Entrapped Styrene Butadiene Polymer Chains by ... - ACS Publications

Sven Wießnera,b, Ulrich Schelera, Kay Saalwächterc,. Petr Formaneka, Gert Heinricha,d, Amit Dasa,e* a. Leibniz-Institut für Polymerforschung, D...
0 downloads 0 Views 5MB Size
Subscriber access provided by READING UNIV

Article

Entrapped Styrene Butadiene Polymer Chains by Sol-Gel Derived Silica Nanoparticles with Hierarchical Raspberry Structures Sankar Raman Vaikuntam, Klaus Werner Stöckelhuber, Eshwaran Subramani Bhagavatheswaran, Sven Wiessner, Ulrich Scheler, Kay Saalwaechter, Petr Formanek, Gert Heinrich, and Amit Das J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.7b11792 • Publication Date (Web): 19 Jan 2018 Downloaded from http://pubs.acs.org on January 19, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Entrapped Styrene Butadiene Polymer Chains by Sol-Gel Derived Silica Nanoparticles with Hierarchical Raspberry Structures Sankar Raman Vaikuntama,b, Klaus Werner Stöckelhubera Eshwaran Subramani Bhagavatheswarana,b, Sven Wießnera,b, Ulrich Schelera, Kay Saalwächterc, Petr Formaneka, Gert Heinricha,d, Amit Dasa,e* a

Leibniz-Institut für Polymerforschung, Dresden e.V., Hohe Strasse 6, 01069 Dresden, Germany b Technische Universität Dresden, Institut für Werkstoffwissenschaft, 01062 Dresden, Germany c Institut für Physik, Martin-Luther-Universität Halle-Wittenberg, 06099 Halle (Saale), Germany d Technische Universität Dresden, Institut für Textilmaschinen und Textile Hochleistungswerkstofftechnik, D-01069 Dresden, Germany e Tampere University of Technology, Korkeakoulunkatu 16, 33101 Tampere, Finland * Corresponding author email: [email protected]

ABSTRACT A sol-gel transformation of liquid silica precursor to solid silica particles was carried out in a one pot synthesis way where a solution of styrene butadiene elastomer (S-SBR) was present. The composites, thus produced, offered remarkable improvements of mechanical and dynamic mechanical performances as compared with precipitated silica. The morphological analysis reveals that the alkoxy-based silica particles resemble a raspberry structure when the synthesis of the silica was carried out in presence of polymer molecules and represent a much more open silica-network structure. However, in absence of polymer the morphology of the silica particles is found to be different. It is envisaged that the special morphology of the insitu synthesized silica particles contributes to the superior reinforcement effects which are associated with a strong silica-rubber interaction by rubber chains trapped inside the raspberry-like silica aggregates. Therefore, the interfaces are characterized in detail by lowfield solid-state 1H NMR spectroscopy,

29

Si solid-state NMR spectroscopy and energy-

dispersive x-ray spectroscopy (EDX). Low-field 1H NMR based double-quantum (DQ) experiments provide a quantitative information about the crosslink density of the silica filled rubber composites and about the influence of silane coupling agent on the chemical crosslink density of the network and correlates well with equilibrium swelling measurements. The special microstructure of the alkoxy-based silica was found to be associated with the interaction between alkoxy-based silica and rubber chains as a consequence of particle growth in presence of rubber chains.

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 31

INTRODUCTION The use of inorganic fillers in polymer matrix has become an important field in materials research1-2. Very recently, the polymer composites consisting with various nanostructured inorganic fillers have been drawn a considerable interest because of their unprecedented mechanical, thermal, and other functional properties3-5. Silica is one of such inorganic materials what has received a great impact in different industrial and engineering field. For example, precipitated silica is now considered one of the most potential fillers for modern tire manufacturing industries as well as becoming a material of commercial and technological importance. Particularly, implementation of saline coupling agent in the silicapolymer system enables tire engineers to develop vehicle tires with enhance grip, reduce fuel consumption and decrease wear to an extent6. These silica particles which are prepared by wet method contain a large number of hydroxyl groups (silanols) on the surface. Due to interparticle hydrogen bonding between two hydroxyl groups, the particles always form aggregates which inhibit the dispersion process in polymer matrix. Simultaneously, by the exploitation of these functional groups with different chemicals the surface properties of the silica particles can be tailored and the dispersion of the particles can be improved to some extent7-8. Nevertheless, the optimum performance of silica based polymer composites can be achieved if the silica particles are uniformly dispersed in the polymer matrix and a strong chemical/ interaction between silica and polymer chains are established. In order to get rid of silica agglomerations and to have a strong rubber filler interaction, in principle, the silica rubber composites can be prepared three different ways, i.e., i) direct blending, ii) in-situ polymerization and iii) sol-gel process 9-14. i) Direct blending: In order to get a homogenous and coherent substance, polymers are melted along with silica at relatively higher temperature and under strong shearing force generated by special kind of machinery like extruder, internal mixer and two-roll mixing mill the silica agglomerates/aggregates are broken into submicron size. Particularly, in rubber system as the rubber has no melting temperature, the process can be ascribed as solid state mixing. At elevated temperature the viscosity is reduced and the efficiency of the silica mixing can be improved in a considerable extent15. However, all this mechanical mixing approaches is not sufficient to exploit the reinforcing character of silica. The use of certain kind of saline coupling agent in the preparation of silica/polymer composites, mostly followed by high temperature mixing process (reactive mixing), partially can solve the above issues but, till date, the industrial applications experienced a strong demand to further enhance the dispersion quality of the silica and to establish a substantial coupling between silica particles 2 ACS Paragon Plus Environment

Page 3 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

with polymer chains16-117. In principle, precipitated silica can be mixed with any kind of elastomers, however, solution polymerized styrene butadiene rubber (SSBR) is found to be suitable with silica when the final mechanical performance and other technical properties are concerned18. Extensive efforts are now given to the natural rubber/silica system to overcome the dispersion problem and to increase rubber-filler interaction by melt mixing technique. The presence of organic protein in natural rubber is supposed to be the main reason behind the poor chemical coupling between the natural polyisoprene chains and silica surface19. ii) In-situ polymerization: Therefore, apart from conventional melt (solid state) mixing process, solution based methods are also being developed. To address these solution methods several reports can be found where hetero-coagulation/hetero-flocculation of the colloid polymer solutions are often followed. To get uniform dispersion of fumed and precipitated silica natural rubber in latex form can be used. Studies also have been done to further assist the dispersion process using ultrasonic bath or agitator bead mill20. Unfortunately, the use of organic solvents, evaporation of this solvent and a complete removal of the solvent molecules are always criticized21-234. In-situ polymerization is also well practiced methods where polymers are synthesized in presence of silica particles24-25. Mostly, in this method the dispersion of hydrophobic silica can be done by mini-emulsification followed by polymerization of the monomer. The mini-emulsion method also further was explored by introducing silica precursor tetra-ethylorthosilicate (TEOS) in the reaction medium which is known as double in situ mini-emulsion polymerization26. A lot of literature could be found where this method is utilized for the preparation of various silica polymer composites and here a detail discussion on this point is out of scope27. iii) In-situ sol-gel silica: Considering the above issues, alkoxy-based silica prepared from a sol-gel process gained also considerable attention

12, 28-33

. Reports can be found that

sol-gel derived alkoxy-based silica offers excellent reinforcement, good dispersion as compared with commercial precipitated silica34-38. The superior mechanical properties of solgel silica are well-known as compared with precipitated silica under similar filler loadings29, 39-41

. It is reported that the silica composites prepared by in-situ sol-gel method comprise with

uniformly distributed single silica nano-particles and these generated particles are free from filler-filler network. Furthermore, the stronger rubber silica interactions are also explained by ‘trapped entanglements’ which results from a bonding of rubber chains with silica surface. In a recent report the higher reinforcing character of sol-gel silica has been explained by the presence of ‘hairy’ silica particles which are formed by a chemical attachment of amine used for sol-gel transformation on the surface of alkoxy silica42. Some of the publications also 3 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

explained the higher reinforcing character of sol-gel silica because of less number of silanol groups resulting a decrease of the hydrophilic character of the silica43-44. Though a lot of investigations were carried out to understand the reinforcing mechanism of in-situ silica, there is still a general lack of information regarding the elucidation of reinforcement process by alkoxy silica in details. On the other hand, as far as the structure of silica particles is concerned in the aspect of silicapolymer hybrid materials, the raspberry-like hierarchal structure is known to be responsible for the surface properties like hydrophobicity etc. of the hybrid materials45. The sol-gel derived silica particles were found to form raspberry-like structure when poly(styrene-cobutyl acrylate) was functionalized with poly-acrylic acid46. The raspberry-like structure of silica particles while using different colloidal polymer solution are well reported in several publications; however a limited number of papers can be found about this kind of structure in rubber based silica composites47-50. In natural rubber latex, when treated with some kind of ions, it was found that silica also can go through a hetero-aggregation process resulting in raspberry-like silica aggregates. The preferential raspberry-like structure was formed when the silica size was d ≈ 30 nm; but when the d ≈100 nm, a dendritic-like structure was favored51. In this paper we report the presence of raspberry like silica particles when the silica was synthesized in colloidal solution of styrene butadiene rubber. Most importantly, we thoroughly discuss how these kind of raspberry-like silica particles contribute to the polymer chain being structural arrest. The extraordinary structure formation of silica particles and possible tethered polymer by the silica particles are explained Low-field 1H double-quantum NMR experiments, Solid state 29Si-NMR experiments, energy dispersive X-ray spectroscopy (EDX) and different microscopic techniques.

EXPERIMENTAL Materials Solution-grade Styrene Butadiene rubber (SSBR) BUNA VSL 2525-0 HM containing 25% vinyl, 25% styrene, the vulcanizing accelerators N-cyclohexyl,2-benzothioazolesulfonamide (CBS-Vulkacit CZ) and diphenylguanidine (DPG) were procured from Lanxess chemical Ltd, Germany. Zinc oxide, stearic acid and sulfur were purchased from Acros Organics, Germany. The silica precursor used here is tetraethoxyorthosilicate (TEOS) was purchased from Sigma Aldrich with a purity of 99 %. Tetrahydrofuran (THF) and n-butyl amine were also supplied 4 ACS Paragon Plus Environment

Page 5 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

by Sigma Aldrich as laboratory reagent grade. Precipitated silica (Ultrasil-VN3 GR) was kindly provided by Evonik Industries AG, Wesseling, Germany, with a density of 2.2 g/cc and BET surface of 175 m2g-1. The silane coupling agents, diethyl dimethoxy silane (DMS), (3-aminopropyl)triethoxysilane

(APTES),

n-octodecyltriethoxsilane

(ODTES),

(3-

mercaptopropyl)trimethoxysilane (MPTES), bis[3-(triethoxysilyl)propyl]disulfide (TESPD) and bis[3-(triethoxysilyl)propyl]tetrasulfide (TESPT) were also kindly supplied by Evonik Industries AG (Essen, Germany) having a purity of 99 %. 3-Octanoylthio 1-propyl triethoxysilane (NXT) was kindly provided by Momentive Performance materials GmbH, Leverkusen, Germany. The chemical structures of the used silane coupling agents are shown in Scheme 1. Sample preparation The alkoxy-based silica (here after abSiO2) composites were prepared by hydrolysis of TEOS using water in presence of n-butylamine as a catalyst. The reaction was carried out in a tetrahydrofuran (THF) solvent medium where SSBR was previously dissolved. The entire mixture was like an emulsion and it was agitated by mechanical stirrer at 60ºC. For filler characterization studies, the abSiO2 powder was separated from the rubber solution by the repeated solvent treatment process and subsequent centrifuging technique. It is expected by this process that only loosely bounded rubber chains will be washed away leaving chemically anchoring of these chains on the surface of silica particles. To prepare the vulcanized rubber composites, the abSiO2/SSBR masterbatch was compounded with other rubber ingredients. The compounding was carried out in a Haake internal mixer at 110°C and 80 rpm for 6 min. Initially, the master batch was added into the internal mixer and simultaneously silane coupling agents (1 parts per hundred rubber (phr) of silane used for 10 phr of silica), zinc oxide (3 phr) and stearic acid (2 phr) were added. The mixed rubber compounds are dumped at 140 to 150 °C. Finally the vulcanizing chemicals CBS (1.4 phr), DPG (1.7 phr) and sulfur (1.4 phr) are added in a two roll mill at 50 ºC for 10 min with 1:1.2 friction ratio. For comparison studies, composites with commercial precipitated silica (Ultrasil-VN3 GR, here after pSiO2) were prepared and the same amount of ingredients was used, following the same addition and mixing sequence. But the mixing time in this case was 10 min. The prepared final rubber compounds of both abSiO2 and commercial silica were allowed to mature for overnight. The matured rubber compounds were subjected to the rheometric study to estimate the optimum cure time of rubber compounds by using a rubber process analyzer (Scarabaeus

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

SIS-V50, Wetzlar, Germany). The rubber compounds were molded into 2 mm thick sheets by using a hydraulic compression molding press. Sample abbreviations used The samples are abbreviated based on the type of silica and followed by the type of silane coupling agents. e.g.: i-30_DMS, x-30_TESPT, etc. ‘i’ and ‘x’ denote the abSiO2 (alkoxybased silica) and pSiO2 (precipitated silica) respectively. The numbers after ‘i’ and ‘x’ refers to the amount of silica presented in rubber (in phr). The abbreviated text for different silane coupling agents and their chemical structures are mentioned below. Characterization of silica particles: Particle size distribution analysis The average silica particle size and its distribution was measured by dynamic light scattering analyzer, Dyna Pro-nanostar (Wyatt instruments, California, USA) with a measurement size range from 0.2 to 2500 nm using 785 nm laser wave length. The experiment was carried out at 25 ºC with a dispersion of silica powders in THF. The light scattering measurements were performed for 3 samples with minimum of 25 scans for each sample.

Scheme 1: Chemical structure of the silane coupling agents used in the study 6 ACS Paragon Plus Environment

Page 7 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Equilibrium swelling method The crosslink density of gum and silica-filled SSBR composites were investigated by the standard equilibrium swelling method. Samples of defined geometry were cut from the vulcanized rubber sheet of 2 mm thickness, weighed and let to swell in toluene (molar volume (Vs) = 106.2 ml/mol, density (ρs) = 0.87 g/cm3) for 72 hours at room temperature (25 ºC) in closed glass vials and kept in dark place to prevent the photo-degradation effect. The samples were then taken out, wiped to remove the excess solvent adsorbed on the rubber surface and weighed in a high precision balance. Finally, the swollen samples were dried at 60 ºC until constant weight is achieved. The initial weight, swollen weight and final dried weight of samples were used to calculate the crosslink density of gum and silica filled rubber composites. The volume fraction of rubber (vr) is determined52 by Eq. 1

υr =

wd − f ins wi ρ r wd − f ins wi ρ r + w0 ρ s

(1)

where, wi, ws and wd are the initial weight of samples before swelling, swollen and dried samples after swelling respectively. w0 is the equilibrium weight of solvent absorbed by the samples. w0 = ws-wd. fins is the weight fraction of insoluble substances in rubber like silica and zinc oxide. The classical Flory-Rehner equation (Eq. 2) is utilized to calculate the crosslinking density (vFR = 1/Mc) of silica filled vulcanized rubber. The modified Flory-Rehner equation with phantom model approximation was used to define the swollen crosslinked rubber networks53. ln(1 − υ r ) + υ r + χυr2 = −

ρr

 2 1 Vs 1 − υ r 3 Μc  f

(2)

where ρr is the density of rubber compound, Vs is the molar volume of toluene, χ is the FloryHuggins interaction parameter (0.413 for SBR-toluene system54 and f is the functionality of crosslink (for phantom model network the functionality is assumed as tetra functional crosslinks, hence f = 4). Low-field 1H double-quantum NMR experiments

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 31

Double-quantum (DQ) NMR experiments were carried out to probe the effect of the incorporation of filler and silane coupling agent on the crosslink density of the rubber. DQ measurements were performed at 110 oC on a Bruker minispec mq20 (Bruker Biospin GmbH, Rheinstetten, Germany) instrument with an operating frequency of about 20 MHz (0.5 Tesla). The 90o and 180o pulse lengths were 3.1 and 5.8 µs, respectively, and the dead time was 15 µs. DQ NMR provides a measure for the crosslink density of vulcanized rubbers and rubber composites by detecting the weak residual dipolar couplings55 existing in polymer networks. The presence of topological constraints (crosslinks and entanglements) in rubber increases the degree of anisotropy of fast segmental motions of rubber chain segments. This leads to the appearance of residual dipole-dipole couplings between the protons in monomeric units56.

The residual dipolar coupling ( ) is proportional to the crosslink density of the

vulcanized network. DQ-NMR measurements generate two main signal functions (Fig. 1). A reference intensity decay curve (Iref) and a DQ build-up curve (IDQ). The sum of both (Iref +

IDQ) reflects the relaxation of the whole magnetization of the entire sample, which means the contributions of dipolar-coupled network chain segments and uncoupled network defects (like sol-gel fraction, dangling chain ends and loops). Networks exhibit faster non-exponential relaxation as compared to more mobile defects, which show slower exponential decay57. To correct the raw IDQ build-up data for the long-time relaxation effect, the data is divided by a suitable relaxation function (I∑MQ) point to point. From this function (Eq. 3) the defect contribution must be subtracted.

∑ =  +   − B exp(− 2  ⁄ )

(3)

The defect fraction B can be determined easily by an exponential fit to Iref in the data range,

where IDQ reached to zero (i.e a DQ evolution time beyond 9 ms).  is the transverse

relaxation time. In permanently crosslinked networks ∑ is used to normalize the DQ

filtered intensity by removing the relaxation effect on the experimental data. Thereby, DQ

build-up data can be normalized, obtaining   =   /∑ . The obtained normalized DQ

intensity (  ) is further fitted assuming a Gaussian distribution (Eq. 4) of residual dipolar couplings according to

  ( ,  ) =  1 − 

' ' )*+, ' )-

"#$%& ( 0 ' ' 2 ./   ( 1 )-

0 34 '1 ')(

5

(4)

8 ACS Paragon Plus Environment

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

rel. intenstiy (a.u)

Page 9 of 31

1,0

Iref IDQ

0,8

IΣMQ=Iref+IDQ-defects

0,6

InDQ=IDQ/IΣMQ InDQ- Gaussian fit

0,4 0,2 0,0 0

1

2

3 4 5 6 7 DQ evolution time (ms)

8

9

10

Fig. 1: Reference intensity (Iref), DQ build-up intensity (IDQ), defect-corrected overall relaxation function (∑ ), and the normalized DQ build-up curve (InDQ) for gum sample (without any filler) as a function of DQ evolution time (τDQ). The dotted line represents the Gaussian fit to determine the dipolar coupling based on Eq. 4.

fit of the normalized DQ (InDQ) curve. The so-obtained residual dipolar coupling  values

Relevant signal functions, Iref, IDQ, IMQ and InDQ are depicted in Fig. 1 along with the Gaussian are directly proportional to the crosslink density of the polymer network,  represents the

distribution of crosslinks throughout the sample.  in rad/s is given as  /26 in the unit of frequency (Hz) for simplicity58. It is well established that  is directly related to the

dynamic chain order parameter (7) given by Eq. 5,

 ∝ 7 =

9:

;
B2 =

(CDEFG H+DIJK GL,, MNLD,KO*+) ' ×R.ST ×U.WW×W'X .PP YZ[\ \[ ] ^[ ^^ < ×W.S



(6)

Here, final weight loss is given in % at 800 ˚C (obtained from thermogravimetric analysis), moisture content in % (estimated by drying of silica powder at 110 °C for 2 hrs by ISO 787-2 standards), Na is the Avogadro number (6.022x1023), and the specific surface area of silica is estimated by BET nitrogen adsorption measurements. The BET nitrogen adsorption for specific surface area of abSiO2 nanoparticles and commercial silica powder was measured at 77.4 K by Autosrob-1, Quantachrome, USA. The silica powders were preconditioned at 80 °C for 2 hrs in a vacuum oven before BET measurements. RESULTS AND DISCUSSIONS The reinforcement effect of abSiO2 silica and pSiO2 can be seen in Fig. 2a. All filled vulcanizates at a given loading of silica (30 phr) show stronger mechanical behavior as compared with pure gum rubber. However, the silanized abSiO2 silica composite shows much higher rubber moduli values (for example, at 200 % or 300 % elongations) than the corresponding pSiO2 composites at same loading. In general, the modulus at certain elongation largely depends on the surface area of the fillers and the rubber filler interactions. For this reason, a more detailed analysis of the microstructure of the abSiO2 is a very important issue here. To get a rough idea about the size of the abSiO2 particles, dynamic light scattering experiments were performed.

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry

a) 12

b) 70

Gum x-30 x-30_TESPT i-30 i-30_TESPT

8

60

intensity (%)

10

Stress (MPa)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 31

6 4

in-situ silica particle commerical silica particle

50 40 30 20

2

10

0 0

100

200

300

400

500

0 100

Strain (%)

1000

10000

Diameter (nm)

Fig. 2: a) The stress-strain curves of vulcanizates obtained from solution styrene butadiene rubber (SSBR) filled with alkoxy-based silica (abSiO2) and precipitated silica (pSiO2), b) particle diameter of commercial Ultrasil-VN3 GR and alkoxy-derived silica measured by DLS method

Fig. 3: Scanning electron microscopic images of (a, c) commercial precipitated silica (pSiO2), and (b, d) alkoxy-based silica (abSiO2). The latter is synthesized under similar conditions that have been used during in-situ silica synthesis in rubber solution. In the present system no rubber was used during the silica synthesis.

12 ACS Paragon Plus Environment

Page 13 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The data obtained from DLS indicate that the size of the abSiO2 particles is much larger than pSiO2 (Fig. 2b). The latter shows a bimodal distribution with broad range of particle sizes from 150 to 1500 nm and each domain has a mean diameter of about 300 and 1500 nm. abSiO2 remains in the narrow range of 200 to 500 nm. These findings also nicely corroborate with scanning electron microscopic (SEM) images as shown in Fig. 3. The precipitated silica is found to form clusters with different sizes, where the size of the primary particles are in the range of 10-20 nm. On the other hand, abSiO2 particles are relatively uniform and spherical in nature. The surface of the particles is not smooth (Fig.3d). Some particles are found to be fused with each other. The size of the abSiO2 found in SEM micrographs also corroborates the DLS data.

a)

b)

c)

d)

b)

Fig. 4: Scanning electron microscopic image of (a) alkoxy-based silica (abSiO2) synthesized by sol-gel method in SSBR-tetrahydrofuran solution without any silane coupling agent; (b) abSiO2 in presence of silane coupling (TESPT) and SSBRtetrahydrofuran solution. Fig (c) and (d) are the enlarged sections of (a) and (b) respectively. After preparation of the silica-rubber masterbatch the un-crosslinked rubber (non-bound rubber) was removed by solvent extraction process and the remaining part was used for this SEM investigation.

The scanning electron microscopic study was also carried out with the abSiO2, which is synthesized in presence of rubber solution (see experimental details). Fig. 4 describes the 13 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 31

morphology of the precipitated silica and a very distinct difference in structure can be observed between the silica, which has been grown in presence and absence of rubber molecules. In the first case, abSiO2 is comprised of very small particles (in the range of a few nanometer) resulting in relatively bigger aggregated structures and, mostly, the aggregates are found to be spherical. These spheres comprise numerous small primary particles with diameters of ~ 20-30 nm and uniform in size (Fig. 4b). However, in the case without rubber the synthesized abSiO2 are found to exhibit similar spherical morphology, but their surface is rather smooth. Primary small particles cannot be traced from this Fig. 4a and 4c. In both cases, a part of the each figure is enlarged and the surface of the big sphere can be easily illuminated (Fig. 4c and Fig. 4d) and difference in the surface characteristics can be easily seen. As seen from Fig. 4d the particles can be compared with a raspberry-like morphology. This kind of structure of the sol-gel silica has been already reported45, 51. Most probably the water molecules reside in droplets and the germination as well as the growth of the silica particles originates from the surface of the water droplets and, finally, the fine silica particles are forming agglomerates with trapped rubber chains. The existence of trapped rubber chains is supported by thermogravimetric studies (Fig. 5). It can be found that abSiO2 (after solvent extraction) contains 4 % bound rubber, which is anchored to the silica particles and cannot be removed. Further studies are required to enlighten more facts behind this unique morphology of silica particles. A schematic presentation visualizes how macromolecular chains are entrapped by the small silica particles (Fig. 6). The BET surface area of precipitated silica (175 m²/g) is much larger as for abSiO2 (32 m2/g) as seen in Table 1. Higher surface is generally enhancing the rubber to filler interactions, but in our case the observations are different. Probably, after separation of abSiO2 particles from the rubber by solvent extraction technique, always a small amount of rubber is permanently impregnated on silica particles and the presence of this rubber may inhibit nitrogen adsorption on silica surface, resulting in lower BET numbers. The discussion is directly indicating some other additional factors like chemical bonds between polymer chains and silica particles or trapping some macromolecular chains by the small primary particles to form a cluster, which could play a major role for higher reinforcing activity of abSiO2. Entrapment of polymer chains by the abSiO2 particles could be the reason for the development of typical surface characteristics, which are different to abSiO2 with smooth surface (synthesized without presence of polymer). This kind of mechanism is not explained so far in the literature and we encourage further studies to understand more details of the mechanism. 14 ACS Paragon Plus Environment

Page 15 of 31

alkoxy silica (blank study) alkoxy silica after solvent extraction

100

Weight loss (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

95

90

85 ∼ 4% rubber adsorbed or traped

80 100

200

300

400

500

600

700

800

Temperature (°C)

Fig. 5: Thermogravimetric analysis of in-situ synthesized silica (in presence of polymer) and without presence of polymer (synthesized maintain similar condition),

Fig. 6.: Schematic presentation of the alkoxy based in-situ silica (abSiO2) morphology and entrapment of the polymer chains by the silica clusters.

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 31

Table 1: Determined properties of different silica materials and silanol density Type of silica

Final weight loss (%)

Amount of moisture in (%)

BET (N2) surface (m2/g)

Si-OH per nm2

Commercial (pSiO2)

7

4.6

175

9.2

Alkoxy (abSiO2)

12

9.4

32

34

The amount of moisture was calculated from the weight loss after the treatment of the sample at 110°C for 2 h.

To elucidate the other reinforcing factors, the formation of direct chemical bonds between silica and rubber chains through classical silane coupling reaction, the number of silanol groups present on the silica surface is very important as these groups are responsible for chemical coupling between silica and polymer chains. The number of the silanol groups are roughly evaluated from thermogravimetric analysis (Eq. 6) and enumerated in Table 1. According to the chemical structure of silica, it contains three different chemical structures namely isolated, vicinal and geminal silanols. The silanol type and density varies according to the preparation condition and the amount of water being present during the synthesis. By knowing the thermal degradation behavior and the specific surface area (BET- nitrogen adsorption number), one can roughly estimate the number of silanol groups per unit area. Surprisingly, a 3 fold higher value can be observed from abSiO2. The presence of larger number of silanol groups on the surface of the alkoxy groups could enhance the coupling reactivity resulting to increasing rubber to filler interactions. Fig. 7 shows the linear relationship between the NMR and equilibrium swelling results in a series of increasing amount of sulfur in SBR rubber. The correlation between NMR and swelling results demonstrates that Dres is directly proportional to the inverse effective molecular weight between successive crosslinks (1/Mc), as shown previously16. The swelling results include a (potentially small) contribution from trapped entanglements. It is important to note that DQ NMR experiments are fully sensitive to both physical (entanglements) and chemical constraints in the bulk, explaining the intercept in Fig. 7. Note that due to the copolymer nature of our sample we did not convert Dres into the actual crosslink density value, because there are no reasonable reference coupling values available52.

16 ACS Paragon Plus Environment

Page 17 of 31

500

unfilled rubber

400 Dres/2π (Hz)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

300 chemical crosslinking

200 100 0 0,0

physical crosslinking 0,1

0,2 0,3 νFR=1/Mc (mol/kg)

0,4

Fig. 7: The correlation between crosslink density of unfilled SSBR with different amounts of sulfur, measured by MQ-NMR experiments and equilibrium swelling method based on the Flory-Rehner equation for evaluation. As mentioned earlier, the particle morphology of the silica particles are entirely different when the abSiO2 is synthesized in presence of rubber, and the final mechanical properties of this composites are rather high. To understand the effect of silica particles on the physical and chemical network, the master line shown in Fig. 7 can be taken as a reference and be compared with results of the composites52. The incorporation of pSiO2 into SSBR tends to reduce the crosslink density, as indicated in Fig. 8a, by the gradual decrease of Dres/2π. The phenomenon is well known and is attributed to the adsorption of polar vulcanizing chemicals like zinc salts, CBS and DPG on the silica surface, thus inhibiting the curing process during vulcanization reaction7. Moreover, the chemical nature of silica is slightly acidic with a pH value of ~6.5, which may interfere the vulcanization reaction. The effects are more pronounced when increasing the volume fraction of pSiO2 in the rubber matrix. The samples x-10 to x-50 (samples filled with 10 to 50 phr of precipitated silica) show significant reduction in crosslink density (downwards shifts) with increasing volume fraction of silica. In some contrast, the positively deviating values from the master line indicate some weak level of filler-polymer interfacial interactions. TESPT is a well-known bifunctional silane coupling agent and widely used in the formulation of silica filled green tires. It is also one of important silica surface modifier, through its silanization reaction of hydroxyl group present on the silica surface and the ethoxy groups of the coupling agent. The silane consists of sulfur bridges (S: 3~4) which can participate in the crosslinking of rubber chains during the vulcanization process. Addition of such bi-functional 17 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 31

coupling agents facilitates interfacial bonding between the silica filler and rubber and one can expect significantly higher crosslinks. In complete analogy to previous observations and interpretations52, in Fig. 8a, with the incorporation of TESPT silane coupling agent into the ppt. silica system, a shift significant right with respect to the master line is observed, while the DQ-NMR displays a negligible increment in the Dres/2π values.The increased apparent 1/Mc values from the swelling measurements indicate clearly that additional crosslinks are provided by the TESPT coupling agent resulting in filler particles acting as giant crosslink points. In Fig. 8b, interesting observations are found for silica particles generated by sol-gel reaction62. The values are found to shift even further towards the right with the increase of abSiO2. Unlike precipitated silica without the presence of TESPT, the abSiO2 did not negatively affect the overall crosslinking density (as seen by NMR) rather, but the swelling values are increased. This finding clearly evidences a higher interaction between rubber chains and the silica filler even with the absence of TESPT coupling agent. Addition of TESPT silane coupling agent to abSiO2 shows an even more significant increase in the rubberfiller attachment. A slight vertical shift indicates that Dres values are slightly increased upon TESPT modified alkoxy silica content. Presence of TESPT has increased the crosslink density, due to better interaction with the active silica filler surface, modified by n-butylamine catalyst used during the synthesis. In our previous study we found that amine modified abSiO2 can enhance the vulcanization efficiency and cleave the sulfur molecules (-S, -S2) providing additional crosslinks to the rubber62. This phenomenon explains the large horizontal shift in 1/Mc values after silanization of abSiO2 composites. One could therefore expect the additional crosslinks as cumulative effects from effective tethering of elastomer chains on the surface of abSiO2 by TESPT coupling agent, entrapment of the rubber chains inside the abSiO2 particles, excess sulfur donation by the silane coupling agent and cleavage of sulfur molecules.

18 ACS Paragon Plus Environment

Page 19 of 31

a)

b)

500

500

400

400 Dres/2π (Hz)

Dres/2π π (Hz)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

300 200

gum rubber (without filler) ppt. silica ppt. silica with TESPT

100 0 0,0

0,1

0,2 νFR=1/Mc (mol/kg)

0,3

0,4

300 200

gum rubber (without filler) alkoxy silica alkoxy silica with TESPT

100 0 0,0

0,1

0,2 0,3 νFR=1/Mc (mol/kg)

0,4

Fig. 8: Plots of crosslink density by DQ-NMR as a function of Flory-Rehner based equilibrium swelling method for (a) precipitated silica (pSiO2) composites with and without TESPT coupling agent (b) alkoxy silica (abSiO2) with and without TESPT coupling agent. The linear fitted line (master line) represents unfilled rubber samples vulcanized with different amounts of sulfur.

Filler-polymer interaction is a major factor deciding the mechanical properties. The properties of silica filled composites depend significantly on the type of silane coupling agent being used as well as the nature of interaction (physical or chemical) between rubber and silica. The final crosslink density of rubber matrix is also influenced by type of functional groups present in the silane coupling agent and their functionality (mono- or bi-functional). Figs. 9a and Fig. 9b depict the influence of various silane coupling agents on the crosslink density of the SSBR matrix based on the NMR and swelling measurements. The values above and below the horizontal line represents the change in network density by physical and chemical coupling. The effect of mono functional silanes (data within the red circles) and bifunctional silanes (data within the blue circles) on the physical and chemical crosslink densities can be easily correlated with the data shown in Fig. 9. Addition of mono functional coupling agents like DMS, APTS, ODTES exhibit a higher value of the chemical crosslink as compared with gum rubber with same dose of sulfur. On the other hand, bi-functional silanes like NXT, MPTS, TESPD and TESPT show significant increase of rubber-filler bonding, as the data shifted more towards right. The plots confirm that the additional crosslinks are mainly contributed by the sulfur presented in the silane coupling agent and are also directly related to the interfacial changes between silica and rubber59.

19 ACS Paragon Plus Environment

The Journal of Physical Chemistry

a)

b)

500

500

400

400

Dres/2π π (Hz)

Dres/2π π (Hz)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 31

300 200 100

Unfilled i-30 i-30_DMS i-30_APTS i-30_ODTES i-30_NXT i-30_MPTS i-30_TESPD i-30_TESPT

300 200 100

0 0,12

0,16 0,20 0,24 νFR=1/Mc (mol/kg)

0,28

0

Unfilled x-30 x-30_DMS x-30_APTS x-30_ODTES x-30_NXT x-30_MPTS x-30_TESPD x-30_TESPT

0,12

0,16

0,20 0,24 νFR=1/Mc (mol/kg)

0,28

Fig. 9: Influence of various coupling agent on the crosslink density of SSBR measured by DQ-NMR as a function of Flory-Rehner based equilibrium swelling method for 30 phr of (a) alkoxy-based silica (abSiO2) composites and (b) precipitated silica (pSiO2) composites.

Finally, comparing the chemical crosslinking contribution from abSiO2 and pSiO2, always a higher chemical crosslinking network can be found with abSiO2. The higher network density can be associated with the trapped rubber chains inside the abSiO2 aggregates as described in the scheme shown in Fig. 6. Furthermore, a higher Dres/2π value of abSiO2 composites (without presence of silane) from NMR measurements, are suggestive of higher physical crosslink density values, probably due to the rubber chain entrapment or the filler surfaces acting as macro-crosslink junctions. With special coupling agents like TESPD, TESPT and MPTS the physical network densities were found to be improved in both pSiO2 and abSiO2. Fig. 10 depicts the direct excited

29

Si magic angle spinning spectrum (MAS) of two silica

powders. The chemical structure of two given silica are different as evident from the MAS spectra. The chemical structure of silica particles are defined by the different silanol groups, namely, isolated, vicinal, geminal silanols and the silica dioxide. Both silica powders are expected to display two important peaks at -103 ppm and -113 ppm, corresponding to monosilanols and silica oxide respectively. In pSiO2 samples, these two main peaks are found at 103 ppm attributed to the mono-silanol group of Si-OH (Q3) and at -113 ppm attributed to the SiO2 (Q4). The peak at -93 ppm of the geminal silanols Si-(OH)2 (Q2) is not detected because of its low c. In the case of abSiO2 powder, the peak at -103 ppm is much stronger and the line width is nearly two times than ppt. silica powders. This explains that abSiO2 surface consists of many mono-silanol groups Si-OH (Q3)63. From the direct excited spectrum of two silica 20 ACS Paragon Plus Environment

Page 21 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

powders, Q4 and Q3 groups are estimated by peak integral analysis to get quantitative information on the chemical structure of in silica powders. pSiO2 powder consists 92 % of Q4 and 8 % of Q3, whereas, abSiO2 consists 57 % of Q4 and 43 % of Q3. Both the silica powders have negligible geminal silanol (Q2) groups. These findings correlate well with the physically estimated silanol density by thermogravimetric analysis and BET surface (in Table 1). To understand the changes in proton density on silica surface in the presence of rubber and silane coupling agent, 29Si-1H cross polarization experiments are performed. Fig. 11 shows the 29

Si-1H cross polarization spectra for SSBR/ abSiO2 and pSiO2 composites. The cross

polarization spectra indicate minor amounts of Q2 groups, which are expected to be present on the surface of the particles. From Table 2, the quantitative peak intensity estimates around about around 22 % of Q3 and 781 % of Q4 groups for the x-silica and around 43 % of Q3 and 57 % of Q4 for the abSiO2 silica powders. Because of the low filler content and the natural abundance of

29

Si of less than 5%

29

Si cross polarization experiments have been performed

for the composite samples. These enhance signals in spatial vicinity of 1H, which is suitable, because the coupling the the protonated rubber is of interest. However, such spectra are not quantitative, they may only be qualitatively compared within a series.

Fig. 10: Direct-polarization29Si MAS spectrum for alkoxy (abSiO2) and precipitated silica (pSiO2) powers

21 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 31

In two of the samples a significant amount of T-groups, indicating the covalent bonds between silica and silane coupling agent, is observed in the cross polarization spectra of the compounds. These signals are in the chemical shift range of -40 to -70 ppm.

x-30B

i-30B

x-30A

i-30A

x-silica

0

-20

-40

-60

-80

-100

-120

-140

-160

i-silica -180

-200

0

-20

-40

-60

-80

-100

-120

-140

-160

-180

-200

δSi [ppm]

δSi [ppm]

Fig. 11: 29Si-1H cross-polarization spectrum for SSBR/alkoxy-based (abSiO2) and precipitated (pSiO2) silica composites

Table 2: Estimated peak integral from 29Si-1H cross polarization experiments

Q3 integral (%)

Q4 integral (%)

i-30

28

72

i-30 TESPT

23

68

x-30

20.

80

x-30 TESPT

16

65

Materials

Fig. 12 shows an EDX spectrum normalized with respect to Si peak acquired form representative areas of few µm2 of each sample. The EDX spectrum shows presence of five elements: Si, C, O, Na, S. The average composition of the two areas in each sample is summarized in Table 3. The average content of C, Si, O of the both specimens in the same within the error margin, however the sulfur content is approximately twice higher in the abSiO2. The higher amount of sulfur atoms detected on the abSiO2 surface indicates its higher reactivity with the silane coupling agent, because of its active and ample hydroxyl groups.

22 ACS Paragon Plus Environment

Page 23 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The small sodium peak observed in the case of abSiO2 and corresponding to ~ 0.2 wt% might be due to the presence of impurities from the precursor. b)

a)

Fig. 12: (a) A main EDX spectrum of precipitated silica (pSiO2) and alkoxy silica (abSiO2 with TESPT silanes). The unreacted rubber or loosely bound rubber was extracted by solvent treatment. (b) Detected sulfur peak on the surface of the pSiO2 and abSiO2 powder by EDX analysis x-30s (blue), i-30s (orange) .‘i’ and ‘x’ denote the abSiO2 and pSiO2 respectively

Table 3. Average composition in weight (%) derived from EDX spectra Element

x-30s

i-30s

C

32 ± 10

36.5 ± 10

Si

33 ± 4

31 ± 7

O

34 ± 10

32.5 ± 7

S

0.2 ± 0.1

0.4 ± 0.1

23 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 31

Fig. 13: Scanning electron microscopic image and elemental maps of precipitated silica (pSiO2) (a-c) and alkoxy-based silica (abSiO2) (d-f). The silica samples were collected from rubber compounds after solvent treatment. a) SEM image, b) color overlay maps of carbon (red), oxygen (green) and silicon (blue) distribution, c) carbon map, d) SEM image, e) color overlay of carbon, oxygen and silicon distribution, f) carbon map. The cyan (cyan = blue + green) areas in (b, e) correspond to silica particles, the blue areas correspond to silicon substrate underneath the particles. (The scale bar is 8 µm.)

Fig.13 shows the SEM images and distribution of different elements of samples x-30s and i30s respectively. In both cases the spherical shape of the aggregates/agglomerate can be seen. The size of the spheres is much smaller in abSiO2 as compared with pSiO2. For specimen i30s the silica particles are surrounded by carbon, probably a thin film of rubber strongly attached to the surface of the particles, indicating rubber chains anchored permanently on the abSiO2 surface, which is supporting the trapped rubber chains proposal as shown in scheme (Fig. 6). Conclusions The considerably higher mechanical reinforcement induced by sol-gel derived alkoxy silica (abSiO2) is explored and discussed, where the main challenge was to study the effects of silica generation in presence of rubber chains in the reaction medium. In this study, it is revealed that silica generated in-situ is forming aggregated, raspberry-like particle structures. The overall crosslinking density is found to be enhanced because of the silica aggregates acting as macroscopic multiple crosslinking sites.

Entrapment of rubber chains inside the silica 24

ACS Paragon Plus Environment

Page 25 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

particles as well as strong mechanical tethering of rubber chains onto the surface of the abSiO2 particles were further supported by low-field proton NMR investigations and EDX experiments. Particularly, low-field NMR along with swelling studies very clearly demonstrates the influence of abSiO2 on the physical as well as overall network density in the system. Elemental mapping of the specially prepared bound rubber samples elaborates the fact that relatively high amount of rubber chains are entrapped within the abSiO2aggregates. Only high-resolution

29

Si NMR was not sensitive enough to further substantiate the entrapment of

polymer chain within the abSiO2. In all, it could be concluded that the silanization efficiency not only depends on processing temperature and time, rather also on the reactive hydroxyl density of abSiO2. Supporting Information Available Fig. S1-S4 designate double-quantum NMR data fitted for the evaluation of the crosslinking density of different rubber samples. Table-S1 and Table- S2 describe the crosslinking density values of the rubber samples measured by solvent swelling and NMR method. Acknowledgements S.R.V. is thankful to IPF Annex fund for providing the PhD fellowship and is grateful to U. Gohs and C. Zschech for helping with the low-field proton NMR instrument. The author is also thankful to A. Mujtaba and A. Vieyres for the discussions on the double-NMR based crosslink density measurements and B. Vieweg for some of the 29Si NMR experiments.

References (1) Shakun, A.; Vuorinen, J.; Hoikkanen, M.; Poikelispää, M.; Das, A. Hard Nanodiamonds in Soft Rubbers: Past, Present and Future–a Review. Composites, Part A 2014, 64, 49-69. (2) Das, A.; Wang, D.-Y.; Stöckelhuber, K. W.; Jurk, R.; Fritzsche, J.; Klüppel, M.; Heinrich, G. Rubber–Clay Nanocomposites: Some Recent Results. In Advanced Rubber Composites, Springer Berlin Heidelberg: 2010; pp 85-166. (3) Rooj, S.; Das, A.; Morozov, I. A.; Stöckelhuber, K. W.; Stocek, R.; Heinrich, G. Influence of “Expanded Clay” on the Microstructure and Fatigue Crack Growth Behavior of Carbon Black Filled NR Composites. Compos. Sci. Technol. 2013, 76, 61-68. (4) Rooj, S.; Das, A.; Stöckelhuber, K. W.; Reuter, U.; Heinrich, G., Highly Exfoliated Natural Rubber/Clay Composites by “Propping‐Open Procedure”: The Influence of Fatty‐Acid Chain Length on Exfoliation. Macromol. Mater. Eng. 2012, 297, 369-383. (5) Rooj, S.; Das, A.; Heinrich, G. Preintercalation of an Organic Accelerator into Nanogalleries and Preparation of Ethylene Propylene Diene Terpolymer Rubber–Clay Nanocomposites. Polym. J. 2011, 43, 285-292. 25 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 31

(6) Evans, M. S. Tyre Compounding for Improved Performance; iSmithers Rapra Publishing, 2002; Vol. 12, p112. (7) Hewitt, N.; Ciullo, P. Compounding Precipitated Silica in Elastomers: Theory and Practice; William Andrew, 2007. (8) Nakajima, N.; Shieh, W. J.; Wang, Z. G. Mixing and Extrusion of High Silica and All SilicaNatural Rubber Compounds. Int. Polym. Process. 1991, 6, 290-296. (9) Zou, H.; Wu, S.; Shen, J. Polymer/Silica Nanocomposites: Preparation, Characterization, Properties, and Applications. Chem. Rev. 2008, 108, 3893-3957. (10) Das, A.; Jurk, R.; Stöckelhuber, K. W.; Heinrich, G. Silica-Ethylene Propylene Diene Monomer Rubber Networking by in-Situ Sol-Gel Method. J. Macromol. Sci., Part A: Pure Appl. Chem. 2008, 45, 101-106. (11) Das, C.; Bansod, N. D.; Kapgate, B. P.; Reuter, U.; Heinrich, G.; Das, A. Development of Highly Reinforced Acrylonitrile Butadiene Rubber Composites Via Controlled Loading of Sol-Gel Titania. Polymer 2017, 109, 25-37. (12) Kapgate, B. P.; Das, C.; Das, A.; Basu, D.; Reuter, U.; Heinrich, G. Effect of Sol–Gel Derived in Situ Silica on the Morphology and Mechanical Behavior of Natural Rubber and Acrylonitrile Butadiene Rubber Blends. J. Sol-Gel Sci. Technol. 2012, 63, 501-509. (13) Meer, S.; Kausar, A.; Iqbal, T. Attributes of Polymer and Silica Nanoparticle Composites: A Review. Polym.-Plas.Technol. Eng. 2016, 55, 826-861. (14) Rahman, I. A.; Padavettan, V. Synthesis of Silica Nanoparticles by Sol-Gel: Size-Dependent Properties, Surface Modification, and Applications in Silica-Polymer Nanocomposites—a Review. J. Nanomater.2012, 2012, 8. (15) Li, Y.; Wang, M.; Zhang, T.; Zhang, F.; Fu, X. Study on Dispersion Morphology of Silica in Rubber. Rubber Chem. Technol. 1994, 67, 693-699. (16) Ten Brinke, J.; Debnath, S.; Reuvekamp, L.; Noordermeer, J. Mechanistic Aspects of the Role of Coupling Agents in Silica–Rubber Composites. Compos. Sci. Technol. 2003, 63, 1165-1174. (17) Stockelhuber, K.; Svistkov, A.; Pelevin, A.; Heinrich, G. Impact of Filler Surface Modification on Large Scale Mechanics of Styrene Butadiene/Silica Rubber Composites. Macromolecules 2011, 44, 4366-4381. (18) Heinrich, G.; Vilgis, T. Why Silica Technology Needs S-SBR in High Performance Tires?: The Physics of Confined Polymers in Filled Rubbers. Kautsch. Gummi Kunstst. 2008, 61, 368376. (19) Le, H. H.; Parsekar, M.; Ilisch, S.; Henning, S.; Das, A.; Stöckelhuber, K. W.; Beiner, M.; Ho, C. A.; Adhikari, R.; Wießner, S. Effect of Non‐Rubber Components of Nr on the Carbon Nanotube (Cnt) Localization in SBR/NR Blends. Macromol. Mater. Eng. 2014, 299, 569-582. (20) Prasertsri, S.; Rattanasom, N. Fumed and Precipitated Silica Reinforced Natural Rubber Composites Prepared from Latex System: Mechanical and Dynamic Properties. Polym. Test. 2012, 31, 593-605. (21) Tiarks, F.; Landfester, K.; Antonietti, M. Silica Nanoparticles as Surfactants and Fillers for Latexes Made by Miniemulsion Polymerization. Langmuir 2001, 17, 5775-5780. (22) Schmid, A.; Armes, S. P.; Leite, C. A.; Galembeck, F. Efficient Preparation of Polystyrene/Silica Colloidal Nanocomposite Particles by Emulsion Polymerization Using a Glycerol-Functionalized Silica Sol. Langmuir 2009, 25, 2486-2494. (23) Zou, H.; Wang, X. Adsorption of Silica Nanoparticles onto Poly (N-Vinylpyrrolidone)Functionalized Polystyrene Latex. Langmuir 2017, 33, 1471-1477.

26 ACS Paragon Plus Environment

Page 27 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

(24) Kutcherlapati, S. R.; Koyilapu, R.; Jana, T. Poly(N-Vinyl Imidazole) Grafted Silica Nanofillers: Synthesis by Raft Polymerization and Nanocomposites with Polybenzimidazole. J. Polym. Sci.,Part A: Polym. Chem. 2018, 56, 365-375. (25) Zhang, J.; Liu, N.; Wang, M.; Ge, X.; Wu, M.; Yang, J.; Wu, Q.; Jin, Z. Preparation and Characterization of Polymer/Silica Nanocomposites Via Double in-Situ Miniemulsion Polymerization. J. Polym. Sci., Part A: Polym. Chem. 2010, 48, 3128-3134. (26) Zhang, J.; Yang, J.; Wu, Q.; Wu, M.; Liu, N.; Jin, Z.; Wang, Y. SiO2/Polymer Hybrid Hollow Microspheres Via Double in-Situ Miniemulsion Polymerization. Macromolecules 2010, 43, 1188-1190. (27) Blin, J.-L.; Stébé, M.-J.; Lebeau, B. Hybrid/Porous Materials Obtained from NanoEmulsions. Curr. Opin. Colloid Interface Sci. 2016, 25, 75-82. (28) Das, A.; Jurk, R.; Stöckelhuber, K. W.; Heinrich, G. Silica-Ethylene Propylene Diene Monomer Rubber Networking by in Situ Sol-Gel Method. J. Macromol. Sci., Part A: Pure Appl. Chem. 2008, 45, 101-106. (29) Sankar Raman, V.; Das, A.; Stockelhuber, K. W.; Eshwaran, S. B.; Chanda, J.; Malanin, M.; Reuter, U.; Leuteritz, A.; Boldt, R.; Wiessner, S. et al. Improvement of Mechanical Performance of Solution Styrene Butadiene Rubber by Controlling the Concentration and the Size of in-Situ Derived Sol-Gel Silica Particles. RSC Adv. 2016, 6, 33643-33655 (30) Kapgate, B. P.; Das, C.; Das, A.; Basu, D.; Wiessner, S.; Reuter, U.; Heinrich, G. Reinforced Chloroprene Rubber by in-Situ Generated Silica Particles: Evidence of Bound Rubber on the Silica Surface. J. Appl. Polym. Sci. 2016, 133, 43717 (31) Bansod, N. D.; Kapgate, B. P.; Das, C.; Das, A.; Basu, D.; Debnath, S. C. Compatibilization of Natural Rubber/Nitrile Rubber Blends by Sol–Gel Nano-Silica Generated by in-Situ Method. J. Sol-gel Sci. Technol. 2016, 80, 548-559. (32) Kapgate, B. P.; Das, C.; Basu, D.; Das, A.; Heinrich, G.; Reuter, U. Effect of Silane Integrated Sol–Gel Derived in-Situ Silica on the Properties of Nitrile Rubber. J. Appl. Polym. Sci. 2014, 131, 40054 (33) Kapgate, B. P.; Das, C.; Basu, D.; Das, A.; Heinrich, G. Rubber Composites Based on SilaneTreated Stöber Silica and Nitrile Rubber: Interaction of Treated Silica with Rubber Matrix. J. Elastomers Plast.2013, 47, 248-261. (34) Kohjiya, S.; Murakami, K.; Iio, S.; Tanahashi, T.; Ikeda, Y. In-situ Filling of Silica onto “Green” Natural Rubber by the Sol-Gel Process. Rubber Chem. Technol. 2001, 74, 16-27. (35) Chaichua, B.; Prasassarakich, P.; Poompradub, S. In-situ Silica Reinforcement of Natural Rubber by Sol–Gel Process Via Rubber Solution. J. Sol-gel Sci. Technol. 2009, 52, 219-227. (36) Ikeda, Y.; Katoh, A.; Shimanuki, J.; Kohjiya, S. Nano-Structural Observation of in Situ Silica in Natural Rubber Matrix by Three Dimensional Transmission Electron Microscopy. Macromol. Rapid Commun. 2004, 25, 1186-1190. (37) Tanahashi, H.; Osanai, S.; Shigekuni, M.; Murakami, K.; Ikeda, Y.; Kohjiya, S. Reinforcement of Acrylonitrile—Butadiene Rubber by Silica Generated in Situ. Rubber Chem. Technol. 1998, 71, 38-52. (38) Kohjiya, S.; Ikeda, Y. Reinforcement of General-Purpose Grade Rubbers by Silica Generated in Situ. Rubber Chem. Technol. 2000, 73, 534-550. (39) Vaikuntam, S. R.; Bhagavatheswaran, E. S.; Stöckelhuber, K. W.; Wießner, S.; Heinrich, G.; Das, A. Development of High Performance Rubber Composites from Alkoxide-based Silica and Solution Styrene–Butadiene Rubber. Rubber Chem. Technol. 2017, 90, 467-486 . (40) Das, C.; Bansod, N. D.; Kapgate, B. P.; Reuter, U.; Heinrich, G.; Das, A. Development of Highly Reinforced Acrylonitrile Butadiene Rubber Composites Via Controlled Loading of Sol-Gel Titania. Polymer 2017, 109, 25-37. 27 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 31

(41) Ikeda, Y.; Kohjiya, S. In-Situ Formed Silica Particles in Rubber Vulcanizate by the Sol-Gel Method. Polymer 1997, 38, 4417-4423. (42) Raman, V. S., et al., Improvement of Mechanical Performance of Solution Styrene Butadiene Rubber by Controlling the Concentration and the Size of in-Situ Derived Sol-Gel Silica Particles. RSC Adv. 2016, 6, 33643-33655. (43) Miloskovska, E.; Friedrichs, C.; Hristova-Bogaerds, D.; Persenair, O.; van Duin, M.; Hansen, M. R.; de With, G. Chemical Mapping of Silica Prepared Via Sol–Gel Reaction in Rubber Nanocomposites. Macromolecules 2015, 48, 1093-1103. (44) Murakami, K.; Iio, S.; Ikeda, Y.; Ito, H.; Tosaka, M.; Kohjiya, S. Effect of Silane-Coupling Agent on Natural Rubber Filled with Silica Generated in-Situ. J. Mater. Sci.2003, 38, 14471455. (45) Bandyopadhyay, A.; Sarkar, M. D.; Bhowmick, A. K. Effect of Reaction Parameters on the Structure and Properties of Acrylic Rubber/Silica Hybrid Nanocomposites Prepared by Sol‐ Gel Technique. J. Appl. Polym. Sci. 2005, 95, 1418-1429. (46) D'Acunzi, M.; Mammen, L.; Singh, M.; Deng, X.; Roth, M.; Auernhammer, G. K.; Butt, H.J.; Vollmer, D. Superhydrophobic Surfaces by Hybrid Raspberry-Like Particles. Faraday Discuss. 2010, 146, 35-48. (47) Zhou, X.; Shao, H.; Liu, H. Preparation and Characterization of Film-Forming RaspberryLike Polymer/Silica Nanocomposites Via Soap-Free Emulsion Polymerization and the Sol– Gel Process. Colloid Polym. Sci. 2013, 291, 1181-1190. (48) Wu, X.; Tian, Y.; Cui, Y.; Wei, L.; Wang, Q.; Chen, Y. Raspberry-Like Silica Hollow Spheres: Hierarchical Structures by Dual Latex− Surfactant Templating Route. J. Phys. Chem. C 2007, 111, 9704-9708. (49) Chen, M.; Wu, L.; Zhou, S.; You, B. Synthesis of Raspberry-Like Pmma/Sio2 Nanocomposite Particles Via a Surfactant-Free Method. Macromolecules 2004, 37, 9613-9619. (50) Reculusa, S.; Mingotaud, C.; Bourgeat-Lami, E.; Duguet, E.; Ravaine, S. Synthesis of DaisyShaped and Multipod-Like Silica/Polystyrene Nanocomposites. Nano Lett. 2004, 4, 16771682. (51) Ming, W.; Wu, D.; van Benthem, R.; De With, G. Superhydrophobic Films from RaspberryLike Particles. Nano Lett. 2005, 5, 2298-2301. (52) Chan, A. J.; Steenkeste, K.; Canette, A.; Eloy, M.; Brosson, D.; Gaboriaud, F.; FontaineAupart, M.-P. Natural Rubber–Filler Interactions: What Are the Parameters? Langmuir 2015, 31, 12437-12446. (53) Valentín, J. L.; Mora-Barrantes, I.; Carretero-González, J.; López-Manchado, M. A.; Sotta, P.; Long, D. R.; Saalwächter, K. Novel Experimental Approach to Evaluate Filler−Elastomer Interactions. Macromolecules 2010, 43, 334-346. (54) Vieyres, A.; Pérez-Aparicio, R.; Albouy, P.-A.; Sanseau, O.; Saalwächter, K.; Long, D. R.; Sotta, P. Sulfur-Cured Natural Rubber Elastomer Networks: Correlating Cross-Link Density, Chain Orientation, and Mechanical Response by Combined Techniques. Macromolecules 2013, 46, 889-899. (55) Valentín, J. L.; Mora-Barrantes, I.; Carretero-González, J.; López-Manchado, M. A.; Sotta, P.; Long, D. R.; Saalwächter, K. Novel Experimental Approach to Evaluate Filler−Elastomer Interactions. Macromolecules 2010, 43, 334-346. (56) Saalwächter, K.; Heuer, A. Chain Dynamics in Elastomers as Investigated by Proton Multiple-Quantum Nmr. Macromolecules 2006, 39, 3291-3303. (57) Mujtaba, A.; Keller, M.; Ilisch, S.; Radusch, H. J.; Thurn-Albrecht, T.; Saalwächter, K.; Beiner, M. Mechanical Properties and Cross-Link Density of Styrene–Butadiene Model

28 ACS Paragon Plus Environment

Page 29 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Composites Containing Fillers with Bimodal Particle Size Distribution. Macromolecules 2012, 45, 6504-6515. (58) Saalwächter, K. Proton Multiple-Quantum Nmr for the Study of Chain Dynamics and Structural Constraints in Polymeric Soft Materials. Prog. Nucl. Magn. Reson. Spectrosc. 2007, 51, 1-35. (59) Saleesung, T.; Reichert, D.; Saalwächter, K.; Sirisinha, C. Correlation of Crosslink Densities Using Solid State Nmr and Conventional Techniques in Peroxide-Crosslinked EPDM Rubber. Polymer 2015, 56, 309-317. (60) Mora-Barrantes, I.; Rodríguez, A.; Ibarra, L.; González, L.; Valentín, J. L. Overcoming the Disadvantages of Fumed Silica as Filler in Elastomer Composites. J. Mater. Chem. 2011, 21, 7381. (61) Valentín, J. L.; Posadas, P.; Marcos‐Fernández, A.; Ibarra, L.; Rodríguez, A. Effect of a Fatty Amine on Processing and Physical Properties of Sbr Compounds Filled with Silane–Silica Particles. J. Appl. Polym. Sci. 2006, 99, 3222-3229. (62) Meier, J. G.; Fritzsche, J.; Guy, L.; Bomal, Y.; Klüppel, M. Relaxation Dynamics of Hydration Water at Activated Silica Interfaces in High-Performance Elastomer Composites. Macromolecules 2009, 42, 2127-2134. (63) Simonutti, R.; Comotti, A.; Negroni, F.; Sozzani, P. 13C and 29Si Solid-State Nmr of RubberSilica Composite Materials. Chem. Mater. 1999, 11, 822-828.

29 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 31

TOC Graphic

30 ACS Paragon Plus Environment

Page 31 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Entrapped styrene butadiene polymer chains by alkoxy silica 254x190mm (300 x 300 DPI)

ACS Paragon Plus Environment