Could Nanoparticles Reinforce Polymer Matrices?

doi:10.4236/msce.2013.15011

Paper Menu >>

Journal Menu >>

Journal of Materials Science and Chemical Engineering, 2013, 1, 53-56

http://dx.doi.org/10.4236/msce.2013.15011 Published Online October 2013 (http://www.scirp.org/journal/msce)

Could Nanoparticles Reinforce Polymer Matrices?

G. Schrodj, M.-P. Hirn, B. Haidar*

Institut de Science de Materiaux de Mulhouse, CNRS-UMR 7361, Mulhouse, France

Email: *[email protected]

Received August 2013

ABSTRACT

In this work, we challenge the idea that “introducing nanoparticles in polymer matrices enhance properties” which is

assumed spreading almost dogmatically. Two series of compounds were prepared: one based on elastomers (solu-

tion-SBR) filled with conventional carbon black particles, CB, the other used the same polymer but filled with carbon

nanotubes, CNT. The results of two types of experiments were compared on the two series. The first is physical based

on the non linear response of filled materials to static deformation, th e second physicochemical, stands on calorimetric

measurements of the polymer heat of adsorption on the solid. Static deformation effect on dynamic mechanical modulus

shows the behavior of the CB filled elas tomers was qualitatively identical to that of glassy polymer reflecting the exi s-

tence of an immobilized fraction of the polymer at the intephase, while that of CNT was found identical to unfilled p o-

lymer indicating a poor filler-polymer interactions. Polymer adsorption measured by flow micro calorimeter showed a

substantial amount of heat exchange on the surface of CB while no heat of adsorption was detected on CNT. The lack

of interaction between the polymer and the CNT, except for a small domain of a narrow polymer molecular weight,

prevent any enhancements of mechanical properties. Other applications may be improved.

Keywords: Polymer; Nanoparticle; Dynamic Modulus; S tatic Deformation; Adsorption; Calorimeter; Interphase

1. Introduction

Expectations of nanoparticles benefits in polymer based

materials rose out of proportion, as it is very often the

case when a new technology immerges. All kind of for-

tunes, and miss fortunes, were attributed to the nano-

aspect of this kind of particles: reinforcing and tough-

ening effects; electrical, thermal and acoustical virtues;

transport properties optical and aging enhancements... [1]

However, when useful real products were late to the

market or appeared less exciting than what it had been

announced, expectations fell as rapidly as it had des-

cended to an overwhelming low level. Only then, thoughts

go back to basics: chemistry, physics and polymer sci-

ence. We most likely are for the time being at this stage

of nanoparticles development in polymer science.

For instance, the driving force behind any effect of

solid particles on the behavior of polymer matrix is the

way polymer chains get in contact with the solid surface.

Adsorption, affinity, interactions, immobilization, inter-

phase buildup… are only a few of many concepts devel-

oped in the literature to emphasize the importance of

polymer/filler contact-adjustment on the properties and

behaviors of the compound.

In this work, we tempted to compare the effect if par-

ticles size-scale on two of the polymers prominent prop-

erties: Dynamic-mechanical and physicochemical.

2. Nonlinear Behavior of Filled Polymer as

Detected by Dynamic Mechanical

Behavior

Nonlinear response of polymer material do deformation

can be demonstrated by different means. One of the most

elegant is that of TL. Smith [2] which consists of apply-

ing at a time, t0, à constant macroscopic strain, λ, on the

polymer specimen (as in relaxation test) and superim-

poses every now and th en on λ a dynamic deformation of

constant frequency, f, and amplitude, ε, small enough be

considered as affiliated to the linear defor ma tion.

2.1. Behavior of Glassy Polymers

In the case of glassy polymers such as polycarbonate,

Poly (methyl methacrylate), polystyrene… dynamic mod-

ulus, e.g. E’ schematically presented in Fig u r e 1, de-

creases suddenly upon deformation from its original val-

ue, E’0, to an undetectable level. Results show linear

modulus-time dependency in a log-log scale, the slope of

which is µ. This behavior was attributed to the nonequi-

librium stat of glassy polymers, the so called “physical

aging” phenomenon and its consequences as a decrease

of the modulus (increase of the polymer segmental mo-

bility, “des-aging”) upon deformation thereafter, and

*Corresponding a uthor.

G. SCHRODJ ET AL.

Figure 1. Schematic presentation of the dynamic mechani-

cal experimen t .

under relaxation conditions, a continues increase of the

modulus because of the decrease of segmental mobility,

“aging” anew. µ therefore is considered as an assessment

of the physical aging rate.

2.2. Behavior of Elastomers

A solution-SBR specimen was crosslinked using 2 phr of

dicumil peroxide. The effect of a static deformation, λ =

l/l0 = 1.03 on dynamic modulus when tested at −100˚C, T

< Tg, de-aging followed by aging processes were clearly

observed, Figure 2. In contrast to the glassy state, the

high segmental mobility of the polymer at the rubbery

state places it in a quasi-thermodynamic equilibrium.

Therefore measurements made at 30˚C under similar

relaxation conditions, the modulus E’ of the same cros-

slinked elastomers is virtually unaffected by a static de-

formation, at least as long as such a deformation remains

moderate as shown also in Figure 2.

2.3. Behavior of Polymer Filled with

Conventional Particles

The same solution-SBR was filled with 45 phr carbon

black N-330 before crosslinking.

CB filled SBR showed a clear de-aging followed by ag-

ing processes, just as the unfilled po lymer behaves at the

glassy state, Figur e 2. This behavior was attributed to the

presence, in filled rubber, of a rubber/filler interfacial

region with a specific behavior [3]. The polymer within

this “interphase” has a low segmental mobility compared

to the free rubber matrix and present, therefore, a glassy

like behavior. Such interphase results most likely form

the interactions that polymer most exchange with the

solid surface. Such a specifically low segmental behavior

of the rubber at the filler interface was already detected

especially by 1H solid NMR mea s ure ments [4] .

2.4. Behavior of Polymer Filled with

Nanoparticles

When filler particles are replaced in SBR based matrix

Figure 2. E’ before, E’0, and after static deformation for

differe nt compounds précis ed on the graph.

by nanoparticles (carbon nanotubes, diameter 25 - 12 nm,

1 - 2 µ length), while all other conditions are kept iden-

tical to the ones adopted for conventional filler, we ob-

tain the same behavior as for unfilled polymer, Figure 3.

This may be a strong indication that no interphase is

created at the nano tubes/polymer interface which pre-

sumably due to a lack of interactions exchange between

the two actors. Under these circumstances one would not

expect a substantial “reinforcing” effect beyond the hy-

drodynamic resulting from the filler addition which is

noticed by the increase of the modulus compare Figures

2 and 3.

3. Calorimetric Evaluation of Polymer-Filler

Interactions

Previous results stress the need for a direct measurement

of filler-polymer interactions. Calorimetric approach

using a flow calorimeter appeared as well adapted to

achieve this goal.

A Flow Micro Calorimeter (FMC) was used to meas-

ure the polymer solution heat of adsorption on the filler.

It operated under a constant flow rate (3.3 ml/h) of carri-

er solvent, THF, at 25˚C. In a pulse mode, a known

amount of the polymer dissolved in the carrier solvent

was injected in the flow path through a calibrated sample

loop. If any thermal exchange occurs when the solute

E’

time

G. SCHRODJ ET AL.

comes into contact with the particles being studied, a

peak was registered (positive or negative corresponding

to exo- or endothermic events respectively, as schema-

tized in Fig u r e 4). The unadsorbed or/and desorbed frac-

tion of the solute, if any, was then detected by a down-

stream detector. The polymer concentration was fixed at

10 g/l. The FMC principles and the experimental setup

were described elsewhere [40-43].

3.1. Heat of SSBR Adsorption on CNT and CB

The peaks of adsorption of a solution of SSBR of CB and

CNT areas presented in Figur e 5.

Figure 3. Effect of static deformation on E’ for CNT filled

SSBR.

Figure 1. Schematic presentation of FMC.

Figure 5. FMC thermogrammes of SSBR solution on CB

and CNT.

In the case of CB we observe an exothermic peak in-

dicating an irreversible adsorption resulting from a clear

interaction exchanged between the polymer and the solid

surface, while in the case of CNT, although a similar

exothermic peak appears first, it is followed right after by

an equivalent endothermic one reflecting reversible ad-

sorption or two successive processes of adsorption and

desorption; the final balance of the two processes is quite

close to zero while for CB the balance is largely exo-

thermic. The estimation of the molar heat of adsorption

on CB gives a value in the range of chemical bounding,

several hundreds of kJ/mol which may explain the irre-

versible aspect of the adsorption in this case.

3.2. The Effect of the Molecular Weight of

Polybutadien on Its Heat of Adsorption on

CNT

The absence of a permanent adsorption on CNT in con-

trast with CB is unexpected and requires more investiga-

tion. The most important difference between CB and

CNT is their forms: aspect ratio and size scale; although

chemical defects and edges of graphite plan are more

prominent in CB than in CNT, both solid surfaces are

essentially formed out graphite plans. Under theses cir-

cumstances it was legitimate to question the reticence of

CNT to adsorb polymer on the base of scale size match-

ing of the two substances. In fact, the size of a CB ag-

gregate is quite huge compared to CNT diameter, which

approaches the radius of gyration of a long polymer

chain itself. The behavior of the polymer chain toward

the two kinds of surfaces might be quite different.

Therefore, we synthesized a series of polybutadienes

with identical micro-structure (1,2-PB content equal to

about 80%, same Tg), various molecular weights (from

thousand to several hundred-thousand g/mol) and a very

narrow molecular weight distribution (Ip ≈ 1.1). Results

obtained by the application of FMC experimental ap-

proach on CNT are show in Figure 6.

Figure 6. Molar heat of adsorption as a function of molecu-

lar weight of polybutadiene.

1.13

1.15

1.17

1.19

1.21

-1 135

log E', MPa

aging time, log s

E'0

E' carbon nanotubes filled SBR at 30°C

measurement cell

4 thermistors

solutefluid

sa mple

thermogramme

refractogramme

050

Time (min)

050

Time (min)

CNT

300

600

900

100100010000 100000 1000000

Heat of Adsorption, kJ/mol

MW, g/mol

G. SCHRODJ ET AL.

It is clear that for linear amorphous polymer adsorp-

tion takes place exclusively in a relatively small window

of MW ’s. Adsorption, within this window, is ir reversible

and associated with substantial amount of heat, whereas

outside its limits adsorption is reversible and heat ex-

change is low. Low molecular weight polymer and oli-

gomers are, expectedly, not adsorbed because of their too

small size which unables them to create a sufficient

number of contact to ensure permanent adhesion; heat of

adsorption in this case is too low to overcome thermal

agitation, kT.

High MW’s molecules do not adsorb either, presumably,

because of their failure under our experimental condi-

tions to wind around the nanotube, contradicting model-

ing predictions [5] or, most likely, because its incapacity

to yield a sufficient number of adsorption contact-points

to guarantee permanent adhesion. In order to do so a po-

lymer chain should undergo a defolding process of its

coiled conformation all along the tube axis. Such process

is highly disadvantaged from an entropic viewpoint.

Therefore, only an intermediate MW’s chain can meet

the compromise of being short enough to defold over the

tube surface without too much entropic penalty but still

able to create enough contact-points to offer an enthalpic

gain high enough to link it permanently to the surface of

the nanotube.

4. Conclusions

As many other new technologies, carbon nanotubes

unique structure and characters have attracted excessive-

ly so much attention so fast that they can hardly live up

to all our earliest expectations namely in polymer appli-

cations. The main reason hides in the fact that the way a

long macromolecule come within reach of nano-sized

solid differs fundamentally from the way it approaches

conventional, relatively “unlimited” surface.

This is clearly demonstrated by the absence of adsorp-

tion of polybutadienes on CNT except through MW’s

narrow window in which the size of the macromolecule

matches the size of the exposed surface. The conse-

quence of such a lack in the establishment of the neces-

sary interactions is to restrain adsorption and therefore,

exclude the formation of a valuable interphase, which is

responsible for the emergence of most of the reinforce-

ment observable facts. This is evidenced by the non- ef-

fect of static deformation on dynamic modulus for CNT

filled SSBR while CB filled one show first, a decrease

then, an increase of the dynamic modulus upon deforma-

tion. The immobilized nature of the interphase is in this

case presumably responsible for such a nonlinear deag-

ing/physical aging behaviors.

REFERENCES

[1] K. Friedrich, S. Fakirov and Z. Zhang, “Poly mer Compo-

sites: From Nano- to Macro-Scale,” Springer, 2005.

[2] B. Haida and T. L. Smith, “Dependance of Segmental

Mobility in Polycarbonate on Time and Deformation,”

Polymer, Vol. 31, 1990, p. 1904.

[3] B. Haida, H. Deradji, A. Vidal and E. Papirer, “Polymers

and Surfaces. A Versatil Combination in Recent Research

Developments in Polymer Science,” In: H. Hommell, Ed.,

(Research Signpost, Trivandrum, India), 1990.

[4] V McBrierty and K. Packer, “Nuclear Magnetic Reso-

nence in Solid Polymers,” Cambridge, 1993.

http://dx.doi.org/10.1017/CBO9780511525278

[5] I. Gurevitch and S. Srebnik, “Conformational Behavior of

Polymers Adsorbed on Wrapping of Nanotubes,” Chemi-

cal Physics Letters, Vol. 444, 2007, p. 96.

http://dx.doi.org/10.1016/j.cplett.2007.06.112