®

Electronically reprinted from November 2016 Molecular Spectroscopy Workbench Characterizing Modified Celluloses Using Raman Spectroscopy

Raman spectra of celluloses modified for use in the pharmaceutical, food, and materials industries are compared and analyzed, with the goal of determining spectroscopic features that can be of use in aiding in the determination of physical and chemical properties. Fran Adar

ellulose is probably the most abun- the goal of then providing some insight restricted. The details of the modifications dant biopolymer on Earth and has into the chemical reactions between cel- (the chemistry of the substituent, percent C been exploited since the dawn of lulose and certain small molecules that can substitution, molecular weight, and so on) human civilization. Wood materials have change the physicochemical properties will determine the temperature of gelation been used in construction, and cotton of the reacted material. After that, we can and the viscosity of the products. Actually, and flax (linen) have been used for cloth- look at the Raman spectra of a few of these hydroxypropyl methyl cellulose (HPMC) ing and other textile applications. During modified celluloses to assess whether the is an interesting example because of the later times paper was manufactured from spectra can be useful in characterizing the variety of ways that it is used. In pharma- cellulosic sources and used for recording final products. ceutical tablets, HPMC can be used as a information. protective coating, and as a rate controlling Chemically, wood, cotton, and linen Cellulose Background release agent in timed-release formula- are all identified as cellulose, but their For the first use of cellulose in a modern tions. HPMC’s high viscosity also enables physical and chemical properties can application, think of microcrystalline its use as an ophthalmic preparation for be quite different. A lot is known today cellulose (MCC), the white powder that dry eyes. about the structure of these materials, and is one of the excipients added to pharma- To understand how these various prod- spectroscopy, in combination with X-ray ceutical tablets to aid in their dissolution ucts are made, we first look at the chemical diffraction (XRD), electron diffraction, in aqueous environments. Then there is and physical structure of cellulose. (I will and nuclear magnetic resonance (NMR), cellulose gum, which is actually methyl depend on a chapter from the Encyclopedia has been used to probe the mystery of cellulose, hydroxypropyl methyl cellulose, of Polymer Science and Technology and these structures. In more modern times, or carboxymethyl cellulose derivatives that two review articles from Comprehensive cellulose has been chemically modified provide a smooth feeling in the mouth in Natural Products Chemistry for this in- for a variety of uses in the pharmaceutical, processed food in which the fat content is formation [1–3].) Cellulose is a linear, un- food, and materials industries. The abil- ity to make these chemical modifications depends, to a large extent, on the history of the cellulose starting material—its purity, crystallinity, and fibrillar dimensions. The CH OH 2 OH CH2OH OH O H H physical measurements have enabled the O O O O O O O cellulose research community to develop HO O O O O H a microscopic model of its form, which in OH CH2OH OH CH2OH turn aids in understanding its reactivity. This column installment is an attempt to summarize this background material with Figure 1: Structure of cellulose chain. branched polymer of glucose, assembled with β-1,4 linkages. Its structure is shown in Figure 1—notice that every other glu- (a) 1100 cose unit is flipped by 180°. For this reason, the monomeric unit of the polymer is often 1000 defined as the dimer, called cellobiose. One 900 curious fact is that even though glucose is a 800 highly soluble molecule, the cellulose poly- 700 mer is virtually insoluble in water. 600

Glucose can also be assembled into 500 another biological polymer—starch. In Hydroxypropyl cellulose 400 fact, starch is a mixture of the linear amy- Intensity (counts/s) 300 Sodium carboxymethyl cellulose lose (20–25%) and branched amylopectin (75–80%). In contrast to cellulose, the 200 Methyl cellulose covalent bond linking the glucose mono- 100 Microcrystalline cellulose α 0 mers is an -1,4 linkage. Also in contrast to 500 1000 1500 2000 2500 3000 3500 4000 the properties of cellulose, starch is highly Raman shift (cm–1) soluble in water and digestible by humans. (b) Cellulose is only digestible in ruminants 800 because their stomachs contain bacteria 1457 700 with enzymes that are capable of attacking 1363

the β-1,4 linkage. 600 1407 Cellulose is actually chemically quite 1415 robust and the challenge to exploit it 500 1368

chemically is to “disassemble” it. In 1321 1460 plants, that means releasing the lignin and 400 1596

hemicellulose (other sugar polymers) that 1457 300 1374 coexist with it. But even when one has Intensity (counts/s) purified cellulose, its chemical reactivity 200 1410 will depend on the packing of the mol- 1382 ecules in the solid. At least four crystal- 100 1479 lographic phases have been identified in 0 cellulose. Cellulose I is the form which is 800 1000 1200 1400 1600 usually found in nature; it, in fact, has two Raman shift (cm–1) subforms noted α and β. Raman spectros- copy shows that the fingerprint spectra (c) 1200 of these forms are virtually identical, but 2875 2934 that there are differences in the OH region (see Figure 11 in reference 1). This points 1000 out the importance of hydrogen bonding 2917 2898 in these materials. In fact, the hydrogen 800 2973 bonding plays a large part in the stability, and, therefore, the chemical resistance of the material. A common treatment of cel- 600 3463 2888 2938

lulose is mercerization. Fibers are clamped 3433 2838 to prevent shrinkage during treatment Intensity (counts/s) 400

with NaOH; the fibers swell, increase their 2985 2981 luster, are easier to dye, and are resistant to 200 3455 2968

mildew, but have an increased tendency to 3357 3280 shrink. The cellulose is converted to form 0 II. There are two other polymorphs of cel- 2600 2800 3000 3200 3400 3600 3800 lulose; they can be generated from either Raman shift (cm–1) cellulose I or II, and curiously they retain “memory” of the initial form from which Figure 2: (a) Raman spectra of microcrystalline cellulose, methyl cellulose, sodium carboxymethyl they were generated. There is an on-going cellulose, and hydroxypropyl cellulose. (b) Spectra enlarged in the fingerprint region to better controversy as to whether there are two define the differences. (c) Spectra enlarged in the CH–OH region to better define the differences. chains per unit cells or one, and discussion of this controversy is certainly beyond the Table I: Calculation of the number of CH and OH functional groups in original cellulose scope of this short article. But there is no (MCC) and reacted celluloses if fully substituted controversy as to the importance of H- CH CH2 CH3 OH bonding in the structures, and it is the OH groups that are available for conversion to MCC 5 1 — 3 chemically modified forms of cellulose. Methyl cellulose 5 1 3 — The important point to recognize is that Sodium carboxymethyl cellulose 5 4 — — the cellulose solid has to be open enough to allow reactions to occur, and therefore Hydroxypropyl cellulose 8 4 3 3 the crystal packing of the starting material will determine the extent of the reaction. CH groups, four methylene groups (one Finally, let’s compare the spectrum of To be semipredictive of what to expect, from the original pendant group, and HPC to MCC. The OH in this spectrum is note that there are three hydroxyl groups potentially three from the conversion of assigned to the OH groups of the substitu- on each glucose molecule, but they are three OH groups to OCH2CO2H), but no tion. The CH stretching region is the most not equivalent. According to Figure 1, one expected OH groups because the proton intense of all the modified species. The hydroxyl group in the ring is hydrogen- would be replaced by sodium. In HPC bands at 2973 and 2934 cm-1 are assigned bonded to an oxygen atom in an adjacent there will be eight CH groups (five from to the asymmetric CH3 and CH2, respec- glucose ring. Another hydroxyl group the ring plus one from each hydroxyl- tively. The band at 2875 cm-1 is assigned to is attached directly to the ring, and the propyl substitution), and four methylene a combination of symmetric stretches for last hydroxyl is attached to the pendant groups (one from the original pendant CH3 and CH2. methylene group. As shown, these last two group, and potentially three from the Note that these are tentative assign- hydroxyls are available for H-bonding be- conversion of three OH groups to OCH- ments. I am guided by the textbooks on tween chains. It is not difficult to see how 2CHOHCH3) and three methyl groups, the interpretation of IR and Raman spec- the availability for H-bonding can affect and then three hydroxy groups from the tra (4). In addition, the use of these spectra and control aggregation, and as the cellu- three hydroxypropyl substitution. This in- for quantitation is still very semiquantita- lose molecule packing becomes denser, the formation is summarized in Table I. tive. In scaling at 1100 cm-1 I attempted to availability for reactions at the hydroxyl In comparing the spectra of microcrys- make the spectra comparable. But in fact, sites will be reduced. One is then led to the talline cellulose to that of methyl cellulose, there is intensity in this region from the conclusion that differences in performance we can see the effects of the additional C-C and C-O bonds of the added func- of nominally identical materials can be methyl groups. What was a small doublet tional groups. My goal is to indicate that attributed to differences in percent conver- CH2 deformation band in MCC near 1479 with enough samples and information sion. It is my goal to examine the spectra cm-1 becomes a much stronger band at on their chemistry, these spectra can be of a few modified celluloses to determine if 1457 cm-1 in methyl cellulose because of calibrated for predictions using multivari- there is information to be acquired. the presence of the additional CH3 groups. ate analysis. Figures 3a–3c show spectra of In the CH stretching region, one can argue nominally similar materials, but with dif- Examining Cellulose Spectra that the bands at 2838 and 2938 can be as- fering amounts of reacted functionality. The spectra in Figure 2 show the results signed to the symmetric and asymmetric These materials have been substituted of measurements (from bottom to top) of CH3 stretches. The apparent intensifica- with hydroxypropyl or methyl groups, by microcrystalline cellulose (MCC), methyl tion of the methine stretch at 2891 cm-1 is varying amounts. At the bottom of Figure cellulose, sodium carboxymethyl cellu- attributed to the wings of the CH3 bands. 3, I show MCC because of the similarity lose, and hydroxypropyl cellulose (HPC). Note that the intensity of the OH band in with the sub-HPC spectrum. My inter- We have scaled the spectra so that the this spectrum is quite low, suggesting that pretation is that enough hydroxypropyl peak intensities in the region of the C-C most of the hydroxyl groups have been groups were substituted to disrupt the and C-O single bonds (near 1100 cm-1) reacted. packing of the cellulose, but clearly not are approximately equivalent. What we Now, let’s compare the spectrum of enough to give the pattern of the HPC will look for is changes in CH stretch- sodium carboxylmethyl cellulose to MCC. where the CH2–CH3 deformation is quite ing (2800–3000 cm-1) and deformation The spectrum is clearly altered, but the strong. (1430–1480 cm-1) intensities relative to the significant intensity in the OH region The next two spectra going up Figure backbone intensities near 1100 cm-1 and should be assigned to OH of the carboxy 3 are those of hypromellose and HPMC, the OH intensity between 3100 and 3600 groups, suggesting that the sodium ion is both of which are hydroxypropyl methyl- cm-1. In MCC there are five CH groups not incorporated. The intensification in cellulose. These spectra are quite similar per glucose unit, one methylene CH2 unit the CH stretching region is assigned to the but, in fact, the HPMC resembles methyl- and three OH groups. In methyl cellulose, additional three CH2 groups. In the region cellulose more than the hypromellose. The if all OH groups are converted, there will of the CH2 and CH3 deformations there is implication is that there are more substitu- be five CH groups, one methylene group a shoulder at 1460 cm-1 but a strong band tions on HPMC of methyl for the proton -1 and three methyl groups. In sodium car- at 1415 cm which is assigned to the CH2 on the hydroxyl than substitution of the boxymethyl cellulose there will be five deformation from the substitution. hydroxypropyl group.

in my collection (with the exception of the sodium carboxymethyl cellulose, which (a)

1200 was kindly sent to me by Philo Morse at

1100 Particle Sciences). Even with this limited

1000 knowledge of the samples’ histories, I was

900 Methocell able to construct a self-consistent story.

800 In particular, this could be of use in un- HPMC 700 derstanding the modes of reactivity of the 600 cellulose. Presumably, the three different Hypromellose 500 hydroxyl groups have the potential for Intensity (counts/s) 400 different reactivities; it is also well known 300 that aggregation of the cellulose has a large 200 Hydroxypropyl cellulose impact on the reactivity, and this type of Sub hydroxypropyl cellulose 100 MCC measurement could help in determining 0 500 1000 1500 2000 2500 3000 3500 4000 the controlling factors of the chemical Raman shift (cm–1) (b) reactions.

1400 1374 1410 1457 Acknowledgments

I would like to thank Professor Rajai Atalla 1200 for the many detailed discussions on the

1000 structure of celluloses, especially as it is

elucidated by Raman spectroscopy. 800 1457 1363 600 1407 References 1372

Intensity (counts/s) (1)R. Atalla, in Comprehensive Natural Prod- 1462 400 ucts Chemistry, Carbohydrates and their

1382 Derivatives Including Tannins, Cellulose 200 1474 and Related Lignins, Volume 3, B. Mario 0 Pinto, Ed. (Elsevier, 1999), pp. 529–598, 800 1000 1200 1400 1600 Raman shift (cm–1) chap. 3.16.

(2) A. French, N.R. Bertoniere, R.M. Brown, (c)

H. Chanzy, D. Gray, K. Hattori, and W. 1800 2888 2938 Glasser, in Encyclopedia of Polymer Sci- 2838 1600 2985 ence and Technology, Volume 5 (John

1400 3455 Wiley and Sons, Inc., 2003).

3462 (3) R.H. Atalla and A. Isogai, in Celluloses 1200 in Comprehensive Natural Products II 2934 1000 2875

3470 Chemistry and Biology, L. Mander and 800 H.W. Lui, Eds. (Elsevier, Oxford, 2010), Intensity (counts/s)

600 2973 pp. 493–539, chap. 6.16. 2889

2936 (4) G. Socrates, Infrared and Raman Char- 400 2976 3463 2891 acteristic Group Frequencies – Tables

200 2968 3357 3280 and Charts (John Wiley and Sons, 2001).

0 2600 2800 3000 3200 3400 3600 Raman shift (cm–1) Fran Adar is the Principal Raman Ap- Figure 3: (a) Raman spectra (from bottom to top) of MCC, sub-hydroxypropyl cellulose, hypromellose plications Scientist for (hydroxypropyl methylcellulose) HPMC, and methyl cellulose. (b) Fingerprint Raman spectra (from Horiba Scientific in Edi- bottom to top) of MCC, sub-hydroxypropyl cellulose, hypromellose (hydroxypropyl methylcellulose) son, New Jersey. Direct correspondence to: HPMC, and methyl cellulose. (c) Raman spectra in the CH–OH region (from bottom to top) of MCC, sub- SpectroscopyEdit hydroxypropyl cellulose, hypromellose (hydroxypropyl methylcellulose) HPMC, and methyl cellulose. @UBM.com.

Conclusions formed material. This small study shows I have presented Raman spectra of some that the spectra do reflect the altered For more information on commonly available modified forms of chemistries. Note that the samples whose this topic, please visit: cellulose to determine if these spectra spectra I recorded were not well-character- www.spectroscopyonline.com can be predictive of the chemically trans- ized. They were simply samples that I have

Posted with permission from the November 2016 issue of Spectroscopy ® www.spectroscopyonline.com. Copyright 2016, Advanstar Communications, Inc. All rights reserved. For more information on the use of this content, contact Wright’s Media at 877-652-5295. 126230