Carbohydrate Polymers 86 (2011) 154–161

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

A “click-chemistry” approach to cellulose-based hydrogels

Andreas Koschellaa , Matthias Hartlieba , Thomas Heinzea,b,∗
a
Friedrich Schiller University of Jena, Institute for Organic Chemistry and Macromolecular Chemistry, Center of Excellence for Polysaccharide Research,
Humboldtstraße 10, D-07743 Jena, Germany
b
Åbo Akademi University, Fibre and Cellulose Technology, Porthansgatan 3, FI-20500 Åbo, Finland

a r t i c l e i n f o a b s t r a c t

Article history: Cellulose derivatives bearing azide- and alkyne moieties were prepared by conversion of cellulose p-
Received 1 March 2011 toluenesulphonic acid ester with sodium azide, on one hand, and propargylamine, on the other. The
Received in revised form 8 April 2011 products obtained were carboxymethylated to yield water soluble multifunctional cellulose derivatives.
Accepted 12 April 2011
Elemental analysis, FTIR- and NMR spectroscopy were applied to prove the structure of the polymers. SEC
Available online 21 April 2011
of the hydrogel components revealed values of the degree of polymerization (DP) between 43 and 200 that
are acceptable values after this multi-step reaction starting from celluloses with DP 600. The copper(I)-
Keywords:
catalyzed 1,3-dipolar cycloaddition reaction (Huisgen-reaction) was applied for the cross-linking. Gel
Carboxymethyl-6-deoxy-6-aminopropargyl
cellulose
formation occurred within 55 and 1600 s after mixing of the aqueous solutions of both components and
Carboxymethyl-6-azido-6-deoxy cellulose copper(I) catalyst. The gelation time was found to depend on both the degree of functionalization and
Click reaction the amount of copper(I) catalyst. FTIR spectroscopy revealed incomplete conversion of the reactive sites.
Hydrogel The gels contain up to 98.4% water. Freeze-drying led to spongy materials with a porous structure as
Morphology visualised by SEM.
Swelling © 2011 Elsevier Ltd. All rights reserved.

1. Introduction lower critical solution temperature (LCST). Moreover, chemical

cross-linking can be achieved by either difunctional molecules
Hydrogels are three-dimensional networks of hydrophilic poly- (e.g. divinyl sulphone) or by radical cross-linking with high-
mers that are cross-linked either chemically via covalent bonds energy irradiation. It is also possible to prepare hydrogels from
or physically by ionic-, van der Waals-, hydrophobic, or hydro- solutions of unmodified cellulose (Chang & Zhang, 2011). Yang
gen bond interactions (Yu & Ding, 2008). Their most important et al. dissolved cellulose in an ionic liquid followed by coagu-
property is the capability of remarkable solvent uptake. Such lation with water yielding a gel that is physically cross-linked
materials play an enormous role in our daily lives as, e.g., super- by hydrogen bonds (Li, Lin, Yang, Wan, & Cui, 2009). Zhang
absorbers in diapers and other hygiene products (Buchholz, 1996). et al. prepared gels by dissolving cellulose in an aqueous 6 wt%
Due to their high water content and tissue like structure, hydro- NaOH solution with 4 wt% urea and subsequent cross-linking
gels are valuable materials for advanced applications in fields with epichlorohydrin (Zhou, Chang, Zhang, & Zhang, 2007). Most
where biocompatible devices in chemical and biochemical appli- of the cellulose-based hydrogels are hybrid systems. Biocom-
cations are needed (Bajpai, Shukla, Bhanu, & Kankane, 2008; patibile cellulose and cellulose derivatives are combined with a
Calvert, 2009; Lee & Mooney, 2001; Park & Park, 1996; Yu & Ding, variety of synthetic polymers, e.g., poly(acrylonitrile) (Pourjavadi,
2008). Zohuriaan-Mehr, Ghasempoori, & Hossienzadeh, 2007), poly(N-
Due to its inherent hydrophilicity the renewable resource cel- isopropylacrylamide) (pNIPAAm) (Ma, Zhang, Fan, Xu, & Liang,
lulose is a valuable starting material for hydrogels. It is already 2008), and poly(vinyl alcohol) (Alupei, Popa, Hamcerencu, &
reported that physical gels from hydroxypropylmethyl cellulose, Abadie, 2002) in order to tune the properties. The disadvantage
methyl cellulose, and hydroxypropyl cellulose possess thermore- of most of these gels is that hazardous cross-linking agents like
sponsive properties and are therefore only stable above the epichlorohydrin or glutaraldehyde have to be used (Shang, Shao, &
Chen, 2008).
An alternative preparation method would be the cross-linking
by 1,3-dipolar cycloaddition reaction of alkynes and azides. One
∗ Corresponding author at: Friedrich Schiller University of Jena, Institute for example is a gel composed of 6-azido-6-deoxy cellulose and
Organic Chemistry and Macromolecular Chemistry, Center of Excellence for Polysac- pNIPAAm-co-hydroxyethylmethacrylate (HEMA). The HEMA block
charide Research, Humboldtstraße 10, D-07743 Jena, Germany.
is modified with an alkyne moiety and the gelation is accomplished
Tel.: +49 3641 948270; fax: +49 3641 948272.
E-mail address: [email protected] (T. Heinze). by the 1,3-dipolar cycloaddition (Zhang, Xu, Wu, Zhang, & Zhuo,

0144-8617/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carbpol.2011.04.031
 A. Koschella et al. / Carbohydrate Polymers 86 (2011) 154–161 155

2009). Because of the LCST of pNIPAAm, the gels produced are obtain md . The maximum water content Qeq was calculated accord-
thermo-sensitive. ing to Eq. (1).
It was interesting to study hydrogels solely composed of cel- ms − md
lulose derivatives that can be prepared in aqueous solution. This Qeq = × 100% (1)
ms
paper describes the synthesis of water-soluble cellulose deriva-
tives capable of undergoing the copper-catalyzed 1,3-dipolar Qeq : maximum water content; ms : mass of the swollen sample; md :
cycloaddition reaction. Hydrogels were prepared and character- mass of the dried sample.
ized regarding their cross-linking- and swelling behaviour as well The gel fraction Fp was calculated according to Eq. (2).
as their morphology in the dried state. mth
Fp = × 100% (2)
md
2. Experimental Fp : gel fraction; mth : theoretical mass of the polymer in the gel
(based on the stock solutions used); md : mass of the dried sample.
2.1. Materials The degree of substitution (DS) of carboxymethyl groups was
determined by HPLC after hydrolysis of the polymer with perchloric
Spruce sulphite pulp was purchased from Fluka. Other chemicals acid according to a previously published method (Heinze & Pfeiffer,
were supplied by Aldrich, Merck and Fluka. Cellulose was dried in 1999).
vacuum at 100 ◦ C for 2 d over KOH and LiCl was dried in vacuum
for 1 d at 150 ◦ C. 2.3. Methods

2.2. Measurements 2.3.1. Synthesis of tosyl cellulose 2

The synthesis was carried out according to (Rahn, Diamantoglou,
FTIR spectra were recorded on a Nicolet Avatar 370 spec- Klemm, Berghmans, & Heinze, 1996).
trometer using the KBr-technique. The 1 H- and 13 C-NMR spectra
were acquired on a Bruker AVANCE 400 spectrometer in dimethyl Yield: 41.1 g (133.8 mmol, 91.6%).
sulphoxide (DMSO)-d6 or D2 O at 70 ◦ C. For 1 H-NMR spectra up to Degree of substitution of tosyl groups, DSTos : 0.94 (based on ele-
200 scans and for 13 C-NMR spectra up to 26,000 scans were accu- mental analysis).
mulated. Degree of substitution of 6-deoxy-6-chloro groups, DSCl : 0.06
Elemental analysis was performed with a Vario EL III from Ele- (based on elemental analysis).
mentaranalysensysteme Hanau (Germany). The DS values were Elemental analysis: 48.21% C, 5.17% H, 9.76% S, 0.7% Cl.
calculated from the elemental composition by an inhouse software. FTIR (KBr): 3479 ( OH), 2887 ( CH2 ), 1359 (as SO2 ), 1173 (s
Both the calculated composition based on the DS values and found SO2 ) cm−1 .
values are given in the experimental part.
Size exclusion chromatography in aq. 0.1 mol/L Na2 HPO4 con- 2.3.2. 6-Azido-6-deoxy cellulose (4)
taining 0.1% NaN3 was measured with a JASCO SEC system Tosyl cellulose 2 (12.01 g, 39 mmol, DSTos 0.94) was dissolved
(degasser DG 980-50, pump PU 980, UV detector 975 working at in 400 mL N,N-dimethyl formamide (DMF) under stirring. Sodium
 = 254 nm, refractive index detector 930, columns Suprema 1000 azide (12.03 g, 185 mmol, 4.8 mol/mol anhydroglucose unit, AGU)
and Suprema 30 from Kromatek, Great Dunmow, Essex, UK, with was added to the viscous solution. The reaction mixture was
eluent flow rate of 1.000 mL/min). The eluent DMSO/LiBr was used allowed to react for 24 h at 100 ◦ C under stirring. After cooling down
in combination with the columns Novema 3000 and Novema 300 to room temperature, the solution was poured into 3 L of ice water.
(Polymer Standards Service, Mainz, Germany). Pullulan- and dex- The precipitate was filtered off, washed with water (4 × 1 L) and
tran standards (for M < 104 g/mol) were used for calibration. Weight ethanol (3 × 500 mL). The white powder was dried in vacuum at
average (M̄w ) and number average molecular mass (M̄n ) as well as 40 ◦ C.
polydispersity index (PDI) were calculated.
Scanning electron microscopic images were obtained from Yield: 6.84 g (37.3 mmol, 95.6%).
a LEO-1450 VP with Cryo-Transfer (BAL-TEC-VCT 100), X-ray- Degree of substitution of 6-azido-6-deoxy groups, DSAzide : 0.81
microanalysis (Oxford-Link Isis) and a resolution of 4 nm. The (based on elemental analysis).
samples were freeze dried gels with the volume of 1 mL. DSTos : 0.01 (based on elemental analysis).
Rheological measurements were performed on a HAAKE Mars SEC: M̄n 53,278 g/mol, M̄w 216,140 g/mol, PDI 4.06.
rheometer with cone and plate geometry at room temperature. The Number average degree of polymerization, DPn : 291.
measurement parameter were  = 10000 Pa and f = 1 s−1 . Elemental analysis: 38.42% C, 4.81% H, 18.51% N, 0.12% S. [theoret-
Ascorbic acid (0.2 mL, 0.1 M) was mixed with 0.2 mL aqueous ical data: 39.65% C, 5.03% H, 18.52% N, 0.17% S].
1 H-NMR: (DMSO-d /LiCl) ı: 7.8–7.4 (H
CuSO4 solution (0.02, 0.03, and 0.04 M). The polymer solution 6 aryl -Tosyl), 5.00–3.14
(0.2 mL, 5%, w/w) was mixed with 0.2 mL of the catalyst solution (HAGU ) ppm.
13 C-NMR (DMSO-d /LiCl) ı: 103.3 (C1), 79.6 (C4), 75.6–72.3 (C2,
and 0.2 mL of the mixture were rapidly transferred to the rheome- 6
ter. The time given in Fig. 2 and Table 2 includes the time starting 3, 5), 60.1 (C6 ), 51.5 (C6) ppm.
from mixing the solutions, transfer to the rheometer and start of FTIR (KBr): 3441 ( OH), 2920 ( CH2 ), 2108 ( azide), 1373 (as
the measurement. The time dependence of the storage modulus SO2 ), 1161 (s SO2 ) cm−1 .
(G ) and the loss modulus (G ) was recorded.
For the determination of the degree of swelling, gels were pre- 2.4. Carboxymethyl-6- azido -6-deoxy cellulose (6)
pared from 1 mL reaction mixture. Immediately after preparation,
the wet gels were placed in tubes equipped with fritted glass plate. 6-Azido-6-deoxy cellulose 4 (6.01 g, 32.8 mmol, DSAzide 0.81)
The hydrogels were allowed to swell in a water bath for 48 h at a was stirred with 100 mL DMSO for 2 h at room tempera-
temperature of 25 ◦ C. Excess of water was wiped off and the tubes ture. Isopropanol (150 mL) was added to the swollen polymer
were centrifuged for 2 min at 2200 rpm and weighed to obtain ms . under stirring. A solution of 7.8 g (195.5 mmol, 6 mol/mol AGU)
The swollen hydrogels were dried for 24 h at 60 ◦ C and weighed to NaOH in 30 mL water was added to the suspension and the
156 A. Koschella et al. / Carbohydrate Polymers 86 (2011) 154–161

mixture was allowed to react for 2 h under stirring. Sodium with ethanol (300 mL), ethanol/water (4:1, v/v, 300 mL), and with
monochloroacetate (22.6 g, 196 mmol) was added and the mix- ethanol (300 mL). The raw product was dissolved in 400 mL water,
ture was allowed to react for 4 h at 55 ◦ C under stirring. Sodium filtrated and precipitated in 1 L ethanol. The product was washed
monochloroacetate (8.5 g, 72 mmol) were added and stirring with ethanol (2 × 500 mL) and dried in vacuum at 40 ◦ C.
was continued for 1 h. After cooling to room temperature, the
product was filtered off and washed with 500 mL ethanol. Purifi- Yield: 2.21 g (6.4 mmol, 31.7%).
cation was carried out by dissolving the crude product in water DSAmine : 0.58 (based on elemental analysis of compound 8).
(120 mL) and reprecipitation in ethanol (700 ml). The product DSCM : 2.05 (based on HPLC analysis of the hydrolyzed polymer).
was washed with ethanol (2 × 500 mL), ethanol/water (5/1, v/v) DSCl : 0.04 (based on elemental analysis).
(2 × 300 mL) and with ethanol (3 × 250 mL) and dried in vacuum SEC: M̄n 14,815 g/mol, M̄w 49,893 g/mol, PDI: 3.37.
at 40 ◦ C. DPn : 43.
Diglycolic acid: 19.7% (based on HPLC analysis of hydrolyzed poly-
Yield: 6.6 g (23.3 mmol, 71%). mer).
DSAzide : 0.81 (based on elemental analysis of compound 4). Glycolic acid: 6.3% (based on HPLC analysis of hydrolyzed poly-
Degree of substitution of carboxymethyl groups, DSCM : 1.57 (based mer).
on HPLC analysis of the hydrolyzed polymer). 13 C-NMR (DMSO-d /LiCl) ı: 176.2–175.4 (C11), 100.5 (C1),
6
SEC: M̄n 45,934 g/mol, M̄w 133,910 g/mol, PDI: 2.92. 79.5–71.8 (C2–5), 69–68.4 (C10), 67 (C6 (CM)), 55.9 (C6 (OH)), 46.7
DPn : 162. (C6 (Amine)), 35.5 (C7) ppm.
Diglycolic acid: 4.4% (based on HPLC analysis of hydrolyzed poly- FTIR (KBr): 3461 ( OH), 3300 ( alkyne, shoulder), 2924 ( CH2 ),
mer). 1601 ( COONa) cm−1 .
Glycolic acid: 8.2% (based on HPLC analysis of hydrolyzed poly-
mer). 2.4.3. Preparation of hydrogels (sample G9)
13 C-NMR (DMSO-d /LiCl) ı: 175.8–175.3 (C8), 100.4 (C1),
6 Carboxymethyl-6-deoxy-6-aminopropargyl cellulose (0.579 g,
79.5–71.4 (C2–5), 70–68.7 (C7), 68.1 (C6 (CM)), 55.8 (C6 (OH)), 1.66 mmol, sample 10) and carboxymethyl-6-azido-6-deoxy cel-
48.8 (C6 (N3 )) ppm. lulose 0.308 g, 1.09 mmol, sample 6) were dissolved in 10 mL
FTIR (KBr): 3439 ( OH), 2926 ( CH2 ), 2115 ( azide), 1607 ( water under stirring. The copper(I)-catalyst was prepared by mix-
COONa) cm−1 . ing aqueous solutions of CuSO4 (5 mL, 0,04 mol/L) and ascorbic
acid (5 mL, 1 mol/L). The solutions were mixed and homogenized
2.4.1. 6-Deoxy-6-aminopropargyl cellulose (8) rapidly. A clear gel of green colour was formed after 2.5 min at room
Tosyl cellulose 2 (12.0 g, 39 mmol, DSTos 0.94) was dissolved in temperature. The compositions of the gels prepared are summa-
500 mL DMSO and propargylamine (20 g, 363 mmol, 9.3 mol/mol rized in Table 2.
AGU) was added drop-wise to the viscous solution. The reaction
mixture was allowed to react for 24 h at 80 ◦ C under stirring.
3. Results and discussion
After cooling down to room temperature, the black solution was
poured into 1.6 L acetone. The brown precipitate was filtered off
3.1. Preparation of the cellulose derivatives
and washed with acetone (4 × 400 mL). The raw product was repre-
cipitated from DMSO (100 mL) into acetone (1.2 L). After filtration
The starting materials for the hydrogel preparation were synthe-
and washing with acetone (2 × 400 mL), the product was dried in
sized from tosyl cellulose (Scheme 1), which can be easily prepared
vacuum at 40 ◦ C.
by conversion of cellulose dissolved in N,N-dimethyl acetamide
(DMA)/LiCl with p-toluenesulphonic acid (tosyl) chloride in the
Yield: 5 g (22.4 mmol, 57.5%). presence of triethylamine according to (Rahn et al., 1996). Tosyl cel-
Degree of substitution of 6-deoxy-6-aminopropargyl groups, luloses with different DS were used (sample 1, DSTos 0.33; sample 2,
DSAmino : 0.58 (based on elemental analysis). DSTos 0.94). A DSTos < 1 was chosen to ensure a preferred tosylation
DSTos : 0.25 (based on elemental analysis). of position 6 with negligible functionalization of the secondary OH
DSCl : 0.04 (based on elemental analysis). groups. The 6-azido-6-deoxy celluloses 3 and 4 were prepared by
Elemental analysis: 50.1% C, 5.96% H, 3.62% N, 3.62% S, 0.6% Cl. conversion of tosyl cellulose (samples 1 and 2) with sodium azide
[theoretical data: 51.15% C, 5.92% H, 3.65% N, 3.6% S, 0.64% Cl]. in DMF solution for 24 h at 100 ◦ C (Heinze, Koschella, Brackhagen,
1 H-NMR (DMSO-d /LiCl) ı: 7.78–7.42 (H
6 aryl -Tosyl), 5.0–2.86 Engelhardt, & Nachtkamp, 2006). The products 3 (DSAzide 0.15)
(HAGU ), 2.3 (CH3 -Tosyl), 2.09 (H9) ppm. and 4 (DSAzide 0.81) were isolated by precipitation in water fol-
13 C-NMR (DMSO-d /LiCl) ı: 129.9–126.1 (C
6 aryl -Tosyl), 103.3 (C1), lowed by washing with water and ethanol. Elemental analysis
83.23 (C8), 80.42 (C9), 75.4-73.9 (C2–5), 61.1 (C6 ), 45.9 (C6), 38.2 revealed an almost complete substitution of the tosylate moieties
(C7), 21.5 (CH3 -Tosyl) ppm. by azide groups (0.88% S in sample 3, DSTos 0.05; 0.12% S in sam-
FTIR (KBr): 3442 ( OH), 3297 ( alkyne, shoulder), 2915 ( CH2 ), ple 4, DSTos 0.01). This was confirmed by 13 C-NMR spectroscopy.
1366 (as SO2 ), 1176 (s SO2 ) cm−1 . The following peaks were assigned: 103.3 ppm (C1), 79.6 ppm (C4),
75.6–72.3 ppm (C2, 3, 5), 60.1 ppm (CH2 OH), 51.5 ppm (CH2 N3 ). Sig-
2.4.2. Carboxymethyl-6-deoxy-6-aminopropargyl cellulose (10) nals of very low intensity belonging to the tosyl groups could only
6-Deoxy-6-aminopropargyl cellulose 8 (4.5 g, 20 mmol, DSAmine be detected in the 1 H-NMR spectrum. The presence of the azide
0.58) was treated with 50 mL DMSO for 2 h at room temperature group was clearly shown by FTIR spectroscopy due to the pres-
and subsequently isopropanol (140 mL) was added to the swollen ence of an absorption band at 2108 cm−1 . Other bands appear at
polymer. A solution of sodium hydroxide (4.88 g, 122 mmol, 3441 cm−1 (␯OH) and 2910 cm−1 (CH). The characteristic bands
6 mol/mol AGU) in 20 mL water was added to the suspension of the tosyl group at 1373 (as SO2 ), 1161 (s SO2 ) cm−1 could be
and the mixture was stirred for 2 h at room temperature. Sodium seen as very weak signals only. The samples are soluble in DMA,
monochloroacetate (8.86 g, 76 mmol) was added and the mixture DMF, and DMSO.
was allowed to react for 4 h at 55 ◦ C under stirring. After cool- The 6-deoxy-6-aminopropargyl celluloses 7 and 8 were pre-
ing to room temperature, the product was filtered off, washed pared by a comparable strategy. Tosyl celluloses 1 and 2 dissolved
 A. Koschella et al. / Carbohydrate Polymers 86 (2011) 154–161 157

Fig. 1. 13 C-NMR spectra of 6-deoxy-6-aminopropargyl cellulose (8, degree of substitution of 6-desoxy-6-aminopropargyl groups, DSAmino 0.58, recorded in dimethyl
sulphoxide-d6 ) and carboxymethyl-6-deoxy-6-aminopropargyl cellulose (10, DSAmino 0.58, DS of carboxymethyl groups, DSCM , 2.05, recorded in D2 O). Impurities (trapped
solvents) are marked with an asterisk.

in DMSO were converted with propargyl amine for 24 h at 80 ◦ C incorrect results will be obtained. Therefore, the DS of CM-groups
according to (Pohl & Heinze, 2008). The products were precipitated was determined by a HPLC method and under the assumption that
in acetone, washed, and dried. The elemental analysis revealed an the other substituents are stable during the carboxymethylation
incomplete nucleophilic displacement of the tosyl groups, i.e., there reaction
is a remaining DSTos of 0.05 (sample 7) and 0.25 (sample 8). The molecular weight of the cellulose derivatives was charac-
The displacement of the tosyl groups became also obvious in terized by means of SEC. However, due to the different solubility
the FTIR spectra: 3442 cm−1 (OH), 3297 cm−1 ( alkyne, shoulder), of the samples, only a rough comparison can be done. 6-Azido-6-
2915 cm−1 (CH2 ), 1366 cm−1 (as SO2 ), 1176 cm−1 (s SO2 ) cm−1 . deoxy cellulose (samples 3 and 4) were measured in DMSO/LiBr
The structure was characterized by 13 C-NMR spectroscopy indi- and possess DPn values of 104 (3) and 291 (4, Table 1). SEC
cating following signals: 129.9–126.1 ppm (Caryl -Tosyl), 103.3 ppm of the carboxymethylated products in aqueous Na2 HPO4 solu-
(C1), 83.23 ppm (C8), 80.42 ppm (C9), 75.4–73.9 ppm (C2–5), tion gave DPn values of 196 (sample 5) and 162 (sample 6). DPn
61.1 ppm (CH2 OH), 45.9 ppm (CH2 NHCH2 C CH), 38.2 (C7), 21.5 values of 140 and 43 were calculated for the carboxymethyl-6-
(CH3 -Tosyl) (Fig. 1). deoxy-6-aminopropargyl cellulose 9 and 10. It can be concluded
The 6-azido-6-deoxy cellulose derivatives 3 and 4 are soluble in that the polymer was more severely degraded during tosylation
DMA, DMF, and DMSO only and 6-deoxy-6-aminopropargyl cellu- and/or nucleophilic displacement reaction compared with the car-
lose derivatives 7 and 8 are insoluble due to their low DS. In order
to get water-soluble cellulose derivatives, the compounds 3, 4, 7,
and 8 were carboxymethylated heterogeneously in an isopropanol
slurry because the ionic moieties may induce water solubility (Pohl,
Morris, Harding, & Heinze, 2009). In order to achieve a complete
conversion of the starting materials (compounds 3, 4, 7, and 8)
it was indispensable to dissolve or at least swell the polymers
in DMSO followed by precipitation in isopropanol. Otherwise, the
inner parts of the particles were inaccessible for the carboxymethy-
lation reagents. After addition of aqueous NaOH, the activated
polymer was allowed to react with sodium monochloroacetate. The
products were washed with aqueous ethanol and reprecipitated
from water into ethanol to remove side-products. HPLC analy-
sis of the carboxymethylated products (samples 5, 6, 9, 10) after
hydrolysis with HClO4 revealed DSCM values of 1.68 (sample 5),
1.57 (sample 6), 1.58 (sample 9), and 2.05 (sample 10). It must
be noticed that by-products of the carboxymethylation (glycolic
acid and diglycolic acid) could not be completely removed from
the samples, even after multiple reprecipitation (Table 1). Fig. 2. Time dependency of storage (G ) and loss modulus (G  ) during the copper-
The DS of azide- and alkyne moieties of the carboxymethylated catalyzed 1,3-dipolar cycloaddition reaction of carboxymethyl-6-azido-6-deoxy
cellulose and carboxymethyl-6-deoxy-6-aminopropargyl cellulose. For the starting
compounds could not be obtained by elemental analysis due to the materials of gels G9 and G12 see Table 2. No data acquisition from 0 to 30 s due to
fact that these ionic compounds do not burn completely and, hence, gel preparation.
158 A. Koschella et al. / Carbohydrate Polymers 86 (2011) 154–161

Scheme 1. Preparation of hydrogels by copper-catalyzed 1,3-dipolar cycloaddition reaction of carboxymethyl-6-azido-6-deoxy cellulose and carboxymethyl-6-deoxy-6-
aminopropargyl cellulose. (The real degree of substitution of the samples is not considered for clarity reasons.).

boxymethylation step. However, it must be taken into account that time required for gel preparation until the start of the measure-
the molar masses measured may not represent the value of a single ment. Storage modulus G and loss modulus G were recorded. At
polymer chain because aggregation of polymer chains is likely to the beginning, i.e., before the gelation occurs, G is lower than G .
occur due to hydrogen bonds. The time where the cross-over is observed was set as gel point,
i.e., the change from sol to gel state (Fig. 2). The cross-linking reac-
3.2. Preparation and characterization of the hydrogels tion is finished when both moduli do not increase further (Sahiner
et al., 2006; Weng, Chen, & Chen, 2007). However, the end of the
Gels were obtained within few minutes by mixing aqueous reaction could not be determined due to drying of the gel in the
solutions of carboxymethyl-6-azido-6-deoxy cellulose (5, 6) and rheometer. Gelation times between 55 and 1594 s were measured
carboxymethyl-6-deoxy-6-aminopropargyl cellulose (9, 10) with (Table 2).
a freshly prepared mixture of aqueous CuSO4 solution and ascor- Increasing copper concentration caused a faster gelation of
bic acid (Table 2). The molar ratio between azide and alkyne was the mixture as exemplified by the comparison of G1 (catalyst:
1 for every sample. Oscillation rheology was applied to monitor 5 mmol/L, gelation time 1594 s) and G3 (catalyst 10 mmol/L, gela-
the cross-linking reaction. Preliminary experiments showed that a tion time 78 s). Increase of the catalyst concentration from 5
frequency of 1 Hz and a stress of 10,000 Pa are appropriate mea- to 10 mmol/L while maintaining the gel composition constant
suring conditions. The solution prepared was rapidly transferred led to an increase of the gelation rate by a factor of 20. More-
to the rheometer. The times given in Fig. 2 and Table 2 include the over, it was found that an increase of the DS of azide- and
 A. Koschella et al. / Carbohydrate Polymers 86 (2011) 154–161 159

Table 1
Cellulose derivatives used for the hydrogel preparation via copper-catalyzed 1,3-dipolar cycloaddition reaction.

Starting material Product

No. Compound No. Degree of substitution of DPn e PDIf Yield [%] Glycolic Diglycolic
acid [%] acid [%]

Azideb Aminoc CMd

a
1 6-Deoxy-6-azido 3 0.15 – – 104 1.96 74.7 – –
cellulose
2a 4 0.81 – – 291 4.06 95.6 – –
3 Carboxymethyl-6- 5 0.15 - 1.68 196 2.76 90.4 2.7 8.5
deoxy-6-azido
cellulose
4 6 0.81 – 1.57 162 2.92 71.0 8.2 4.4
1 6-Deoxy-6- 7 – 0.19 – – – 80.8 – –
aminopropargyl
cellulose
2 8 – 0.58 – – – 57.5 – –
7 Carboxymethyl-6- 9 – 0.19 1.58 140 5.54 52.9 2.4 50.3
deoxy-6-
aminopropargyl
cellulose
8 10 – 0.58 2.05 43 3.37 31.7 6.3 19.7
a
Cellulose p-toluenesulphonic acid ester with degree of substitution of 0.33 (sample 1) and 0.94 (sample 2).
b
6-Deoxy-6-azido groups.
c
6-Deoxy-6-aminopropargyl groups.
d
Carboxymethyl cellulose.
e
Number average degree of polymerization calculated from the number average molar mass determined by SEC.
f
Polydispersity index.

aminopropargyl groups accelerated the gelation time signifi- could be determined. Gravimetric investigations gave also infor-
cantly. It was also observed that the gelation of a mixture of mation on the amount of gel fraction (Fp ). Surprisingly low
carboxymethyl-6-azido-6-deoxy cellulose with DSN 0.81 (sample amounts in the range from 34 to 60% were measured indicating
6) and carboxymethyl-6-deoxy-6-aminopropargyl cellulose with a comparably low yield of the cross-linking reaction. The high-
DSAmino 0.19 (sample 9) occurred very fast (sample G4, gel point est gel fraction and, hence, a higher yield was found for samples
421 s). Both components are mixed in equimolar amounts with with high DS of cross-linking groups. Noticeable, the maximum
regard to azide- and alkyne groups. water content decreased with increasing gel fraction. An expla-
The freshly prepared gels are transparent with a green-brownish nation is that a higher cross-linking density leads to a more
colour (Fig. 3). They are capable of further swelling in water rigid network with a lower absorption capacity than more flex-
but lose their mechanical stability, i.e., the gels disintegrate. ible gels (Po, 1994). The gels were freeze-dried prior to further
Therefore, it was impossible to monitor the swelling kinetics. How- analysis obtaining a solid spongy mass. All structural features
ever, maximum water content (Qeq ) in the range of 96.8–98.4% could be detected in the FTIR spectra (Fig. 4). The character-

Table 2
Composition and properties of hydrogels prepared from carboxymethylated 6-deoxy-6-azido cellulose and carboxymethylated 6-deoxy-6-aminopropargyl cellulose via
copper-catalyzed 1,3-dipolar cycloaddition reaction.

Conditions Result

6-Deoxy-6-azido 6-Deoxy-6- Functional group Cu2+ [mmol/L] Gel Gel fraction [%] Max. water content [%] Gel point [s]
carboxymethyl aminopropargyl [mmol/L]
cellulosea carboxymethyl
cellulosea

Sample Conc. [wt%] Sample Conc. [wt%]

5 6.15 9 5.0 30.5 5.0 G1 n.d. n.d. 1594

5 6.15 9 5.0 30.5 7.5 G2 n.d. n.d. 253
5 6.15 9 5.0 30.5 10.0 G3 34.3 98.4 78
6 1.07 9 5.0 30.5 5.0 G4 n.d. n.d. 421
6 1.07 9 5.0 30.5 7.5 G5 n.d. n.d. 96
6 1.07 9 5.0 30.5 10.0 G6 42.0 98.3 ∼55
6 2.89 10 5.0 82.5 5.0 G7 50.1 97.7 500
6 2.89 10 5.0 82.5 7.5 G8 59.9 96.8 292
6 2.89 10 5.0 82.5 10.0 G9 54.2 96.9 158
5 6.15 10 1.8 30.5 5.0 G10 n.d. n.d. 341
5 6.15 10 1.8 30.5 7.5 G11 n.d. n.d. 198
5 6.15 10 1.8 30.5 10.0 G12 45.0 98.45 131

Not determined (n.d.).

a
The functional groups were mixed in equimolar ratio.
160 A. Koschella et al. / Carbohydrate Polymers 86 (2011) 154–161

Fig. 3. Freshly prepared hydrogels obtained by copper-catalyzed 1,3-dipolar cycloaddition reaction of carboxymethyl-6-azido-6-deoxy cellulose and carboxymethyl-6-
deoxy-6-aminopropargyl cellulose. For starting materials of the gels see Table 2.

istic band of the azide moiety at 2120 cm−1 was still present,
which indicated that not all azide groups were consumed during
the 1,3-dipolar cycloaddition reaction. However, remaining azide-
or alkyne groups can be used to attach further functional moi-
eties.
SEM images of G3, G6, G9, and G12 were taken to visu-
alise the porous structure of the dried hydrogels (Fig. 5). G3
and G6 exhibit an irregular but closed structure while G9 and
G12 show porous topology as usual for freeze dried hydro-
gels. This might be caused by the high amount of diglycolic
acid in compound 10. This side product closes the pores of the
gels. However, due to the good water solubility of glycolic acid
and diglycolic acid they do not influence the swelling proper-
ties of the gels remarkably. The comparison of three gels with
same composition but different amounts of catalyst is shown in
Fig. 5. Obviously, the structure became more regular with an
increasing copper concentration (G7 5 mmol/L, G8 7.5 mmol/L,
Fig. 4. FTIR spectra of selected freeze dried gels obtained by copper-catalyzed
and G9 10 mmol/L). The pore size of the samples was nearly the
1,3-dipolar cycloaddition reaction of carboxymethyl-6-azido-6-deoxy cellulose and
carboxymethyl-6-deoxy-6-aminopropargyl cellulose. For starting materials of the same.
gels see Table 2.

Fig. 5. SEM images of selected freeze dried gels obtained by copper-catalyzed 1,3-dipolar cycloaddition reaction of carboxymethyl-6-azido-6-deoxy cellulose and
carboxymethyl-6-deoxy-6-aminopropargyl cellulose. The surface of the samples is shown. For starting materials of the gels see Table 2.
 A. Koschella et al. / Carbohydrate Polymers 86 (2011) 154–161 161

4. Conclusions Chang, C., & Zhang, L. (2011). Cellulose-based hydrogels: present status and appli-
cation prospects. Carbohydrate Polymers, 84, 40–53.
Heinze, T., Koschella, A., Brackhagen, M., Engelhardt, J., & Nachtkamp, K. (2006).
It was shown that copper mediated 1,3-dipolar cycloaddition Studies on non-natural deoxyammonium cellulose. Macromolecular Symposia,
of water soluble cellulose derivatives decorated with azide- and 244, 74–82.
alkyne functions yields novel cellulose hydrogels. The gels were Heinze, T., & Pfeiffer, K. (1999). Studies on the synthesis and characterization of
carboxymethylcellulose. Angewandte Makromolekulare Chemie, 266, 37–45.
investigated via FTIR spectroscopy, oscillation rheology, SEM; also, Lee, K. Y., & Mooney, D. J. (2001). Hydrogels for tissue engineering. Chemical Reviews,
the degree of swelling were measured. The gelation time can be 101, 1869–1879.
controlled by the DS of azide- and alkyne groups as well as by Li, L., Lin, Z. B., Yang, X., Wan, Z. Z., & Cui, S. X. (2009). A novel cellulose hydrogel
prepared from its ionic liquid solution. Chinese Science Bulletin, 54, 1622–1625.
the catalyst concentration. FTIR spectroscopy revealed remaining
Ma, J., Zhang, L., Fan, B., Xu, Y., & Liang, B. (2008). A novel sodium
alkyne- and azide groups in the gels that might be used to bind carboxymethylcellulose/poly(N-isopropylacrylamide)/clay semi-IPN nanocom-
bioactive molecules or other substances in the gels. The limiting posite hydrogel with improved response rate and mechanical properties. Journal
of Polymer Science, Part B: Polymer Physics, 46, 1546–1555.
factor for applications is the low mechanical stability, which is
Park, H., & Park, K. (1996). Hydrogels in bioapplications. ASC Symposium Series, 627,
probably related to the low yield of the cross-linking reaction. Fur- 2–10.
ther investigations will be carried out to improve the mechanical Po, R. (1994). Water-absorbent polymers: a patent survey. Journal of Macromolecular
stability of the gels. Moreover, it would be interesting to study the Science, Reviews Macromolecular Chemistry and Physics, C34, 607–662.
Pohl, M., Morris, G. A., Harding, S. E., & Heinze, Th. (2009). Studies on the molecular
response of the gels to different stimuli, e.g., temperature and pH flexibility of novel dendronized carboxymethyl cellulose derivatives. European
of the aqueous medium. Polymer Journal, 45, 1098–1110.
Pohl, M., & Heinze, T. (2008). Novel biopolymer structures synthesized by den-
dronization of 6-deoxy-6-aminopropargyl cellulose. Macromolecular Rapid
Acknowledgements Communications, 29, 1739–1745.
Pourjavadi, A., Zohuriaan-Mehr, M. J., Ghasempoori, S. N., & Hossienzadeh, H. (2007).
The research leading to these results has received funding Modified CMC. V. Synthesis and super-swelling behavior of hydrolyzed CMC-g-
PAN hydrogel. Journal of Applied Polymer Science, 103, 877–883.
from the European Community’s Seventh Framework Programme Rahn, K., Diamantoglou, M., Klemm, D., Berghmans, H., & Heinze, T. (1996).
(FP7/2007-2013) under grant agreement no. 214653 (surface func- Homogeneous synthesis of cellulose p-toluenesulphonates in N,N-
tionalization of cellulose matrices using coatings of functionalized dimethylacetamide/LiCl solvent system. Angewandte Makromolekulare
Chemie, 238, 143–163.
polysaccharides with embedded nano-particles, SurFunCell). Cel- Sahiner, N., Singh, M., John, De Kee, D., Vijay, T., & McPherson, G. L. (2006). Rheolog-
lulose p-toluenesulphonic acid ester (tosyl cellulose) 1 with a DS of ical characterization of a charged cationic hydrogel network across the gelation
0.33 was a kind gift of M. Schöbitz (Thuringian Institute for Textile boundary. Polymer, 47, 1124–1131.
Shang, J., Shao, Z., & Chen, X. (2008). Electrical behavior of a natural polyelectrolyte
and Plastics Research, Rudolstadt, Germany).
hydrogel: chitosan/carboxymethylcellulose hydrogel. Biomacromolecules, 9,
1208–1213.
Weng, L., Chen, X., & Chen, W. (2007). Rheological characterization of in situ
References crosslinkable hydrogels formulated from oxidized dextran and N-carboxyethyl
chitosan. Biomacromolecules, 8, 1109–1115.
Alupei, I. C., Popa, M., Hamcerencu, M., & Abadie, M. J. M. (2002). Superabsorbent Yu, L., & Ding, J. (2008). Injectable hydrogels as unique biomedical materials. Chem-
hydrogels based on xanthan and poly(vinyl alcohol). 1. The study of the swelling ical Society Reviews, 37, 1473–1481.
properties. European Polymer Journal, 38, 2313–2320. Zhang, J., Xu, X. D., Wu, D. Q., Zhang, X. Z., & Zhuo, R. X. (2009). Synthesis of ther-
Bajpai, A. K., Shukla, S. K., Bhanu, S., & Kankane, S. (2008). Responsive polymers in mosensitive P(NIPAAm-co-HEMA)/cellulose hydrogels via “click” chemistry.
controlled drug delivery. Progress in Polymer Science, 33, 1088–1118. Carbohydrate Polymers, 77, 583–589.
Buchholz, F. L. (1996). Superabsorbent polymers: an idea whose time has come. Zhou, J., Chang, C., Zhang, R., & Zhang, L. (2007). Hydrogels prepared from unsub-
Journal of Chemical Education, 73, 512–515. stituted cellulose in NaOH/urea aqueous solution. Macromolecular Bioscience, 7,
Calvert, P. (2009). Hydrogels for soft machines. Advanced Materials, 21, 743–756. 804–809.