Molecular Nuances Governing the Self-Assembly of 1,3:2,4

Article pubs.acs.org/Langmuir

Molecular Nuances Governing the Self-Assembly of 1,3:2,4Dibenzylidene‑D‑sorbitol Andrew Singh,† France-Isabelle Auzanneau,‡ Maria G. Corradini,§ Girishma Grover,∥ Richard G. Weiss,∥ and Michael A. Rogers*,† †

Department of Food Science and ‡Department of Chemistry, University of Guelph, Guelph, Ontario N1G2W1, Canada § Department of Food Science, University of Massachusetts Amherst, Amherst, Massachusetts 01003, United States ∥ Department of Chemistry, Georgetown University, Washington, District of Columbia 20057-1227, United States S Supporting Information *

ABSTRACT: 1,3:2,4-Dibenzylidene-D-sorbitol (DBS) is the gold-standard for low-molecular-weight organogelators (LMOGs). DBS gels a wide array of solvents, as illustrated by the large Hansen sphere representing gels (2δd = 33.5 MPa1/2, δp = 7.5 MPa1/2, and δh = 8.7 MPa1/2; radius = 11.2 MPa1/2). Derivatives of DBS have been synthesized to isolate and determine molecular features essential for organogelation. In this work, π−π stacking and hydrogen bonding are the major noncovalent interactions examined. The importance of π−π stacking was studied using 1,3:2,4 dicyclohexanecarboxylidene-D-sorbitol (DCHS), which eliminates possible π−π stacking while still conserving the other structural aspects of DBS. The replacement of the benzyl groups with cyclohexyl groups led to a very a poor gelator; only one of the several solvents examined, carbon tetrachloride, formed a gel. 1,3:2,4-Diethylidene-D-sorbitol (DES), another DBS analogue incapable of π−π stacking but with very different polarity, gelated a large Hansen space (2δd = 34.0 MPa1/2, δp = 10.9 MPa1/2, and δh = 10.8 MPa1/2; radius = 9.2 MPa1/2). DES gels solvents with higher δp and δh values than DBS. To assess the role of hydrogen bonding, DBS was acetalated (A-DBS), and it was found that the Hansen space gelated by A-DBS shifted to less polar solvents with higher hydrogen-bonding Hansen solubility parameters (HSPs) (2δd = 33.8 MPa1/2, δp = 6.3 MPa1/2, and δh = 9.6 MPa1/2; radius = 11.1 MPa1/2) than for DBS. These systematic structural modifications are the first step in exploring how specific intermolecular features alter aspects of Hansen space corresponding to positive gelation outcomes.

■

intermolecular interactions such as hydrogen-bonding, π−π stacking, electrostatic interactions, and van der Waals interactions.4,15,16 These noncovalent interactions promote one-dimensional (1D) fibrillar growth, with the 1D fibers interacting to form temporary and permanent (i.e., formed by crystallographic mismatches giving rise to daughter fibers) junction zones, resulting in a self-assembled fibrillar network (SAFiN).17 The SAFiN entraps solvent molecules on a microscopic level, through noncovalent interactions, and on a macroscopic level, through capillary forces and surface tension, giving rise to solid-like rheological properties.18 These interactions can be disrupted by heating the gel above the sol−gel transition temperature, but they will re-form once the sol is cooled below the transition temperature. Hansen solubility parameters (HSPs) are used to quantify the potential of gelator−solvent combinations to form SAFiNs because of the interplay between the two components. HSPs,

INTRODUCTION Sorbitol-derived molecules have garnered attention as lowmolecular-mass organogelators (LMOGs) with prospective applications in the food, cosmetic, and pharmaceutical industries.1−5 Organogels made from LMOGs are thermoreversible and form gels at low concentrations.6 In addition, molecular gelators must be nontoxic for the aforementioned applications, making biocompatible carbohydrate-based gelators excellent candidates for medical and edible applications.3,7 Although sorbitol derivatives are under investigation, 1,3:2,4dibenzylidene-D-sorbitol (DBS) attracts extremely broad interest because it gels an unusually wide array of organic solvents.3,4,8,9 The gelation capacity of DBS has been attributed to its ability to make intermolecular hydrogen bonds between free hydroxyl groups and π−π stacks between aromatic groups, as well as its “butterfly-like” shape,10−13 whereby sorbitol acts as the body and the benzylidene rings act as wings. The hydrophobic benzylidene wings are responsible for DBS being slightly soluble in numerous organic solvents.14 Molecular gelators are driven to self-assemble into threedimensional (3D) networks consisting of fibers by noncovalent © XXXX American Chemical Society

Received: June 27, 2017 Revised: September 18, 2017 Published: September 19, 2017 A

DOI: 10.1021/acs.langmuir.7b02191 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir originally developed to identify solvents capable of dissolving polymers, have been adapted to the study of the assembly of small molecular gelators.19 Raynal and Bouteiller were the first to recognize that, barring a few exceptions, solvents gelled by a specific gelator have similar HSPs.19 Other approaches, such as the Hildebrand solubility parameters, were unable to account as well for the specific intermolecular interactions that drive selfassembly.19 A gelator must be both sufficiently soluble in a solvent that it does not precipitate from solution and sufficiently insoluble to facilitate self-assembly/crystallization, producing a gel.20−22 Because a gelator must balance noncovalent interactions between the solvent and other gelator molecules to produce a gel, molecular gelators capable of gelling solvents tend to reside within a confined, specific region of HSPs. HSPs can be used as effective a priori tools for predicting the abilities of small molecules to self-assemble into SAFiNs for specific solvents.19,23 HSPs are derived from the total cohesive energy density or the negative energy of vaporization per cm3 of sample, corresponding to the Hildebrand parameter (δt) (eq 1). HSPs contain three components: dispersive (2δd), polar (δp), and hydrogen-bonding (δh) forces24 δt 2 = δd 2 + δp2 + δ h 2

Scheme 1. Chemical Structures of Sorbitol Derivatives Including 1,3:2,4-Dibenzylidene-D-sorbitol (DBS), 1,3:2,4Dicyclohexanecarboxylidene-5,6-diacetyl-D-sorbitol (ADCHS), 1,3:2,4-Dicyclohexanecarboxylidene-5,6-diacetyl-Dsorbitol (A-DCHS), 1,3:2,4-Dicyclohexanecarboxylidene-Dsorbitol (DCHS), and 1,3:2,4-Di-O-ethylidene-D-sorbitol (DES)

(1)

where δd is the dispersive HSP, δp is the polar HSP, and δh is the hydrogen bonding HSP.25 A greater affinity between gelator and solvent exists when their Hansen space coordinates (eq 1) are similar, thus leading to the two compounds being miscible. Already, HSPs have provided tremendous insights into why gelators are capable of assembling in certain solvents and not others.19 However, only limited studies have used HSPs to explain which gelators will gel a solvent.19 In part, this is because the few studies have involved the systematic modification of the chemical structures of the gelator to observe the gelation outcome in a wide array of solvents.26,27 Although work has been amassed to determine how the molecular features of the LMOG influence the gel properties of DBS,28 there has yet to be a study examining how these modifications alter the Hansen space that corresponds to gel formation.

■

EXPERIMENTAL SECTION

Materials. 1,3:2,4-Dibenzylidene-D-sorbitol (98%, BOC Science, New York), mp 221.1 °C (Figure S1), and 1,3:2,4-di-O-ethylidene-Dsorbitol (DES) (95%, Sigma-Aldrich, Oakville, ON, Canada), mp 167.1 °C (Figure S1), were used as received. 1,3:2,4-Dibenzylidene5,6-diacetyl-D-sorbitol (A-DBS), 1,3:2,4-dicyclohexanecarboxylidene-Dsorbitol (DCHS), and 1,3:2,4-dicyclohexanecarboxylidene-5,6-diacetyl-D-sorbitol (A-DCHS) were synthesized as presented in the Supporting Information. The solvents used for gelation tests included salicylaldehyde (98%), dimethyl sulfoxide (≥99.9%), o-xylene (97%), isobutyl alcohol (≥99%), triethylene glycol (99%), butylamine (99.5%), hexanoic acid (≥99.5%), α,α-dichlorotoluene (≥95%), tetramethylurea (99%), carbon tetrachloride (≥99.5%), benzyl methacrylate (96%), acetophenone (99%), chloroform (≥99%,), triethylamine (≥99.5%), hexanes (≥98.5%), and benzene (≥99%) were obtained from Sigma-Aldrich, Oakville, ON, Canada. N,NDimethylformamide (≥99%) was obtained from Acros Organics (Morris Plains, NJ). Acetone (HPLC-grade), acetonitrile (HPLCgrade), toluene (≥99.5%), pyridine (≥99%), and methylene chloride (≥99.5%) were obtained from Fisher Scientific (Ottawa, ON, Canada). Ethyl alcohol (95%) was obtained from Commercial Alcohols (Brampton, ON, Canada). Gelation Tests. Five different derivatives of sorbitol were tested for their gelation abilities (Scheme 1). Each 5% wt gelator/solvent

combination was combined, and the solid was dissolved by heating the sample to 10 °C above its melting temperature in a 2 mL screw-top glass vial (Sigma-Aldrich, Oakville, ON, Canada) for 30 min until a transparent sol formed. The concentration of 5 wt % was selected to ensure that the concentration was above the critical gelator concentration of the gelator in each solvent; this ensured that no false negatives (i.e., formation of a solution when, at slightly higher concentrations, a gel would have been seen) were observed, which would dramatically effect the Hansen spheres. The vials were cooled to room temperature (20−22 °C) and allowed to stand for 24 h. Sample B


Article

Langmuir Table 1. Gelation Outcomesa of 5 wt % Sorbitol-Derived Gelators in a Wide Range of Solvents solvent

DBS

A-DBS

DES

DCHS

A-DCHS

acetonitrile acetone hexane pyridine dimethyl sulfoxide dimethylformamide benzene toluene o-xylene triethylamine ethanol hexanoic acid dichlorotoluene dichloromethane chloroform carbon tetrachloride tetramethylurea butylamine acetophenone salicylaldehyde triethylene glycol isobutyl alcohol benzyl methacrylate

gel gel precipitate solution solution solution gel gel gel gel gel gel gel gel gel gel solution solution gel gel gel gel gel

solution precipitate precipitate solution precipitate solution precipitate precipitate gel precipitate gel gel gel solution solution gel solution precipitate gel gel gel gel gel

gel precipitate precipitate gel solution gel precipitate precipitate precipitate precipitate gel solution precipitate precipitate precipitate precipitate precipitate solution gel gel gel precipitate precipitate

solution solution precipitate solution solution solution solution solution solution solution solution solution solution solution solution gel solution solution solution solution solution solution solution

precipitate solution solution solution solution precipitate solution solution solution solution precipitate solution solution solution solution solution solution solution solution solution gel solution solution

Samples were classified as gels when no flow was observed, solutions when they exhibited flow but were transparent, and precipitates when there was a crystalline mass at the bottom of the vial (i.e., they were macroscopically phase-separated) or they were opaque (i.e., they contained weakly interacting crystallites but, again, were macroscopically phase-separated). a

vials were then inverted for 30 min and examined to see if flow occurred. For our tabletop rheological approach, which is required because of the volume of samples, the materials must be crudely defined. As such, a precipitate, according to our designation, has a crystalline mass at the bottom of the vial (i.e., is macroscopically phase-separated) or is opaque (i.e., the crystallites are weakly interacting but, again, macroscopically phase-separated). Samples were classified as gels when no flow was observed and as solutions they exhibited flow but were transparent. Hansen Spheres. Hansen solubility parameters for the solvent were obtained using HSPiP software (Hansen-Solubility.com, Denmark), and the HSPs for the gelators were calculated using the group contribution methods in the aforementioned software package. A globally constrained optimization procedure in Mathematica 9 (Wolfram Research, Champaign, IL) was used to calculate minimal enclosing spheres that contained all of the points pertaining to each category (i.e., sol, gel, or precipitate). The optimization procedure was implemented using the “NMinimize” built-in function in Mathematica to obtain the sphere center in terms of Hansen coordinates while solving for the smallest possible radius. The NMinimize function was used to obtain the global optimization problem numerically, and a direct search method, differential evolution, was selected as the numerical algorithm to reach a numerical global optimum solution.29 The selection of this direct search method was based on its robustness despite being computationally more expensive.30 Four effective digits of precision were sought in the final results; these criteria were used to halt the iteration process. Two- and three-dimensional renditions of the resulting spheres were plotted. Optical Microscopy. A small portion of the gel was taken from a glass vial after 24 h of storage at room temperature and placed on a 75 mm × 25 mm glass microscope slide (Fisher Scientific, Ottawa, ON, Canada), and then a 25 mm × 25 mm glass coverslip (Fisher Scientific, Ottawa, ON, Canada) was placed on top of the sample. A Nikon Eclipse Ti−S inverted light microscope (Nikon Instruments, New York) equipped with a QIMAGING Retiga 2000R color camera (QImaging, Surrey, BC, Canada, and a Nikon Plan Apo 10X/0.45 DIC N1 lens (Nikon Instruments, New York) and a Nikon Plan Apo 40x/

0.95 DIC M/N2 40X lens (Nikon Instruments, New York) were used to acquire polarized light micrographs. Scanning Electron Microscopy. Scanning electron microscopy (SEM) of the xerogels was performed after the samples had been dried from acetonitrile. A small aliquot of the gelator−acetonitrile gel was placed on an SEM stub and placed into an oven (Fisher Scientific, Isotemp, Fair Lawn, NJ) at 35 °C for 10 min, allowing the acetonitrile to evaporate. The sample on the SEM stub was mounted on a sputter coater (Emscope K550 sputter coater, Ashford, Kent, U.K.) and coated with gold using a 20 mA deposition current and a 7 nm min−1 deposition rate for 2 min. The sample was then transferred to a specimen holder on the SEM stage (Hitachi S-570, Tokyo, Japan). Images were taken using Quartz PCI Imaging software (Quartz Imaging Corp., Vancouver, BC, Canada). Differential Scanning Calorimetry. For differential scanning calorimetry (DSC) measurements, 10−12 mg of each gelator was transferred into an Alod-Al hermetically sealed DSC pan. The DSC chamber (Q2000, TA Instruments, New Castle, DE) was precooled to 20 °C before the sample was placed into the chamber, which was continually flushed with nitrogen (0.5 mL/min). The samples were heated at 2 °C/min from 20 to 250 °C to determine the peak melting temperature, and then data storage was turned off before the DSC cell was cooled to 20 °C. X-ray Diffraction. A Rigaku multiflex powder X-ray diffractometer (Rigaku, Tokyo, Japan) with a 1/2° divergence slit, 1/2° scatter slit, and a 0.3-mm receiving slit was set at 40 kV and 44 mA to determine the polymorphic form of the network. Scans were performed from 1° to 30° with a 0.02° step at 1° min−1. Computational Modeling. The structures used for the calculations were optimized using Gaussian 16 software31 and the DFT/B3LYP32 method with the 6-31G33 basis set. There were no imaginary frequencies. The reported distances are between atoms at the extreme ends of the X, Y, and Z axes, to which the van der Waals radii of the terminal atoms have been added. The van der Waals radii used for hydrogen and oxygen are 1.2 and 1.52 Å, respectively.34 C


Article

Langmuir

■

RESULTS AND DISCUSSION Assessments of the gelation abilities of the five sorbitol-derived molecules revealed that the substitution of benzyl groups for cyclohexyl groups, eliminating π−π stacking and imparting steric impediments to highly ordered packing, prevents gelation in all solvents except for carbon tetrachloride (Table 1). By defining gelator−solvent combinations as solutions, precipitates, or gels and by selecting solvents to cover a wide range of HSPs, we sought an adequate representation of Hansen space for each outcome (Figure 1). The advantage of using this

coordinates for the center of the sphere and its radius that can be compared between gelators to observe the global effects on gelation. Qualitatively, the process of self-assembly in molecular gels is intricate and must balance parameters influencing solubility with the contrasting forces that govern epitaxial growth into axially symmetric elongated aggregates.16,35,36 Hansen solubility parameters can quantitatively define the balance required for these parameters, and the HSPs of the gelators were calculated using the group contribution method (Table 2).24,25 For example, when DBS (2δd = 35.3 MPa1/2, δp Table 2. HSPs of the Various Gelators Calculated with the Group Contribution Method Using HSPiP Software gelator

δd (MPa1/2)

δp (MPa1/2)

δh (MPa1/2)

DBS A-DBS DCHS A-DCHS DES

17.6 17.6 16.8 16.9 17

8.3 7.1 7.4 6.4 9.6

10.1 5.8 8.5 4.8 13.4

= 8.3 MPa 1/2 , and δ h = 10.1 MPa 1/2 ) is added to tetramethylurea (2δd = 33.4 MPa1/2, δp = 8.2 MPa1/2, and δh = 11.0 MPa1/2), a solution is obtained. The distance between two molecules of differeng types (Rij) can be calculated using the equation R ij =

4(δdi − δdj)2 + (δpi − δpj)2 + (δ hi − δ hj)2

(2)

where i represents the HSPs for the gelator and j represents those for the solvent and the 4 before the dispersive component allows for a spherical fit. For DBS in tetramethylurea, the Rij value is 2.01 MPa1/2. When the solvent and potential gelator are very close in Hansen space, they have similar interactions, and the gelator is typically solvated by the solvent. Conversely, when DBS is added to hexane (2δd = 29.8 MPa1/2, δp = 0 MPa1/2, and δh = 0 MPa1/2), the Rij value is 14.1 MPa1/2, and DBS precipitates from solution. Gelation, which requires a meticulous balance between contrasting parameters including solubility and the intermolecular forces that control epitaxial growth into axially symmetric elongated aggregates, occurs at intermediate Rij values. Of the gelators tested, DBS gelled the most solvents (17 of 23 solvents) and had the largest gelation sphere, with a radius of 11.2 MPa1/2 (Table 3). The distance between the center of the solubility sphere and the Hansen coordinates for DBS is 1.72 MPa1/2, indicating that, when the solvent and gelator are in close proximity in Hansen space, the solvation of the gelator results. Clearly, for DBS, the solution sphere is confined within the gelation sphere, and the only solvent that resulted in a precipitate was located just outside the gelation sphere (Figure 1 and Figure S18). Compared to DBS (2δd = 35.3 MPa1/2, δp = 8.3 MPa1/2, and δh = 10.1 MPa1/2), DES (2δd = 34 MPa1/2, δp = 9.6 MPa1/2, and δh = 13.4 MPa1/2) was the most dissimilar of the tested potential gelators. In the case of DES (Figure 1 and Figure S19), the center of the solution sphere is located 3.71 MPa1/2 from the coordinates of DES. Unlike for DBS, for which the precipitate was located outside the solution and gel spheres, the solution sphere for DES is enclosed within the gelation sphere, but the precipitate sphere and the gelation sphere have considerable overlap (Figure 1). Because there was considerable overlap between the spheres, we examined the directionality of the solvent relative to the gelator to observe

Figure 1. Three-dimensional Hansen space for solution spheres (blue), precipitate spheres (green), and gelation spheres (red) for sorbitol-derived gelators.

approach is that it provides an encompassing view of how these potential gelators behave globally. To illustrate this point, if carbon tetrachloride had been the only solvent examined, then it would have been concluded that DBS, A-DBS, and DCHS were all good gelators whereas A-DCHS and DES were not able to form SAFiNs, which does not adequately represent the global findings for these molecular changes. Each gelator was tested in 23 solvents for the gelation outcomes (i.e., solution, precipitate, or gel), and the results were plotted in Hansen space (Figure 1). The HSPs of the solvents were used to create the minimal enclosing spheres for each outcome. This method results in the generation of D


Article

Langmuir

In trying to determine why considerable overlap between the spheres is observed for DES, we plotted the differences (e.g., δp(gelator) − δp(solvent)) between the Hansen coordinates (Figure 2). Unlike Rij calculations, calculations of the difference in Hansen coordinates preserves the directionality of the tensor distance. For example, solvents that are below or to the left of the (0,0) value have a greater HSP than the gelator. The differences in Hansen coordinates between the solvents and DBS (Figure 2A−C) illustrate the expected result [i.e., solvents that are close to (0,0) in Figure 2 for DBS result in solutions, those at intermediate distances form gels, and the most distant points represent precipitates]. DES does not show the same trend (Figure 2D−F); specifically, Figure 2D illustrates that gels result only when the solvent point is below Δδp = 0 MPa1/2 whereas precipitate states exist above this line. In comparing the intermolecular interactions that would drive DBS to gel versus DES, DBS can utilize both hydrogen-bonding and π−π stacking interactions, whereas DES relies only on hydrogen bonding. It can be concluded that DES will gel only in solvents that are more polar than the gelator; however, the exact reason for this behavior has not been elucidated. Modifying DBS by acetylating the primary and secondary hydroxyl groups did not have a major effect on the gel Hansen sphere (Figure 1, Figure S20). The radius of the gelation sphere was 11.1 MPa1/2 compared to a radius of 11.2 MPa1/2 for DBS. Although acetylating DBS did not have a major effect on the size of the Hansen space, it did shift the gelation sphere to a lower δp value and a greater δh value (Table 3). Upon acetylation, A-DBS could no longer gel benzene, toluene,

Table 3. Coordinates of the Centers of the Gelation, Solution, and Precipitate Spheres in Hansen Space and Their Radii outcome

2δd (MPa1/2)

precipitate gel sol

NA 33.5 34.6

precipitate gel sol

33.3 33.8 33.1

precipitate gel sol

NA NA 33.3

precipitate gel sol

31.1 NA 34.9

precipitate gel sol

33 34 33.4

δp (MPa1/2) DBS NA 7.5 10.5 A-DBS 8.2 6.3 10.6 DCHS NA NA 7.7 A-DCHS 13.4 NA 7.2 DES 5.2 10.9 10.3

δh (MPa1/2)

radius (MPa1/2)

NA 8.7 9.1

NA 11.2 6.4

5.1 9.6 5.9

10.3 11.1 7.9

NA NA 8.8

NA NA 11

12.8 NA 7.3

8.1 NA 9.9

7 10.8 9.8

9.3 9.2 7

whether there was an influence of the position of the molecules’ HSPs relative to each other on the gelation outcome (Figure 2).

Figure 2. Two-dimensional projections of the distances between the polar (Δδp), dispersive (Δδd), and hydrogen-bonding (Δδh) Hansen solubility parameters of the solvents and (A−C) DBS, (D−F) DES, and (G−I) A-DBS. Solutions are represented by blue diamonds, gels by red circles, and precipitates by green squares. E


Article

Langmuir

Figure 3. Polarized light micrographs of gels containing 5 wt % gelator: (A−E) DBS in (A) acetone, (B) chloroform, (C) isobutyl alcohol, (D) oxylene, and (E) benzene; (F−J) A-DBS in (F) benzyl methacrylate, (G) acetophenone, (H) hexanoic acid, (I) benzene, and (J) salicaldehyde; and (K−O) DES in (K) dimethylformamide, (L) triethylene glycol, (M) hexanoic acid, (N) pyridine, and (O) acetophenone. Scale bars are 10 μm.

stacking, DCHS, gelled only carbon tetrachloride. It is also possible that the planar nature of the benzylidene groups on DBS allowed for superior self-assembly in comparison to the cyclohexyl groups of DCHS, which would adopt a bulkier chair conformation. The idea of the bulkier cyclohexane group affecting the ability to self-assemble is reinforced by the comparison of DES to DCHS. Neither DES nor DCHS can π−π stack, but both can form hydrogen bonds; nevertheless, DES is a much more efficient gelator. Therefore, it is possible that the less bulky ethylidene acetal group featured on DES does not physically constrain gelation. Polarized light microscopy (Figure 3) was used to examine the effects of the gelator and solvent structures on the morphologies of the SAFiNs constituting the supramolecular networks of the gels. It is evident that DBS, A-DBS, and DES all form SAFiNs within the gelation sphere. As the distance in Hansen space, Rij, increases, the aspect ratios of the fibers decrease, resulting in wider, denser fibers, an indication that they are becoming less soluble. Except for DBS, an Rij value greater than 10 MPa1/2 leads the fibrillar morphology to

acetonitrile, or acetone. Gels are thought to form at intermediate Rij values, where the gelators are slightly soluble but still able to crystallize into SAFiNs entrapping solvent. It is peculiar that A-DBS in toluene (Rij = 6.89 MPa1/2) and benzene (Rij 8.21 MPa1/2) forms precipitates whereas DBS forms gels in both toluene (Rij = 10.67 MPa1/2) and benzene (Rij = 11.70 MPa1/2) (Table S1). It was expected that A-DBS should be soluble in these solvents because it has smaller Rij values than DBS, which forms gels, and the solvent should easily be able to interact with the phenyl groups of A-DBS. Neither DCHS (Figure 1 and Figure S18) nor A-DCHS (Figure 1 and Figure S20) were efficient gelators in the solvents tested, with each forming a gel in a single solvent. DCHS and A-DCHS both lack the ability to π−π stack and have bulkier configurations than DBS and A-DBS, where the aromatic phenol groups are planar. It is obvious that π−π stacking plays a more predominant role in gel formation than hydrogen bonding. However, removing π−π stacking from DBS had a much more pronounced effect than expected. DBS could gel 17 of 23 solvents, but a DBS analogue that cannot achieve π−π F


Article

Langmuir transition to a platelet morphology. One reason for the remarkable ability of DBS to form SAFiNs in a broad array of solvents is that self-assembly can be driven by hydrogen bonding of the primary and/or secondary hydroxyl groups and/ or by π−π stacking of the benzylidene groups.4 In the case of DBS, irrespective of the noncovalent interactions that drive gelation, both functional groups are attached to a chiral carbon, which is a well-established requisite for SAFiNs with high aspect ratios.37−40 In the case of A-DBS (Figure 3F−J), the fibers are much thicker than for DBS, and this morphology does not appear to be simply a product of solubility (i.e., Rij). Clearly, A-DBS in acetophenone (Figure 3G) has the lowest aspect ratio below an Rij value of 10 MPa1/2. A-DBS gels in aromatic solvents (Figure 3F,G,J) are much less fibrillar and have platelet-like morphologies, so a potential molecular mechanism might be that the gelator−gelator π−π stacking is impeded because the solvent is capable of π−π stacking with itself. This is likely why Rij, albeit important, does not explain certain aspects of SAFiNs, and it would be very useful if HSPs could be further broken down to account for each noncovalent interaction. In the case of DES, as Rij increases, so does the fiber thickness. The morphologies of sorbitol-derived organogels were also examined by SEM for DBS, DES, and A-DBS in acetonitrile (Figure 4). The morphology of DBS shows a very fine fibrillar network. DES also exhibits a fibrillar morphology when solvent is removed. However, the fibers are much wider than those of DBS. A-DBS in acetonitrile forms a transparent solution with a very slight haze that is caused by very fine aggregates (∼10-μm crystals; Figure 4) that are suspended; this gelator−solvent combination could be equally well described as a precipitate, but irrespective of the classification, this does not impact our Hansen spheres because of the location of this point and overlap of the solution and precipitate spheres. The morphology of A-DBS xerogels shows what appear to be micellar structures or spherulitic crystals. Powder X-ray diffraction (XRD) shows dramatic differences between the crystalline structures of the various gelators (Figure 5). DBS gels in isobutyl alcohol have higher-order reflections and a diffraction peak that corresponds to a Bragg distance of 45.94 Å, a value much greater than the largest dimension of a single molecule (11.74 Å) based on computational modeling (Figure 5). The unit cell must be composed of an arrangement of DBS molecules; a tetramer is a reasonable suggestion for the unit cell that can facilitate its unusual gelation behavior. However, other organizations are certainly possible, and additional data must be collected to determine the actual molecular packing. If the gelator molecules adopt a highly order structure, such as a tetramer, this would facilitate the functional groups involved in the noncovalent interactions adopting the same configuration, thereby decreasing the likelihood of crystallographic mismatches and leading to epitaxial growth. The higher-order reflections for DBS dried organogels were previously reported using DBS in silicone− poly(ethylene glycol)−poly(propylene glycol) terpolymer;13 however, in those spectra, the peaks are much broader than observed in this study, likely because of the complexity of the solvent system used in that case. Irrespective of the molecular feature modified (i.e., the alcohol groups being converted to an acetyl group or benzyl moieties being substituted for cyclohexyl moieties), important features of the nanoassembly (most notably, higher-order reflections) are lost. For A-DBS, DES, and A-DCHS, the first peak of the diffraction profile

Figure 4. SEM images of DBS, A-DBS, and DES xerogels obtained by drying 5 wt % gelator in acetonitrile gels.

corresponds to a Bragg distance that is slightly larger than the shortest dimension of the molecule, and there is a lack of detectable higher-order peaks. These observations are indicative of lower degrees of crystallinity for the DBS derivatives/ analogues.

■

CONCLUSIONS The sorbitol derivatives DCHS and DES were synthesized to evaluate the importance of π−π stacking in SAFiN formation, G


Article

Langmuir

Figure 5. Long spacings for neat gelators in isobutyl alcohol obtained by powder X-ray diffraction with the d spacings, full widths at half-maximum, and domain sizes calculated using the Scherrer equation. Molecular dimensions were calculated, independent of the XRD data, using the Gaussian software g09/g1631 and the DFT/B3LYP32 method with the 6-31G33 basis set.

whereas A-DBS was used to assess the importance of hydrogen bonding. The results indicate that π−π stacking plays a significant role in the formation of DBS SAFiNs. The cyclohexyl rings featured in DCHS hinder self-assembly because of their lack of planarity and inability to support π−π stacking. From plots of the HSPs in Hansen space, gelation, solubility, and precipitate spheres have been constructed. They visualize and quantify differences in the intermolecular noncovalent interactions that are significant in these gelator−solvent systems. Overall, this study helps to identify the effects of different intermolecular interactions on SAFiN formation in this very important gelator system.

■

■

profiles for the pure D-sorbitol-based gelators heated at 2 °C/min; 1H and 13C NMR, correlation spectroscopy (COSY), and heteronuclear single-quantum coherence (HSQC) spectra of A-DBS, 5,6-diacetyl-D-sorbitol, ADCHS, and DCHS; 2D projections of DCHS, DBS, DES, 5,6-diacetyl-1,3:2,4-dicyclohexanecarboxylidene-Dsorbitol, and 5,6-diacetyl-1,3:2,4-dibenzylidene-D-sorbitol; and calculated distances (Rij) between the gelator and solvent (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected].

ASSOCIATED CONTENT

ORCID

S Supporting Information *

Richard G. Weiss: 0000-0002-1229-4515 Michael A. Rogers: 0000-0003-0079-4309

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.7b02191. Details of the syntheses of A-DBS, 5,6-diacetyl-D-sorbitol, A-DCHS, and DCHS and procedures for NMR and optical rotation measurements; depiction of the pathway for synthetizing DCHS starting from DBS; DSC melting

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS M.A.R. acknowledges the generous support of the Canadian Foundation for Innovation, NSERC Discovery, and Canadian H


Article

Langmuir

(20) Rohner, S. S.; Ruiz-Olles, J.; Smith, D. K. Speed versus stability − structure−activity effects on the assembly of two-component gels. RSC Adv. 2015, 5, 27190−27196. (21) Smith, D. K. Molecular gelsunderpinning nanoscale materials with organic chemistry. Tetrahedron 2007, 63, 7283−7284. (22) Muro-Small, M. L.; Chen, J.; McNeil, A. J. Dissolution Parameters Reveal Role of Structure and Solvent in Molecular Gelation. Langmuir 2011, 27, 13248−13253. (23) Diehn, K. K.; Oh, H.; Hashemipour, R.; Weiss, R. G.; Raghavan, S. R. Insights into organogelation and its kinetics from Hansen solubility parameters. Toward a priori predictions of molecular gelation. Soft Matter 2014, 10 (15), 2632−2640. (24) Hansen, C. M. 50 years with solubility parameterspast and future. Prog. Org. Coat. 2004, 51, 77−84. (25) Hansen, C. M. Hansen Solubility Parameters. 2nd ed.; CRC Press: Boca Raton, FL, 2007. (26) Bonnet, J.; Suissa, G.; Raynal, M.; Bouteiller, L. Organogel formation rationalized by Hansen solubility parameters: influence of gelator structure. Soft Matter 2015, 11, 2308−2312. (27) Lan, Y.; Corradini, M. G.; Rogers, M. A. Do Molecular Gelators Cluster in Hansen Space? Cryst. Growth Des. 2014, 14, 4811−4818. (28) Shen, H.; Niu, L.; Fan, K.; Li, J.; Guan, X.; Song, J. Application of solubility parameters in 1,3:2,4-bis(3,4-dimethylbenzylidene)sorbitol organogel in binary organic mixtures. Langmuir 2014, 30, 9176−9182. (29) Storn, R.; Price, K. Differential Evolution − A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces. J. Global Optim. 1997, 11, 341−359. (30) Champion, B.; Strzebonski, A. Constrained Optimization. In Wolfram Mathematica Tutorial Collection [Online]; Wolfram Research, Inc: Champaign, IL, 2008; http://reference.wolfram.com/language/ tutorial/ConstrainedOptimizationOverview.html, accessed May 30, 2017. (31) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A. V.; Bloino, J.; Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J. V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams-Young, D.; Ding, F.; Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.; Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.; Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J. J.; Brothers, E. N.; Kudin, K. N.; Staroverov, V. N.; Keith, T. A.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J. Gaussian 16, revision A.03.; Gaussian, Inc.: Wallingford, CT, 2016. (32) Andersson, M. P.; Uvdal, P. New Scale Factors for Harmonic Vibrational Frequencies Using the B3LYP Density Functional Method with the Triple-ζ Basis Set 6-311+G(d,p). J. Phys. Chem. A 2005, 109, 2937−2941. (33) Rassolov, V. A.; Pople, J. A.; Ratner, M. A.; Windus, T. L. 631G* basis set for atoms K through Zn. J. Chem. Phys. 1998, 109 (4), 1223−1229. (34) Bondi, A. J. van der Waals Volumes and Radii. J. Phys. Chem. 1964, 68, 441−451. (35) Lan, Y.; Corradini, M. G.; Liu, X.; May, T. E.; Borondics, F.; Weiss, R. G.; Rogers, M. A. Comparing and Correlating Solubility Parameters Governing the Self-Assembly of Molecular Gels Using 1,3:2,4-Dibenzylidene Sorbitol as the Gelator. Langmuir 2014, 30 (47), 14128−42. (36) Gao, J.; Wu, S.; Rogers, M. A. Harnessing Hansen Solubility Parameters to Predict Organogel Formation. J. Mater. Chem. 2012, 22, 12651−12658. (37) Rogers, M. A.; Weiss, R. G. Systematic modifications of alkanebased molecular gelators and the consequences to the structures and properties of their gels. New J. Chem. 2015, 39 (2), 785−799.

Research Chairs programs. G.G. and R.G.W. thank the U.S. National Science Foundation (Grant CHE-1502856) for support of the portion of the research conducted at Georgetown University.

■

REFERENCES

(1) Patel; Velikov, K. P. Colloidal delivery systems in foods: A general comparison with oral drug delivery. LWT-Food Science and Technology 2011, 44 (9), 1958−1964. (2) Zulim Botega, D. C.; Marangoni, A. G.; Smith, A. K.; Goff, H. D. Development of formulations and processes to incorporate wax oleogels in ice cream. J. Food Sci. 2013, 78 (12), C1845−51. (3) Grassi, S.; Carretti, E.; Dei, L.; Branham, C. W.; Kahr, B.; Weiss, R. G. D-Sorbitol, a structurally simple, low molecular-mass gelator. New J. Chem. 2011, 35, 445−452. (4) Okesola, B. O.; Vieira, V. M.; Cornwell, D. J.; Whitelaw, N. K.; Smith, D. K. 1,3:2,4-Dibenzylidene-D-sorbitol (DBS) and its derivatives–efficient, versatile and industrially-relevant low-molecularweight gelators with over 100 years of history and a bright future. Soft Matter 2015, 11 (24), 4768−87. (5) Sagiri, S. S.; Singh, V. K.; Pal, K.; Banerjee, I.; Basak, P. Stearic acid based oleogels: a study on the molecular, thermal and mechanical properties. Mater. Sci. Eng., C 2015, 48, 688−99. (6) Smith, J. M.; Katsoulis, D. E. Gelation of Silicone Fluids using 1,3:2,4-Dibenzylidene Sorbitol. J. Mater. Chem. 1995, 5 (11), 1899− 1903. (7) Sangeetha, N. M.; Maitra, U. Supramolecular gels: Functions and uses. Chem. Soc. Rev. 2005, 34 (10), 821−36. (8) Xie, C.; Rieth, L. R.; Danielson, T. D. (Milliken & Company). Dibenzylidene Sorbitol (DBS)-based Compounds, Compositions and Methods for Using Such Compounds. U.S. Patent 7,662,978 B2, Feb 16, 2010. (9) Liu, S.; Yu, W.; Zhou, C. Solvents effects in the formation and viscoelasticity of DBS organogels. Soft Matter 2013, 9, 864−874. (10) Wilder, E. A.; Spontak, R. J.; Hall, C. K. The molecular structure and intermolecular interactions of 1,3:2,4-dibenzylidene-D-sorbitol. Mol. Phys. 2003, 101, 3017−3027. (11) Krishnan, A. S.; Vargantwar, P. H.; Spontak, R. J. Thermorheological behavior of coexisting physical networks: combining SAFIN and SAMIN organogels. Soft Matter 2012, 8 (48), 12025. (12) Basrur, V. R.; Guo, J.; Wang, C.; Raghavan, S. R. Synergistic Gelation of Silica Nanoparticles and a Sorbitol-Based Molecular Gelator to Yield Highly-Conductive Free-Standing Gel Electrolytes. ACS Appl. Mater. Interfaces 2013, 5, 262−267. (13) Mercurio, D. J.; Spontak, R. J. Morphological Characteristics of 1,3:2,4-Dibenzylidene Sorbitol/Poly(propylene glycol) Organogels. J. Phys. Chem. B 2001, 105, 2091−2098. (14) Wilder, E. A.; Braunfeld, M. B.; Jinnai, H.; Hall, C. K.; Agard, D. A.; Spontak, R. J. Nanofibrillar Networks in Poly(ethyl methacrylate) and Its Silica Nanocomposites. J. Phys. Chem. B 2003, 107, 11633− 11642. (15) Zhu, G.; Dordick, J. S. Solvent Effect on Organogel Formation by Low Molecular Weight Molecules. Chem. Mater. 2006, 18, 5988− 5995. (16) Lan, Y.; Corradini, M. G.; Weiss, R. G.; Raghavan, S. R.; Rogers, M. A. To Gel or Not to Gel: Correlating Molecular Gelation with Solvent Parameters. Chem. Soc. Rev. 2015, 44, 6035−6058. (17) Terech, P.; Furman, I.; Weiss, R. G. Structures of Organogels Based upon Cholesteryl 4-(2-Anthryloxy)butanoate, a Highly Efficient Luminescing Gelator: Neutron and X-ray Small-Angle Scattering Investigations. J. Phys. Chem. 1995, 99 (23), 9558−9566. (18) Weiss, R. G.; Térech, P. Introduction. In Molecular Gels: Materials with Self-Assembled Fibrillar Networks, Weiss, R. G.; Térech, P., Eds.; Springer: Dordrecht, The Netherlands, 2006; pp 1−13. (19) Raynal, M.; Bouteiller, L. Organogel Formation Rationalized by Hansen Solubility Parameters. Chem. Commun. 2011, 47, 8271−8273. I


Article

Langmuir (38) Grahame, D. A. S.; Olauson, C.; Lam, R. S. H.; Pedersen, T.; Borondics, F.; Abraham, S.; Weiss, R. G.; Rogers, M. A. Influence of Chirality on the Modes of Self-Assembly of 12-Hydroxystearic Acid in Molecular Gels of Mineral Oil. Soft Matter 2011, 7, 7359−7365. (39) Brizard, A.; Berthier, D.; Aime, C.; Buffeteau, T.; Cavagnat, D.; Ducasse, L.; Huc, I.; Oda, R. Molcular and Supramolecular Chirality in Gemini-Tartrate Amphiphiles Studies by Electronic and Vibrational Circular Dichroisms. Chirality 2009, 21, E153−E152. (40) Brizard, A.; Oda, R.; Huc, I. Chirality Effects in Self-Assembled Fibrillar Networks. Top. Curr. Chem. 2005, 256, 167−215.

J


Molecular Nuances Governing the Self-Assembly of 1,3:2,4

Recommend Documents