Document Detail

Quantifying the effect of polymer blending through molecular modelling of cyanurate polymers.
Jump to Full Text
MedLine Citation:
PMID:  22970230     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
Modification of polymer properties by blending is a common practice in the polymer industry. We report here a study of blends of cyanurate polymers by molecular modelling that shows that the final experimentally determined properties can be predicted from first principles modelling to a good degree of accuracy. There is always a compromise between simulation length, accuracy and speed of prediction. A comparison of simulation times shows that 125ps of molecular dynamics simulation at each temperature provides the optimum compromise for models of this size with current technology. This study opens up the possibility of computer aided design of polymer blends with desired physical and mechanical properties.
Authors:
Alasdair O Crawford; Ian Hamerton; Gabriel Cavalli; Brendan J Howlin
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't     Date:  2012-09-06
Journal Detail:
Title:  PloS one     Volume:  7     ISSN:  1932-6203     ISO Abbreviation:  PLoS ONE     Publication Date:  2012  
Date Detail:
Created Date:  2012-09-12     Completed Date:  2013-03-07     Revised Date:  2013-07-11    
Medline Journal Info:
Nlm Unique ID:  101285081     Medline TA:  PLoS One     Country:  United States    
Other Details:
Languages:  eng     Pagination:  e44487     Citation Subset:  IM    
Affiliation:
Department of Chemistry, University of Surrey, Guildford, Surrey, United Kingdom.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Models, Molecular*
Molecular Dynamics Simulation
Polymers / chemistry*
Triazines / chemistry*
Chemical
Reg. No./Substance:
0/Polymers; 0/Triazines; H497R4QKTZ/cyanuric acid
Comments/Corrections

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Full Text
Journal Information
Journal ID (nlm-ta): PLoS One
Journal ID (iso-abbrev): PLoS ONE
Journal ID (publisher-id): plos
Journal ID (pmc): plosone
ISSN: 1932-6203
Publisher: Public Library of Science, San Francisco, USA
Article Information
Download PDF
Copyright: 2012 Crawford et al
License:
Received Day: 30 Month: 3 Year: 2012
Accepted Day: 3 Month: 8 Year: 2012
collection publication date: Year: 2012
Electronic publication date: Day: 6 Month: 9 Year: 2012
Volume: 7 Issue: 9
E-location ID: e44487
ID: 3435312
PubMed Id: 22970230
Publisher Id: PONE-D-12-09388
DOI: 10.1371/journal.pone.0044487

Quantifying the Effect of Polymer Blending through Molecular Modelling of Cyanurate Polymers Alternate Title:Modelling of Polymer Blends
Alasdair O. Crawfordaff1
Ian Hamertonaff1
Gabriel Cavalliaff1
Brendan J. Howlinaff1*
Emanuele Paciedit1 Role: Editor
Department of Chemistry, University of Surrey, Guildford, Surrey, United Kingdom
University of Leeds, United Kingdom
Correspondence: * E-mail: b.howlin@surrey.ac.uk
[conflict] Competing Interests: The authors have declared that no competing interests exist.
Contributed by footnote: Conceived and designed the experiments: AOC BJH IH GC. Performed the experiments: AOC. Analyzed the data: BJH IH GC. Contributed reagents/materials/analysis tools: BJH. Wrote the paper: BJH AOC IH.

Introduction

Cyanate esters constitute a family of addition cured high performance, thermosetting polymers, which occupy a niche intermediate between high glass transition temperature (Tg), tetrafunctional epoxy resins and bismaleimides (BMIs) [1]. The combination of favourable thermal and mechanical performance (dry Tg values of 270–300°C are common with a strain at break of over 5–8%) coupled with low dielectric loss properties (a low dielectric constant of ca. 2.7 with a loss tangent of 0.003 is typical [1]) make cured cyanates attractive and able to offer a unique property profile. Although requiring toughening for some engineering applications, cyanates can be combined with inherently tough engineering thermoplastics (HexPly 954-2A, GIC = 250 J/m2) [2] or elastomers (HexPly 953-3, GIC = 450 J/m2) [3] to yield impressive enhancements. In this form they typically find application as matrices in advanced composites (either in combination with epoxy resins in aerospace applications [4]) or with BMIs as dielectric polymers in the microelectronics industry [5]. The rationale for the preparation of the binary blends presented here is to examine the potential for deriving improved properties (physical and mechanical) through the co-reaction of chemically compatible monomers. To date, aside from isolated studies [6], it appears that little systematic work has been carried out to examine the potential benefits of blending and co-curing polycyanate monomers. In the current study, we have examined three common cyanate species and the binary blends thereof by simulating selected properties using atomistic modelling in our drive to improve our predictive capability.

Molecular modelling of polymers is a growing area and it has been used in a wide variety of polymeric systems. By far the most effort has been concentrated on epoxies, owing to their general usefulness. Reports have predicted the structure, mechanical properties and moisture diffusion in epoxy resins [7], [8][13]. Other thermosetting polymers have also been modelled including, polycyanurates [14], polybenzoxazines [15], polyimides (in particular gas permeation across polyimide membranes) [16][18] and cyclohexanone formaldehyde resins (plastic printing) [19]. Non thermosetting polymers have included polyethylene oxides [20], polysiloxanes (glass transition temperature) [21], [22] and polyethylene terepthalate (gas diffusion) [23]. Recently the field has moved into the modelling of nanocomposites with carbon nanotube reinforced composites becoming of interest [24], [25].

It is fundamental to our approach that the simulations that are performed are always supported by empirical data, either single crystal data in the formation of structures or from physical or mechanical measurements when determining properties for the final polymer. In this paper we report a systematic study of the blending of the three cyanates by molecular simulation using the software suite Materials Studio [26]. This method is generally applicable to all crosslinked resin systems. There is always a compromise between how long the simulation takes and the accuracy of the predicted results. If it takes much longer to simulate the system than it takes to synthesize and measure the properties predicted then the method will be of limited usefulness in property prediction. This is a particular problem with synthetic polymers as there is an on-going debate on whether it is possible to equilibrate a long polymer chain by molecular mechanics, see for example [27].


Methods
Materials

The dicyanate ester monomers: 2,2-bis(4-cyanatophenyl)propane (1), hereafter termed BADCy 1,1-bis(4-dicyanatophenyl)ethane (2) (LECy), and the oligomeric phenolic cyanate (average value of n = 1) (3) (PT30) were the three monomers simulated in this work. (Figure 1). Experimental values of the materials were also obtained by the methods described in Crawford et. al, 2012 [28]. The blends simulated are presented in Table 1. The monomers polymerise by cyclotrimerisation, where three cyanate groups combine to form a six membered cyanurate ring, so the resulting polymer is a cross linked network composed of cyanurate rings joined by the bridging groups attached to the cyanate moieties (Figure 2). N.B. cured blends are denoted by the use of square brackets to differentiate them from the corresponding monomer blend i.e. monomer blend (190–210) becomes polycyanurate [190–210] following cure.

Simulation

The Materials Studio molecular modelling suite (Accelrys Inc.) was utilised in this work [26] using in house PCs (e.g. a Dell PowerEdge 1950, 2×Quad Core Intel Xeon E5140 2.33 GHz, 8 GB RAM, 500 GB HDD). The Discover module was used for general simulation requirements, such as geometry optimisation and molecular dynamics as well as molecular mechanical analysis to predict values for tensile, bulk modulus, shear modulus, Poisson’s ratio and the Lamé constants. The Amorphous Cell module was used to build amorphous, homogenous 3D cells composed of molecules that were drawn in silico. It also has a number of protocols designed to make greater use of the Discover module, of specific interest is the temperature cycling protocol, which can be used for Tg prediction. All simulations were performed in the bulk state, i.e. without the addition of solvent as there is no added solvent in epoxy resin cure. Potential energies were calculated using the Polymer Consistent Force Field (PCFF) [29]. An attempt was made to use the COMPASS [30] forcefield but this implementation does not cope with the cyanate group. The Legacy module within Materials Studio was used to construct an amorphous cell. The amorphous cell contained an average of 4480 atoms. The simulations were also run with cell sizes double, thrice and four times this number. A unit cell of the appropriate volume was packed randomly with the correct number of monomers, either as single monomers as in the neat resin work or in the correct mole fraction in the case of the blends. The cyanate groups were inspected and those within 4.5 Å of each other were bonded manually to form 1,3,5-triazine rings and a minimization of 5000 iterations was carried out to relieve the strain using the Discover minimization module to a convergence of 1000 kcal mol−1 Å−1. A Conjugate Gradient algorithm [31] was carried out to a convergence of 100 kcal mol Å−1. As all monomers were assumed to be equally coreactive the nature of the final structure depended ultimately on the initial random arrangement of the monomers. Hence it was possible to generate homogenous networks or blocked semi-interpenetrating networks depending on the starting conditions. Bonds were reacted through the faces of the unit cell to neighbouring periodic boxes to simulate an infinite three-dimensional highly cross-linked network. After the final cyclotrimerization the system was submitted to a final minimization of 10,000 iterations. The NPT ensemble with a time step of 1 fs was utilized with the Andersen thermostat [32] at a Pressure of 0.1 MPa under the Parrinello Barostat [33]. PCFF was used with the atomic Van der Waals summation, a cut-off of 10.00 Å, a spline width of 3.00 Å and a buffer width of 1.00 Å. 51 MD simulations were run between 773 K (500°C) and 273 K (0°C) in decrements of 10 K; at each temperature stage a 125 ps MD simulation was carried out. In accord with the problem of equilibrating long polymer chains, the simulations were also run for 1 ns at each temperature stage. A plot of calculated cell density was plotted against simulation temperature to determine both the Tg and the degradation onset temperature.


Results and Discussion
Effect of Model Size

Increasing the model size had the effect of improving the statistics of the runs but made the Tg value more difficult to determine. In essence we are providing more degrees of freedom for the system with the larger models and thereby increasing the error (by comparison with experiment). The equilibration problem increases with larger model sizes. A survey of the data produced by varying cell size is given in Table S1.

Effect of Simulation Length

The simulations that were run for 1 ns showed that, the statistics (in terms of the scatter of points) were much improved. However, the determination of the Tg value was more problematic, with a shift in the predicted value compared to the experiment. Additionally, the longer timescale served to remove the high temperature transition, which we have assigned to the thermal degradation of the material.

Hysteresis of the Model

It has been noticed in previous work by ourselves and others that in order to get results that correlate with experiment, it is necessary to run the simulation by cooling down from the high temperature regime to the lower. If the simulation is run by heating up the system, the Tg value determined can vary by 30 degrees. This hysteresis is a function of the equilibration problem mentioned above and the cooling down is thought to serve to freeze out high energy motions thereby providing a better correlation with experiment.

Neat Polymers

The plot of simulated density versus temperature from the MD simulation for the neat polycyanurate [BADCy] is shown in Figure 3. The black crosses are the equilibrated volume at that simulated temperature and the red solid line is the derivative of the plot using the method outlined in Hall et al. [34]. There are several features apparent in the plot which correlate well with empirical data. At lower temperatures (80°C), a small drop in the density is apparent, which we have termed the γ transition. This transition corresponds to the melt endotherm in Differential Scanning Calorimetry data of the uncured monomer. We have observed this experimentally at 82°C [27]. A second drop is also observed at 175°C corresponding to the β transition observed in the DMTA data for this polymer [28]. The glass transition (α transition) is marked by the larger fall at 260°C. The large drop in density at 350°C and 420°C may correspond to the thermal degradation of the polymer. The group bridging the two aromatic rings in this polymer has two pendant methyl groups.

The simulation of the second neat polycyanurate [LECy] (Figure 4) shows a higher γ transition at 195°C, a higher β transition at 235°C and a higher α transition at 310°C with the main change in structure being the replacement of one of the pendant methyl groups in the bridging group by a hydrogen. This small change in atomic architecture has larger implications for the thermo- mechanical properties of the resulting polymer. The simulation of the third neat polycyanurate [PT30] (Figure 5)shows the γ transition at 90°C, the β transition at 260°C and the highest α transition (Tg) at 360°C, with the Degradation (Td) at ca. 400°C and 480°C. The higher glass transition temperature results from the much reduced flexibility of this polymer with both of the pendant groups effectively reduced to hydrogens.

Effects of Parameterisation

In previous work we reparameterised the Dreiding force field [35] specifically for cyanurate rings [36]. A study of the available x-ray crystal structure data for cyanurate rings showed that the bond lengths are not equal and that the ring was slightly puckered. The new parameterization improved the simulation results significantly. The second generation force field used in this work is generally better all round than the first generation one used earlier but there is still the option to use alternating single and double bonds in the cyanurate rings or to use delocalized aromatic rings. By default Materials Studio chooses atom types as alternating single and double bonds but we tested the aromatic rings by running the simulations with wholly aromatic rings (Figure 6). An example is shown below for the 70∶30 blend of monomer 1 and 2 (Figures 7 and 8). There is more scatter on the data for the aromatic parameterization and the thermal events are different, i.e. α transition (Tg) = 250°C, β transition = 350°C, γ transition = 110°C, degradation (Td) = i) 400°C ii) 430°C for the alternating single and double bonds to α transition (Tg) = 295°C, β transition = 210°C, γ transition = 90°C, degradation (Td) = i) 420°C ii) 480°C for the aromatic parameterization. The experimentally determined values of the α transition (Tg) = 292°C and the degradation (Td) = i) 416°C. Hence the aromatic parameterization gives the better comparison with experiment, so was the one used.

Binary Blends

The corresponding data for the binary blend (50% of the first two polycyanurates) comprises elements of both networks: displaying a flatter response at lower temperatures (where the behaviour of monomer 1 predominates evidenced by the lower γ transition at 80°C similar to the first monomer, but appears more similar to monomer 2 at higher temperature with α transition (Tg) at 295°C and the β transition at 220°C. Interestingly the blending has pushed the degradation onset to 390°C with the second stage at 440°C, which is more similar to the third monomer.

A systematic study of the blends of the three monomers was carried out from 50∶50 of each of the monomers to 90∶10 and the alpha, beta, gamma and degradation onset temperatures for each were simulated. These results are presented in Table 2. It can be seen that there is a general trend towards the thermal data peaking for the 80∶20 and 90∶10 blends, although this is not clear cut in all cases. The predicted alpha transitions and degradation onset temperatures were compared with actual experimental values for all cases (Table 3). A visual comparison of the experimental TGA and DMTA data plotted against the simulated data is given in supporting information, Table S2. As noticed for the binary blend discussed earlier there is a general trend to favour the behaviour of the less thermally stable monomer at lower temperatures and the more thermally stable monomer at higher temperatures. There is good agreement in almost all cases except for the blend of 70∶30 of monomer 3 and 1, where there are 2 inflexions in the simulated data at 280 and 370 K and the experimental data gives a value of 380 K. The three dimensional structure of this blend is shown in Figure 9. This blend also shows an anomaly at the 50∶50 blend, again of monomer 3 and 1 (Figure 10). Here there are also 2 inflexions but also 2 inflexions in the experimental data, which are close to the experimental values. This blend seems to be behaving as 2 independent networks both in the simulation and also in the experiment.

Molecular simulation of cyanurate polymers is shown to reproduce with a good degree of accuracy the thermal events occurring experimentally for three different cyanate containing monomers. Furthermore it is also possible to simulate the thermal events of binary blends of the three monomers, again with good agreement to experiment. These thermal events, particularly the α transition, more commonly known as the Tg, are important in selecting materials for end use applications. The Tg determines the thermal softening point of the polymer and is crucially important in e.g. aerospace applications where it is not desirable for the polymer to exceed its Tg in use. Therefore molecular simulation can be used as a tool to screen and design polymers and blends thereof for end use applications.


Supporting Information Table S1

A survey of the data produced by varying cell size.

(DOCX)


Click here for additional data file (pone.0044487.s001.docx)

Table S2

A visual comparison of the experimental TGA and DMTA data plotted against the simulated data.

(DOCX)


Click here for additional data file (pone.0044487.s002.docx)


We thank Lonza AG for funding this work in the form of a PhD studentship (AOC) and Stefan Ellinger and Gaetano La Delfa for useful discussions.


References
1. Hamerton I (Ed.) (1994) Chemistry and Technology of Cyanate Ester Resins. Glasgow: Blackie A&P.
2. Tao QS,, Gan W,, Yue Y,, Wang MH,, Tang X,, et al. (Year: 2004) Viscoelastic effects on the phase separation in thermoplastics modified cyanate ester resin. Polymer45: 3505–3510
3. Nair CPR, Mathew D, Ninan K (2001) Cyanate Ester Resins, Recent Developments. In: New Polymerization Techniques and Synthetic Methodologies, Springer: Berlin, Vol: 155, 1–99.
4. Karad SK,, Attwood D,, Jones FR, (Year: 2002) Moisture absorption by cyanate ester modified epoxy resin matrices. Part III. Effect of blend composition. Composites Part A33: 1665–1675
5. Nair CPR,, Vijayan FT,, Krishnan K, (Year: 1999) Sequential interpenetrating polymer networks from bisphenol A based cyanate ester and bimaleimide: Properties of the neat resin and composites J. Appl. Polym. Sci. 74: 2737–2746
6. Guenthner AJ,, Lamison KR,, Vij V,, Reams JT,, Yandek GR,, et al. (Year: 2012) New Insights into Structure-Property Relationships in Thermosetting Polymers from Studies of Cocured Polycyanurate Networks Macromolecules. 45: 211–220
7. Pan R,, Liu X,, Zhang A,, Gu Y, (Year: 2007) Molecular simulation on structure-property relationship of polyimides with methylene spacing groups in biphenyl side chain, Computational Materials Science. 39(4): 887–895
8. Fan HB,, Yuen MMF, (Year: 2007) Material properties of the cross-linked epoxy resin compound predicted by MD simulation, Polymer. 48(7): 2174–2178
9. Tack JL,, Ford DM, (Year: 2008) Thermodynamic and mechanical properties of epoxy resin DGEBF crosslinked with DETDA by molecular dynamics. Journal of Molecular Graphics and Modelling26(8): 1269–127518222717
10. Gou J,, Minaie B,, Wang B,, Liang Z,, Zhang C, (Year: 2204) Computational and experiment study of interfacial bonding of single-walled nanotube reinforced composites, Computational Materials Science. 31(3–4): 225–236
11. Wu C,, Xu W, (Year: 2006) Atomistic Molecular modelling of crosslinked epoxy resin, Polymer. 47: 6004–6009
12. Wu C,, Xu W, (Year: 2007) Atomistic molecular simulations of structure and dynamics of crosslinked epoxy resin. Polymer48: 5802–5812
13. Lin YC,, Xu C, (Year: 2005) Investigation of moisture diffusion in epoxy system: Experiments and MD simulations. Chemical Physics Letters412: 322–326
14. Hamerton I,, Howlin BJ,, Klewpatinond P,, Shortley HJ,, Takeda S, (Year: 2006) Developing predictive models for polycyanurates through a comparative study of molecular simulation and empirical thermo-mechanical data. Polymer47: 690–698
15. Hamerton I,, Howlin BJ,, Mitchell AL, (Year: 2006) Developing poly(bis-benzoxazines) with improved fracture toughness. 1: Using molecular simulation to determine and predict structure-property relationships. Reactive & Functional Polymers66: 21–39
16. Hofmann D,, Fritz L,, Ulbrich J,, Paul D, (Year: 2000) Molecular simulation of small molecule diffusion and solution in dense amorphous polysiloxanes and polyimides. Computational and Theoretical Polymer Science10: 419–436
17. Rui P,, Xikui L,, Zhang A,, Gu Y, (Year: 2007) Molecular simulation on structure-property relationship of polyimides with methylene spacing groups in biphenyl side chain. Computational Materials Science39: 887–895
18. Neyertz S,, Douanne A,, Brown D, (Year: 2006) A molecular dynamics simulation study of surface effects on gas permeation in free-standing polyimide membranes. Journal of Membrane Science280: 517–529
19. Ren H,, Zhang O,, Chen X,, Zhao W,, Zhang J,, et al. (Year: 2007) A molecular simulation study of a series of cyclohexanone formaldehyde resins: Properties and applications in plastic printing. Polymer48: 887–893
20. Rigby D,, Sun H,, Eichinger B, (Year: 1997) Computer Simulations of Poly(ethylene oxide): Force Field, PVT Diagram and Cyclization Behavior. Polymer International44: 311–330
21. Sun H,, Rigby D, (Year: 1997) Polysiloxanes: Ab Initio Force Field and Structural, Conformational and Thermophysical Properties. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 53: 1301–1323
22. Fried JR,, Li B, (Year: 2001) Atomistic simulation of the glass transition of di-substituted polysilanes. Computational and Theoretical Polymer Science. 11: 273–281
23. Pavel D,, Shanks R, (Year: 2005) Molecular dynamics simulation of diffusion of O2 and CO2 in blends of amorphous poly(ethylene terephthalate) and related polyesters. Polymer. 46: 6135–6147
24. Gou J,, Minaie B,, Wang B,, Liang Z,, Zhang C, (Year: 2004) Computational and experiment study of interfacial bonding of single-walled nanotube reinforced composites. Computational Materials Science. 31: 225–236
25. Grujicic M,, Sun YP,, Koudela KL, (Year: 2007) The effect of covalent functionalization of carbon nanotube reinforcements on the atomic-level mechanical properties of poly-vinyl-ester-epoxy. Applied Surface Science. 253: 3009–3021
26. Materials Studio, Accelrys.com. Available: http://accelrys.com/products/materials-studio/ Accessed 14, March 2012.
27. Auhl R,, Everaers R,, Grest GS,, Kremer K,, Plimpton SJ, (Year: 2003) Equilibration of Long Chain Polymer Melts in Computer Simulations. Journal of Chemical Physics. 119: 12718–12728
28. Crawford AO,, Howlin BJ,, Cavalli G,, Hamerton I, (Year: 2012) Examining the thermo-mechanical properties of novel cyanate ester blends through empirical measurement and simulation. Reactive & Functional Polymers. 72: 596–605
29. Maple JR,, Dinur U,, Hagler AT, (Year: 1988) Derivation of force fields for molecular mechanics and dynamics from ab initio energy surfaces. Proc. Natl. Acad. Sci. 855350–5354
30. Sun H, (Year: 1998) COMPASS: An ab Initio Force-Field Optimized for Condensed-Phase Applications - Overview with Details on Alkane and Benzene Compounds. J. Phys. Chem. B. 102: 7338–7364
31. Powell MJD, (Year: 1977) Restart procedures for the conjugate gradient method, Mathematical Programming. 12: 241–254
32. Andersen HC, (Year: 1980) Molecular dynamics simulations at constant pressure and/or temperature. J. Chem. Phys. 72: 2384–2394
33. Parinello M,, Rahman A, (Year: 1981) Polymorphic Transitions in Single-Crystals - A New Molecular-Dynamics Method. J. Appl. Phys. 52: 7182–7190
34. Hall SA, Howlin BJ, Hamerton I, Baidak A, Billaud C, et al. (2012) Solving the Problem of Building Models of Crosslinked Polymers: An Example Focussing on Validation of the Predicted Properties of Crosslinked Epoxy Resins. PLOSONE, In press.
35. Mayo SL,, Olafson BD,, Goddard WA, (Year: 1990) Dreiding- A Generic Force Field for Molecular Simulations. J. Phys. Chem. 94: 8897–8909
36. Allington RD,, Attwood D,, Hamerton I,, Hay JN,, Howlin BJ, (Year: 2003) New force-field parameters for molecular simulations of s-triazine and cyanurate-containing systems. 2–Application and comparison with different simulation methods. Polymer44: 793–799

Figures

[Figure ID: pone-0044487-g001]
doi: 10.1371/journal.pone.0044487.g001.
Figure 1  Cyanate ester structures studied in this work. (N.B., in this instance n = 1 in structure (3)).

[Figure ID: pone-0044487-g002]
doi: 10.1371/journal.pone.0044487.g002.
Figure 2  Scheme of cyclotrimerisation to triazine.

[Figure ID: pone-0044487-g003]
doi: 10.1371/journal.pone.0044487.g003.
Figure 3  Results of Molecular dynamics simulation of density versus temperature for the homopolymer of BADCy.

[Figure ID: pone-0044487-g004]
doi: 10.1371/journal.pone.0044487.g004.
Figure 4  Results of molecular dynamics simulation of density versus temperature for the homopolymer of LECy.

[Figure ID: pone-0044487-g005]
doi: 10.1371/journal.pone.0044487.g005.
Figure 5  Results of molecular dynamics simulation of density versus temperature for the homopolymer of PT30.

[Figure ID: pone-0044487-g006]
doi: 10.1371/journal.pone.0044487.g006.
Figure 6  Image of delocalised triazine ring formed through polymerisation.

[Figure ID: pone-0044487-g007]
doi: 10.1371/journal.pone.0044487.g007.
Figure 7  Results of molecular dynamics simulation of density versus temperature for the blend of monomer 1 and 3 in a 70∶30 ratio using alternating single and double bonds.

[Figure ID: pone-0044487-g008]
doi: 10.1371/journal.pone.0044487.g008.
Figure 8  Results of molecular dynamics simulation of density versus temperature for the blend of monomer 1 and 3 in a 70∶30 ratio using aromatic bonds.

[Figure ID: pone-0044487-g009]
doi: 10.1371/journal.pone.0044487.g009.
Figure 9  Image of [370

130].



[Figure ID: pone-0044487-g010]
doi: 10.1371/journal.pone.0044487.g010.
Figure 10  Results of molecular dynamics simulation of density versus temperature for the blend of BADCy:LECy in a 50∶50 blend.

Tables
[TableWrap ID: pone-0044487-t001] doi: 10.1371/journal.pone.0044487.t001.
Table 1  Designation of monomers and blends examined in this work. The data represent the percentage of each monomer in the blend.
Sample Designation (1) (2) (3)
(1) 100 0 0
(2) 0 100 0
(3) 0 0 100
(150–250) 50 50 0
(160–240) 60 40 0
(170–230) 70 30 0
(180–220) 80 20 0
(190–210) 90 10 0
(350–150) 50 0 50
(360–140) 40 0 60
(370–130) 30 0 70
(380–120) 20 0 80
(390–110) 10 0 90
(350–250) 0 50 50
(360–240) 0 40 60
(370–230) 0 30 70
(380–220) 0 20 80
(390–210) 0 10 90

[TableWrap ID: pone-0044487-t002] doi: 10.1371/journal.pone.0044487.t002.
Table 2  Thermal data from molecular simulation (Transition and onset temperatures in °C).
Model Composition No. of atoms α Transition (Tg) β Transition γ Transition Degradation onset (Td)
(1) 4778 260 175 80 350 420
(2) 5553 310 235 195 405
(3) 5230 360 260 90 400 480
[(1)50–(2)50] 3580 295 220 80 390 440
[(1)60–(2)40] 4470 290 195 100 395 460
[(1)70–(2)30] 4604 250 350 110 400 430
[(1)80–(2)20] 4276 280 310 75 380 420
[(1)90–(2)10] 4380 295 220 80 400 440
[(3)50–(2)50] 5348 350 270 80 395 420
[(3)60–(2)40] 4704 350 250 110 405 420
[(3)70–(2)30] 5658 300 350 230 105 400 420
[(3)80–(2)20] 5920 370 260 120 410 460
[(3)90–(2)10] 5552 290 360 180 95 410 450
[(3)50–(1)50] 5409 305 360 175 75 400 450
[(3)60–(1)40] 5798 305 360 160 80 420 460
[(3)70–(1)30] 6016 280 370 205 80 480
[(3)80–(1)20] 6224 260 360 110 75 400 450
[(3)90–(1)10] 6040 310 370 120 60 405 490

[TableWrap ID: pone-0044487-t003] doi: 10.1371/journal.pone.0044487.t003.
Table 3  Comparison of predicted and experimental thermal data (all +/−5°C).
Blend Predicted Tg(1/2) Actual Tg Predicted Td(1/2) Actual Td
[(1)50–(2)50] 295 291.83 390/440 414
[(1)60–(2)40] 290 291.59 395/460 411
[(1)70–(2)30] 250 292.72 400/430 416
[(1)80–(2)20] 280 286.3 380/420 413
[(1)90–(2)10] 295 287.76 400/440 400
[(3)50–(2)50] 350 336.9 395/420 420
[(3)60–(2)40] 350 331.83 405/420 420
[(3)70–(2)30] 300/350 331.72 400/420 408
[(3)80–(2)20] 370 325.27 410/460 420
[(3)90–(2)10] 290/360 410/450 418
[(3)50–(1)50] 305/360 311.82/343.83 400/450 416
[(3)60–(1)40] 305/360 307.82 420/460 419
[(3)70–(1)30] 280/370 380.62 480 420
[(3)80–(1)20] 260/360 325.27 400/450 420
[(3)90–(1)10] 310/370 405/490 419


Article Categories:
  • Research Article
Article Categories:
  • Chemistry
    • Computational Chemistry
      • Molecular Dynamics
      • Molecular Mechanics
    • Polymer Chemistry
      • Monomers
      • Polymerization
    • Theoretical Chemistry
Article Categories:
  • Materials Science
    • Material by Structure
      • Polymers
        • Thermosetting Polymers
    • Materials Design


Previous Document:  Nicorandil prevents right ventricular remodeling by inhibiting apoptosis and lowering pressure overl...
Next Document:  Dietary monounsaturated fatty acids intake and risk of skin photoaging.