Analytical Approaches Towards Understanding Structure-Property

LLNL-CONF-578735

Analytical Approaches TowardsUnderstanding Structure-PropertyRelationships in End-Linked Model PDMSNetworks

J. P. Lewicki , R. L. F. Albo, C. T. Alviso, M.Ashmore, S. J. Harley, J. A. Finnie, R. S. Maxwell

September 4, 2012

MODEST 2012Prauge, Czech RepublicSeptember 2, 2012 through September 7, 2012

Disclaimer

This document was prepared as an account of work sponsored by an agency of the United States government. Neither the United States government nor Lawrence Livermore National Security, LLC, nor any of their employees makes any warranty, expressed or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States government or Lawrence Livermore National Security, LLC. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States government or Lawrence Livermore National Security, LLC, and shall not be used for advertising or product endorsement purposes.

LLNL-PRES-XXXXXXThis work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344. Lawrence Livermore National Security, LLC

James P. Lewicki, Rebecca L. F. Albo, Cynthia T. Alviso, Jasmine Finnie, Stephen J. Harley, Michael Ashmore and

Robert S. Maxwell

Lawrence Livermore National Laboratory LLNL-PRES-xxxxxx2

Δαβɣ-H2OLOAD

Bulk

Netw

ork structure

Chem

ical identity

Structural complexity over a range of size-scales

Complex engineering silicones

Aging and degradation mechanisms

operate over a broad range of

size scales

Macro

Micro

Molecular

Suffer aging and degradation

Aging and degradation over a range of size-scales


Solid state NMR

Therm. Ana.

Mech. Ana. BDS sorption

studies


Pyrolysis-Gas Chromatography/Mass Spectrometry

Interrogative thermal analysis of siloxane network architecture

• Sample• He carrier

gas

Probe head

• Δ• Analytes

evolvedFurnace

• Analyteseparation

• MS detection

GCMS

Small sample size (0.05mg):

• Non-diffusion limited regime

Versatile probe designs:

• ‘low & slow’ out-gassing • thermal degradation

Cryo-GC/MS:

• High sensitivity• Efficient separation

• broad mass detection


What Information can Py-GCMS provide?Pyrolysis-evolved gas analysis (Py-EGA)

(20 ⁰C/min) with GC column in a non-eluting mode, Carbon NT/PDMS composites

250 500 750 10000

1000000

2000000

3000000

4000000

5000000

6000000

7000000

8000000

9000000

MS

D R

espo

nse

Temperature /oC

CA 0_2 (0% SWNT) CA 1_2 (6% SWNT) CA 2_2 (13% SW NT) CA 10_2 (63%% SW NT)

Py-EGA gives us ‘bulk’ degradation profiles:

• Onset & peak degradation temp.

• ‘Yield’ of volatile degradation products


5 10 15 20 25 30 350

2000000

4000000

6000000

8000000

10000000

12000000

14000000

16000000M

SD

det

ecto

r res

pons

e

Retention time (min)

Unfilled

(1000 ⁰C/min) with GC column in an eluting mode

Ballistic PY yields chemical information

• Speciation of degradation products

• yields of individual species

• Informs mechanistic models

What Information can Py-GCMS provide?Ballistic pyrolytic analysis


Siloxanes thermally degrade via well established chemistries to form a series of cyclic oligomers

We’ve shown that network structure can influence the stability of a PDMS network in the bulk*.

?

1) Lewicki JP, Mayer, BP, Alviso CT, and Maxwell, RS.” Journal of Inorganic and Organometallic Polymers and Materials (2012) 22:636–645).


OSi

O

SiO

Si

1 Lewicki JP, Mayer, BP, Alviso CT, and Maxwell, RS.” Journal of Inorganic and Organometallic Polymers and Materials (2012) 22:636–645).2) Lewicki JP, Liggat JJ and Patel M. Polymer Degradation and Stability 2009, 94, 1548-1557.

Using Quantitative GC-Pyrolysis we have shown that the D3 cyclic remains the principle product of degradation, irrespective of network

structure1,2

Studied Architectures:

• Mono-modal• Bi-Modal• Free-chain end systems• Above and below

entanglement• (liquid precursors)


• The thermodynamic product of thermal degradation in PDMS is not sensitive to the subtleties of network architecture.

• Using a statistical Analysis technique we can examine the relationship between network architecture and minor product formation

• We observe clear clustering of degradation profiles as a function of network structure

Quantifiable relationships exist between basic network architectures and the distributions of degradation derived species in PDMS networks.

Lewicki JP, Mayer, BP, Alviso CT, and Maxwell, RS.” Journal of Inorganic and Organometallic Polymers and Materials(2012) 22:636–645).


Our analysis suggests that at shorter chain lengths/high x-link densities larger cyclicsare favored. However at low x-link densities and long chain lengths smaller cyclicsdominate

A) Short chains high X-link density

B) long chains low X-link density

• The additional constraint of high x-link density networks makes the formation of larger cyclics somewhat more conformationalyfavorable

• In low x-link density systems above entanglement random coil behavior dominates and the majority of a chain does not ‘see’ a x-link site, the main driver is now thermodynamics and small cyclic formation dominates


• Analytical pyrolysis coupled with PCA can map the degradation profiles of chemically similar engineering silicone elastomers

• Materials group by Network modality and catalyst chemistry

• Neither aging or irradiation significantly alter a materials pyrolysis degradation profile.

Lewicki, JP, Albo RLF, Alviso, CT and Maxwell, RS. “Pyrolysis-gas chromatography/mass spectrometry for the forensic fingerprinting of silicone engineering elastomers” Journal of Analytical and Applied Pyrolysis (accepted)


• As with the model systems, architectural features such as phenyl free chain ends or high x-link densities ‘skew’ the profile of minor degradation products providing a fingerprint

Lewicki, JP, Albo RLF, Alviso, CT and Maxwell, RS. “Pyrolysis-gas chromatography/mass spectrometry for the forensic fingerprinting of silicone engineering elastomers” Journal of Analytical and Applied Pyrolysis (accepted, in press)


The synthesis of well defined end-linked model PDMS networks allows us to decouple and investigate the effects of individual architectural motifs on the properties of a elastomer system

X-link density, chain length, modality and level of FCE’s all influence the degradation chemistry of a silicone material

Here we have demonstrated the utility of interrogative thermal analytical analysis methodologies for the study of otherwise intractable materials


Thank you.



Matrix of data

Variables

Sam

ples

(n)

(m)

(X)

Principle component analysis

Small number of new variables or “components” that maintain structure of the data (i.e., the basic information) but that can better capture the data’s variance.

These components give information on:-How samples are (or aren’t) related to one another-The degree to which variables influence the overall structure of the data.

COMPLEX = scale, similarity, etc.Examples of variables (n): Pressure, temperature, absorbance, intensity, etc.Examples of samples (m): Data samples at specific time points, # of samples taken in a study.


Ex.: Elliptically distributed x-y-z data.

Variable 2Variable 1

Varia

ble

3

Component 2

Component 1

Principle Component 1:- Accounts for the largest variance (spread) in the data.

Principle Component 2:-Orthogonal to PC1, accounts for next largest variance.

Principle Component 3:-Direction forced by # of variables (3)-Not necessary!!! – little variance in last remaining new dimension

Dimension Reduction by discarding unnecessary components…

…unimportant ‘variables’…spectral noise…etc.

PCA can also identify outliers and discriminate between outliers within the model and outside the model.


Loadings identify the variables that contribute to the grouping and/or separation.

Scores describe the relationship of the samples – how the samples group to (or separate from) each other.

Matrix of data

Variables

Sam

ples

Scores

AAA

AB

B B

CC

CC

D

DD

D

*

13

14 15

16

17 12

18

1920

21

22

Loadings


-20000 0 20000 40000 60000 80000100000120000140000160000180000

4.5

4.6

4.7

4.8

4.9E

ps'

Time [s]

Model ExpDec1Equation y = A1*exp(-x/t1) + y0

Reduced Chi-Sqr

1.91314E-6

Adj. R-Square 0.99935Value Standard Error

Eps' y0 4.48995 3.10282E-4Eps' A1 0.41781 0.00103Eps' t1 51045.60943 189.84508


Purity & polydispersity of the starting materials

3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5

0.0

0.5

1.0

1.5

2.0 68K 8K 10K 32K 54K 136K

WF/

dLog

M

log(mol Wt.)

Si O Si

CH3

CH3

O OH

CH3

CH3

Si

CH3

CH3

OHSi[O(CH2)2CH3]4+

SnII[O2C(C2H5)CH(CH2)3CH3]

n

+CH2

CH2

CH3OH

x4

Hydroxy-terminated PDMS Crosslinked PDMS

2H2O R-Sn(II)[OH]x + OH

O

CH3

CH3

Tin(II)ethylhexanoate

Si-Si-O-PDMS-O-SiO

OSi-O-PDMS-O-Si-

OSi-O-PDMS-O-Si-

-Si-O-PDMS-O-SiO

Active catalyst

Ethylhexanoic acid

Tetrapropylorthosilicate (TPOS) crosslinker

Unwanted side reactions

Complex cure kinetics


1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.00.00000

0.00002

0.00004

0.00006

0.00008

0.00010

0.00012

0.00014

0.00016

0.00018

0.00020

0.00022

0.00024

Ave

. XLD

stoic eqv. on silane

0.0

0.2

0.4

0.6

0.8

1.0

Sol.

frac

tion

2 3 4 5 6 7 8-0.000005

0.000000

0.000005

0.000010

0.000015

0.000020

0.000025

0.000030

0.000035

0.000040

0.000045

0.000050

Ave

. XLD

stoic eqv. on silane

0.0

0.2

0.4

0.6

0.8

1.0

Sol.

frac

tion

0 1 2 3 4 5 6 7 8

0.000000

0.000002

0.000004

0.000006

0.000008

0.000010

0.000012

0.000014

Ave

. XLD

Stoic eqv. on silane

0.0

0.2

0.4

0.6

0.8

1.0So

l fra

ctio

n

Assessment of crosslink density with the application of F-H network theory.

~6000 g mol-1 ~28,000 g mol-1

~115,000 g mol-1


4.6 4.7 4.8 4.9 5.0

0

10000

20000

30000

8:17.5:1

6.5:16:1

5.5:15:1

ppm

1:1

5.0 5.5 6.0 6.5 7.0 7.50

5000

10000

2:1

ppm

1:1

1H MAS – Can directly observe the excesses of functional groups in partially gelled systems

Decreasing vinyl signal

Increasing levels of Silanol FCE’s

Magic Sandwich Echo allows protons associated with regions of differing

mobility to be assessed

Under-cure Vinyl

excess

Fully cured

network

Under-cure silane

excess

Processed data yields distributions of protons with distinct relaxation times

A homogenous network structure is approached

=> distinct pops. converge

The progress of a structure towards a homogenous, ordered network can therefore be followed using convenient, non-invasive and easy to implement solid state NMR

methodologies

Maus et. al Macro. Chem. Phys. 207, 13, 2006


Water sorption studies yield insight into the dynamics

of siloxanenetworks

• In contrast to commercial and end-use formulations, the uptake behavior of the optimized model systems is linear and single component.

• Residual functionality and void volume in Sylgard-184 and the LLNL end-use formulation (M97) lead to significant Langmuir and pooling sorption modes – absent in the model system.

• In the single component Henrys law behavior of the model network, we are observing the fundamental sorption behavior of the fully dense network

Void occupation

Interaction withFG’s

Simple networkoccupation

LLNL’s M97

Sylgard 184

Monomodal Model Network


0.01 0.1 1 10 100 1000 10000

0.0

0.2

0.4

0.6

0.8

1.0 68K Mono 8 & 68K Bimodal 68K 20% FCE's

T2 /ms

Solid state NMR can be tasked as a powerful tool for the analysis of otherwise intractable polymeric network materials

Basic T2 measurements have been shown to be

sensitive network changes in the solid state

Multiple quantum NMR yield NMR derived distributions of MW through <DQ> - the residual

dipolar coupling constant. MQ NMR is also sensitive to chain entanglements and Mc

Mayer, Lewicki ,Weisgraber et al. MACROMOLECULES 44 8106 2011

Documents

Analytical Approaches Towards Understanding Structure-Property