reading

Chapter 2: Engineering of Small-Molecule Gels Based on the Thermodynamics and Kinetics of Fiber Formation

Chapter 7: Functionalization of Colored/Fluorescent Silkworm Silk Fibrous Materials

Chapter 8: Flexible Nanogenerator and Nano-Pressure Sensor Based on Nanofiber Web of PVDF and its Copolymers

Introduction to Soft Matter - Synthetic and Biological Self-Assembling Materials Revised

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at <http://dnb.d-nb.de>.

All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form — by photoprinting, microfilm, or any other means — nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

Preface

Nowadays, the advance of modern sciences and technologies depends to a large extent on the step changes in materials science. The research and engineering of materials have become one of the most exciting areas across physics, chemistry, biology, and engineering. Soft matter is a subfield of condensed matter comprising a variety of physical states that are easily deformed by thermal stresses or thermal fluctuations or under normal stress. They include liquids, colloids, polymers, foams, gels, granular materials, and a number of biological materials. These materials share an important common feature in that predominant physical behaviors occur at an energy scale comparable with room-temperature thermal energy.

In the area of materials science and engineering, three trends are of major research interest (Figure 1).

Figure 1 Three major trends in materials science and engineering.

The ultrafunctional materials refer to those having some extraordinary properties. The materials entirely or partially appear to be superhard, superhydrophobic, superhydrophilic, superconducting, and so on. Spider dragline silk fibers can be considered one of the toughest materials in terms of energy versus density. It was estimated that a spider silk string a pencil-width thick can stop a Boeing 747 in flight. Lotus leaves turn out to be one of the common examples of superhydrophobicity, with the capability of self-cleaning. Multifunctional materials correspond to those having more than one major in-use properties/functions, an example being fluorescent silk. Smart and responsive materials are those that respond to some external stimuli in the way that particular properties of the materials change drastically and/or in opposition to conventional materials. Under some external stimuli, the color, optical properties, or conductivity of the materials change correspondingly. Shear responsive fluids, thermal responsive gels, and such materials belong to this type.

In comparison with conventional “hard materials,” soft materials play a more important role in contemporary science and technology. It is the current tendency that many conventional “hard materials” are gradually replaced by soft materials due to the excellent performance, light weight, and broader applications. Subject to structural characteristics, the aforementioned three trends of research can be implemented more easily in soft materials. The increasing demand and broad applications of various special fibers and complex materials can be regarded as two such examples.

Soft materials display combined solid and liquid properties, the so-called rheological properties. Correspondingly, the structures are normally complicated. This can be because soft materials consist of certain network structures. In many cases, these are fibrous network structures, ranging from nanoscale to microscopic scales. Therefore, the understanding on the formation of fibrous networks is the key to fabricate and engineer materials of this type.

This book deals with this type of important soft functional materials. We will take this opportunity to demonstrate a principle: the elegant engineering of materials should be built on decent understanding, which can be illustrated by the so-called engineering triangle (Figure 2). More specifically, the engineering of materials with some particular properties can be implemented by fabricating the structure of the materials, which can be achieved by controlling the formation kinetics. In detail, this concerns the establishment of the correlation between the structure and performance of the materials and the acquirement of formation kinetics of the materials. The latter should allow us to control the structure in order to acquire the materials with particular functionalities. In this regard, our aim is to demonstrate that based on the understanding of the formation mechanism of the materials, one can design and fabricate the materials of new functions and smartperformance/ultraperformance. The approaches described in this book will provide the readers with comprehensive knowledge and feasible approaches in designing and refining performance by tuning the network structure of the materials.

Figure 2 Engineering triangle from understanding to engineering: the engineering of materials with some particular in-use properties/performance is implemented by fabricating the structure of the materials. This can be achieved by controlling the formation kinetics.

The book covers subjects related to important soft functional materials that have fibrous network structures. The materials include small-molecule physical gels, polymer gels, natural silk fibrous materials, and network materials based on nanofibrils, with respect to both the fundamentals, and the development and engineering methods. Their applications will concern drug delivery, home and personal care, macromolecule separation, catalysis, templating, tissue engineering, sensing, technical textiles and so on. It provides the reader with the necessary knowledge regarding chemical and physical formation mechanisms of these materials and demonstrates that one can rationally design and tune fibrillar networks so that the resulting materials exhibit desired functionalities. It also shows how materials from Nature, such as spider silk, can be adapted and functionalized for man-made applications and even mimicked in the laboratory.

The uniqueness of this book lies in the combination of the fundamentals of materials formation, engineering principles and approaches, and product design. The basic principles and sciences behind the technical approaches will be discussed in detail so that it is suitable to be adopted as a textbook for graduate students or specialists in this field. Numerous examples of applications and formulation based on the above engineering criteria are highlighted. Therefore, it can also serve as a comprehensive reference for the scientists and engineers working in related fields.

Chapter 1

Molecular Gels and their Fibrillar Networks

1.1 Introduction

This chapter will review, in a non-comprehensive fashion, the formation and properties of objects with very high aspect ratios [essentially one-dimensional (1D) objects at the micron or larger distance scales] made from organic molecules [topologically zero-dimensional (0D) objects at micron-range distance scales] which are not linked covalently and aggregate upon separation from dilute organic solutions or sols [1]. It will stress those 1D structures which undergo further assembly into 3D networks [self-assembled fibrillar networks (SAFINs)] that entrap the liquid in which they form. It remains largely unknown how and why many small organic molecules with very different shapes and functionalities [2] are able to separate from dilute organic (NB, leading to organogels) or aqueous (NB, leading to hydrogels) solutions or sols in the form of objects with very high aspect ratios [1].

The general name given to such materials is “molecular gels”, and the molecules that constitute them are referred to as low-molecular-mass organic gelators (LMOGs), although many of the materials may not meet the strict rheological definition of a gel as required by their viscoelastic properties [3]. The smallest known LMOG is N,N′-dimethylurea, 88 DA [4], and the largest are limited arbitrarily at < 2000 Da (although with some “poetic license”). The range of small molecules that can lead to gels via fiber and SAFIN formation is now in the hundreds, if not more than one thousand [1]. Because the molecules are aggregated but not linked covalently, the disassembly of the 1D objects (and their 3D networks) can be accomplished by application of heat, dilution, shear, or other perturbations which will be discussed.

The history of gels made from LMOGs may go back as far as the fourteenth century, although this example remains unsubstantiated and controversial [5]. The first formal description of a hydrogel of which we are aware, employing lithium urate, was reported by Lipowitz in 1841 [6]. A description of gels with the well-known and widely used LMOG, 1,3:2,4-di-O-benzylidene-d-sorbitol (1), was published in 1891 [7]. However, it was not until the middle of the twentieth century that scientists began to confront the intricacies of SAFINs and different forms of gels. In his “structural classification of gels,” Flory included those starting with 0D molecules as an afterthought, naming them “particulate, disordered structures”! [8] Although much has been learned during the last decade about the supramolecular assembly of polymeric chains (topologically 1D objects) into a variety of 2D and 3D objects [9, 10], much less is known about the initial steps that take 0D molecules to 1D objects, such as fibers, rods, tapes, and nanotubes (Figure 1.1); for the purposes of this chapter, all of these high-aspect-ratio objects will be designated as “fibers”, regardless of the details of their shape, unless specified otherwise for purposes of differentiation.

Figure 1.1 Cartoon representation of the steps in the evolution of LMOGs (0D objects; tear drops) to fibers (1D objects) and, in some cases, to SAFINs (3-D objects) in liquids (wavy lines). Lower left is a freeze-fracture electron micrograph of a SAFIN.

This type of 1D aggregation is distinguished from other types of self-assembly [12] that do not lead to fibrous networks and may involve plates, multilayered objects [13], and even bulk crystals as the basic units [14]. In many cases, the micro-phase separation of the 1D objects leads to organogels when the liquid is organic or hydrogels when it is aqueous. In both cases, there is an evolution of the aggregate structures which is controlled by very complex dynamics.

To date, the vast majority of studies of molecular gels has concentrated on structural and rheological aspects of their properties. In fact, the number of detailed studies treating both structure and kinetics of fiber formation in SAFINs is relatively small [15–19]. As a result, many questions remain about how small aggregates of LMOGs (still topologically 0-D objects at submicron length scales) form and then become (topologically) 1D objects. There are many important gaps in our knowledge as well about how 1D fibers transform into 2D or 3D objects, how 1D fibers of a SAFIN revert to 3D (microcrystalline) objects [20–23], how they undergo Ostwald ripening [21], and what controls their thixotropic behavior [24]. SAFINs may form as depicted in Figure 1.1 or by a completely different series of events, depending on the structure of the gelator, its concentration, the liquid component, and the protocol to transform the solution/sol to the gel. For example, in an alternative mode, new grains may develop on the sides of fibers or by tip-splitting (i.e., branching at the ends of growing fibers), giving rise to branched structures that lead to branched networks or spherulites [19]. Most of the systems discussed here undergo microphase separation by nucleation phenomena rather than by spinodal decomposition mechanisms [25].

Because the LMOG molecules in fibers are not attached covalently, the relevant intermolecular interactions include H-bonding, π-π-stacking, dipolar interactions, and London dispersion forces [1, 26]. In fact, the manner in which 1D objects, especially those composed of unbranched polymeric chains (i.e., objects in which one dimension of aggregation is due to covalent bonds) [27], convert to 2D and 3D objects [28] has received much more attention than the 0D → 1D transformations (i.e. those involving LMOGs) because experimental observations become much easier as the objects under scrutiny increase in size. Many of the polymeric gel networks are not disassembled by the same stimuli mentioned above; instead, they undergo conformational changes or separate otherwise physically from other polymer chains without losing their 1D status. For both 1D objects composed of LMOGs and polymer chains, additional interactions are needed to make them into 3D networks. Those interactions can be chain entanglements, branching, or inter-object associations involving “junction zones” of various types. Branching of the 1D objects made from LMOGs can be thought of as a consequence of defective growth during the 0D → 1D process [19c]. Junction zones occur at points of intersection between two 1D objects, and the participating molecules are frequently more disordered than within the “undisturbed” parts along the object. Alternatively, a junction zone may consist of abutting segments of two objects.

In some of the 1D objects, the constituent molecules are packed in a crystalline fashion whereas others, such as giant worm-like micelles, are not. The amount of detailed packing information potentially available about the crystalline objects is greater than about the amorphous (non-crystalline) ones. Yet, the ability of gels made with the amorphous (non-crystalline) 1D objects to recover their viscoelastic properties after cessation of severe shearing [24b] is much greater because many of them are in dynamic equilibria which allows self-annealing with time.

The study of 1D objects, especially those composed of LMOGs, and their gels requires multidisciplinary approaches among chemists, physicists, chemical engineers, biologists, and theoreticians. Research in this area, a branch of supramolecular chemistry, is important because systems based upon 1D objects and their assemblies, especially if the keys to designing them de novo can be discovered, can yield fundamental understanding of complex and highly selective catalytic processes, useful devices, and new ways to exploit systems available in nature. It can also shed light on the evolution and function (or malfunction) of systems of important biomolecular fibers that are involved with blood clotting and neurodegenerative diseases such as Alzheimer's, mad cow disease, and sickle cell anemia [29]. Also, fiber aggregates of small molecules are used to modify the mechanical properties of polymers [30] and food-related oils [31]. Ingenious manipulation of gelators in sols can lead to monodomains of 1D viscoelastic objects which are centimeters long [32] and may be useful in biological applications.

The questions of “How” and “Why” molecules with such diverse structures organize into 1D objects – fibers, tapes, nanotubes, and so on, with very high aspect ratios – remain largely unanswered. Although there are several theoretical [33–36] and experimental approaches [9, 15–17b, 19c,d,e, 37–40] to explain such aggregation and growth and even some predictive models for molecules with specific structures [19a,b, 35, 41–43], a generally applicable set of rules for when 1D objects will form is not available. It is likely that more than one basic mechanism controls the aggregation of molecules into the 1D objects, and the specific mechanism depends on the structure of the molecules, the nature of the solvent in which aggregation occurs [44], and the mode by which the sol phase is transformed into a gel [45]. Besides the need for strong attractive interactions along the long axis of the objects [46], there seem to be no real unifying principles. Although this chapter cannot present solutions to the parts of this science that remain unresolved, it can, is intended to, and hopefully will present a current picture of the state of the art in ways that allow the reader to discern where fruitful approaches to solutions may lie.

1.2 Advances and Perspectives for Design of Gelators

1.2.1 Analyses of Structure Packing via X-Ray, Synchrotron, and Other Techniques, Including Spectroscopic Tools

Elucidation of the molecular packing within the fibers formed during organogelation remains a challenging task. However, this information can provide key insights into the design of better gelators. Typically, fibers are characterized in the gel state (native gel, henceforth) or the dried gel state (xerogel). Microscopic characterization techniques such as polarized optical microscopy (POM), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and atomic force microscopy (AFM) have provided pictures of fiber morphologies in xerogels. Various characterization methods, as described in this section, can yield detailed information on the structures of SAFINs of gels at different length scales. However, information from a single characterization method is usually insufficient to reveal all aspects of a gel structure. Complementary tools should be employed and data from them used to build a cohesive picture of gel structure, including fiber morphology, molecular packing, intermolecular interactions, and so on.

A caveat noted by many others is reiterated here: the morphology of a xerogel does not represent necessarily that of the native gel because fiber damage or secondary assembly may occur during the drying process [47]. To minimize the possibility of such complications, freeze-fracture/etching SEM, and cryo-TEM techniques have been employed to visualize SAFIN structures of native gels. For example, the 3D network of 1D fibers of a steroid LMOG (2) in cyclohexane gel was revealed by a freeze-etching replication, electron-microscopy method (Figure 1.2a) [48]. Albeit less well resolved, POM can also provide the SAFIN structure of an organogel. Figure 1.2b demonstrates the POM image of a native gel of n-hexatriacontane (3) in silicone oil prepared in a flattened, sealed glass capillary [49].

Figure 1.2 (a) Electron micrograph of a gel (2) replica with a carbon film overlay showing the filament gel network protected (Scale bars are 100 nm for the main image and 10 nm for the inset); each filament is imaged as three layers and consists of a core (region A) sandwiched between a more electron-transparent coating, the outer edges of which are marked with arrows at B. Reprinted with permission from Ref. [48]. Copyright 1986 Elsevier.) (b) Optical micrographs of a 0.01 M 3/silicone oil gel at room temperature viewed through crossed polars; sample thickness = 0.8 mm.

Fibrillar structures can be clearly visualized from xerogels by SEM and AFM techniques, as shown in Figure 1.3 with gelator 4 [50]. However, as mentioned above, a correlation between such images and those of the gel itself should be made only if supported by additional characterization techniques, such as small angle scattering (SAS) [51], which relate the SAFIN structure and xerogels.

Figure 1.3 Structure of gelator 4 (a), SEM (b) and AFM (c) images from xerogels of 4.

SAS, including X-rays (SAXS) and neutrons (SANS), is a powerful technique to provide structure information about native gels. It has been used to provide insights into many gel structures [52]. As a result of their high intensity, synchrotron sources can enable characterization of native gels better than conventional X-ray sources. To perform SANS experiments, either deuterated gelators or deuterated solvents (or other contrasting liquids) are required. The difficulty to deuterate significant portions of most gelator structures has resulted, as expected, in the vast majority of studies being conducted with deuterated liquid components.

SAS is a model-based approach involving extensive mathematical operations; fortunately, many fitting programs are available. When SAS profiles of a native gel are obtained, an appropriate model needs to be chosen (e.g., rigid-rod, tubule, ribbon, or cylinder). Then, comparison is made between the simulated and experimental SAS profiles to validate the chosen model after the fitting parameters for size, persistence length, and so on, have been optimized. Terech and co-workers have reported many SAS investigations on gels, revealing the morphology of fibers as well as their junction zones [53]. For example, gelator 5 [53c] in decane formed hexagonally packed bundles (from structure factor analysis at large-angle scattering) of cylinders (from form factor analysis at low-angle scattering) (Figure 1.4). In addition, a solvent-dependent morphology change to more rectangular ribbon-shaped objects was observed in 1-alkanols.

Figure 1.4 (a) Structure of gelator 5. (b) SAXS cross-sectional intensity (QI/C) versus Q for organogels of 5 in 1-octanol (1 (•), C = 0.0017 g cm⁻³) and in decane (2 (), C = 0.0043 g cm⁻³); the lines are based on a theoretical fitting: 1, full line, R₀ (the geometric radii) = 72 Å, (the cross-sectional radial dispersity) = 0.1; 2, dotted line, R₀ = 75 Å, = 0.2; the vertical bar is a visual guideline between the low-angle and high-angle parts of the scattering data sets.

Sakurai et al. have employed synchrotron SAXS to support a previously proposed model [54] for molecular arrangement in a helical fiber of an azobenzene–cholesterol-based gelator (6) [52a]. A hollow cylinder model exhibited better agreement with the experimental SAXS profile than a solid cylinder model, suggesting that higher-electron-density azobenzene moieties are located at the exterior of the fibers while lower-electron-density cholesterol moieties are at the core of the fibers (Figure 1.5).

Figure 1.5 (a) Structure of gelator 6. (b) Proposed molecular packing in the fiber. Reprinted with permission from Ref. [54]. Copyright 1994 American Chemical Society. (c) Comparison of the measured SAXS data and the particle scattering functions of the hollow cylinder model with r (radius) = 65 and 55 Å.

Reproduced from Ref. [53a] with permission of The Royal Society of Chemistry. http://dx.doi.org/10.1039/B005470O

Wide-angle X-ray diffraction (WAXD) [55] has been utilized to investigate molecular packing within the crystalline fibers of gelators. Solving the crystal structure from single-crystal X-ray crystallography is a desired method to identify molecular packing. However, growing single crystals of LMOGs suitable for diffraction has been quite challenging; many form fibers or crystallize in a morph that is different from the one in the gel fibers. When a single crystal from a gelator is available and the X-ray powder diffraction (XRD) patterns from the crystal (or simulated XRD patterns from the crystal structure) and from the organogel are the same, molecular packing in the fiber can be elucidated [9]. As mentioned, the morphs of gelator fibers and bulk crystals may be either identical [4, 56] or different [57]. Ostuni et al. have demonstrated that XRD patterns of gelator fibers in a native gel (5/1-octanol) can be isolated by subtracting those of the solvent [57a]. It was found that the fibers had molecular packing closer to that of the neat solid cooled from the melt than to crystals isolated by precipitation from solution. The gelator (R/S)-7 (Figure 1.6a) also exhibited a similar behavior [57b]. The solvent subtracted XRD of its decane gel showed a pattern closer to that of the sublimed solid than to the bulk crystal (Figure 1.6b). Based on the single-crystal data of the sublimed solid, molecular packing in the fibers has been proposed (Figure 1.6c).

Figure 1.6 (a) Structure of gelator (R/S)-7. (b) Comparison of simulated XRD patterns of both known solid-state morphologies of (R/S)-7 (1: single-crystal data from solution, 2: single-crystal data from sublimed material) to XRD patterns of (R/S)-7 (3: sublimed solid, 4: solid cooled from neat melt, 5: solvent subtracted decane gel). (c) Aggregation model for (R/S)-7 in gel fibers.

Dastidar et al. have used molecular packing in gel fibers and in bulk crystals as obtained from XRD data to design effective gelators by identifying supramolecular synthons capable of 1D (and 2D) hydrogen-bonding (HB) networks that promote anisotropic fiber growth [9]. For example, dicyclohexylammonium 4-nitrocinnamate 8 (Figure 1.7a) gelates a few organic liquids such as benzene, toluene, xylene, and even gasoline [58].

Figure 1.7 (a) The structure of gelator 8 and 1D HB network of ion pairs in the crystal structure of 8. (b) XRD patterns under various conditions for 8. (c) The structure of nongelator 9 and 0D HB network of ion pairs in the crystal structure of 9.

Reproduced from Ref. [58] with permission of The Royal Society of Chemistry. http://dx.doi.org/10.1039/B504969E

A single-crystal packing structure of the organic salt 8 shows that one-dimensional HB is the most important intermolecular interaction responsible for the molecular arrangement. As shown in Figure 1.7b, XRD patterns simulated from the single-crystal data are nearly superimposable with those from the bulk solid, indicating the same molecular packing. Also, xerogels from benzene and p-xylene gels of 8 showed XRD patterns nearly identical to that of the pattern simulated from the single-crystal data. This result indicates that fibers in the xerogels adopt the same molecular arrangements found in the single crystal and the bulk solid. However, the molecular packing in the gel state could not be directly correlated with that in the single crystal since the XRD patterns from the gel was difficult to obtain due to the strong scattering from the solvent. Importantly, salt 9 was unable to form a gel, and only 0D HB networks were identifiable in the single crystal (Figure 1.7c).

Additional spectroscopic tools, including nuclear magnetic resonance (NMR), Fourier-transform infrared (FT-IR), UV–vis absorption, fluorescence (FL), and circular dichroism (CD), are available to monitor the changes in physical properties of aggregates during gelation. These techniques are able to identify different aspects of intermolecular interactions which contribute to gelation. CD spectroscopy, limited to chiral gelators or liquids, is discussed in Section 1.2.2.

A comprehensive description of NMR investigations of gels has been presented in a recent review by Shapiro [59]. Upon transition from a sol to a gel phase, proton resonances in ¹H NMR spectra experience significant broadening or disappear completely due to limited molecular motion [53d, 60]. For example, aromatic, vinylic, and some aliphatic protons of compound 10 cannot be observed in gel state spectra, but are clearly seen in the solution/sol phase spectra at high temperature where the system is a solution/sol (Figure 1.8) [60b].

Figure 1.8 (a) Structure of gelator 10. (b) Variable temperature ¹H NMR spectra of 10 mg/ml 10 in C₆D₆ in its gel phase (25 and 35 °C) and its solution/sol phase (45 and 55 °C).

Gels where solvent molecules are incorporated within fibers do allow more proton signals from gelator molecules to be observed, although some line broadening and shifts in proton resonances occur [52b, 61]. For example, a gel of 11 (Figure 1.9a) [61c] in toluene-d₈ exhibited a downfield shift of the N-H protons (H_a and H_b) in the gel state, indicating the presence of HB in the fibers (Figure 1.9b). The aromatic H_c signal appeared as a doublet in the solution/sol state and as two overlapping doublets in the gel state due to their inequivalence as packed in the fibers (Figure 1.9c). The spectra also indicate significant π-π stacking in the gel fibers.

Figure 1.9 (a) Structure of leucine-based gelator 11. (b) Superimposition of 1 wt% 11 in toluene-d₈ ¹H NMR spectra at different temperatures (diphenylmethane as internal standard). (c) Change of signals from the two H_c protons upon heating (in the rectangle). Intensities are normalized.

Gelators with chromophores or fluorophores typically suffer spectral changes as sol–gel transitions occur [62]. Cofacial (H-aggregate) and off-face stacking (J-aggregate) of chromophores induce a blue [63] or redshift [64], respectively, in absorption spectra. J-aggregate formation is more common and, in many cases, induces enhancement of emission intensities [60, 65], whereas H-aggregates frequently lead to decreased emission intensities. However, at this point, there are too few examples and inadequate theoretical understanding to conclude that these observations are universal. For example, the xerogel of 12 (Figure 1.10a) has an FL quantum efficiency (Φ_F) nearly two orders of magnitude higher than that of a dilute chloroform solution (Figure 1.10b) [65b]. The redshift in the emission maximum in the gel state (439 nm at 25 °C) from the solution state (402 and 423 nm at 80 °C) indicates J-aggregate formation as the cause of the emission enhancement (Figure 1.10c). Aggregation-induced emission enhancement can also be induced by restriction of molecular motion in the gel fibers, which decreases the rates of internal conversion and/or freezes in more planar and conjugated conformations [66].

Figure 1.10 (a) Structure of gelator 12. (b) Absorption (A) and emission (B) spectra of 1.0 × 10⁻⁵ mol L^–1 12 in chloroform; emission spectra (λ_ex = 320 nm) of a 0.1 wt% 12 gel in chloroform (C) and xerogel from chloroform (D). (c) Photoluminescence spectra (λ_ex = 320 nm) of 1.2 mg mL⁻¹ 12 in 1,2-dichloroethane at 80 °C, partial gel from 45 to 75 °C, and gel state below 40 °C.

FT-IR spectroscopy is a valuable tool to identify certain intermolecular interactions in the gel fibers, especially HB [61c, 65c]. Temperature-dependent FT-IR spectra of gelator 11 [61c] clearly shows the existence of intermolecular HB between N–H and C=O groups. In the sol and gel phases, at 90 and 30 °C, respectively, N–H and C=O stretching peaks were shifted to lower frequencies as a result of HB formation (Figure 1.11).

Figure 1.11 Temperature-dependent FT-IR spectra of 11 in toluene-d₈ as gels and solutions/sols (0.8 wt%).

The investigation of molecular organization in gel fibers by optical properties such as linear birefringence and FL dichroism has received little attention thus far although its potential utility is very high. In one example, fibers of gelator 13 (2,3-bis-n-decyloxyanthracene, Figure 1.12a) were aligned perpendicular to the direction of a magnetic field of 20 T (due to the diamagnetism of the LMOG) that was applied during the gelation process [67].

Figure 1.12 (a) Structure of gelator 13. (b) Polarized fluorescence spectra of a magnetically aligned gel of 13 in butanol; excitation at 340 nm was parallel to the long axis of the fibers and emission was either parallel (solid line) or perpendicular to the fiber long axis (dotted line); no changes in the unpolarized fluorescence spectra were detected before and after application of the magnetic field. (c) Calculated birefringence curves (solid lines) and experimental data points; the different fitting curves correspond to different stacking geometries of the fiber structure shown in (c); see text for explanation.

A higher emission intensity was observed when the detection was parallel to the fiber direction, which is also parallel to the optical transition dipole moment of molecules of 13 (Figure 1.12b); the transition dipole is perpendicular to the long molecular axis and in the plane of the aromatic ring. From this experiment, it was deduced that molecules in the fibers align with an angle (0 ≤ δ ≤ 54.7 °) relative to the original magnetic field. δ is defined as the usual polar angle in polar coordinates which describes the orientation of the long molecular axis with respect to the direction of magnetic field. This result is in good agreement with the birefringence data in which the alignment angle was estimated to be 0 ≤ δ ≤ 45 °. Furthermore, possible molecular models were provided, and these agreed well with the calculated birefringence and experimental data (Figure 1.12c). The field-induced birefringence from structures of I and II are positive (curve a), which is inconsistent with the experimental data. Both III and IV produce negative field-induced birefringence: III was inconsistent with fiber alignment direction from SEM; IV overestimates the birefringence (curve c). Only structures V and VI agree well with the experimental birefringence. In addition, the fiber and molecular arrangement directions in these models are consistent with SEM and FL dichroism results.

In another recent and elegant report, FL dichroism of nanofibers in some white-light-emitting multicomponent gels has been utilized to understand fiber structure [68]. The gels consist of 0.012 equiv. of green-emitting (14) and red-emitting (15) energy transfer (ET) acceptors (Figure 1.13a,b) added to the matrix of blue-emitting gelator 13 (Figure 1.12a for structure and Figure 1.13b for FL in gel) in DMSO (Figure 1.13c). The anisotropy of individual fibers in the white (W)-gel was analyzed with confocal FL polarization (P, Equation 1.1 where the intensity of linearly polarized emission is measured parallel to the excitation beam and the intensity of polarized emission is measured on the perpendicular axis) imaging under linearly-polarized laser excitation.

Figure 1.13 (a) Structures of LMOG energy transfer (ET) acceptors (14 and 15). (b) Fluorescence spectra (λ_ex = 365 nm) of 13 (2 mM gel in dimethylsulfoxide (DMSO), blue), 14 (10 µM solution in tetrahydrofuran (THF), green), and 15 (10 µM solution in THF, red). (c) Cuvette with 2.0 mM 13 and 0.012 equiv. of 14 and 15 in DMSO (W-gel) under UV light, λ_ex = 365 nm. (d,e) Fluorescence polarization confocal images (30 × 30 µm) under horizontally polarized excitation (horizontal, λ_ex = 385 nm) of a W-gel: (d) 400 nm < λ_em < 450 nm and (e) λ_em > 500 nm. Color codes correspond to the polarization P (from negative to positive: blue, green, and red). (f) Polar plots of average P for individual fiber segments (2–5 µm long) as a function of the angle θ of the fiber axis with respect to the laser polarization (horizontal) with λ_ex = 385 nm: (gray circles) 13-fibers λ_em > 405 nm; (blue circles) W-fibers, 400 nm < λ_em < 450 nm (orange squares) λ_em > 500 nm.

Selective absorption of the linearly polarized light occurs when the transition dipole for absorption is aligned parallel to the axis of the excitation beam; the dipoles of 13–15 are along their short molecular axes. P can range from −1 to +1, and the strong variation of P vs θ (the angle of nanofibers with respect to the orientation of the laser beam polarization) indicates a preferential orientation of the molecules within the nanofibers: in W-fibers, component 13 showed a positively polarized emission with P = 0.25 for fibers with θ = 90 ± 5 ° relative to the laser polarization (Figure 1.13d, red color code) and a negative polarization for fibers with θ = 0 ± 5 ° (P = −0.10, blue color code; 400 nm < λ_em < 450 nm). The large change of polarized emission with the angle θ indicates a high degree of molecular order and a preferential average orientation of molecules of 13 within the W-fibers. Because this result is very similar to that from fibers of 13 in the absence of the ET acceptors (Figure 1.13e, −0.08 ≤ P ≤ +0.22), it can be concluded that the self-assembly of 13 is not affected by the presence of 14 and 15. When sensitized by 13 (λ_ex = 385 nm, λ_em > 500 nm; Figure 1.13e,f), the FL from the 14 and 15 components in the W-fibers is almost non-polarized. Additional experiments revealed that 13 and 14 have similar preferential orientations, while 15 is more randomly oriented.

1.2.2 Chirality as a Tool – Comparisons between Optically Pure and Racemic Gelators and Optically Pure and Racemic Liquids

LMOGs with stereogenic centers have been studied extensively [69]. Enantio-pure gelators have enhanced our understanding of the gelation process by virtue of their ability to create helical supramolecular assemblies with a single handedness. Upon gelation, these helical assemblies are typically characterized using CD spectroscopy [69a,b] coupled with other microscopic techniques that help visualize fiber morphology [47, 69a]. In the solution state, chiral molecules generally exhibit very weak CD signals. Upon gelation of enantio-pure or enantio-enriched systems, significantly enhanced CD effects are commonly observed as a result of helical structure formation.

Typically, racemic mixtures of chiral gelators either do not form gels or they form unstable ones that degenerate easily into precipitates or bulk-separated crystals [69a, 70]. However, there have been some interesting exceptions in which a racemate produces stronger gels than their enantio-pure counterparts [70, 71].

The gelation of 12-hydroxyoctadecanoic acid (or 12-hydroxystearic acid) 16 has been studied as a model system based upon its structural simplicity. Tachibana, T. et al. initially investigated the gelation abilities of (R)-16 (d-16) as compared to its racemic mixture (dl-16) (Figure 1.14) [72]. The gels of enantio-pure 16 in CCl₄ exhibited CD maxima at 350 nm. Because this LMOG possesses no chromophores which absorb in this region, the origin of this band was hypothesized to be from preferential reflection of circularly polarized light of one sense by the gel. Interestingly, this effect is solvent dependent; the CD band shifted to 480 nm in benzene. Also, the racemic mixture, dl-16, did not form a gel at comparable concentrations.

Figure 1.14 CD spectra of enantiomerically pure 16 in CCl₄ gels: 25.7 mmol L⁻¹ for the l-acid; 35.7 mmol L⁻¹ for the d-acid.

Recent work by Grahame et al. has demonstrated the relationship between the gelation ability of 16 and its enantio-purity in mineral oil [73]. Thus, the critical gelator concentration (CGC) of enantio-pure 16 was less than 1.0 wt%, while racemic dl-16 required ∼ 2 wt%. The morphologies of the crystalline objects in SAFINs were drastically different as well. The gel of enantio-pure d-16 produced long, twisted fibers (Figure 1.15a), while the racemic mixture exhibited platelet crystallites (Figure 1.15b). When the ratio of d:l content in 16 was systematically varied, the FT-IR spectra of the resultant gels in mineral oil could be interpreted according to two different modes of crystallization. The analyses focused on the hydroxyl and carbonyl stretching regions. Fitting the area of hydroxyl HB peaks to the Avrami model [74] indicated (i) platelet-like crystals and sporadic nucleation (or spherulitic crystals and instantaneous crystallization) at d:l ratios below 80 : 20 and (ii) fiber-like crystal growth and sporadic nucleation at d:l ratios above 80 : 20. From analysis of the carbonyl stretching region, it was found that equal amounts of cyclic and acyclic dimers, formed between carboxylic acids, were present at d:l ratios below 80 : 20. At d:l ratios higher than 80 : 20, significantly more cyclic dimers were present.

Figure 1.15 Bright-field micrographs of (a) d-16 and (b) 50 : 50 d:l-16. (bar = 20 µm). Schematic representations of d-16 packing (c) and dl-16 packing (d) in SAFINs of mineral oil gels.

http://dx.doi.org/10.1039/C1SM05757J