Effects of moisture on structure and electrophysical properties of a ferroelectric composite from nanoparticles of cellulose and triglycine sulfate

Effects of Moisture on Structure and Electrophysical Properties of a Ferroelectric Composite from Nanoparticles of Cellulose and Triglycine Sulfate CONDENSED MATTER Effects of Moisture on Structure an[.]

CONDENSED MATTER

Effects of Moisture on Structure and Electrophysical Properties

of a Ferroelectric Composite from Nanoparticles of Cellulose

and Triglycine Sulfate

Bich Dung Mai1

&Hoai Thuong Nguyen2,3

&Dinh Hien Ta4

Received: 6 November 2018 / Published online: 8 April 2019

# Sociedade Brasileira de Física 2019

Abstract

In this study, a novel ferroelectric composite consisting of triglycine sulfate and cellulose nanoparticles at different weight composition ratios was successfully synthesized A comparative study on structure and electrophysical properties for dried and wet composite samples was carried out The measurements of electrophysical parameters were performed from 10 to

120 °C under a weak electric field with an amplitude of 1 V cm−1at low and infra-low frequencies (10−3–103

Hz) under different relative humidities of 0, 30, 60, 80, and 100% The characterization results showed a significant impact of moisture on crystal-linity and features of functional groups in the composite Besides, phase transition temperature of the composite increased by 3 to

63 °C higher than those for single crystal of triglycine sulfate (+ 49 °C) in dependence on cellulose content in the composite Along with a significant increase in dielectric constant, dielectric loss, and dielectric dispersion in the composite due to high conductivity caused by moisture, the water molecules on sample surface led to the appearance of addition peaks in temperature dependences of dielectric constant and dielectric loss tangent in the initial stage of heating All the anomalies can be explained by the strong interaction through hydrogen bonds between triglycine sulfate and cellulose components as well as between these components and water molecules in the composite

Keywords Nanocomposites Ferroelectrics Humidity Phase transition Cellulose

1 Introduction

Recently, one of the most urgent global issues is related to the

ever-growing amount of electronic waste (e-waste) discharged

from out-of-order electronics devices with about 50 million

tons forecasted to reach by 2018 [1,2], causing serious

prob-lems for ecosystems and human health The reason is

associated with the fact that most of materials used to manu-facture electronics systems are originated from inorganic sub-stances, which are toxic to environment after service life In the context, the inspiration from nature has encouraged re-searchers to create biodegradable forms called as Bgreen electronics^; for that, natural abundant materials are preferred

to be used In this regard, cellulose with low cost, light weight, high electrical stability, and biodegradability is considered as a promising candidate [3,4] With these advantages, cellulose can be a perfect substrate for preparing transistors [3,5] and OLEDs [6], an ideal support material for photovoltaic cells [7] and the main material for processing of highly flexible, sus-tainable optoelectronic devices [2] However, to achieve the stable and optimal performance of cellulose in electronics de-vices, it is needed to overcome several shortcomings caused

by its high adsorption capacity towards moisture In this re-gard, understanding of effects of humidity on electrophysical properties of materials is extremely important for preparation

of cellulose-containing electrical and electronic equipment Among advanced electrical and electronics materials, fer-roelectric nanocomposites are promising for manufacturing

* Hoai Thuong Nguyen

nguyenhoaithuong@tdtu.edu.vn

1

Institute of Biotechnology and Food Technology, Industrial

University of Ho Chi Minh City, Ho Chi Minh City, Vietnam

2

Division of Computational Physics, Institute for Computational

Science, Ton Duc Thang University, Ho Chi Minh City, Vietnam

3 Faculty of Electrical & Electronics Engineering, Ton Duc Thang

University, Ho Chi Minh City, Vietnam

4

Faculty of Electrical and Electronics Engineering Technology, Ho

Chi Minh City University of Food Industry, Ho Chi Minh

City, Vietnam

https://doi.org/10.1007/s13538-019-00658-5

modern electronics devices with several beneficial properties

thanks to the size effects of ferroelectric fillers at nanoscale

level [8,9] In the role of nanosized ferroelectric fillers,

t r i g l y c i n e s u l f a t e ( T G S ) w i t h c h e m i c a l f o r m u l a

(NH2CH2COOH)3·(H2SO4) is one of the most popular

mate-rials and has been used to synthesize state-of-the-art

ferroelec-tric nanocomposites [4,10–12] because of its valuable

ferro-electric properties and high wettability Because TGS is a

hydrogen-containing ferroelectric, the presence of moisture

may lead to the appearance of several effects such as the

change of pyroelectric coefficients and coercive fields, the

higher sensitivity of dielectric relaxation towards frequencies

of measuring external fields, and the significant increase of

dielectric permittivity and conductivity [13,14] The influence

of humidity on properties of nanocomposites from porous

cellulose and triglycine sulfate was already investigated [15]

However, a huge drawback of porous nanocomposite is

relat-ed to difficulties in controlling the amount of inclusions

em-bedded into nanopores Besides, in the study [15], the impact

of moisture on structure of cellulose-containing composite

was not reported yet In this regard, the present study is

de-voted to clarifying the effects of moisture on structure and

electrophysical properties of a novel composite prepared from

cellulose nanoparticles (CNP) and TGS The composite was

prepared by precisely controlling the composition weight of

CNP and TGS Then, various techniques as particle size

anal-ysis, scanning electron microscopy (SEM), X-ray powder

dif-fraction (XRD), and Fourier-transform infrared spectroscopy

(FTIR) were utilized to investigate the change of composite

structure under different relative humidities The

electrophysical properties of the composite were tested under

a weak electric field in temperature range of 10–120 °C at low

and infralow frequencies (10−3–103

Hz)

2 Samples and Experiments

Triglycine sulfate used to prepare composite samples was

pur-chased from Merck supplier in the form of reagent-grade

chemicals that can be utilized for preparation process without

further purification Nanoparticles of cellulose were obtained

from waste cotton using the method of enzymatic hydrolysis,

sonification treatment, and free drying according to

proce-dures described in [16] The scheme for preparation of the

composite from CNP and TGS is shown in Fig.1 Firstly, a

determined amount taken from each of TGS powder and CNP

was mixed using a magnetic stirrer in distilled water at 40 °C

The weight of CNP and TGS was chosen at different

compo-sition weight ratios of xCNP + (1− x)TGS with x changed

from 0.2 to 1 The stirring was kept for 1 h at 40 °C in a sealed

container and then in open air to obtain a solid mixture, which

was separated, heated at 120 °C for 2 h to remove residual

water, cooled down at room temperature, and crushed in

mortar to get the mixture well mixed The obtained mixture was compressed into tablets of 6 mm in diameter and 1 mm thick under a pressure of 20 MPa For testing the influence of humidity on electrophysical properties of the composite, the synthesized samples were divided into groups and stored in different containers under different relative humidity (RH) of

0, 30, 60, 80, and 100% for 24 h at room temperature For dielectric measurements, silver leaf electrodes were applied on the prepared samples using a conductive glue

The size of nanoparticles and their morphology were ex-amined by a Zetasizer analyzer with a detection range of 0.3 nm–10 μm and a FE-SEM S4800 HITACHI scanning electron microscope with an accelerating voltage of 10.0 kV, respectively The structure of dried and wet samples was

test-ed using a Rigaku Ultima IV X-ray diffractometer (Japan) using Cu Kα radiation (λ = 1.5406 Å) at a voltage of 40 kV and current of 30 mA in the scanning range 2θ of 5–70° with a step size of 0.02° The information of functional groups in the synthesized samples were analyzed by Fourier transform in-frared spectroscopy on a Bruker Tensor 37 spectrophotometer (USA) at a scanning range over 400–4000 cm−1with resolu-tion of 2 cm−1 The phase transition and dielectric relaxation

of the composite were measured by a model GW Instek

LCR-821 meter at 1 kHz and by Solartron 1260 impedance analyzer connected to an expanded model of Solartron 1296 dielectric interface at frequencies from 1 mHz to 1 kHz, respectively The study on electrophysical properties was conducted under

a weak electric field with an amplitude of 1 V cm−1

3 Experimental Results and Discussion

3.1 Materials Characterization and Influence of Water

The characterization results for size distribution and morphol-ogy of cellulose particles are presented in Fig.2 We can see that the size of cellulose particles was formed in near-spherical shape with size ranged in 40–80 nm The similar results were reported in several studies [17,18]

The XRD patterns and FTIR spectra for pure TGS, dried CNP, and composite samples of 0.5 CNP + 0.5 TGS stored at different RH of 0, 30, 60, 80, and 100% are shown in Figs.3

and4, respectively It is seen in Fig.3that the detected peaks for pure TGS were in good agreement with the ICPDS XRD data (00-015-0947), while for dried samples of CNP, three strong peaks of Iβ crystalline phase at 14.7° (101), 16.3°

101
, and 22.5° (002) were clearly revealed [19–21], indi-cating the formation of crystalline structure in cellulose Correspondingly, several characteristic adsorption peaks in FTIR spectra for dried CNP and pure TGS were also ob-served For examples, for TGS, a broad band detected in the range of 3300–2800 cm−1corresponds to the asymmetric and

symmetric N–H (NH3), C–H (CH2), and O–H (COOH)

stretching [22] Besides, the two small adsorption peaks at

1706 and 1621 cm−1could refer to the stretching of C=O

bonds and symmetric stretching modes of COOˉ groups,

re-spectively In addition, several peaks observed from 1128 to

909 cm−1can be assigned for SO4 −of sulfate groups In the

case of dried CNP samples, a broad band centered at

3273 cm−1is related to O–H stretching and the flexural

vibra-tion of intra- and intermolecular hydrogen bonds of cellulose

[23] Besides, two peaks detected at 2900 and 2840 cm−1are

originated from CH2asymmetric vibrations, while the peaks

at 1433 and 1378 cm−1are related to the OCH in-plane and

CH deformation vibrations, respectively [16, 24]

Additionally, the strong peaks observed at 1635 and

707 cm−1correspond to OH stretching and OH out of plane

bending [17,25] Finally, the amorphous region which always

exists in any cellulose structures was also detected at

899 cm−1, corresponding to the COC, CCO, and CCH defor-mation modes as well as stretching vibrations of the C5 and C6 atoms [16] The above obtained results indicated that the reagents for preparation of the composite are adequate Before analyzing the influence of moisture on structure of the synthesized composite, it is needed to characterize the obtained composite samples in dry form to confirm the suc-cessful synthesis For dried composite samples after heat treat-ment, their XRD pattern (Fig.3) and FTIR spectrum (Fig.4) contain almost all characteristic peaks for CNP and TGS com-ponents as described above However, several anomalies were detected For examples, the intensity of XRD peaks at 14.7° (101), 16.3° 101

, and 22.5° (002) characteristic for cellulose was lower than those for CNP without TGS, i.e., the crystal-linity of cellulose decreased after becoming a component of the composite The similar phenomenon was also reported for the composite from porous nanocellulose and TGS [4] In

Fig 1 Scheme for preparation of

xCNP + (1 − x)TGS composite

Fig 2 Particle size distribution (a) and SEM image (b) for cellulose suspension

addition, in FTIR pattern, the broad band at wavelength higher

than 2800 cm−1is broadened as compared to those of TGS and

CNP The reason for that might be related to the strong

ad-sorption of functional groups characteristic for both TGS (the

asymmetric and symmetric NH3, CH2, COOH stretching) and

CNP (O–H stretching and hydrogen bonds) as mentioned

above Moreover, the combination of TGS and CNP could

lead to the formation of new hydrogen bonds in the composite

due to the presence of OHˉ hydroxyl groups in cellulose and

TGS As known [16,24,26] that the change in the number and

strength of hydrogen bonds brings the change in intensity and

the width of the related bands As a result, it is worth to

as-sume that the expansion of this adsorption band is mainly due

to the increase in number of hydrogen bonds in the composite

Along with the expansion of the mentioned band, a slight deviation of peak position for the composite from that of its components and overlapping of peaks were also detected For instance, the peaks at 1433, 1164, and 707 cm−1 in FTIR spectrum of CNP were shifted to 1424, 1164, and 708 in the dried composite, or the strong peak at 1631 cm−1is a result of overlapping the two peaks at 1621 and 1706 cm−1of TGS and

1635 cm−1of CNP Overall, the obtained results suggested that the mixed composite consisting of cellulose nanoparticles and triglycine sulfate was successfully synthesized

The presence of moisture in the synthesized composite led

to the appearance of several anomalies in XRD patterns (Fig

3) and FTIR spectra (Fig.4) Firstly, the intensity of charac-teristic peaks for cellulose at 14.7° (101), 16.3° 101

, and 22.5° (002) for the composite increased with increasing RH, while the intensity of peaks for TGS component almost

Fig 3 XRD patterns for dried samples of TGS and cellulose

nanoparticles and for composite 0.5 CNP + 0.5 TGS under different

conditions of relative humidity

Fig 4 FTIR spectra for dried samples of TGS and cellulose nanoparticles and for composite 0.5 CNP + 0.5 TGS under different conditions of relative humidity

remained unchanged and the shift of position of XRD peaks

for CNP as well as TGS was not detected (Fig.3) Secondly,

the adsorption peaks at 1631 and 1151 cm−1corresponding to

OH stretching of cellulose and SO4 −of sulfate groups of TGS

were shifted to lower wavelengths with increasing RH At the

same time, the adsorption intensity of these peaks increased

In addition, the broad band of 2700–3800 cm−1centered at

3340 cm−1 and small peaks at 1547 and 707 cm−1became

deeper under the influence of moisture All the obtained

changes are obviously related to the interaction between

func-tional groups of the composite and adsorbed water molecules

3.2 Electrophysical Properties and Influence

of Moisture

The results for investigation of phase transition in xCNP + (1

− x)TGS composite in dry form and at different composition

weight ratios with x varied from 0.2 to 1 are presented in

Fig.5a, b Herein, the temperature dependences of dielectric

constantɛ′(T) and dielectric loss tangent tgδ(T) were

mea-sured at 1 kHz According to the results for all composition

weight ratios, the phase transition peaks were strongly

smeared as always found in most of ferroelectric composites [4,10,27–30], and the phase transition point was shifted to the region of higher temperatures as compared to those of single crystal TGS (+ 49 °C) Furthermore, the rise in the content of cellulose in the composite led to increasing of phase transition temperatures (Table1), while reducing the maximum of di-electric constant In addition, for samples with ratios higher than 70% (x < 0.7) of TGS mass content (curves 1, 2 in Fig

5a), there was an additional peak observed at higher tempera-tures as compared to those for the lower-temperature one This anomaly was probably associated with the structural changes

of TGS nanocrystals in the composite [31] This was also observed for the mixed composite of SiO2and TGS [10] It should be noted that in the case of matrix composite prepared from TGS embedded into nanochannels of porous cellulose [4], there was only one phase transition point detected under the same heating process and applied electric field In the case

of CNP without TGS (x = 1), the values of dielectric constant are quite small (less than 10) in the entire temperature range (Fig.5a) The temperature dependences of tgδ(T) have a sim-ilar shape as forɛ′(T) with the values of dielectric loss tangent not higher than 0.5 (Fig.5b)

Fig 5 Temperature dependences of dielectric constant (a, c) and

dielectric loss tangent (b, d) for dried (a, b) and wet (RH = 80%) (c, d)

samples of the composite at different composition weight ratios at 1 kHz.

The inset shows the change of dielectric constant and dielectric loss tangent at the initial stage of heating

After storing the composite samples under RH = 80% for

24 h, both dielectric constant and dielectric loss tangent

sig-nificantly increased and the transition peaks became more

smeared as compared to that of dried ones (Fig.5c, d) Even

so, the shift of phase transition in the composite at all

compo-sition weight ratios was not detected However, the second

peaks ofɛ′(T) around 100–110 °C for samples containing

TGS content higher than 70% as mentioned above

disap-peared Instead, there was only a rising tendency detected

for dielectric constant with increasing temperature in this

tem-perature range Besides, in the initial stage of heating process

for wet composite samples, an addition peak in ɛ′(T) and

tgδ(T) appeared in the range of 10–20 °C (insets in Fig

5c, d) It should be noticed that these peaks can be removed

by heat treatment This anomaly has been reported in previous

studies for various ferroelectric composites from TGS [14,

15] The similar character ofɛ′(T) and tgδ(T) for the

compos-ite was also observed under other RH

To clarify the effects of humidity on dielectric relaxation

for the synthesized composite, frequency dependences of the

realɛ′( f ) and imaginary ɛ″( f ) parts of the complex dielectric

permittivityɛ*( f ) = ɛ′( f ) + i ɛ″( f ) under different RHs of 0

(dried samples), 30, 60, 80, and 100% at low and infralow

frequencies of 10−3–103

Hz at room temperature were inves-tigated The samples of 0.5 CNP + 0.5 TGS were chosen for

these experiments The results are shown in Fig.6a, b It is

seen in Fig.6a, bthat the higher the moisture content is, the

higher the values ofɛ′ and ɛ″ are Moreover, the dependences

ofɛ′ and ɛ″ on frequency obey the universal power-law of

dielectric relaxation (ɛ′, ɛ″ ~ f−n), where n varied between 0 to

1 [32,33] In our case, the values of n tend to increase from

0.51 to 0.92 with increasing RH from 0 to 100% It can be

considered as an evidence of increased conductivity in the wet

composite samples Indeed, the AC conductivity can be

cal-culated by the following formula:

where f—electric field frequency, ɛo—vacuum dielectric

con-stant, andɛ″—the imaginary part of complex dielectric

per-mittivity The obtained results are shown in Fig.6c As seen,

the values of conductivity increase drastically with increasing

frequency at low RH of 0 and 30% At higher RH, plateau

regions were observed in the plots and expanded with increas-ing RH, indicatincreas-ing the presence of static conductivity at infralow frequencies

3.3 Discussion

The anomalies of characterization results could be explained

as follows Firstly, the increase in crystallinity level of cellu-lose component in the wet composite can be referred to the fact that the presence of moisture helps to recover partial col-lapse of cellulose structure in dry samples [34] It means that the dehydration process for cellulose can lead to partially decrystallizing cellulose structure and increasing strains as well as twist deformation [34] Besides, the water molecules introduced into the composite samples as analyzed above can interact with functional groups through hydrogen bonds, resulting in changes of adsorption intensity and the shift of adsorption peaks

The shift of phase transition point in the synthesized com-posite towards higher temperatures as compared to those of single crystals TGS can be explained by the fixation of polar-ization in TGS inclusion due to the strong interaction between TGS and cellulose nanoparticles through hydrogen bonds The interaction could become more stronger with increasing the cellulose content in the composite because of more effec-tive isolation of TGS particles surrounded by cellulose In addition to that when TGS particles were isolated by cellulose, they could interact with each other through dipole–dipole mechanism Based on the Landau–Ginzburg–Devonshire the-ory [34], the phase transition temperature Toof a heteroge-neous system from bounded particles can be determined by the following formula:

To¼ Tc−α1

o∑

i

p*i

!

E*i

ƒ!

ð2Þ

where Tois Curie point of the bulk of ferroelectric particles,αo

is a positive constant, p!*i

is intrinsic dipole moment of a fer-roelectric particle, Eƒ!*i

is an effective field acting on ith dipole from nearest neighbors, and the term of∑

i

p*i

!

E*i

ƒ!

is the energy

of dipole–dipole interaction It is obviously seen in formula (2) that if the energy of dipole–dipole interaction is negative, i.e., the dipoles in TGS particles were oriented by the way so that their fields can be compensated by each other, the phase transition temperature increases Thus, it allows us to assume that the dipole–dipole interaction could be another reason and contribute to the above shift of phase transition point in the composite

The presence of additional peaks inɛ′(T) and tgδ(T) around 10–20 °C in the initial stage of heating process for the wet composite samples could be related to the evaporation of

Table 1 The increase in values of phase transition temperature in

composite samples at different composition weight ratios as compared

to those of TGS single crystals

xCNP + (1 − x)TGS

ΔT = T o − T c , T o —phase transition temperature of composite, T c = +

49 °C —the Curie point of TGS single crystals

water molecules from the sample surface, on which water

molecules were loosely connected to the sample and therefore

easily to be thrown out The similar behavior has been

report-ed in previous studies for the composites basreport-ed on triglycine

sulfate [14,15]

The strong dielectric dispersion in the composite at 10−3–

103Hz under influence of moisture was obviously associated

with the presence of conductivity according to the Maxwell–

Wagner–Sillars interfacial polarization mechanism This

be-havior was also observed in matrix composite from porous

nanocellulose and TGS [33]

4 Conclusion

The obtained results confirmed the successful synthesis of

the composite consisting of triglycine sulfate and

nanocellulose particles with transition temperatures higher

than those for single crystals TGS by 3 to 63 °C

depend-ing on the cellulose content in the composite The

in-crease in phase transition is explained by the fixation of

polarization in TGS crystals due to strong interaction be-tween TGS and cellulose nanoparticles through hydrogen bonds Besides, the phase transition temperature could be increased by increasing cellulose content due to

effective-ly isolating TGS particles covered by cellulose In this case, the dipole–dipole interactions between TGS parti-cles cannot be negligible The presence of moisture strongly affected the structure of composite as the change

of crystallinity level of cellulose component and the change of intensity or adsorption position for several functional groups In addition to that, a significant in-crease in dielectric constant, dielectric loss, and dielectric dispersion in the composite was detected because of high conductivity Interestingly, the water molecules on sample surface led to the appearance of addition peaks in temper-ature dependences of dielectric constant and dielectric loss tangent The low and infra-low dispersion of dielec-tric permittivity was governed by Maxwell–Wagner– Sillars interfacial mechanism Overall, the obtained results could be useful in controlling dielectric properties of fer-roelectric materials for practical applications

Fig 6 Frequency dependences of real (a) and imaginary (b) parts of dielectric permittivity and of conductivity (c) for 0.5 CNP + 0.5 TGS composite under different conditions of relative humidity in the range of 10−3–10 3 Hz at room temperature

1 C.P Baldé, F Wang, R Kuehr, J Huisman, The Global E-Waste

Monitor – 2014, (United Nations University, IAS – SCYCLE,

Bonn, Germany, 2015)

2 I.-V Mihai, E.D Glowacki, N.S Sariciftci, S Bauer, Green

Materials for Electronics, (Wiley-VCH Verlag GmbH & Co.

KGaA, 2018) Print ISBN:9783527338658 |Online ISBN:

9783527692958 | https://doi.org/10.1002/9783527692958 , https://

onlinelibrary.wiley.com/doi/book/10.1002/9783527692958

3 B Peng, P.K.L Chan, Flexible organic transistors on standard

print-ing paper and memory properties induced by floated gate electrode.

Org Electron 15, 203–210 (2014) https://doi.org/10.1016/j.orgel.

2013.11.006

4 N.H Thu’o’ng, A.S Sidorkin, S.D Milovidova, Dispersion of

di-electric permittivity in a nanocrystalline cellulose–triglycine sulfate

composite at low and ultralow frequencies Phys Solid State 60,

559–565 (2018) https://doi.org/10.1134/S1063783418030320

5 S Thiemann, S.J Sachnov, F Pettersson, R Bollström, R.

Österbacka, P Wasserscheid, J Zaumseil, Cellulose-based ionogels

for paper electronics Adv Funct Mater 24, 625–634 (2014).

https://doi.org/10.1002/adfm.201302026

6 E.F Gomez, A.J Steckl, Improved performance of OLEDs on

cellulose/epoxy substrate using adenine as a hole injection layer.

ACS Photonics 2, 439–445 (2015) https://doi.org/10.1021/

ph500481c

7 M.C Barr, J.A Rowehl, R.R Lunt, J Xu, A Wang, C.M Boyce,

S.G Im, V Bulovi ć, K.K Gleason, Paper Adv Mater 23, 3500–

3505 (2011) https://doi.org/10.1002/adma.201101263

8 S Li, J.A Eastman, Z Li, C.M Foster, R.E Newnham, Size effects

in nanostructured ferroelectrics Phys Let A 212, 341–346 (1996).

https://doi.org/10.1016/0375-9601(96)00077-1

9 M.F Vladimir, Phys.-Usp 49, 193 (2006) https://doi.org/10.1070/

PU2006v049n02ABEH005840

10 S.D Milovidova, A.S Sidorkin, O.V Rogazinskaya, E.V.

Vorotnikov, Dielectric properties of the mixed nanocomposites:

triglycine sulfate - silica Ferroelectrics 497, 69–73 (2016).

https://doi.org/10.1080/00150193.2016.1162620

11 Y Yang, H.L.W Chan, C.L Choy, Properties of triglycine sulfate/

poly(vinylidene fluoride-trifluoroethylene) 0 –3 composites.

Frontiers of Ferroelectricity ((Springer, Boston, 2006)

12 A Plyushch, J Macutkevic, V Samulionis, J Banys, D Bychanok,

P Kuzhir, S Mathieu, V Fierro, A Celzard, Polym Compos.

(2018) https://doi.org/10.1002/pc.24932

13 V.E Khutorsky, S.B Lang, Very strong influence of moisture on

pyroelectric and dielectric properties of triglycine sulfate-gelatin

films J Appl Phys 82, 1288–1292 (1997) https://doi.org/10.

1063/1.365900

14 O.M Golitsyna, S.N Drozhdin, A.E Gridnev, Influence of the

humidity on dielectric characteristics of porous aluminum oxide

with inclusions of triglycine sulfate Phys Solid State 54, 1961–

1965 (2012) https://doi.org/10.1134/S1063783412100149

15 H.T Nguyen, A.S Sidorkin, S.D Milovidova, O.V Rogazinskaya,

Influence of humidity on dielectric properties of nanocrystalline

cellulose – triglycine sulfate composites Ferroelectrics 501, 180–

186 (2016) https://doi.org/10.1080/00150193.2016.1204866

16 T Fattahi Meyabadi, F Dadashian, G Mir Mohamad Sadeghi, H.

Ebrahimi Zanjani Asl, Spherical cellulose nanoparticles preparation

from waste cotton using a green method Powder Technol 261,

232 –240 (2014) https://doi.org/10.1016/j.powtec.2014.04.039

17 P Lu, Y.-L Hsieh, Preparation and properties of cellulose

nanocrystals: rods, spheres, and network Carbohydr Polym 82,

329–336 (2010) https://doi.org/10.1016/j.carbpol.2010.04.073

18 J Zhang, T.J Elder, Y Pu, A.J Ragauskas, Facile synthesis of spherical cellulose nanoparticles Carbohydr Polym 69, 607–611 (2007) https://doi.org/10.1016/j.carbpol.2007.01.019

19 P.B Filson, B.E Dawson-Andoh, D Schwegler-Berry, Enzymatic-mediated production of cellulose nanocrystals from recycled pulp Green Chem 11, 1808 (2009) https://doi.org/10.1039/B915746H

20 S.-S Wong, S Kasapis, Y.M Tan, Bacterial and plant cellulose modification using ultrasound irradiation Carbohydr Polym 77,

280 –287 (2009) https://doi.org/10.1016/j.carbpol.2008.12.038

21 P Satyamurthy, P Jain, R.H Balasubramanya, N Vigneshwaran, Preparation and characterization of cellulose nanowhiskers from cotton fibres by controlled microbial hydrolysis Carbohydr Polym 83, 122–129 (2011) https://doi.org/10.1016/j.carbpol 2010.07.029

22 N Sinha, S Bhandari, H Yadav, G Ray, S Godara, N Tyagi, J Dalal, S Kumar, B Kumar, Performance of crystal violet doped triglycine sulfate single crystals for optical and communication ap-plications Cryst Eng Comm 17, 5757–5767 (2015) https://doi org/10.1039/C5CE00703H

23 Y Cao, H Tan, Structural characterization of cellulose with enzy-matic treatment J Mol Struct 705, 189–193 (2004) https://doi org/10.1016/j.molstruc.2004.07.010

24 S.Y Oh, D.I Yoo, Y Shin, G Seo, FTIR analysis of cellulose treated with sodium hydroxide and carbon dioxide Carbohydr Res 340, 417–428 (2005) https://doi.org/10.1016/j.carres.2004 11.027

25 L Wang, Y Zhang, P Gao, D Shi, H Liu, H Gao, Changes in the structural properties and rate of hydrolysis of cotton fibers during extended enzymatic hydrolysis Biotechnol Bioeng 93, 443–456 (2006) https://doi.org/10.1002/bit.20730

26 H Zhao, J.H Kwak, Z.C Zhang, H.M Brown, B.W Arey, J.E Holladay, Studying cellulose fiber structure by SEM, XRD, NMR and acid hydrolysis Carbohydr Polym 68, 235–241 (2007).

https://doi.org/10.1016/j.carbpol.2006.12.013

27 S.V Baryshnikov, A.Y Milinskiy, E.V Charnaya, A.S Bugaev, M.I Samoylovich, Dielectric studies of ferroelectric NH4HSO4 nanoparticles embedded into porous matrices Ferroelectrics 493, 85–92 (2016) https://doi.org/10.1080/00150193.2016.1134174

28 S.D Milovidova, O.V Rogazinskaya, A.S Sidorkin, H.T Nguyen, E.V Grohotova, N.G Popravko, Dielectric properties of compos-ites based on nanocrystalline cellulose with triglycine sulfate Ferroelectcrics 469, 116–506 (2014) https://doi.org/10.1134/ S1063783415030178

29 A.Y Milinskii, S.V Baryshnikov, H.T Nguyen, Dielectric proper-ties of nanocomposites based on potassium iodate with porous nanocrystalline cellulose Ferroelectrics 524, 181–188 (2018).

https://doi.org/10.1080/00150193.2018.1432830

30 H.T Nguyen, A.S Sidorkin, S.D Milovidova, O.V Rogazinskaya, Investigation of dielectric relaxation in ferroelectric composite nanocrystalline cellulose – triglycine sulfate Ferroelectrics 498,

27 –35 (2016) https://doi.org/10.1080/00150193.2016.1166835

31 T.R Volk, S.V Mednikov, L.A Shuvalov, Unipolarity of Tgs-crystals induced in paraelectric phase Ferroelectrics 47, 15–23 (1983) https://doi.org/10.1080/00150198308227816

32 A.K Jonscher, The ‘universal’ dielectric response Nature 267,

673 –679 (1977) https://doi.org/10.1038/267673a0

33 H.T Nguyen, A.S Sidorkin, S.D Milovidova, O.V Rogazinskaya, Electrophysical properties of matrix composites nanocrystalline cellulose – triglycine sulfate Ferroelectrics 512, 71–76 (2017).

https://doi.org/10.1080/00150193.2017.1349900

34 V.L Ginzburg, Theory of ferroelectric phenomena Usp Fiziol Nauk 38, 390 (1949)

Publisher ’s Note Springer Nature remains neutral with regard to juris-dictional claims in published maps and institutional affiliations.