See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/322098251

Vapor and pressure sensors based on cellulose nanofibers and carbon nanotubes
aerogel with thermoelectric properties

Article in JOURNAL OF RENEWABLE MATERIALS · January 2018

DOI: 10.7569/JRM.2017.634182

CITATIONS READS
26 1,681

6 authors, including:

Rajendran Muthuraj Abhishek Sachan

Scott Bader Company Ltd European University of Brittany
42 PUBLICATIONS 2,200 CITATIONS 12 PUBLICATIONS 205 CITATIONS

SEE PROFILE SEE PROFILE

Mickael Castro Jean-Francois FELLER

Université Bretagne Sud Université Bretagne Sud (UBS)
180 PUBLICATIONS 5,285 CITATIONS 238 PUBLICATIONS 5,093 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Rajendran Muthuraj on 02 January 2018.

The user has requested enhancement of the downloaded file.

 Scrivener First

1
2
Vapor and Pressure Sensors Based on Cellulose Nanofibers
3 and Carbon Nanotubes Aerogel with Thermoelectric
4
5 Properties
6
7 Rajendran Muthuraj, Abhishek Sachan, Mickael Castro*, Jean-François Feller, Bastien Seantier* and Yves Grohens
8
Dupuy de Lôme Research Institute (IRDL), FRE CNRS 3744, University of Southern Brittany, Rue de Saint Maudé, 56321 Lorient
9
Cedex, France
10
11 Received June 20, 2017; Accepted November 14, 2017
12
13
14 ABSTRACT: 
In this work, thermally insulating and electrically conductive aerogels were prepared from cellulose
15 nanofibers (CNF) and carbon nanotubes (CNTs) by environmentally friendly freeze-drying process. The
16 thermal conductivity of neat CNF aerogel is 24 mW/(m·K) with a density of 0.025 g/cm3. With the addition of
17 CNTs into CNF aerogel, the electrical conductivity was significantly increased while the thermal conductivity
18 was increased to 38 mW/(m·K). Due to these interesting properties, the Seebeck coefficient and the figure of
19 merit (ZT) of the CNF/CNTs aerogels were measured and showed that CNF/CNTs aerogel thermoelectric
properties can be improved. The compressibility and electrical resistance of the CNF/CNTs aerogel
20
highlighted its pressure-responsive property. A set of volatile organic compounds (VOCs) were exposed to
21
aerogels to monitor the resistance change. The CNF/CNTs aerogel showed high sensitivity and good response
22 to both nonpolar and polar vapors due to the absorption by both CNF and CNTs networks. The prepared
23 CNF/CNTs aerogel is therefore a good candidate for thermal insulation, thermoelectric material, VOCs
24 sensing, and pressure-sensing applications.
25
26 KEYWORDS: CNF/CNTs aerogel, thermal conductivity, pressure sensing, thermoelectric materials, VOCs sensing
27
28
29
30 1 INTRODUCTION the conduction through the materials and the radiation
31 of electromagnetic waves through the material. In gen-
32 Aerogels are lightweight and highly porous with a eral, the convection has a negligible effect as the pores
33 nanometer-scale porosity and with high specific sur- are very small [18]. Concerning the conduction, two
34 face area. Therefore, aerogels have drawn increasing heat transfers can be observed. On one hand, there is
35 attention for use in various fields [1–3], including ther- the heat transfer due to the conduction via phonons in
36 mal insulation [4–6], capacitors [7, 8], fire retardant the solid phase (later referred to as solid conduction).
37 [9], and sensor applications [10–12]. Various inorganic On the other hand, the heat transfer can be due to the
38 and organic materials have been used to produce aero- collision between gas molecules (referred to as gas con-
39 gels either by freeze-drying or by supercritical drying duction) [19]. Finally, as the material is semi-transparent
40 [13, 14]. Among the organic aerogels, cellulose-based to electromagentic waves, some heat is transferred via
41 aerogels are attractive materials because of their good these radiations [16]. It will later be referred to as “radi-
42 mechanical properties, sustainability, and renewability ation” contribution. In the case of aerogels, the pores
43 [15]. However, the effective thermal conductivity of are quite large. Therefore, the material has a huge effect
44 cellulose aerogels (0.029–0.030 W/(m·K)) is relatively on gas conduction and radiation. That is the reason
45 high. This can be explained by the parallel flux model why they usually have relatively high effective ther-
46 [16, 17]. In fact, the measured thermal conductivity, mal conductivity. Recently, CO2 supercritical drying
47 called effective thermal conductivity, is the combina- method allowed addressing this drawback. For exam-
48 tion of three contributions: the convection in the pores, ple, the thermal conductivity of amorphous cellulose
49 aerogels prepared by this method reached lower values
50 than that of air. This is the reason why these materials
*Corresponding author: bastien.seantier@univ-ubs.fr; mickael.
51 are called thermal superinsulating material [20]. Lately,
castro@univ-ubs.fr
52 Kobayashi et al. [5] reported the lowest thermal conduc-
53 tivity for cellulose nanofiber (CNF) aerogels prepared

J. Renew. Mater. © 2017 Scrivener Publishing LLC   1

Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

by supercritical drying (0.020 W/(m·K)). CNF aerogels conductivity, electrical conductivity, thermoelectric 1
produced by freeze-drying are also promising sustain- property, VOCs sensing, and pressure-sensing prop- 2
able materials for thermal insulation because of their erties were investigated. 3
low thermal conductivity (0.028 W/(m·K)) [21]. With 4
the aim of reducing aerogel thermal conductivity, it is 5
possible to use thermal opacifiers. These particles are 2 MATERIALS AND METHODS 6
able to absorb photons and to reduce the material radia- 7
tions contribution of the effective thermal conductivity. 2.1 Materials 8
Carbonaceous materials have already been deposited A viscous 1 wt% CNF water suspension with 1.3 9
on silica aerogels as opacifiers [22]. Therefore, carbon mmol/g surface charges content was procured from 10
nanotubes (CNTs) addition in CNF aerogel could play EMPA (Dübendorf, Switzerland). It was prepared via 11
a role to decrease the thermal conductivity of resulting TEMPO-oxidation pretreatment of wood pulp [35]. 12
aerogels [23]. It can also make the aerogels more electri- In order to achieve a required concentration of CNF, 13
cally conductive and create thermoelectric (TE) materi- osmotic concentration method was used to concentrate 14
als [24]. Until now, inorganic compounds (e.g., Bi2Te3 the CNF dispersion [36]. The diameter and length of the 15
or bismuth telluride-based alloys) have been used as CNF are 3.85 ± 0.98 nm and 450 nm, respectively [36]. 16
good TE materials at room temperature. However, Commercially available aqueous dispersion of unfunc- 17
their limited availability and complex fabrication tionalized multiwall carbon nanotubes (MWCNTs) 18
process limit their use for industrialization [25, 26]. were purchased from Nanocyl. The concentration of 19
Organic aerogels with electrically conductive and ther- MWCNT in the aqueous dispersion was 1 mg/100 µl. 20
mally insulative properties can provide a new route to Hereafter, MWCNTs will be referred to as CNTs. 21
produce TE materials. Very few researchers have stud- 22
ied aerogel-based material for TE applications. Gordon 23
et al. [27] proposed intrinsically electrically conductive 2.2 Aerogel Preparation 24
poly(3,4-ethylenedioxythiophene):poly(4-styrenesul- 25
A previous study [36] showed that the aerogel pre-
fonate) (PEDOT:PSS) aerogels as potential TE materi- 26
pared from 2 wt% CNF suspension has better thermal
als. Graphene and multiwall carbon nanotube-based 27
insulation property. Therefore, in this study we have
aerogels have also been studied for such application 28
selected 2 wt% CNF suspension to prepare aerogels.
[28]. These studies concluded that the performance of 29
Carbon nanotubes (CNTs) were used as thermal opaci-
the resulting aerogels is not suitable for practical TE 30
fier in CNF aerogels. Different concentrations of CNTs
application. 31
related to CNF (1, 5, 10, 15, 20, and 25 wt%) were added
With unique features such as porous structure, 32
in CNF suspension to prepare hydrogels. After incor-
electrical conductivity and high specific surface area, 33
poration of CNTs into CNF suspension, the mixture
aerogels can be ideal candidates for gas- and pres- 34
was sonicated in a Branson 3510 device (100 W, 40 kHz)
sure-sensing applications [29–31]. Cellulose/CNTs 35
for 30 min at 25 °C. The sonicated formulations were
aerogel has been used as volatile organic compound 36
poured into an aluminum mold (2 × 2 × 0.39 cm). They
(VOCs) sensors [10]. Noremberg et al. [32] recently 37
have been designed to fit the device used for the ther-
attempted cellulose/multiwall CNTs-based com- 38
mal conductivity measurement having a volume of
posite film for gas sensors. However, combining the 39
1.578 cm3. All these samples were frozen at –80 °C for
CNF aerogels with the CNTs is not easy. Therefore, 40
24 h. Subsequently, a freeze-drying process was per-
Qi et al. [33] dispersed CNTs in cellulose dissolved in 41
formed in a Christ Alpha 1–2 LD Plus freeze dryer at
aqueous alkaline-urea solution to prepare CNT/cel- 42
–55 oC with 0.1 mbar pressure. After 48 h, the aerogels
lulose aerogel. The electrical resistance of the CNF/ 43
are formed. Similary, neat CNTs and neat CNF aerogels
CNTs aerogel was reduced under pressure because 44
were prepared to compare with CNF/CNTs aerogels.
of the nanotubes percolation [34]. Moreover, they 45
At least four samples were prepared for each formu-
showed that the electrical conductivity of CNF/ 46
lation. The produced monolithic aerogels are repre-
CNTs aerogels is increased with increasing CNTs con- 47
sented in Figure 1. They are lightweight (Figure 1a) and
tent. The maximum electrical conductivity (2 × 10–3 48
mechanically stable (Figure 1b). The morphology will
S/cm) was observed with 25 wt% CNTs [12]. These 49
be described more precisely in the coming sections.
studies showed the feasibility of developing cellu- 50
lose aerogel-based sustainable materials for various 51
applications. The present study was aimed at prepar- 2.3 Density and Porosity Measurements 52
ing CNF/CNTs aerogel with different CNTs concen- 53
trations by environmentally friendly freeze-drying The samples bulk density (rbulk) was determined 54
method. The prepared CNF/CNTs aerogels thermal by dividing their weight by their volume. The 55

2  J. Renew. Mater. © 2017 Scrivener Publishing LLC

  Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

1 U and V are measured with a multimeter, while R is a

g/cm3
2 Density 0.025 known resistance equal to 50 Ω. Calibration of the hot
3 200 g strip device is needed to obtain the thermal condcutiv-
4 ity. It was performed using wood, cork, polyurethane
0.18 g
5 foams and polycarbonate samples having thermal con-
6 ductivity already determined by other techniques. The
7 (a) (b) detailed description of the thermal conductivity equip-
8 ment setup and calibration is given elsewhere [36, 38].
Figure 1 Images of monolithic CNF/CNTs aerogels. These
9 aerogels are: (a) lighweight, as they do not damage the
10 dandelion and (b) mechanically stable, as their shape does 2.6 Sample Electrical Resistance
11 not change under a compresive load of 200 g.
12 In order to measure the electrical properties, a silver
13 lacquer was applied on the opposite side of each sam-
14 measurements were performed with a digital caliper ple. The electrical resistance of the material was meas-
15 and a Denver Instrument (Model SI-114) balance. The ured by a Keithley 6517B electrometer at room temper-
16 porosity (ε) of the aerogels was calculated by Equation ature. The resistance (Relec) was measured three times
17 1, where the relative density is rbulk/rskeletal. The skeletal on each sample and the electrical resistivity (relec) was
18 density (rskeletal) of the aerogel has been estimated from calculated by Equation 3:
19 the proportion of CNF and MWCNT and their densi-
20 ties. They come from the literature and their values are relec = Relec × S/L(3)
21 1.6 [36] and 2.6 g/cm3 [37], respectively. where S is cross section of the sample (0.789 cm²) and L
22 is length of the sample (2 cm). The electrical resistivity
ε(%) = (1– (rbulk/rskeletal)) × 100 =
23 was used to calculate the electrical conductivity of the
(1– (rbulk / (%CNF × rCNF + %CNT × rCNT)) × 100(1) materials. The electrical conductivity (s) is equal to the
24
25 inverse of the electrical resistivity (relec).
26 2.4 Scanning Electron Microscopy
27
2.7 Thermoelectric Measurements
28 Morphology of the aerogel was observed with a JEOL
29 JSM 6460LV scanning electron microscope (SEM). It A thermoelectric (TE) material works either with a dif-
30 was used with an accelerating voltage of 20 kV. The ference of temperature creating an electrical potential
31 images made are cross sections of the samples. They or an electrical potential generating a temperature
32 were obtained after cryo-fracture of the sample in liquid difference [39]. The TE materials performance can be
33 nitrogen. Prior to observing the morphology, the sam- measured with the dimensionless figure of merit, ZT,
34 ples were sputtered with a thin gold layer. To get the defined by Equation 4:
35 pore size distribution, SEM images were analyzed with
36 ImageJ version 1.47. At least 100 pores were measured. ZT= s × S2 × T/λ(4)
37 where s is the electrical conductivity (S/cm), S is the
38 Seebeck coefficient (µV/K) which is equal to the ratio
2.5 Thermal Conductivity
39 of ΔV over ΔT, T is the absolute temperature (K), and
40 Thermal conductivity (λ) of the prepared aerogels l is the thermal conductivity (W/(m·K)) [40]. The
41 was measured with the hot strip technique. The con- Seebeck coefficient was determined via a homemade
42 ductance (Κ) of the aerogel is first calculated by the device. The aerogels’ thermoelectric temperature sens-
43 ratio of the dissipated heat through the sample over ing was measured by holding them between two cop-
44 the temperature difference in both sides of the sample per bars. Both sides of the samples were heated using
45 (Equation 2). two Peltier modules. The temperature difference (ΔT)
46 across the sample was monitored with two thermocou-
47
Κ = Q/DT = U × I / (T2 – T1) = (U × V)/ ples attached to each side. The voltage (ΔV) generated
48 (R × (T2 – T1))(2) by temperature difference between the hot and cold
49 More specifically, the cell which contained the mate- side was measured over time. The detailed experimen-
50 rial was kept at constant temperature of 19 °C (T1). The tal setup is provided in a previous publication [41].
51 other side of the sample had a temperature of T2. T2
52 was due to a temperature difference coming from the 2.8 Pressure Sensing
53 heating of a strip. The dissipated heat is calculated by
54 measuring the Joule effect, i.e., the tension (U, in Volts) Simultaneously measuring the electrical conduc-
55 times the intensity (V/R, in Amps) passing the strip. tivity and compressibility of the aerogels results in

J. Renew. Mater. © 2017 Scrivener Publishing LLC   3

Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

evaluating their pressure-responsive property. A for VOCs. The sensitivity of sensors is defined as the 1
cylindrically shaped aerogel with a diameter of 15 mm increase in sensors amplitude when a unit change is 2
and a height of 25 mm was prepared for compression applied in analyte concentration [42]. 3
test. The compression tests were performed with an In order to assess the efficiency of the sensors with 4
Anton Paar MCR 301 rheometer equipped with paral- low concentration of vapor molecules, the signal-to- 5
lel plates. The samples were placed between the two noise ratio (SNR) of the sensors was calculated during 6
circular electrodes and compressed to different strains. sensing of VOCs [43]. The SNR of sensors is calculated 7
Then, the electrical conductivity was measured at each by using Equation 6: 8
strain. 9
SNR = AR/ σbaseline(6) 10
where AR is response of aerogel obtained after expo- 11
2.9 Dynamic Vapor Sensing
sure to solvent and σbaseline represents the standard 12
The CNF/CNTs aerogel was tested for chemoresis- deviation in baseline resistance of sensor before the 13
tive properties using dynamic vapor testing setup. analyte delivery. The concentration of the solvents can 14
The conducting areogel sensor is developed to work at be detectable by the sensors if its SNR value is higher 15
room temperature. Therefore, the sensor has not been than 3 (SNR > 3). 16
tested at higher temperatures. The aerogel connected 17
with copper wires using silver paint was kept inside 18
3 RESULTS AND DISCUSSIONS
a closed rectangular chamber (10 × 10 cm). Aerogel 19
was first exposed to nitrogen bubbling in organic 20
3.1 Aerogel Structure and Morphology
solvents to generate saturated gas for sensing. Then, 21
an alternate cycle of dry nitrogen was carried out for The cross-sectional morphology of the prepared 22
the cleaning of the chamber from solvent vapors. The aerogels is shown in Figure 2. It can be observed that 23
VOCs selected are common toxic solvent vapors pre- the CNF aerogel morphology was slightly affected 24
sent in the environment. Each cycle was carried out after incorporation of different concentrations of 25
for 5 min at room temperature (20 °C) with relative CNTs. CNF aerogel (Figure 2a) showed a filmlike 26
humidity of 57%. structure which is commonly observed for CNF 27
Sensing device consisted of a mass flow controller, aerogels prepared with more than 0.5 wt% concen- 28
solvent bubbler, programmable electronic valves and tration [36]. The CNF aerogels with carbon nano- 29
Keithley 6517B electrometer connected with LabVIEW tubes (Figure 2b and 2c) showed similar sheetlike 30
software. The change in the resistance of an aerogel structures compared to neat CNF aerogel. The insert 31
was measured by multimeter and data were recorded in Figure 2b shows that some CNTs are absorbed 32
by software. Flow rate for dry nitrogen was 500 mL/ at the surface of CNF filmlike structure. The SEM 33
min and 100 mL/min for saturated solvent vapor; image analysis revealed that the neat CNF aerogels 34
both were controlled by mass flow controller. Relative have an average pore size of 26 ± 11 µm. Concerning 35
change in aerogel resistance (AR) was calculated from the CNF/CNTs aerogels, the pore size is 23 ± 11 µm. 36
Equation 5 to compare its response towards different This slight decrease in the pore size could be due 37
solvent vapors: to the aggregation of carbon nanotubes on the CNF 38
films (see insert in Figure 2b). The preparation of 39
AR = (R – R0) / R0(5) CNT aerogels highlights another possibility. Figure 40
where R0 is initial resistance of sensors in dry nitrogen 2d represents the SEM images of the cross section of 41
stream and R is resistance in the presence of solvent CNT aerogels. They showed that CNTs are also able 42
vapor. to form film via freeze-drying. Therefore, the reduc- 43
At saturated conditions, the concentration of VOCs tion of the pore size can also be due to the formation 44
cannot be controlled effectively. Therefore, experi- of new pore walls with CNTs. 45
ments were carried out using an Owlstone OVG-4 The density of the aerogels versus the CNT concen- 46
vapor generator and LabVIEW software to test the per- tration is presented in Figure 3. The density of the CNF 47
formance of CNF/CNTs aerogel at low concentration aerogel is 0.025 g/cm3 while the density of CNTs aero- 48
of water vapor and benzene vapor. The flow rate of gel is 0.035 g/cm3. Due to the higher density of CNTs, 49
the vapor was 100 mL/min at 83 °C for both solvents. the density of CNF/CNTs aerogel increased with 50
The concentrations of both solvent vapors were varied increasing CNTs content. The porosity of the neat CNF 51
from 100 to 17 ppm during the experiment using a spit aerogel and CNTs aerogel, calculated from Equation 1, 52
flow control valve of an OVG-4 oven. is 98.43 and 98.65%, respectively. This high porosity is 53
The slope of the curves representing AR versus in complete agreement with the large pore sizes mea- 54
the concentration gives the sensitivity of the sensor sured on SEM images. When different concentrations 55

4  J. Renew. Mater. © 2017 Scrivener Publishing LLC

  Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

1
2
3 1 µm 1 µm
4
5
6
7
100 µm 100 µm
8
9 (a) (b)
10
11
12 1 µm
13
14
15
16
100 µm 100 µm
17
18 (c) (d)
19
20 Figure 2 SEM images of the aerogel cross section: (a) neat CNF, (b) CNF/CNTs (95/5 wt%), (c) CNF/CNTs (75/25 wt%), and
21 (d) neat CNTs. The inserts are 100 times magnification of the pore wall. They reveal the fine interaction between CNF and CNTs.
22 In Figure 2a, smooth pore walls are observed. In Figure 2b, CNTs are present at the surface of CNF pore walls. In Figure 2d, the
pore walls are formed by CNT entanglement.
23
24
25
55
26 0.04
27
Thermal conductivity (mW/m.K)

50
28
29
45
30 0.03

Density (g/cm3)
31
32 40
33
34 35
35 0.02
36 30
37
38 25
39 0.01
40 0 5 10 15 20 25 100 105
Concentration of CNT (wt%)
41
42 Figure 3 Thermal conductivity and density of CNF/CNTs aerogels versus the CNT content.
43
44
45
46 of CNTs were incorporated into CNF aerogel, the 3.2 Thermal Conductivity
47 porosity of the resulting aerogel was not affected
48 significantly. The crosslinked polysaccharide-based The thermal conductivity of the aerogels is represented
49 aerogels have lower porosities (85–41%) and higher in Figure 3. For neat CNF aerogel, the value is around
50 densities (0.25–0.85 g/cm3) because, when crosslink- 24 mW/(m·K), which is slightly lower than the ther-
51 ing, such a network can undergo a large shrinkage, mal conductivity of air [27]. The observed CNF ther-
52 i.e., 20–60% [15, 36]. The absence of crosslinking in mal conductivity is similar to pectin aerogels while
53 our system can explain why it has higher porosity and superior to some other polysaccharide aerogels [44–
54 lower density compared to crosslinked polysaccharide 46]. It was expected that the CNTs could act as thermal
55 aerogels. opacifier in CNF aerogels. CNTs were used to reduce

J. Renew. Mater. © 2017 Scrivener Publishing LLC   5

Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

the radiation contribution of the thermal conductiv- into CNF aerogel, there was a significant improvement 1
ity in CNF/CNTs aerogels. However, the thermal in electrical conductivity of the resulting aerogel. The 2
conductivity of the CNF/CNTs aerogel was insignifi- increased electrical conductivity of CNF/CNTs aero- 3
cantly increased from 2 to 25 wt% CNTs incorporation gels is attributed to the charge carriers injected by 4
(Figure 3). Indeed, it has been commonly observed CNTs. It was observed that the CNF/CNTs aerogel 5
that a higher density for aerogels induces higher solid with 25 wt% CNTs had maximum electrical conduc- 6
conduction. Thus, the thermal conductivity of the tivity (0.029 S/cm). But it is still lower than the neat 7
resulting aerogel increases [47]. The maximum ther- CNTs aerogel electrical conductivity (0.045 S/cm). The 8
mal conductivty (38 mW/(m·K)) was observed with observed electrical conductivity of CNF/CNTs aero- 9
20 wt% CNTs incorporation. This observed significant gel is much higher than the literature cellulose-func- 10
deviation in the thermal conductivity of the CNF/ tionalized CNTs composite aerogels [12, 34] as well as 11
CNTs aerogel could be due to the reasons previously intrinsically conductive PEDOT:PSS/acid treated car- 12
discussed in the SEM image analysis. On one hand, bon nanotube aerogels [48]. This could be due to the 13
the formation of CNT pore walls may cause increased better dispersion of CNTs in the CNF matrix. 14
solid conduction. On the other hand, the aggregation 15
of the CNT on CNF pore walls can induce higher solid 16
conduction because of the intrinsic CNT thermal con- 3.4 Thermoelectric Properties 17
duction (Figure 2). Moreover, the first quick increase Because of previous results, we thought it would be 18
to 5 wt% of CNT followed by a slight increase of the interesting to examine the Seebeck coefficient of these 19
thermal conductivity may be attributed to the perco- aerogels to evaluate their potential as TE material. Due 20
lation thresholds of the CNT in the resulting aerogels to their p-type characteristic, all the aerogels showed 21
(Figure 2). From the thermal conductivity result, it positive Seebeck coefficient values. The Seebeck coef- 22
can be concluded that in order to achieve CNF/CNTs ficient value of neat CNTs aerogel was around 11 23
aerogels with a low thermal conductivity, a low con- μV/K. There was no significant difference in CNF/ 24
centration of CNT should be incorporated into the CNTs aerogel Seebeck coefficient when increasing 25
CNF aerogels. the CNTs content. The maximum Seebeck coefficient 26
(15 μV/K) value was observed for CNF/CNTs aerogel 27
with 25 wt% CNTs at room temperature. This Seebeck 28
3.3 Electrical Resistance and Electrical 29
coefficient is comparable to intrinsically conductive
Conductivity 30
PEDOT:PSS aerogels [27] while being lower than the
The resistance and the electrical conductivity of the n-type polyacetylenes/polyanilines value [25]. Using 31
aerogels as a function of the CNTs content are given the observed Seebeck coefficient values, ZT was evalu- 32
in Figure 4. The CNF/CNTs aerogel with 1 wt% CNTs ated. Its value was slightly increased with increasing 33
showed electrically insulating behavior with a high CNTs content up to 25 wt%. Yet, the value of the ZT 34
resistance value. When 5 wt% CNTs was incorporated was quite low. The maximum ZT value (0.0547 × 10–8) 35
36
37
0.1 38
39
1E9
40
Electrical conductivity (S/cm)

41
0.01
42
43
Resistance (Ohm)

44
1000 45
46
47
100 48
49
1E-9 50
10 51
0 5 10 15 20 25 100 52
Concentration of CNT (wt%)
53
54
Figure 4 Resistance and electrical conductivity of CNF/CNTs aerogels versus the CNT content. 55

6  J. Renew. Mater. © 2017 Scrivener Publishing LLC

  Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

1 was observed for CNF/25 wt% CNTs aerogel while it including water. Average maximum responses for 4
2 was 0.0343×10-8 for neat CNTs aerogel. cycles are shown in Figure 5a. Aerogel exhibited a
3 maximum response for water and chloroform. The
4 maximum response was observed towards chloro-
5
3.5 Vapor Sensing form. The second highest response was found once
6 Though the TE performances were not good enough exposed to water. In order to better understand
7 for applications, the developed aerogels exhibit a these results, we hypothesized possible VOC sens-
8 certain electrical conductivity. Therefore, CNF/CNTs ing mechanisms for CNF/CNTs sensor, which is
9 areogels have been tested for vapor-sensing proper- schematically represented in Figure 6 [49]. One pos-
10 ties. Chemoresistive response of CNF/CNTs aero- sibility is the swelling of the CNF structure (big red
11 gel is presented in Figure 5b for 10 different VOCs, circle in Figure 6). It induces the disconnection of the
12
13
14 Water
Acetone
15 0.10
CNF/CNT (5%)
Ethanol
Butanone
16 5000
Chloroform
17 Hexene
Tetra decane
0.08
18 4900 Benzene
Toluene
19
(Relative resistance)

Resistance (ohm)
Benzyl alcohol
4800
20 0.06

21
AR

22 0.04
4700

23
4600
24 0.02
25
4500
26
0.00
27 0 300 600 900 1200 1500 1800 2100 2400
28 (a) ter e ol e ne e e
Wa Aceton Ethan tanon roformHexe ecane enzen Toluen lcoho
l (b) Time (s)
Bu Chlo ra d B la
29 Te t
Be n z y

30
31 Figure 5 VOCs sensing response of the CNF/CNTs (5%) aerogel: (a) representation of the mean relative resistance of the sensor
32 over the 4 cycles represented in Figure 5b. Error bars are the standard deviation. (b) Row data of the sensor resistance for 10
VOCs over the 4-cycle experiment.
33
34
35
36
37
38
39
40
41
42
43
44
45
46 CNF film-like structure
47
48 CNT electrically
49 conducting network
50 Solvent molecules
51
52 Figure 6 Mechanism of CNF/CNTs aerogel vapor sensing in the presence of solvent vapors. If the solvent has preferential
53 interactions with CNFs, then the structure may swell (big red circle) and induce the disconnection of the CNTs (small red circles).
54 If the solvent interacts more with CNTs, it may disrupt the CNT/CNT connections (white circles). In both case, it induces a
55 decrease of the electrical conduction.

J. Renew. Mater. © 2017 Scrivener Publishing LLC   7

Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

CNTs (small red circles in Figure 6) and an increase responses are lower because of their low interaction 1
of the electrical resistivity of the aerogel. Another with cellulosic fibers. 2
possibility is the disturbance of CNT/CNT connec- To study the sensitivity of the sensor, the response 3
tion because of the presence of solvent molecules of one non-interacting (benzene) and one interacting 4
(white circles in Figure 6). With these hypotheses, (water) solvent was carried out at 4 different concen- 5
we speculate the following to explain our results. On trations (at 100, 50, 25 and 17 ppm). The raw signals 6
one hand, for chloroform, the maximum response is are presented in Figure 7b and 7c for benzene and 7
due to its high concentration present at room tem- water respectively. For benzene, fewer variations were 8
perature. The chloroform is known to have good observed in the response when concentration was 9
swelling effects on polymers. Thus, chloroform varied compared to water. To evaluate the difference 10
absorption at CNF surface increased the resistance of in behavior between both solvents, AR is plotted with 11
conducting network very rapidly, as shown in Figure respect to concentrations (Figure 7a). It can be seen that 12
5b [50–52]. On the other hand, the hydrophilic nature the sensitivity of CNF/CNT aerogel was 1.642 × 10-5 13
of CNF leads to the absorption of water molecules ppm-1 and 8.443 × 10-5 ppm-1 for benzene and water, 14
at conducting junctions of CNTs [53]. The CNF film respectively. This sensitivity of CNF/CNTs aerogel is 15
swelling and the disconnection of CNTs due to the higher for moisture. Some studies have already shown 16
presence of water molecules induces a decrease of that sensors may have enhanced response when sol- 17
the electrical conduction. For other nonpolar VOCs, vent vapors are mixed with a large amount of water 18
19
20
0.016
4590 100 ppm Water 21
22
4580
23
0.014
y=0.00766+(1.64E-5)x
4570
50 ppm 24
4560 25 ppm 25
Resistance (ohm)

0.012
4550
17 ppm 26
27
AR

4540
0.010 28
4530 29
0.008
4520 30
y=0.00702+(1.64E-5)x
4510
31
Benzene 32
0.006
Water
4500
33
0 300 600 900 1200 1500 1800 2100 2400 2700 34
20 40 60 80 100 120
Concentration (ppm)
Time (s) 35
(a) (b)
36
4550
Benzene
37
100 ppm 38
50 ppm
4540 25 ppm
39
17 ppm 40
41
Resistance (ohm)

4530
42
43
4520 44
45
4510
46
47
48
4500
49
2100
50
0 300 600 900 1200 1500 1800 2400
51
(c) Time (s)
52
Figure 7 (a) Sensitivity curve of aerogel at various concentrations (100 to 17 ppm) of water and benzene. It corresponds to the 53
evolution of the relative resistance of the sensor versus the concentration of the solvent vapors. (b, c) The raw resistance of the 54
sensor versus time for benzene and water, respectively, at various concentrations (100 to 17 ppm). 55

8  J. Renew. Mater. © 2017 Scrivener Publishing LLC

  Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

1 vapors [54]. However, due to the high affinity of CNF of the CNF/CNTs aerogel gradually decreased with
2 towards water, a large amount of water can disturb increasing compressive strain and stress. This behav-
3 the conducting network of CNTs in CNF/CNTs aero- ior is attributed to an increasing number of contacts in
4 gel. Therefore, the sensing response of the CNF/CNTs the continuous conductive CNTs network. The lowest
5 aerogels with high humidity was not performed in relative electrical resistance was observed when the
6 this study. Though not very selective for any specific strain and stress reached 70% and ~177 kPa, respec-
7 solvent vapors, CNF/CNTs sensor can detect a wide tively. This observation was in good agreement with
8 range of VOCs present as pollutants in the environ- aerographite relative electrical resistance change with
9 ment. Besides thermoelectric properties, the CNF/ respect to compressive strain [56]. Zhou et al. [57] also
10 CNTs sensor can be used to monitor the closed envi- observed such a behavior in CNF/polypyrrole/silver
11 ronments such as inside buildings, offices and homes. nanoparticles aerogels with increasing compressive
12 Finally, the efficiency of the sensor was evaluated. strain up to 90%. Therefore, this kind of electrically
13 SNR values were 4.6 and 5.7 for water and benzene, conductive aerogels under compressive strain could
14 respectively. It illustrates the ability of CNF/CNTs be used for pressure-sensing material applications.
15 aerogels to correctly sense down to the concentration
16 of 15 ppm for both solvents. However, it was difficult
17 to sense VOCs below 15 ppm with this aerogel. Yet, 4 CONCLUSIONS
18 this benzene detection limit (15 ppm) is much lower
19 In conclusion, an efficient, low-cost, and environ-
than that of the human nose (60 ppm) in air [55]. The
20 mentally friendly process was used to produce CNF/
stability of signals was also calculated from the stan-
CNTs aerogels. The thermal conductivity, themoelec-
21 dard deviation of the baseline. The σbaseline was around
22 trical property, electrical conductivity, chemical vapor-
0.002, which is very small as compared to the signals.
23 sensing, and pressure-sensing properties were inves-
The maximum deviation was less than 10% of its aver-
24 tigated. With the addition of CNTs into CNF aerogel,
age signal value. Thus, the signals generated by CNF/
25 the thermal conductivity and density of the resulting
CNTs areogel were quite stable at room temperature.
26 aerogels were slightly increased as compared to neat
27 CNF aerogel. At the same time, electrically conductive
28 3.6 Pressure Sensing aerogels were obtained. Therefore, the CNF/CNTs
29 aerogels were tested for TE properties. The Seebeck
Pressure-responsive properties of the aerogel were and ZT values appeared too low for the real applica-
30 obtained from their combined compressible properties
31 tion requirements. Therefore, these values should be
and electrical resistance. The relative electrical resist- effectively modified by increasing electrical conduc-
32 ance, compressive stress, and compressive strain of the
33 tivity and reducing the thermal conductivity of CNF/
CNF/CNTs (95/5 wt%) aerogel are shown in Figure 8. CNTs aerogels. Yet, the CNF/CNTs aerogels showed
34 Ordinarily, electrical conductivity increases when the
35 relative electrical resistance change towards chemi-
contact between the electrically conductive particles cal vapors because of their porous structure, which
36 increases. As expected, the relative electrical resistance
37 can efficiently and directly interact with vapors. The
38 electrical conductivity of the aerogel was increased
39 under compressive stress because of the formation of
40 0.00
continuous CNT networks. Finally, all these results
41 illustrate a facile and intriguing preparation strategy
aerogel

42
wire 150 for aerogels for multifunctional material applications.
–0.6
43 The prepared CNF/CNTs aerogel could be used as
a platform to develop a new class of TE material by
Relative resistance

44
45
100 incorporating intrinsically conductive polymers and/
Stress (kPa)

46 or by using a reductive treatment.

–0.8
47
50
48
49 REFERENCES
Compressed aerogel wire
50 1. K.S. Mikkonen, K. Parikka, A. Ghafar, and M. Tenkanen,
0
51 –1.0
Prospects of polysaccharide aerogels as modern
0 10 20 30 40 50 60 70
52 advanced food materials. Trends Food Sci. Technol. 34(2),
Strain (%)
53 124–136 (2013).
54 Figure 8 Relative resitance and stress of CNF/CNTs (95/5 2. H. Maleki, L. Durães, C.A. García-González, P. del
55 wt%) aerogels with respect to strain. Gaudio, A. Portugal, and M. Mahmoudi, Synthesis and

J. Renew. Mater. © 2017 Scrivener Publishing LLC   9

Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

biomedical applications of aerogels: Possibilities and 19. J. Fricke and A. Emmerling, Aerogels—Preparation, 1
challenges. Adv. Colloid Interface Sci. 236, 1–27 (2016). properties, applications. Struct. Bond. 77, 37–87 (1992). 2
3. A. Demilecamps, C. Beauger, C. Hildenbrand, A. Rigacci, 20. F. Ayadi, B. Martín-García, M. Colombo, A. Polovitsyn, 3
and T. Budtova, Cellulose–silica aerogels. Carbohydr. A. Scarpellini, L. Ceseracciu, I. Moreels, and A. 4
Polym. 122, 293–300 (2015). Athanassiou, Mechanically flexible and optically trans-
5
4. S. Takeshita and S. Yoda, Chitosan aerogels: Transparent, parent three-dimensional nanofibrous amorphous aero-
6
flexible thermal insulators. Chem. Mater. 27(22), 7569– cellulose. Carbohydr. Polym. 149, 217–223 (2016).
7572 (2015). 21. D. Bendahou, A. Bendahou, B. Seantier, Y. Grohens, 7
5. Y. Kobayashi, T. Saito, and A. Isogai, Aerogels with 3D amd H. Kaddami, Nano-fibrillated cellulose-zeolites 8
ordered nanofiber skeletons of liquid-crystalline nano- based new hybrid composites aerogels with super ther- 9
cellulose derivatives as tough and transparent insula- mal insulating properties. Ind. Crops Prod. 65, 374–382 10
tors. Angew. Chem. Int. Ed. 53(39), 10394–10397 (2014). (2015). 11
6. C. Rudaz, R. Courson, L. Bonnet, S. Calas-Etienne, H. 22. D. Lee, P.C. Stevens, S.Q. Zeng, and A.J. Hunt, Thermal 12
Sallée, and T. Budtova, Aeropectin: Fully biomass-based characterization of carbon-opacified silica aerogels. J. 13
mechanically strong and thermal superinsulating aero- Non-Cryst. Solids 186, 285–290 (1995). 14
gel. Biomacromolecules 15(6), 2188–2195 (2014). 23. H. Yu, D. Liu, Y. Duan, and Z. Yang, Applicability of the 15
7. G. Zu, J. Shen, L. Zou, F. Wang, X. Wang, Y. Zhang, and effective medium theory for optimizing thermal radia-
16
X. Yao, Nanocellulose-derived highly porous carbon tive properties of systems containing wavelength-sized
17
aerogels for supercapacitors. Carbon 99, 203–211 (2016). particles. Int. J. Heat Mass Transf. 87, 303–311 (2015).
8. J. Tian, D. Peng, X. Wu, W. Li, H. Deng, and S. Liu, 24. M.A. Kamarudin, S.R. Sahamir, R.S. Datta, B.D. Long, 18
Electrodeposition of Ag nanoparticles on conductive M.F. Mohd Sabri, and S. Mohd Said, A review on the 19
polyaniline/cellulose aerogels with increased synergis- fabrication of polymer-based thermoelectric mate- 20
tic effect for energy storage. Carbohydr. Polym. 156, 19–25 rials and fabrication methods. Sci. World J. DOI: 21
(2017). 10.1155/2013/713640 (2013). 22
9. B. Wicklein, A. Kocjan, G. Salazar-Alvarez, F. Carosio, 25. B. Russ, A. Glaudell, J.J. Urban, M.L. Chabinyc, and R.A. 23
G. Camino, M. Antonietti, and L. Bergström, Thermally Segalman, Organic thermoelectric materials for energy 24
insulating and fire-retardant lightweight anisotropic harvesting and temperature control. Nat. Rev. Mater. 1, 25
foams based on nanocellulose and graphene oxide. Nat. 16050 (2016). 26
Nano. 10(3), 277–283 (2015). 26. N. Massonnet, A. Carella, O. Jaudouin, P. Rannou, G.
27
10. H. Qi, J. Liu, J. Pionteck, P. Pötschke, and E. Mäder, Laval, C. Celle, and J.-P. Simonato, Improvement of the
28
Carbon nanotube–cellulose composite aerogels for Seebeck coefficient of PEDOT:PSS by chemical reduc-
vapour sensing. Sens. Actuators B Chem. 213, 20–26 (2015). tion combined with a novel method for its transfer using 29
11. J. Zou, J. Liu, A.S. Karakoti, A. Kumar, D. Joung, Q. Li, S.I. free-standing thin films. J. Mater. Chem. C 2(7), 1278283 30
Khondaker, S. Seal, and L. Zhai, Ultralight multiwalled (2014). 31
carbon nanotube aerogel. ACS Nano 4(12), 7293–7302, 27. M.P. Gordon, E.W. Zaia, P. Zhou, B. Russ, N.E. Coates, A. 32
(2010). Sahu, and J.J. Urban, Soft PEDOT:PSS aerogel architec- 33
12. M. Wang, I.V. Anoshkin, A.G. Nasibulin, R.H. Ras, J. tures for thermoelectric applications. J. Appl. Polym. Sci. 34
Laine, E.I. Kauppinen, and O. Ikkala, Electrical behav- 134(3), 44070 (2017). 35
iour of native cellulose nanofibril/carbon nanotube 28. L. Zhao, X. Sun, Z. Lei, J. Zhao, J. Wu, Q. Li, and A. 36
hybrid aerogels under cyclic compression. RSC Adv. Zhang, Thermoelectric behavior of aerogels based on 37
6(92), 89051–89056 (2016). graphene and multi-walled carbon nanotube nanocom-
38
13. H. Maleki, Recent advances in aerogels for environmen- posites. Compos. Part B Eng. 83, 317–322 (2015).
39
tal remediation applications: A review. Chem. Eng. J. 300, 29. M. Knite, V. Teteris, A. Kiploka, and J. Kaupuzs,
98–118 (2016). Polyisoprene-carbon black nanocomposites as tensile 40
14. S. Nardecchia, D. Carriazo, M.L. Ferrer, M.C. Gutierrez, strain and pressure sensor materials. Sens. Actuators 41
and F. del Monte, Three dimensional macroporous archi- Phys. 110(1), 142–149 (2004). 42
tectures and aerogels built of carbon nanotubes and/or 30. M. Knite, K. Ozols, G. Sakale, and V. Teteris, Polyisoprene 43
graphene: Synthesis and applications. Chem. Soc. Rev. and high structure carbon nanoparticle composite for 44
42(2), 794–830 (2013). sensing organic solvent vapours. Sens. Actuators B Chem. 45
15. F. Fischer, A. Rigacci, R. Pirard, S. Berthon-Fabry, and P. 126(1), 209–213 (2007). 46
Achard, Cellulose-based aerogels. Polymer 47(22), 7636– 31. K.M. Tripathi, T. Kim, D. Losic, and T.T. Tung, Recent 47
7645 (2006). advances in engineered graphene and composites for 48
16. H.-P. Ebert, Thermal properties of aerogels, in Aerogels detection of volatile organic compounds (VOCs) and
49
Handbook, pp. 537–564, Springer, New York (2011). non-invasive diseases diagnosis. Carbon 110, 97–129
50
17. J. Fricke, Aerogels—A fascinating class of high-perfor- (2016).
mance porous solids, in Aerogels, pp. 2–19, Springer, 32. B.S. Noremberg, R.M. Silva, O.G. Paniz, J.H. Alano, 51
Berlin, Heidelberg (1986). M.R.F. Gonçalves, S.I. Wolke, J. Labidi, A. Valentini, and 52
18. M. Venkataraman, R. Mishra, T.M. Kotresh, J. Militky, N.L. V Carreño, From banana stem to conductive paper: 53
and H. Jamshaid, Aerogels for thermal insulation in high- A capacitive electrode and gas sensor. Sens. Actuators B 54
performance textiles. Text. Prog. 48(2), 55–118 (2016). Chem. 240, 459–467 (2017). 55

10  J. Renew. Mater. © 2017 Scrivener Publishing LLC

  Muthuraj et al.: Vapor and Pressure Sensors Based on Cellulose Nanofibers and Carbon Nanotubes Aerogel

1 33. H. Qi, E. Mader, and J. Liu, Electrically conductive aero- alginate aerogels’ production. J. Non-Cryst. Solids
2 gels composed of cellulose and carbon nanotubes. J. 443,112–117 (2016).
3 Mater. Chem. A 1(34), 9714–9720 (2013). 46. T.O.J. Blomfeldt, F. Nilsson, T. Holgate, J. Xu, E.
4 34. M. Wang, I. V Anoshkin, A.G. Nasibulin, J.T. Korhonen, Johansson, and M.S. Hedenqvist, Thermal conductivity
J. Seitsonen, J. Pere, E.I. Kauppinen, R.H.A. Ras, and O. and combustion properties of wheat gluten foams. ACS
5
Ikkala, Modifying native nanocellulose aerogels with Appl. Mater. Interfaces 4(3), 1629–1635 (2012).
6
carbon nanotubes for mechanoresponsive conductiv- 47. O.A. Madyan, M. Fan, L. Feo, and D. Hui, Physical prop-
7 ity and pressure sensing, Adv. Mater. 25(17), 2428–2432 erties of clay aerogel composites: An overview. Compos.
8 (2013). Part B Eng. 102, 29–37 (2016).
9 35. T. Saito, Y. Nishiyama, J.-L. Putaux, M. Vignon, and 48. X. Zhang, J. Liu, B. Xu, Y. Su, and Y. Luo, Ultralight con-
10 A. Isogai, homogeneous suspensions of individual- ducting polymer/carbon nanotube composite aerogels.
11 ized microfibrils from TEMPO-catalyzed oxidation Carbon 49(6), 1884–1893 (2011).
12 of native cellulose. Biomacromolecules 7(6), 1687–1691 49. J.F. Feller, N. Gatt, B. Kumar, and M. Castro, Selectivity
13 (2006). of chemoresistive sensors made of chemically func-
14 36. C. Jiménez-Saelices, B. Seantier, B. Cathala, and Y. tionalized carbon nanotube random networks for vola-
15 Grohens, Spray freeze-dried nanofibrillated cellu- tile organic compounds (VOC). ChemoSensors 2, 26–40
lose aerogels with thermal superinsulating properties. (2014).
16
Carbohydr. Polym. 157, 105–113 (2017). 50. B. Kumar, M. Castro, and J.F. Feller, Tailoring the chemo-
17
37. M.F.L. De Volder, S.H. Tawfick, R.H. Baughman, and A.J. resistive response of self-assembled polysaccharide-
18 Hart, Carbon nanotubes: Present and future commercial CNT sensors by chain conformation at tunnel junctions.
19 applications. Science 339(6119), 535–539 (2013). Carbon 50(10), 3627–3634 (2012).
20 38. B. Seantier, D. Bendahou, A. Bendahou, Y. Grohens, 51. B. Kumar, M. Castro, and J.F. Feller, Controlled conduc-
21 and H. Kaddami, Multi-scale cellulose based new bio- tive junction gap for chitosan–carbon nanotube quan-
22 aerogel composites with thermal super-insulating and tum resistive vapour sensors. J. Mater. Chem. 22(21),
23 tunable mechanical properties. Carbohydr. Polym. 138, 10656–10664 (2012).
24 335–348 (2016). 52. B. Kumar, J.F. Feller, M. Castro, and J. Lu, Conductive
25 39. D.D.L. Chung, Composites get smart. Mater. Today 5(1), bio-polymer nano-composites (CPC): Chitosan-carbon
26 30–35 (2002). nanotube transducers assembled via spray layer-by-
40. A. Dey, O.P. Bajpai, A.K. Sikder, S. Chattopadhyay, and layer for volatile organic compound sensing. Talanta
27
M.A. Shafeeuulla Khan, Recent advances in CNT/gra- 81(3), 908–915 (2010).
28
phene based thermoelectric polymer nanocomposite: 53. B. Kumar, M. Castro, and J.F. Feller, Controlled conduc-
29 A proficient move towards waste energy harvesting. tive junction gap for chitosan-carbon nanotube quan-
30 Renew. Sustain. Energy Rev. 53, 653–671 (2016). tum resistive vapour sensors. J. Mater. Chem. 22(21),
31 41. T.T. Tung, C. Pham-Huu, I. Janowska, T. Kim, M. Castro, 10656–10664 (2012).
32 and J.-F. Feller, Hybrid films of graphene and carbon 54. S. Nag, A. Sachan, M. Castro, V. Choudhary, and J.F.
33 nanotubes for high performance chemical and tem- Feller, Spray layer-by-layer assembly of POSS function-
34 perature sensing applications. Small 11(28), 3485–3493 alized CNT quantum chemo-resistive sensors with tune-
35 (2015). able selectivity and ppm resolution to VOC biomarkers.
36 42. J. Li, Y. Lu, Q. Ye, M. Cinke, J. Han, and M. Meyyappan, Sens. Actuators B Chem. 222, 362–373 (2016).
37 Carbon nanotube sensors for gas and organic vapor 55. Http://desnhgovorganizationcommissionerpipfactsheetsard-
detection. Nano Lett. 3(7), 929–933 (2003). documentsard-Ehp-3pdf. (2017).
38
43. S. Nag, L. Duarte, E. Bertrand, V. Celton, M. Castro, V. 56. S. Garlof, M. Mecklenburg, D. Smazna, Y.K. Mishra, R.
39
Choudhary, P. Guegan, and J.F. Feller, Ultrasensitive Adelung, K. Schulte, and B. Fiedler, 3D carbon networks
40 QRS made by supramolecular assembly of functional- and their polymer composites: Fabrication and electro-
41 ized cyclodextrins and graphene for the detection of mechanical investigations of neat aerographite and aero-
42 lung cancer VOC biomarkers. J. Mater. Chem. B 2(38), graphite-based PNCs under compressive load. Carbon
43 6571–6579 (2014). 111, 103–112 (2017).
44 44. G. Tkalec, Z. Knez, and Z. Novak, Formation of polysac- 57. S. Zhou, M. Wang, X. Chen, and F. Xu, Facile template
45 charide aerogels in ethanol. RSC Adv. 5(94), 77362–77371 synthesis of microfibrillated cellulose/polypyrrole/sil-
46 (2015). ver nanoparticles hybrid aerogels with electrical conduc-
47 45. G. Tkalec, R. Kranvogl, A.P. Uzunalić, Ž. Knez, and tive and pressure responsive properties. ACS Sustain.
48 Z. Novak, Optimisation of critical parameters during Chem. Eng. 3(12), 3346–3354 (2015).
49
50
51
52
53
54
55

J. Renew. Mater. © 2017 Scrivener Publishing LLC   11

View publication stats