Cocrystal or Salt: Solid State-Controlled Iodine Shift in Crystalline Halogen-Bonded Systems

Cocrystal or Salt: Solid State-Controlled Iodine Shift in CrystallineHalogen-Bonded SystemsOlena Makhotkina,† Julien Lieffrig,† Olivier Jeannin,† Marc Fourmigue,*,† Emmanuel Aubert,‡

and Enrique Espinosa*,‡

†Institut des Sciences Chimiques de Rennes (ISCR), Universite Rennes 1, UMR CNRS 6226, Campus de Beaulieu, 35042 Rennes,France‡Laboratoire CRM2, UMR CNRS 7036, Institut Jean Barriol, Universite de Lorraine, BP 70239, 54506 Vandoeuvre-les-Nancy, France

*S Supporting Information

ABSTRACT: The distinction between cocrystals and salts isusually investigated in hydrogen-bonded systems as A−H···B⇆ [A]−···[H−B]+, where the position of the hydrogen atomactually defines the ionicity of the complex. The samedistinction, but in halogen-bonded systems, is addressedhere, in complexes formed out of N-iodoimide derivatives ashalogen bond donors, and pyridines as halogen-bond accept-ors, anticipating that the position of the iodine atom in theseA−I···B ⇆ [A]−···[I−B]+ systems will also define their degreeof ionicity. We show that the crystalline halogen-bondedcomplexes of N-iodosuccinimide (NIS) with pyridine, 4-methylpyridine, and 4-dimethylaminopyridine can be described as“close-to-neutral” cocrystals while the crystalline halogen-bonded complex of N-iodosaccharin (NISac) with 4-dimethylaminopyridine adopts a “close-to-ionic” structure. Theoretical calculations were performed (i) in gas phase on isolatedNIS···Py-NMe2 and NISac···Py-NMe2 complexes, and (ii) on the periodic crystal phases, and combined with the topologicalanalysis of the electron density distribution ρ(r). We demonstrate unambiguously that the crystal environment actually plays acrucial role in the stabilization of the “close-to-ionic” structure of the NISac···Py-NMe2 complex. An external homogeneouselectric field ε applied to this complex (all atoms frozen at the crystalline geometry, except iodine) in either gas phase (ε = 3.7GV m−1) or periodic pseudo-isolated configuration (ε = 2.8 GV m−1) is able to shift the iodine atom at the crystal geometry,miming the polarization effect induced by the local crystal electric field. The strong influence of the crystalline environment onthe iodine position is demonstrated by using plane wave DFT periodic calculations on optimized NIS·Py-NMe2 and NISac·Py-NMe2 crystal structures, as well as by applying this plane wave basis set formalism to a hypothetical solid where the halogen-bonded complexes are pushed apart from each other within a periodic system.

■ INTRODUCTION

The definition of cocrystals and salts is currently the subject ofan active debate,1−3 in connection with applications in thepharmaceutical field.4,5 Indeed, the modifications of the solidstate properties (solubility, morphology, hygroscopicity, and soforth) of a given active pharmaceutical ingredient (API) ofteninvolve the formation of ionic salts between hydrogen-bondedacids and bases. When the ΔpKa (ΔpKa = pKa(base) −pKa(acid)) exceeds 2 or 3, a salt is indeed formed, while when itis smaller than 0, a neutral cocrystal is most often isolated.6,7

The distinction between cocrystals and salts become moredifficult in the intermediate ΔpKa region,

8 but in all situations,one is faced with a hydrogen-bonded system, A−H···B ⇆[A]−···[H−B]+, where the position of the hydrogen atombetween A and B atoms actually defines the nature (neutral orionic) of the bimolecular complex.9−11

Another intermolecular interaction, namely, halogen bond-ing,12−14 has been shown to compare in many ways withhydrogen bonding.15 Halogen bonding interactions with Lewisbases find their origin in the anisotropic charge density

distribution around a covalently bound halogen atom, with anelectrophilic zone which develops in the prolongation of theC−X bond, called the σ-hole.16−20 A recent definition ofhalogen bonding is now available from IUPAC.21 Thisinteraction was investigated in the 1950s by Hassel andothers,22,23 from the reaction of dihalogens with Lewis bases.More recently, combined theoretical and experimental effortshave given to halogen bonding a strongly renewedinterest,24−27 as illustrated by numerous reviews andbooks.28−33 This interaction today finds uses in many differentdomains, such as crystal engineering,34−36 molecular materials(conducting,37−42 magnetic,43−45 photoactive,46,47 liquid crys-tals,48,49 gels,50 and polymers51), supramolecular chemistry,52 aswell as organic catalysis,53−58 molecular biology, and drugdesign.59−62 From a thermodynamic point of view, thisinteraction has been shown to compare in many instances

Received: April 17, 2015Revised: May 25, 2015Published: June 9, 2015

Article

pubs.acs.org/crystal

© 2015 American Chemical Society 3464 DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

with normal hydrogen bonds, and a halogen bond basicity scaletoward I2 (acting as XB donor) has been developed63 andcorrelated to structural data.64 Considering these closeanalogies between hydrogen and halogen bonding, and themany examples of competitive situations associating bothinteractions,65−73 we wondered if a situation comparable to thehydrogen-bonded cocrystal vs salt system (Scheme 1a) couldbe extended to an analogous, halogen-bonded, cocrystal vs saltsystem, where the moving hydrogen atom would be replaced bya moving halogen atom, according to Scheme 1b.

This situation is essentially observed only in the product ofthe reaction of dihalogens such as I2, Br2, Cl−I, or Br−I withthiones74−77 or pyridines,78−81 affording for example the ionicN-iodopyridium salts, halogen bonded to the halide anion(Scheme 2a). Other related examples involve strongly covalent,

symmetrical systems, but with charge conservation (Scheme2b), as found for example with the central iodine atom in thetriiodide anion, or in bis(pyridine)iodonium cations.82−85

Our strategy to favor an ionic form through halogen transferbetween the neutral halogen-bonded cocrystal and the ionic,halogen-bonded salt, that is, A−I···B ⇆ [A]−···[I−B]+, is basedon the analogy with the hydrogen bonded systems. The abilityto delocalize the positive charge brought by the moving iodineshould indeed play an important role in stabilizing the [I−B]+species. Similarly, the stability of the [A]− species after iodinetransfer is another important point.86 These combinedrequirements led us to consider the solid state properties ofN-iodoimide derivatives in their interaction with pyridines ofvarying Lewis base character. Indeed, as noted by Rissanen,87

N-iodosuccimide (NIS) and N-iodosaccharin (NiSac) (Scheme3) are powerful halogen bond donors: “the structures presented(with these N-iodoimides) contain some of the strongesthalogen bonds found up to now. The strong interactions are aconsequence of the strong polarization of the nitrogen-boundhalogen by the imide group”.

In crystalline NIS,88 a short N−I···OC XB is indeed found(2.58 Å, 26% reduction with respect to the sum of the van derWaals radii), while several halogen-bonded cocrystals are alsoreported with halides,89 amines,90 and imines.91 Similarly,NiSac crystallizes with water, THF, pyridine, or pyrazine, withinevery structure a very short and linear N−I···O or N−I···N′halogen bond.92 We therefore postulated that such stronghalogen bond donors, when faced with Lewis bases such aspyridines, also known to exist as N−iodonium cations, could bethe most appropriate candidates to eventually observe, in thesolid state, a continuous evolution from cocrystal to salt, withan iodine shift analogous to a proton shift, according to theequation: A−I···B ⇆ [A]−···[I−B]+. We describe below ourfirst results along these lines, from a combination of structuraland theoretical investigations.

■ RESULTS AND DISCUSSIONCocrystals of NIS and NISac were investigated with threepyridines of different halogen acceptor strength, namely,pyridine itself (Py), 4-picoline (Py-Me), and 4-dimethylamino-pyridine (DMAP, Py-NMe2). It is indeed anticipated that theelectron releasing substituents in the para position of pyridinewill increase the charge on the sp2 nitrogen atom of thepyridine derivative and favor a stronger halogen bonding.64

With NIS, we systematically obtained halogen-bondedcomplexes with a 1:1 stoichiometry (Figure 1), a very short

and linear N−I···N′ halogen bond (Table 1), corresponding toan averaged reduction ratio (relative to van der Waals radiitaken at 1.55 for N and 1.98 Å for I)93 of 0.70 with Py to 0.68with Me2N-Py. This strengthening can be noted also with othercomplexes of Py-NMe2, for example, with perfluorodiodoben-zene derivatives.94,95 Altogether, the three NIS systems behaveat first sight as strong but “normal” halogen-bonded cocrystals.Moving now to the NISac complexes, the pyridine derivative,

NISac•Py, had already been reported earlier,92 but the authorsnoted that “the modest quality of the crystal of NISac•Pyresulted in large R1 and wR2 residualsLarge residual electrondensity was found in the final difference electron density maps”.Our own investigations on single crystals of NISac•Py confirmthe reported unit cell parameters but showed a very complexdiffraction pattern with two sets of incommensurate super-structures.96 Clearly thus, the reported structure cannot be fullytrusted, albeit it indicates a tendency toward a much strongerhalogen bond than with NIS (Table 2). In order to clarify thispoint, the NISac complexes with Py-Me and Py-NMe2 werealso prepared and structurally characterized (Figure 2). Asshown in Table 2, the I···N′ distances are now extremely short(2.2−2.3 Å), when compared with the NIS complexes (2.4−2.5Å). Also, the N···N′ distances are much shorter (4.50 Å) thanin the NIS complexes (4.55−4.60 Å). This is also accompaniedby a concomitant lengthening of the initially “covalent” N−Ibond of N-iodosaccharin. All data confirm that the N-

Scheme 1

Scheme 2. Ionic, Halogen-Bonded Systems

Scheme 3. Structures of the Strong Halogen Bond Donors

Figure 1. Detail of the structure of the halogen-bonded NIS···Py-NMe2 complex, as a representative example of the NIS complexes (seeTable 1). N···I distances are given in Å.

Crystal Growth & Design Article

DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

3465

iodosaccharin is a much stronger halogen bond donor than N-iodosuccinimide.Furthermore, we note also that in NISac•Py-NMe2, the so-

called I···N′ halogen bond is actually shorter than the so-calledN−I “covalent” bond of NISac. In other words, the iodine atomis closer to the sp2 nitrogen atom (N′) of Py-NMe2 than to thenitrogen atom of saccharin, providing a very appealing situationwith a potentially predominant ionic form, when comparedwith the halogen-bonded neutral cocrystal form (Scheme 4).The two interatomic distances N···I (d1) and I···N′ (d2) have

been correlated, for the six systems except NISac•Py, with theSteiner-Limbach relationship (eq 1)

+ = + − +

+ − − +

d d d d d b

d d d d b

2 ( ) 2

ln[1 exp{( )/ }]1 2 02 1 2

01 02 1 2 (1)

where d1, d2, (d1 + d2), and (d1 − d2) are distances given inTables 1 and 2, and d01, d02, and b are adjustable parameters.This correlation, which follows from the bond valence-bondlength concept well established in crystallography,97 wasoriginally developed for hydrogen-bonded systems.98 Recently,the same expression has been extended to theoreticallycalculated halogen-bonded systems.99,100 While very goodcorrelations were found for F−Cl···CNX (R2 = 0.994)99 andF−Cl···CNY (R2 = 0.989)100 complexes involving chlorineatom as halogen bond donor, only rough and bad correlationswere respectively observed for bromo- and iodo complexes F−Br···CNY (R2 = 0.945) and F−I···CNY (R2 = 0.804)complexes. Attempts to improve the quality of the fit switchingfrom the original model (eq 1) with a four-parameter modelafforded only side improvements.100 Here, the Steiner-Limbachplot for the five complexes (Figure 3) leads to an experimentaldata fitting having a very good correlation coefficient (R2 =0.986, with b = 0.37(6) Å, d01 = 2.00(3) Å, and d02 = 2.01(6)Å), indicating that eq 1 can be extended to halogen-bondedsystems even when involving highly polarizable halogen atoms.To our knowledge, this is the first experimental assessment inthe halogen bonding context. The fitted reference values of d01and d02 would correspond to the N−I and I−N′ distances inisolated N-iodoimide and N-iodopyridinium. In the monomersoptimized with DFT calculations (see Supporting Informationfor full details), the latter distances are found to be 2.046 Å inNISac, 2.059 Å in NIS, 2.094 Å in N-iodopyridinium, 2.091 Åin N-iodo-4-methylpyridinium, and 2.080 Å in N-iodo-4-dimethylaminopyridinium, respectively.This situation is strongly reminiscent of the cocrystal-salt

distinction found in hydrogen-bonded systems. For example,solid state associations of saccharin itself acting as a Brønstedacid with pyridine derivatives adopt almost invariably an ionicform,101−103 while the only reported example of an associationof succinimide with pyridine derivatives is a neutral, hydrogenbonded system.104 As stated by Bond,5 this distinction “can alsobe problematic, especially for chemically similar crystals wherethere might be a transition from neutral to charged molecular

Table 1. Room Temperature Geometrical Distances (Å) and Angles (deg) of the NNIS−I···N′R‑Py Halogen Bond in theCocrystals with NISa

d1 (N−I) d2 (I···N′) d1 − d2 N−I···N′ d1 + d2

Pyb 2.116(5) 2.493(8) −0.377(9) 180 4.609(9)2.144(8) 2.430(8) −0.286(11) 180 4.574(11)

Py-Meb 2.116(4) 2.483(4) −0.367(6) 180 4.599(6)2.142(4) 2.428(4) −0.286(6) 180 4.570(6)

Py-NMe2 2.146(4) 2.407(4) −0.261(6) 178.9(1) 4.553(6)ad1 = N−I distance, d2 = I···N′ distance, d1 + d2 = N···N′ distance (the interaction is linear). bTwo crystallographically independent NIS•Py orNIS•Py-Me complexes, both located on twofold axes.

Table 2. Geometrical Distances (Å) and Angles (deg) of the NNISac−I···N′R‑Py Halogen Bond in the Cocrystals with NISaca

T (K) d1 (N−I) d2 (I···N′) d1 − d2 N−I···N′ d1 + d2

Pyb 293 2.254(11) 2.279(11) a 174.5(4) 4.52(2)Py-Me 293 2.220(3) 2.304(3) −0.084(4) 178.2(1) 4.523(4)

150 2.223(4) 2.301(4) −0.078(6) 178.1(1) 4.523(6)Py-NMe2 293 2.292(1) 2.228(1) +0.064(1) 178.8(1) 4.520(1)

150 2.292(2) 2.218(2) +0.074(3) 178.5(1) 4.509(3)aDistances are defined as in Table 1. bFrom ref 92. X-ray data quality is poor, due to the presence of superstructures in the diffraction figure. See text.

Figure 2. Details of halogen-bonded complexes obtained betweenNISac and (a) 4-picoline (Py-Me) and (b) 4-dimethylaminopyridine(Py-NMe2), at room temperature. N···I distances are given in Å.

Scheme 4. Analogy between Iodine Transfer in theNISac•Pyridine System and Hydrogen Transfer in aCarboxylic Acid•Pyridine System

Crystal Growth & Design Article

DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

3466

components partway through a series, or for cases whereproton transfer might be dynamic”. This has also beendescribed as a “salt−cocrystal continuum”, to consider thepossibility for continuously variable degrees of proton transferin hydrogen-bonded molecular crystals.105 This possibility ofcontinuous halogen transfer has been theoretically investigatedin the prototype halogen-bonded system F−Cl···CN−X, with Xranking from strongly electron-withdrawing to stronglyelectron-releasing groups (X = CN, NC, NO2, F, CF3, Cl, Br,CCF, CCH, CH3, SiH3, Li, Na).

99 Depending on the nature ofX, the halogen bond was described as evolving from traditionalto “chlorine-shared” to ion pair.In order to rationalize the experimental observations

described above, Density Functional Theory calculations wereconducted with the NIS and NISac complexes with thestrongest halogen bond acceptor, namely, Py-NMe2, aspotential prototypes of the cocrystal and salt systems,respectively (see details in Supporting Information). Thesecalculations were done in three different ways: (i) in gas phaseon isolated NIS and NISac molecules, (ii) in gas phase onisolated NIS•Py-NMe2 and NISac•Py-NMe2 bimolecularcomplexes as extracted from the crystalline phase, and (iii)on periodic crystal phases, aimed at reflecting the actual effectsof the whole crystalline environment on the bimolecular motifand iodine transfer ability.In the gas phase DFT calculations of isolated halogen-

bonded NIS•Py-NMe2 and NISac•Py-NMe2 complexesextracted from the crystal phases, the experimental geometrieswere fixed for all atoms but iodine. As shown in Table 3,whatever the nature of the density functionals used, the iodineatom was systematically found close to the N-imide moleculesfor both systems, at variance with the experimental observationof the opposite situation in NISac•Py-NMe2. This demon-strates that the full picture of the relationships betweeninteraction energies, bonding distances, and properties of theelectron distribution in such crystal phases cannot be simplyderived from calculations on isolated systems, but dependsstrongly on the environment, as a polar solvent or a crystal. Inhydrogen-bonded systems, it has been shown that theseenvironmental effects can be successfully described in termsof a homogeneous external field ε parallel to the hydrogen

bond.106−111 Accordingly, the iodine position in the NISac•Py-NMe2 outlier was also optimized upon the influence of varyingexternal ε fields oriented along the N···N′ direction (−3 < ε <+5 GV m−1), showing that this atom behaves as a positivecharge following the applied electric field. The linear depend-ence of the Ndonor···I distance with ε is clearly established fromFigure 4, and this trend can be brought close to the observed

effect of an external electric field on the equilibrium distancedeq(F···H) of the hydrogen bonded dimer H−F···H−F, forwhich the linear dependence of deq with ε was analyticallydemonstrated from the interaction potential Ei[d(F···H), ε]developed in the study.109 It is noteworthy that the calculatedexternal electric field needed to place the iodine atom at theexperimental geometry in NISac•Py-NMe2 (ε = 3.7 GV m−1)falls well in the range of values felt by molecules in crystals(∼3−20 GV m−1),112−114 and is close to those observed inproteins and enzymes (∼5 GV m−1),115,116 indicating that thecrystal environment indeed plays a crucial role in the actualcharge distribution within the bimolecular halogen-bondedcomplexes.Theoretical calculations performed on the frozen NIS•Py-

NMe2 and NISac•Py-NMe2 complexes at the experimentalgeometry (without iodine position optimization) have been

Figure 3. Steiner-Limbach relationship applied to the halogen-bondedN···I···N′ systems described in this work (see room temperature datain Tables 1 and 2). d1 = N···I distance; d2 = I···N′ distance. Error barsrepresent the standard uncertainties of (d1 + d2) and (d1 − d2) inTables 1 and 2. Dashed lines represent the confidence band at 95%level.

Table 3. N−I and I···N′ Distances with Optimized Positionof the Iodine Atom in NIS•Py-NMe2 and NISac•Py-NMe2Bimolecular Complexes from Various DFT Functionals andat Experimental Geometries

functional N−I (Å) vs I···N′ (Å)With NIS B3LYP-D3 2.1263 < 2.4260

Experimental 2.146(4) < 2.407(4)With NISac B3LYP-D3 2.1625 < 2.3468

TPSSTPSS-D3 2.1697 < 2.3395PBE1PBE-D3 2.1509 < 2.3582PBEPBE-D2 2.1724 < 2.3369Experimental 2.292(2) > 2.218(2)

Figure 4. NNISac−I bond distance as a function of the amplitude ε ofthe applied external electric field. Black squares: gas phase isolatedNISac•Py-NMe2 bimolecular complex (PBEPBE-D2 functional withthe aug-cc-pVTZ basis set); open diamonds: periodic pseudoisolatedNISac•Py-NMe2 bimolecular complexes (Castep calculations). Ex-perimental NNISac−I bond distance and corresponding ε values areindicated as dashed lines. Black lines represent the least-squares linearfitting against each data set. Linear fittings are given with thecorresponding equations and correlation coefficients, as well as theconfidence band at 95% level (dot-dashed lines).

Crystal Growth & Design Article

DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

3467

also used to evaluate their partial charges by applying thetopological analysis of the electron density distribution ρ(r) inthe framework of the Quantum Theory of Atoms in Molecules(QTAIM) methodology.117 As shown in Figure 5, the iodine

atom indeed bears in both cases a positive charge, varying from+0.42e (NIS complex) to +0.45e (NISac complex). Thepartition of charges among the moieties in the complexes(Figure 5) depends on which of them the iodine atom isbelonging to, a conclusion that leads to the definiteclassification (cocrystal vs salt) of the corresponding crystalforms, as explained hereafter.The nature of both bonding interactions within the N···I···N′

motif can be further investigated from the topologicalproperties of ρ(r) at each N···I and I···N′ bond critical point(bpc). To undertake this analysis, it is suitable to refer first tothe topological properties of the N−I bonds in the fullyoptimized halogen bond donor alone, namely, NISac, NIS, andiodo-benzene, the latter being the archetypal C−I covalentbond of iodine. The polarization of ρ(r) along the X−I (X = C,N) covalent bond in these molecules increases fromiodobenzene to NIS and NISac, as shown by theircorresponding Laplacian values at bcp’s (∇2ρbcp = −1.07,+2.40 and +2.72 eÅ−5, respectively). Along the same series, theelectron density values remain closely similar (ρbcp = 0.84, 0.85,and 0.86 e Å−3), indicating that in the (C,N)−I bonding region,a similar amount of charge is progressively more depleted. Inaddition, the observed iodine net positive charge (+0.06, +0.39,and +0.43 e, respectively) shows the increase of its acidiccharacter along the same series (see also Table S1 and FiguresS3−S5). Hence, all ρ(r) properties together point to NISac asthe best halogen bond donor in the calculated series ofmolecules. The difference in the N−I distance from NIS toNISac optimized complexes [Δdcomplex(N−I) = dNIS − dNISac =

0.013 Å] is opposite in sign and significantly different than thatexperimentally observed [Δdcrystal(N−I) = dNIS − dNISac =−0.146 Å]. Furthermore, as previously pointed out, whiled(N···I) > d(I···N′) is true for NISac•Py-NMe2, the opposite isfound for NIS•Py-NMe2. These features seem to indicate thatiodine has been transferred and should be therefore associatedwith the Py-NMe2 molecule in the crystal phase of the NISaccomplex, while in the case of NIS, even though iodine is farfrom the nitrogen atom [ΔdNIS(N−I) = dcrystal − dmonomer =0.087 Å], the intermolecular distances indicate that the transferis not already done. The topological properties ρbcp and ∇2ρbcpat both N···I and I···N′ bonding interactions parallel theprevious conclusion (Figure 5). Indeed, whereas the relativevariation of those quantities in the NISac•Py-NMe2 complexare respectively of +14.5% and +28.4% from N···I to I···N′(both simultaneous increases point the strengthening of I···N′with respect to N···I, in a region of charge depletion that ischaracterized by ∇2ρ > 0), in the NIS•Py-NMe2 complex, thecorresponding values (−40.9% and −0.4%) indicate theopposite.118 In addition, from the isolated halogen bonddonors to the corresponding complexes, the topological valuescalculated at the N···I bcp indicate a relative variation of ρbcpand ∇2ρbcp of −38.5% and −13.1% for NISac, and of −16.1%and +8.4% for NIS. Thus, while the negative variation of ρbcpcorresponds to the weakening of the N−I bonding interactionfor both NISac and NIS, the negative variation of ∇2ρbcp forNISac and the positive one for NIS indicates that iodine isinvolved in a N···I interaction with pure-like closed-shellcharacter in NiSac only (both ρbcp and ∇2ρbcp magnitudesdecrease with the increase of the internuclear distance),pointing out a geometry where iodine could be alreadytransferred.The transfer of iodine in NISac•Py-NMe2 and the weakening

of the N···I bonding interaction only in NIS•Py-NMe2 isdefinitely supported by the value of the potential energy densityat bcp (Vbcp), which can be interpreted as the pressure exertedon the electrons to accumulate charge between nuclei at theinteratomic surface,119 and is therefore a quantitative measureof the bonding. Indeed, whereas Vbcp is significantly larger inmagnitude for N···I than for I···N′ in NIS•Py-NMe2 (−326 and−152 kJ/mol/a0

3, respectively), the opposite is observed forNISac•Py-NMe2 (−204 and −268 kJ/mol/a03, respectively), asfound in the case of the proton shift between two fluorineatoms in the [F···H···F]− system.118 It should be noted thatwhether transfer takes place (in NISac•Py-NMe2) or not (inNIS•Py-NMe2), both N···I and I···N′ interactions exhibit apartially covalent character, as measured by the 1 < |Vbcp|/Gbcp< 2 descriptor118 (Gbcp being the kinetic energy density at bcp)in both complexes (N···I and I···N′: 1.52 and 1.53, and 1.64and 1.37, for NISac•Py-NMe2 and NIS•Py-NMe2, respec-tively). The asymmetry of the Vbcp and |Vbcp|/Gbcp values forN···I and I···N′ interactions in NISac•Py-NMe2 is reduced withrespect to those in NIS•Py-NMe2, pointing that iodine transferhas just been taking place in the former complex while iodine isclearly trapped in an intermediate position of a starting transferprocess, yet belonging to NIS, in the latter. In conclusion, thesecalculations on isolated halogen-bonded complexes extractedfrom the crystal phases show that the NISac•Py-NMe2 complexcan indeed be described as an ionic salt with a charge transfer ofq = 0.68e, whereas NIS•Py-NMe2 should still be ratherconsidered as a cocrystal with q = 0.16e.In order to analyze the influence of the crystal environment

on the transfer of iodine, geometry optimizations were also

Figure 5. Laplacian maps derived from PBEPBE-D2 aug-cc-pVTZcalculations in the Ndonor···I···Nacceptor planes for (a) NISac•Py-NMe2and (b) NIS•Py-NMe2 isolated complexes at frozen experimentalgeometries. Topological QTAIM charges (Q) are indicated for thevarious fragments. At the Ndonor···I and I···Nacceptor bcp’s (green dots)the topological values are (a) ρ = 0.53 and 0.61 eÅ−3, and ∇2ρ = 2.36and 3.03 eÅ−5; and (b) ρ = 0.71 and 0.42 eÅ−3, and ∇2ρ = 2.60 and2.59 eÅ−5. Positive Laplacian contours are shown as continuous bluelines and negative contours as dashed red lines.

Crystal Growth & Design Article

DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

3468

performed using plane wave DFT periodic calculations120

(CASTEP 7.01), either by atom position optimization in a fixedunit cell or with simultaneous optimizations of unit cell andatom positions. The PBE functional,121 completed with D2Grimme dispersion correction,122 was used, leading tocalculation parameters similar to those used for the study ofisolated complexes (see above). With or without celloptimization, results are very close to each other (detailedvalues are given in Table S2 of SI). The main conclusions arecollected in Table 4 and show that such calculations now

reproduce very well the difference between the “close toneutral” NIS•Py-NMe2 system and the “close to ionic”NISac•Py-NMe2 system, demonstrating unambiguously thatthe solid state organization within the crystal plays a crucial rolein the actual charge delocalization within these systems.In order to further check the crystal environment effect on

the stabilization of the iodine transfer, iodine positionoptimizations (all other atoms being kept fixed) were alsoperformed by applying the plane wave basis set formalism to apseudo-isolated NISac•Py-NMe2 complex. With this aim, ahypothetical crystal composed of a unique halogen-bondedcomplex centered in a rectangular unit cell of increasingparameter values was considered. It corresponds to ahypothetical solid where halogen-bonded complexes are pushedapart from each other within a periodic system (see SI for fulldetails). As shown in Table S3, the increase of the distancebetween the halogen-bonded complexes leads to a shift of theiodine atom back to the imide side, converging toward the samegeometry observed in the molecular calculations of the isolatedcomplex (see above), therefore indicating that this geometricalcharacteristic is not an artifact of the computational methodused. It is noteworthy that, using similar periodic calculationswith the largest unit cell considered, the application of anelectric field along the molecular axis (ε = 2.79 GV m−1, Figure4) is again able to recover the experimental position of iodine.A clear illustration of the effects of the crystal packing and

environment on the iodine position in NISac•Py-NMe2 is givenin Figure 6, which shows the evolution of the calculated N···Idistance vs the expansion factor f of the unit cell. Thesecalculations started with the optimized crystal structure ofNISac•Py-NMe2 and then the original a, b, and c unit-cellparameters were multiplied by the same f value. In suchexpansions, the halogen-bonded complexes are progressivelypushed away from each other, whereas the donor and acceptorentities within each complex are kept mutually fixed (see SI for

full details). Thus, while f = 1 corresponds to the actual periodicsituation with the iodine atom closer to the acceptor moleculeat the crystal geometry, for f = 3 the situation is very close tothat found in the case of the isolated bimolecular complexcalculations with the iodine atom closer to the halogen bonddonor molecule. This trend unambiguously demonstrates thestrong influence of the molecular packing or, conversely, thehigh sensitivity of the iodine position toward the molecularenvironment in this system.Looking at the crystal structures, one can see (Figure 7) that

the iodine atom of NISac is also engaged in a contact with one

oxygen atom of an inversion-related NISac molecule. Thisadditional OCO···I contact may be a key point in theenvironmentally assisted iodine atom shifting toward Py-NMe2. Albeit, the structural characteristics of this “secondaryinteraction” are not very favorable [d(OCO···I) = 3.632 Å vsΣvdW = 3.5 Å and α(NPy‑NMe2···I···OCO) = 58.70°], theintegrated atomic charges derived from isolated molecules[Q(OCO) = −1.07; Q(I) = +0.43] show that an attractiveelectrostatic interaction can indeed occur between these twoatoms, the oxygen atom assisting the displacement of iodinefrom NISac toward Py-NMe2. In the NIS•Py-NMe2 crystalstructure, such a OCO···I contact is also present but in a leastfavorable way: the charge of the iodine atom is lower [Q(I) =+0.39] and the larger contact angle is less effective for OCO toattract the I atom [d(OCO···I) = 3.622 Å; α(NPy‑NMe2···I···OCO) = 74.61°].123

■ CONCLUSIONThe distinction between cocrystals and salts in crystallinehydrogen-bonded systems, formulated as A−H···B ⇆ [A]−···[H−B]+, has been addressed here in analogous halogen-bondedsystems, i.e., A−I···B ⇆ [A]−···[I−B]+, through structural andtheoretical investigations of the solid state associations between

Table 4. Optimized Structural Characteristics of NIS•Py-NMe2 and NISac•Py-NMe2 Based on Periodic DFTCalculationsa

N−I (Å) vs I···N′ (Å) N···N′ (Å)NIS···Py-NMe2Optimization A 2.214 < 2.360 4.574Optimization B 2.214 < 2.373 4.587Experimental 2.146(4) < 2.407(4) 4.553(6)NISac•Py-NMe2Optimization A 2.314 > 2.247 4.560Optimization B 2.312 > 2.253 4.564Experimental 2.292(2) > 2.218(2) 4.509(4)

aOptimization A: Unit-cell and structural parameters are refined.Optimization B: Unit-cell is fixed while structural parameters arerefined.

Figure 6. NNISac···I bond distance as a function of the expansion factorapplied to the NISac•Py-NMe2 crystal unit cell.

Figure 7. Geometry of the secondary intermolecular OCO···I contactin NISac•Py-NMe2.

Crystal Growth & Design Article

DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

3469

N-iodoimide derivatives as halogen bond donors, and pyridinesas halogen-bond acceptors. We have shown that the crystallinehalogen-bonded complexes of N-iodosuccinimide (NIS) withpyridine, 4-methylpyridine, and 4-dimethylaminopyridine canbe described as “close-to-neutral” cocrystals, while thecrystalline halogen-bonded complex of N-iodosaccharin(NISac) with 4-dimethylaminopyridine adopts a “close-to-ionic” salt structure. An estimation of the charge transferbetween the moieties is deduced from theoretical calculationswith isolated NIS•Py-NMe2 and NISac•Py-NMe2 complexes atfrozen experimental geometries to amount to qtransfer = 0.16 and0.68 e, respectively. Theoretical calculations and topologicalanalysis of the electron density distribution ρ(r) demonstrateunambiguously that the intermolecular interactions betweenneighboring molecules in the crystal actually play a crucial rolein the stabilization of the “close-to-ionic” structure of theNISac···Py-NMe2 complex. An external homogeneous electricfield ε applied along the N···N′ direction of the NISac•Py-

NMe2 complex in either gas phase or in periodic pseudo-isolated configuration is able to shift the iodine atom at thecrystal geometry (ε = 3.7 or 2.8 GV m−1, respectively), mimingthe polarization effect induced by the local crystal electric field.The linear dependence of the distance Ndonor···I with ε clearlyindicates the quantitative effect of a polarizing environment onthe I atom position, leading to the salt formation. Finally, thestrong influence of the crystalline environment on the iodineposition has been demonstrated by using plane wave DFTperiodic calculations on optimized NIS•Py-NMe2 andNISac•Py-NMe2 structures, as well as by applying the planewave basis set formalism to a hypothetical solid where thehalogen-bonded complexes are pushed apart from each otherwithin a periodic system.

■ EXPERIMENTAL SECTIONCrystal Growth. Starting materials are commercially available and

were used as received.

Table 5. Crystallographic Data

aTwinned data (see cif file for details).

Crystal Growth & Design Article

DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

3470

NIS•Py. NIS (5 mg, 0.02 mmol) was dissolved in CH2Cl2 (2 mL).Pyridine (0.003 mL, 2.9 mg, 0.03 mmol) was added to the solution.The solution was filtered to remove nondissolved particles. Crystalswere obtained by slow evaporation over 24 h. White crystals wereobtained. Mp 128 °C. Elem. Anal. Calcd for C9H9IN2O2: C 35.5, H2.98, N 9.21%. Found C 33.6, H 2.77, N 8.49%.NIS•Py-Me. As above from NIS (5 mg, 0.02 mmol) and 4-picoline

(0.02 mL, 2 mg, 0.02 mmol) in CH2Cl2. Mp 93−94 °C. Elem. Anal.Calcd for C10H11IN2O2: C 37.76, H 3.49, N 8.91%. Found C 39.24, H3.69, N 9.01%.NIS•Py-NMe2. NIS (13 mg, 0.057 mmol) and 4-dimethylaminopyr-

idine (7 mg, 0.057 mmol) were dissolved in ethyl acetate and thesolution was left to evaporate over 1 week. White crystals werecollected by filtration. Mp 85 °C. Elem. Anal. Calcd for C11H14IN3O2:C 38.06, H 4.06, N 12.1%. Found C 34.5, H 3.7, N 10.4%.NISac•Py. Slight modification of the previously described

procedure:92 N-iodosaccharin (5 mg, 0.016 mmol) was dissolved in1 mL of ethyl acetate. Pyridine (0.002 mL, 1.9 mg, 0.02 mmol) wasadded and the mixture was left for slow evaporation for 1 week. Mp152 °C. Elem. Anal. Calcd for C12H9IN2 O3S: C 37.13, H 2.34, N7.22%. Found: C 37.51, H 2.23, N 7.02%.NISac•Py-Me. The solution was made from N-iodosaccharin (10

mg, 0.03 mmol) dissolved in CH2Cl2 (3 mL) and 4-picoline (0.003mL, 0.03 mmol). The crystals were obtained by slow evaporation after1 week. Mp 110 °C. Elem. Anal. Calcd for C13H11IN2 O3S: C 38.82, H2.76, N 6.96%. Found: C 42.77, H 3.18, N 7.07%.NISac•Py-NMe2. A solution of N-iodosaccharin (5 mg, 0.016

mmol) in ethyl acetate (0.5 mL) was transferred in a long thin tube(internal diameter 5 mm), layered with pure ethyl acetate (0.5 mL)and then with a solution of 4-dimethylaminopyridine (1.9 mg, 0.015mmol) dissolved in ethyl acetate (0.2 mL). Thin needle-like crystalswere formed by slow diffusion. Mp 173 °C. Elem. Anal. Calcd forC14H14 IN3O3S: C 38.99, H 3.27, N 9.74%. Found: C 47.16, H 4.68, N8.65%.X-ray Crystallography. X-ray crystal structure collections were

performed on a Nonius FR590 diffratometer or on an APEXII Bruker-AXS diffractometer equipped with a CCD camera and a graphite-monochromated Mo Kα radiation source (λ = 0.71073 Å). Details ofthe structural analyses are summarized in Table 5. Absorptioncorrections were performed with SADABS. Structures were solvedby direct methods using the SIR97 program,124 and then refined withfull-matrix least-squares methods based on F2 (SHELXL-97)125 withthe aid of the WINGX program.126 All non-hydrogen atoms wererefined with anisotropic atomic displacement parameters. H atomswere finally included in their calculated positions.Theoretical Calculations. Density Functional Theory calculations

on isolated NISac•Py-NMe2 and NIS•Py-NMe2 complexes wereperformed using Gaussian09 software (revision D.01; Frisch et al.,2013).127 Geometry of the XB donor•acceptor couple was taken fromthe experimental X-ray structures, and hydrogen atoms were moved tostandard neutron distances. The triple-ζ quality basis set aug-cc-pVTZaugmented with diffuse functions was used for all elements; in the caseof iodine atom the core electrons were modeled through thecorresponding pseudopotential.128 Different functionals were used inthe framework of these DFT calculations, as detailed in SupportingInformation. Periodic Density Functional Theory calculations wereperformed within the Castep code (v 7.0.1)120 using plane wave basisand built-in ultrasoft pseudopotentials. The PBE functional wasused,121 as completed with the semiempirical dispersion correction ofGrimme.122 All details are available in SI.

■ ASSOCIATED CONTENT*S Supporting InformationDetails on the theoretical calculations and coordinates ofoptimized structures, cif files for the single crystal X-raydiffraction experiments. Crystallographic information files arealso available from the Cambridge Crystallographic DataCenter (CCDC) upon request (http://www.ccdc.cam.ac.ukCCDC deposition numbers 1060256−1060262). The Support-

ing Information is available free of charge on the ACSPublications website at DOI: 10.1021/acs.cgd.5b00535.

■ AUTHOR INFORMATIONCorresponding Authors*E-mail: [email protected].*E-mail: [email protected] authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was supported at its start by ANR (France) undercontract n°ANR-08-BLAN-0091-02. The authors thank theCINES/CEA CCRT for allocation of computing time (projectc2015087449 and the CDIFX (Rennes, France) for access to X-ray facilities.

■ REFERENCES(1) Desiraju, G. R. CrystEngComm 2003, 5, 466−467.(2) Dunitz, J. D. CrystEngComm 2003, 5, 506−506.(3) Aitipamula, S.; Banerjee, R.; Bansal, A. K.; Biradha, K.; Cheney,M. L.; Choudhury, A. R.; Desiraju, G. R.; Dikundwar, A. G.; Dubey, R.;Duggirala, N.; Ghogale, P. P.; Ghosh, S.; Goswami, P. K.; Goud, N. R.;Jetti, R. R. K. R.; Karpinski, P.; Kaushik, P.; Kumar, D.; Kumar, V.;Moulton, B.; Mukherjee, A.; Mukherjee, G.; Myerson, A. S.; Puri, V.;Ramanan, A.; Rajamannar, T.; Reddy, C. M.; Rodriguez-Hornedo, N.;Rogers, R. D.; Guru Row, T. N.; Sanphui, P.; Shan, N.; Shete, G.;Singh, A.; Sun, C. C.; Swift, J. A.; Thaimattam, R.; Thakur, T. S.;Thaper, R. K.; Thomas, S. P.; Tothadi, S.; Vangala, V. R.; Variankaval,N.; Vishweshwar, P.; Weyna, D. R.; Zaworotko, M. J. Cryst. GrowthDes. 2012, 12, 2147−2152.(4) Aakeroy, C. B.; Salmon, D. J. CrystEngComm 2005, 7, 439−448.(5) Bond, A. D. CrystEngComm 2007, 9, 833−834.(6) Huang, K. S.; Britton, D.; Etter, M. C.; Byrn, S. R. J. Mater. Chem.1997, 7, 713−720.(7) Bhogala, B. R.; Basavoju, S.; Nangia, A. CrystEngComm 2005, 7,551−562.(8) Childs, S. L.; Stahly, G. P.; Park, A. Mol. Pharmaceutics 2007, 4,323−338.(9) Arunan, E.; Desiraju, G. R.; Klein, R. A.; Sadlej, J.; Scheiner, S.;Alkorta, I.; Clary, D. C.; Crabtree, R. H.; Dannenberg, J. J.; Hobza, P.;Kjaergaard, H. G.; Legon, A. C.; Mennucci, B.; Nesbitt, D. J. Pure Appl.Chem. 2011, 83, 1637−1641.(10) Steiner, T. Angew. Chem., Int. Ed. 2002, 41, 48−76 andreferences therein.(11) Laurence, C.; Brameld, K. A.; Graton, J.; Le Questel, J.-Y.;Renault, E. J. Med. Chem. 2009, 52, 4073−4086.(12) Metrangolo, P.; Meyer, F.; Pilati, P.; Resnati, G.; Terraneo, G.Angew. Chem., Int. Ed. 2008, 47, 6114−6127.(13) Fourmigue, M. Curr. Opin. Solid State Mater. Sci. 2009, 13, 36−45.(14) Legon, A. C. Phys. Chem. Chem. Phys. 2010, 12, 4543−4551.(15) Corradi, E.; Meille, S. V.; Messina, M. T.; Metrangolo, P.;Resnati, G. Angew. Chem., Int. Ed. 2000, 39, 1782−1786.(16) Clark, T.; Hennemann, M.; Murray, J. S.; Politzer, P. J. Mol.Model. 2007, 13, 291−296.(17) Politzer, P.; Murray, J. S.; Concha, M. C. J. Mol. Model. 2007, 13,643−650.(18) Murray, J. S.; Lane, P.; Politzer, P. Int. J. Quantum Chem. 2007,107, 2286−2292.(19) Politzer, P.; Murray, J. S.; Lane, P. Int. J. Quantum Chem. 2007,107, 3046−3052.(20) Mohajeri, A.; Pakiari, A. H.; Bagheri, N. Chem. Phys. Lett. 2009,467, 393−397.(21) Desiraju, G. R.; Ho, P. S.; Kloo, L.; Legon, A. C.; Marquardt, R.;Metrangolo, P.; Politzer, P.; Resnati, G.; Rissanen, K. Pure Appl. Chem.2013, 85, 1711−1713.

Crystal Growth & Design Article

DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

3471

(22) Hassel, O.; Roemming, C. Q. Rev., Chem. Soc. 1962, 16, 1−18.(23) Bent, H. A. Chem. Rev. 1968, 68, 587−648.(24) Tsuzuki, S.; Wakisaka, A.; Ono, T.; Sonoda, T. Chem.Eur. J.2012, 18, 951−960.(25) Lieffrig, J.; Jeannin, O.; Frąckowiak, A.; Olejniczak, I.; Swietlik,R.; Dahaoui, S.; Aubert, E.; Espinosa, E.; Auban-Senzier, P.;Fourmigue, M. Chem.Eur. J. 2013, 19, 14804−14813.(26) Cabot, R.; Hunter, C. A. Chem. Commun. 2009, 2005−2007.(27) Sarwar, M. G.; Dragisic, B.; Salsberg, L. J.; Gouliaras, C.; Taylor,M. S. J. Am. Chem. Soc. 2010, 132, 1646−1653.(28) Metrangolo, P.; Resnati, G., Eds. Halogen Bonding: Fundamentalsand Applications; Springer-Verlag: Berlin, 2008.(29) Cavallo, G.; Metrangolo, P.; Pilati, T.; Resnati, G.; Sansotera,M.; Terraneo, G. Chem. Soc. Rev. 2010, 39, 3772−3783.(30) Politzer, P.; Murray, J. S.; Clark, T. Phys. Chem. Chem. Phys.2010, 12, 7748−7757.(31) Cavallo, G.; Metrangolo, P.; Pilati, T.; Resnati, G.; Terraneo, G.Cryst. Growth Des. 2014, 14, 2697−2702.(32) Pennington, W. T.; Resnati, G.; Taylor, M. S. CrystEngComm2013, 15, 3057−3057.(33) Meyer, F.; Dubois, P. CrystEngComm 2013, 15, 3058−3071.(34) Metrangolo, P.; Pilati, T.; Terraneo, G.; Biella, S.; Resnati, G.CrystEngComm 2009, 11, 1187−1196.(35) Lieffrig, J.; Jeannin, O.; Fourmigue, M. J. Am. Chem. Soc. 2013,135, 6200−6210.(36) Mukherjee, A.; Tothadi, S.; Desiraju, G. R. Acc. Chem. Res. 2014,47, 2514−2524.(37) Yamamoto, H. M.; Yamaura, J.-I.; Kato, R. J. Am. Chem. Soc.1998, 120, 5905−5913.(38) Yamamoto, H. M.; Kosaka, Y.; Maeda, R.; Yamaura, J.-i.; Nakao,A.; Nakamura, T.; Kato, R. ACSNano 2008, 2, 143−155.(39) Fourmigue, M.; Batail, P. Chem. Rev. 2004, 104, 5379−5418.(40) Fourmigue, M. Struct. Bonding (Berlin) 2008, 126, 181−207.(41) Shin, K.-S.; Jeannin, O.; Brezgunova, M.; Dahaoui, S.; Aubert,E.; Espinosa, E.; Auban-Senzier, P.; Swietlik, R.; Frąckowiak, A.;Fourmigue, M. Dalton Trans. 2014, 43, 5280−5291.(42) Devic, T.; Evain, M.; Moelo, Y.; Canadell, E.; Senzier, P.;Fourmigue, M.; Batail, P. J. Am. Chem. Soc. 2003, 125, 3295−3301.(43) Herve, K.; Cador, O.; Golhen, S.; Costuas, K.; Halet, J.-F.;Shirahata, T.; Muto, T.; Imakubo, T.; Miyazaki, A.; Ouahab, L. Chem.Mater. 2006, 18, 790−797.(44) Clemente-Juan, J. M.; Coronado, E.; Minguez Espallargas, G.;Adams, H.; Brammer, L. CrystEngComm 2010, 12, 2339−2342.(45) Nassirinia, N.; Amani, S.; Teat, S. J.; Roubeau, O.; Gamez, P.Chem. Commun. 2014, 50, 1003−1005.(46) Chen, Y. J.; Yu, H. F.; Zhang, L. Y.; Yang, H. A.; Lu, Y. F. Chem.Commun. 2014, 50, 9647−9649.(47) Saccone, M.; Terraneo, G.; Pilati, T.; Cavallo, G.; Priimagi, A.;Metrangolo, P.; Resnati, G. Acta Crystallogr. 2014, B70, 149−156.(48) Nguyen, H. L.; Horton, P. N.; Hursthouse, M. B.; Legon, A. C.;Bruce, D. W. J. Am. Chem. Soc. 2004, 126, 16−17.(49) McAllister, L. J.; Prasang, C.; Wong, J. P.-W.; Thatcher, R. J.;Whitwood, A. C.; Donnio, B.; O’Brien, P.; Karadakov, P. B.; Bruce, D.W. Chem. Commun. 2013, 49, 3946−3948.(50) Houbenov, N.; Milani, R.; Poutanen, M.; Haataja, J.;Dichiarante, V.; Sainio, J.; Ruokolainen, J.; Resnati, G.; Metrangolo,P.; Ikkala, O. Nature Commun. 2014, 5 (4043), 1−8.(51) Virkki, M.; Tuominen, O.; Forni, A.; Saccone, M.; Metrangolo,P.; Resnati, G.; Kauranen, M.; Priimagi, A. J. Mater. Chem. C 2015, 3,3003−3006.(52) Jungbauer, S. H.; Bulfield, D.; Kniep, F.; Lehmann, C. W.;Herdtweck, E.; Huber, S. M. J. Am. Chem. Soc. 2014, 136, 16740−16743.(53) Bruckmann, A.; Pena, M. A.; Bolm, C. Synlett 2008, 6, 900−902.(54) Walter, S. M.; Kniep, F.; Herdtweck, E.; Huber, S. M. Angew.Chem., Int. Ed. 2011, 50, 7187−7191.(55) Kniep, F.; Jungbauer, S. H.; Zhang, Q.; Walter, S. M.; Schindler,S.; Schnapperelle, I.; Herdtweck, E.; Huber, S. M. Angew. Chem., Int.Ed. 2013, 52, 7028−7032.

(56) Jungbauer, S. H.; Schindler, S.; Kniep, F.; Walter, S. M.; Rout,L.; Huber, S. M. Synlett 2013, 24, 2624−2628.(57) Jungbauer, S. H.; Walter, S. M.; Schindler, S.; Rout, L.; Kniep,F.; Huber, S. M. Chem. Commun. 2014, 50, 6281−6284.(58) He, W.; Ge, Y.-C.; Tan, C.-H. Org. Lett. 2014, 16, 3244−3247.(59) Auffinger, P.; Hays, F. A.; Westhof, E.; Ho, P. S. Proc. Natl. Acad.Sci. U.S.A. 2004, 101, 16789−16794.(60) Trogdon, G.; Murray, J. S.; Concha, M. C.; Politzer, P. J. Mol.Model. 2007, 13, 313−318.(61) Voth, A. R.; Hays, F. A.; Ho, P. S. Proc. Natl. Acad. Sci. U.S.A.2007, 104, 6188−6193.(62) Voth, A. R.; Khuu, P.; Oishi, K.; Ho, P. S. Nat. Chem. 2009, 1,74−79.(63) Laurence, C.; Graton, J.; Berthelot, M.; El Ghomari, M. J.Chem.Eur. J. 2011, 17, 10431−10444.(64) Le Questel, J.-Y.; Laurence, C.; Graton, J. CrystEngComm 2013,15, 3212−3221.(65) Mukherjee, A.; Tothadi, S.; Desiraju, G. R. Acc. Chem. Res. 2014,47, 2514−2524.(66) Zapata, F.; Caballero, A.; Molina, P.; Akorta, I.; Elguero, J. J.Org. Chem. 2014, 79, 6959−6969.(67) Lemouchi, C.; Yamamoto, H. M.; Kato, R.; Simonov, S.; Zorina,L.; Rodriguez-Fortea, A.; Canadell, E.; Wzietek, P.; Iliopoulos, K.;Gindre, D.; Chrysos, M.; Batail, P. Cryst. Growth Des. 2014, 14, 3375−3383.(68) Gilday, L. C.; Beer, P. D. Chem.Eur. J. 2014, 20, 8379−8385.(69) Takemura, A.; McAllister, L. J.; Hart, S.; Pridmore, N. E.;Karadakov, P. B.; Whitwood, A. C.; Bruce, D. W. Chem.Eur. J. 2014,20, 6721−6732.(70) Awwadi, F. F.; Taher, D.; Haddad, S. F.; Turnbull, M. M. Cryst.Growth Des. 2014, 14, 1961−1971.(71) Aakeroy, C. B.; Wijethunga, T. K.; Haj, M. A.; Desper, J.;Moore, C. CrystEngComm 2014, 16, 7218−7225.(72) Aakeroy, C. B.; Baldrighi, M.; Desper, J.; Metrangolo, P.;Resnati, G. Chem.Eur. J. 2013, 19, 16240−16247.(73) Aakeroy, C. B.; Panikkattu, S.; Chopade, P. D.; Desper, J.CrystEngComm 2013, 15, 3125−3136.(74) Koskinen, L.; Jaaskelainen, S.; Hirva, P.; Haukka, M. Cryst.Growth Des. 2015, 15, 1160−1167.(75) Ouvrard, C.; Le Questel, J.-Y.; Berthelot, M.; Laurence, C. ActaCrystallogr. 2003, B59, 512−526.(76) Aragoni, M. C.; Arca, M.; Devillanova, F. A.; Isaia, F.; Lippolis,V.; Mancini, A.; Pala, L.; Slawin, A. M. Z.; Woollins, J. D. Chem.Commun. 2003, 2226−2227.(77) Aragoni, M. C.; Arca, M.; Demartin, F.; Devillanova, F. A.;Garau, A.; Isaia, F.; Lelj, F.; Lippolis, V.; Verani, G. Chem.−Eur. J.2001, 7, 3122−3133.(78) Batsanov, A. S.; Howard, J. A. K.; Lightfoot, A. P.; Twiddle, S. J.R.; Whiting, A. Eur. J. Org. Chem. 2005, 1876−1883.(79) Romming, C. Acta Chem. Scand. 1972, 26, 1555−1560.(80) Aragoni, M. C.; Arca, M.; Devillanova, F. A.; Hursthouse, M. B.;Huth, S. L.; Isaia, F.; Lippolis, V.; Mancini, A.; Ogilvie, H. R.; Verani,G. J. Organomet. Chem. 2005, 690, 1923−1927.(81) Aragoni, M. C.; Arca, M.; Coles, S. J.; Devillanova, F. A.;Hursthouse, M. B.; Coles (nee Huth), S. L.; Isaia, F.; Lippolis, V.;Mancini, A. CrystEngComm 2011, 13, 6319−6322.(82) Carlsson, A.-C. C.; Grafenstein, J.; Laurila, J. L.; Bergquist, J.;Erdelyi, M. Chem. Commun. 2012, 48, 1458−1460.(83) Carlsson, A.-C. C.; Uhrbom, M.; Karim, A.; Brath, U.;Grafenstein, J.; Erdelyi, M. CrystEngComm 2013, 15, 3087−3092.(84) Carlsson, A.-C. C.; Grafenstein, J.; Budnjo, A.; Laurila, J. L.;Bergquist, J.; Karim, A.; Kleinmaier, R.; Brath, U.; Erdelyi, M. J. Am.Chem. Soc. 2012, 134, 5706−5715.(85) Georgiou, D. C.; Butler, P.; Browne, E. C.; Wilson, D. J. D.;Dutton, J. L. Aust. J. Chem. 2013, 66, 1179−1188.(86) Murray, J. S.; Macaveiu, L.; Politzer, P. J. Comput. Chem. 2014,5, 590−595.(87) For a review, see: Troff, R. W.; Makela, T.; Topi, F.; Valkonen,A.; Raatikainen, K.; Rissanen, K. Eur. J. Org. Chem. 2013, 1617−1637.

Crystal Growth & Design Article

DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

3472

(88) K. Padmanabhan, K.; Paul, I. C.; Curtin, D. Y. Acta Crystallogr.1990, C46, 88−92.(89) Ghassemzadeh, M.; Dehnicke, K.; Goesmann, H.; Fenske, D. Z..Naturforsch. 1994, B49, 602−608.(90) Raatikainen, K.; Rissanen, K. CrystEngComm 2011, 13, 6972−6977.(91) Castellote, I.; Moron, M.; Burgos, C.; Alvarez-Builla, J.; Martin,A.; Gomez-Sal, P.; Vaquero, J. J. Chem. Commun. 2007, 1281−1283.(92) Dolenc, D.; Modec, M. New J. Chem. 2009, 33, 2344−2349.(93) Bondi, A. J. Phys. Chem. 1964, 68, 441−451.(94) Roper, L. C.; Prasang, C.; Kozhevnikov, V. N.; Whitwood, A. C.;Karadakov, P. B.; Bruce, D. W. Cryst. Growth Des. 2010, 10, 3710−3720.(95) Prasang, C.; Whitwood, A. C.; Bruce, D. W. Cryst. Growth Des.2009, 9, 5319−5326.(96) Zerdane, S.; Toudic, B.; Collet, E. personal communication.(97) Brown, I. D. Acta Crystallogr. 1992, B48, 553−572.(98) Ramos, M.; Alkorta, I.; Elguero, J.; Golubev, N. S.; Denisov, G.S.; Benedict, H.; Limbach, H. H. J. Phys. Chem. A. 1997, 101, 9791−9800.(99) Del Bene, J. E.; Alkorta, I.; Elguero, J. J. Phys. Chem. A 2010,114, 12958−12962.(100) McAllister, L. J.; Bruce, D. W.; Karadakov, P. B. J. Phys. Chem.A 2011, 115, 11079−11086.(101) Banerjee, R.; Bhatt, P. M.; Ravindra, N. V.; Desiraju, G. R.Cryst. Growth Des. 2005, 5, 2299−2309.(102) Banerjee, R.; Saha, B. K.; Desiraju, G. R. CrystEngComm 2006,8, 680−685.(103) Sudhakar, P.; Kumar, S. V.; Vishweshwar, P.; Moses Babu, J.;Vyas, K. CrystEngComm 2008, 10, 996−1002.(104) Gou, G.-Z.; Kou, J.-F.; Zhou, Q.-D.; Chi, S.-M. Acta Crystallogr.2013, E69, o153−154.(105) Childs, S. L.; Stahly, G. P.; Park, A. Mol. Pharmacol. 2007, 4,323−338.(106) Jordan, M. J. T.; Del Bene, J. E. J. Am. Chem. Soc. 2000, 122,2101−2115.(107) Del Bene, J. E.; Jordan, M. J. T. J. Phys. Chem. A 2002, 106,5385−5392.(108) Rozas, I.; Alkorta, I.; Elguero, J. Chem. Phys. Lett. 1997, 275,423−428.(109) Mata, I.; Molins, E.; Alkorta, I.; Espinosa, E. J. Chem. Phys.2009, 130, 044104−1−011104−16.(110) Mata, I.; Alkorta, I.; Espinosa, E.; Molins, E. Chem. Phys. Lett.2011, 507, 185−189.(111) Mata, I.; Molins, E.; Alkorta, I.; Espinosa, E. J. Phys. Chem. A2011, 115, 12561−12571.(112) Spackman, M. A.; Munshi, P.; Jayatilaka, D. Chem. Phys. Lett.2007, 443, 87−91.(113) Kahn, R.; Lara, E. C. D.; M?ller, K. D. J. Chem. Phys. 1985, 83,2653−2660.(114) Aubert, E.; Porcher, F.; Souhassou, M.; Lecomte, C. J. Phys.Chem. Solids 2004, 65, 1943−1949.(115) Lehle, H.; Kriegl, J. M.; Nienhaus, K.; Deng, P.; Fengler, S.;Nienhaus, U. Biophys. J. 2005, 88, 1978−1990.(116) Suydam, I. T.; Snow, C. D.; Pande, V. S.; Boxer, S. G. Science2006, 313, 200−204.(117) Bader, R. F. W. Atoms in Molecules: A Quantum Theory; OxfordUniversity Press: Oxford, 1990.(118) Espinosa, E.; Alkorta, I.; Elguero, J.; Molins, E. J. Chem. Phys.2002, 117, 5529−5542.(119) Espinosa, E.; Molins, E.; Lecomte, C. Chem. Phys. Lett. 1998,285, 170−173.(120) Clark, S. J.; Segall, M. D.; Pickard, C. J.; Hasnip, P. J.; Probert,M. J.; Refson, K.; Payne, M. C. Z. Kristallogr. 2005, 220, 567−570.(121) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996,77, 3865−3868.(122) Grimme, S. J. Comput. Chem. 2006, 27, 1787−1799.(123) To further investigate the origin of this trend, we have alsoconsidered a hypothetical polymorph of NIS•Py-NMe2 having the

same P-1 space group symmetry than NISac•Py-NMe2. This crystalstructure was optimized with Castep and was found to be 29.6 kJmol−1 less stable (and less dense, by 6.6%) than the originalorthorhombic structure (Table S5). In that triclinic polymorph, theNNIS···I distance increased up to 2.256 Å, a value intermediate betweenthe bond distance observed in its orthorhombic crystal structure(2.214 Å) and in the NISac•Py-NMe2 triclinic crystal structure (2.314Å), whereas the NNIS···N′ distance is slightly shortened to 4.559 Å(corresponding values for NIS•Py-NMe2 and NISac•Py-NMe2 crystalstructures are 4.574 Å and 4.560 Å). In the triclinic polymorph ofNIS•Py-NMe2, the OCO···I contact is weakened compared toNISac•Py-NMe2, in line with the smaller elongation of the NNIS···Ibond distance (d(OCO···I) = 3.994 Å; α(NPy‑NMe2···I···OCO) =54.05°). This demonstrates again the influence of the molecularpacking on the positioning of the iodine atom in these systems.(124) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G.;Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;Spagna, R. J. Appl. Crystallogr. 1999, 32, 115−119.(125) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.(126) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837−838.(127) Frisch, M. J. et al. Gaussian 09, revision D.01; Gaussian, Inc.,Wallingford, CT, 2013.(128) Peterson, K. A.; Shepler, B. C.; Figgen, D.; Stoll, H. J. Phys.Chem. A 2006, 110, 13877−13883.

Crystal Growth & Design Article

DOI: 10.1021/acs.cgd.5b00535Cryst. Growth Des. 2015, 15, 3464−3473

3473