composites of inorganic nanotubes and polymersmacromolchem.com/papers/paper43.pdfpart of topical...

Part of Topical Section onComposites of Inorganic Nanotubes and Polymers

Influence of inorganic nanoparticleson the glass transitions of polyureaelastomers

Marius Reinecker*,1, Armin Fuith1, Viktor Soprunyuk1, Antoni Sánchez-Ferrer2, Aleš Mrzel3, Renato Torre4,and Wilfried Schranz**,1

1 Faculty of Physics, University of Vienna, Boltzmanngasse 5, 1090 Vienna, Austria2 Department of Health Sciences and Technology, ETH Zurich, Institute of Food, Nutrition and Health, Food and Soft Materials ScienceGroup, Schmelzbergstrasse 9, LFO, E29, 8092 Zurich, Switzerland

3 J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia4 European Laboratory for Non-Linear Spectroscopy, Universitá di Firenze, via Carrara 1, 50019 Sesto Fiorentino, Italy

Received 10 May 2013, revised 7 August 2013, accepted 9 August 2013Published online 9 September 2013

Keywords composites, elasticity, glass transitions, inorganic materials, nanoparticles, polymers

* Corresponding author: e-mail [email protected], Phone: þ43 1 4277 51171, Fax: þ43 1 4277 51181** e-mail [email protected], Phone: þ43 1 4277 51171, Fax: þ43 1 4277 51181

The properties of polymers can be significantly changed byincorporating nanoparticles, which yields a great potential forapplications. In the present study, we use nanocomposites ofnew polyurea elastomers filled with 0.1, 0.5, and 1wt.% MoS2nanotubes. Using dynamic mechanical analysis (DMA)measurements we show, that the glass transition temperatures

Tg of the nanocomposites are increased by small amountsof inorganic nanotubes. In line with results from computersimulations of polymer melts with nanoscopic particles, weexplain the observed shift of Tg to be due to a gradual slowingdown of polymer chain dynamics in the proximity ofnanotubes.

� 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1 Introduction Composites of nanoparticles (NPs)and polymers are attractive materials with a high potentialfor applications [1], due to their flexibility and superiorelectrical, magnetic, mechanical, thermoelectric, and opticalproperties, etc. [2]. The properties of polymers can be tunedby adding NPs of different shapes, i.e., spheres, tubes, orsheets. Since NPs have large surface areas their performancedepends strongly on the properties of the interfacial regionsbetween the polymer and the NPs, i.e., on the type ofinteraction (attractive or repulsive) between the NPs and thepolymer matrix, the degree of dispersion of the NPs, etc.

Polyurea networks synthesized by sol/gel chemistry [3]are relatively new elastomers with extreme resistance tochemical abrasion and service temperatures between �60and þ200 8C.

In the present work, we have to deal with a problem ofmany scales of length (and time) for the following reason.Polyurea elastomers exhibit a phase separated structure [3]with rigid urea domains (hard domains) embedded in amatrix of flexible polymer chains (soft domains). The

average size of the hard domains is of the order of severalnanometers and the same holds for the average distancebetween them [3]. As we shall show below, the averagedistance between MoS2 nanotubes is of the order of severalmicrometers, i.e., thousand times larger.

The mechanical properties of polyurea elastomers can betuned by the relative amount of hard and soft domains –

average molecular weight between crosslinking points – andalso by the loading with NPs [4]. Pure polyurea elastomersexhibit two glass transition temperatures at Tg1��65 8Cand Tg2� Tg1þ 25 8C [9] attributed to cooperative motionsof molecular segments in the soft domains of pure polyurea(Tg1) and to regions near the hard nanodomains (Tg2).

Although there are several techniques available toevaluate Tg, one of the most sensitive techniques is dynamicmechanical analysis (DMA). Here we present new DMAexperiments where the presence of inorganic MoS2nanotubes [5] in different PU elastomer nanocompositesinfluences the polymer chain dynamics and the glasstransition temperatures of the polymer network.

Phys. Status Solidi A 210, No. 11, 2320–2327 (2013) / DOI 10.1002/pssa.201329317 p s sapplications and materials science

a

statu

s

soli

di

www.pss-a.comph

ysi

ca

� 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

2 Experimental2.1 Samples For the present study, the linear hydro-

phobic diamino-terminated polyetheramine Jeffamine D-2000 (Mn¼ 2060 gmol�1, r¼ 991 kgm�3) and the hydro-philic Jeffamine ED-2003 (Mn¼ 2300 gmol�1, r¼ 1068 kgm�3), from Huntsman International LLC, and the triiso-cyanate crosslinker Basonat HI-100 (mw¼ 504 gmol�1,r¼ 1174 kgm�3), from BASF SE, were used as received.Details of the synthesis of these PU elastomers are welldescribed in the literature [4].

PU polymers are formed by the chemical reactionbetween a diisocyanate and a diamine to build urea moieties,as shown in Scheme 1. When one of these components – theisocyanate- or amino-containing chemicals – has more thantwo reacting groups, then networks are obtained.

In these PU networks, intermolecular hydrogen bondsare formed between the active hydrogen atoms from one ureagroup (–NH–) and the oxygen coming from the carbonylgroup (>CO) from another urea moiety (Scheme 2),bringing additional properties to the system.

At room temperature polyurea shows microphasesegregation, with hard segments dispersed in a soft matrix(Fig. 1). From SAXS and WAXS experiments [4], theaverage size of the hard domains and the distance betweenthem were determined for the Jeffamine D-2000-based PUnetwork, with values in the order of few nanometers and6.4 nm, respectively.

Two reference elastomeric samples JD-2000 and JEDwere synthesized when crosslinking the two linear diamino-terminated polyetheramine Jeffamine D-2000 and Jeffamine

ED-2003, respectively, with the triisocyanate crosslinkerBasonat HI-100. The volume fraction of hard domains forboth PU systems was fx¼ 0.12.

A first PU nanocomposite containing around 1wt.% ofMoS2 nanotubes (JD-2000/MoS2) was synthesized aftercrosslinking the corresponding Jeffamine D-2000 [8, 9].Moreover, two sets of nanocomposites filled with 0.1,0.5, and 1wt.% of MoS2 multiwall-nanotubes (MoS2/170:80–100 nm diameter, 10mm length) obtained from thedesulfurization of MoS7 nanotubes (JED-C1-X) and ofMoS2 multiwall-nanotubes (MoS2/628: 250 nm diameter,100mm length) obtained from the decomposition ofmultiwall Mo6S2I8 nanowires (JED-C2-X) were preparedafter crosslinking the Jeffamine ED-2003.

2.2 Experimental methods DMA is a technique tomeasure the low frequency elastic moduli and damping ofmaterials [7]. A static force Fstat is sinusoidally modulated bya dynamic force Fdyn exp(ivt) at a chosen amplitude andfrequency. The elastic and anelastic response of the sampleleads to a change in length u and phase shift d between forceand amplitude, which is registered via inductive coupling.Normal resolutions for the length and phase shift areDu� 10 nm and Dd� 0.18, respectively. The knowledge ofthe sample length u and the phase shift d allows determiningthe real and imaginary parts of a certain component of thecomplex elastic compliance tensor S

�ii

Sii ¼ S0ii cos d and S

00ii ¼ Sii sin d: ð1Þ

The corresponding storage (E0) and loss modulus (E00)are defined as E

0:= 1=S

0ii and E

00:= 1=S

00ii, respectively.

DMA experiments were performed on a Perkin ElmerDMA 7 or on the Perkin Elmer Diamond DMA. In bothsetups the complex elastic compliance can be measured in

Scheme 1 Chemical reaction between a diisocyanate and adiamine yielding the corresponding polyurea compound (in thered oval, the urea moiety is shown).

Scheme 2 Intermolecular hydrogen bonds between two ureamotifs.

Figure 1 Phase separated structure of polyurea elastomers.

Phys. Status Solidi A 210, No. 11 (2013) 2321

www.pss-a.com � 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Original

Paper

a frequency range between 0.01 and 100Hz, and fortemperatures between 80 and 850K. The minimal force thatcan be applied is 10�3N with a resolution of 10�5 N. Themaximal forces are 2.5N for DMA 7 and 10N for theDiamond DMA. In both instruments the force is transmittedvia a quartz- (DMA 7) or steel- (DMA 7 and DiamondDMA) rod, which also probes the sample size. The presentexperiments were performed in tensile stress (TS) geometry(Fig. 2). For more details concerning the DMA method, see,e.g., Ref. [7]. Typical sample size was l¼ 9mm, b¼ 2mm,and t¼ 0.2mm. The samples were cooled down to about�183 8C and heated again with a heating rate between 0.5and 2 8Cmin�1. Up to five heating/cooling cycles wereperformed. In the present work, we always compare data forpure and filled samples that were taken from the sameheating sequence.

3 Results and discussion In Figs. 3–5, optical andscanning electron microscopy images of polyurea JD-2000filled with �1wt.% MoS2 (Fig. 3) and JED-C1 filled with0.1wt.% MoS2/170 (Fig. 4) and 0.5 wt.% MoS2/170 (Fig. 5)are shown for comparison.

The nanotubes are well dispersed, with an averagedistance between the NPs of several micrometers (Figs. 3and 4), i.e., several tens of the diameter (�100 nm) of amultiwall-nanotube. The JED-C1 samples filled with0.5wt.% show “clouds” of agglomerated nanotubes withregions between where the NPs are relatively well dispersed(Fig. 5). A similar inhomogeneous distribution of NPswas also observed for the 1wt.% JED-C1 samples. Forcomparison, pristine MoS2 nanotubes usually form bundlesor ropes of up to 106 individual tubes [5] which can be up tomany tens of micrometers long.

In order to study the effect of NPs in the polymermatrices, DMA experiments as a function of temperature andfrequency were performed.

Figure 6 displays the temperature and frequencydependencies of tan d¼E00/E0 of pure polyurea JD-2000,showing two glass transitions at Tg1��65 8C andTg2� Tg1þ 25 8C.

The glass transition at Tg1 is attributed to the regions inthe soft domains whereas the higher Tg2 originates fromregions near the hard domains [9]. Such a two glasstransition behavior can be observed when sufficiently strongparticle enthalpic interactions lead to a permanent attach-ment of chain segments to the NPs [24], which obviously isthe case in the vicinity of the hard domains in polyurea. Adiscussion of this two glass transition behavior as well asdetailed measurements for polyureas with different molecu-lar weight of the polyetheramines (soft domains) will be thesubject of a forthcoming paper.

The storage E0 and loss E00 moduli and tan d as functionof temperature are compared for the pure polyurea elastomerJD-2000 and the elastomeric nanocomposite JD-2000/MoS2(�1wt.% of MoS2) in Fig. 7. Several methods are

Figure 2 Tensile stress measurement geometry (left). The tworight-hand pictures show two JED samples with 0.1 and 0.5wt.%MoS2, respectively. Sample size is about 2mm� 9mm and0.2mm.

Figure 3 Optical (top) and scanning electron microscopy(bottom) [6] images of JD-2000 filled with �1wt.% MoS2.

2322 M. Reinecker et al.: Influence of inorganic nanoparticles on the glass transitions of polyurea elastomers

� 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.pss-a.com

ph

ysic

a ssp stat

us

solid

i a

commonly used to determine the glass transition temperatureTg from DMA measurements [10]. They either relate Tg tothe onset-temperature of E0 or to the temperature of theE00-peak or the tan d peak, respectively. Irrespectively whichmethod we use to locate Tg, we obtain a shift of Tg1 tohigher temperatures by the addition of the inorganic NPs. For1wt.%MoS2 we obtain a shift of Tg1 of aboutþ8 8C (Fig. 7).Averaging ten successive temperature scans for pure andfilled JD-2000, respectively we obtain DTg1¼ 8� 3 8C. ForJD-2000 filled with 0.1 wt.% MoS2 the change in Tg1 wassmaller than 0.8 8C. In contrast, Tg2 is not affected by thepresence of inorganic NPs.

It is important to note, that all these measurementsshown here were performed on samples that were notannealed at high temperatures, i.e., the intermolecularhydrogen bonds between polyurea chains were not removed.Starting from room temperature they were cooled down to

low temperatures and then heated until the hydrogennetwork broke down.

Figure 4 Optical (top) and scanning electron microscopy(bottom) of JED-C1 filled with 0.1 wt.% MoS2/170. On theSEM picture separated multiwall-nanotubes are visible as “sticks”of about10mm length.

Figure 5 Optical microscopy (top) and SEM (bottom) images ofJED-C1 filled with 0.5wt.% MoS2, showing “clouds” of MoS2aggregates (top) with relatively well-dispersed regions in between(bottom). Individual multiwall-nanotubes of about 100 nm thick-ness and 10mm length are perfectly discernible.



Original

Paper

The inset of Fig. 7 displays the high temperature part ofE0(T) drawn on a larger scale to show the changes in Young’smoduli due to the breakdown of the hydrogen bondednetwork. One clearly observes an increase of the annealingtemperature due to the presence of MoS2 NPs.

On the other hand, samples that were annealed at about80 8C for at least 2 h prior to measurements, did not showany significant effects of NPs on Tg.

To study the influence of inorganic nanofillers foranother polymer we have performed DMAmeasurements ofthe elastomer JED (empty as reference system) and JEDfilled with 0.1, 0.5, and 1wt.% MoS2/170 (called JED-C1)and MoS2/628 (called JED-C2). Pure JED exhibits a glasstransition around Tg��51 8C. Similar as for JD-2000 wefound an increase of the glass transition for increasingconcentration of MoS2/170 NPs. (Fig. 8) for not annealedsamples. Also here the effect of NPs on Tg vanishes withannealing.

The maximum shift of Tg¼þ7 8C was observed in notannealed samples for 1wt.% of MoS2. DTg�þ3 8C for0.1wt.% and þ4 8C for 0.5wt.% of MoS2. For the samplesfilled with larger nanotubes (JED-C2) no significant shiftin Tg was observed (Fig. 9). In both types of composites(JED-C1 and JED-C2) an increase in E0 (reinforcement)was observed at least at temperatures above Tg (inset ofFigs. 8 and 9). It is also notable that in both composites theannealing temperature for removing the hydrogen bonds isinfluenced by the presence of nanotubes. With increasingconcentration it shifts to higher temperatures by an amountup to 5 8C for 1wt.% nanotubes (inset of Figs. 8 and 9). ForJD-2000 the shift was more than 20 8C (inset of Fig. 7).

Such changes of Tg in nanocomposites are not unusual.It is well established, that many properties of polymernanocomposites depend crucially on the polymer–NPinterface [11, 13]. For example, Bansal et al. [12] haveshown, that the thermomechanical properties of polymernanocomposites are critically affected by the polymer–NPwetting behavior. They used SiO2 NPs with a diameter of

14� 4 nm dispersed in a polysterene matrix. For untreatedsilica surfaces, which are non-wetting to polysterene, theyfound a rather moderate decrease in Tg. For 1wt.% SiO2

they obtain |DTg(1%)|� 1.5K, for 5wt.% the shift isabout 4K. For silica NPs grafted with dense polystyrenebrushes (“hairy” NPs), silica surfaces provide intimatesurface–polymer contact and Tg is shifted to highertemperatures, i.e., up to 4K for 5wt.% of grafted (wetting)NPs.

-100 -80 -60 -40 -20 00,0

0,2

0,4

0,6

0,8

1,0

1,2

Tg2

Tg1 0.1 Hz 0.5 Hz 2 Hz 5 Hz 50 Hzta

nδ

T (°C)

Figure 6 Temperature and frequency dependence of tan d of thepure polyurea elastomer JD-2000. The measurements wereperformed at a heating rate of dT/dt¼ 1 8Cmin�1.

-120 -100 -80 -60 -40 -20 0 20

0

1

2

3

4

5

6

JD-2000

JD-2000/MoS2

f=1 Hz

E' (G

Pa)

T (°C)

0 20 40 60 80 100 1200,000

0,005

0,010

0,015

JD-2000

JD-2000/MoS2

E' (G

Pa)

T (°C)

-90 -80 -70 -60 -50 -40 -30 -20-0,10,00,10,20,30,40,50,60,70,80,91,0

JD-2000

JD-2000/MoS2

E'' (

GP

a)

T (°C)

Figure 7 Storage E0, loss E00 moduli and tan d of pure polyurea JD-2000 compared with the nanocomposite JD-2000/MoS2 (1wt.% ofMoS2). The measurements were performed at the heating ratedT/dt¼ 1.5 8Cmin�1 and frequency of 1Hz.



ph

ysic

a ssp stat

us

solid

i a

The details of the mechanisms giving rise to the changesof Tg measured in polyurea–NP composites are difficult toresolve, since the glass transition mechanism itself is not yetfully understood [16–20]. A possibility to explain theobserved increase of Tg is based on the results of extensivemolecular dynamics simulations of polymers with nano-scopic particles [15]. According to these results the polymerdynamics is affected in the proximity of the NPs. Dependingon the type of interaction between the NPs and the polymers,the chain relaxation dynamics can accelerate (for non-wetting conditions, i.e., repulsive interactions) or slow downconsiderably (for attractive interactions, i.e., wetting or forneutral interfaces).

In the present case, the prevalent polymer–nanotubeinteraction is most probably attractive [28] occurring dueto bonding of the hydrogens with the S-atoms of MoS2 [14](Fig. 10).

Because the size of CRRs increases for T! Tg, thedynamics of polymer chains at certain distances away fromthe NP surface will slow down, which then will reduce thedynamics of the whole sample, rationalizing the observedincrease of the glass transition temperature. This picture isin good agreement with our observations (Fig. 7), showingthat only the glass transition in the soft domains at Tg1 is

influenced by the addition of inorganic NPs, whereas theregions near the hard domains yield a glass transition at Tg2that is independent of the presence of NPs: the moleculesnear the hard domains are already strongly (chemically)attached, so that their dynamics is substantially slowed downand thus Tg2 is not expected to be much influenced by theinteraction with inorganic NPs.

A similar decrease of mobility was recently observed innatural rubber–silica nanocomposites [21]. Even in theabsence of specific polymer–filler interactions, polymersegments within a few nanometers of the filler particlesexhibit relaxation times up to 2–3 orders of magnitudeslower leading to an increase of the glass transitiontemperature of few Kelvin compared to bulk natural rubber.

In PMMA-C60 nanocomposites with NPs up to 6wt.% amaximal increase of Tg of about 4K was detected [22], andwas also explained in terms of slowed down interfacialpolymer/NP regions.

This picture of slowed down interfacial regions nearMoS2 NPs may be complemented by the observation, thatthe annealing temperatures for removing the hydrogen bondsare increased with increasing NP concentration (Figs. 7–9).It suggests that the NPs support the formation ofintermolecular hydrogen bonding between the polymerchains (Scheme 2). Such an increase in cross-links betweenpolyurea chains could explain the observed increase in

-160 -140 -120 -100 -80 -60 -40 -20 0 20 40 60

0123456789

10

JED JED-C2_0.1% JED-C2_0.5% JED-C2_ 1%

0.1 Hz

E' (G

Pa)

T (°C)

0 5 10 15 20 25 30 35 400,0

0,1

0,2

0,3

0,4

0,5

E' (G

Pa)

T (°C)

-160 -140 -120 -100 -80 -60 -40 -20 0 20 40 600,0

0,2

0,4

0,6

0,8

1,0

Tg


0.1 Hz

E'' (

GP

a)

T (°C)

Figure 9 Storage E0 and loss modulus E00 of pure elastomer JEDcompared with JED filled with 0.1, 0.5, and 1wt.% MoS2/628.The measurements were performed at a heating rate ofdT/dt¼ 1.5 8Cmin�1.

-180 -160 -140 -120 -100 -80 -60 -40 -20 0 20 40 600123456789

101112


E' (

GP

a)

T (°C)

0 5 10 15 20 25 30 35 400,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

E' (

GP

a)T (°C)

-180-160-140-120-100 -80 -60 -40 -20 0 20 40 600,0

0,1

0,2

0,3

0,4

0,5Tg

E" (G

Pa)


1 Hz2 Hz5 Hz

T (°C)

Figure 8 Storage E0 and loss modulus E00 of pure elastomer JEDcompared with JED filled with 0.1, 0.5, and 1wt.% MoS2/170.The measurements were performed at the heating rate dT/dt¼ 1 8Cmin�1 with frequencies of 1, 2, and 5Hz.



Original

Paper

tensile strength and would also explain the observed reducedchain mobility in the vicinity of NP induced crosslinkingpoints and the resulting increase in Tg. Similar observationshave been made very recently by Dodiuk et al. [27].The authors have studied the influence of WS2 NPs on thethermo-mechanical and adhesive peel properties of polyure-thane. They also came to the conclusion, that the WS2 NPsmay induce formation of intermolecular hydrogen bondingwithin the polyurethane.

An interesting observation has been made recently byTaschin et al. [6] with transient grating experiments ofpolyurea JD-2000 filled with �1wt.% MoS2 nanotubes.An intermediate dynamics was clearly detected at roomtemperature at the time scale of t� 5� 10�5 s. Since thissignal was absent in the pure samples, the authors argued thatthe effect is coming from the interfacial polymer chains inthe vicinity of the nanotube surfaces. Taking into accountthat the relaxation time ta of the alpha-relaxation in purepolyurea [21] is of the order of ta� 10�7 s at roomtemperature, one could speculate that the observedintermediate dynamics becomes visible in the filled samplesdue to an increase of local relaxation times from 10�7 s in thebulk to 10�5 s close to the NPs. This explanation is ingood agreement with our findings. However, to drawmore specific conclusions detailed investigations, like, e.g.,temperature dependent transient grating experiments or

local dielectric spectroscopy measurements [26] have to beperformed on polyurea/MoS2 nanocomposites.

4 Conclusions We have presented extensive DMAmeasurements of pure polyurea elastomers and nano-composites of polyurea filled with various concentrationsof inorganic nanotubes MoS2. An upshift of Tg of several 8Cwas observed together with a slight increase of the Young’smoduli of the composites. The increase of Tg is explainedin terms of slowing down of local polymer chain dynamics,i.e., for polymer chains in the vicinity of the nanotubes. Aclear cut analytic theory concerning such a suppression ofdynamics is still missing. However, molecular dynamicssimulations [15, 25] have shown a gradual slowing down ofthe polymer dynamics approaching the surface of NPs, thatdo interact with the polymer matrix. In the present case,such an interaction between polyurea chains and MoS2 NPscould be mediated via hydrogen bonding. Indeed, recentcomputer simulations [23] have shown that hydrogen mayform stable chemical bonds with some low Miller-indexededges of MoS2. First principles calculations on a MoS2monolayer [28] yielded that hydrogen atoms prefer to bondto S atoms with a length of 1.41Å and a formation energyof 1.91 eV. Although the detailed local geometry of thepolymer–nanotube interface cannot be resolved at themoment, it seems clear, that the observed slowing downcomes from such an interaction between the hydrogen atomsand sulfur atoms of MoS2 nanotubes. Also our observation,that for annealed samples (hydrogen bonds are removed) theeffect of nanotubes on Tg vanishes, is in accordance withsuch a picture.

As we have shown here, in the absence of permanentattachments of the polymer chains to the MoS2 NPs thesystem yields a single, but larger average Tg, due to theincrease of the longest relaxation time. On the other hand,polyurea opens a unique possibility to study also the effect ofvery strong NP/chain enthalpic interactions. Due to its phaseseparated nanostructure into so called hard and soft domains(Fig. 1), two glass transitions appear which are related tocooperative motions of molecular segments in the softdomains (Tg1) and to regions near the hard nanodomains(Tg2), respectively. Due to the strong attractive NP–polymerchain interactions the polymer chain motions near the hardnanodomains are considerably slowed down and as a resultTg2 turns out to be much larger than Tg1.

More detailed results on the corresponding two glasstransition behavior (Fig. 3) in polyureas with variousmolecular weights of the polyetheramines will be presentedin a forthcoming paper.

Acknowledgements The present work was performed inthe frame of the COST Action MP0902 (COINAPO – Compositesof Inorganic Nanotubes and Polymers). Financial support bythe Austrian Science Fund (FWF) P23982-N20 is gratefullyacknowledged. Support from Prof. R. Mezzenga (member ofthe COINAPO Swiss team) is greatly appreciated. We sincerelythank S. Puchegger (Faculty Center for Nanostructure Research,University of Vienna) for providing the SEM images.

Figure 10 Top: Polymer chain relaxation time t(r) as a function ofdistance r from the nanoparticle surface, at various temperaturesapproaching Tg. The arrow indicates the spatial increase of theslowed down regions due to the increase of correlation length withT!Tg. Bottom: Schematic illustration of the bonding between thehydrogen atoms of polyurea chains and the sulfur atoms of MoS2nanotubes as proposed in Ref. [28].



ph

ysic

a ssp stat

us

solid

i a

References

[1] T. Hanemann and D. V. Szabó, Materials 3, 3468 (2010).[2] A. C. Balazs, T. Emrick, and T. P. Russell, Science 314, 1107

(2006).[3] A. Sánchez-Ferrer, D. Rogez, and P. Martinoty, Macromol.

Chem. Phys. 211, 1712 (2010).[4] A. Sánchez-Ferrer, R. Mezzenga, and H. Dietsch, Macromol.

Chem. Phys. 212, 627 (2011).[5] M. Remskar, A.Mrzel, Z. Skraba, A. Jesih,M. Ceh, J. Demsar,

P. Stadelmann, F. Lévy, and D. Mihailovic, Science 292, 479(2001).

[6] A. Taschin, P. Bartolini, A. Sánchez-Ferrer, R. Mezzenga,A. Mrzel, and R. Torre, Mater. Sci. Forum 714, 79 (2012).

[7] W. Schranz, Phase Transitions 64, 103 (1997).[8] A. Taschin, P. Bartolini, A. Sánchez-Ferrer, R. Mezzenga,

A. Mrzel, and R. Torre, Sens. Transducers 12, 46 (2011).[9] A. Fuith, M. Reinecker, A. Sánchez-Ferrer, R. Mezzenga,

A. Mrzel, M. Knite, I. Aulika, M. Dunce, and W. Schranz,Sens. Transducers 12, 71 (2011).

[10] E. A. Turi, Thermal Characterization of Polymeric MaterialsVol. I, 2nd ed. (Academic Press, New York, 1997), p. 980.

[11] H. Oh and P. F. Green, Nature Mater. 8, 139 (2009).[12] A. Bansal, H. Yang, C. Li, B. C. Benicewicz, S. K. Kumar,

and L. S. Schadler, J. Polym. Sci. B 44, 2944 (2006).[13] S. Srivastava and J. K. Basu, Phys. Rev. Lett. 98, 165701

(2007).[14] G. Seifert, H. Terrones, M. Terrones, G. Jungnickel, and

T. Frauenheim, Phys. Rev. Lett. 85, 146 (2000).

[15] F. W. Starr, T. B. Schrøder, and S. Glotzer, Macromolecules35, 4481 (2002).

[16] G. Adam and J. H. Gibbs, J. Chem. Phys. 43, 139(1965).

[17] E. Hempel, G. Hempel, A. Hensel, C. Schick, and E. Donth,J. Phys. Chem. B 104, 2460 (2000).

[18] H. Huth, M. Beiner, and E. Donth, Phys. Rev. B 61, 15092(2000).

[19] B. M. Erwin and R. H. Colby, J. Non-Cryst. Solids 307–310,225 (2002).

[20] J. Koppensteiner, W. Schranz, and M. A. Carpenter, Phys.Rev. B 81, 024202 (2010).

[21] D. Fragiadakis, L. Bokobza, and P. Pissis, Polymer 52, 3175(2011).

[22] J. M. Kropka, K. W. Putz, V. Pryamitsyn, V. Ganesan, andP. F. Green, Macromolecules 40, 5424 (2007).

[23] M. V. Bollinger, K. W. Jacobsen, and J. K. Norskov, Phys.Rev. B 67, 085410 (2003).

[24] G. Tsagaropoulos and A. Eisenberg, Macromolecules 28, 396(1995).

[25] G. D. Smith, D. Bedrov, L. Li, and O. Byutner, J. Chem. Phys.117, 9478 (2002).

[26] M. Labardi and D. Prevosto, J. Vac. Sci. Technol. B 28,C4D11 (2010).

[27] H. Dodiuk, O. Kariv, S. Kenig, and R. Tenne, J. AdhesionSci. Technol., DOI: 10.1080/01694243.2013.819554(2013).

[28] E. W. K. Koh, C. H. Chiu, Y. K. Lim, Y.-W. Zhang, andH. Pan, Int. J. Hydrogen Energy 37, 14323 (2012).



Original

Paper

composites of inorganic nanotubes and polymersmacromolchem.com/papers/paper43.pdfpart of topical...

Documents