Phys., 2014, 16, 6283–6293.

26 I. W. Hamley, Angew. Chem., Int. Ed., 2003, 42, 1692–1712.27 B. Olenyuk, J. A. Whiteford, A. Fechtenkotter and P. J. Stang,

Nature, 1999, 398, 796–799.28 X. Huang, Y. Jeong, B. K. Moon, L. Zhang, D. H. Kang and

I. Kim, Langmuir, 2013, 29, 3223–3233.29 A. Patra, N. Hebalkar, B. Sreedhar, M. Sarkar, A. Samanta

and T. P. Radhakrishnan, Small, 2006, 2, 650–659.30 J. B. Birks, Photophysics of Aromatic Molecules, Wiley,

London, 1970.31 C. Reichardt, Chem. Rev., 1994, 94, 2319–2358.32 J. Riedl, R. Pohl, N. P. Ernsting, P. Orsag, M. Fojta and

M. Hocek, Chem. Sci., 2012, 3, 2797–2806.33 J. Stambasky, M. Hocek and P. Kocovsky, Chem. Rev., 2009,

109, 6729–6764.34 D. E. Wetzler, C. Chesta, R. Fernandez-Prini and P. F. Aramendıa,

J. Phys. Chem. A, 2002, 106, 2390–2400.35 D. Mandal, A. Datta, S. K. Pal and K. Bhattacharyya, J. Phys.

Chem. B, 1998, 102, 9070–9073.36 M. Sajadi, T. Obernhuber, S. A. Kovalenko, M. Mosquera,

B. N. Dick and P. Ernsting, J. Phys. Chem. A, 2009, 113,44–55.

37 S. Sen, D. Sukul, P. Dutta and K. Bhattacharyya, J. Phys.Chem. A, 2001, 105, 10635–10639.

38 D. Noukakis and P. Suppan, J. Lumin., 1991, 47, 285–295.39 R. Wang, C. Hao, P. Li, N. G. Wei, J. Chen and J. Qiu,

J. Comput. Chem., 2010, 31, 2157–2163.40 W. R. Ware, S. K. Lee, G. J. Brant and P. P. Chow, J. Chem.

Phys., 1971, 54, 4729–4737.41 T. Soujanya, R. W. Fessenden and A. Samanta, J. Phys.

Chem., 1996, 100, 3507–3512.42 G. Saroja, B. Ramachandram, S. Saha and A. Samanta,

J. Phys. Chem. B, 1999, 103, 2906–2911.43 Y. Zhao, N. E. Schultz and D. G. Truhlar, J. Chem. Theory

Comput., 2006, 2, 364–382.44 Y. Zhao and D. G. Truhlar, Acc. Chem. Res., 2008, 41, 157–167.45 P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973,

28, 213.46 M. J. Frisch, et al., Gaussian 09 Rev 09.A0, Gaussian, Inc.,

Wallingford, CT, 2009.47 A. K. Jissy and A. Datta, J. Phys. Chem. B, 2010, 114, 15311–15318.48 A. Nijamudheen, D. A. Jose, A. Shine and A. Datta, J. Phys.

Chem. Lett., 2012, 3, 1493–1496.49 R. E. Stratmann, G. E. Scuseria and M. J. Frisch, J. Chem.

Phys., 1998, 109, 8218–8224.50 (a) M. Kasha, H. R. Rawls and M. A. El-Bayoumi, Pure Appl.

Chem., 1965, 11, 371–392; (b) A. Datta and S. K. Pati, J. Chem.Phys., 2002, 118, 8420.

51 (a) H. Yuning, W. Y. Jacky and Z. T. Ben, Chem. Commun.,2009, 4332–4353; (b) H. Yuning, W. Y. Jacky and Z. T. Ben,Chem. Soc. Rev., 2011, 40, 5361–5388.

Paper PCCP

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online

This journal is© the Owner Societies 2014 Phys. Chem. Chem. Phys., 2014, 16, 18349--18359 | 18359

52 A. M. Durantini, R. D. Falcone, J. D. Anunziata, J. J. Silber,E. B. Abuin, E. A. Lissi and N. M. Correa, J. Phys. Chem. B,2013, 117, 2160–2168.

53 T. Li, P. W. Ayers, S. Liu, M. J. Swadley and A.-M. Clare,Chem. – Eur. J., 2009, 15, 361–371.

54 M. Sarkar, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2008,64, O1654.

55 A. Panda, G. Mugesh, H. B. Singh and J. B. Ray, Organo-metallics, 1999, 18, 1986–1993.

56 Y. Zhao, N. E. Schultz and D. G. Truhlar, J. Chem. TheoryComput., 2006, 2, 364–382.

57 Y. Zhao and D. G. Truhlar, Acc. Chem. Res., 2008, 41,157–167.

58 S. A. Abraham, D. Jose and A. Datta, ChemPhysChem, 2012,13, 695–698.

59 A. K. Jissy, U. P. M. Ashik and A. Datta, J. Phys. Chem. C,2011, 115, 12530–12546.

60 D. Jose and A. Datta, Cryst. Growth Des., 2011, 11, 3137–3140.61 R. E. Stratmann, G. E. Scuseria and M. J. Frisch, J. Chem.

Phys., 1998, 109, 8218–8224.62 G. B. McGaughey, M. Gagne and A. K. Rappe, J. Biol. Chem.,

1998, 273, 15458–15463.63 C. R. Martinez and B. L. Iverson, Chem. Sci., 2012, 3,

2191–2201.64 A. Datta and S. K. Pati, Chem. – Eur. J., 2005, 11, 4961–4969.65 A. Datta, F. Tarenziani and A. Painelli, ChemPhysChem,

2006, 7, 2168–2174.66 A. Datta and S. K. Pati, J. Chem. Phys., 2003, 118, 8420–8427.67 A. Datta and S. K. Pati, J. Phys. Chem. A, 2004, 108, 320–325.68 A. Datta and S. K. Pati, Chem. Soc. Rev., 2006, 35, 1305–1323.

PCCP Paper

Publ

ishe

d on

14

July

201

4. D

ownl

oade

d by

Nat

iona

l Ins

titut

e of

Sci

ence

Edu

catio

n &

Res

earc

h on

26/

04/2

015

13:1

7:50

. View Article Online