Upload
duongthuy
View
242
Download
1
Embed Size (px)
Citation preview
Polymer Characterization:Past, Present and Future
Guy C. Berry
Department of Chemistry
Carnegie Mellon University
Penn-Ohio Border SectionAm. Chem. Soc
February 2001
A copy of the Lecture Notes may be found at the URL:
http://www.chem.cmu.edu/Berry
F. W. Billmeyer Jr. (1976):J. Polym. Sci.: Symp. (1976) 55: 1-10
"…characterization of polymers is inherentlymore difficult than that of other materials.Polymers are roughly equivalent incomplexity to, if not more complex than,other materials, at every physical level oforganization from microscopic tomacroscopic…"
"We would wish, ideally, to characterize allaspects of a polymer structure in enoughdetail to predict its performance from firstprinciples. I seriously doubt that this will everbe possible, and I am sure that even if itwere, it would never be economicallyfeasible."
Some Reference MaterialNew developments in polymer analytics I & II
M. Schmidt, ed. Adv. Polym. Sci., ed. Abe; Vol. 150-1, Springer: Berlin, 2000, 190.
Light Scattering, Classical: Size and Size Distribution ClassificationG. C. Berry, In Encyclopedia of Analytical Chemistry; R. A. Meyers, ed.; John Wiley & Sons Ltd:Chichester, UK, 2000; 5413-48.
Static and dynamic light scatteringG. C. Berry; P. M. Cotts, In Experimental methods in polymer characterization; R. A. Pethrick; R. S.Stein, eds.; John Wiley & Sons Ltd.: Sussix, UK, 1999.
Separation and Analysis of Peptides and Proteins Andreas F. R. Huehmer; Gabi I. Aced; Melissa D. Perkins; R. Neslihan Guersoy; D. S. SeetharamaJois; Cynthia Larive; Teruna J. Siahaan; Christian Schoeneich, Anal. Chem. 69(12), 29R-57R (1997).
Optimization of liquid chromatography-NMR spectroscopy. II-saturation and flow in on-flow liquidchromatography-NMR spectroscopyLee Griffiths, Magn. Reson. Chem. 35(4), 257-61 (1997).
The development and application of coupled HPLC-NMR spectroscopyJohn C. Lindon; Jeremy K. Nicholson; Ian D. Wilson, Adv. Chromatogr. 36, 315-82 (1996).
SEC Absolute Molar Mass Detection by Online Membrane OsmometryU. Lehmann; W. Koehler; W. Albrecht,Macromolecules 29(9), 3212-15 (1996).
Analysis of Synthetic Polymers and RubbersCharles G. Smith; Patrick B. Smith; Andrew J. Pasztor, Jr.; Marianne L. McKelvy; David M. Meunier;Stephen W. Froelicher, Anal. Chem. 67(12), 97-126 (1995).
Chromatographic characterization of polymers: Hyphenated and multidimensional techniquesTheodore Provder; Howard G. Barth; Marek W. Urban, eds. Advances in Chemistry Series, ed. R. J.
Alaimo et al; Vol. 247, American Chemical Society: Washington, DC, 1995.
Polymer Analysis and CharacterizationA. Abe et al., ed. Advances in Polymer Science, ed. A. Abe et al.; Vol. 114, Springer-Verlag: New
York, 1995.
Modern analytical ultracentrifugation: Acquisition and interpretation of data for biological andsynthetic polymer systems
Todd M. Schuster; Thomas M. Laue, eds., ed. Birkhauser: Boston, 1994.
Hypenated and multidimensional techniquesT. Provder; M. W. Urban; H. G. Barth, eds. ACS Symposium series, ed. R. J. A. e. al; Vol. 581,
American Chemical Society: Washington, DC, 1994.
Hyphenated techniques in polymer characterization: Thermal-spectroscopic and other methodsTheodore Provder; Marek W. Urban; Howard G. Barth, eds. ACS Symposium Series, ed. Vol. 581,
American Chemical Society: Washington, DC, 1994.
Polymer analysis and characterizationA. Abe et al, ed. Adv. Polym. Sci., ed. Vol. 114, Springer-Verlag: New York, 1994, 233-90.
Modern methods of polymer characterizationHoward G. Barth; Jimmy W. Mays, eds. Chemical analysis, ed. Vol. 113, J. Wiley: New York, 1991.
Polymer characterization by liquid chromatography,Gottfried Glöckner. Journal of chromatography library. Vol. 34; Elsevier: New York, 1987.
Light scatteringG. C. Berry, Encycl. Polym. Sci. Eng. 8, 721-94 (1987).
Microscopic Characterization Needed at ManyResolutions:
AABA
B B A A B A BBAABA B B
C C CC C
Characterization Needed for a Variety ofMacroscopic Structures:
◊ The Amorphous State (including glasses,fluids and gels) over a wide range of lengthscales
◊ The (Partially) Crystalline State (includingfibers, biaxially oriented films and bulksamples)
◊ The Liquid Crystalline State (including main-chain and side-chain mesogenic polymers)
◊ Polymers in Confined Environments
◊ Copolymers, including the morphology ofmicro-phase separated systems
A Stipulation:Given the very large span of topicsencompassed in Characterization, this shortpresentation will be purposely limited to anarrow range of topics:
A subset of the characterization methodsapplicable to techniques involving dilutesolutions, recognizing that this neglects aWorkshop full of important topics.
The presentation will be structured aroundthree major sections:
◊ A Reminder of Times Past--The principal methods for characterizingthe distribution of certain molecularproperties available cira 1950-70
◊ A Survey of the Present--Some of the principal methods in currentuse to characterize molecular properties,including molecular weight andcomposition distribution, etc.
◊ Some Thoughts on the Future
A Reminder of Times Past◊ Liquid-Liquid Separations
⇒ Separation by molecular weight for linearhomopolymers
⇒ Complex separation for branchedhomopolymers
⇒ Complex separation for linear orbranched copolymers
⇒ Often problems with crystallizablepolymers
⇒ Slow and labor intensive; large volumesof solvent
⇒ Not amenable to column separations
◊ Liquid-Solid Separations⇒ Most useful for crystallizable polymers
and sometimes for copolymers⇒ Separations often complex
◊ Miscellaneous Methods⇒ Refractive index, Infra-red and UV
spectroscopies, specific volume, end-group analysis
◊ Ultracentrifugation⇒ Mw, Mz, Mz+1, … by equilibrium methods⇒ Distribution of hydrodynamic radii RH by
sedimentation velocity (e.g., MWD forlinear homopolymers)
⇒ Sometimes requires complex analysis⇒ Complicated by branching and
copolymer distributions⇒ Requires an expert analyst and complex
equipment
◊ Osmotic Pressure⇒ Determination of Mn and A2,Π⇒ Mn independent of branching, copolymer
composition, etc.⇒ Requires a suitable membrane; limited to
solute neither too high nor too low inmolecular weight components
◊ Vapor Phase Osmometry⇒ A reasonable approximation to Mn for
very low molecular weight polymers (nota thermodynamic method)
⇒ Subject to errors difficult to control andassess
◊ Light Scattering (Static)⇒ Determination of MLS, RG,LS and A2,LS
⇒ MLS = MW for homopolymer (independentof branching)
⇒ MLS, RG,LS and A2,LS complicated forcopolymers
⇒ Method requires a tedious, andsometimes difficult, optical clarification ofthe solution
⇒ The values of MLS and RG,LS may beseverely biased by a minor componentof aggregated species
⇒ Requires an expert analyst andspecialized equipment
◊ Viscometry⇒ Determination of [η] and k'⇒ The relation M[η] = Φ'RG
3 providesanalysis procedure ([η] = KMv
a for mostlinear homopolymers, with Mn < Mv < Mw
for flexible chains)⇒ Analysis complicated for branched chain
and copolymers⇒ Measurements possible with minimal
experience and simple equipment
◊ Chromatographic Separations⇒ Size exclusion chromatography using a
detector for eluent concentration⇒ Column separations based on solubility--
usually for crystalline polymers
An Example: The characterization of long-chainbranching in a homopolymer.
◊ No method was available to effect aseparation according to long-chain content ormolecular weight; Liquid-liquid separationgives a complex mixture according to theseattributes
◊ Sedimentation velocity ultracentrifugationprovided a weighted distribution of RH =h(RH)lin for the sample or fractions thereof,where h ≤ 1 depends on the long-chainbranch content (expected to vary with M insome samples)
◊ Equilibrium ultracentrifugation provided Mw,Mz, Mz+1, … for the sample or fractionsthereof
◊ Static light scattering provided Mw and RG,LS
for the sample or fractions thereof; RG =g1/2(RG)lin, where g1/2 ≈ h ≤ 1 depends on thelong-chain branch content (expected to varywith M in some samples)
◊ Osmometry provided Mn for the sample orfractions thereof (unless Mn is too large)
◊ Viscometry provided [η] for the sample orfractions thereof, where [η] = g'[η]lin, with g' ≈gh ≈ g3/2 ≤ 1 for some types of branching
◊ Analysis required assumption of some modelfor h, g and g', along with approximations ofthe effects on the measured RG,LS and [η] ofthe (a priori unknown) distributions in branchcontent and M that characterized any realsample (unless a model compound)
◊ Uncertainties in the analysis generally madedefinitive determination of branchingimpossible for a given sample, even with thedetermination of a number of parameters
A Survey of the Present◊ Size Exclusion Chromatography
⇒ Separation according to molecular size andshape
⇒ Separation essentially according tomolecular weight for linear homopolymers,including crystallizable chains
⇒ Amenable to the use of multiple detectors toanalyze the eluent and computer-aided datacollection and analysis
⇒ Complex separation for branchedhomopolymers
⇒ Complex separation for linear or branchedcopolymers
⇒ May be automated for enhanced efficiency
◊ Chromatography Based on Composition⇒ Separation by preferential adsorption⇒ Separation by melting⇒ Separation by preferential solubility⇒ Often complementary to SEC separation,
and may be followed or preceded by aseparation via SEC
Chain Length
Composition/Block Length
End Groups
Architecture
Molar Mass
Poly
mer
Con
cent
rati
on
All Components, mixed in an unknown w
A problem in the characterization of mixtures of polymeric compone
Molar Mass
Pol
ymer
Con
cent
ratio
n
Pol
ymer
Con
cent
ratio
n
Molar Mass
Pol
ymer
Con
cent
ratio
n
Molar Mass
"Two-dimensional chromatographic separation"
Composition
Pol
ymer
Con
cent
ratio
n
Pol
ymer
Con
cent
ratio
n
Molar Mass
Pol
ymer
Con
cent
ratio
n
Molar Mass
Separation by composition
Flow Control
Separation by Molecular Size
ConcentrationDetector
◊ Other Chromatographic Methods⇒ Field Flow Fractionation methods (flow
under the influencing of an external fieldorthogonal to the flow, e.g., centrifugalfield, cross-flow, temperature gradient,etc.)
⇒ Hydrodynamic chromatography(separation via the velocity gradient incapillary flow)
Flow In Flow out
Flow in annular region
Carrier Flow
Centrifugal Field
◊ Ultracentrifugation⇒ Only modest changes from prior art⇒ Not widely used with synthetic
polymers--more often used withbiological macromolecules. Cannot beused as an on-line SEC detector
◊ Osmotic Pressure⇒ Only modest changes from prior art⇒ Not widely used; some attempts to
develop an on-line detector
◊ Vapor Phase Osmometry⇒ No change from prior art⇒ Rarely used; cannot be used as an on-
line SEC detector
◊ Light Scattering (Static)⇒ Principles of data analysis only modestly
advanced in recent years⇒ Instrumentation advance by the use of
lasers for incident light (smaller scatteringvolumes possible), and the use ofcomputer-aided data analysis
⇒ Used as an on-line detector with SEC;optical clarification of the solutionautomatic in this case (provided thecolumn does not shed debris)
⇒ Instrumentation for on-line use with SECis often more primitive than desirable(e.g., some commercial models neglector approximate the angular dependence)
◊ Light Scattering (Dynamic)⇒ Determination of RH,LS and a weighted
distribution of RH
⇒ The distribution of RH coupled with theangular dependence of the staticscattering is often useful in characterizingsupramolecular association
⇒ Usually not suitable for on-line SECdetector (owing to sampling timerequirements)
⇒ Method requires a tedious, andsometimes difficult, optical clarification ofthe solution
⇒ Requires an expert analyst andspecialized equipment
◊ Viscometry⇒ Only minor changes from prior art except
for development of on-line instruments⇒ Used as an on-line detector for SEC⇒ Analysis in terms of molecular weight
distribution requires a molecular model
◊ High Resolution NMR⇒ Provides analysis of the statistics of
stereoregular and copolymer placements(both averaged over the sample studied)
⇒ Difficult to apply as an on-line SECdetector owing to sensitivity, butprototypes are under development
◊ Mass Spectroscopy⇒ Determination of a weighted (ideally, a
number) distribution of molecularweights; very low to moderately high M
⇒ Matrix-Assisted Laser-DesorptionIonization--Time of Flight MS(MALDI-TOF);Useful for a variety of polymers, up tomoderately high molecular weight(polymer degradation sometimesoccurs);Difficult to use as an on-line SECdetector, but prototypes have beentested
⇒ Electrospray Ionization-MS is suitable forlower molecular weight solute; use as anon-line SEC detector has beendemonstrated
◊ Atomic Force Microscopy & Related Methods⇒ A new and rapidly developing method to
explore surface features and moleculesat surfaces
⇒ A new and rapidly developing method,that is presently most effectively used bya specialist
S.S. Sheiko, Adv. Polym. Sci. (2000) 151:61-174
Example 1: Comparison of the conformationalproperties of homopolymers
◊ SEC permits separation of the sample into aneluent with approximately common RG orcommon M[η] at any given elution time
◊ On-line multiple angle static light scatteringand refractive index detectors permitmeasurement of Mw ≈ M, RG,LS ≈ RG and soluteconcentration as functions of the elution time.
1
10
100
1000
RG
/nm
; o
M/1
04;
o, o
Elution Volume
Ref
ract
ive
Ind
ex
5
20000
1000×RG2/Mvs√M
poly(di-n-hexylsilane) & poly(phenyl-n-hexylsilane)P. M. Cotts J. Poly. Sci.: Pt. B: Polym. Phys. (1994) 32:771-8.
Example 2: The characterization of long-chainrandom branching.
◊ SEC permits separation of the sample into eluentswith approximately common RG = g1/2(RG)lin orcommon M[η] = Mg'[η]lin at any given elution time
◊ On-line multiple angle static light scattering andrefractive index detectors permit measurement ofMw, RG,LS and solute concentration as functions ofthe elution time. Information on RG,lin as a functionof M, and the assumption thatRG,LS ≈ RG then permits an estimate of g for eachM, and a measure of the long-chain branchinggiven a model for g.
◊ On-line small angle static light scattering andrefractive index detectors permit measurement ofMw and solute concentration as functions of theelution time. Off-line measurement of RH,LS ≈ RH = hRH,lin for samples of known elutiontime then permits an estimate of h giveninformation on RH,lin as a function of M, and ameasure of the long-chain branching given amodel for h ≈ g1/2
◊ On-line viscosity and refractive index detectorspermit measurement of [η] and soluteconcentration as functions of the elution time. Theassumption that elution takes place with respectto M[η] (the "universal calibration") and informationon [η]lin as a function of M then permits anestimate of g' for each M, and a measure of thelong-chain branching given a model for g'.
log M
g’ a
nd g
log M5 6 7
100
200
400
60
20
[η]/mL·g-1
RG/nm
Branched & Linear SEC Eluent
5 6 7
1.0
0.8
0.6
0.4
0.2
Randomly Branched PMMAC. Jackson, Y.-J. Chen, J. W. Mays,J. Appl. Polym. Sci. (1996) 59: 179
Example 3: The characterization ofcomposition and MWD.
◊ Cross (or 2D or Orthogonal) method withSEC analysis of each of several fractionsprepared by Temperature Rising ElutionChromatography (TREF)
◊ TREF separation is based on the dependenceof the melting temperature on the polymercomposition
◊ The method has been used, for example, toelucidate the composition and MWD incopolymers of ethylene and propylene:
To Detector
Solvent
Injection
SEC
TREF
Solvent
Effluent vs
Temp.
Elution Volume
SE
C R
esp
on
se
for
each
TR
EF
Fra
ctio
n
TR
EF
Tem
per
atu
re
Analysis of poly(ethylene-co-propylene)T. Usami et al. J. Appl. Polym. Sci.: Appl. Polym. Symp. (1993)
52: 145 (ISPAC-5)
Example 4: The use of mass spectrometry tostudy the MWD of low Mpolymers
◊ UV and Electrospray Ionization-MS Detectionon eluent of a low molecular weightoligomeric solute (doped with a sodium salt toprovide charge)
◊ The ESI-MS method used includes responsesfrom singly, doubly and triply chargedoligomers
◊ An analysis was made on a mixture ofoligomers with the
Det
ecto
r R
esp
on
se
Elution Volume
Single ChargeDouble ChargeTriple Charge
Electrosospray-MS Response
0 10 20
20 22 24
Elution VolumeM
ol. W
t.
2000
1000
4000
polystyrene
W. J. Simonsick Jr. and L. Prokai, in Chromatographic Characterization ofPolymers, T. Provider, H. G. Barth and M. W. Urban, Eds. Adv. Chem. Series
247, (1995), p. 41
A Matrix Assisted Method
X. Fei and K. K. Murray (1996) Anal. Chem. 68:3560
Example 5: Dynamic light scattering on gels
Time averaged scattering photon count auto-correlation function:
Correlation function for an ergodic system with no heterodyne source:
g(2)(τ.q) =⟨nq(t + τ)nq(t)⟩t
⟨nq(t)⟩t =
⟨nq(t + τ)nq(t)⟩E⟨nq(t)⟩E
= 1 + γ|g(1)(τ.q)|2
For a system undergoing a sol-gel transition g(2)(τ,q) is similar to that for therheological function G'(ω):
-4 -3 -2 -1 0
Solution
Gel
75
65
64.2
63.6
55
25
-1
-0.8
-0.6
-0.4
-0.2
0
log τ/s
(1/2
)log
{[g(
2)(τ
,q)
– 1]
/γ}
(Fixed q)
G. Merkle and W. Burchard Macromolecules 29: 3574-8 (1996)
{g(2)(τ.q) – 1}/γ = 2X(1 – X){f(τ.q) – f(∞,q)} + X2{f2(τ.q) – f2(∞,q)}
Heterodyne with static scattering nstatic:
X = nstatic/⟨nq(t)⟩t; f(∞,q) = 0
Non-ergodic medium with ensemble average scattering ⟨nq(t)⟩E:
X = ⟨nq(t)⟩E,t/⟨nq(t)⟩t; 1 ≥ f(∞,q) ≥ 0; f(∞,q) = exp(-q2⟨δ⟩2)
-1
-0.8
-0.6
-0.4
-0.2
0
-4-6 -2 20
log τ/s
(Fixed q)
f(τ,
q)
From g(2)(τ,q) as non-ergodic
From g(2)(τ,q) as ensemble average
From g(2)(τ,q) as particle heterodyne
-1
-0.8
-0.6
-0.4
-0.2
0
-4-6 -2 20
log τ/s
2%3%
4%
5%
10%
(Fixed q)
f(τ,
q)
J. G. H. Joosten, J. L. McCarty and P. N. Pusey Macromolecules 24: 6690-9 (1991)
Example 6: Branched Chain Polymers (Concentrated or undiluted)
η(M, c, T) ≈ ηLOC(T) F(M, c, T)
ηLOC(T) ≈ [ηLOC(T)]LIN; Rare exceptions to this known
F(M, c, T) ≈ 1 + [η](c)c
[η](c) = πNA(â/3) gM
F(M, c, T) ≈ 1 + ~XE(
~X/
~Xc );
~X = [η](c)c
E(~X/
~Xc ) = {1 + B(g, MBR/Mc)(
~X/
~Xc )
4.8}1/2
B(g, MBR/Mc) ≈ 1 unless the branch molecular MBR > Mc
~Xc = πNA(â/3)ρMc ≈ 100 for many polymers
Branched
Linear
Slope 1
Log (ΜΜΜΜw)
0 2-2
Slope 3.4
Slope >3.4L
og
((((ηηηη
//// ηηηηc)
~~
0
-2
2
4
6
Branched
Linear
Slope 1
Log (gΜΜΜΜw)
0 2-2
Slope 3.4
Slope >3.4L
og
((((ηηηη
//// ηηηηc)
~~
0
-2
2
4
6
Approximate Representation of Enhancement Effect
η~ = η/ηLOC
(c) ≈ X̃E(X̃/X̃c)F(σX̃/X̃c)
The function F(σX̃/X̃c) ≥ 1 is intended to account for theenhancement effect, with σ the ratio of the molecular weightof a branch to the total molecular weight, e.g., for a combbranch molecule with n branches, σ = (1 - λ)/n.
An empircial relation suggested for F(σX̃/X̃c):
F(σX̃/X̃c) ≈ exp{b[(σX̃/X̃c) – 1)
(σX̃/X̃c) ≥ 1; b ≈ [2.16 + 0.848λ](1 – g)
(σX̃/X̃c) < 1; b ≈ 0
Reptation models provide a qualitative description for thebehavior, but not as yet a quantitative one.
Some Thoughts on the Future
◊ Some easy predictions⇒ Methods will become increasingly
computer intensive, both in the control ofthe experiment and in the analysis of theresults
⇒ Chromatographic methods will becomeincreasingly complex, with multipledetectors, and with combinations ofchromatographic separations in a "single"analysis
⇒ Methods will become faster, with moreinformation issuing from the experiment;information "overload" will become moreof a problem
⇒ Methods will become increasingly proneto misinterpretation, with the analystbecoming further removed from the "rawdata" of the experiment
⇒ There will be a drive to increase thethroughput of experimental data, perhapsusing methods adapted from the"Genome Project"
◊ Some Critical Needs⇒ Methods to characterize the distribution
of properties within a single SEC elutionfraction, e.g., branch nature, copolymercomposition distribution
⇒ Methods to characterize the sequencedistribution in copolymers; existingmethods are specialized and usually notamenable to SEC on-line use
⇒ Methods (chemical or enzymatic) tosystematically disassemble complexmacromolecules
⇒ Robust instrumentation and methodsapplicable to process control applications
◊ Instrumental Developments⇒ Development of robust, "TURN KEY"
instruments will continue for multipledetector SEC chromatography, includinginstruments suitable for process controlapplications
⇒ The escalating cost of sophisticatedinstrumentation will present a barrier toits "every-day" use; increased relianceon computer-aided methods may helpcontrol this trend
⇒ The implementation of serial multipleseparation methods will expand, but willlikely remain a tool for the researchlaboratory and the specialist to beapplied for critical problems
◊ People issues⇒ Analysts will be required to continually
upgrade their skills in instrumentation,separation methods, and the applicationof computer-aided methods
⇒ The challenges and opportunities ofmaterials characterization must betransmitted to the student population atall levels
Stretching a molecule using a bead manipulatedin a calibrated optical trap
Science 283: 1689 (1999)
Stretching a molecule using a calibratedcantilever beam
Science 283: 1727 (1999)
Stretching a molecule in a flow field
Fluorescent labeled DNA
Science 283: 1724 (1999)
Total Internal Reflectance Microscopy
The vertical movement of the bead providesinformation on the mean potential of the bead
and the surface
Adsorbed polymersElectrostatic interactions
D. C. Prieve, Adv. Colloid Interface Sci. 1999.