Weathering Reactions on Phosphate Glass...
1
Weathering Reactions on Phosphate Glass Surfaces Weathering Reactions on Phosphate Glass Surfaces B. C. Tischendorf and R. K. Brow B. C. Tischendorf and R. K. Brow B. C. Tischendorf and R. K. Brow University of Missouri University of Missouri University of Missouri-Rolla Rolla Rolla Rolla MO 65401 Rolla MO 65401 Rolla MO 65401 Introduction Introduction Discussion Discussion Acknowledgements Acknowledgements Future Work Future Work ppm -30 -25 -20 -15 -10 -5 0 5 10 intensity (arbitrary units) WKMAP WKBAP WKAP Q2 Q1 Q0 Dissolution Activation Energies Dissolution Activation Energies Weathering Data Weathering Data Corrosion Model Corrosion Model Weathering Surface Species Weathering Surface Species Experimental Procedure Experimental Procedure This work is funded by Lawrence Livermore National Laboratory. Work was performed under the auspices of the US Department of Energy by of Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48. Column Retention Time (min) 0 5 10 15 20 25 30 35 40 45 50 Solute Conductivity (uS) KAP Weathered 0 Days KAP Weathered 14 Days P 1 P2 P1 P2 P3 P 3m P 3 P4m P4 P5m P5P6 P7P8 ... KBAP 10 Days Corrosion 80%RH at 50ºC 1/T (*1000 ºK) 0.0027 0.0028 0.0029 0.0030 0.0031 0.0032 0.0033 0.0034 23.0 (+/- 2.8) KAP 19.2 (+/- 2.9) MAP 18..4 (+/- 1.3) BAP 11.2 (+/-0.73) KBAP 9.9 (+/- 2.3) KMAP Ea (kcal/mole) Glass 23.0 (+/- 2.8) KAP 19.2 (+/- 2.9) MAP 18..4 (+/- 1.3) BAP 11.2 (+/-0.73) KBAP 9.9 (+/- 2.3) KMAP Ea (kcal/mole) Glass Bulk Dissolution Data Bulk Dissolution Data Weathered Layer Structure Weathered Layer Structure Chain Distribution of Weathering Layer Chain Distribution of Weathering Layer ?Microscope images of the glass surface before and after weathering show accelerated corrosion at surface defects. 15% K 2 O 15% MgO KMAP 15% K 2 O 15% BaO KBAP 30% MgO MAP 30% BaO BAP 30% K 2 O KAP 15% K 2 O 15% MgO KMAP 15% K 2 O 15% BaO KBAP 30% MgO MAP 30% BaO BAP 30% K 2 O KAP 0 Days 3 Days 7 Days Student Poster Student Poster There are a series of steps that occur during corrosion of the glass surface: ?Adsorption of water onto the glass surface ?Diffusion of this water into the bulk glass ?Hydration of the metal cations and phosphate chains ?In aqueous systems, removal of these chains into solution ?Hydrolysis of the chains either in solution or directly on the glass surface ?Crystallization of orthophosphate species Weathering reactions in the ambient environment affect the long-term performance of phosphate glass amplifiers. These weathering reactions appear to have a compositional dependence that is different from that observed in aqueous corrosion experiments. To date, corrosion studies have concentrated on metaphosphate compositions like those used for laser amplifiers. (1) More recent studies have focused on very durable pyrophosphate glasses, including those compositions under development for encapsulating nuclear waste. (2) The goal of this research is to determine and model the underlying mechanisms that control these reactions to explain these compositional effects. 1. Bunker, B. C.,Arnold, G. W.,Wilder, J. A. J. Non-Cryst. Solids , 64, 1984, 291-316. 2. Reis, S. T., Karabulut, M.,Day, D. E. Journal of Non-Crystalline Solids , 292, 2001, 150-157. Glass Compositions ?Five aluminophosphate glass compositions were prepared for this study. All of the compositions contained 10 mol% Al 2 O 3 and 60% P 2 O 5 . The remaining components are as follows: Weathering Experiments ?Rectangular samples were polished and suspended in a humidity chamber. The samples were removed from the chamber periodically, the surfaces ultrasonically cleaned to remove the reacted layer, and the weight change measured. Powdered glass samples were also weathered and the reaction products were analyzed by Solid and Liquid NMR as well as Chromatography. All samples were weathered at 50ºC and 80% RH. Aqueous Dissolution ?Samples were polished and suspended in DI water in PP bottles with V/SA ratios of ~12. The bottles were then placed into number of different temperature ovens and were removed periodically, dried, and the weight loss due to corrosion measured. Dissolution time (days) 0 5 10 15 20 25 30 pH Change -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 KAP KBAP KMAP Dissolution Time (days) 0 5 10 15 20 25 Weight Loss (mg/cm 2 ) -6 -5 -4 -3 -2 -1 0 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 KAP KMAP ?The trends of pH vs. time show a small decrease in pH with corrosion time. ?From the graph of weight loss vs. time for KAP and KMAP examples, it appears that we are in the linear region at least 2 weeks into the experiment. 1.2± 1 x10-8 1.2 ± 4 x10-8 1.2 ± 3 x10 -8 7.9± 0.5 x10 -9 1.2 ± 0.3 x10 -7 50 ºC 1.0 ±0.3 x10-7 3.1 ± 1.8 x10-9 KBAP 3.7 ±1.6 x10-8 1.6 ± 2 x10-9 KMAP 5.6 ±3 x10 -7 5.0x10-8 7.5 ± 4 x10 -9 MAP 1.8 ±0.6 x10 -7 5.0x10 -8 BAP 6.1 ±0.4 x10 -6 3.2x10 -6 1.1 ± 0.15 x10 -7 5.2 ± 1.5 x10 -9 KAP 90º C 70º C* 4 0ºC 2 3ºC Glass 1.2± 1 x10-8 1.2 ± 4 x10-8 1.2 ± 3 x10 -8 7.9± 0.5 x10 -9 1.2 ± 0.3 x10 -7 50 ºC 1.0 ±0.3 x10-7 3.1 ± 1.8 x10-9 KBAP 3.7 ±1.6 x10-8 1.6 ± 2 x10-9 KMAP 5.6 ±3 x10 -7 5.0x10-8 7.5 ± 4 x10 -9 MAP 1.8 ±0.6 x10 -7 5.0x10 -8 BAP 6.1 ±0.4 x10 -6 3.2x10 -6 1.1 ± 0.15 x10 -7 5.2 ± 1.5 x10 -9 KAP 90º C 70º C* 4 0ºC 2 3ºC Glass *E.Metwalli&R. K. Brow, JNCS, 289 (2001) 113-122 ?Activation energies for the temperature dependence of the aqueous corrosion rates for one week in DI water were calculated using the Arrhenius equation ?Table shows the dissolution rates (g/cm 2 *min) after 1 week for the studied glass compositions. Temperature (ºC) 50 100 150 200 250 300 350 400 450 Weight Loss (%) 86 88 90 92 94 96 98 100 5 Days 10 Days 14 Days 0 Days 7 hours 3 Hours 1 Day ?Thermogravimetric analysis shows the presence of two different weight loss regions, one of which can be removed with vacuum drying at 50ºC. ?The first weight loss continues to increase with weathering time. The second weight loss reaches a maximum of ~4 wt% and remains constant at this level with corrosion time. Adsorbed Water Reacted Water ?The table shows the normalized weight changes for the glasses after 5 days of weathering as well as the thickness this layer is equivalent to. -1.8x10 -3 -1.3x10 -4 -1.6x10-5 -7.8x10-7 +5.7x10 -6 5 Day Weight Change (g/cm 2 ) 750 51 5.6 0.24 -- Thickness of layer removed (nm) KAP MAP KBAP BAP KMAP Glass -1.8x10 -3 -1.3x10 -4 -1.6x10-5 -7.8x10-7 +5.7x10 -6 5 Day Weight Change (g/cm 2 ) 750 51 5.6 0.24 -- Thickness of layer removed (nm) KAP MAP KBAP BAP KMAP Glass Temperature (ºC) 50 100 150 200 250 300 350 400 450 500 550 Weight Loss (%) 92 93 94 95 96 97 98 99 100 5 Days Corrosion Vacuum Oven Dried Ambient Oven Dried Adsorbed Water Reacted Water 2Theta (º) 10 20 30 40 50 60 70 Diffracted Intensity (cps) BAP KBAP ?Crystals were observed to grow on the glass surfaces after extended periods of weathering time on all glasses. An SEM image of crystals on the surface of KBAP is shown to the right. EDS analysis indicate the composition to be similar to that of the bulk glass. ?XRD shows that KH 2 PO 4 grows on the weathered surface of KAP glass samples, but monoclinic AlPO 4 forms on the surface of KBAP and BAP. A second crystal also forms on the BAP surface. ? 31 P MAS spectra show the growth of multiple phosphorus species with increasing corrosion time. ?2D correlation spectra show that each new peak interacts with proton species from adsorbed water.(~5ppm) The 31 P peak at ~20ppm also shows a strong correlation with a second proton environment. ?Liquid NMR spectra show the distribution of Q n units in the weathered after 7 days at 50ºC and 80% RH for three of the glasses. The spectra show that the KAP surface layer has undergone the most hydrolysis, followed by KMAP and KBAP. ?Liquid Chromatography gives similar results, as well as the distribution of phosphate chain lengths for the glass surface. At right is a figure showing the distribution of phosphate chains in KAP before and after 2 weeks of weathering. ?The dissolution and weathering rates for a series of phosphate glasses have been studied and appear to depend on the modifying cations. ?The corrosion rates of these glasses can be compared using a free energy of hydration model for the modifying cations: ?In aqueous dissolution it follows the free energy of aqueous hydroxide species giving the expected order of stability by Mg>Ba>K. (3-5) ?In weathering experiments, however, the Ba glasses appear to be more durable than the Mg glasses. If we look at the free energy of the crystalline hydroxide species, the trend changes to Ba>Mg>K. (6) 5. Watanabe, M. Bull. Chem. Soc. Jpn., 55,1982, 3766-9. 4. Parks, J. R.,VanWazer, J. R. J. Am. Chem. Soc. , 79,1957 , 4890 -7. 3. Thilo, E.,Wieker , W. Journal of Polymer Science, 53, 1961, 55-58. 6. Paul, A. J. Mater. Sci., 12,1977, 2246-68. ?Further determinations of the reaction species on the glass surfaces using solid and liquid state NMR and Chromatography measurements of the glass structure. ?Studies of pyrophosphate glass systems to determine the mechanism behind the anomalous durability ?pH stat titration of the glass systems to determine the concentration of surface reaction sites, the kinetics of the reactions, and the effect of pH on the reactions. ?Weathering experiments to determine the effect of both temperature and RH on the corrosion mechanisms as well as the effect of surface finish on corrosion. ?The data from these techniques and others will be applied to the preliminary model to help develop a mechanistic understanding of phosphate glass corrosion.
Transcript of Weathering Reactions on Phosphate Glass...
Weathering Reactions on Phosphate Glass SurfacesWeathering Reactions on Phosphate Glass SurfacesB. C. Tischendorf and R. K. BrowB. C. Tischendorf and R. K. BrowB. C. Tischendorf and R. K. Brow
University of MissouriUniversity of MissouriUniversity of Missouri---RollaRollaRollaRolla MO 65401Rolla MO 65401Rolla MO 65401
IntroductionIntroduction
DiscussionDiscussion
AcknowledgementsAcknowledgements
Future WorkFuture Work
ppm-30-25-20-15-10-50510
inte
nsity
(ar
bitr
ary
units
)
WKMAP WKBAP WKAP
Q2Q1Q0
Dissolution Activation EnergiesDissolution Activation Energies
Weathering DataWeathering Data
Corrosion ModelCorrosion Model
Weathering Surface SpeciesWeathering Surface Species
Experimental ProcedureExperimental Procedure
This work is funded by Lawrence Livermore National Laboratory. Work was performed under the auspices of the US Department of Energy by of Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48.Column Retention Time (min)
0 5 10 15 20 25 30 35 40 45 50
So
lute
Co
nd
uct
ivity
(uS
)
KAP Weathered 0 DaysKAP Weathered 14 Days
P1
P2P1
P2
P3
P3m
P3
P4m
P4
P5m
P5P6 P7P8...
KBAP10 Days Corrosion
80%RH at 50ºC
1/T (*1000 ºK)0.0027 0.0028 0.0029 0.0030 0.0031 0.0032 0.0033 0.0034
23.0 (+/- 2.8)KAP
19.2 (+/- 2.9)MAP
18..4 (+/- 1.3)BAP
11.2 (+/- 0.73)KBAP
9.9 (+/- 2.3)KMAP
Ea (kcal/mole)Glass
23.0 (+/- 2.8)KAP
19.2 (+/- 2.9)MAP
18..4 (+/- 1.3)BAP
11.2 (+/- 0.73)KBAP
9.9 (+/- 2.3)KMAP
Ea (kcal/mole)Glass
Bulk Dissolution DataBulk Dissolution Data
Weathered Layer StructureWeathered Layer Structure
Chain Distribution of Weathering LayerChain Distribution of Weathering Layer
?Microscope images of the glass surface before and after weathering show accelerated corrosion at surface defects.
15% K2O15% MgO
KMAP
15% K2O15% BaO
KBAP30% MgOMAP
30% BaOBAP
30% K2OKAP
15% K2O15% MgO
KMAP
15% K2O15% BaO
KBAP30% MgOMAP
30% BaOBAP
30% K2OKAP
0 Days
3 Days
7 Days
Student PosterStudent Poster
There are a series of steps that occur during corrosion of the glass surface:
?Adsorption of water onto the glass surface
?Diffusion of this water into the bulk glass
?Hydration of the metal cations and phosphate chains
?In aqueous systems, removal of these chains into solution
?Hydrolysis of the chains either in solution or directly on the glass surface
?Crystallization of orthophosphate species
Weathering reactions in the ambient environment affect the long-term performance of phosphate glass amplifiers. These weathering reactions appear to have a compositional dependence that is different from that observed in aqueous corrosion experiments. To date, corrosion studies have concentrated on metaphosphate compositions like those used for laser amplifiers.(1) More recent studies have focused on very durable pyrophosphate glasses, including those compositions under development for encapsulating nuclear waste. (2) The goal of this research is to determine and model the underlying mechanisms that control these reactions to explain these compositional effects.
1. Bunker, B. C.,Arnold, G. W.,Wilder, J. A. J. Non-Cryst. Solids , 64, 1984, 291-316.2. Reis, S. T.,Karabulut, M.,Day, D. E. Journal of Non-Crystalline Solids, 292,2001, 150-157.
Glass Compositions?Five aluminophosphate glass compositions were prepared for this study. All of the compositions contained 10 mol% Al2O3 and 60% P2O5. The remaining components are as follows:
Weathering Experiments?Rectangular samples were polished and suspended in a humidity chamber. The samples were removed from the chamber periodically, the surfaces ultrasonically cleaned to remove the reacted layer, and the weight change measured. Powdered glass samples were also weathered and the reaction products were analyzed by Solid and Liquid NMR as well as Chromatography. All samples were weathered at 50ºC and 80% RH.
Aqueous Dissolution?Samples were polished and suspended in DI water in PP bottles with V/SA ratios of ~12. The bottles were then placed into number of different temperature ovens and were removed periodically, dried, and the weight loss due to corrosion measured.
Dissolution time (days)
0 5 10 15 20 25 30
pH C
hang
e
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
KAP KBAP KMAP
Dissolution Time (days)
0 5 10 15 20 25
Wei
gh
t L
oss
(m
g/c
m2 )
- 6
- 5
- 4
- 3
- 2
- 1
0
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
KAP KMAP
?The trends of pH vs. time show a small decrease in pH with corrosion time.
?From the graph of weight loss vs. time for KAP and KMAP examples, it appears that we are in the linear region at least 2 weeks into the experiment.
1.2± 1 x10-8
1.2 ± 4 x10 -8
1.2 ± 3 x10 -8
7.9± 0.5 x10-9
1.2 ± 0.3 x10-7
50ºC
1.0 ± 0.3 x10-73.1 ± 1.8 x10-9KBAP
3.7 ± 1.6 x10-81.6 ± 2 x10-9KMAP
5.6 ± 3 x10-75.0x10-87.5 ± 4 x10-9MAP
1.8 ± 0.6 x10-75.0x10-8BAP
6.1 ± 0.4 x10-63.2x10-61.1 ± 0.15 x10-75.2 ± 1.5 x10-9KAP
90ºC70ºC*40ºC23ºCGlass
1.2± 1 x10-8
1.2 ± 4 x10 -8
1.2 ± 3 x10 -8
7.9± 0.5 x10-9
1.2 ± 0.3 x10-7
50ºC
1.0 ± 0.3 x10-73.1 ± 1.8 x10-9KBAP
3.7 ± 1.6 x10-81.6 ± 2 x10-9KMAP
5.6 ± 3 x10-75.0x10-87.5 ± 4 x10-9MAP
1.8 ± 0.6 x10-75.0x10-8BAP
6.1 ± 0.4 x10-63.2x10-61.1 ± 0.15 x10-75.2 ± 1.5 x10-9KAP
90ºC70ºC*40ºC23ºCGlass*E.Metwalli&R.K. Brow, JNCS, 289 (2001) 113-122
?Activation energies for the temperature dependence of the aqueous corrosion rates for one week in DI water were calculated using the Arrhenius equation
?Table shows the dissolution rates (g/cm2*min) after 1 week for the studied glass compositions.
Temperature (ºC)
50 100 150 200 250 300 350 400 450
Wei
gh
t L
oss
(%
)
86
88
90
92
94
96
98
100
5 Days10 Days14 Days
0 Days
7 hours3 Hours
1 Day
?Thermogravimetric analysis shows the presence of two different weight loss regions, one of which can be removed with vacuum drying at 50ºC. ?The first weight loss continues to increase with weathering time. The second weight loss reaches a maximum of ~4 wt% and remains constant at this level with corrosion time.
AdsorbedWater Reacted
Water
?The table shows the normalized weight changes for the glasses after 5 days of weathering as well as the thickness this layer is equivalent to.
-1.8x10- 3
-1.3x10- 4
-1.6x10- 5
-7.8x10- 7
+5.7x10- 6
5 Day Weight Change (g/cm2)
750
51
5.6
0.24
--
Thickness of layer removed (nm)
KAP
MAP
KBAP
BAP
KMAP
Glass
-1.8x10- 3
-1.3x10- 4
-1.6x10- 5
-7.8x10- 7
+5.7x10- 6
5 Day Weight Change (g/cm2)
750
51
5.6
0.24
--
Thickness of layer removed (nm)
KAP
MAP
KBAP
BAP
KMAP
Glass
Temperature (ºC)
50 100 150 200 250 300 350 400 450 500 550
Wei
ght
Loss
(%
)
92
93
94
95
96
97
98
99
100
5 Days CorrosionVacuum Oven DriedAmbient Oven Dried
AdsorbedWater
ReactedWater
2Theta (º)10 20 30 40 50 60 70
Diff
ract
ed In
tens
ity (
cps)
KAP
BAP
KBAP
?Crystals were observed to grow on the glass surfaces after extended periods of weathering time on all glasses. An SEM image of crystals on the surface of KBAP is shown to the right. EDS analysis indicate the composition to be similar to that of the bulk glass.
?XRD shows that KH2PO4 grows on the weathered surface of KAP glass samples, but monoclinic AlPO4 forms on the surface of KBAP and BAP. A second crystal also forms on the BAP surface.
?31P MAS spectra show the growth of multiple phosphorus species with increasing corrosion time.
?2D correlation spectra show that each new peak interacts with proton species from adsorbed water.(~5ppm) The 31P peak at ~20ppm also shows a strong correlation with a second proton environment.
?Liquid NMR spectra show the distribution of Qn units in the weathered after 7 days at 50ºC and 80% RH for three of the glasses. The spectra show that the KAP surface layer has undergone the most hydrolysis, followed by KMAP and KBAP.
?Liquid Chromatography gives similar results, as well as the distribution of phosphate chain lengths for the glass surface. At right is a figure showing the distribution of phosphate chains in KAP before and after 2 weeks of weathering.
?The dissolution and weathering rates for a series of phosphate glasses have been studied and appear to depend on the modifying cations.?The corrosion rates of these glasses can be compared using a free energy of hydration model for the modifying cations:?In aqueous dissolution it follows the free energy of aqueous hydroxide species giving the expected order of stability by Mg>Ba>K.(3-5)
?In weathering experiments, however, the Ba glasses appear to be more durable than the Mg glasses. If we look at the free energy of the crystalline hydroxide species, the trend changes to Ba>Mg>K.(6)
5. Watanabe, M. Bull. Chem. Soc. Jpn., 55,1982, 3766-9.
4. Parks, J. R.,Van Wazer, J. R. J. Am. Chem. Soc., 79,1957, 4890-7.
3. Thilo, E.,Wieker , W. Journal of Polymer Science, 53,1961, 55-58.6. Paul, A. J. Mater. Sci., 12,1977, 2246-68.
?Further determinations of the reaction species on the glass surfaces using solid and liquid state NMR and Chromatography measurements of the glass structure.?Studies of pyrophosphate glass systems to determine the mechanism behind the anomalous durability?pH stat titration of the glass systems to determine the concentration of surface reaction sites, the kinetics of the reactions, and the effect of pH on the reactions.?Weathering experiments to determine the effect of both temperature and RH on the corrosion mechanisms as well as the effect of surface finish on corrosion.?The data from these techniques and others will be applied to thepreliminary model to help develop a mechanistic understanding of phosphate glass corrosion.
