Demystifying Cellulose

A primer outlinines the inherent challenges for biofuel production using biomass
By Scott Mowrey | October 11, 2013

Raw cellulose from plant biomass is composed of six classes of materials with varying percentages and widely different physical and chemical properties: cellulose, hemicelluloses (five carbon sugar chains), lignins, ash, protein and water.  

Plants harvested later in maturity will have lower protein content and crop residues will have the lowest of all. Since cellulose is the glucose polymer, it is sometimes referred to as glucan. The hemicelluloses are polymer chains of the five (pentose) and six (hexose) carbon sugars of various orientations and are designated by the base sugar name plus the “an” ending. Lignins are not at all chemically related to carbohydrates and are composed of various forms of propanyl substituted phenolic ring compounds. A common analysis for switchgrass composition is shown in the accompanying table.

Air-dried forage crops would usually equilibrate in open air around 10 percent. The hemicellulose components, arabinan to mannan, comprise over 25 percent of the dry basis matter available and 45 percent of the fermentable material available and are therefore an important component to include in the process mix.  

Structure of Cellulose

Reducing cellulose to its raw components is a challenging venture. The natural function of plant biomass to provide structure and environmental resistance to the elements is the same characteristic that makes it difficult to process. Several layers of protection have evolved over the course of time. At the center of the biomass strand complex are cellulose polymer chains composed of end-to-end glucose monomers attached in a 1-4 ring configuration. Starch, which is easier to hydr      zolyze, has both 1-4 and 1-6 ring linkages creating a noncrystalline (amorphous) structure. This opens sites for enzyme activity, hence the ability of endoenzymes such as alpha-amylase to effect the shotgun approach to breaking up and liquefying the starch molecule. Additional strength to the cellulose strand is gained in the 1 to 4 linkage via a tighter  and less flexible angle than what exists in starch.

Strong Hydrogen Bonds 

This rigid and consistent pattern creates a crystalline structure with repeated patterns that lead to the ability for these chains to form sheets of cellulose strands. These rigid sheets allow cross-sheet hydrogen bonding to occur between the hydrogen and oxygen atoms of different layers creating a very tight and formidable bond strength between sheets. This same hydrogen bonding is the property that gives water its unique physical properties (such as high boiling point, high latent heat of vaporization, high freezing point and many others) as compared to other hydrogen compounds in the oxygen family. An example of the strength of the hydrogen bond is the expanding force created by water as it freezes to ice, a phenomenon that occurs due to a reduction in the hydrogen bonds that hold the same mass of water in a smaller volume. These same hydrogen bonds hold the sheets of cellulose together so perfectly that experimental observation with hydrogen isotopes have shown that only hydrogen ions can transport themselves to the inner layers of the cellulose molecule. This is especially amazing to contemplate when you consider that hydrogen ions are composed of only one proton.  A very small space, indeed! 

Cellulose strands consisting of glucose monomers are also referred to as glucans. 

Weakly Attached Hemicelluloses

The hemicellulose component of the biomass complex is composed of various five- and six-carbon sugars in chains of varying lengths and composition. They do not possess the regular pattern of cellulose and are wrapped loosely around the cellulose microfibril strands by way of glycosidic bonds and to the lignin complex via ether and ester bond sites with the lignin molecule. The hemicelluloses are the most weakly attached of the biomass components and are often easily removed with water in a first step to process the biomass. As with glucose and glucan in cellulose, the hemicellulose chains are designated with the base sugar and the “an” ending to denote their composition. Examples include arabinose/arabinan, xylose/xylan, mannose/mannan, etc.

Megamolecule Lignin

Attached at various sites, along only the outside layer of the cellulose strand complex are the very amorphous megamolecules of lignin. Lignin is hydrocarbon in structure comprised of carbon and hydrogen in aromatic ring structures and short three-carbon (propanyl) aliphatic chains, possessing some double bonds but the majority being of a single-bonded structure. Oxygen is also present in the megamolecule in internal and external ether bonds, external ester bonds and in internal hydroxyl groups substituted on the ring structures and propanyl chains. The presence of oxygen on the order of 25 percent of the mass of the lignin molecule is what leads to poor Btu content when compared to traditional fuel sources.   

Lignin can be typified in general as aromatic substituted propyl alcohols. Gymnosperm lignin (pine species) contain mostly coniferyl alcohols, angiosperms contain coniferyl and sinapyl alcohols, and grass lignin is comprised of coniferyl, sinapyl and paracoumaryl alcohols. The lignin molecule is attached to the hemicellulose strand via either ether or ester chemical bonds.  Plant species vary as to which is the more prevalent, for example corn stover possessing more ester bonded lignin and wheat straw possesing more ether bonded lignin. 

Pretreatment Approaches 

Acid hydrolysis, the oldest of pretreatment methods, breaks the ether bond by inserting a water molecule in the gap. Sulfuric acid is the acid of choice due to its ionization strength, industrial availability and price competitiveness. The difficulties in the acid hydrolysis process are from the unselective nature of the reaction, due to the fact that not only are the ether bonds between lignin and cellulose cleaved, but the potential to hydrolyze the 1-4 bond within the cellulose and hemicellulose strands is also present. The resultant mixture often adds challenges with the varying degrees of polymer chain monomerization that have occurred. In addition, the mix of water soluble, water insoluble and unconverted biomass further complicates the process stream. Finally, the need to neutralize the mixture for additional downstream processing adds expense and further complicates the components carried in the mix.  

Bases can also be used to hydrolyze the complex and tend to be more selective in the site attacked. Bases seem to attack only the ester bonded components. This also has been demonstrated successfully, but like acid hydrolysis still has the lingering problem of needing to neutralize the process stream for further downstream processing. Time and temperature control can optimize the conversion in an industrial setting. Both of these methods have their limitations.  

The field is ripe for a new idea and a fresh approach to the pretreatment process.

(Editor’s Note: The excerpt above was printed as a contribution in the November issue of Ethanol Producer Magazine. Following is the remainder of the article.)


Separation Pretreatment Methods

Many schemes to pretreat the cellulose biomass for processing have been proposed.  In general, all pretreatment methods attempt to separate components in the biomass matrix and aid any additional downstream processing required.   Processes are either physical in nature (grinding, steam/ammonia explosion, heat/pressure cooking, dissolving) or chemical (acid/base hydrolysis, enzymatic conversion).  Any one or combination of these treatments processes may rise to the top in the future, but the best scheme will meet the following criteria:

1) Allow cellulose, hemicellulose and lignin to be isolated from the balance of the biomass.

2) Hydrolyze cellulose to glucose.

3) Hydrolyze hemicelluloses to their respective sugars.

4) Create product side income streams.

5) Minimize waste streams.  

6) Be able to recover any unconsumed components (solvents, acids, bases, etc.) to decrease input costs of operation.

7) Pass downstream only trace amounts of process chemicals that are enzyme and fermentation positive or, at least, neutral.

8) Pass downstream only trace amounts of process chemicals that are environmentally neutral.

9) Demonstrate reaction conversion rates high enough to maintain a level of production efficiency necessary for a plant process.

Many pure research articles in the literature identify cellulose solvents that have little fit for industrial ramp-up because of either cost or poor environmental fit.  The research is none-the-less still important. Other articles have considered mimicking systems utilized in nature. The difficulty in adopting such schemes is that nature has the benefit of time to accomplish some of these observed phenomena and is satisfied with low conversion rates, as long as enough is produced. However, if those processes can ramp up to production scale, then they may be quite viable in the long run.  One such example is the understanding of cellulose enzymes in the gut of ruminants and the potential for those bacteria to be grown in large cultivation tanks for enzyme production. 

Lignin and Its Properties

Cellulose of woody species origin will for the most part have slightly higher lignin content than herbaceous species. These species have developed the need for longer term protection and resistance to environmental factors and have therefore evolved to a lignin of more complex nature.  Woody lignin, therefore, has a higher level of water resistance and also contributes more to the structural integrity of the species.   Coniferous trees are of higher lignin content than deciduous species with the former ranging from 26 to as high as 32 percent in hemlock and deciduous species ranging from 19 to 24 percent. (2)   The the woody lignin complex has some molecular masses as high as 10,000 u (water has a molecular mass of 18). The basic building block is composed of three carbons of propyl alcohol attached to a substituted aromatic ring.

Herbaceous species lignin is less complex than woody species and exhibit less side branching and cross-linking between base lignin groups. This simpler structure has led to greater ease in removing the lignin in pretreatment schemes.  Herbaceous species, therefore, have even been found to dissolve in some common solvents. For example, Sherman et al (1) have proposed a viable pretreatment scheme to remove lignin using ammonium hydroxide.

The procedure is comprised of the following steps:

1) Separation of lignin strand with aqueous ammonia.

2) Cellulose filtrate/lignin leachate separation.

3) Ammonia removal from the leachate via distillation.

4) Lignin coagulation with acid.

5) Lignin purification with water washing.

The procedure comes from the older process where strong bases were used to free the lignin from the complex, that required larger amounts of acid to neutralize the lignin leachate and therefore has economical drawbacks.  Another advantage to using aqueous ammonia is the ability to strip the solvent ammonia from the dissolved mixture.  In addition, any residual ammonia not able to be stripped is beneficial downstream as a fermentation supplement.

Cellulose Supply

Aqueous ammonia as a means to separate the lignin is reported to work well for living herbaceous plants, but there is no mention as to its effectiveness on crop residues that are an attractive feedstock supply. Buranov and Mazza (5) report the annual world supply of wheat straw as 529 million tons and rice residue at  525 million tons. Corn stover supplies in the U.S. alone are estimated at 133.6 million tons annually.

Herbaceous species biomass not only has the advantage of slightly lower lignin content on a w/w percent basis, but also a much larger advantage of being less complex in structure.  Lignin appears to be added in the later stages of plant development as the traits related to survival develop first, such as root, leaf and seed development. Therefore, it follows that herbaceous species harvested earlier in the life cycle of the plant would contain less lignin content on a dry basis.  This idea coincides with the fact that early harvested forages are more highly digestible to ruminants (higher total digestible nutrients) than more mature plants.  However, the plant is still adding cellulose mass when it reverts to the reproductive stage of development and plant protein levels are higher at the earlier stages.  There may be an argument for growing sterile plants which never go in to a reproductive stage and therefore do not slow down in adding mass in their latter life cycle.  This is already accomplished in a number of pure forage crops.  In addition, these species would not encourage problems later as an invasive species.  Some companies have recognized this as an opportunity for future biomass market development.     

The Lignin Market Challenge

If lignin is to be removed from the plant structure complex, there must be a use for it in the market place or in some capacity at the processing facility. The development of lignin as a potential income stream parallels the development of the DDGS and wet distillers grain markets. It is believed that as the biomass industry grows, it will be necessary to find markets for lignin.  

Lignin when purified has a heat value of 20 Mj/kg compared to lignite, coal, and natural gas at around 11, 29, and 37 Mj/kg, respectively. (3,4) So in the short run, until a market is developed, the new processing facility may use extracted lignin for fuel.  Eventually, the much more attractive market will be as a petrochemical replacement.  Pyrolysis has the potential to produce lighter weight components with the heavier components maybe resembling asphalt.  This use has extremely low economic potential for developing any large market for lignin in the short term.  Sherman et al (1) report extraction costs higher than the fuel value offset by the lignin produced.  

Lignin is a hydrocarbon-rich organic polymer, second only to cellulose in abundance.  The viable biomass project must determine how to derive an income stream from a source that comprises at least 20 percent of the weight of the feedstock. 

A Curious Reaction for Cellulose

In the previously referenced work by Matthews (6), much time is spent emphasizing the closeness of the cellulose layers.  Computer modeling shows the spacing between these sheets is so close that only hydrogen ions are small enough to pass through to inside layers.  This provides an explanation to the mixed success that acid hydrolysis techniques have had with breaking cellulose to its glucose monomer. 

On the opposite end of the pH scale, strong bases have also had some degree of success breaking down cellulose for a different reason.  Clearly, the sodium or hydroxide ion is far too large to penetrate the cellulose crystal.  This is especially understandable when you consider the hydrogen ion is simply composed of one proton. It’s difficult to be much smaller than that on the level of reaction chemistry. 

A curious reaction occurs when cellulose is subjected to strong bases and even liquid anhydrous ammonia.  Fibers are reported to swell in a process called mercerization (7), which alters the chemical structure of the cotton fiber. The structure of the fiber inter-converts from alpha-cellulose to a thermodynamically more favorable beta-cellulose polymorph. Mercerizing results in the swelling of the cell wall of the cotton fiber.  In the textile industry this swelling effect is important to allow the penetration of dyes on the cotton fiber.  Matthews clearly identifies forms 2 and 3 as the more strongly H-bonded cellulose structure and the 1 alpha and 1 beta forms as those that can slide over each other.  It appears that ammonia has the ability to weaken hydrogen bonds and open up the molecular strand for further action.  Since cotton fibers are 90 percent cellulose, little lignin is present to shield the cellulose from chemical action.

Complex Cellulose Molecules

Polymerization Index is the term used to describe by the number of monomer units in a cellulose chain.  In a laboratory setting, however, viscosity is typically used to determine the effectiveness of a treatment, lower viscosity denoting shorter length chains. 

Glucose monomer units in cellulose are linked only through the 1-4 linkage of the carbons on the glucose ring.  Natural starches are present in two different forms.  Amylose, the simpler form, is linked exclusively via 1-4 linked monomers and can have a polymerization index up to 1000.  Amylopectin, the more common form and typically 90 percent of natural starch, is linked at both the 1-4 and the 1-6 junctions and can have polymerization indices up to 10,000.  This irregular pattern in amylopectin gives it an amorphous shape well accessible to enzymes. 

There is one difference to the glucose linkage in amylose as compared to cellulose that is the cause of the widely different physical and chemical properties. The angle in amylose allows the chain to be parallel to the plane of the glucose ring where the 1-4 carbon bond in cellulose causes the chain to be at a angle to the plane of the glucose ring.  This allows the cellulose monomer chains to fit in to each other tighter allowing cross-chain hydrogen bonding to occur.  Measured cellulose polymerization indices are similar to those of the longest amylopectin chains, but the ability for chains to hydrogen bond with adjacent strands adds greatly to the complexity of the cellulose molecule complex. 

James Matthews and others (6), in work conducted for the U.S. DOE National Renewable Energy Labs, Golden, Colo., extensively illustrated the different forms of cellulose as exhibited in nature.  Four forms of cellulose are discussed, namely 1-alpha, 1-beta, cellulose 2 and cellulose 3.  The main differences in these four forms of cellulose are the amount of hydrogen bonding occurring within the cellulose chains.  1-alpha and 1-beta demonstrate only interchain bonding with no cross-link hydrogen bonding from one chain to the other.  Forms 2 and 3 exhibit heavy cross-link bonding. 

Enzyme Utilization

Cellulase enzymes exist in nature, but are very slow.  Ligninases are slower yet and less efficient than cellulose enzymes.  Crawford (9) reports that less than 20 percent of lignin present was degraded by an Actinomycetes species. Kirk and Farrell (8) report the presence of an extracellular lignin and manganese peroxidases present in white rot fungi used to initiate lignin degradation.  Future process may involve similar models to these found in nature. Xylanases are already in use to utilize the typical main component in hemicellulose, xylan/xylose.  

Separation of lignin from the balance of biomass components has the potential to greatly increase cellulose hydrolysis conversion efficiencies.  Lignin can be pictured as a protective coating around the cellulose micro fibrils.  As a larger percent of the lignin is removed more of the cellulose is exposed to the materials used in pretreating the cellulose. 

Rope Analogy

One way to envision the cellulose-lignin complex is as strands of rope representing the micro fibrils of cellulose, each micro fibril consisting of many chains of glucose laced with one another.  Around each of these micro fibrils of cellulose is a thin coating of random ordered hemicelluloses wrapped around the cellulose strands.  The hemicellulose component of the matrix is much less organized, amorphous in structure, and adds little additional strength to the matrix. As such, the hemicellulose component seems to be an intermediate in the evolution of the more efficient and functional cellulose strands.  Finally, the cellulose and hemicellulose components are wrapped in an irregular fashion (think of a poor wrapping job with black electrical tape) connected to both the cellulose and hemicellulose exterior exposed strands, the black tape representing the final covering by the very complex, but amorphous lignin molecule. 


During the course of this discussion, several reasons have been stated for the stability of the cellulose complex.  Reviewing, starting from the inside out:

1) Cellulose is made up of glucose monomer units linked together in a very rigid 1-4 glycosidic bond, the angle of that bond eliminating the flexibility observed in the 1-4 amylose bond.

2) The rigidity of this single line of glucose molecules is further bolstered by the presence of 1-alpha and 1-beta intra-strand hydrogen bonds that reach from one glucose ring to another.

3) This dual source rigidity allows long strands of cellulose to fit in a consistent manner very close to each other.

4) That closeness allows cross-strand hydrogen bonding to occur, drawing the layers so close to each other that only hydrogen ions can traverse the space between molecules.

5) Accumulated cellulose chains, called microfibrils, are then encircled by hemicellulose strands, adding marginally to the overall strength of the complex

6) Finally, the complex is encircled with an amorphous, environmentally resistant hydro-carbon mega-molecule of lignin which is attached at random points to the hemicellulose and cellulose strands.

 With this illustration in mind, it would seem that the most successful long-term  schemes to utilize cellulose would first separate the lignin component from the process mix. Comprising some 25 to 30 percent of the dry weight of the biomass processed, it would also create fermentation space in the downstream process and increase overall production efficiencies.

The simplicity of the cellulosic structure means that only a small number of enzymes would be required to deal with a pure cellulose stream.  The complexity lies in the additional components that are included in the biomass matrix: some water soluble, such as xylose; some water insoluble, such as lignin; some easily fermented, such as glucose; and some with slowed difficulty, such as xylose (11); and, the utilization of all of the process streams in a closed loop system producing only usable and marketable product. 

Cellulose is the most abundant of organic compounds, comprising roughly 33 percent of all plant matter.  Availability of feedstock has never been an issue, but it will be absolutely necessary to the longevity of an upcoming industry to develop a means to provide low-cost feedstock to the processing facility while still rewarding  the supplier and without competing with current crops.  Of equal importance is an efficient processing method that utilizes to the fullest all process streams involved. 

Conventional energy sources continue to exert price pressure on renewable fuels as additional oil and gas resources are found.  Such pressure is long-term beneficial to the development of viable projects as requirements to be profitable are integrated into the early development of any competitive technologies. The biomass industry continues to be ripe for new ideas and is in no means focused in any one direction.  Technologies utilized 10, maybe 20 years from now may not even be envisioned at this moment in time.  The possibilities are exciting.

Author: Scott Mowrey
Alexandria, Ohio


1) Sherman et al., A New Process Developed for Separation of Lignin from Ammonium Hydroxide Pretreatment Solutions, (2011), American Institute of Chemical Engineers.

2) Encyclopedia of Chemical Technology, Lignin, pp 297, (1981), John Wiley and Sons, Inc.

3) Young, R.A. & Akhtar, M. (Ed.) (1997). Environmentally Friendly Technologies for the Pulp and Paper Industry.  New York, NY: Wiley.

4) Laaksometsa, C., Axelsson, E., Berntsson, T. & Lundstrum, A.  (2009).  Energy savings combined with lignin extraction for production increase:  Case study as at eucalyptus mill in Portugal, Clean Technologies and Environmental Policy, 11-77-82. 

5) Buranov, Anvar U. & Mazza, G., Lignin in Straw of Herbaceous Plants, (2008), Industrial Crops and Products, pp 237-259.  

6) Matthews, James F., High-Temperature Behavior of Cellulose 1, (2011), Journal of Physical Chemistry, pp. 2155-2166.

7) Wikipeadia,

8) Kirk, T.K. and R.L. Farrell. 1987. Enzymatic “combustion”:  the microbial degradation of lignin.  Annu. Rev. Microbiol. Vol. 41, pp 465-505.

9) Crawford, D.L. 1986. The role of actinomycetes in the decomposition of lignocellulose. FEMS Syposium, Vol. 34, pp715-728.

10) Wikipeadia,

11)  Kotter, Peter and Michael Ciriacy. 1993. Xylose fermentation by   Saccharomyces cerevisiae.  Applied Microbiology and Biotechnology. Volume 38, Issue 6, pp. 776-783.

12) Milller, P.S., and P.H. Blum. “Extremophile-Inspired Strategies for Enzymatic Biomass Saccharification.” Environmental Technology, 31.8-9 (2010): pp.1005-15.