Nanocellulose or nano-structured cellulose is general term used for cellulose fibers with diameters around 5–20 nm and varying degree of length (depending on source and processing parameters). Generally, nanocellulose includes three categories of cellulose, which are 1) Cellulose Nano Fibrils (CNF)/Nano Fibrillated Cellulose (NFC), 2) Cellulose Nano Crystals (CNC)/Cellulose Nano Whiskers (CNWs), and 3) Bacterial Nano Cellulose (nano structured cellulose of bacterial origin, BNC) (Klemm et al., 2011).

The classical approach to synthesize CNC is by acid hydrolysis (mostly sulfuric acid; phosphoric acid and hydrochloric acid are also reported) (Abitbol et al., 2016). Recently, two new methods have also been patented by the companies American Process Inc. (Nelson et al., 2014) and Blue Goose Biorefineries Inc. (Olkowski and Laarveld, 2013). These two methods were based on acid and solvent-based pretreatment to minimize mechanical energy consumption and transition metal-based nano-catalyst for biorefinery, respectively. CNF is mainly produced by mechanical processing assisted by enzymatic or chemical pretreatments. Mechanical methods to produce CNF include high pressure homogenization (HPH), microfluidization, grinding, cryocrushing, and high intensity ultrasonication (Abdul Khalil et al., 2014). HPH involves passing of cellulose slurry through a fine nozzle into a vessel. Size reduction takes place mainly due to shear forces generated at high pressure, velocity, and impact forces (Figure 1). First reports of the HPH process are as old as 1983 (Herrick et al., 1983). Microfluidizer is similar to HPH with an additional pump to generate high pressure streams. Interaction chamber here collides the streams and walls to create the fibrillating forces (Ferrer et al., 2012). Grinding involves passage of cellulose slurry between the static and mobile grinding stones, and motion of the stone provides the fibrillating force to get nanoscale fibers (Wang et al., 2012). In cryocrushing, water swollen cellulosic fibers frozen by liquid nitrogen are crushed using a mortar and pestle. The crushing impact forces and force exerted by ice crystals during the process create the liberating force for cellulose fibers from the cell wall of plant materials (Siró and Plackett, 2010). During high intensity ultrasonication, cavitation leads to powerful mechanical oscillating power involving formation, expansion, and implosion of microscopic gas bubbles due to absorption of ultrasonic energy by molecules (Chen et al., 2013).

Bacterial nanocellulose or biocellulose is the cellulose synthesized by the action of microorganisms. It is synthesized mainly by the bacterium Gluconacetobacter xylinus (also named as Acetobacter xylinus). Some other microorganisms also exhibit the ability to synthesize the cellulose, such as other species of Gluconacetobacter, Agrobacterium tumefaciens, Rhizobium spp., and Gram-positive Sarcina ventriculli (Tanskul et al., 2013; Mohammadkazemi et al., 2015; Jozala et al., 2016). G. xylinus is the primary microbe producing the biocellulose and commonly studied as a model system for the study of the biosynthetic mechanisms of cellulose synthesis. For the production of cellulose, G. xylinus secretes a nanofibrillar film with a denser lateral surface and a gelatinous layer on the opposite side (Kurosumi et al., 2009). The process of cellulose synthesis by G. xylinus can be explained in three steps: 1) polymerization of glucose residues, 2) extracellular secretion of linear chains, and 3) organization and crystallization of glucan chains (from Step 1) in a hierarchy into fibrils/strips (Klemm et al., 2011).

In regard of the established applications of the above three varieties of nanocelluloses, bacterial nanocellulose based biomedical products (wound care and wound healing) are already being marketed. Botanical-biomass based celluloses are mainly used in reinforcement and composite applications. There are several reasons for easy acceptability of bacterial nanocellulose in biomedical applications. First is its inherent purity. It is free of lignin, hemicelluloses, pectin, and other plant-based phenolic compounds (Chawla et al., 2009), and therefore it is simpler to purify as compared to botanical-NC. Second, bacterial-NC possesses a highly porous structure with high water absorption capacity that can help in absorbing exudates from wounds (Ullah et al., 2016). Finally, the lack of impurity provides a higher number of hydroxyl groups for surface modification, harmless degradation products (only glucose), negligible amount of endotoxin (approved by FDA for surgical sheets and tissue reinforcements), higher complement activation parameters as compared to alternates, slower blood-coagulation in comparison to clinically available materials, and lack of skin irritation potential (Fink et al., 2011; Petersen and Gatenholm, 2011; Almeida et al., 2014) and hence improved compatibility properties. On the other hand, the botanical-biomass based nanocellulose carry remnants of chemicals from processing steps, which can be explained as a main reason for their less acceptability in applications requiring a cleaner surface. Although new reports of botanical-biomass based nanocellulose drug delivery and wound care system are showing promising results, this area needs more reports and tests to reach a reasonably strong conclusion.

Cellulose is a linear polymer of D-glucopyranose units linked by β-1,4-glycosidic bonds. At molecular levels, it is composed of carbon (44.44%), hydrogen (6.17%), and oxygen (49.39%). Cellulose can also be represented by a basic chemical formula of (C₆H₁₀O₅)n; where n is called as the degree of polymerization (DP), attributed to the number of glucose groups, ranging from hundreds to thousands, varying from source to source and also affected by the method of processing for obtaining the cellulose. Traditionally, β-cellulose and γ-cellulose (Table 1) constitute industrial hemicellulose, and holocellulose refers to all the carbohydrates (cellulose and hemicellulose).

In the chemical structure of cellulose (Figure 2), the primary structure (linear chain and stereo-chemical structure) shows a cellobiose unit, reducing end (open ring structure), and non-reducing end (close ring structure). Within the linear chain, every glucose unit can be seen rotated by 180° and new glucose units are added at non-reducing ends deciding the direction of chain growth. The intramolecular hydrogen bonding can be seen in hydroxyl groups attached to the 2nd and 6th carbons, hydroxyl groups attached to the 3rd carbon, and adjacent oxygen from adjacent molecules. The intramolecular hydrogen bonds and van der Waal forces come into play during stacking of cellulose chains one above the other, between hydroxyl groups attached to the 3rd and 6th carbons of stacking chains. Surface characteristics of cellulose are most important for surface modifications, which in turn play a pivotal role in final product applications. Starting from the molecular level, each anhydro-glucopyranose unit does not lie in the plane of their structure but they assume a chair form (lowest energy conformational isomer) and each next molecule is rotated by 180° about the molecular axis with hydroxyl group at the equatorial position (Habibi et al., 2010). Each cellulose unit has three hydroxyl groups in cellulose, at the 2nd, 3^rd, and 6th positions (Figure 2). At the 6th position, the hydroxyl group behaves as a primary carbon but at the 2nd and 3rd positions it acts as a secondary alcohol. Comparing the reactivity, the hydroxyl group at the 6th position shows 10 times more reactivity than the other two and the hydroxyl group at the 2nd position is twice as reactive as that at the 3rd position.

However, in the case of surface characteristics of nanocellulose, the chemical nature of the reactive agents and solvents also plays a major role in determination of reactivity. The reactivity of the hydroxyl group toward the nucleophilic attack was reported as almost the same for the 6th and 2nd carbon, which is more than reactivity of the hydroxyl group at the 3rd carbon (de la Motte et al., 2011; Lin and Dufresne, 2014).

In the specific case of CNC, it is produced by hydrolysis of an amorphous region of cellulose using sulfuric acid, leaving behind the crystalline part. This process results in negatively charged sulfate esters (condensation esterification/sulfation) on the CNC surface, which determines most of the surface characteristics of resultant material. As the surface charge is mainly due to sulphate esters, it is affected by duration and temperature of sulfation reaction (hydrolytic reaction).

Several methods based on XRD and ¹³C NMR spectra have been reported to study the crystallinity of cellulose. Based on XRD data, there can be a peak height method (Segal et al., 1959), peak deconvolution method (Hult et al., 2003; Garvey et al., 2005; He et al., 2008), and amorphous subtraction method (Ruland, 1961). In the NMR-based method used is C₄ peak separation, the peak height method was based on a ratio of heights of 002 peak and minimum between 002 and 101 peak. This method worked well for comparative studies but had several limitations, too. Limitations include underestimation of a minimum due to its non-alignment with amorphous peak (Park et al., 2010); due to choice of only peak one out of at least four peaks, one specific orientation gets more attention than others; peaks in the cellulose spectrum are very wide and peak height cannot be used as an exact measure (Garvey et al., 2005). The amorphous subtraction method involves subtracting the amorphous spectra from diffraction pattern and crystallinity index (CI) is calculated as the ratio of crystalline area and total area. In this case, the challenge was to choose the right amorphous standard. Peak deconvolution method requires a software to separate different peaks. Four to five peaks have been separated in different cases. This method assumes an amorphous component as the main contributor to peak broadening, but other factors like crystalline size and non-uniform strain with the sample can also lead to the same results. Another aspect is that cellulose peaks are very broad and are resolved only with peak overlaps.

In the ¹³C NMR method, the ratio of the area under C₄ peaks (assigned to crystalline cellulose at 89 ppm) and total areas of C₄ peaks (assigned to amorphous cellulose at 84 ppm and crystalline peak both) are considered as CI values. Based on ¹³C CP/MASS NMR spectra reports in which C (4) and C (6) of cellulose show different signals (Horii et al., 1982), cellulose polymorphs can be divided into two groups: Cellulose (I, III_I, and IV_I) and Cellulose (II, III_II, and IV_II). Cellulose I (the native cellulose) has cellulose chains arranged in such a way that glucopyranose rings are parallel to the “bc” plane of crystal (Gardner and Blackwell, 1974), hence exposing the glucopyranose ring from one crystal to another one. This property has been supported by evidence of binding enzymes on glucopyranose structures. The native cellulose (cellulose I) also shows two polymorphs called I_α and I_β. The former is dominant in cellulose produced by primitive organisms whereas the latter dominates in higher plants. I_α has triclinic one chain unit cell in which chains are stacked by van der Waal forces and I_β has monoclinic two chain unit cells stacking with alternating shear. The I_α gets converted into I_β upon hydrothermal treatment and some solvents. Cellulose II is the result of alkali treatment of cellulose I. Cellulose II has cellulose chains rotated by 30° from the parallel to “ab” crystal phase (most applicable for 2 chain model but some cellulose requires 8 chain models). Additionally, intermolecular hydrogen bonding is significantly complex in crystalline parts of cellulose II as compared to cellulose I. The main difference lies in the configuration of the 6th carbon. For cellulose I this arrangement is trans-gauche (tg), whereas for cellulose II it is gauche-trans (gt) (Gardiner and Sarko, 1985). The additional feature of hydrogen bonding in cellulose II over cellulose I is due to intersheet bonding of C (2) hydroxyl group of corner chain and oxygen of the central chain, which is absent in native cellulose. Cellulose II has two chain monoclinic unit cells stacked by opposite polarity (antiparallel structure). Cellulose III_I and III_II can be reversibly transformed into cellulose I and II, respectively, using ammonia. Cellulose III_I is a monoclinic one chain unit cell where parallel cellulose chains are stacked with no stagger along the chain axis. Cellulose IV_I and IV_II are possible transformed products of cellulose I and II, respectively. Although some reports suggest that direct transformation of cellulose I and II to IV_I and IV_II, respectively, does not happen in a step, there is possible formation of III_I and III_II (Wada et al., 2004). In the diffraction study of microfibrils, alternate light and dark regions have been found, which are explained as crystalline and amorphous cellulose, respectively. Amorphous cellulose has been explained by various postulates, including isotropically distributed straight chains (Fink et al., 1987), and bent and twisted chains (Paakkari et al., 1989). Furthermore, it was also postulated that light and dark bands could be because of a slight curvature which makes in and out of Bragg’s diffraction conditions (Gautam et al., 2010). More details on cellulose crystallinity are beyond the scope of this article. Authors recommend an excellent work on this topic by Park et al. (2010).

Cellulose does not get digested in the human gut due to the absence of β-1-4 glycosidic bond cleaving enzymes, which is present in ruminating animals. This property of cellulose is used for its utilization as excipient in drug delivery formulations. In the environment, microorganism degrading cellulose (called cellulases) produce two types of cellulases, namely, endoglucanases and cellobiohydrolases (CBH). Endoglucanases hydrolyze internal bonds (preferably the amorphous regions) releasing new terminal ends. CBH (exo-1,4-b-glucanases) act on the existing or endoglucanase-generated chain ends. Amorphous cellulose can be degraded by both types of enzymes, but crystalline cellulose is efficiently degraded only by CBH. Both cellulases release cellobiose molecules as the result of hydrolytic cleavage. Also, breakdown of cellobiose requires β-glucosidases, which converts it into two molecules of glucose. Further hydrolysis of glucose results in carbon and energy sources for cellulolytic microorganisms (Pérez et al., 2002).

Cellulose is relatively stable to UV absorption and decomposition as compared to lignin (Mishra and Wimmer, 2017). For the toxicity studies, each type of cellulose needs to be considered individually as the method of production affects the surface characteristics significantly, which in turn plays a significant role in toxicity. For purified cellulose, the production steps result in trace organochlorine contamination originating from the chemical reactions in the purification process. However, the chronic ingestion did not show any increase in spontaneous disease or neoplasia. Additionally, any promotional activity in the mammary gland, colon, or bladder of the rats was not reported and any negative impact on the absorption or the metabolism of dietary components was not concluded. Therefore, no adverse health effects in humans were suggested from exposure to purified cellulose (Anderson et al., 1992). For bacterial cellulose produced by Gluconacetobacter xylinus, Jeong et al. reported a study of toxicity of bacterial cellulose nanofibers in human umbilical vein endothelial cells (HUVECs) using viability and flow cytometric assays, and in C57/Bl6 mice. The absence of toxicity in vitro and in vivo supported the view that bacterial cellulose may be used as a tissue engineering biomaterial (Jeong et al., 2010). In a study done on respirable cellulose fibers, short-term inhalation of cellulose caused an inflammatory lung response which resolved despite continuing exposure. Intraperitoneal injection of cellulose fibers induced sarcomas rather than mesotheliomas at the highest dose (10⁹ fibers), while the two middle doses (10⁷ and 10⁸ fibers) each produced a mesothelioma (Cullen et al., 2002). In a review published on toxicity of CNC, it was mentioned that oral and dermal toxicity assessment of CNCs have shown a lack of adverse health effects, whereas studies on the pulmonary and cytotoxicity have yielded discordant results. Authors suggested the need for additional studies to support the general conclusion that CNCs are nontoxic on ingestion or contact with the skin and to determine whether CNCs have adverse health effects on inhalation or elicit inflammatory or oxidative stress responses at the cellular level.

Nanocellulose has gained great popularity as a drug carrier in the past few years. Owing to its several favorable characteristics, including nanosize, high surface area, bioavailability, biocompatibility, and surface tunable chemistry, different drug delivery systems could be explored. There have been hydrogels, gels, microparticles, membranes, scaffolds, and films. Not only the diversity in drug delivery systems, but also nanocellulose possesses diversity in route of administration. There are oral, transdermal, mucoadhesive, and even injectable formulations too. Further, nanocellulose has been used to deliver drugs locally, as well as systemically (Salimi et al., 2019).

One of the main applications of nanocellulose in drug delivery is the controlled and sustained release of the active pharmaceutical ingredient. Reportedly, nanocellulose sustains the drug release by forming a tight fiber network around the incorporated drug entities (Kolakovic et al., 2012). Also, the hydrogen bonding that nanocellulose forms with various drugs improves the stability and cohesion of the biopolymer matrix leading to a sustained release (Guo et al., 2017). Apart from controlled release of drugs, nanocellulose has served other purposes too in drug delivery, such as targeting, improved stability, better bioavailability, and increased solubility (Table 2). Nanocellulose can even be made to carry an imaging agent inside the body and thus aid in diagnostics (Colombo et al., 2015; Zhou et al., 2015).

Another very common biomedical application of nanocellulose is in tissue engineering. The unique three-dimensional network formed by cellulose makes it an ideal candidate for a variety of tissue engineering applications. Other properties of nanocellulose, such as its mechanical strength and biocompatibility also add to its suitability for tissue engineering (Jorfi and Foster, 2015).

Tissue engineering is a novel field where cells, biomaterials, and growth factors are combined to produce engineered-organs and tissues for replacement of damaged tissues in the human body (Moroni et al., 2014). The required properties of scaffolds to be used for tissue engineering include good mechanical properties to sustain cell proliferation for more than 4 weeks, ability to support its differentiation into specialized structured tissues, and porous gel structure to allow gases transport and promote vascularization. The growth tissue, after sufficient development, is to be incorporated into the organism, expecting a minimal inflammatory response. Further, the scaffold should ideally degrade naturally into the body, without the need of any invasive procedure for its removal (Drury and Mooney, 2003). Nanocellulose fulfills all these requirements and thus is a suitable tissue engineering material (Curvello et al., 2019).

The most common type of tissue engineering application that nanocellulose is used for is wound healing. Chronic wounds pose an increasingly significant worldwide economic burden (over £1 billion per annum in the United Kingdom alone). With the escalation in global obesity and diabetes, chronic wounds will increasingly be a significant cause of morbidity and mortality (Jack et al., 2017). Wound dressing materials should have a porous network, ability to swell, a specific elasticity, and the ability to retain moisture and pH over time (Zhong et al., 2010). Since there always are high chances of infections, it is desired for dressings to also have antibacterial properties to minimize the bacterial growth during the healing process (Curvello et al., 2019). Nanocellulose has been extensively explored for scaffold and wound dressing applications. Nanocellulose is highly versatile and can be tailored with specific physical properties to produce an assortment of three-dimensional structures (hydrogels, aerogels, or films), for subsequent utilization as wound dressing material. It can also be loaded with antimicrobial agents to prevent infection in the wound (Kupnik et al., 2020).

After wound healing, bioprinting is another innovative tissue engineering strategy which nanocellulose is commonly used for. 3D bioprinting is emerging as a powerful tool for the construction of highly structured tissue engineering scaffolds. Through this technique, a 3D bioprinter is able to precisely dispense materials in three dimensions while moving in X, Y, and Z directions, enabling the production of complex structures from the bottom up (Markstedt et al., 2015). Cells are encapsulated within the printed material gel in homogeneous density and quantity. This feature makes 3D bioprinting better than the traditional two-step process where cells are inoculated into pre-made scaffolds, leading to heterogeneous cell distribution (Mandrycky et al., 2016).

It is often challenging to identify a bioink that supports cell growth, tissue maturation, and the successful formation of functional grafts to be used in regenerative medicine. In important research, a mitogenic hydrogel system based on alginate sulfate was identified as a bioink as it supports chondrocyte phenotype, but it was not printable due to its unfavorable rheological properties. To convert alginate sulfate into a printable bioink, the researchers combined it with nanocellulose, which has been shown to possess very good printability (Müller et al., 2017). The ability to spatially control the placement of cells, biomaterials, and biological molecules makes the nanocellulose a suitable tool for bioprinting (Martínez Ávila et al., 2016).

The important studies performed on biomedical applications of nanocellulose are summarized in Table 2.

In addition to renewability, better carbon footprint, sustainability, indispensability to bioeconomy, and environmentally benignity of nanocellulose, especially in comparison to petroleum-based products, the recent developments in science and technology have placed nanocellulose in the categories of future material of innovation and production. To understand the future scope of nanocellulose, we will consider different fields individually. In material science, nanocellulose based and/or reinforced composites for materials application in food packaging involving active packaging have been reported numerous times and some commercial products are already available. Active or smart packaging, which is a developing branch of food packaging technology, can be seen as a promising application of cellulose. The transparent paper and its applications in thin film transistors, organic light-emitting-diode, organic photovoltaic devices, printed foldable antenna, and resistive paper touch screen can give some idea of future applications of cellulose in high value areas. In the biomedical field, newly reported applications in tissue engineering, implants, prosthetics, drug delivery, patches, and wound healing using films, aerogels, and hydrogels, 3D printed structures add a very specific dimension to the high value applications of cellulose. Additionally, the already marketed bio-cellulose based products for tissue engineering, wound care, and wound healing are relatively new entrants to the catalog of cellulosic biomedical products. In light of the above arguments, the future of cellulosic materials in interdisciplinary and especially in high-value applications seems very bright and these materials can be assumed to be major contributors to future sustainable bio-economies at the global scale.

Utilization of cellulose in human society is as old as utilization of paper. It has been a traditional product of the pulp and paper industry. Along with the development in the field of science, the understanding of structure and properties of cellulose has improved, which in turn has opened the doors for its newer applications. Better understanding of delignification process of wood and other lignocellulosic biomass has led to greener, economical, and environmentally favorable processes of cellulose production. Recent reports on biological compatibility and toxicity properties of cellulose have pulled cellulose from just being a tableting excipient and pH sensitive coating agent to being a drug carrier molecule and more. Cellulose is now being used in its nano form as nanocrystals and nanofibers, which are now extensively explored as drug carriers, tissue engineering material, in wound healing, 3D bioprinting, etc. A good number of pilot plant and semi-commercial scale demonstration and production units are already working across the globe and bodes well for the future of nanocellulose-based materials.

There is no specific legislative act in EU legislation regulating nanocellulose from plants. Thus, only the general REACH regulation, which regulates the manufacture, import, and registration of chemicals, can be applied. Several other regulations that follow the content of the REACH regulation will also be applied. For the registration of nanocellulose from plants, it will be necessary to submit the necessary documents for registration, in particular the exact identification of the substance. The condition of substance identification will also be necessary if nanocellulose from plants is exempted from registration for scientific and research purposes; however, this purpose must always be demonstrated by the importer or manufacturer.