The morphological changes upon cryomilling of cellulose and concurrent generation of mechanoradicals

Highlights

• Quantitative analysis of radicals produced by cellulose ball-milling is possible by reacting them with DPPH radical scavenger.

• The amount of radicals produced in ball-milling depends on the morphology of cellulose.

• If the crystalline domains disintegrate, more radicals can be produced.

Abstract

With mechanical input, chemical bonds in polymers can be broken. Recently, it was shown that reactive ends formed by homolytic cleavage, so-called mechanoradicals, can be used in driving further chemical reactions or in making new composite materials. Cellulose, the most abundant polymer on earth, can also be subjected to mechanical input via ball-milling to produce mechanoradicals. Despite many reports on morphological changes in cellulose upon milling, there is only a limited understanding on how these changes affect the mechanoradical production, i.e., in which domains of cellulose the bonds are broken to produce the mechanoradicals. Here we show, the effect of the initial morphology of cellulose (cotton or microcrystalline cellulose) and the mode of grinding (dry or solvent-assisted) on the amount of generated cellulose mechanoradicals. The morphological and the chemical changes taking place upon milling of cellulose are monitored by SEM, XRD, and ATR, and the number of mechanoradicals is determined by a first-time quantitative analysis of cellulose mechanoradicals using radical scavenger DPPH. Our findings can help in efficient mechanofunctionalization of cellulose and to make useful mechanochemical reactions of cellulose using mechanoradicals, which stand as a promising economic and environment-friendly alternative to the conventional solvent-assisted chemistry of cellulose.

Introduction

Cellulose is the most abundant biopolymer in nature, which contains linear chains of repeating d-glucose molecules connected by C–O–C bonds named as β-1,4-glycosidic linkages [1]. Over the past decades, due to growing interest in sustainability and green chemistry, cellulosic materials have received attention, since cellulose is highly abundant, lightweight, strong, biodegradable material, and because composite materials including cellulose can be environment-friendly, biocompatible, low cost, low weight, and multifunctional. For functionalization of cellulose and obtaining large-scale cellulose composites, mechanochemistry stands as a straightforward and green alternative [[2], [3], [4], [5], [6], [7], [8]]. Since 1921 it is known that when cellulose is subjected to mechanical input, its fibrils disintegrate physically. Cellulose mechanochemistry, on the other hand, breaks the chemical bonds in cellulose by mechanical input [[9], [10], [11]]. When cellulose is ball-milled, 1) the intermolecular hydrogen bonds and 2) covalent bonds (C–O–C glycosidic bonds and C–C bonds) are broken, and mechanoradicals are produced. There are several examples of mechanofunctionalization by using the generated ‘broken ends’; for example, by breaking intermolecular hydrogen bonds in cellulose and using free OH groups formed, esterification of cellulose was achieved [[2], [3], [4]]. On the other hand, by breaking the covalent bonds of cellulose, mechanoradicals can be produced (Fig. 1a), which can be verified and analyzed by Electron Spin Resonance spectroscopy (ESR) (Fig. 1b) [[5], [6], [7], [8], [9], [10], [11]]. Such mechanoradicals formed by the bond cleavages of polymers upon ball milling were firstly examined with ESR spectroscopy by Butyagin et al. in 1964 [12]. Later, by Sakaguchi and Sohma, a special ball-milling apparatus was designed for ESR analysis of polymer mechanoradicals [13]. In this setup, there is a direct connection between an ESR sample tube and the milling chamber under vacuum, which helps to proceed to the following ESR measurement without any sample exposure to air. Also, samples can be milled at cryo-conditions; the temperature of the sample tube is not let to increase before and during the ESR measurement, which is essential for detection of the less stable radicals. Using this method, they showed that alkyl (carbon-centered), alkoxyl (oxygen-centered), and peroxyl radicals (if the milling chamber is exposed to oxygen atmosphere during milling) are produced upon covalent bond breaking in cellulose. These mechanoradicals were then used as initiators to polymerize several monomers such as methyl methacrylate [5,6], hydroxyethyl methacrylate [7], styrene [8], and to obtain synthetic polymer-cellulose copolymers.

As mentioned above, mechanochemistry of cellulose provides green, up-scalable access to cellulose functionalization and cellulose composites, however, surprisingly, so far it has only been sparsely used for these purposes. In our opinion, the reason for this rarity is the lack of studies on the mechanoradical quantification, i.e., knowing the number of produced mechanoradicals is vital for driving reactions other than polymerization, especially when the mechanoradicals should be used in stoichiometric amounts. It is also essential to know the effect of the concurrent morphological changes on the production of mechanoradicals in the cellulose matrix. With this study, we try to provide answers to these points. First, we analyze the formation of cellulose mechanoradicals in two morphologically different samples of cellulose, i.e., cotton, which has long (up to mm-length) fibers, and microcrystalline cellulose, which has fibers of a few hundred microns. We use cryomilling to generate mechanoradicals in cellulose and determine their amount through reaction with a radical scavenger, 2,2-Diphenyl-1-picrylhydrazyl (DPPH), which loses its deep violet color (abs max. 519 nm) upon reacting with radicals [[14], [15], [16], [17], [18]]. In this study, we do not use inert atmosphere conditions in the production and the detection of the mechanoradicals since these conditions are not applicable in most practical instances, upon up-scaling or industrial mechanochemical processes.

We have shown in our previous studies, the generation of mechanoradicals in some common synthetic polymers, their further chemical reactions [14,19], and how they affect the generation and stabilization of static charges on polymer surfaces [15,16]. In those studies, we have adapted the literature procedure given for alumina and quartz [17,18] and frequently used ‘DPPH-radical scavenging’ as a reliable method for quantification of mechanoradicals in polymers generated during/after mechanoradical treatment (Fig. 1c) [[14], [15], [16]]. Using a similar procedure, in this study, we determined the number of mechanoradicals produced in cellulose upon solvent-assisted milling (a method that does not cause a significant change the crystallinity of the samples, because of ‘dampening action’ [20]) and ‘dry’ milling (upon which percent crystallinity of the samples are changed drastically). In parallel, we monitored the morphological changes in cellulose during mechanical treatment with SEM, FTIR-ATR, and XRD - common tools used for tracking cellulose and polymer mechanodegradation. We show, even without a significant change in crystallinity upon milling, cellulose samples can produce significant number of radicals-because of the bond-breakages occurring predominantly at the amorphous domains. On the other hand, a greater number of mechanoradicals are produced when crystalline domains are transformed into amorphous regions during (dry) milling.