NC State
Hubbe, M. A., Trovagunta, R., Zambrano, F., Tiller, P., and Jardim, J. (2023). “Self-assembly fundamentals in the reconstruction of lignocellulosic materials: A review,” BioResources 18(2), 4262-4331.


This review article considers processes by which the main components of wood have been reported to arrange themselves into various kinds of organized structures, at least to a partial extent.  The biosynthesis of wood provides the clearest examples of such self-organization.  For example, even before a cellulose macromolecule has been completely synthesized in a plant organism, the leading parts of the polymer chains already will have assembled themselves into organized crystals, i.e., nano-fibrils.  This review then considers a challenge that faces industrial engineers:  how to emulate the great success of natural systems when attempting to achieve favorable materials properties, process efficiency, and environmental friendliness when developing new engineered wood structures, barrier films, and other desired products composed of lignocellulosic materials.  Based on the reviewed literature, it appears that the main chemical components of wood, even after they have been isolated from each other, still have a remnant of their initial tendencies to come back together in a somewhat non-random fashion, following mechanisms that can be favorable for the production of engineered materials having potentially useful functions.

Download PDF

Full Article

Self-assembly Fundamentals in the Reconstruction of Lignocellulosic Materials: A Review

Martin A. Hubbe,a,* Ramakrishna Trovagunta,b Franklin Zambrano,b Phoenix Tiller,a and Juliana Jardim c

This review article considers processes by which the main components of wood have been reported to arrange themselves into various kinds of organized structures, at least to a partial extent. The biosynthesis of wood provides the clearest examples of such self-organization. For example, even before a cellulose macromolecule has been completely synthesized in a plant organism, the leading parts of the polymer chains already will have assembled themselves into organized crystals, i.e., nano-fibrils. This review then considers a challenge that faces industrial engineers: how to emulate the great success of natural systems when attempting to achieve favorable materials properties, process efficiency, and environmental friendliness when developing new engineered wood structures, barrier films, and other desired products composed of lignocellulosic materials. Based on the reviewed literature, it appears that the main chemical components of wood, even after they have been isolated from each other, still have a remnant of their initial tendencies to come back together in a somewhat non-random fashion, following mechanisms that can be favorable for the production of engineered materials having potentially useful functions.

DOI: 10.15376/biores.18.2.Hubbe

Keywords: Thermodynamics; Trapped non-equilibrium states; Self-organization; Cellulose crystal formation; Hydrogen bonds; Surface activity; Nanostructures; Bonding; Composites

Contact information: a: Department of Forest Biomaterials, North Carolina State University, Campus Box 8005, Raleigh, NC 27695-8005; b: Solenis L.L.C., 500 Hercules Road, Wilmington, DE 19808; c: Simplifyber, 625 Hutton ST, Suites 105-106, Raleigh, NC 27606;

* Corresponding author:


Self-assembly can be defined as a tendency of molecules or other substructures to come together in an organized fashion, ideally resulting in a structure that embodies desired characteristics and behaviors. The actions of the DNA molecules, and indeed of life itself, can be regarded as the world’s premier example of self-assembly. Remarkably, a set of DNA molecules has the potential to direct the assembly of an entire living organism – not just for a single generation but also for the second generation, and then the next. Numerous other examples of self-assembly take place in a wide variety of existing processes, and in many cases they may be under-appreciated. For instance, when cellulose is formed into a solution and then rendered insoluble again by changing the solvent mixture, it has a strong tendency to form fiber-like structures (Fink et al. 2001; Zhang et al. 2018). Thus, engineers who work with lignocellulosic materials may ask themselves what they can do to take advantage of such a powerful principle in the manufacture of materials and products for usage in modern society. The goal of this review is to consider both the opportunities and constraints when self-assembly is considered as a potential tool for development and production of items based on the components of lignocellulosic source materials. This article focuses on fundamentals. A series of future articles are planned to go more deeply into literature findings related to self-assembly phenomena involving lignin, cellulose, hemicellulose, and wood extractives.


This review article aims to discuss basic concepts related to self-assembly strategies and to envision ways in which such strategies are or can be applied in biomass-related technologies, which can include papermaking, composites, and the formation of bio-based films. As should become clear from articles cited in this review, there are many challenges associated with such efforts. Accordingly, it makes sense to consider a simple question, “Why should a modern engineer, involved with product development, even consider self-assembly as a possible tool to use in preparing new materials and products from lignocellulosic resources?” Four reasons, to be described below, can be briefly labeled as “eco-friendliness,” efficiency,” “performance,” and “tailored properties”.


There is not necessarily a direct correlation between self-assembly and eco-friendliness. But on the converse, many of the preparation steps used presently in present industry yield large proportions of wasteful and often harmful byproducts and emissions, even when utilizing bio-based materials (Weiss et al. 2012; Hahladakis et al. 2018; Ramesh et al. 2020). Production of steel requires, for example, large amounts of fuel to reduce the ore to iron, leaving behind tons of low-valued slag (Norgate et al. 2007; Olmez et al. 2016). Traditional organic synthesis of drugs requires multiple procedural steps, many of which have relatively low yields, with the generation of toxic solvent-containing mixtures (Jahangirian et al. 2017). Plastic beverage containers are a prime example when considering products that compete with cellulose-based products. In addition to inherent inefficiencies and environmental harm associated with the initial production (Hahladakis et al. 2018), the proportion of recycled plastics is relatively low (Ragaert et al. 2017), and their remnants as litter and microparticles contaminate the environment and fail to decompose in practical lengths of time (Horton et al. 2017).

The term “biomimicry,” i.e., the imitation of nature, describes a promising strategy to achieve environmental compatibility. The logic is that nature itself has evolved a balanced, self-sustaining cycle of production, decay, and reutilization of byproducts. A modern engineer can choose to take advantage of some aspects of this self-sustaining web of processes in the production of items used by society. For instance, wood can be utilized as lumber, as oriented strands for strandboard, or as particles in the preparation of particleboard (Chen et al. 2020). At the same time, the engineer may choose to avoid toxic or non-biodegradable components or to avoid processing aids that do not fit with natural degradative and regenerative cycles (Hemmila et al. 2017; Hubbe et al. 2018).

An approach based on self-assembly may be expected to contribute to eco-friendly outcomes if and when the inherent self-organizing characteristics of the substances and procedures make it possible to decrease the usage of energy, toxic materials, or the generation of waste materials during preparation of an item needed by society. Consider, for example, the preparation of particleboard. In conventional practice, particleboard is made by coarse grinding of either freshly cut or used wood, followed by gluing it all back together in the presence of formaldehyde-containing resins, heat, pressure, and time (Owodunni et al. 2020). There has been extensive research aimed at finding conditions of pressing and formulation to allow the production of “binderless” board (Hubbe et al. 2018; Owodunni et al. 2020). This can include reliance on natural binder-like materials (e.g. tannins or lignin itself) that are already present in the plant materials (Tajuddin et al. 2016; Nasir et al. 2019). Such binderless or natural-binder approaches can be called self-assembly if and when the woody particles or resinous substances have an inherent tendency to organize themselves into a desired structure upon application of suitable processing conditions.


Enzymes can be regarded as premier examples of self-assembly tools. They can direct chemical reactions toward preferred pathways. For example, the enzyme amylase can be used to convert starch to sugar at room temperature and with very high yield (van der Maarel et al. 2002). Likewise, it was shown recently that the lipase enzyme can be used for surface esterification of cellulose nanocrystals (CNC) with palmitoyl groups, thus providing a hydrophobic surface that is more compatible with a hydrophobic polymer matrix (Yin et al. 2020). Enzymes often can direct chemical reactions toward a single desired outcome, thus minimizing the generation of unwanted byproducts.

An inherent challenge faces modern engineers when considering the use of approaches based on the principles of self-assembly. While enzymes and other self-assembly processes are known from nature, the resulting cellulose, hemicellulose, or lignin materials, in the form that they are readily available to the modern industrialist, may be lacking in key features that may have been present at an earlier stage of biosynthesis or freshness of those materials. Some of the biosynthesis steps in the natural generation of lignocellulose have at least partially irreversible character. For instance, once cellulose chains come together as crystals, during their biosynthesis, there are relatively few solvent conditions capable of separating them (Rosenau et al. 2001; Chen et al. 2019; Verma et al. 2019). Likewise, when technologists attempt to dissolve lignin through pulping technologies or other isolation methods, the created substructures often undergo partial condensation reactions, leading to substantial differences in chemical structure relative to native lignin (Kim and Kim 2018).


One of the first well-known efforts in nanotechnology involved the one-by-one placement of atoms onto a crystal surface, thus spelling out the initials of a corporation (Bayda et al. 2020). Though such an outcome was noted as a great achievement, it is clear that different approaches need to be employed to achieve rates of production suitable for mass production. Ideally, the building blocks of the desired product would be capable of organizing themselves. And even if the spelling of a corporate logo by placement of atoms remains a CEO’s dream, one also can aim for more practical goals, such as using self-assembly principles to create barrier films for packaging (Gokhale and Lee 2014; Hubbe 2021).

Oriented monolayer formation on smooth surfaces is achievable by optimized treatment with suitable surface-active compounds in solution (Senaratne et al. 2005). The contrasting affinities of the head groups and tail groups of the surfactant molecules drive the well-known adsorption tendency for such compounds. As shown in Fig. 1, there are often three main contributions to the stability of contiguous adsorbed monolayers in aqueous systems: between a hydrophobic tail group and the solid material; between adjacent hydrophobic tail groups; and between a hydrophilic head group and the aqueous solution. A main contribution to the so-called hydrophobic effect is the dominant effect of hydrogen bonding among water molecules (Tanford 1980). Those forces favor systems in which non-polar entities self-associate and thereby get out of the way of potential hydrogen bonding. In this way, self-association among non-hydrogen-bonding groups and compounds contributes to the free energy of the system. As suggested in Fig. 1, the interactions between the head groups and the aqueous phase can include not only hydrogen bonding but also the polar forces associated with optional ionized groups (Martinez et al. 2011). When the amount of surface-active molecules is less than what is needed for a fully compressed monolayer, as shown on the left of the figure, thermodynamics is likely to favor a conformation in which the molecules’ hydrophobic groups associate with a cellulosic surface. Though the latter can be described as “hydrophilic,” it is less so compared to an aqueous phase. The hydrophobic effect, having hydrogen bonding as its key factor, also benefits from the van der Waals-London component known as dispersion forces (Bowen and Jenner 1995).

Fig. 1. Illustration of three main contributions to the stability of certain surfactant monolayer films self-assembled onto solid surfaces from aqueous solution

Oriented multilayer formulations also can be assembled. Layer-by-layer structures can achieve a certain degree of regularity because of their opposite charge in comparison to the net charge of each preceding layer (Xiong et al. 2017a; Kramer et al. 2019). By alternating the application of cationic and anionic polyelectrolyte solutions, with each such application separated by a rinsing (and optional drying) step, it is possible to prepare highly regular multilayer structures (Tong et al. 2012; Ariga et al. 2014; Keeney et al. 2015). Though the process is very slow, it is possible to achieve practical effects in the laboratory such as increases in the fiber-to-fiber bonding strength within paper structures (Aulin et al. 2010b). In addition, the knowledge gained in forming such multilayers may lead to more practical strategies for forming paper with large gains in bonding strength at commercially competitive speeds of formation (Hubbe 2014a).

Engineered materials with tailored properties

To compete in the current marketplace, a manufactured material or product needs to accurately fulfill many requirements simultaneously. Likewise, natural materials, all of which are developed through biological self-assembly processes, display a high level of complexity (Sanchez et al. 2005; Mendes et al. 2013). These considerations imply that extensive trial and error may be needed when one’s goal is to use self-assembly as a tool in product development.

One promising approach by which to achieve a moderate level of complex structuring, while at the same time not exceeding what can be reasonably expected in terms of reliance on self-assembly, consists of three-dimensional printing, which is also called additive manufacturing (Mitchell et al. 2018; Ee and Li 2021). For example, Greenhall and Raeymaekers (2017) employed 3D printing to achieve a general placement of substances; ultrasonic vibrations were then used to enable the microscale self-organization of particles. Likewise, researchers have 3D-printed living biological cells into desired shapes or films, then relying on their natural self-organizing ability to finish the job of forming a biological tissue (Marga et al. 2012).

Reasons to be cautious

While the potential advantages of eco-friendliness, efficiency, performance, and tailoring of properties provide reasons for researchers and industrialists to consider production strategies based on self-assembly, there are also reasons to be cautious. Some of these already have been mentioned. A more complete list is as follows:

  • The amazing self-assembly capabilities associated with DNA and life itself have taken very long spaces of time to evolve.
  • The biosynthesis of lignocellulosic matter involves some steps that are at least partly irreversible.
  • The chemical pulping of wood, which is an example of a process that separates lignocellulose into its main components, is expected to involve irreversible changes to those components.
  • The organizing principles of self-assembly, such as charge attractions, hydrogen bonds, and differing affinities, may provide insufficient strength and durability of the prepared structures relative to their intended uses.
  • Industrial processes such as papermaking often involve the recirculation of water, which will be contaminated with diverse substances that may interfere with various details of self-assembly.

Some Definitions Related to Self-assembly


The definition of self-assembly provided at the very beginning of this article is not the only one. The textbook by Pelesko (2007) considers such definitions in detail. The book also provides numerous examples of self-assembly research, including both of ultra-simple and complex systems, ranging all the way to living organisms and ambitious philosophical questions. Pelesko (2007) has proposed the following general definition, in an attempt to encompass both the simple and the much more complex examples of self-assembly: “Self-assembly refers to the spontaneous formation of organized structures through a stochastic process that involves pre-existing components, is reversible, and can be controlled by proper design of the components, the environment, and the driving forces.” By the term “driving forces,” Pelesko is emphasizing the fact that self-assembly usually requires work to be done, and a completely static system, with no force fields or gradients in levels of energy, will lack a reason to become organized.

Fortunately, except at an absolute zero temperature, thermal energy can be relied upon to cause diffusion, i.e., Brownian motion (Li et al. 2008). The random motions can often be sufficient motivation, acting together with other features, to allow the components to sample different possible patterns of organization on the way to finding a favorable self-assembled pattern. Accordingly, the next definition given by Pelesko is “static self-assembly,” meaning that the random process has reached a stable equilibrium. The converse is a “dynamic self-assembly,” which means that the resulting structure is stuck in a non-equilibrium state, which is determined by kinetic factors. Also there is programmed self-assembly, in which the components each embody some form of instructions regarding how they are supposed to ultimately fit together. Lindsey (1991) uses the term “strict self-assembly,” which appears to be the same as what Pelesko (2007) would call “static self-assembly,” with the further stipulation by Pelesko that there is no human intervention. This is consistent also with a definition used by Whitesides and Grzybowski (2002). In contrast, one can define “directed self-assembly” to cover cases of interest in which the human influences the outcome by adjusting various conditions as a self-assembly process is taking place, such as to favor a certain outcome (Lindsey 1991). The cited authors also notes that the main interest is often in forming a higher-order structure in comparison to the level of complexity of the initial molecules or substructures. In its most highly developed form, one might apply an ultimate definition for self-assembly as having “stunning features” such as convergence, control of disassembly and reassembly, error-checking and recovery, and high accuracy (Lindsey 1991). From the foregoing discussion it should be clear that the definition of self-assembly can be fashioned in diverse ways to describe many different situations.


By carefully observing certain processes associated with living systems, scientists may gain insight into what is possible to achieve. Such a focus is especially appropriate when considering ways to manufacture products such as biodegradable packaging items. Biological materials having a lignocellulosic nature are well known for such attributes as toughness, high strength-to-weight ratio, and biodegradability (Bond et al. 1995). In addition, rather than breaking in a catastrophic fashion, woody materials often fail in a graceful manner, meaning that the loss of strength is gradual (Bond et al. 1995). However, the same authors note that some of the premier biobased structures, including bone, wood, and insect cuticles, have highly refined low-density structures that would be inherently difficult for a modern engineer to replicate in an industrial process. Although one might consider preparing the detailed structure of wood by means of three-dimensional printing, a recent article suggests that such a goal still remains beyond the capabilities of current 3D printing technology (Aydin and Yilmaz Aydin 2022). It remains an unresolved question as to whether developers will be able to match the properties of wood when using just the components, e.g., cellulose, hemicellulose, and lignin in non-native forms. Gradwell et al. (2004) showed that at least part of such an effort could be achieved. They utilized pullulan abietate – roughly representing lignin and polysaccharide, according to those authors – as a matrix phase in the presence of cellulosic fibers. The amphiphilic pullulan abietate adsorbed strongly onto the cellulose, as a purported first step in making a wood-like material.

Bottom-up preparation of nanomaterials

When discussing different ways to obtain nanomaterials, especially in the case of cellulose nanomaterials, the term “bottom-up” means that fibril-like materials are assembled by biosynthesis (Mendes et al. 2013). One can envision other nanomaterials, such as colloidal silica particles, being formed as a result of sticking collisions among primary particles, followed by their fusion into persistent structures (Iler 1979). It has been shown that, by careful adjustment of both pH and concentrations of the starting materials, the primary particles will experience a relatively high probability of sticking collisions only to the ends of chains, rather than onto the sides of existing chains or groups of such primary particles. Thereby, chain-like growth may be preferred. The resulting particles have been called “gels.” In contrast, the unchained “sol” form of such particles will form under conditions in which there are sufficiently strong electrostatic repulsions between particles such that they grow individually (Iler 1979). The mechanism is illustrated in Fig. 2.

Fig. 2. Mechanism to explain preferential end-wise deposition of primary colloidal silica particles, thus leading to chain formation in the presence of an intermediate level of electrostatic component of repulsive forces between the surfaces. A: A higher energy barrier is encountered when approaching from the side; B: A lower energy barrier is encountered when approaching from the end.

When a primary particle of colloidal silica (typically about 1 to 5 nm in diameter) approaches the side of a chain of such particles (as illustrated in part A), the approaching particle is affected by electrostatic repulsion originating at the surfaces of several particles in the chain simultaneously. By contrast, if the individual particles approaches from the end, it experiences repulsion mainly from just one particle, and therefore a successful collision is much more likely. Once the particles stick together, the chain essentially becomes sintered together due to ongoing interactions at the molecular scale. Such a bottom-up self-assembly could be considered for the components of lignocellulosic materials, and possibly such behavior might be demonstrated in the future.


Hypothesis One:

It is proposed that the essentially irreversible character of certain sub-processes in both the biosynthesis of woody materials and in the isolation of woody materials provides barriers to full implementation of self-assembly procedures during attempted re-assembly of the components. In other words, it is proposed, for instance, that the mixing of cellulose, hemicellulose, lignin, and minor components of wood will not be able to re-create material with properties corresponding to those of wood.

Irreversible attributes of cellulose

Various research findings can be cited in general support of this first hypothesis. The insolubility of cellulose in most solvent systems is well known (Rosenau et al. 2001; Chen et al. 2019; Verma et al. 2019). The explanation for cellulose’s reluctance to dissolve in most solvents apparently has to do with a two-fold mechanism involved in its crystallization. First, it is well known that cellulose undergoes highly regular and extensive intra- and intermolecular hydrogen bonding (Bergenstrahle et al. 2010). In addition, in the 1-10 crystal plane, the dominant interaction appears to be van der Waals attractions (Yamane et al. 2006). Any solvent system for cellulose needs to deal with both issues at once, or alternatively, it needs to form a derivative of cellulose (Medronho et al. 2012; Medronho and Lindman 2015). Thus, it is valid to say that cellulose’s initial crystallization and insolubilization during its biosynthesis are essentially irreversible.

Drying of cellulosic materials can lead to irreversible changes, especially in cases where lignin and some of the hemicelluloses have been removed by pulping (Stone and Scallan 1966; Weise and Paulapuro 1999; Pönni et al. 2012). The effect, which is often called hornification, appears to involve the closing up of gaps between cellulosic surfaces (Pönni et al. 2012). Such closing up is favored by strong capillary forces as water is being evaporated (Campbell 1959; Page 1993). The effect occurs to a minor degree even during the wet-pressing of cellulosic pulp fibers (Maloney et al. 1997). One could say that as the tiny gaps between cellulosic surfaces become healed (Pönni et al. 2012), larger sizes of crystals are achieved, though one can expect there to be defects in such coalesced crystal structures (Scallan 1998).

Irreversible attributes of hemicellulose

Hemicelluloses, despite their tendency to swell in water, are mainly insoluble in aqueous media. This insolubility is achieved despite the molecules’ somewhat irregular molecular structures (having side groups) and having two or three –OH groups per sugar monomeric unit. Based on these considerations, it can be concluded that a pattern of multiple, simultaneous hydrogen bonds is sufficient to keep hemicelluloses in solid form within a lignocellulosic or holocellulosic structure in the presence of water having moderate pH values. Only at pH values greater than about 14 does the hemicellulose fully dissolve (TAPPI Method T 429). There appears to be a need for further research to clarify such irreversible tendencies in hemicellulose aqueous systems, including a fuller explanation of hemicelluloses’ limited solubility in water.

Irreversible attributes associated with removal of lignin and hemicellulose

Another aspect of irreversibility can be blamed on the dominant pulping process that is being employed worldwide to liberate the cellulosic fibers from wood. The kraft pulping process not only breaks down and removes much of the lignin from the wood or nonwood lignocellulosic raw material, but it removes a substantial part of the hemicellulose as well (Fardim and Tikka 2011). It follows that one cannot expect the resulting cellulose fiber and directly reassemble as a wood-like material.

Irreversible attributes associated with lignin

Lignin, as it is being separated from lignocellulose in the course of pulping, is susceptible to condensation reactions (Schutyser et al. 2018). These reactions entail the essentially irreversible formation of carbon-carbon single bonds between aromatic groups. In principle, such condensed byproducts of lignin will be expected to have less ability to behave like native lignin. In an effort to minimize undesired condensation of lignin, researchers have developed a variety of strategies, often involving chemical derivatization of the starting material (Shuai et al. 2016; Renders et al. 2017; Schutyser et al. 2018). Such chemical details make it reasonable to question whether or not certain kinds of self-assembly, involving lignin moieties, may no longer be available after the completion of ordinary pulping and bleaching processes. Readers are encouraged to consult review articles for more details regarding pulping processes, the nature of the resulting lignin structures, and their possible modifications (Matsushita 2015; Trovagunta et al. 2021b).

Hypothesis Two:

It is proposed that developmental researchers and engineers will be able to improve their chances of achieving self-assembly due to the fact that they can select from a wide range of temperature, pressure, and chemical conditions, in comparison to the rather narrow ranges that are mandated by biological systems. Though living organisms have evolved to occupy niches in highly diverse environments, the aqueous conditions within the living cells generally are believed to approximate the salinity of an ancient ocean, in which the key initial steps of evolution presumably took place (Monnard and Deamer 2002). Under such conditions, DNA-based systems provide a high level of self-assembly in the perpetuation of life forms. But on the other hand, the modern developmental researcher does not have millions of years in which to wait for a desired evolutionary change to fully take place. As noted by Whitesides and Grzybowski (2002), a developmental researcher’s motivation to employ some sort of self-assembly process may be greatest when considering the nano-scale of product development. As has been noted (Bayda et al. 2020), it is very difficult to manipulate a nano-scale object on an individual basis. Thus, the present review article will consider various evidence that more effective and efficient organization may be achieved by taking advantage of certain propensities of lignocellulosic components to organize themselves, especially if the engineer is able to prescribe the temperatures, pressures, aqueous conditions, and other factors. It is proposed that present and future engineers can widen their options by employing ranges of temperature and pressure that are not available during biosynthesis. They can also pre-plan various chemical conditions and orders of addition. It is proposed that such options can provide opportunities for advances in materials processing that take advantage of self-assembly phenomena.

Some Key Sources

Table 1 provides a chronological listing, together with key themes and authors of review articles, other key articles, and one book dealing with self-assembly from various perspectives. Rather than going deeply into the subject matter of such articles, readers are encouraged to read the originals.

Table 1. Key Literature Related to Self-assembly


As has already been noted, this article is about instances in which separated components of lignocellulose appear to self-arrange themselves. It is reasonable, in such cases, to consider a null hypothesis to the effect that the apparent self-organization is merely random or illusory. It has been suggested that humans have an inherent tendency to look for elements of organization in any situation that first appears to be chaotic (Whitesides and Grzybowski 2002). Thus, the first order of business is often to look carefully at each set of evidence.

In cases where evidence of self-assembly seems persuasive, the next step is to provide an explanation. In general terms, three things need to be present as a precondition for self-assembly to occur, namely some kind of information, some kind of interactive force, and some form of chaotic input, such as thermal diffusion or vibrations, etc. As discussed in the subsections below, all of these influences are required to follow the constraints imposed by thermodynamics.


It has been proposed that organized assembly will require some form of information, which can take various forms. It can be embedded, for instance, in an amphiphilic nature of the parts to be assembled, such that they will prefer to go together in a certain manner. These tendencies could be based on shapes, on degrees of hydrophilicity, on the signs of electrical charge, or on details of molecular structure.

Amphiphilic character

Especially in aqueous media, a certain degree of self-organization can be expected when the dispersed entities (e.g., molecules or particles) have hydrophilic parts and hydrophobic parts. Some examples are listed in Table 2. Of particular interest here are the findings that support the presence of different faces of native cellulose crystalline entities, giving rise to different affinities (Glasser et al. 2012; Khazraji and Robert 2013). Such affinities are important to keep in mind as potential contributing factors, even when researchers may be paying attention to different issues in their investigations.

Table 2. Publications Providing Examples in Which Amphiphilic Character Appeared to Promote Self-organization


The attraction between opposite electrical charges is well known to provide a driving force for various self-assembly phenomena. For example, the adsorption of cationic surfactants onto anionic substrates such as lignocellulosic material is known to benefit from such interactions (Biswas and Chattoraj 1997; Wågberg 2000; Szilagyi et al. 2014). Electrostatic forces between surfaces have been found to be especially prominent after oxidation of the cellulose to increase the amount of anionic groups on such surfaces (Soboyejo and Oki 2013).

Molecular structure

The fact that the details of molecular structure can affect organization at the nano-scale is already apparent from studies of DNA and its functions (Pelesko 2007). Another example comes from studies involving the partial modification of xylans (Kaya et al. 2009). The cited authors showed that different levels of hydroxypropylation of xylan influence its solubility in interesting ways. Although the hydroxypropyl groups are somewhat hydrophobic, low levels of hydroxypropylation favored water-solubility. At higher levels of substitution, the products of the reaction tended to agglomerate and separate themselves from water, as would be expected based on an increased hydrophobic character. It seems likely that the opposite effect at low substitution levels can be attributed to a break-up of regularity. This can be taken as evidence that the initial insolubility of the xylan is dependent on the regularity within its chain structure. Studies have shown that xylans, despite their strongly hydrophilic character, have a tendency to agglomerate in aqueous solution (Saake et al. 2001).


Some of the most rudimentary empirical studies to confirm various aspects of self-assembly theories have involved the shapes of the yet-to-be-assembled items (Pelesko 2007). For instance, Li et al. (2008) showed that even monodispersed spheres tend to pack into neat face-centered cubic arrangements under the influence of gravity. Collino et al. (2015) showed that particle shape affected the organization of particles exposed to acoustic excitation. Rods, bricks, and bowtie-shaped objects yielded different types of “brick and mortar” structures having regular spacing.

For lignocellulosic materials, the most notable examples of shape-related organization are associated with cellulose nanocrystals (Habibi et al. 2010; Lavoine et al. 2019). Thus, Han et al. (2013) showed that a wide variety of CNCs and other forms of cellulosic nanomaterials could form various regular patterns in aqueous suspension, depending on the details. Lagerwall et al. (2014) explained such phenomena based on a competition between the formation of a glass-like organization and liquid crystal self-assembly. By adjusting various details, it is possible to achieve striking visual effects (Parker et al. 2018). Kawakatsu et al. (2006) used computer simulations to show the mechanistic feasibility of the initial self-organization of cellulose molecules shortly after their biosynthesis, which depends on their shape.


Some examples that already have been considered show that various types of self-organization can be achieved in the laboratory that do not involve close-range forces. This would be true, for instance, in the case of packing relatively large hard spheres in a gravity field (Li et al. 2008). But in many cases the subsequent usefulness of a self-organized object depends on the action of forces of cohesion or adhesion. As indicated in the subjections that follow, these can involve van de Waals-type attractions, electrostatic forces, hydrogen bonding, and some related effects such as π-type bonding between aromatic groups associated with lignin structures.

Some general principles have become established regarding the relative magnitude of short-range forces relative to processes of self-assembly. Successful self-organization often requires that there be a suitable relationship between the magnitudes of attractive forces relative to the strength of chaotic effects such as Brownian diffusion or applied vibrations during the assembly process (Suo and Lu 2000). Especially if the aim is to create highly regular arrangements, it will be important for the aggregated system to have sampled a large number of arrangements. Each time, the structure is able to partially disassociate and then reassociate itself along a path leading to a more organized arrangement.

van der Waals forces

When the term “van der Waals forces” is used, it most often refers to the London dispersion component. This component arises due to mutually-induced momentary dipoles associated with the whirling motions of electrons in their orbits. In nature, such transient dipoles align themselves in such a way as to minimize free energy. Progress has been achieved in understanding such interactions in terms of electron density functions (Klimes and Michaelides 2012). Figure 3 presents a schematic illustration of three key components of force that are often involved with self-assembly in the absence of covalent bonding.

Fig. 3. Three key components of short-range interactive forces that are often involved with self-assembly phenomena that do not depend on the development of covalent bonds

For most systems of interest to scientists working with lignocellulosic materials, often in aqueous systems or in air media, the dispersion forces are attractive, acting to draw all objects – whether at the molecular level or larger – toward each other. Articles that have discussed the role of van der Waals (London) dispersion forces in the context of self-assembly are listed in Table 3. Of particular relevance for cellulosic systems are the articles showing the dominance of dispersion forces relative to certain aspects of cohesion between adjacent cellulose chains (Saxena and Brown 2005; Khazraji and Robert 2013; Wang et al. 2013). In addition, Besombes and Mazeau (2005a,b) noted a tendency for the relatively hydrophobic lignin entities to adsorb onto the hydrophobic (200) crystalline face of cellulose.

Table 3. Publications Providing Examples in Which van der Waals Dispersion Forces Appear to be Involved with Self-assembly

Electrostatic forces

Electrostatic forces were already mentioned in a previous subsection having to do with forms of information. Thus, the arrangements of charge on a molecule or on a particle could help to explain a resulting self-organizing arrangement. Table 4 lists some articles in which the action of electrostatic forces has been cited to explain self-assembled systems. A key advantage of electrostatic forces, from the standpoint of product development, is that one can envision strategies based on orders of addition. This is especially relevant when using polyelectrolytes, since the summative effect of multiple charge interactions can favor a stable adsorbed state (Dobrynin and Rubinstein 2005; Szilagyi et al. 2014). Such strategies are commonly employed in the paper industry, which will be considered near the end of this article.

Table 4. Publications Providing Examples in Which Electrostatic Forces Appear to be Involved with Self-assembly

Hydrogen bonding

In view of the numerous and highly regular occurrence of hydroxyl groups in both cellulose and hemicellulose chains, there is reason to expect that hydrogen bonds will play a dominant role relative to the self-assembly that involves those components in aqueous systems. Table 5 lists articles in which hydrogen bonding was mentioned as a likely organizing factor in self-assembly.

Table 5. Publications Providing Examples in Which Hydrogen Bonding Appears to be Involved with Self-assembly

In addition to articles that attribute polysaccharide self-assembly effects to hydrogen bonding, there are additional articles that further bolster the accumulated evidence that such self-association has been broadly observed (Mora et al. 1986; Lang and Burchard 1993; Westbye et al. 2007; Grigoray et al. 2014). Whereas individual hydrogen bonds between polymers in an aqueous system can be easily displaced by water molecules, extensive and regular sheets of them can hold tenaciously together. Mora et al. (1986) demonstrated such effects through the usage of agents designed to disrupt hydrogen bonding.

Pi (π) bonding

A common variety of pi (π) bonding involves an attraction arising due to the delocalized electron cloud of an aromatic ring or rings. Figure 4 provides an illustration, based on a figure provided by Proteopedia (n.d.), using benzene as an example.