Exploitation of Reactivity and Selectivity in Cellulose Functionalization

limited accessibility of the reactive groups, induced before converting the polymer, were investigated.2-6 In the course of these studies new media we...
0 downloads 0 Views 932KB Size
Biomacromolecules 2001, 2, 1124-1132

1124

Exploitation of Reactivity and Selectivity in Cellulose Functionalization Using Unconventional Media for the Design of Products Showing New Superstructures Tim F. Liebert and Thomas J. Heinze*,† Institut fu¨r Organische Chemie und Makromolekulare Chemie, Friedrich-Schiller-Universita¨t Jena, Humboldtstrasse 10, D-07743 Jena, Germany Received March 26, 2001; Revised Manuscript Received August 20, 2001

A variety of new cellulose solvents was investigated toward their potential as media for the functionalization of the polyglucane. Thus, mixtures of dimethyl sulfoxide (DMSO)/tetrabutylammonium fluoride trihydrate (TBAF), N-methylmorpholine-N-oxide (NMMNO)/DMSO, melts of LiClO4‚3H2O, and aqueous solutions of Ni(tren)(OH)2 [tren ) tris(2-aminoethyl)amine] were applied as reaction media. In case of the new solvent, DMSO/TBAF its usefulness for derivatization reactions including the etherification with sodium monochloroacetate and the acylation with vinyl esters of carbonic acids was studied. The structural features of the products were analyzed by means of 1H NMR spectroscopy (after depolymerization or peresterification), 13C NMR spectroscopy, and HPLC after complete hydrolytic chain degradation. The results were compared with those obtained from derivatives prepared using the solvent N,N-dimethylacetamide (DMAc)/LiCl and conventional, heterogeneous synthesis. It can be shown that in case of carboxymethylation reactions the reaction medium applied has a drastic influence both on the course of reaction and on the structural features of the products. A highly efficient tool was found to be atomic force microscopy (AFM), showing remarkable differences in the superstructures of the differentially synthesized derivatives. Introduction In recent years, a number of approaches was established for a tailored functionalization of polyglucanes.1 Besides the search for new regioselective paths using time-consuming, multistep protecting group techniques, the application of unconventional reagents and the development of modification methods exploiting the native superstructure as well as limited accessibility of the reactive groups, induced before converting the polymer, were investigated.2-6 In the course of these studies new media were applied. An important approach to new cellulosics was found to be the reaction via an induced phase separation (in reactive microstructure) obtained with solid alkali hydroxide particles, e.g.7-9 Starting from a solution of cellulose in N,N-dimethylacetamide (DMAc)/LiCl or from organo-soluble cellulose derivatives dissolved in dimethyl sulfoxide (DMSO), subsequent etherification and esterification yield products of unconventional functionalization patterns and properties.10,11 Thus, we investigated new types of solvents capable to extend the synthesis paths or to yield new structural parameters and can, in addition, be applied for the reaction of cellulose with new reagents. Besides our own work on the determination of substitution patterns of cellulose derivatives on the level of the repeating unit as well as along the polymer chain, attempts were made to establish structure-property relationships. A remarkable influence on the polymer behavior was observed by intro† Current address: Bergische Universita ¨ t Wuppertal, Gauss Strasse 20, D-42097 Wuppertal, Germany.

ducing concentration gradients of functionalization along the chain by applying reactions in reactive microstructure. Thus, derivatives with new colloidal and rheological properties were accessible.11 From these findings, the existence of different superstructures was concluded. Up to now the most detailed data concerning association and aggregation in solution were acquired by means of light scattering, angular dependence of the scattered light, TEM, and birefringence measurements.12,13 The surprising result is that cellulose derivatives with a total degree of substitution below three form various aggregates, so-called fringed micelles, caused by intermolecular and intramolecular hydrogen bonding of the remaining hydroxyl groups independent of the chemical nature of the functional groups introduced. A similar behavior was observed for polyelectrolytes as well as hydrophobically and hydrophilically modified polymer chains. We have found that atomic force microscopy (AFM) is a useful tool to investigate these types of superstructures. The paper reports about the synthesis of cellulose derivatives in different new solvents and the structural features of the products obtained. Furthermore, it deals with the investigation of superstructures of cellulose derivatives in particular carboxymethylcellulose by means of AFM. Experimental Section Materials. Avicel (Fluka, degree of polymerization, DP 330), spruce sulfite pulp (Papierfabrik Weissenborn, Weissenborn, Germany, DP 680) and dissolving pulp (Rosenthal, Germany, DP 950) after irradiation treatment with electron

10.1021/bm010068m CCC: $20.00 © 2001 American Chemical Society Published on Web 10/11/2001

Cellulose Functionalization

Biomacromolecules, Vol. 2, No. 4, 2001 1125

Figure 1. Typical chromatogram of a hydrolytically degraded CMC sample (crude product; not dialyzed). Peak assignment: (1) inorganic salts; (2) 2,3,6-tri-O-carboxymethylglucose; (3) 2,3-, 2,6-, and 3,6-di-O-carboxymethylglucose; (4) 2-, 3-, and 6-mono-O-carboxymethylglucose; (5) glucose; (6) diglycolate; (7) not assigned; (8) glycolate.

beam were applied as starting polymers. Carboxymethylcelluloses (DSCMC ) 1.89 and 1.72) for AFM measurements were prepared with sodium monochloroacetate in aqueous NaOH/isopropyl alcohol or in the solvent DMAc/LiCl according to ref 9. DMSO, tetrabutylammonium fluoride trihydrate (TBAF), LiClO4‚3H2O, and the acylating reagents were purchased from Fluka and were used without further purification. NaOH and sodium monochloroacetate were dried in a vacuum for 24 h prior to use. Solutions of cellulose in N-methylmorpholine-N-oxide (NMMNO) and Ni(tren)(OH)2 were prepared according to refs 14 and 15. Carboxymethylation of Cellulose in DMSO/TBAF. The solution was prepared by addition of 4 g cellulose (Avicel) to a mixture of 140 mL of DMSO and 26 g of TBAF. Dissolution occurred within 15 min. The cellulose solution was treated with a suspension of 9.9 g of NaOH powder in 50 mL of DMSO and 14.3 g of sodium monochloroacetate under vigorous stirring. The temperature was raised to 70 °C. After 4 h the mixture was cooled to room temperature and was precipitated into 700 mL methanol. The precipitate was filtered off, dissolved in 100 mL of distilled water, neutralized with acetic acid, and reprecipitated into 500 mL of ethanol. After filtration the product was washed with ethanol and dried in a vacuum at 50 °C. The sample (1d) was dialyzed against running water for 5 days. Degree of substitution, DS ) 2.09 (revealed by HPLC analysis8,9) mole fractions of the different repeating units, see Figure 1); IR (KBr) 1620, 1410 cm-1 (CdO, carboxylate group); 13C NMR spectroscopy (D2O): 175.9-179.2 ppm (CdO), 60.5-102.5 ppm (cellulose backbone); 1H NMR spectroscopic analysis after hydrolysis led to a partial DS at O-6 ) 0.96, O-2 ) 0.65, and O-3 ) 0.51 resulting in a total DS ) 2.12. Tosylation of Cellulose in DMSO/TBAF. First, 3 g cellulose (Avicel) were dissolved in the mixture DMSO (100 mL)/TBAF (20 g) and were treated with 10.4 g of tosyl

chloride in the presence of 11.2 g of triethylamine for 24 h at room temperature. Isolation and work up was carried out as described in ref 16. DSTosyl ) 1.10 (sulfur analysis: 10.58%); IR (KBr) 3027 (C-H, aromate), 1598, 1500, 1453 (C-C, aromate), 1359, 1178 cm-1 (SO2); 13C NMR spectroscopy (DMSO-d6): 127.4-144.5 (C-Haromat), 60.3-102.5 (cellulose backbone), 20.9 ppm (CH3). Acetylation of Cellulose in DMSO/TBAF. For a typical conversion, a solution of 1 g cellulose (Avicel) in 33 mL of DMSO and 6.6 g of TBAF was treated with 1.2 g (0.014 mol) of vinyl acetate for 70 h at 40 °C (other reagents see Table 1). The product was isolated by precipitation into 200 mL of isopropyl alcohol, addition of 50 mL of water (removal of inorganic impurities) and filtration. After being washed with 200 mL of isopropyl alcohol, the product was dried in a vaccum at 50 °C (sample 2b). DSacetate ) 1.04 (determined by means of 1H NMR spectroscopy after perpropionylation, see ref 17); IR (KBr) 1752 cm-1 (CdO); 13C NMR spectroscopy (DMSO-d ): 169.1-169.9 ppm (Cd 6 O), 60.3-102.5 ppm (cellulose backbone). Carboxymethylation in LiClO4‚3H2O. First, 0.5 g of cellulose (dissolving pulp) was added to a melt (at 95-100 °C) of 30 g of LiClO4‚3H2O. The cellulose dissolves within a few minutes. This mixture was treated stepwise with 1.23 g of solid NaOH and 1.07 g of sodium monochloroacetate. During this procedure gelation occurred. The mixture was kept for 4 h at 95-100 °C. Isolation was carried out by precipitation into 300 mL of methanol, dissolution of the precipitate in 10 mL of water, neutralization with acetic acid, and reprecipitation into 100 mL of ethanol. This crude product was purified by reprecipitation of an aqueous solution into ethanol, washing 3 times with 100 mL aqueous ethanol (90%, v/v), and drying in a vacuum at 50 °C (sample 8c). DS ) 0.69, determined by HPLC after chain degradation; IR (KBr) 1646, 1419 cm-1 (CdO, carboxylate group); 13C

1126

Biomacromolecules, Vol. 2, No. 4, 2001

Liebert and Heinze

Table 1. Conditions and Results of Reactions of Cellulose Performed in Dimethyl Sulfoxide (DMSO)/Tetrabutylammonium Fluoride (TBAF)d cellulose derivative CMC CMC CMC CMC CMC CMC CMC CA CA CA cellulose butyrate cellulose laurate cellulose benzoate cellulose tosylate

no. 1a 1b 1c 1d 1e 1f 1g 2a 2b 2c 3 4 5 6

reagents SCA/NaOH SCA/NaOH SCA/NaOH SCA/NaOH SCA/NaOH SCA/NaOH SCA/NaOH acetic anhydride vinyl acetate vinyl acetateb vinyl butyrate vinyl laurate vinyl benzoate tosyl chloride/TEA

molar ratio 1:5:10a 1:5:10a 1:5:10a 1:5:10a 1:5:10a 1:10:20a 1:10:20a 1:2.3 1:2.3 1:10 1:2.3 1:10 1:2.3 1:3:6

time (h) 0.5 1 2 4 16 4 48 70 70 70 70 70 70 24

temp (°C)

deg of substitution

70 70 70 70 70 70 70 40 40 40 40 40 40 8

1.82c 2.06c 2.02c 2.09c 1.91c 2.07c 1.89c 0.83 1.04 2.72 0.86 2.60 0.95 1.10

solubility insoluble water water water water water water insoluble DMSO DMSO insoluble pyridine, THF, CHCl3 only swelling DMSO, DMF

a Anhydroglucose unit/NaOH/sodium monochloroacetate. b Catalyst applied was a mixture of KH PO and Na HPO yielding a pH of 7.1 in water. 2 4 2 4 Determined by HPLC after depolymerization; deviation (0.01. d Key: CMC, carboxymethyl cellulose; CA, cellulose acetate; SCA, sodium monochloroacetate; TEA, triethylamine; THF, tetrahydrofuran; DMF, N,N-dimethyl formamide.

c

NMR spectroscopy (D2O): 175.8-180.0 (CdO), 60.3102.5 ppm (cellulose backbone). Homogeneous Carboxymethylation in Ni(tren)(OH)2. For a typical preparation, 0.5 g of cellulose (Avicel) were dissolved in 5 mL of Ni(tren)(OH)2. Then 2.46 g of NaOH in 8 mL of water and 3.58 g of sodium monochloroacetate in 10 mL of water were added dropwise. During this procedure no regeneration or precipitation of the polymer occurred. After 3 h at 80 °C, the solution was cooled to room temperature and was precipitated into 100 mL of methanol. The precipitate was filtered off, dissolved in 20 mL of distilled water, neutralized with acetic acid and reprecipitated into 80 mL of ethanol. After filtration the product (sample 9c) was washed with ethanol and dried in a vacuum at 50 °C. DS ) 0.54, determined by HPLC after chain degradation; IR (KBr) 1630, 1410 cm-1 (CdO, carboxylate group); 13C NMR (D2O): 176.4-180.3 ppm (CdO), 60.3-102.5 ppm (cellulose backbone). Carboxymethylation in NMMNO. Cellulose (spruce sulfite pulp) was dissolved in NMMNO (11% w/w cellulose) as described.15 It was stored as solidified material at room temperature prior to use. Then 10 g of this sample was heated to 85 °C and 20 mL of DMSO were carefully added. After the mixture was stirred at 80-90 °C for about 1 h, a homogeneous system was obtained. Then 5.5 g of NaOH in 20 mL of DMSO and 8.0 g of sodium monochloroacetate in 10 mL of DMSO were added under vigorous stirring. This mixture was kept at 80 °C for 2 h. It was carefully precipitated into 400 mL of methanol. The precipitate was filtered off, dissolved in 40 mL of distilled water, neutralized with acetic acid, and reprecipitated into 200 mL of ethanol. After filtration, the product was washed four times with 200 mL of aqueous ethanol (90%, v/v) and dried in a vacuum at 50 °C (sample 7b). DS ) 1.26, determined by HPLC after chain degradation; IR (KBr) 1641, 1412 cm-1 (CdO, carboxylate group); 13C NMR spectroscopy (D2O): 175.2179.3 (CdO), 60.3-102.5 ppm (cellulose backbone). Measurements. The HPLC analysis of the CMC samples was carried out as described by Heinze et al. 1994.9 However, the samples were hydrolyzed with perchloric acid. 0.1 g of

CMC were dispersed in 2 mL of HClO4 (70%) and after 10 min at room-temperature diluted with 18 mL of distilled water. This mixture was kept at 100 °C for 16 h. The solution obtained was carefully neutralized with 2 N KOH and kept at 4 °C for 1 h to precipitate the KClO4. The salt was filtered off and washed three times with distilled water. The obtained solution was reduced to approximately 3 mL and diluted with distilled water to give exactly a 5 mL sample. Chromatographic experiments were carried out at 65 °C with 0.01 N H2SO4 as eluent with a flow rate of 0.5 mL/min. The column used was a Bio-Rad Aminex HPX-87 H. The 1H NMR analyses of CMC were carried out as described in reference.18 For this purpose the CMCs were hydrolyzed with a mixture of D2SO4/D2O (25% v/v) within 5 h at 90 °C. The spectra were acquired on a Bruker AMX 400 spectrometer. 1H NMR spectra of cellulose esters were acquired after perpropionylation as described in reference.17 FTIR spectra were measured on a Bio-Rad FTS 25 PC using the KBr pellet technique. AFM Studies. A stock solution of CMCs of 1 mg/mL was diluted to various concentrations in the range from 1 to 10 µg/mL. Drops (about 2 µL) of these solutions were deposited onto freshly cleaved mica and allowed to dry for 15 min on air. The samples were then imaged in a liquid cell under butanol in both contact and tapping mode. The AFM measurements were carried out with a MultiMode SPM and Nanoscope IIIa control system (Digital Instruments, Santa Barbara, CA). The tip used was a NPSTT type (nitride probes oriented twin tip) with a spring constant (nominal) of 0.1 N/m. The force during contact mode measurements was in the range of 0.5-1 nN, the amplitude in tapping mode experiments was about 4 nm (damping 1%). Results and Discussion Reactions in Dimethyl Sulfoxide (DMSO)/Tetrabutylammonium Fluoride Trihydrate (TBAF). A mixture of DMSO/TBAF represents an efficient cellulose solvent. It dissolves the polymer with a degree of polymerization (DP)

Cellulose Functionalization

of up to 650 within 15 min completely. No pretreatment is necessary. The solutions applied in this study contained 16.6% (w/v) TBAF and 2.9% (w/v) cellulose. In 13C NMR spectra signals appear at 102.7 (C-1), 78.4 (C-4), 75.6 (C5), 75.0 (C-3), 73.5 (C-2), and 59.9 ppm (C-6) only.19 In comparison with spectral data obtained for cellulose in N,Ndimethylacetamide (DMAc)/LiCl, which is a typical nonderivatizing cellulose solvent,20 it is obvious that cellulose is dissolved without covalent interactions in the new medium. Thus, it was exploited for a number of conversions including acylation, tosylation, and carboxymethylation reactions. A summary of reaction conditions and degrees of substitution (DS) of the products obtained is given in Table 1. In a first series of experiments, the cellulose dissolved in DMSO/TBAF was treated with acetic anhydride (for conditions, see Experimental Section). If a molar ratio of 2.3:1.0 (acylation reagent/anhydroglucose unit, AGU) is applied, a DS value of 0.83 was determined. Comparable reaction conditions were used for the acetylation with vinyl acetate as acylating reagent. In case of the same molar ratio, a DS of 1.04 can be achieved, which is due to the formation of acetic aldehyde during this conversion shifting the equilibrium toward the product side. Moreover, the lower DS in case of the application of acetic anhydride is caused by the comparably fast hydrolysis of the reagent due to the water content of the solvent. A wide variety of vinyl carbonic acid esters can be exploited for this type of conversion (see Table 1). It can be seen that a stoichiometric control of the DS values is possible via the amount of reagent added. A remarkable result was a DS of 2.6 for cellulose laurate indicating that this homogeneous esterification path is efficient for the preparation of fatty acid esters of cellulose. Besides the tosylation (conditions see Table 1), which is possible by conversion of cellulose dissolved in DMSO/ TBAF with tosyl chloride in the presence of triethylamine yielding a derivative with a DS of 1.1, the etherification with sodium monochloroacetate was investigated. Here the derivatization in reactive microstructure was studied. This type of reaction was described for conversions starting from both cellulose dissolved in DMAc/LiCl and from solutions of cellulose derivatives with limited hydrolytic stability in common organic solvents, e.g., DMSO.7-9 It is characterized by the addition of a solid catalyst or reagent in the first step resulting in the formation of a highly reactive interface solid reagent/polymer which is called “reactive microstructure”. The conversion to the cellulose derivative succeeds preferably in that microstructure. Thus, cellulose dissolved in DMSO/ TBAF was treated with solid NaOH dispersed in DMSO and with sodium monochloroacetate. The DS values of the products obtained are summarized in Table 1. A remarkable finding was that a DS higher than 2 was already achieved after a reaction time of about 1 h. It can be seen that a maximum DSCMC of 2.09 is accessible. A prolongation of the reaction time and the increase of the amounts of reagents did not result in an increasing degree of functionalization. The carboxymethylcellulose (CMC) samples gave clear solutions in water starting at DS values of 1.89. Besides standard 13C NMR experiments structure analysis of CMCs

Biomacromolecules, Vol. 2, No. 4, 2001 1127

obtained was carried out by means of 1H NMR spectroscopy and HPLC both after hydrolytic depolymerization of the samples. The partial degrees of substitution on the three reactive sites of the AGU are accessible by 1H NMR spectroscopy after polymer degradation.18 The hydrolysis was carried out by treating the samples in D2SO4/D2O for 5 h at 90 °C. Before hydrolysis, the samples have to be dialyzed to remove glycolic acid (one of the main impurities of CMC) which gives a NMR signal at 4.19 ppm, i.e., in the range of chemical shift of the carboxymethylated O-6 within the AGU. From these measurements a distribution of substituents on the level of the AGU in the order position 6 > 2 g 3 (e.g., sample 1d possess a partial DS at position 6 ) 0.96, at position 2 ) 0.65, and at position 3 ) 0.51 resulting in a total DS ) 2.12) was concluded. An analogous distribution of functional groups within the AGU was evaluated for CMCs prepared via the induced phase separation technique in DMAc/LiCl or using cellulose intermediates with limited hydrolytic stability in DMSO.8 The characterization of the CMCs including 1H NMR studies showed no evidence for the occurrence of oxidative processes during the reaction in DMSO. For HPLC measurements, the controlled depolymerization of the derivatives was achieved by solvolysis with perchloric acid (see Experimental Section). A typical chromatogram of a crude CMC (not dialyzed) is shown in Figure 1. Besides some impurities of the reaction mixture, e.g., glycolate and diglycolate, which can be removed by dialysis, only traces (