State of the art of size- and interaction based …...and LS provides a wealth of information from...
Transcript of State of the art of size- and interaction based …...and LS provides a wealth of information from...
![Page 1: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/1.jpg)
FOR INTERNAL USE ONLY
State of the art of size- and
interaction based separations for industrial polymers
Harry Philipsen, Tiny Frijns
International Symposium on GPC/ SEC and related techniques
Amsterdam, September 27, 2016
![Page 2: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/2.jpg)
Page
FOR INTERNAL USE ONLY
1
Outline: the industrial perspective in analytical polymer separations
Introduction DSM, DSM-Resolve.
Some of our recent progresses and faced challenges in polymer separations:
o How to improve precision and how accurate can
we get in industrial practice?
o Short-chain and long-chain branching in practice.
o Extremely high molar masses in SEC.
o Current status of APC.
o Current relevance of 2D-LC for polymers?
![Page 3: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/3.jpg)
Page
FOR INTERNAL USE ONLY
Chemical synthesis & biotechnology
Energy, chemistry & polymer technology
Biotechnology
Hoffman La Roche’s Vitamins (1930s)
DSM (1902)
Gist-Brocades (1869)
Vitamins Omega’s
Carotenoids Premixes for food & feed
Enzymes Minerals
Cultures & Yeasts Nutraceuticals
Pharmaceuticals Cellulosic bioethanol Biomedical materials
Bio-plastics High Performance Plastics
Coating Resins Functional Materials
Solar – advanced surfaces
Life Sciences
Materials Sciences DSM 1902
DSM: Building on an impressive history
![Page 4: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/4.jpg)
Page
FOR INTERNAL USE ONLY
Annual revenues ca. 20M EUR, +50% third parties.
~ 80 scientists (40% PhD, 60% Bachelor).
ISO 9001-2008 certified, fit for ISO 9001-2015.
IP Security - Quote by well recognized external audit team: ‘Intellectual Property’ security above standard.
100% DSM.
DSM Resolve – Facts & Figures
![Page 5: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/5.jpg)
Page
FOR INTERNAL USE ONLY
DSM Resolve - Unique analytical capabilities Integrate micro- & spectroscopy to probe the distribution & organization of polymers in (Bio) materials.
Molecular Imaging
Materials & Chemical processing
Catalyst characterization
Macromolecular characterization
Small molecules profiling
Create profound understanding in design & upscaling by integrating off-in- or on-line monitoring strategies with chemo-metrics.
Correlate morphology of Catalysts materials to performance, fouling & reaction kinetics.
Relate details in molecular microstructures of poly-disperse (bio)polymers to materials properties.
Employ state-of-the-art hyphenated separation technologies to characterize complex mixtures.
![Page 6: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/6.jpg)
Page
FOR INTERNAL USE ONLY
5
Developments in Polymer Analysis @ DSM
o Corporate program on (Bio)Macro-molecular Analysis: o Chemical heterogeneity. o Very high molar masses. o 3D structures. o Quantification aspects. o Screening of additives. o Enhanced extractions, sample prep.
o Analytical technologies involved: o GPEC, 2DLC (off-line, on-line). o AF4. o DOSY-NMR, HR NMR. o HR-MS, IMS-MS, MALDI.
![Page 7: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/7.jpg)
Page
FOR INTERNAL USE ONLY
6
SEC in industrial applications - 2016
o Gap between academic research and industrial practice on polymer separations.
o SEC: one of the most used techniques for polymer characterization in industry.
o Considered as simple, but still many pitfalls.
o Real life accuracy and precision often cumbersome – how come?
![Page 8: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/8.jpg)
Page
FOR INTERNAL USE ONLY
7
SEC; accuracy and/ or precision?
o Often: precision more important than accuracy: o Trends. o Quality control. o Polymer users.
o Accuracy: o Polymer manufacturers: modelling,
understanding. o Polymer users do not want to be to
far “off”. o Improved accuracy plus information
density by DV and LS detection – but at cost of precision.
![Page 9: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/9.jpg)
Page
FOR INTERNAL USE ONLY
2. Column stability
o Mechanical and chemical. o Polymer versus silica.
o Eluent type, e.g. THF worse than HFIP.
4. Sample and standards
o Standards batch to batch variations or different manufacturers.
o Sample: molar mass level, concentration (depends on MM),
dn/dc.
8
Precision in modern SEC
1. Flow stability
o For conventional and universal calibration. Use flow marker. o Not (much less) for 3SEC.
3. Detector stability
o RI is known ‘unstable’, especially in e.g. HFIP and high temperatures.
Precision is worst for universal calibration Better for conventional calibration (only one detector, no
accurate weighing) Also better for Light Scattering/ 3SEC (no calibration).
![Page 10: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/10.jpg)
Page
FOR INTERNAL USE ONLY
1. 3SEC in THF (PS, Mw = 200 kDa)
Within one run (n = 10, RSD, 1σ) Mn: 5% - Mw: 2% - Mz: 3% - [η]: 1%.
2. 3SEC in HFIP (PA, Mw 60 kDa)
Short term (5 weeks, n = 26, RSD, 1σ) Mn: 3% - Mw: 3% - Mz: 3%
Long term (2 years, RSD, 1σ)
Mn: 10% - Mw: 10% - Mz: 10%
3. HT-SEC – DV (NBS1475A, Mw = 56 kDa)
Short term (1 run, n = 10, 2σ) Mn: 10% - Mw: 8% - Mz: 11% - [η]: 4%.
Long term (2 years, n = 220, 2σ)
Mn: 15% - Mw: 10% - Mz: 13% - [η]: 6%
4. HT-SEC – LS (NBS1475A, Mw = 56 kDa)
Long term (2 years, n = 220, 2σ) Mn: 12% - Mw: 2% - Mz: 10%
9
Some precision numbers from our lab
![Page 11: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/11.jpg)
Page
FOR INTERNAL USE ONLY
10
Base line issues affecting precision
Improvement of baseline stability on transferring high temperature SEC (solvent: TCB, T: 160 C) from RI to IR detection (broad band filter)
![Page 12: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/12.jpg)
Page
FOR INTERNAL USE ONLY
11
SPC sample (NBS1475A, linear PE) PolymerChar Infrared detection
Viscotec Refractive Index detection
Mn Mw Mz Mw/Mn Mz/Mw IVw
Average 15.3 55.2 166.4 3.61 3.01 1.03
SD 0.82 1.91 7.92 0.13 0.14 0.03
RSD (%) 5% 3% 5% 4% 5% 3%
Mn Mw Mz Mw/Mn Mz/Mw IVw
Average 14.0 56.3 182.3 4.03 3.24 1.02
SD 0.51 1.80 6.81 0.12 0.07 0.02
RSD (%) 4% 3% 4% 3% 2% 2%
Improved precision via more stable detection IR versus RI detection in HT-SEC
o Improvement of precision.
o Some systematic differences in Mn and Mz – which can (and must!) be explained.
![Page 13: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/13.jpg)
Page
FOR INTERNAL USE ONLY
12
Correlation study IR versus RI detection for PE
Mn Mw
Mz IVw
![Page 14: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/14.jpg)
Page
FOR INTERNAL USE ONLY
13
Example: multi detector approach with less emphasis on accuracy
o Polymer users perspective:
o Precision more important than accuracy. o Characterization and comparison of ‘unknown’ products and formulations.
o Multi detector approach: SEC coupled to PDA-UV/VIS, RI, DV
and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information on branching/ conformational differences (DV). o Rough impression of chemical composition (PDA-UV). o Presence of gels (DV) and particles (LS). o Information on blending (PDA-UV, RI).
![Page 15: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/15.jpg)
Page
FOR INTERNAL USE ONLY
Comparison of UV and RI: no blend of chemically different compounds.
UV-DAD: o No significant impurities, I.e. no
blend of chemically different compounds.
o Except one peak eluting before total exclusion area.
Example: multi detector approach..
![Page 16: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/16.jpg)
Page
FOR INTERNAL USE ONLY
5. Study of competitors products by multi detector SEC
From UV-DAD: o Classification polymer type:
polyester (1, 2). o Detection and classification of
dye (3). o Indication of gel particles (4).
Example: multi detector approach..
![Page 17: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/17.jpg)
Page
FOR INTERNAL USE ONLY
5. Study of competitors products by multi detector SEC
3SEC (RI-DV-LS): o Molar mass averages. o Information on branching
indicates highly branched structures.
o LS indicates gel particles.
Example: multi detector approach..
![Page 18: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/18.jpg)
Page
FOR INTERNAL USE ONLY
17
o Real accuracy way more difficult to achieve than ‘good’ precision.
o Most important error sources:
o Concentration: o Moisture, solvents, additives. o Exceeding overlap concentration.
o dn/dc (copolymers, polyelectrolytes). o Enthalpic effects (non-ideal SEC). o Pore mismatch. o Shear and/ or slalom effects. o Calibration and alignment of the detectors
(‘peak parameters’). o Data-analysis: integration, side products. o Scattering angle (LALS, RALS, MALS).
Accuracy in modern SEC
![Page 19: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/19.jpg)
Page
FOR INTERNAL USE ONLY
18
Revisiting calibration of SEC in HFIP
o Our case: o slightly wrong dn/dc ratio
of PMMA to Nylon. o Mw value of PMMA standard
appeared not correct.
o New broad PMMA standards (NIST), correct dn/dc (ratio): correct values for Nylons calculated via universal calibration.
o Work still in progress..
Coinciding universal calibration curves for (broad) Nylons and (narrow) PMMA’s
![Page 20: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/20.jpg)
Page
FOR INTERNAL USE ONLY
19
Accuracy: pore mismatch Obvious example: (wrong) combination of 104 and 100 Å column
More resolution in the low molar mass region leads to distortion of peak shape, also effecting accurate MMD.
![Page 21: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/21.jpg)
Page
FOR INTERNAL USE ONLY
20
Accuracy: pore mismatch Less obvious example: linear PFG columns (1000 + 100 Å)
Non-perfect linearity, only observed from accurate ‘universal’ calibration curves deduced from 3SEC or DV/RI
with many points.
Pore mismatch: still often overlooked
but affects accuracy! Use good quality, linear
columns!
![Page 22: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/22.jpg)
Page
FOR INTERNAL USE ONLY
o Concentration, signal to noise.
o Concentration signal is a function of molar mass up to M ≈ 2000.
o Dealing with different detectors.
o Which components to include?
> Influences Mn due to invisibility of lowest molar mass parts.
> Effect on accuracy Mn.
> Comparing Mn from universal calibration and Goldwasser method (only using viscosity signal and total sample amount). Experience: use DV signal for determining integration end.
> Depending on the polymerization model. E.g.: for polycondensations (low molar mass) cyclics not included in (most) models!
21
Accuracy: end of integration, Mn issue
![Page 23: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/23.jpg)
Page
FOR INTERNAL USE ONLY
22
Accuracy: practice compared to theory – Flory plots
0
0.2
0.4
0.6
0.8
1
1.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
2 2.5 3 3.5 4 4.5 5 5.5 6
resi
dual
Wf/
d(lo
gM)
log M
-0.04
-0.03
-0.02
-0.01
0
0.01
0.02
0.03
0
0.2
0.4
0.6
0.8
1
1.2
1.4
2 2.5 3 3.5 4 4.5 5 5.5 6
resi
dual
Wf/
d(lo
gM)
log M
-0.05
-0.04
-0.03
-0.02
-0.01
0
0.01
0.02
0.03
0
0.2
0.4
0.6
0.8
1
1.2
1.4
2 2.5 3 3.5 4 4.5 5 5.5 6
resi
dual
Wf/
d(lo
gM)
log M
-0.04
-0.03
-0.02
-0.01
0
0.01
0.02
0.03
0
0.2
0.4
0.6
0.8
1
1.2
1.4
2 2.5 3 3.5 4 4.5 5 5.5 6
resi
dual
Wf/
d(lo
gM)
log M Red: calculated MMD. Blue: Flory fit for a linear polyamide.
Purple: residuals, end of integration determined by DV = 0 Blue: residuals, end of integration determined by RI = 0 (PLUS correcting for mass (peak area) taken into account).
Green: residuals, end of integration determined by RI = 0 (NOT correcting for mass (peak area) taken into account).
High quality SEC: excellent agreement (Flory) theory
and SEC practice!
![Page 24: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/24.jpg)
Page
FOR INTERNAL USE ONLY
23
Quantitative branching information still challenging.
o Spectroscopy provides averages and
often lacks sensitivity or resolution.
o Rheology relies on models.
o IMS-MS still in its infancy.
o SEC relatively strong for both LCB and SCB, quantification needs attention.
SEC for branching information
![Page 25: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/25.jpg)
Page
FOR INTERNAL USE ONLY
24
HT-SEC coupled to IR detection for SCB quantification in polyolefins o Quantification of SCB: via ratio of the CH3 and CH2 band signals. o Calibration with standards with known amounts of CH3/1000 C. o CH3/CH2 ratio plotted as function of CH3/1000 C. o Correction for CH3 end groups.
Recent progress on SCB quantification
IR5 detector (PolymerChar)
![Page 26: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/26.jpg)
Page
FOR INTERNAL USE ONLY
25
o For non-C3 branches up to 60 CH3/1000C one (linear) calibration line applicable, (+/-2.7 CH3/1000C), irrespective of co-monomer type.
o Calibration lines up to 100 wt% co-monomer (C3 and C4) show quadratic relation. (C3; +/- 9 CH3/1000C (valid 50 – 335 CH3/1000C).
SCB quantification via SEC-IR
Calibration curves for C3 and C4-C6
Results from: Frijns-Bruls, T., Ortin, A., Weusten, J., Geladé, E., Macromolecular Symposia, 356, 87 – 94, 2015
![Page 27: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/27.jpg)
Page
FOR INTERNAL USE ONLY
26
o Calibration from copolymers or
from blends leads to identical results. Ease of calibration!
o Bandwidth of the detector filters critically determine shape of calibration curve.
o Simulated calibrations excellently correspond to practice. Detector now very well understood!
SCB quantification via SEC-IR
Despite very good analytical performance:
We need a better LOD! (down to 0,1 CH3/1000C)
![Page 28: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/28.jpg)
Page
FOR INTERNAL USE ONLY
27
Quantification of LCB
o LCB can be detected relatively sensitive via SEC-DV.
o Accurate quantification of LCB not trivial.
o Zimm-Stockmayer theory related to Rg rather than Rh and holds for theta conditions.
o Correlation between Rg and Rh: draining factor ε, not exactly known.
![Page 29: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/29.jpg)
Page
FOR INTERNAL USE ONLY
28
o “Late co-elution”. Effect increases with decreasing dp and
increasing flow rate. o Effect can be somewhat suppressed but not completely
prevented.
SEC for very high molar mass polymers
----1ml/min ----0.5ml/min ----0.2 ml/min
RI and LS chromatograms of LDPE tubular on PL Mixed A, (dp 20 μm)
![Page 30: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/30.jpg)
Page
FOR INTERNAL USE ONLY
29
High molar mass molecules need to reorganize and slalom
through the beads. Highest M will elute last !!!
Random coil transition
stretch
Slalom effects, coil-stretch transition
![Page 31: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/31.jpg)
Page
FOR INTERNAL USE ONLY
30
Slalom chromatography
Chromatograms of PS standards in slalom mode. Acquity UPLC BEH C18 Column 100 × 2.1 mm I.D; 1.7-μm particles. Flow rate 0.7 mL/min; mobile phase THF; ambient temperature.
Data from E.V. Uliyanchenko, PhD thesis, Chapter 4.
![Page 32: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/32.jpg)
Page
FOR INTERNAL USE ONLY
31
Alternative: AF4?
Molecular characterization of UHMW polymers still
needs further development.
![Page 33: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/33.jpg)
Page
FOR INTERNAL USE ONLY
32
Yes:
o Much faster results.
o Higher resolution, especially in low molar mass range.
o Less waste.
o Flexibility in solvent composition (rigidity of column packing).
But also a few catches..
Fast SEC – APC
MMD of PS 19,9 Da – by SEC and APC (conventional calibration).
![Page 34: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/34.jpg)
Page
FOR INTERNAL USE ONLY
33
o Increasing elution volume on increasing flow rate, for higher masses (from 105 kDa).
o Slalom effects?
o P affecting coil density?
APC: flow rate affects elution volume
Effect of flow rate (0,2 – mL/min) on elution volume. Solvent: NMP.
Increasing flow rate
![Page 35: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/35.jpg)
Page
FOR INTERNAL USE ONLY
34
APC: varying molar mass compared to SEC
Mp values for 8 saturated polyester samples in NMP (PS calibration).
0
10000
20000
30000
40000
50000
60000
A B C D E F G H I J K L M N O P
Mp (APC) Mp (SEC)
0
5000
10000
15000
20000
25000
30000
A B C D E F G H
Mp (APC) Mp (SEC)
Mp values for 20 non-functional acrylate samples in NMP (PS
calibration).
![Page 36: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/36.jpg)
Page
FOR INTERNAL USE ONLY
35
o Sample dependent results indicate non-SEC effects. No other column packing types (yet) available to help preventing this. Solvent triangle!
o Changing molar masses values requires effort for implementation in e.g. QA/ QC.
o Observation: column clogging for various semi-crystalline polyesters. Not observed in SEC.
o LS detector available nowadays but DV not (yet).
o Use of high flow rates needs further attention!
APC: flow rate affects elution volume
APC is interesting for its speed and flexibility but needs further attention!
![Page 37: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/37.jpg)
Page
FOR INTERNAL USE ONLY
36
o Separation of various polyamide types by Gradient Polymer Elution Chromatography (GPEC).
o Separation based on chemical composition plus molar mass.
Increasing retention (x-axis) from increasingly non-polar polyamides.
Chemical composition separation by GPEC
![Page 38: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/38.jpg)
Page
FOR INTERNAL USE ONLY
37
Chemical composition separation by GPEC
Following blending of polymers by GPEC: changing CCD.
o Next step: 2D-LC, e.g. SEC-GPEC OR GPEC-SEC:
o More details, (small) differences.
o Quantification of CCD?!
![Page 39: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/39.jpg)
Page
FOR INTERNAL USE ONLY
On-line Advantages • No laborious sample (fraction) prep. • (Coupled to the above): suited for
large series and standardized applications.
• ‘Sexy’..
Disadvantages • Interfacing still cumbersome due to
e.g. solvent mismatch issues (strength, viscosity, miscibility).
• Time-consuming setting-up per application.
• No free choice of order of dimensions.
• Second dimension resolution limited. (SEC..).
Off-line Advantages • No ‘complex’ interfacing. • For incidental applications and
freedom to choose separation order. • Maximum separation power in both
dimensions. • Fractions also available for e.g. NMR
and titrations.
Disadvantages • Relatively laborious (sample prep). • Critical to pollution (eluent!). • Sometimes unexpected effects
(adsorption at tubing etc.).
38
2D-LC for polymers: on-line or off-line?
![Page 40: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/40.jpg)
Page
FOR INTERNAL USE ONLY
39
2D-LC for polymers: in practice
o Despite developments since ~15 years: use of on-line 2DLC for polymers seems rare.
o Quantification in 2DLC seems completely neglected in research. E.g. accuracy, precision of 2nd dimension SEC?
![Page 41: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/41.jpg)
Page
FOR INTERNAL USE ONLY
40
Our practice: working on both on-line and off-line coupling
o On-line 2DLC o Striving for a few ‘standardized’
systems. o Collaboration with Universities to
speed-up implementation. Projects on e.g. fast method optimization.
o Off-line 2DLC o Streamlining fractionations. o Developing sample prep. Strategies
for e.g. HFIP-KTFA systems.
![Page 42: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/42.jpg)
Page
FOR INTERNAL USE ONLY
41
Example on-line 2D chromatography
On-line 2D-LC separation (RPLCxHILIC) of oligomers with
different end groups. Structured Chromatograms.
First results LCxSEC of polymer blending. Little SEC resolution.
![Page 43: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/43.jpg)
Page
FOR INTERNAL USE ONLY
42
o Issues on setting-up SEC
fractionation from HFIP/ KTFA:
o Adsorption to Teflon tubing.
o Preventing contaminations.
o Removal of salt by SPE: o Adsorption of KTFA to C18:
optimization of desorption conditions.
o Drying and re-dissolution procedure on the C18 cartridge.
o Now ready for further work. Already
new details visible on polymers.
Example off-line 2D chromatography DSM - the Life Sciences and Materials Sciences Company
Low polydispersity fractions from SEC.
First results SEC-GPEC for a polyamide.
![Page 44: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/44.jpg)
Page
FOR INTERNAL USE ONLY
43
Conclusions
o Value of modern SEC can hardly be over estimated but..
o Establishing accurate and precise systems is still not trivial (for many users).
o Need for better, more precise and sensitive branching characterization.
o Need for better methods for very high molar masses especially at high temperatures.
o Fast SEC (APC): need for other column packings; beware of too high speeds.
o 2D LC for polymers is still in its infancy: quantification!
![Page 45: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/45.jpg)
Page
FOR INTERNAL USE ONLY
44
DSM - the Life Sciences and Materials Sciences Company
Acknowledgements
Tiny Frijns
Sandra van der Flier
Artur Henke
Sander Kok
Joep Vanderfeesten
![Page 46: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/46.jpg)
Page
FOR INTERNAL USE ONLY
45
Thank You!
DSM - the Life Sciences and Materials Sciences Company
![Page 47: State of the art of size- and interaction based …...and LS provides a wealth of information from one single analysis: o ‘Rough’ molar mass distributions (DV/ LS). o Information](https://reader030.fdocuments.us/reader030/viewer/2022011903/5f0f97dd7e708231d444eea3/html5/thumbnails/47.jpg)
46