Poly(vinyl alcohol) hydrogels in medicine Structure and mechanics
Gavin Braithwaite
Cambridge Polymer Group, 56 Roland Street, Suite 310 Boston, MA 02129
7-17 Presentation (10/1/2010)
Poly (vinyl alcohol) • • •
First synthesized by Hermann and Haehnel in 1924 First commercialized in US by Dupont in 1939 Not polymerized directly – Vinyl acetate formed through catalyzed addition of acetic acid to acetylene – Radical polymerization to PVAc – Converted to PVA by hydrolyzation in alcohol in presence of catalyst
•
Conversion rates influence solubility and application
2015 SCI Rideal Meeting
2
Cambridge Polymer Group
PVA uses • • • • • • • • •
Market (2006) > 1 million metric tons Precursors - Reacted to create polyvinyl acetals Solutions - Emulsion polymerization aid and rheology modifier Fiber production - Clothing, Concrete reinforcement Paper products - Adhesives, Sizing (preparation) Barriers - Water soluble and CO2 barrier films Medicine – Sponges, Eye drops, Embolization, Nerve guides, cartilage Children’s toys - slime Sport - fishing
2015 SCI Rideal Meeting
3
Cambridge Polymer Group
PVA structure •
Solubility and mechanical properties variable – Hydrolysis rates 87-99+% – Naturally atactic but syndiotactic can be obtained – Broad range of molecular weights
•
Hydroxyl groups – – – –
•
Crystal structure similar to PE Water soluble Does not melt (decomposes) Hydrogen bonds
Biocompatible – – – –
Cells grow happily No known toxicity Not naturally degraded Generally not persistent
2015 SCI Rideal Meeting
4
Cambridge Polymer Group Billmeyer (1984)
Crosslinking of PVA •
Chemical crosslinking – Formaldehyde, Glutaraldehyde etc – Radiation
•
Physical crosslinking – – – – –
Hydrogen bonding between hydroxyls Mediated by water molecules Borate ion (“slime”) Formation of crystallites in water Solutions not long-term stable
2015 SCI Rideal Meeting
5
Cambridge Polymer Group
Physical Association •
Freeze-thaw process – “Cryogels” first proposed in 1971 (Peppas) – Repeated freeze-thaw cycles • Properties influenced by – Cycles, concentration, molecular weight, hydrolysis, time of freezing
– Freezing process • Liquid-liquid phase separation • Polymer-poor regions freeze first
2015 SCI Rideal Meeting
6
Cambridge Polymer Group Holloway et al. 2013
Hydrogen bonded crystals •
Hydrogen bonds form crystals – Onset ~55 °C, Peak ~70 °C, H 1-5 J/g – (solid PVA 161 J/g, Tm ~230 °C)
2015 SCI Rideal Meeting
7
Cambridge Polymer Group Holloway et al. 2013
1 FT
PVA cryogel structure •
Croygel structure driven by formation of ice crystals
200 kDa 99.9+% PVA in DI water 8 hour freeze, 8 hour thaw
– Fine pores form during melting of polymer rich regions – Structure “sharpens” with increasing freeze-thaw – Phase separation process – Hydrogen bonded “crosslinks”
Scale bar: 10 m
2 FT
5 FT
5% PVA
10% PVA
15% PVA
25% PVA
Confocal microscopy. All images and gels MGH.
2015 SCI Rideal Meeting
8
Cambridge Polymer Group Confocal Images: Jeeyoung Choi (MGH)
Gelation of PVA •
Coarseness of gel depends on – – – – –
•
Molecular weight Number of cycles Concentration Age Hydrolysis
Crystal junctions – Not a conventional gel – Polymer-rich “bridges”
2015 SCI Rideal Meeting
9
Cambridge Polymer Group Bercea et al. 2013
Solvent-driven gelation •• • •
• •
An entangled polymer solution a good the When the temperature is indropped solvent is stable and homogeneous hydrogen-bonded crystals can form in If the solvent quality is dropped the polymer-rich regions The solution enters an unstable condition A physical where there iscrosslinked coexistence ofhydrogel a polymer is rich and polymer poor region formed The polymer can phase separate into two concentration regions as the solvent quality changes
20 m 2015 SCI Rideal Meeting
10
Cambridge Polymer Group Image: ESEM CPG
Simulation unknown
Theta point • • •
Freeze-thaw PVA hydrogel - Swell Ratio at equilibrium in mixed solvent (measurements using CPG SRT) 0% PEG “good” solvent, 15% PEG “poor” solvent, 28% PEG “bad” solvent Solvency approximately inversely related to temperature for these systems
Theta solvent swelling line
2015 SCI Rideal Meeting
11
Cambridge Polymer Group
Thetagel structure •
Names defined as (wt% PVA-wt% PEG 400) relative to water – 200 kDa 99.9+% hydrolyzed PVA
•
Distinctive pore structure – Order 10 m diameter – Size and morphology depends on PVA concentration 10-28 DP
2015 SCI Rideal Meeting
15-28 DP
25-28 DP
12
Cambridge Polymer Group Scale bar 20 m
PVA in water •
15% PVA in DI water is essentially stable up to 93 °C 1000
100 90
80 C
80 70
100
60 50
40 C
Temp [C]
G', G" [Pa]
80 C
40
10
30
93 C
G' Cool only G'' Cool only G' Heat and cool G'' Heat and cool
89 C
20 10
1
0 0
2015 SCI Rideal Meeting
50
100 150 Time after start [min] 13
200
250
Cambridge Polymer Group Theta temperature for PVA in water is 97 °C (Polymer Handbook, Brandrup)
Effect of solvent quality • • •
15-x: 15% PVA in x% PEG in water 15% PEG behaves qualitatively the same as DI at all temperatures 28% PEG is unstable below approximately 50 °C 15-0 G* 15-0 delta
1000
15-15 G* 15-15 delta
15-28 G* 15-28 delta
100 80
G* [Pa]
70 60 100
50 40 30 20 40 ºC
Cool to 40 ºC
80 ºC
Phase angle (d) [deg]
90
10
10
0 0
20
2015 SCI Rideal Meeting
40
60 80 Time after start [min] 14
100
120
140
Cambridge Polymer Group
Importance of temperature 15-28 (15% PVA in 28% PEG400/water) Temperature below 55 °C drives gelation
100000
100 Cooling starts
40C
60C
20C
90 80
10000
70 60 50
G* [Pa]
G* 15-28 hold at 60C G* 15-28 cool to 40C G* 15-28 hold at 20C delta 15-28 hold at 60C delta 15-28 cool to 40C delta 15-28 hold at 20C
1000
40 30
100
20 10
Phase angle (delta) [deg]
• •
0
10 0
10 20 30 40 50 60 70 80 90 100 110 120 130 140 Time after start [min]
2015 SCI Rideal Meeting
15
Cambridge Polymer Group
PVA hydrogel creep properties • • •
“Diurnal cycle” 0.5 MPa on, 0.05 MPa off Higher molecular weight implies stiffer High creep strain and poor creep recovery 1 0.9
Strain [mm/mm]
0.8 0.7 0.6 0.5 0.4 0.3
20-28 250PVA
20-28 100PVA
0.2
15-28 250PVA
15-28 100PVA
0.1
25-28 250PVA
25-28 100PVA
0 0
2
2015 SCI Rideal Meeting
4
6
8 Time [hr] 16
10
12
14
16
Cambridge Polymer Group
Elastic Modulus 20%-65%* Loading Curve E 3
n=5 Elastic Modulus [MPa]
2.5
2
1.5
1
0.5
0
DP
DP
DP
10-28
15-28
25-28
2015 SCI Rideal Meeting
AG RSA3 AG RSA3 AG RSA3- AG RSA3PEG PEG 15-28
25-28
17
15-28
25-28
1FT
5FT
5FT
15-28
15-28
15%
Cambridge Polymer Group
Cartilage ~ 10MPa
EWC vs. max creep strain 1.000
n=3 in all cases. Scale bar 20 m
Maximum creep strain after 8 hrs at 0.5 MPa [mm/mm]
0.900
0.800
0.700
0.600
0.500 10-28 DP 15-28 DP 25-28 DP 15-28 1FT 15-28 5FT 15% 5FT 15-28 AG RSA3 15-28 AG RSA3-PEG 25-28 AG RSA3 25-28 AG RSA3-PEG
0.400
0.300
0.200
0.100
0.000 0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
EWC [mass water/mass hydrogel]
2015 SCI Rideal Meeting
18
Cambridge Polymer Group
Injecting PVA •
Phase separation not instantaneous – “window” of useful time for injections or manipulation
•
Allows injection of PVA hydrogel without toxic crosslinkers 1000
90 70
G* [Pa]
60 50
100
40
45 degree phase angle
30 G* 15-28 10C/min G* 15-28 30C/min delta 15-28 10C/min delta 15-28 30C/min
10 -50
-30
2015 SCI Rideal Meeting
-10 10 30 Time after reached 40C [min] 19
20 10
Phase angle (delta) [deg]
80
0
50 Cambridge Polymer Group
Mitral Regurgitation •
Common complication of coronary artery disease – Doubles risk of late death
• •
Heart attack results in compromised muscle Over time changes geometry of chamber – Wall thins and distorts
• • •
Deforms mitral valve tethers Results in reversed flow through valve Chronic problem, usually fatal
•
Current solutions inadequate – Ring annuloplasty • Invasive and complex
2015 SCI Rideal Meeting
Cambridge Polymer Group Work performed in collaboration with MGH under NIH 1R01HL092101-01A1
Tissue bulking •
Mitigation of Mitral Regurgitation (MR) – Inject bulking agent to displace muscle wall
Echo Transducer Apex
Biomaterial
Ischemic LV
Biomaterial PM
AO
Coapting Surface
AO
MR LA Leaflet Tenting
2015 SCI Rideal Meeting
Restored Leaflet Coaptation
21
Cambridge Polymer Group
MR reduction in ovine model
Figure 2
2015 SCI Rideal Meeting
22
Cambridge Polymer Group
Load-bearing applications •
Applications – – – –
•
Spine Mosaicplasty Interpositional Devices Resurfacing
Requirements – High fatigue resistance – Good recovery – Loads over 3 kN intermittently
•
Thetagels (and cryogels) do not have sufficient properties – Cartilage ~ 10MPa compressive modulus – Chemical crosslinking – “Toughening” by annealing
2015 SCI Rideal Meeting
23
Cambridge Polymer Group
Toughening of thetagels – “annealing” •
Annealing of PVA gels (dehydration) changes morphology 15-28
25-28
2015 SCI Rideal Meeting
15-28 annealed
25-28 annealed
24
•
For 15-28, annealing procedure results in thicker struts and a more closed pore structure.
•
For 25-28, pores appear to shrink or collapse
Cambridge Polymer Group
Elastic Modulus 20%-65%* Loading Curve E 3
n=5 Elastic Modulus [MPa]
2.5
2
1.5
1
0.5
0
DP
DP
DP
10-28
15-28
25-28
2015 SCI Rideal Meeting
AG RSA3 AG RSA3 AG RSA3- AG RSA3PEG PEG 15-28
25-28
25
15-28
25-28
1FT
5FT
5FT
15-28
15-28
15%
Cambridge Polymer Group
Strain Recovery as a % of strain at 10 N 100 90
n=5
Recovered Strain [%]
80 70 60 50 40 30 20 10 0
DP
DP
DP
10-28
15-28
25-28
2015 SCI Rideal Meeting
AG RSA3 AG RSA3 AG RSA3- AG RSA3PEG PEG 15-28
25-28
26
15-28
25-28
1FT
5FT
5FT
15-28
15-28
15%
Cambridge Polymer Group
Total creep strain after 8 hr at 0.5 Mpa [%]
Total Creep Strain 100 n=5
90 80 70 60 50 40 30 20 10 0 DP
DP
DP
10-28
15-28
25-28
2015 SCI Rideal Meeting
AG RSA3 AG RSA3
15-28
25-28
27
AG RSA3PEG
AG RSA3PEG
1FT
5FT
5FT
15-28
25-28
15-28
15-28
15%
Cambridge Polymer Group
EWC vs. max creep strain 1.000
n=3 in all cases Maximum creep strain after 8 hrs at 0.5 MPa [mm/mm]
0.900
Scale bar 20 m
0.800
0.700
0.600
0.500 10-28 DP 15-28 DP 25-28 DP 15-28 1FT 15-28 5FT 15% 5FT 15-28 AG RSA3 15-28 AG RSA3-PEG 25-28 AG RSA3 25-28 AG RSA3-PEG
0.400
0.300
0.200
0.100
0.000 0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
EWC [mass water/mass hydrogel]
2015 SCI Rideal Meeting
28
Cambridge Polymer Group
2015 SCI Rideal Meeting
29
Cambridge Polymer Group Plot: MGH OBBL
CoF on a rheometer Spring
Annulus
Surface
1
Kavehpour and McKinley “Tribo-rheometry: from gap-dependent rheology to tribology” Tribology Letters (2004) 17(2) 327-335
2015 SCI Rideal Meeting
30
Cambridge Polymer Group
Hydrogel coefficient of friction •
CoCr against hydrogel flats 3.0 15-28 DP, CPG 15-28 AG RSA3, CPG 15-28 DP, MGH 15-28 AG RSA3, MGH
2.5
COF
2.0 1.5 1.0 0.5 0.0 0.0 2015 SCI Rideal Meeting
10.0 20.0 30.0 40.0 Nominal contact pressure [kPa] 31
50.0
Cambridge Polymer Group
Where next? •
PVA appealing – Readily available – Biocompatible – Already in use
•
PVA thetagels – injectable
•
Limited application – – – –
Poor mechanicals Tissue bulking? Degradables? Mucoadhesives?
2015 SCI Rideal Meeting
32
Cambridge Polymer Group
Functional proteins • •
Mucins - Terminal groups cysteine-rich and naturally gel-forming Lubricin (PRG4) - present in synovial fluid as lubricant
2015 SCI Rideal Meeting
33
Cambridge Polymer Group
Thiolation of PVA •
Conversion of some OH groups to thiol groups adds thiol pendant groups directly to the PVA backbone – Mercaptopropionic acid reacted with PVA – Thiol groups react with cysteine residues in proteins to form disulfide bonds
2015 SCI Rideal Meeting
34
Cambridge Polymer Group
1H NMR of TPVA •
1H NMR of converted product indicate presence of mercaptopropionic ester fragment – Degree of modification ~3%
2015 SCI Rideal Meeting
35
Cambridge Polymer Group
Molecular Weight Distribution of PVA and TPVA •
Gel Permeation Chromatography indicates a small fraction of higher molecular weight species 9
Wn (Log M) * 10-1
8
PVA
7 6 5 4 3 2
TPVA
1 0 2
3
4
5
6
7
Log M 2015 SCI Rideal Meeting
36
Cambridge Polymer Group
T-PVA/Mucin reaction •
Mixing of TPVA with mucin and tracking rheology response proves molecular interactions
|n*| (Pa.s)
– Complex viscosity of TPVA (green), mucin (blue) and TPVA combined 1.000 with mucin (red) measured at 25 °C.
0.1000
0.01000 0
10.0
2015 SCI Rideal Meeting
20.0
30.0 time (min) 37
40.0
50.0
60.0
Cambridge Polymer Group
Synthesis of TPVA/PEGDA Hydrogels •
TPVA crosslinking with difunctional poly (ethylene glycol) – – – –
thiol-reaction forms hydrogel through Michael-Type addition reaction Thiol groups control crosslink density PEGDA chain length control molecular weight between crosslinks Physiologically benign reaction
TPVA
b
+ pH 7.4 1xPBS PEGDA
2015 SCI Rideal Meeting
38
Cambridge Polymer Group
Gelation kinetics 10000
1000
1000
100.0
100.0
10.00
10.00
1.000
1.000
0.1000
0.1000
0.01000
G'' (Pa)
G' (Pa)
10000
0.01000
1.000E-3 0
2.5
5.0
7.5 time global (min)
Polymer
1.000E-3 15.0
12.5
Temperature, °C
concentration, % [w/v]
10.0
25 °C
37 °C
Gelation
G’
G’’
Gelation
G’
G’’
time, [min]
[Pa]
[Pa]
time, [min]
[Pa]
[Pa]
3.0
23.3
803
5
4.2
3607
480
4.5
9.2
6440
133
3.0
9860
280
2015 SCI Rideal Meeting
39
Cambridge Polymer Group
Degradability •
Esters are relatively unstable bonds – hydrolyzable 60
Gel disintegration onset
40
b
Swelling, %
50
30 20 10 0 0
5
10 Time, days
2015 SCI Rideal Meeting
15
40
Cambridge Polymer Group
TPVA Degradation Products by GPC Analysis •
Cleaving of ester bonds yields species with TPVA and PEGDA molecular weights
Signal MV * 10
15 14 TPVA
13 12
TPVA/PEGDA Degradation products
11 PEGDA
10 3
2015 SCI Rideal Meeting
5
7 9 11 Elution Volume (mL) 41
13
15
Cambridge Polymer Group
Thank you Cambridge Polymer Group is a contract research laboratory specializing in polymers and their applications. We provide outsourced research and development, consultation and failure analysis as well as routine analytical testing and custom test and instrumentation design. Cambridge Polymer Group, Inc. 56 Roland St., Suite 310 Boston, MA 02129 (617) 629-4400 http://www.campoly.com
[email protected]
2015 SCI Rideal Meeting
42
Cambridge Polymer Group