Cellulose Aggregation under Hydrothermal Pretreatment Conditions

Jun 14, 2016 - Michael S. Reid , Stephanie A. Kedzior , Marco Villalobos , and Emily D. ... Umesh P. Agarwal , Sally A. Ralph , Richard S. Reiner , Ca...
0 downloads 0 Views 3MB Size
Subscriber access provided by George Mason University Libraries & VIVA (Virtual Library of Virginia)

Article

Cellulose Aggregation Under Hydrothermal Pretreatment Conditions Rodrigo L. Silveira, Stanislav R. Stoyanov, Andriy Kovalenko, and Munir S. Skaf Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.6b00603 • Publication Date (Web): 14 Jun 2016 Downloaded from http://pubs.acs.org on June 22, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biomacromolecules is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Cellulose Aggregation Under Hydrothermal Pretreatment Conditions

Rodrigo L. Silveira,† Stanislav R. Stoyanov,‡,#, ┴,§ Andriy Kovalenko#,┴ and Munir S. Skaf†,* †

Institute of Chemistry, University of Campinas, Caixa Postal 6154, Campinas CEP 13083-970, São Paulo, Brazil #

National Institute for Nanotechnology, 11421 Saskatchewan Drive NW, Edmonton, Alberta T6G 2M9, Canada ‡

Department of Chemical and Materials Engineering, 9107 - 116 Street, University of Alberta, Edmonton, Alberta T6G 2V4, Canada



Department of Mechanical Engineering, 4-9 Mechanical Engineering Building, University of Alberta, Edmonton, Alberta T6G 2G8, Canada

§

CanmetENERGY-Devon, Natural Resources Canada, 1 Oil Patch Drive, Devon, Alberta T9G 1A8, Canada *Corresponding author. E-mail: [email protected]

ACS Paragon Plus Environment

1

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 40

Abstract Cellulose, the most abundant biopolymer on Earth, represents a resource for sustainable production of biofuels. Thermochemical treatments make lignocellulosic biomaterials more amenable to depolymerization by exposing cellulose microfibrils to enzymatic or chemical attacks. In such treatments, the solvent plays fundamental roles in biomass modification, but the molecular events underlying these changes are still poorly understood. Here, the 3D-RISM-KH molecular theory of solvation has been employed to analyze the role of water in cellulose aggregation under different thermodynamic conditions. The results show that, under ambient conditions, highly structured hydration shells around cellulose create repulsive forces that protect cellulose microfibrils from aggregating. Under hydrothermal pretreatment conditions, however, the hydration shells lose structure and cellulose aggregation is favoured. These effects are largely due to a decrease in cellulose-water interactions relative to those at ambient conditions, so that cellulose-cellulose attractive interactions become prevalent. Our results provide an explanation to the observed increase in the lateral size of cellulose crystallites when biomass is subject to pretreatments, and deepen the current understanding of the mechanisms of biomass modification.

Keywords: Cell wall; Biomass; Cellulose; 3D-RISM; Entropy; Hydration

ACS Paragon Plus Environment

2

Page 3 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Introduction

Cellulose – the most abundant biopolymer on Earth and the main component of plant cell walls – remains largely underutilized due to the natural protection of plant cell walls against enzymatic and chemical attacks. Cellulose consists of linear chains of glucose residues linked together by β(1→4) glycosidic bonds and, in nature, cellulose-degrading microorganisms accomplish its cleavage by a large variety of glycoside hydrolases.1 By means of hydrogen bonding and van der Waals interactions, individual cellulose chains pack together to form crystalline supramolecular complexes,2 which occur in plants as two polymorphs, Iα and Iβ, with a predominance of the latter.3,4 Such supramolecular complexes take the form of microfibrils and constitute the main load-bearing component of the plant cell wall architecture.5-7 A noncrystalline matrix of hemicellulose and lignin surrounds the cellulose microfibrils and creates a barrier which protects the cellulose from enzymatic degradation.1,8 As the enzymes only act over single cellulose chains taken from crystalline microfibrils,9 two barriers need to be overcome for cellulose hydrolysis: the non-cellulosic matrix and the crystalline structure of the cellulose. These natural protective barriers generate the so-called plant biomass recalcitrance. Thermochemical pretreatments such as hot water, dilute acid and steam explosion processes1 expose cellulose to facilitate its hydrolysis by solubilizing hemicellulose,10 cleaving lignin-carbohydrate complexes,1 modifying and redistributing lignin,11,12 and increasing cell wall porosity.13,14 Moreover, pretreatments induce morphological alterations in cellulose, such as microfibril aggregation into larger bundles12 and, in some cases, alter the cellulose polymorph.1517

These modifications contribute to decreasing the biomass recalcitrance and improving its

ACS Paragon Plus Environment

3

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 40

digestibility, but the thermodynamic drivers of such effects remain largely unexplored, especially owing to the complexity of such materials across many length scales. Despite the fact that individual cellulose microfibrils are synthesized as single structural units (microfibrils) by cellulose synthase complexes in plant cells, cellulose microfibril aggregation can happen naturally to produce larger structural units (macrofibrils).5-7,18-20 In addition, cellulose microfibril aggregation can be enhanced by hydrothermal treatments.17,18,21,22 In fact, studies based on NMR23, neutron scattering and X-ray diffraction techniques12,17,21,22,24 indicate that aggregation is the main alteration regarding cellulose organization in hydrothermal treatments of lignocellulosic biomass. For instance, powder X-ray diffraction of sugarcane biomass after hydrothermal, dilute acid or steam explosion treatments shows a sharper (200) diffraction peak of cellulose in relation to the native material, which is indicative of an increase of the average crystallite width.21 Similar effects have been observed in aspen17,22 and switchgrass.12 Thus, bundles of cellulose microfibrils are ubiquitous in pretreated plant biomass and, therefore, constitute the main target of enzymatic attacks. A common feature of these thermochemical processes is the active role of solvent in shaping the free energy landscape over which cellulose modification takes place. To this extent, computational molecular simulations have been successfully employed to study many aspects of cellulose from a molecular point of view.25-32 For instance, the work of decrystallization of chains from the cellulose surface has been computationally evaluated in different solvents, revealing that this process is nonspontaneous in water (∆G > 0),27,29 but spontaneous in ionic liquids (∆G < 0).26,31 As such, fully accounting for solvent effects is of utmost importance for accurately modeling the effective forces that govern the cellulose behavior during pretreatment. Molecular dynamics simulations are able to account for these effects by explicit sampling of

ACS Paragon Plus Environment

4

Page 5 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

solvent configurations around the cellulose, but are time-consuming, and the convergence of thermodynamic observables is a challenge. Molecular theory of solvation is an alternative approach for exploring solvent-mediated effects in plant biomass and other biomolecular systems.33-35 Also known as 3D-RISM-KH (three-dimensional reference interaction site model with the Kovalenko-Hirata closure approximation), this theory is based on the first principles of statistical mechanics and has been applied to practical problems. Starting from spatial coordinates and interaction potentials (force field), the 3D-RISM-KH method provides full ensemble-averaged 3D distribution functions (normalized local densities) of solvent molecules around a solute macromolecule, from which thermodynamic observables, such as free energy of solvation, are computed analytically.33-36 Thus, the 3D-RISM-KH theory directly connects atomistic information to thermodynamics. Recently, we have applied the 3D-RISM-KH method to unveil hemicellulose- and ligninmediated nanoscale forces present in plant cell walls and provide a molecular-level picture of such

interactions.37,38

Here,

we

employ

3D-RISM-KH

to

examine

water-mediated

thermodynamic forces that govern cellulose microfibril aggregation under ambient and hydrothermal treatment conditions and provide a detailed analysis of the thermodynamic drivers of such a process.

ACS Paragon Plus Environment

5

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 40

Methods

3D-RISM theory 3D-RISM is an integral equation theory of liquid state that allows computation of solutesolvent correlation functions carrying ensemble-averaged structural information for a system containing a solvated macromolecule. Here, an overview of the theory is presented. Details can be found in several other references.33-36 The 3D-RISM integral equation reads ℎ  = ∑  d    | −  |,

(1)

where hγ(r) and cγ(r) are, respectively, the 3D total and direct correlation functions of the solvent site γ around the solute macromolecule (here, α and γ are indexes that indicate solvent sites). The solvent-solvent susceptibility between the sites α and γ, χαγ(|r - r’|), carries information about the solvent interactions (water-water interactions) in absence of the solute (cellulose), and determines how the local structure of the solvent responds to the immersion of the solute into the bulk solvent. The solvent susceptibility functions are computed beforehand the 3D-RISM calculations using the dielectrically consistent one-dimensional RISM theory (DRISM).33,35,39. The 3D distribution functions gγ(r) (normalized local density distribution relative to the bulk distribution) are obtained from the total correlation functions as gγ(r) = hγ(r) + 1.33-36 To be solved, the 3D-RISM (and DRISM) equation must be complemented with another equation relating hγ(r) and cγ(r), referred to as a closure relation. The exact closure relation has a non-local form and is represented as an infinite diagrammatic series in terms of multiple integrals which are computationally intractable, so that approximations are needed. The Kovalenko-Hirata (KH) closure employed in this study nontrivially couples the hypernetted chain (HNC) and the

ACS Paragon Plus Environment

6

Page 7 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

mean spherical approximation (MSA) closures, and accurately describes strongly associative liquid systems.33-36 The 3D version of the KH closure reads exp −   + ℎ  −   ,   ≤ 1    =  ,  ! 1 −   + ℎ  −  ,   > 1  

(2)



where uγ(r) is the interaction potential specified with a force field, T is the solution temperature and kB is the Boltzmann constant. Once the 3D-RISM equations (1) and (2) are numerically solved for the correlation functions hγ(r) and cγ(r), the solvation thermodynamics can be obtained through analytical expressions. In the KH approximation, the solvation free energy ∆µ (excess chemical potential) is expressed in a closed form as:33-36

∆$ = ∑  d % &' ( )+ ℎ+ Θ -−ℎ+ . −   − + ℎ   /, *

*

(3)

where Θ is the Heaviside step function, ργ is the density of solvent site γ, and the sum goes over all solvent sites. The solute-solvent interaction energy ∆Euv (between solute cellulose denoted by u and solvent water denoted by v) can be obtained by

∆0 1 = ∑ %  d 2 ,

(4)

and the hydration entropy by the thermodynamic relation ∆3 = − - 4 . 4∆5

6

.

(5)

The derivative in Eq. (5) is computed using the finite difference approximation ∆3 = −

∆57∆8∆58∆ +∆

.

(6)

From these expressions, one can obtain the solvent-solvent reorganization energy by ∆0 11 = ∆$ + (∆3 − ∆0 1 .

ACS Paragon Plus Environment

(7)

7

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 40

From the solvation free energy ∆µ of two solute molecules in solution as a function of their separation d, the potential of mean force (PMF) can be obtained as:

PMF 0 and ∆ETotal > 0 or ≈0, demonstrating that CN aggregation is entropy driven.

ACS Paragon Plus Environment

20

Page 21 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 6. Decomposition of the thermodynamic functions of aggregation of (a) hydrophobic faces and (b) hydrophilic faces under ambient (blue) and hydrothermal treatment (red) conditions.

Water density over the CN surfaces The lower penalties associated with ∆ECN-Wat during aggregation of CNs under hydrothermal treatment conditions indicate that these conditions lead to the hydration shells containing water molecules less tightly bound to cellulose. To understand this point, we solved the 3D-RISM-KH equations for an isolated 36-chain cellulose microfibril with both hydrophilic

ACS Paragon Plus Environment

21

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 40

and hydrophobic faces exposed to water. From the 3D distribution function of water oxygen g(r), we obtained the normalized density profile g(z) along the direction normal to the hydrophilic and hydrophobic faces of the cellulose microfibril by averaging g(r) over the corresponding cellulose faces. The g(z) functions obtained using 3D-RISM-KH are similar to those obtained by molecular dynamics simulations under ambient conditions.54 The g(z) functions represented in Figure 7a (and in Figures S5 and S6 for the entire range of thermodynamic conditions) show that upon heating and with decreasing density, the hydration shell over the cellulose faces becomes progressively less structured and with little ordering beyond the first shell (Figure 4). This can also be seen in Figure 7b, which shows the two-dimensional hydration structure around the cellulose microfibril obtained by taking the average of g(r) only over the microfibril length. From the comparison of the hydration structure and the

ACS Paragon Plus Environment

22

Page 23 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 7. (a) Distribution function of water oxygen g(z) along the direction orthogonal to the hydrophobic (blue) and hydrophilic (red) faces of a 36-chain cellulose microfibril, averaged over the corresponding cellulose elementary fibril face, under ambient (300 K, 1.0 g cm-3) and hydrothermal treatment conditions (500 K, 0.5 g cm-3). (b) Two-dimensional hydration structure over the cellulose elementary fibril faces obtained by averaging g(r) over the length of the microfibril.

PMF at several conditions, we note that the degree of hydration shell structure correlates with the energetic barrier that separates the two PMF minima, indicating that highly structured hydration shells are more difficult to disrupt. Finally, we observe that the first hydration shell over the hydrophobic surface exhibits higher density than over the hydrophilic surface (Figure 7a). On the other hand, the second hydration shell exhibits similar density over both surfaces. Thus, the second hydration shell represents a higher contribution, relative to the first shell, to the cellulose-water interactions over the hydrophilic surface than over the hydrophobic, explaining why water molecules in the second hydration shell interacts more with the hydrophilic surface than with the hydrophobic and, therefore, why ∆ECN-Wat (and ∆µ and the PMF) starts increasing at ~4–5 Å during the hydrophilic aggregation.

ACS Paragon Plus Environment

23

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 40

Discussion

We have employed the 3D-RISM-KH molecular theory of solvation to study hydration effects on CN-CN effective interactions under a wide range of thermodynamic conditions that include those present in common hydrothermal treatments, including hot water, steam explosion and dilute acid processes. Our results demonstrate the major role of hydration forces in determining the effective CN-CN interactions measured by the PMF of aggregation. While van der Waals and hydrogen bonding interactions ensure that the CN-CN interactions are always attractive, the aggregation free energy strongly depends on the thermodynamic conditions of the surrounding water that hydrates the CNs. The energetic cost for surface dehydration determines the strength of the CN-CN interactions. Our work differs from those by Sinko et al.,55-57 reporting the mechanical work needed to separate cellulose microfibrils in the absence of water or in the presence of a small number of water molecules at the cellulose-cellulose interface (to simulate moisture). Here, we consider fully solvated CNs and focus on the role played by water. The combination of high temperature and low solvent density favors CN aggregation. This result explains, at least in part, the increased crystallite width observed experimentally in cell walls subjected to hydrothermal treatment.17,21,22,58 The trends of the aggregation free energies (Figure 4b) are consistent with the fact that steam affects crystallite width more than hot water and dilute acid,17 and that drying processes also lead to cellulose aggregation.59,60 These aggregation processes have lower desolvation penalties. Time-resolved neutron scattering shows that in steam explosion treatment, aggregation of cellulose microfibrils into macrofibrillar bundles occurs during the heating phase, and that after cooling the macrofibrils do not disaggregate.58 A similar picture was obtained by molecular

ACS Paragon Plus Environment

24

Page 25 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

dynamics simulations starting from a solvent-separated bundle of cellulose microfibrils similar to that shown in Figure 1.22 According to our results, during the heating phase the hydration shells around the cellulose microfibrils become progressively less structured and so weaker forces are necessary to overcome the repulsive hydration forces needed to establish cellulose-cellulose contacts. After aggregation at high temperature, the system gets trapped in the first minimum of the PMF (corresponding to crystalline contact) and remains there after cooling back to the ambient temperature, thereby coalescing the microfibrils into macrofibrils. Another aspect observed in this study is the possibility of water-mediated CN-CN contact, especially between hydrophobic faces under ambient conditions. Such configuration corresponds to the second minimum of the PMF and the hydration forces help keeping the CNs from aggregating. This has been observed by neutron scattering.53 We also noticed that the CNCN interface is able to confine water molecules when the CNs are close to contact, which could lead to imperfect aggregation. This is what seems to happen in the simulations by Langan et al. in which the number of water molecules inside the interfibrillar region is not null after coalescence of the microfibrils at high temperature.22 According to our 3D-RISM-KH calculations, the hydration entropy increases when the CNs aggregate. This indicates that the cellulose surfaces constrain water molecules, as it has previously been suggested.25,61 Our results corroborate the conclusions of Miyamoto et al., who performed a detailed analysis of the hydration structure over the hydrophobic face of cellulose, according to which the limited hydrophobic surface area of cellulose (comprised of low-polarity C-H bonds of the glucose rings) behaves like small spherical hydrophobic objects in water, which induces water ordering around themselves and whose aggregation is entropy-driven.32 Although entropy drives aggregation, the effects of hydrothermal treatment conditions are

ACS Paragon Plus Environment

25

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 40

mainly due to decreasing cellulose-water interactions, and not due to an increasing entropic component of the aggregation free energy. In addition to solvent-related effects, other factors may affect cellulose aggregation in plant cell walls during hydrothermal treatments. It has been suggested that twisting of cellulose microfibrils could affect cellulose aggregation in native cell walls, since such a twisting would impair the alignment of microfibrils.53 According to molecular dynamics simulations, cellulose microfibrils untwist at high temperature,28 which could help ease the aggregation observed during pretreatments. Another factor is the presence of hemicellulose chains that coat cellulose microfibrils and prevent their coalescence in native cell walls,62 but hemicellulose dissolution in pretreatments1 can set the microfibrils free to aggregate and also provide thermodynamic forces to drive this process.37 The relative importance of each of these factors in controlling cellulose aggregation during pretreatments remains elusive so far. A thorough understanding of cellulose aggregation has implications for comprehending not only the fundamentals of biomass thermal treatments, but also cell wall growth63 and biomechanics.64 It has been established that these processes depend on how the cellulose microfibrils are organized within cell walls and how susceptible their native arrangements are to internal (turgor pressure) and external (tensile stress) forces.

ACS Paragon Plus Environment

26

Page 27 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Conclusions

We have employed the 3D-RISM molecular theory of solvation with the KH closure relation to study water-mediated forces that modulate effective interactions between cellulose microfibrils under a wide range of thermodynamic conditions. Although CN-CN interactions are always attractive and contribute to cellulose aggregation, the hydration shells surrounding the CNs strongly influence how effectively microfibrils interact with each other. Repulsive, shortrange hydration forces act on the CNs as they approach one another, arising from the highly structured hydration shells surrounding the CN surfaces. However, with increased temperature and decreased density – conditions commonly present in hydrothermal treatments – solvent effects are decreased, resulting in a propensity of the cellulose microfibrils to directly interact with each other and aggregate. In addition, we found that cellulose aggregation is an entropydriven process and that decreasing of cellulose-water interactions are the main reasons for enhanced aggregation under hydrothermal treatment conditions. The results presented here reveal the thermodynamic forces underlying the experimental findings that high-temperature treatments lead to an average increase of cellulose microfibril diameter and help understanding the physico-chemical aspects involved in biomass pretreatments, with implications for advancing plant biomass conversion technologies.

ACS Paragon Plus Environment

27

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 40

Associated Content Supporting Information. Details of the 3D distribution function at CN separations corresponding to the maximum and second minimum of the PMF, g(z) functions, and water dielectric constants for the entire range of thermodynamic conditions considered in the study. This material is available free of charge via the Internet at http://pubs.acs.org.

Author Information Corresponding author *E-mail: [email protected] Notes The authors declare no competing financial interest.

Acknowledgements This work was supported by the Sao Paulo Research Foundation FAPESP (grants 2013/08293-7 and 2014/10448-1). R.L.S. thanks Caroline Simões for critically reading the manuscript. S.R.S thanks John M. Villegas for discussion. The computations were performed on the highperformance computing facilities of Compute/Calcul Canada WestGrid and at the Center for Computational Engineering and Sciences (CCES) at Unicamp, Campinas, Brazil. A.K. thanks the Natural Science and Engineering Research Council of Canada (NSERC) and National Research Council of Canada (NRC) for support to this work.

ACS Paragon Plus Environment

28

Page 29 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

References (1)

Chundawat, S. P. S.; Beckham, G. T.; Himmel, M. E.; Dale, B. E. Annu. Rev. Chem. Biomol. Eng. 2011, 2, 121-145.

(2)

O’Sullivan, A. C. Cellulose 1997, 4, 173-207.

(3)

Nishiyama, Y.; Langan, P.; Chanzy, H. J. Am. Chem. Soc. 2002, 124, 9074-9082.

(4)

Nishiyama, Y.; Sugiyama, J.; Chanzy, H.; Langan, P. J. Am. Chem. Soc. 2003, 125, 1430014306.

(5)

Ding, S. -Y.; Himmel, M. E. J. Agric. Food Chem. 2006, 54, 597-606.

(6)

Ding, S. -Y.; Liu, Y. S.; Zeng, Y.; Himmel, M. E.; Baker, J. O.; Bayer, E. A. Science 2012, 338, 1055-1060.

(7)

Ding, S. -Y.; Zhao, S.; Zeng, Y. Cellulose 2014, 21, 863-871.

(8)

Himmel, M. E.; Ding, S. -Y.; Johnson, D. K.; Adney, W. S.; Nimlos, M. R.; Brady, J. W.; Foust, T. D. Science 2007, 315, 804-807.

(9)

Divne, C.; J. Ståhlberg, J.; Teeri, T. T.; Jones, T. A. J. Mol. Biol. 1998, 275, 309-325.

(10) Kumar, F.; Mago, G.; Balan, V.; Wyman, C. E. Biores. Technol. 2009, 100, 3948-3962. (11) Donohoe, B.; Decker, S.; Tucker, M.; Himmel, M. E.; Vinzant, T. Biotechnol. Bioeng. 2008, 101, 913-925. (12) Pingali, S. V.; Urban, V. S.; Heller, W. T.; McGaughey, J.; O’Neill, H.; Foston, M.; Myles, D. A.; Ragauskas, A.; Evans, B. R. Biomacromolecules 2010, 11, 2329-2335. (13) Meng, X.; Foston, M.; Leisen, J.; DeMartini, J.; Wyman, C. E.; Ragauskas, A. J. Biores. Technol. 2013, 144, 467-476. (14) Chundawat, S. P. S.; Vismeh, R.; Sharma, L. N.; Humpula, J. F.; Sousa, L. C.; Chambliss, C. K. Jones, A. D.; Balan, V.; Dale, B. E. Biores. Technol. 2010, 101, 8429-8438. (15) Igarashi, K.; Wada, M; Samejima, M. FEBS J. 2007, 274, 1785-1792. (16) Chundawat, S. P. S.; Bellesia, G.; Uppugundla, N.; Sousa, L. C.; Gao, D.; Cheh, A. M.; Agarwal, U. P.; Bianchetti, C. M.; Phillips, G. N.; Langan, P., Balan, V.; Gnanakaran, S.; Dale, B. E. J. Am. Chem. Soc. 2011, 133, 11163-11174.

ACS Paragon Plus Environment

29

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 40

(17) Nishiyama, Y.; Langan, P.; O’Neill, H.; Pingali, S. V.; Harton, S. Cellulose 2014, 21, 1015-1024. (18) Ioelovitch, M. Acta Polym. 1992, 43, 110-113. (19) Donaldson, L. Wood Sci. Technol. 2007, 41, 443-460. (20) Thomas, L. H.; Forsyth, V. T.; Šturvá, A.; Kennedy, C. J.; May, R. P.; Altaner, C. M.; Apperley, D. C.; Wess, T. J.; Jarvis, M. C. Plant Physiol. 2013, 161, 465-476. (21) Driemeier, C.; Pimenta, M. T. B.; Rocha, G. J. M.; Oliveira, M. M.; Mello, D. B.; Mazieiro, P.; Gonçalves, A. R. Cellulose 2011, 18, 1509-1519. (22) Langan, P.; Petridis, L.; O'Neill, H. M.; Pingali, S. V.; Foston, M.; Nishiyama, Y.; Schulz, R.; Lindner, B.; Hanson, B. L.; Harton, S.; Heller, W. T.; Urban, V.; Evans, B. R.; Gnanakaran, S.; Ragauskas, A. J.; Smith, J. C.; Davison, B. H. Green Chem. 2014, 16, 6368. (23) Sun, Q.; Foston, M.; Sawada, D.; Pingali, S. V.; O’Neill, H. M.; Li, H.; Wyman, C. E.; Langan, P.; Pu, Y.; Ragauskas, A. J. Cellulose 2014, 21, 2419-2431. (24) Cheng, G.; Zhang, X.; Simmons, B.; Singh, S. Energy Environ. Sci. 2015, 8, 436-455. (25) Bergenstråhle, M.; Wohlert, J.; Himmel, M. E.; Brady, J. W. Carbohydr. Res. 2010, 345, 2060-2066. (26) Cho, H. M.; Gross, A. S.; Chu, J. W. J. Am. Chem. Soc. 2011, 133, 14033-14041. (27) Beckham, G. T.; Matthews, J. F.; Peters, B.; Bomble, Y. J.; Himmel, M. E.; Crowley, M. F. J. Phys. Chem. B 2011, 115, 4118-4127. (28) Matthews, J. F.; Bergenstråhle, M.; Beckham, G. T.; Himmel, M. E.; Nimlos, M. R.; Brady, J. W.; Crowley, M. F. J. Phys. Chem. B 2011, 115, 2155-2166. (29) Payne, C. M.; Himmel, M. E.; Crowley, M. F.; Beckham, G. T. J. Phys. Chem. Lett. 2011, 2, 1545-1550. (30) Matthews, J. F.; Himmel, M. E.; Crowley, M. F. Cellulose 2012, 19, 297-306. (31) Gross, A. S.; Bell, A. T.; Chu, J. -W. J. Phys. Chem. B 2013, 117, 3280-3286. (32) Miyamoto, H.; Schnupf, U.; Brady, J. W. J. Agric. Food Chem. 2014, 62, 11017-11023. (33) Kovalenko, A. Pure Appl. Chem. 2013, 85, 159-199.

ACS Paragon Plus Environment

30

Page 31 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

(34) Ratkova, E. L.; Palmer, D. S.; Fedorov, M. V. Chem. Rev. 2015, 115, 6312-6356. (35) Kovalenko, A. In Springer Handbook of Electrochemistry; Breitkopf, C.; Swider-Lyons, K., Eds.; Springer, 2017. (36) Kovalenko, A. In Molecular Theory of Solvation; Hirata, F., Ed.; Kluwer Academic: Dordrecht, The Netherlands, 2003; pp 169-275. (37) Silveira, R. L.; Stoyanov, S. R.; Gusarov, S.; Skaf, M. S.; Kovalenko, A. J. Am. Chem. Soc. 2013, 135, 19048-19051. (38) Silveira, R. L.; Stoyanov, S. R.; Gusarov, S.; Skaf, M. S.; Kovalenko, A. J. Phys. Chem. Lett. 2015, 6, 206-211. (39) Perkyns, J. S.; Pettitt, B. M. Chem. Phys. Lett. 1992, 190, 626-630. (40) Gomes, T. C. F.; Skaf, M. S. J. Comput. Chem. 2012, 33, 1338-1346. (41) Kovalenko, A.; Ten-no, S.; Hirata, F. J. Comput. Chem. 1999, 20, 928-936. (42) Guvench, O; Hatcher, E.; Venable, R. M.; Pastor, R. W.; MacKerell Jr., A. D. J. Chem. Theory Comput. 2009, 5, 2353-2370. (43) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. J. Chem. Phys. 1983, 79, 926-935. (44) Uematsu, M.; Frank, E. U. J. Phys. Chem. Ref. Data 1980, 9, 1291-1306. (45) Ducker, W. A.; Senden, T. J.; Pashley, R. M. Nature 1991, 353, 239-241. (46) Butt, H. J.; Cappella, B.; Kappl, M. Surf. Sci. Rep. 2005, 59, 1-152. (47) Israelachvili, J.; Min, Y.; Akbulut, M.; Alig, A.; Carver, G.; Greene, W.; Kristiansen, K.; Meyer, E.; Pesika, N.; Rosenberg, K.; Zeng, H. Rep. Prog. Phys. 2010, 73, 036601. (48) Valle-Delgado, J. J.; Molina-Bolívar, J. A.; Galisteo-González, F.; Gálvez-Ruiz, M. J. Curr. Opin. Colloid Interface Sci. 2011, 16, 572-578. (49) Kilpatrick, J. I.; Loh, S. -Y.; Jarvis, S. P. J. Am. Chem. Soc. 2013, 135, 2628-2634. (50) Rau, D. C.; Parsegian, V. A. Science 1990, 249, 1278-1281. (51) Leikin, S.; Parsegian, V. A.; Rau, D. C. Annu. Rev. Phys. Chem. 1993, 44, 369-395. (52) Parsegian, V. A.; Zemb, T. Curr. Opin. Colloid Interface Sci. 2011, 16, 618-624.

ACS Paragon Plus Environment

31

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 40

(53) Fernandes, A. N.; Thomas, L .H.; Altaner, C. M.; Callow, P.; Forsyth, V. T.; Apperley, D. C.; Kennedy, C. J.; Jarvis, M. C. Proc. Natl. Acad. Sci. USA 2011, 108, E1195-E1203. (54) Oehme, D. P.; Downton, M. T.; Doblin, M. S.; Wagner, J.; Gidley, M. J.; Bacic, A. Plant Physiol. 2015, 168, 3-17. (55) Sinko, R.; Mishra, S.; Ruiz, L.; Brandis, N.; Keten, S. ACS Macro Lett. 2014, 3, 64-69. (56) Sinko, R.; Keten, S. Appl. Phys. Lett. 2014, 105, 243702. (57) Sinko, R.; Keten, S. J. Mech. Phys. Solids 2015, 78, 526-539. (58) Pingali, S. V.; O’Neill, H. M.; Nishiyama, Y.; He, L.; Melnichenko, Y. B.; Urban, V.; Petridis, L.; Davison, B.; Langan, P. Cellulose 2014, 21, 873-878. (59) Szcześniak, L.; Rachocki, A.; Tritt-Goc, J. Cellulose 2008, 15, 445-451. (60) Quiévy, N.; Jacquet, N.; Sclavons, M.; Deroanne, C.; Paquot, M.; Devaux, J. Polym. Degrad. Stab. 2010, 95, 306-314. (61) Selig, M. J.; Thygesen, L. G.; Felby, C. Biotechnol. Biofuels 2014, 7, 159. (62) Cosgrove, D. J. Curr. Opin. Plant Biol. 2014, 22, 122-131. (63) Cosgrove, D. J. Nat. Rev. Mol. Cell Biol. 2005, 6, 850-861. (64) Cosgrove, D. J.; Jarvis, M. C. Front. Plant Sci. 2012, 3, 204.

ACS Paragon Plus Environment

32

Page 33 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Table of Content (TOC) Graphics

ACS Paragon Plus Environment

33

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. Cellulose Iβ macrofibril comprising 7 cellulose elementary microfibrils. Cellulose nanocrystallites (CNs) at hydrophobic and hydrophilic interfaces (highlighted in bold colors) were considered in the study of cellulose-cellulose effective interactions. 40x20mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 34 of 40

Page 35 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 2. (a) PMF, hydration free energy (∆µ), interaction energy (ECN-CN), and its van der Waals and electrostatic components along the CN-CN separation coordinate for hydrophobic and hydrophilic interfaces. Water oxygen 3D distribution function around the (b) hydrophobic and (c) hydrophilic interfaces at separations corresponding to the first and second minima and the first maximum of PMF. Isosurfaces of g(r) = 3.5. 65x52mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. (a) Mean force acting on the CNs, and its components due to hydration and CN-CN interaction. (b) Hydration force and hydration free energy in logarithmic scale show the oscillatory profile and the exponential decay at short distances. 52x33mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 36 of 40

Page 37 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 4. (a) PMF (continuous lines) and hydration free energy (dashed lines) along the CN separation coordinate for the hydrophobic and hydrophilic faces under ambient (300 K, 1.0 g cm-3) and hydrothermal treatment (500 K, 0.5 g cm-3) conditions. (b) Free energy of aggregation of the hydrophobic and hydrophilic faces as a function of water density at temperatures 300, 400, 500 and 600 K. Connecting lines are plotted for the eye guidance. 61x46mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. Decomposition of the PMF of CN-CN aggregation into energetic and entropic components for hydrophobic aggregation under (a) ambient and (c) hydrothermal treatment conditions, and for hydrophilic aggregation under (b) ambient and (d) hydrothermal treatment conditions. 57x40mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 38 of 40

Page 39 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 6. Decomposition of the thermodynamic functions of aggregation of (a) hydrophobic faces and (b) hydrophilic faces under ambient (blue) and hydrothermal treatment (red) conditions. 104x137mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 7. (a) Distribution function of water oxygen g(z) along the direction orthogonal to the hydrophobic (blue) and hydrophilic (red) faces of a 36-chain cellulose microfibril, averaged over the corresponding cellulose elementary fibril face, under ambient (300 K, 1.0 g cm-3) and hydrothermal treatment conditions (500 K, 0.5 g cm-3). (b) Two-dimensional hydration structure over the cellulose elementary fibril faces obtained by averaging g(r) over the length of the microfibril. 60x44mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 40 of 40