Carbohydrate Research 487 (2020) 107861
Contents lists available at ScienceDirect
Carbohydrate Research journal homepage: www.elsevier.com/locate/carres
2D NMR assisted structure elucidation of three cyanoethylated cellulose derivatives and correlated with their properties
T
Yan Zhang, Xiaofei Shen, Hao Qian, Lei Song, Kaiyun Xie, Mingtao Zhang, Huiqing Wang∗ Department of Polymer Science and Engineering, School of Chemistry and Chemical Engineering, Hefei University of Technology, Hefei, Anhui Province, 230009, China
ARTICLE INFO
ABSTRACT
Keywords: Cyanoethyl cellulose 2D NMR COSY HSQC
Three cellulose derivatives CEC, CEHEC, and CEGEC prepared by cyanoethylation of cellulose, hydroxyethyl cellulose (HEC), and glycerol ether cellulose (GEC), respectively, with similar degrees of substitution, exhibited different properties involving their dielectric application. But their structures were hard to distinguish by FTIR, 1 H NMR, or 13C NMR spectroscopy. Here, 2D NMR techniques (COSY, HSQC) assisted by partial hydrolysis and full acetylation were applied and demonstrated to be useful for structural elucidation, differentiated their peaks on 2D NMR map. Furthermore, GPC-MALLS revealed that CEHEC has highest branching ratio, while CEC was strictly linear with similar molecular weights. Branching CEHEC exhibited a lowest Tg, while CEGEC showed better dielectric properties due to the relative mobility of the cyano groups at the end of the glycerol moieties. So this work open the way of exploring structure of carbohydrate polymer derivatives assisting by 2D NMR techniques COSY and HSQC.
1. Introduction
that these materials exhibited different properties, even with similar degrees of cyanoethyl substitution. Therefore, it is necessary to explore their structures to explain their properties. Many efforts have been made to explore the structures of cellulose derivatives. 1D and 2D NMR spectroscopy has been used to reveal the partial DS at the O-2, O-3 and O-6 positions with the assistance of partial or total hydrolysis by acid such as sulphuric acid [9] or methanol-hydrochloric acid [10] or by hydrolysing enzymes such as exoglucanases [11], endoglucanases and β-glucosidases [12]. NMR studied numerous cellulose derivatives, e.g. carboxy-methylcellulose CMC, hydroxyethyl cellulose HEC [13,14], hydroxypropyl cellulose HPC [9,15], methylcellulose [16], ethyl cellulose [17] cellulose acetate [18] and cyanoethyl cellulose [19,20]. However, 1D NMR methods are less useful for the structural analysis of mixed ether cellulose such as our derivatives CEHEC and CEGEC. Mixed ether cellulose usually show better properties in applications than single ether derivatives. To date, only the molecular structure of hydroxypropyl methyl cellulose (HPMC), an important abundantly used mixed ether cellulose derivative, has been explored by 1H-and 13C NMR [10,21]. COSY and HSQC NMR [22] had been used to analyze the structure of cellulose acetate [23], 3-O-(2-methoxyethyl)cellulose [24], 3-monoO-hydroxyethyl cellulose [25], 3-mono-O-ethyl cellulose [26] 3-Oethylene glycol cellulose [27], sulfoethylated cellulose [28], methylcellulose [29]. Here, we have used 2D NMR techniques to explore the
Cellulose is an abundant natural resource that is available from plants, wood, cotton, seaweeds, or Ascidiacea, with structure of linear 1,4-β-d-glucan with three hydroxyl groups at the 2, 3, and 6 positions per anhydroglucose units (AGU). Cellulose derivatives by performing reactions on the three hydroxyl groups of the AGUs can endow them with new properties in application. Cyanoethyl cellulose (CEC) [1] is one important cellulose derivative prepared via the michael addition reaction between cellulose and acrylonitrile, exhibits improved antiacid property, microbiological and moisture resistance at low cyanoethyl substitution [2,3], and shows high dielectric constant and low dielectric loss factor when substituent degree is more than 2, because of the strong electron-attracting cyanogen group, that could be used as high dielectric film and insulating medium in miniature capacitors [3], high-temperature capacitors [4] electroluminescent materials [5]. We explored solvent-media technology [6] to obtain high substituent cellulose derivatives including CEC, which show better dielectric properties with lower etherification agent consumption than those using the traditional heterogenerous method [7] or the homogeneous method [8], we also applied solvent-media technology to prepare mixed cyanoethyl ethers such as cyanoethyl hydroxyethyl cellulose (CHEC) and cyanoethyl glycerol cellulose (CGEC) by cyanoethylation of hydroxyl ethyl cellulose and glycerol ether cellulose, respectively. It was found
∗
Corresponding author. E-mail address:
[email protected] (H. Wang).
https://doi.org/10.1016/j.carres.2019.107861 Received 3 September 2019; Received in revised form 17 October 2019; Accepted 31 October 2019 Available online 02 November 2019 0008-6215/ © 2019 Elsevier Ltd. All rights reserved.
Carbohydrate Research 487 (2020) 107861
Y. Zhang, et al.
structures of cellulose mixed-ether CEHEC, CEGEC and CEC. In addition their molecular weight distributions, branching ratios, glass transition temperatures, and dielectric properties were compared in order to investigate the correlation between their properties and their molecular structures.
2.5. Other characterizations ATR-FTIR spectra of dry samples were recorded from 500 to 4000 cm−1 using an L128-0099 PerkinElmer Spectrometer (Waltham, MA, USA). Measurements of molecular weights, molecular weight distributions, and branching analyses were performed using a GPC-MALLS system using N,N-dimethylacetamide/lithium chloride (0.9% w/v, DMAc/LiCl) as an eluent. All samples were dissolved in DMAc/LiCl(9% w/v) (DMAc/LiCl) before measurement. Thermal properties were tested using a NETZSCH differential scanning calorimeter (DSC), with samples first being heated to 150 °C at 20 °C/min to clean up their thermal history, then cooled to −100 °C, and finally their thermal curves were measured up to 150 °C at 10 °C/min. Dielectric properties were tested using an Agilent 4294A impedance analyzer, with samples consisting of 0.5 cm × 0.5 cm thin films of derivatives coated with a conductive silver paste adhesive on both surfaces.
2. Methods and materials 2.1. Materials CEC, CEHEC, and CEGEC samples were all prepared by starting from raw cotton linter (M100, DP ~1000). CEC with DSCN = 2.4, CEHEC was prepared by cyanoethylation of hydroxyethyl cellulose with MSHE = 0.5, MSCN = 2.46. CEGEC was prepared from glyceryl ether cellulose with MSGE = 0.2, MSCN = 2.56. All DS and MS values were measured by elementary analysis methods.
3. Results and discussion
2.2. Partial hydrolysis of cyanoethylated derivatives
3.1. Comparison by ATR-FTIR
Samples (50 mg) were placed in a 7 mL vial containing 6 mL of 2 M trifluoroacetic acid for 15 min at 120 °C to allow partial hydrolysis to occur. After cooling to room temperature, the solvent was evaporated and residual acid was removed by five rounds of co-distillation with toluene.
FTIR spectra of pure cotton, hydroxyethyl cellulose (HEC), and glycerol ether cellulose (GEC) and their corresponding cyanoethyl derivatives (CEC, CEHEC, and CEGEC, respectively) were compared, as shown in Fig. 1A-C. They exhibited similar signals, with the free –OH peaks at ~3500 cm−1 significantly decreasing and a strong signal that is characteristic of the CN group appearing at 2240-2260 cm−1 after cyanoethylation of all materials. In addition, the C–H bending vibration peak at 1200–1400 cm−1 moved to a lower frequency as adjacent OH groups were substituted with cyanoethyl groups. However, it is very difficult to find other obvious differences between the FTIR signals of the derivatives. To overcome this lack of structural information, NMR spectroscopic techniques were used.
2.3. Fully acetylation of cyanoethylated derivatives Unsubstituted hydroxyl groups of the derivatives were fully acetylated after partial hydrolysis. The derivatives were treated with 1 mL of acetic anhydride in 3 mL of pyridine at 90 °C for 3 h. Saturated NaHCO3 solution was added, then the product was washed with water and dissolved in acetone, then precipitated into ethanol, filtered, and dried. The acetylation procedure was repeated until no free hydroxyl absorption was detected in the Fourier transform infrared (FTIR) spectrum, with the final samples being labeled as ACEC, ACEHEC, and ACEGEC, and their structures were illustrated as Scheme 1.
3.2. Effect of pretreatment on NMR spectrum For the two-dimensional NMR spectroscopic analysis, low molecular weight cellulose samples was necessary to obtain high resolution and high signal/noise ratios [30]. Derivatives were pretreated by partial hydrolysis and full acetylation before NMR spectra measures Hydrolysis can reduce the molecular weight of samples, while acetylation can separate substituent group peaks and improve solubility in the NMR solvent [31], resulting in better NMR signal resolution. Full acetylation was proved by the complete disappearance of the free –OH absorption band near 3500 cm−1 in the FTIR spectra (Fig. S1) and the emergence of new peaks at 1739 cm−1 and 1230 cm−1 related to C]O and C–O bonds. However, the characteristic CN peaks at 2240-2260 cm−1 did not become weaker or shifted during the pretreatment, indicating that no substitutions were destroyed. As shown in Fig. 2, peaks in the 13C NMR spectrum became sharper after pretreatment of the ACEGEC sample. The signal of C-1 at 102 ppm increased in intensity without shifting, so the neighboring C-2 structure was still intact. The CN groups displayed three sharp high resolution peaks in the region 118–121 ppm that belong to the cyanoethyl groups on positions O-6, O-3, and O-2, and at the end of the glycerol ether of ACEGEC (labeled as *), whereas only two broad peaks were present before pretreatment. In addition, the methylenes adjacent to the CN groups led to the presence of three peaks at 17–19 ppm after pretreatment. Therefore, it can be seen that pretreatment is a necessary and helpful chemical method for detecting NMR spectroscopy. AGU carbon signals in the region of 60–85 ppm caused by C-2, C-3, C-4, C-5, and C-6 can separate as shown in Fig. 2, and new signals at 79–82 ppm, 74–75 ppm, 71–72 ppm, and 64–66 ppm were observed, with these most likely belonging to the glycerol ether, methylene carbon, and second cyanoethyl methylene carbon resonance signals, but they are hard to be determined only by 13C NMR spectrum.
2.4. Determination of degree of substitution by NMR The pretreated derivatives were dissolved in DMSO‑d6 at concentrations of 20 mg/mL for 1H NMR measurements and 100 mg/mL for 13C NMR measurements.1H,1H COSY and HSQC-DEPT were recorded at 60 °C using a Bruker Avance II 400 with a 5 mm broadband observe probe head equipped with z-gradient and standard Bruker programs. Quantitative 13C NMR spectra were recorded with a pulse angle of 4530°, a relaxation time of 30 s, and 1H decoupling only during the acquisition time. Chemical shifts were calibrated using DMSO‑d6 as an internal reference (1H = 2.49 ppm and 13C = 39.6 ppm).
Scheme 1. Structural formula of acetylated cyanoethylated cellulose derivatives. 2
Carbohydrate Research 487 (2020) 107861
Y. Zhang, et al.
Fig. 1. ATR-FTIR spectra of: (A) cotton linter (a) and CEC (b); (B) HEC (a) and CEHEC (b); (C) GEC (a) and CEGEC (b).
3.3. Comparison by 1D NMR spectroscopy
assigned to the cellulose AGU units, hydroxyethyl group of ACEHEC, and glyceryl group of ACEGEC. The 85-60 ppm was magnified in Fig. 3B (inset), ACEHEC presented one more signal than ACEC at 70 ppm, likely belonging to hydroxyethyl carbons C-9 to C-12 according to the literature [30]. On the other hand, ACEGEC showed two more signals, with one peak appearing at 70 ppm (glyceryl C-9) and the other at 65 ppm (glyceryl C-10). Note that glyceryl branching was still not confirmed. 13C NMR spectroscopy showed more information than 1H NMR spectroscopy, but it is difficult to confirm the structures of CEHEC and CEGEC solely based on 13C NMR spectroscopy.
The structures of the derivatives of ACE, ACEHEC, and ACEGEC were investigated by NMR spectroscopy, and their 1H-and 13C NMR spectra after pretreatment are shown in Fig. 3. The samples had similar 1 H spectra, as shown in Fig. 3. A, with signals at 2.6–2.9 ppm for the CH2–CH2–CN protons (the cyanoethyl methylene protons), 1.8–2.1 ppm for the acetyl group, and 3.0–5.5 ppm for the glucose units, with the CH2–CH2–CN H-7 signal and the methylene protons of the hydroxyethyl groups being in this range as well. Therefore, it is impossible to identify their structures by comparing their 1H NMR spectra. The 13C NMR spectra of ACEC, ACEHEC, and ACEGEC were compared in Fig. 3B, expressed many same signals as their similar structures, such as peaks at 168–171 ppm for CH3–COO, 120-118 ppm for CN, 105-102 ppm for C-1, 16–18 ppm for CH2–CH2–CN, and 18–21 ppm for CH3–COO. Some differences existed in the 60–85 ppm region, which
Fig. 2.
13
3.4. Elucidation of derivative structures using 2D NMR It is not possible to assign of the signals of the substituent only with COSY NMR spectra, so here combined COSY and HSQC-DEPT NMR techniques to analyze CEC, CEHEC, and CEGEC samples after
C NMR spectra of ACEGEC (a) before and (b) after pretreatment. 3
Carbohydrate Research 487 (2020) 107861
Y. Zhang, et al.
Fig. 3. 1H NMR (A) and
13
C NMR (B) spectra of (a) ACEC, (b) ACEHEC, and (c) ACEGEC.
Fig. 4. 1H,1H COSY (a) and HSQC-DEPT (b) NMR spectraS of ACEC.
Fig. 5. 1H,1H COSY (a) and HSQC-DEPT (b) NMR spectra of ACEHEC.
pretreatment in order to provide more structural information, as shown in Figs. 4–6. The 1H–1H COSY NMR spectrum of CEC after pretreatment is shown in Fig. 4 a. H-1 peak was found at 4.30–4.55 ppm as a reference, from the off-diagonal signals, three signals at 2.83, 3.03, and 3.25 ppm belonging to H-2 were found. The signal at 3.34 ppm belongs to H-3 where cyanoethyl substitution had occurred, while another H-3 signal appears at 5.05 ppm because of acetylation of the free hydroxyl on C-3. And H-4 was located at 3.76 to 3.57 ppm, while the signals at
3.57–3.77 ppm being attributed to the H-5 and H-6 protons. The signal at 2.72 ppm relates to H-8 (CH2–CH2–CN), while the H-7 (CH2–CH2–CN) signal is overlapped with H-6 proton signal in COSY NMR. By using HSQC-DEPT NMR spectroscopic technique, it is possible to differentiate the overlapped peaks. HSQC NMR spectrum of ACEC (Fig. 4b) shows distinct peaks identified as C-6 at 65.60 ppm, and C-5 at 73.04 ppm, with their corresponding protons being found at 3.48–3.70 ppm (H-6), and 3.55–3.80 ppm (H-5). Two peaks were 4
Carbohydrate Research 487 (2020) 107861
Y. Zhang, et al.
Fig. 6. 1H,1H COSY (a) and HSQC-DEPT (b) NMR spectra of ACEGEC.
distinguished for C7 at 67.35 ppm and C8 at 18.45 ppm with crosspeaks to H-7 (3.75 ppm), and H-8 (2.72 ppm) respectively, of the cyanoethyl methylene groups in ACEC. All signals are good agreement with HSQC map of reported cyanoethyl cellulose [19]. Fig. 5 shows 1H,1H COSY (a) and HSQC-DEPT (b) NMR spectroscopy of CEHEC after partial hydrolysis and fully acetylation. In Fig. 5a, H-1 was found at 4.3–4.6 ppm, then from cross peaks, the signals for H-2 were determined at 2.85, 3.03, and 3.24 ppm as a result of substitution by cyanoethyl, hydroxyethyl and acetyl in position 2. whereas H-3 signals appeared at 3.4 ppm (by cyanoethylation in position 3) and 4.95 ppm (by acetylation in position 3). So the peaks for H-4 located at 3.75 to 3.6 ppm because of the substitution pattern (cyanoethylation or acetylation) at neighboured position 3 [30]. The signals in the 3.57–3.77 ppm region belong to H-5 and H-6, while H-8 (CH2–CH2–CN) was found at 2.72 ppm. It is not possible to assign of the signals H-7 of cyanoethyl and hydroxyl ethyl substituent H-9 to H-12 only with COSY NMR spectra. By using HSQC-DEPT NMR spectroscopic technique, it is possible to differentiate the peaks for the hydroxyethyl and cyanoethyl substituent. In Fig. 5b, C-5 signal was located at 72.96 ppm with cross-peaks to H-5 3.54 ppm, C-6 was found at 65.95 ppm with cross-peaks to H-6 at 3.40–3.70 ppm. Two peaks were distinguished for C7 at 66.83 ppm and C8 at 18.25 ppm with cross-peaks to H-7 (3.82 ppm), and H-8 (2.70 ppm) respectively, of the cyanoethyl methylene groups in ACEHEC. In the region of the hydroxyethyl methylene carbons
Fig. 7. Calculation of the DSCN of CEGEC by the C-1 method.
Fig. 8. Calculation of the DSCN of CEHEC in the 1H NMR spectrum.
Fig. 9. 1H,1H COSY (a) and HSQC-DEPT (b) spectra of AGE with MSGE = 0.2. 5
Carbohydrate Research 487 (2020) 107861
Y. Zhang, et al.
Fig. 10. GPC curves (a) and aggregate sizes in solution (b) of CEC, CEHEC, and CEGEC.
field region around 70 ppm, meaning that no branching occurred on GEC. This is evident from the 2D NMR of GEC and the DS calculation for CEGEC later.
Table 1 GPC results for the three cyanoethylated cellulose derivatives. Sample CEC2.5 CEHEC2.73 CEGEC 2.72
Mn (g/mol) 4
5.187 × 10 6.792 × 104 6.477 × 104
Mw (g/mol) 5
1.076 × 10 1.207 × 105 1.092 × 105
Mz (g/mol)
Mw/Mn 5
1.843 × 10 1.812 × 105 1.762 × 105
2.075 1.777 1.686
3.5. Degree of substitution Estimation of the cyanoethyl substitution of three derivatives was carried out by 1H and 13C NMR spectroscopic analyses after the partial hydrolysis and fully acetylation. As cyano groups showed unique signals at 118–120 ppm in the 13C NMR spectra of all three derivatives, that separated from other carbon signals, allowing the total degree of cyanoethyl substitution to be calculated by calculating the ratio of the cyano carbon (δ118-120) integrated intensities and the C-1 carbon signals, known as the C-1 method. The cyanoethyl DS results were 2.50, 2.73, and 2.72 for CEC, CEHEC, and CEGEC, respectively, from their 13 C NMR spectra in Figs. S3 and S4, and Fig. 7, respectively. The degree of substitution (DS) of cyanoethyl groups could also be calculated from the ratios of the 1H integrated intensities in acetyl methyl CH3–COO at 1.9–2.1 ppm and cyanoethyl methylene CH2–CN (δ2.54–2.89) protons in 1H NMR spectrum. Since the CH2–CN proton signal a signals did not overlap with any other methylene protons such as the hydroxyethyl or glycerol ether protons as seen in their HSQC NMR spectras, therefore the quantification of DSCN using these peaks is acceptable. As shown in Fig. S5, DSCN of CEC was 2.46, while DSCN of CEHEC was 2.56 as shown in Fig. 8, with these values being determined by the 1H method. Calculating the DS is much more complicated for CEGEC because each glycerol ether group could branch into two or more hydroxyl groups. So the structure of GEC was first analyzed, 2DCOSY and HSQC NMR spectra of GEC after pretreatment are shown in Fig. 9. The substituted glycerol ether C-8 and C-9 signals did not overlap with the C-6 signal of AGU that has been reported in the literature [31]. As no branching of this GEC occurred, with MSGE = 0.2, the degree of substitution can be calculated by the same method as used before, meaning that the DSCN of CEGEC is 2.66, as shown in Fig. S6. The DS calculated by the 1H method is slightly higher than EA method, but lower than C-1 method, the reason could be seen in Fig. S7
substituent in neighbourhood of ether functions (66.3–72.01 ppm), signals for C-atoms were detected, which were assigned to C9, C10, C11, whereas, peaks for C12 (71.5 ppm) were determined in the region of the signals of the methylene protons in neighbourhood of cyanoethyl functions. Indicated by the crosspeaks, the signals of the protons at C9, C10, C11 were detected at 3.42–3.89 ppm as a result of an ether function in the neighbourhood. The protons at C12 was observed at 3.75 ppm. The signals for Hydroxyethyl substituent are similar as reported 2,3-O-Hydroxyethyl Cellulose [30] and 3-mono-O-hydroxyethyl cellulose [32]. COSY and HSQC NMR spectroscopy were also applied to clarify the structure of ACEGEC. As shown in Fig. 6 a, the 1H–1H COSY NMR spectrum, based on the H-1 signal at 4.30–4.50 ppm, H-2 peaks were located at 2.82, 3.03, and 3.24 ppm as C-2 was substituted by cyanoethyl or glycerol, while no low field signals above 4.00 ppm, indicating that the DS on C-2 was 1. The H-3 signal was found at 3.30 ppm and 4.93 ppm, while H-4 was located from 3.78 to 3.55 ppm. H-5 and H-6 were both in the region 3.50–3.80 ppm. The HSQC NMR spectrum of ACEGEC is shown in Fig. 6 b, (H-5, C5) signals being found at (3.50–3.65 ppm, 72.8 ppm) and (H-6, C-6) peaks at (3.40–3.75 ppm, 65.4 ppm), and two peaks were distinguished for C7 at 67.10 ppm and C8 at 18.20 ppm with cross-peaks to H-7 (3.84 ppm), and H-8 (2.85 ppm) respectively, of the cyanoethyl methylene groups in ACEGEC, which are the similar as the C7, C8 signals in ACEC and ACEHEC samples, too. The glycerol ether methyl C-10 peak was found at 64.60 ppm. While 70.30 ppm being attributed to glycerol methylene C-9. But no further signals were observed in the low
Fig. 11. (a) Branching ratio vs molar mass and (b) long chain branching vs molar mass for CEHEC and CEGEC. 6
Carbohydrate Research 487 (2020) 107861
Y. Zhang, et al.
Fig. 12. DSC curves of (a) CEC, (b) CEHEC, and (c) CEGEC.
small branching ratio due to branch-like hydroxyethyl grafting of the cellulose backbone, and the branching ratios of both CEHEC and CEGEC were higher at higher molecular weights. However, long chain branching predominantly occurred at relatively low molecular weights, with CEGEC exhibiting long chain branching at a molar mass of 104 g/ mol, while almost no long chain branching was observed for either CEGEC or CEHEC at molar masses of 105 g/mol, which is consistent with the HSQC NMR results. 3.7. Comparison of glass transition temperatures The glass transition temperature (Tg) is the temperature at which polymer chain segments begin to move, and so a lower Tg indicates a more flexible chain. DSC curves of the three cyanoethylated cellulose derivatives were compared, as shown in Fig. 12. In CEC, a secondary transition occurred at Tβ = 49.2 °C related to CEC chain segments, followed by complete segmental movement Tg at 55.3 °C. CEGEC chain segment motion began at Tβ = 44.2 °C and Tg = 49.8 °C, while CEHEC chain segment motion began at the lowest temperature observed, Tβ = 36.6 °C and Tg = 41.2 °C. This was due to the existence of hydroxy ethyl substituents (MS = 0.51) with long chain segments that contributed to good flexibility, which is similar as reported literature of hydroxyalkyl cellulose ether nitrate [18].
Fig. 13. Dielectric constants of CEC, CEHEC, and CEGEC.
of HSQC spectrum. CH2–CN proton partially overlapped with proton H2, induced increased value than EA method, but, the CH2–CN proton signal partially cut by the DMSO solvent signal in the HSQC NMR spectrum, resulting lower value than C-1 method. It should be noted that here the proposed ‘H-1’ and ‘C-1’ methods are semi-quantitative methods. So we further analyzed the CH2–CN proton and CH3–COO signals shape area in HSQC map, the DSCN of CEC was estimated to be 2.25, while DSCN of CEHEC was 2.36 and DSCN of CEGEC was 2.44, all the substitution values are lower than 1H or 13C NMR method since the overlapped area was removed.
3.8. Comparison of dielectric properties The substituent CN groups endowed the CEC with dielectric character [33], here the dielectric property of cyanoethylated cellulose derivatives with similar substitution were compared. As shown in Fig. 13, the dielectric constants decreased in trend of CEGEC2.72 > CEHEC2.73 > CEC2.50, the dielectric constants of CEC, CEHEC, and CEGEC were 15, 16, and 20 at a frequency of 100 Hz, respectively. Dielectric constant depends on the degree of polarization, with polar groups on side positions having more freedom to be polarized than those on the polymer backbone, thereby inducing higher dielectric constants. Asymmetric polar structures show higher dielectric constants than symmetric molecules. Accordingly, CEGEC exhibited the highest dielectric constant due to the relative mobility of the cyano groups at the end of the glycerol substituents.
3.6. Comparison of molecular weights and degrees of branching GPC-MALLS measurement provided not only average molecular weight, molecular weight distributions, but also aggregate size and branching information. The data from GPC with DMAc/LiCl (0.9% w/v) as an eluent were shown in Fig. 10 a, and Table 1. The three samples showed similar molecular weights as the derivatives used the same batch of refined cotton as their raw material. The average molecular weight (Mw) of CEC was 1.076 × 105 g/mol, increasing to 1.207 × 105 g/mol for CEHEC and to 1.092 × 105 g/mol for CEGEC, their corresponding molecular weight distribution (Mw/Mn) decreased from 2.075 to 1.777 and 1.686 respectively. As more reaction and more purification steps led to narrow distributions, as well higher degree of substitution lead to improved solubility, enabling more uniform substitution and narrower molecular weight distributions. The aggregate size of polymer chains in solution reflects the interaction between the polymer chains and the solvent. If the chains stretch out, larger aggregates are observed, while if the chains repel each other, then the aggregates become smaller. As shown in Fig. 10b, the aggregate sizes of the three cyanoethyl derivatives in DMAc/LiCl (0.9% w/v) increased in the following order: CEC < CEHEC < CEGEC. This indicates that chains of mixed ethers stretched the most, with the glycerol ether derivative CEGEC stretching the most in solution. Branching ratio and long chain branching information can also be obtained from the GPC-MALLS curves, as shown in Fig. 11. No branching data could be recorded for CEC, indicating that the chains were linear, while the other two samples were branched. CEHEC had a
4. Conclusion Various techniques such as FTIR, 1H NMR, 13C NMR, and 2D NMR (COSY, HSQC) were used to determine structure of three cyanoethylated cellulose derivatives: CEC, CEHEC, and CEGEC. 2D NMR (COSY, HSQC) proved to be most useful as it could distinguish between the complicated structures of the mixed ether derivatives, which was impossible when using FTIR, 1H NMR and 13C NMR spectroscopy. Degree of substitution were measured by 1H and C-1 methods. Furthermore their molecular weight distributions, branching ratio, glass transition temperatures, and dielectric constants were compared, and the differences were explained by corresponding structures. This work showed 2D NMR analysis is a useful tool to explore structure of mixed ether cellulose derivatives for explaining specific properties. 7
Carbohydrate Research 487 (2020) 107861
Y. Zhang, et al.
Declaration of competing interest We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled.
[13] [14] [15] [16]
Acknowledgements
[17]
The authors would like to thank the financial supports of the National Natural Science Foundation of China (51603059), Fundamental Research Fund for the Central Universities of China (1064115100027). China Postdoctoral Science Foundation and Anhui Provence Postdoctoral Science Foundation are thankfully acknowledged.
[18] [19] [20]
Appendix A. Supplementary data
[21]
Supplementary data to this article can be found online at https:// doi.org/10.1016/j.carres.2019.107861.
[22] [23]
References
[24]
[1] G.C. Daul, R.M. Reinhardt, J.D. Reid, The preparation of partially cyanoethylated cotton with acrylonitrile, Text. Res. J. 25 (3) (1995) 246–253. [2] A.K. Saha, B.C. Mitra, Studies on cyanoethylation of jute fiber, J. Appl. Polym. Sci. 62 (5) (1996) 733–742. [3] S.P.S. Yen, et al., Cyanoresin, Cyanoresin/cellulose Triacetate Blends for Thin Film, dielectric capacitors, US, 1996. [4] D.Q. Tan, High-temperature Capacitors and Methods of Making the Same, (2010) US. [5] S. Yasuhiro, Ein-Chip-Mikrocomputer, (1993). [6] H.Q. Wang, Z.Q. Shao, S.Y. Lv, New facile method for the hetergeneous synthesis of cyanoethyl cellulose- solvent Method,and comparing with traditional method, Adv. Mater. Res. 239–242 (2011) 2084–2090. [7] B.D. aniel, C.W. Lewis, Elektrisches Schaltgeraet, (1963). [8] J. Zhou, Q. Li, Y. Song, A facile method for the homogeneous synthesis of cyanoethyl cellulose in NaOH/urea aqueous solutions, Polym. Chem. 1 (10) (2010) 1662–1668. [9] D.S. Lee, A.S. Perlin, Use of methanolysis in the characterization of O -(2-hydroxypropyl)cellulose by 13 C-n.m.r. spectroscopy, Carbohydr. Res. 124 (1) (1983) 172–175. [10] Y. Tezuka, K. Imai, M. Oshima, T. Chiba, Determination of substituent distribution in cellulose ethers by 13 C- and 1 H-n.m.r. studies of their acetylated derivatives: O -(2-hydroxypropyl)cellulose, Carbohydr. Res. 196 (10) (1990) 1–10. [11] R.A. Gelman, Characterization of carboxymethylcellulose: distribution of substituent groups along the chain[J], J. Appl. Polym. Sci. 27 (8) (1982) 2957–2964. [12] B. Saake, S. Horner, T. Kruse, J. Puls, Detailed investigation on the molecular
[25] [26] [27] [28] [29] [30] [31] [32] [33]
8
structure of carboxymethyl cellulose with unusual substitution pattern by means of an enzyme-supported analysis, Macromol. Chem. Phys. 201 (15) (2000) 1996–2002. Y. Tezuka, Determination of substituent distribution in cellulose ethers by means of a 13 C nuclear magnetic resonance study on their acetylated derivatives: 3 Hydroxyethylcellulose, Polymer 30 (12) (1989) 2288–2291. P. Zadorecki, T. Hjertberg, M. Arwidsson, Characterization of cellulose ethers by 13C NMR, 2. Structural determination, Makromol. Chem. 188 (3) (2003) 513–525. Y. Tezuka, K. Imai, M. Oshima, Determination of substituent distribution in cellulose ethers by means of a 13C NMR study on their acetylated derivatives, 4. Omethyl-O-hydroxyalkylcelluloses, Macromol. Chem. Phys. 191 (3) (1990) 681–690. Y. Tezuka, K. Imai, M. Oshima, Determination of substituent distribution in cellulose ethers by means of a carbon-13 NMR study on their acetylated derivatives. 1. Methylcellulose, Macromolecules 20 (10) (1987) 2413–2418. J. Reuben, Analysis of the carbon-13 n.m.r. spectrum of methanolyzed O -ethylcellulose: monomer composition and models for its description, Carbohydr. Res. 161 (1) (1987) 23–30. S.J. Lee, Determination of the substituent distribution along cellulose acetate chains as revealed by enzymatic and chemical methods, Carbohydr. Polym. 54 (3) (2003) 353–362. E. Nakayama, J.I. Azuma, Substituent distribution of cyanoethyl cellulose, Cellulose 5 (3) (1998) 175–185. J. Zhou, A facile method for the homogeneous synthesis of cyanoethyl cellulose in NaOH/urea aqueous solutions, Polym. Chem. 1 (10) (2010) 1662–1668. H. Schagerlöf, M. Johansson, S. Richardson, Substituent distribution and clouding behavior of hydroxypropyl methyl cellulose analyzed using enzymatic degradation, Biomacromolecules 7 (12) (2006) 3474–3481. K. Hiroyuki, A. Hikaru, Ha Hisaho, S. Yuuichi, 13C-detection two-dimensional nmr approaches for cellulose derivatives, Cellulose 22 (5) (2015) 2927–2942. H. Kono, Chemical shift assignment of the complicated monomers comprising cellulose acetate by two-dimensional NMR spectroscopy, Carbohydr. Res. 375 (12) (2013) 136–144. T. Heinze, A. Koschell, Water-soluble 3-O-methoxyethyl cellulose: synthesis and characterization, Carbohydr. Res. 343 (4) (2008) 668–673. D. Fenn, T. Heinze, Novel 3-mono- O -hydroxyethyl cellulose: synthesis and structure characterization, Cellulose 16 (5) (2009) 853–861. A. Koschella, Water soluble 3-mono- O -ethyl cellulose: synthesis and characterization, Polym. Bull. 57 (1) (2006) 33–41. B.N. Batiabenaroya, K. Johnf, Synthesis and structural characterization of 3-Oethylene glycol functionalized cellulose derivatives, Carbohydr. Polym. 76 (1) (2009) 60–67. Zhang, Synthesis and characterization of low sulfoethylated cellulose, Carbohydr. Polym. 83 (2) (2011) 616–622. K. Hiroyuki, Two-dimensional NMR data of a series of methylcellulose with different degrees of substitution, Data in Brief 18 (2018) 1088–1098. K. Petzold-Welcke, M. Kötteritzsch, D. Fenn, A. Koschella, Study on synthesis and NMR characterization of 2,3‐O‐hydroxyethyl cellulose depending on synthesis conditions, Macromol. Symp. 294 (2) (2010) 133–140. Y. Tezuka, K. Imai, M. Oshima, Determination of substituent distribution in cellulose ethers by means of a 13C NMR study on their acetylated derivatives, 4. Omethyl- O-hydroxyalkylcelluloses, Macromol. Chem. Phys. 191 (3) (1990) 681–690. F. Dominik, H. Thomas, Novel 3-mono-O-hydroxyethyl cellulose: synthesis and structure characterization, Cellulose 16 (5) (2009) 853–861. Z.Q. Shao, Y.D. Zhang, F.F. Yang, Molecular design and synthesis of hydroxyalkyl cellulose ether nitrate as novel energetic adhesive, Chin. J. Energetic Mater. 16 (5) (2008) 609–613.