Professional Documents
Culture Documents
80Th Conference On Glass Problems S K Sundaram Full Chapter
80Th Conference On Glass Problems S K Sundaram Full Chapter
K. Sundaram
Visit to download the full and correct content document:
https://ebookmass.com/product/80th-conference-on-glass-problems-s-k-sundaram/
80th Conference
on Glass Problems
80th Conference
on Glass Problems
Ceramic Transaction, Volume 268
A Collection of Papers Presented at the 80th Conference on
Glass Problems
Greater Columbus Convention Center, Columbus, Ohio
October 28-31, 2019
Edited by
S. K. Sundaram
This edition first published 2021
© 2021 The American Ceramic Society
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by
any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain
permission to reuse material from this title is available at http://www.wiley.com/go/permissions.
The right of S K Sundaram to be identified as the author of the editorial material in this work have been asserted in
accordance with law
Registered Office
John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA
Editorial Office
111 River Street, Hoboken, NJ 07030, USA
For details of our global editorial offices, customer services, and more information about Wiley products visit us at
www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard
print versions of this book may not be available in other formats.
10 9 8 7 6 5 4 3 2 1
Contents
Foreword vii
Preface ix
Acknowledgments xi
PLENARY SESSION
Peculiar Wear Behavior of Soda Lime Silicate Glass in Humid Air And Its Implications 13
Nisha Sheth and Seong H. Kim
Electric Boosting and Hybrid Furnaces (Practical Application of Higher Levels of Electric 53
Heat Input)
W. Kuhn, A. Reynolds, P. Molcan, and B. Malphettes
Carbon Reduction with Super Boosting and Advanced Energy Management using 71
Renewable Resources
H. P. H. Muijsenberg, Hans Mahrenholtz, Petr Jandacek, Stuart Hakes, and Christoph Jatzwauk
Bag Filter And Catalyst (SCR) – Does This Fit Together? 109
Ruediger Margraf
REFRACTORIES
New Tuckstone Refractory Solution for Long Life Glass Furnace Superstructure 133
Isabelle Cabodi, Pierrick Vespa, Thibaut Chuffart, and Michel Gaubil
Optimization and Energy Savings Especially in Container Glass Production by Using a 141
Refractory Coating
Rolf Weigand and Heiko Hessenkemper
153
Paulus Schreuders
Refractories
Larry McCloskey, Anchor Acquisition, LLC, Lancaster, OH and Eric Dirlam,
Ardagh Group, Muncie, IN
This volume is a collection of papers presented at the 80th year of the Glass
Problems Conference (GPC) in 2019. The GPC continues the tradition of
publishing the papers that goes back to 1934. The manuscripts included in this
volume are reproduced as furnished by the presenting authors, but were reviewed
prior to the presentation and submission by the respective session chairs. These
chairs are also the members of the GPC Advisory Board.
As the Program Director of the GPC, I am thankful to all the presenters at the 80th
GPC. This year’s meeting was record breaking in many sense. We had a total of
570 registered attendees including 40 students from across the country. I appreciate
all the support from the members of Advisory Board. Their volunteering sprit,
generosity, professionalism, and commitment were critical to the high quality
technical program at this Conference. I also appreciate continuing support and
strong leadership from the Conference Director, Mr. Robert Weisenburger Lipetz,
Executive Director of GMIC and excellent support from Ms. Donna Banks of
GMIC in organizing the GPC. I look forward to continuing our work with the
entire team in the future.
Please note that The American Ceramic Society and myself did minor editing and
formatting of these papers. Neither Alfred University nor GMIC is responsible
for the statements and opinions expressed in this volume.
S. K. Sundaram
Alfred, NY
March 2020
Finally, I am indebted to Donna Banks, GMIC for her patience, support, and
attention to detail in making this conference a big success and this Proceedings
possible.
FUTURE OF OXY-FUEL GLASS MELTING: OXYGEN PRODUCTION, ENERGY
EFFICIENCY, EMISSIONS AND CO2 NEUTRAL GLASS MELTING
Hisashi Kobayashi
Praxair, Inc.
Danbury, CT 06810
ABSTRACT
Over 300 commercial glass melting furnaces have been successfully converted to oxy-fuel
firing worldwide since 1991 when the first full oxy-fuel conversion of a large container glass
furnace took place. The main benefits of oxy-fuel conversion are fuel reduction, glass quality
improvement, emissions reduction (CO2, CO, NOx, SO2, particulates), and productivity
improvements. Significant changes in the melting and fining behaviors were observed under
oxy-fuel firing. Most furnaces required some batch modifications to optimize the glass fining
chemistry and to control foam. Improved oxy-fuel burner and furnace designs have reduced
alkali volatilization and silica crown corrosion. Silica crown is expected to last for a full
furnace campaign, especially with new no-lime silica bricks. Today most of high-quality
specialty glass products such as LCD display glass and fiber glass are melted in oxy-fuel fired
glass furnaces. Oxy-fuel conversion of large soda lime glass furnaces, however, has been
limited to about sixty container and ten float/flat glass furnaces due to the additional cost of
using oxygen. Key factors to improve the economics of oxy-fuel fired such as efficiency of air
separation technology and waste heat recovery are reviewed. The potential of using hydrogen
and renewable fuels with oxygen to reduce CO2 emissions is also discussed.
(key words: oxy-fuel, glass melting, CO2 reduction, hydrogen combustion)
INTRODUCTION
In 1988, the U.S. Department of Energy awarded a program to Praxair, Inc. (a member of
the Linde group now) to demonstrate the use of oxy-fuel combustion in a large commercial glass
furnace using an on-site vacuum-pressure swing adsorption (VPSA) technology. A container glass
furnace at Gallo Glass Company was rebuilt in 1991 as the first large scale oxy-fuel fired
furnace1. The successful conversion of the furnace and the demonstration of significant fuel
savings (15%) and emissions reduction (80% reduction in NOx and CO, and 30% particulates)
stimulated the glass industry to adopt the new technology at a rapid rate. By 1996 about 90
commercial glass furnaces were converted to oxy-fuel firing worldwide2. Although the rate of
oxy-fuel firing conversions slowed down since then, over 300 commercial glass furnaces are fired
with oxygen today. Most of specialty glass furnaces such as LCD glass furnaces are fired with
oxygen as high glass melting temperature, relatively small furnace size and the high glass quality
requirement made oxy-fuel firing more economic. Over one hundred insulation and reinforcing
glass fiber furnaces have been converted to oxygen firing as large fuel savings are achieved when
air fired recuperative furnaces are converted to oxy-fuel firing. About fifty container glass
furnaces and about ten float/flat glass furnaces have been converted for NOx reduction,
production rate increase, and capital cost reduction.
Most of fuel efficiency gains of oxy-fuel fired furnaces come from the elimination of
nitrogen contained in combustion air (i.e., about 78% N2 and 1% Ar by volume) and the
corresponding reduction in the flue gas sensible heat loss3. Fuel savings of 5 to 50% have been
achieved without using any flue gas heat recovery systems under oxy-fuel firing as compared with
8 0th Conference on Glass Problems, Ceramic Transaction Volume 26 8 , First Edition. Edited by S. K. Sundaram.
© 2021 The American Ceramic Society. Published 2021 by John Wiley & Sons, Inc.
various air fired furnaces. Fuel savings achievable by oxy-fuel conversion depend on the type of
heat recovery systems used in the air fired furnaces and their conditions. About 10 to 15% fuel
savings have been achieved on the furnace campaign average for large container and float glass
furnaces equipped with efficient regenerators to preheat combustion air to about 1300C. The
efficiency of regenerators deteriorates with furnace age due mainly to deposits build up in the
regenerator passage and to increase in air infiltration4. For example, specific fuel consumption for
an air fired regenerative furnace may increase by 16% over 12 years (i.e., 1.35% per year)5, while
that for oxy-fuel fired furnace without heat recovery may increase only by 6% over 12 years. Thus,
fuel savings by oxy-fuel firing is relatively small in early furnace campaign and increases as the
furnace ages. For fiber glass furnaces with metallic recuperators fuel savings by oxy-fuel conversion
are typically in a range of 30 to 50%. Metallic recuperators can preheat combustion air only up to
about 800C and the furnace energy efficiency is significantly lower than the furnaces equipped with
regenerators. For small specialty glass furnaces operating at high temperatures, fuel savings over
50% have been achieved in some furnaces since small recuperators and regenerators are not very
efficient.
Reduction of NOx emissions was an important benefit and an economic driver for oxy-
fuel conversion, especially in the U.S.. Due to the high furnace temperature required for glass
melting significant “thermal NOx” is formed in the flame region. The rate of formation of thermal
NOx is strongly temperature dependent and approximately proportional to the concentration of
nitrogen in the furnace. The conversion of an air fired furnace to oxy-fuel firing typically results in
NOx reduction by 80 to 90% as the nitrogen concentration in the furnace is reduced from about
70% in the air fired furnace to about 5 to 10% in typical oxy-fuel fired furnaces. Other key factors
influencing NOx emission are oxy-fuel burner design which influences the peak flame temperature,
excess oxygen and batchs niter content 6.
Melting and fining behaviors change significantly under oxy-fuel firing due to the
interaction between the furnace atmosphere and glassmelt and changes in the heat transfer
characteristics. The concentration of water vapor in the furnace atmosphere is about 16-18% in
the air-natural gas fired furnace, which increases to 50-55% in the oxy-fuel fired furnace. Higher
water vapor concentration increases water dissolution into glassmelt and enhances fining
reactions7. Extensive laboratory studies and mathematical modeling have been conducted to
investigate heat transfer, glass fining, alkali volatilization and refractory corrosion mechanisms
under oxy-fuel firing. Most furnaces required some batch modifications to optimize the glass
fining chemistry8. Although accelerated silica crown corrosion was experienced in early
conversions, improved burner and furnace designs and the development of new silica crown
materials with low of no lime extended the life of the silica crown close to that of a conventional
air fired furnace9-10. A review paper11 describes technical differences between oxy-fuel fifing and
the air-firing in more details.
Recent advances in oxygen production and oxy-fuel technology aim to make oxy-fuel glass
melting a more cost-effective solution to meet the sustainability goal of CO2 reduction. For example,
the efficiency of air separation technology has improved and the power consumption to produce
oxygen has decreased significantly. Advanced waste heat recovery technologies for oxy-fuel fired
furnaces have also been developed to reduce the fuel and oxygen requirement for oxy-fuel furnaces12-
14
. Oxy-hydrogen combustion is considered a leading option for glass melting of future. This paper
reviews the key economic factors of oxy-fuel glass melting and discusses the future of oxy-fuel
combustion for CO2 neutral glass melting.
Figure 1. fuel and oxygen costs for a generic 300 mtpd (metric ton per day) container with 50%
cullet and no electric boostingg
Figure 2. Historical trend of specific electric power consumption to make VPSA oxygen
Figure 3. Potential improvements in specific fuel consumption for a 300 mtpd container furnace with
50% cullet.
Figure 4. Comparison of CO2 emissions from 300 tpd container glass furnaces at 50% cullet
In order to achieve 90% reduction in CO2 emissions in the future all these sources of CO2
emissions have to be substantially reduced. Raw materials have to be mostly free of carbonate materials
such as cullet. For container glass manufacture cullet ratios as high as 90% are already used in some
furnaces and the trend for higher cullet uses will continue. Electric power has to be either generated from
renewable sources or nuclear plants and CO2 emissions from fossil fuel burning power plants have to be
captured and sequestered. For fuel fired glass melting furnaces biomass derived fuels or hydrogen made
without CO2 emissions need to be used.
Table 1. Fuel and oxygen consumption and cost of biomass derived fuels relative to natural gas
for an oxy-fuel fired 300 mtpd container furnace at 50% cullet without electric boost
APPROXIMATE APPROXIMATE
SPECIFIC FUEL SPECIFIC OXYGEN FUEL COST FUEL COST
CONSUMPTION (%) CONSUMPTION (%) ($/GJ in EU) ($/T GLASS in EU)
Natural
Gas 100 100 8 31
Hydrogen 101 85 15 58
Digester
Gas 110 111 6 26
Ethanol 102 97 18 71
Pyrolysis
Oil 121 127 8 38
Dry
Wood 119 117 8 37
Torrefied
Wood 100 88 9 35
Solid fuels such as wood chips, wood char and torrefied wood can be fired with oxygen
in glass furnaces. Since pulverized petroleum coke has been used in glass melting furnaces, there
is no fundamental technical barriers to use solid biomass fuels as long as ash quality is
compatible with the quality of glass produced. Solid fuel firing, however, requires a
pulverization step and a special solid fuel feeding and distribution system and the conversion
from a natural gas fired system is more involved. Dry wood with 5% moisture requires about
20% more fuel and oxygen as compared with natural gas. Torrefied wood is produced in a mild
roasting process and is an excellent dry fuel. Fuel requirement is the same as natural gas and the
oxygen requirement is reduced by about 12%.
SUMMARY
Over 300 oxy-fuel fired glass melting furnaces operate worldwide today and provide the benefits of
fuel reduction, productivity increase and NOx reduction. For wider adoption of oxy-fuel firing the
reduction of fuel and oxygen consumption and the cost of oxygen generation is required. Several
oxy-fuel fired furnaces have been equipped with flue gas heat recovery systems and specific fuel
consumption as low as 3 GJ/t (2.8 MMBtu/ston) have been demonstrated in container furnaces. The
goal of 2.5 GJ/t (2.4 MMBtu/ston) is considered to be technically feasible with 50% reduction in
wall heat loss combined with flue gas waster heat recovery. For CO2, neutral glass melting of future
REFERENCES
1. Tuson, G. B., Higdon, R., and Moore, D., "100% Oxygen Fired Regenerative Container
Glass Melters," Glass 91 52nd Conference on Glass Problems, University of Illinois at
Urbana - Champaign, IL, November 12 to 13, 1991.
2. Ronald W. Schroeder of Praxair, Inc., and Allan E. Zak of Corning, Inc., "Oxy-Fuel
Economics Update Based on Case Histories” the 56th Conference on Glass Problems -
October 1995 at the University of Illinois/Champaign-Urbana
3. Kobayashi, H., "Oxygen Enriched Combustion System Performance Study," Vol. I Technical
and Economic Analysis. Prepared for U. S. Dept. of Energy Idaho Operations Office, Idaho
Falls, ID, Report No. DOE/ID/12597, March 1987.
4. Beerkens, R.G.C.,”The Most Energy Efficient Glass Furnace: Energy Efficiency
Benchmarking and Energy Saving Measures for Industrial Glass Furnaces”, A.T.I.V.
Conference - Parma (Italy), 2006
5. Beerkens, R.G.C., H. Van Limpt, ”Energy Efficiency of Glass Furnaces” , Presented at
GMIC Workshop on Evolutionary and Revolutionary Strategies for Keeping Glass Viable
through the 21st Century”, July 30-31, 2003, Rochester, NY
6. Kobayashi, H., G. B. Tuson and E. J. Lauwers, "NOx Emissions from Oxy-Fuel Fired Glass
Melting Furnaces," European Society of Glass Science and Technology Conference on
Fundamentals of the Glass Manufacturing Process, Sheffield, England, September 9-11,
1991.
7. Beerkens, R.G.C., Laimbock, P., A.J. Faber, and Kobayashi, H., "Interaction between
furnace atmospheres and sulfate fined glass melts", Proceedings of XVIII International
Congress on Glass, July 5-10, 1998, San Francisco, CA, USA
8. Kobayashi, H., and Beerkens, R.G.C., "Reduction of SO2 Emissions of Oxy-Fuel Firing –
Water Enhanced Sulfate Fining," Fifth International Conference on Advances in the Fusion
and Processing of Glass, Toronto, Canada, July 27-31, 1997
9. Whittemore, D.S., R.F. Spaulding, H.E. Wolfe, and J.T. Brown, “New Silica Refractory for
Oxy-Fuel Glass Melting”, J. Int. Glass, vol.102, pp120-124, 1999
10. Gonzalez, A.R., J.T. Brown, R.P. Weilacher, and M.A. Nelson, “Review of Improved Silica
Crown Refractory and Practices for Oxy-Fuel Fired glass Melters”, 64th Conference on
Glass Problems, University of Illinois at Urbana – Champaign, IL, October 28-29, 2003.
11. H. Kobayashi, “Advance in Oxy-Fuel Fired Glass Melting Technology”, Proceedings of XX
International Congress on Glass, Sep. 26-Oct.1, 2004, Kyoto Japan
12. M. van Valburg, E. Sperry, S. Laux, R. Bell, A. Francis, and H. Kobayashi, “Design and
Implementation of OPTIMELT™ Heat Recovery
for an Oxy-Fuel Furnace at Libbey Leerdam”, 78th Conference on Glass Problems,
Columbus, OH, October, 2017.
13. M. van Valburg, F. Schuurmans, E. Sperry, S. Laux, R. Bell, A. Francis, S. Chakravarti and
H. Kobayashi, “Operating Experience with the OPTIMELT™ Heat Recovery Technology on
a Tableware Glass Furnace”, 79th Conference on Glass Problems, Columbus, OH, October,
2018.
Nisha Sheth
Vitro Architectural Glass
Cheswick, PA, 15024, USA
Seong H. Kim
The Pennsylvania State University
University Park, PA 16802, USA
ABSTRACT
Soda lime silicate glass shows unusual wear and scratch behavior as a function of humidity.
Under moderate wear conditions, the surface damage produced is entirely through the interfacial
shear along a direction tangential to the surface. In contrast, severe wear tends to occur at much
larger contact pressures resulting in damage due to indentation and shear stress (e.g. scratching).
In dry environments under moderate wear conditions, soda lime glass is easily abraded by harder
counter surface materials. Yet, in humid environments, soda lime silicate glass has unusually high
wear resistance to counter-surface materials that are significantly harder or more chemically
durable. This high wear resistance of soda lime glass can be attributed to mechanically induced
chemical reactions that may suppress shear-induced damage. In the past, the high wear resistance
of soda lime glass has been attributed to leachable sodium ions that may participate in suppressing
shear-induced damage. Yet, recent work casts doubt on this hypothesis. Heat strengthened soda
lime glass is highly susceptible to wear and scratching, indicating that wear behavior cannot be
exclusively attributed to leachable sodium ions. Instead, it was hypothesized that the strained Si-
O network becomes more susceptible to shear-induced hydrolysis resulting in greater damage
under moderate wear conditions at high humidity. Under severe wear conditions in the presence
of water, heat strengthened glass is also more susceptible to time-delayed chipping compared to
annealed glass. It was hypothesized that the more open silicate network structure of heat
strengthened glass may increase the transport kinetics of molecular water to critical sub-surface
flaws. This may result in sub-critical crack growth of sub-surface lateral cracks resulting in time-
delay chipping along the width of the scratch.
INTRODUCTION
In the last century, glass for window applications has gained increased utility due to
technological advancements in processing. Strengthening treatments, such as thermal tempering,
have addressed durability and safety concerns of brittle and fragile glass windows. Thermal
tempering also increased the strength to weight ratio for glass materials that can lead to lower costs
associated with transportation. Recent technological advancements in coatings have enabled
architectural glass manufacturers to more precisely control visible light transmission through
buildings, filter solar radiation, and provide better thermal insulation allowing for more energy-
efficient buildings and increased comfort for building occupants. However, anecdotal and
empirical evidence shows that heat strengthened glasses have a greater sensitivity to scratching
than annealed glasses.[1,2] Since scratches can act as stress concentrators and may grow larger
under applied load, damage-resistance is critical for applications that require high strength.
8 0th Conference on Glass Problems, Ceramic Transaction Volume 26 8 , First Edition. Edited by S. K. Sundaram.
© 2021 The American Ceramic Society. Published 2021 by John Wiley & Sons, Inc.
Furthermore, for value-added glazing products, such as low emissivity insulated glazing units
(IGU), surface damage and the associated replacement can be costly.
For much of the architectural glazing product lifecycle (Figure 1), glass products are
frequently exposed and become vulnerable to scratching defects that can compromise their
aesthetics, functionality, and strength. During construction and use, glass is exposed to both
organic debris (such as adhesives) and inorganic debris (such as concrete overspray and welding
spots). Post-construction cleaning to remove accumulated construction debris is often
challenging—requiring harsher cleaning techniques at the risk of scratching the glass surface. This
is especially true for tempered glass that appears to be highly scratch-sensitive.
The Glass Committee of the International Window Cleaning Association (IWCA), a trade
organization for the health and safety of professional window cleaners, suggested protocols for
cleaning architectural glass post-construction and recommend glass scratch liability waivers for
all window cleaning contracts, especially for cleaning tempered glass. Since anecdotal and
empirical evidence shows that tempered glass could be scratched more easily than annealed glass,
many window cleaners concerned about post-construction cleaning of tempered glass requested
glass scratch liability waivers. Even then, making scratches on heat strengthened glass could cost
millions of dollars due to the replacement of damaged glass products and lawsuits. At times,
window cleaners are blamed for the damages that are attributed to metal scrapers and harsh
cleaning techniques. Other times, fabricators may also be blamed for the damages attributed to
hard-to-see “abrasive fabrication debris” that are difficult to clean with a standard scraper without
scratching the glass.
Currently, there are no viable solutions to this issue. Professional window cleaners claim
that there are no suitable alternatives to metal scrapers to remove paint, stickers, or adhesives due
to health, time, and ergonomic concerns. Furthermore, they argue that metal scrapers are typical
and infrequently scratch glass. Instead, window cleaners often suggest that the air-side and tin-side
“feel” different with one side “feeling” and “sounding” grittier and having more “drag”.
Fabricators claim that there is no method to test for fabrication debris that may become fused to
the glass as it is in contact with the rollers in the tempering furnace. Microscope images have not
been able to properly capture any supposed fabrication debris.
Table 1. Summary of moderate wear behavior of the glass substrate in high humidity (RH 90%)
environments for various tribosystems.
Counter-surface Stainless Al2O3 Si3N4 Pyrex
Steel H: 20 H: 15 GPa borosilicate
Substrate H: 7 GPa GPa H: 5.5 GPa
G: 26 GPa
Soda lime silicate Wear Wear Wear Wear
Hardness (H): 5.4 GPa resistant [7] resistant resistant [7] resistant
Shear modulus (G): 19.6 - 34.2 GPa [7] [1,7–9,11–
13]
Na+ K+ ion-exchanged soda lime - - - Damaged
silicate [11]
Na+ K+ Na+ ion-exchanged - - - Wear
soda lime silicate resistant [11]
Thermally poled soda lime silicate - - - Wear
(cathode-side) resistant [12]
H: 5.8 GPa
Thermally poled soda lime silicate - - - Damaged
(anode-side) [12]
H: 5.4 GPa
Hydrothermally treated soda lime - - - Damaged
silicate [13]
H: 5.7 GPa
Fused quartz - - - Damaged [8]
H: 5.0 GPa
G: 32 GPa
Alkali borosilicate (Schott BF33) Wear - - Damaged [9]
H: 7.3 GPa resistant [7]
G: 26.3 GPa
Barium boroaluminosilicate - - - Damaged [9]
(Schott AF45)
G: 26.7 GPa
Sodium aluminosilicate Wear - - Damaged
H: 5.8 GPa resistant [7] [10]
Ion-exchanged sodium Wear - - Damaged
aluminosilicate (Corning Gorilla resistant [7] [10]
2)
H: 7.1 GPa
G: 28.5 GPa
80th Conference on Glass Problems 16
Several hypotheses have been proposed attributing the peculiar wear resistance of soda
lime glass at the near-saturation vapor pressure of water in the environment to highly mobile
sodium ions in the adsorbed water layer. [1,5,8–14] One hypothesis suggested that the sodium ions
in soda lime glass may alter the structure and reactivity of the adsorbed water layer influencing
wear outcomes in humid environments.[8,11,12,15]
When sodium ions were exchanged with potassium ions, the Na+-K+ ion-exchanged soda
lime glass lost its high wear resistance and became damaged during moderate wear testing at 90%
RH against a Pyrex borosilicate counter-surface. When the ion-exchanged soda lime glass was
back-exchanged (Na+-K+ - Na+), the glass regained the high wear resistance at 90% RH. [11] While
ion-exchanged soda lime glass has a high compressive stress casing that increased its mechanical
strength under normal indentation conditions, the glass still became damaged easily during
moderate wear testing at 90% RH. These results indicate that hardness, mechanical strength and
fracture toughness of glass do not necessarily correlate with the moderate wear behavior especially
in humid conditions. Furthermore, this data supported the hypothesis that wear resistance is
sensitive to the modifiers in the glass.
Thermal poling is a process where an electrical bias is applied to the glass at elevated
temperatures (~200 ˚C) allowing for modifier cations (e.g. Na+) to migrate from the anode to the
cathode-side. When soda lime glass was thermally-poled, the sodium accumulation side (cathode-
side) demonstrated high wear resistance while the anode side with sodium depletion lost its wear
resistance.[12] When soda lime glass is treated hydrothermally producing a surface that was
hydrated and depleted in sodium, the glass became damaged more easily during moderate wear at
high humidity.[13]
Yet, it is unknown how sodium may alter the structure and reactivity of the adsorbed water
layer on the glass surface and how that may influence wear outcomes. The hydrogen-bonding
network structure of the adsorbed water film depends on the glass composition and surface
conditions.[16] Non-linear optical spectroscopy techniques, such as sum frequency generation
(SFG) vibration spectroscopy, can be used to probe the hydrogen bonding network structure of the
adsorbed water on the glass surface. While the structure of the adsorbed water film was shown to
be distinct depending on glass composition and surface treatment, the SFG spectral features are
complex and a correlation between SFG spectra and wear behavior in humid environments has yet
to be established.
Figure 2. The wear depth of soda lime float glass as a function of humidity for the air-side (a) and
the tin-side (b). Four thermal processing conditions were compared: as-received (○), annealed (□),
heat strengthened ( ), thermally tempered (∆). The dashed lines are drawn by connecting the
annealed surface data as a guide to the eye. The error bars represent standard error.
PATH FORWARD
A vast majority of literature on the mechanical behavior of glasses focuses on mechanical
strength, crack propagation and indentation hardness. Yet, shear-induced damage, such as
scratching, is a poorly understood failure mode that has critical implications for glass products.
ACKNOWLEDGEMENTS. This work was supported by the National Science Foundation (Grant
No. DMR-1609107) and the Usable Glass Strength Consortium (UGSC). During this study, NS
was supported by the NSF Graduate Research Fellowship Program (Grant No. 1255832).
REFERENCES
[1] N. Sheth, A. Howzen, A. Campbell, S. Spengler, H. Liu, C.G. Pantano, S.H. Kim, Effects
of tempering and heat strengthening on hardness, indentation fracture resistance, and wear
of soda lime float glass, Int. J. Appl. Glas. Sci. (2019) ijag.13507. doi:10.1111/ijag.13507.
[2] J. Schneider, S. Schula, W.P. Weinhold, Characterisation of the scratch resistance of
annealed and tempered architectural glass, Thin Solid Films. 520 (2012) 4190–4198.
doi:10.1016/j.tsf.2011.04.104.
[3] A. Alazizi, A.J. Barthel, N.D. Surdyka, J. Luo, S.H. Kim, Vapors in the ambient - A
complication in tribological studies or an engineering solution of tribological problems?,
Friction. 3 (2015) 85–114. doi:10.1007/s40544-015-0083-5.
[4] Z. Chen, X. He, C. Xiao, S. Kim, Effect of Humidity on Friction and Wear—A Critical
Review, Lubricants. 6 (2018) 74. doi:10.3390/lubricants6030074.
[5] H. He, S.H. Kim, L. Qian, Effects of contact pressure, counter-surface and humidity on
wear of soda-lime-silica glass at nanoscale, Tribol. Int. 94 (2016) 675–681.
doi:10.1016/j.triboint.2015.10.027.
[6] P. Bandyopadhyay, A.K. Mukhopadhyay, Role of shear stress in scratch deformation of
soda-lime-silica glass, J. Non. Cryst. Solids. 362 (2013) 101–113.
doi:10.1016/j.jnoncrysol.2012.11.019.
[7] H. He, L. Qian, C.G. Pantano, S.H. Kim, Effects of humidity and counter-surface on
tribochemical wear of soda-lime-silica glass, Wear. 342–343 (2015) 100–106.
doi:10.1016/j.wear.2015.08.016.
[8] L.C. Bradley, Z.R. Dilworth, A.L. Barnette, E. Hsiao, A.J. Barthel, C.G. Pantano, S.H.
Kim, Hydronium ions in soda-lime silicate glass surfaces, J. Am. Ceram. Soc. 96 (2013)
458–463. doi:10.1111/jace.12136.
[9] H. He, L. Qian, C.G. Pantano, S.H. Kim, Mechanochemical Wear of Soda Lime Silica
Glass in Humid Environments, J. Am. Ceram. Soc. 97 (2014) 2061–2068.
doi:10.1111/jace.13014.
James Uhlik
Toledo Engineering Co., Inc.
Toledo, OH 43606
INTRODUCTION
This paper will discuss the resolution of a glass melting problem that caused a quantity of
lost production at a float glass plant.
Significant reduction in pack yield was experienced for an extended period of time due to a
degradation quality from a melting and/or refining problem. After remedial efforts were made
by the plant, followed by additional support provided by several outside consultants to find and
resolve the problem, TECO was asked to quickly respond as an added resource to try to find the
solution.
This objective of this paper will be to illustrate the various actions and tools used to
troubleshoot the problem, and how, in this age of Factory 4.0, there remains the need to ‘place
hands upon the sick’ to heal them (well, it…).
TECO was contacted in mid-December, and given the following five points of information:
1. The plant is facing some difficulty with a bubbles problem in the last 3 months. There are
high concentration of small bubbles in the left side of the ribbon (around 1400), edge lights
are around 30. The content of the bubbles are about 50/50 CO2 and N2, also Ar was found
in small quantities. Pressure is about 15 kPa. Analysis of the surface next to the bubble
points in the direction of refractory corrosion (sample contain alumina and zirconia
mostly).
2. From experiments done, the problem is probably in the furnace (working end and
upstream), sand and cullet were eliminated as a source. Also, as far as can be told, this is
not a mechanical problem (waist coolers, tweel or stirrers).
3. The only weird thing found is that when turning all stirrers to the left –the bubbles spread
throughout the whole ribbon, and while when facing them to the right, the bubbles remain
on the left side, as always (attached is a diagram to explain).
4. Recently there was a refractory problem between port 3 and 4 at the left side, there were 2
events of glass penetration through the vision block located under the patch block. There
is suspicion that the problem lies in the patch blocks – some of it interacts with the glass
(and perhaps parasitic air) causing these bubbles. It may be required to either put up a
second row of patch blocks or replace the first row.
5. Attached are the reports of the bubbles and surface glass from Company 1 and Company
2, as well as an Excel listing of actions done in the last couple of months.
In TECO’s perspective, Clue 1 is that there is a root cause located on, or influenced by, the
left side of the furnace. In the float process, all modeling illustrates the bilateral nature of this
type of furnace. Here is the defect system output received:
8 0th Conference on Glass Problems, Ceramic Transaction Volume 26 8 , First Edition. Edited by S. K. Sundaram.
© 2021 The American Ceramic Society. Published 2021 by John Wiley & Sons, Inc.
Fig. 1 – Defect level on October 1
There are 34 spaces; in 24 hours, that’s approx. 42.35 minutes between spikes, which is
about one of the ‘one side to the other’ firing reversals. Left side of the ribbon is still worse, but
more defect generation occurs when firing over to one side versus the other. Therefore, the source
appears to have some firing/combustion/reversal source or interface.
It should also be added to the initial problem statement, that in on-site discussions and
Q&A with the plant staff, it was learned that there had occurred a color transition and a batching
problem at the start of this defect issue. It is necessary to add into the evaluation any operational
changes or process upset that is coincident with the beginning of a production problem. Therefore,
possibly, a Clue #3.
The main purpose of a site visit and furnace audit is to make an assessment of the physical
structure of the furnace and its systems, and the most important part of this is to assess the key to
melting in any regenerative furnace – the radiant heat transfer. How do the fires look? Excess
oxygen and combustibles data along with the developing capabilities to thermally image and
monitor the process are providing better control of this paramount process, especially when the
process is fine-tuned to a consistent setting. But operational changes to the process, varieties of
combustion systems and port layouts, along with maintenance issues, still require an experienced
eye in visual evaluation of the fuel use efficiency.
Small bubbles and seeds in float glass can have several sources. Most common can be poor
refining, contamination and devitrification. What will the furnace inspection reveal?
THE FIRES
Inspection of each ports’ firing on each side revealed flame characteristics that looked
higher off of the top of glass than what is common, shown here:
As a result of this visual inspection, as well as review of the combustion system settings,
the fires were adjusted over the next day or two. This was done to provide improvement to the
downstream heat transfer to the glass, to improve the refining of the glass, lessen the flame short-
circuiting and refractory impingement, and stop the reversal characteristic evident in the defect
pattern.
THE GLB
The glass level bowl inspection yielded minor areas of concern, where the superstructure
had developed some condensation degradation. The plant quickly sealed this area.
As a result of all these steps (GLB sealing, waist work), during which there was some upset
to the furnace, the periodicity of the defects stopped (adjustments to the furnace firing). As yet, no
exact single root cause of the defect source had been distinctly identified, and as the start of the
problem was concurrent with a batching upset (and possible contamination), while these other
actions were being undertaken, it was discussed and decided to also lower the glass level slightly.
This easily reversible action might release any material or contaminant located near the sidewalls
and producing seeds. After all these actions, the waist coolers were then removed, cleaned and
reinserted, to put the furnace in the best possible stable and steady-state process situation, and
allowed to settle out into steady state production.
Over the next days, due to float furnace glass flows taking up to three days to stabilize, the
defects slowly diminished:
From the first day or two of inspection, the suspect causes and actions identified were:
1. Non-optimized furnace flame characteristics. Possible source of insufficient refining of the
melted glass. Fires were adjusted.
2. Possible left side contamination source. Key areas of the glass level bowl and waist areas
were sealed.
3. Possible left side sidewall upset contamination source. Glass level temporarily dropped.
What was the root cause? No single source was identified, but even though furnace
instrumentation provides the proper appearance of the operation, when a physical inspection is
performed and certain non-optimized furnace parameters are identified and improved, the glass
quality responds!
Tools used were 1) experience in furnace design and operations, 2) extensive modeling
knowledge and 3) eyeballs and sweat. Even with a fully instrumented furnace, measured and
recorded temperatures, flow rates, pressures and other inputs/outputs, still, detailed physical visual
inspection is often required. Even though cameras and thermal systems are increasingly affordable,
even today somebody’s got to watch and/or analyze them (until AI…), and you can’t sensor every
corner. Beyond this, it’s not just the sensor, it’s the engineering, wiring, signal I/O, HMI – and
STILL someone experienced has to assess it!
ACKNOWLEDGMENT
TECO wishes to thank their customer for the use of this information.
Marc Bermard
BERNARD BONNEFOND
Saint-Etienne, FRANCE
ABSTRACT
Use of electric power increases in glass melting. There are several technologies available to
adjust electric power delivery in glass melters : off-load or on-load transformers, SCR (Silicon
Carbide Rectifiers) or IVR (Induction voltage regulators) and VARIVOLT. BERNARD
BONNEFOND founded in 1925 in France is the only manufacturer of VARIVOLT in the world.
Since 1968, more than 1700 VARIVOLT have been delivered, and more than 1300 are used in
glass industry. Every year BERNARD BONNEFOND manufactures between 70 and 90
VARIVOLT, of which 80% are sold to glass producers and glass engineering companies.
This is the reason why BERNARD BONNEFOND developped some new options to allow for
more frequent power variations and drastically reduce the reaction time of the
VARIVOLT through :
Frequency drive allows to use full speed of the variation when the voltage is far from the setpoint
and low speed when approaching. This solution can easily be implemented on existing or new
instalaltions.
Variable reactor, designed for 5 to 10% of the maximum voltage, installed in serie with the
VARIVOLT allows to make a vernier regulation around the setpoint without consequences on
the VARIVOLT itself. The advantages of the VARIVOLT remain. The number of changes on
the VARIVOLT are not impacted. The power or current regulation is immediate. This variable
reactor can be added on existing system and can be by-passed easily.
CONCLUSION:
Although VARIVOLT power adjustment is suited for all type of glasses processed in the glass
industry (E glass, Composite fibers, container glass etc.) BERNARD BONNEFOND can
propose technical solutions to improve power adjustment frequency and reaction time on existing
or new VARIVOLT UNITS on.
8 0th Conference on Glass Problems, Ceramic Transaction Volume 26 8 , First Edition. Edited by S. K. Sundaram.
© 2021 The American Ceramic Society. Published 2021 by John Wiley & Sons, Inc.
SYNCHRONIZED OXY-FUEL BOOST BURNERS FOR ZERO-PORT PERFORMANCE
OPTIMIZATION IN FLOAT GLASS MELTING FURNACES
ABSTRACT
Zero-port oxy-fuel boost burners have become widely accepted in float glass melting furnaces as
a valuable means for increasing glass production and/or improving efficiency. However, boost
burner effectiveness is often limited by flame interaction with the highly turbulent air-fuel flames.
Moreover, both the strength and direction of these oxy-fuel flames versus air-fuel flame
interactions are dramatically shifted following each regenerator reversal cycle. The result of these
effects can include overheating of the charge wall and “snubbing” of the flame, causing
concentrated heat release close to the breast wall and/or flame lofting toward the crown. By
understanding the nature of these interactions, Air Products researchers have developed an
advanced burner technology that is capable of automatically adjusting flame properties
(particularly length, luminosity and momentum) with each regenerator reversal to avoid negative
effects, while maximizing oxy-fuel performance benefits. This development combines Air
Products Process Intelligence technology with the recently commercialized Cleanfire® HRx™
burner. Both the methodology and beneficial results of field implementation of synchronized oxy-
fuel boost burners are presented in this paper.
INTRODUCTION
Oxy-fuel boost burners have been employed in air-fired regenerative furnaces for a few decades
and the benefits are well known [1,2]. The primary benefits include higher furnace efficiency
and/or lower fuel consumption, higher productivity, improved glass quality, and lower NOx [3].
While oxy-boosting has its benefits, there are also some challenges which include maintaining a
consistent, highly luminous flame that can withstand the high levels of turbulence created inside
air-fired regenerative glass furnaces. The oscillatory nature of the firing direction within a
regenerative furnace creates changing air currents and turbulence patterns that can cause the boost
burner flames to loft, deflect, and appear inconsistent. The inconsistent performance of the boost
burner flames can result in overheating of the nearby charge wall. Also, premature flame
shortening or “snubbing” can cause a concentrated heat release close to the breast wall and/or
flame lofting toward the crown, releasing heat in the upper part of the furnace and away from the
glass surface. This is especially true with staged oxy-fuel burners, where the mixing of oxygen and
fuel is delayed by redirecting a portion of the oxygen above or below the flame in order to create
a longer more luminous flame. Staged burners tend to lose flame momentum as staging levels
increase, leading to increased susceptibility to lofting or snubbing due to turbulence. Many
furnaces deal with the problem of boost burner flame turbulence by choosing flame stability over
optimal heating, that is, they choose conservative burner settings (i.e. no staging) so the boost
8 0th Conference on Glass Problems, Ceramic Transaction Volume 26 8 , First Edition. Edited by S. K. Sundaram.
© 2021 The American Ceramic Society. Published 2021 by John Wiley & Sons, Inc.
burner flames look acceptable during each reversal cycle. This leads to poor flame optimization
and the benefits of oxy-boosting are greatly diminished.
To maximize the benefits of oxy-boosting and overcome the present challenges, Air Products
developed a synchronized boosting system that is capable of automatically adjusting flame
properties (particularly length, luminosity and momentum) with each regenerator reversal. The
synchronized boosting system allows furnace engineers to customize each burner’s settings for
each firing direction and local atmosphere conditions. Such a system can ensure that each burner’s
flame quality is maximized to overcome the negative effects of turbulence generated by the
regenerator reversal cycles. Figure 1 depicts potential gas recirculation zones that can develop in
the area between the Port 1 air-fired burners and the Port 0 oxy-fuel boost burners and negatively
impact burner performance.
Figure 1. A zone of recirculated gases can develop between the charge wall and the Port 1 air-fuel
burners. This can cause the boost burner flames to deflect or be drawn into the nearby Port 1 air-
fuel flame.
These recirculation zones can cause the boost burner flames to become entrained into the flow of
the nearest air-fired burner (when firing on the same side), or it can alternately impede or deflect
the flow of a boost burner firing in the opposing direction. This situation was modeled using Ansys
Fluent computational fluid dynamic (CFD) simulation tools. In the simulation, a typical air-fired
regenerative furnace size and geometry was replicated based on the production rate of 650 tons
(U.S.) per day. Figure 2 shows CFD modeling results that validate this observation.
Christmas before the war. There never will be another in any land,
with any peoples, like the Christmas of 1859—on the old plantation.
Days beforehand preparations were in progress for the wedding at
the quarters, and the ball at the “big house.” Children coming home
for the holidays were both amused and delighted to learn that Nancy
Brackenridge was to be the quarter bride. “Nancy a bride! Oh, la!”
they exclaimed. “Why Nancy must be forty years old.” And she was
going to marry Aleck, who, if he would wait a year or two, might
marry Nancy’s daughter. While the young schoolgirls were busy
“letting out” the white satin ball dress that had descended from the
parlor dance to the quarter bride, and were picking out and
freshening up the wreath and corsage bouquet of lilies of the valley
that had been the wedding flowers of the mistress of the big house,
and while the boys were ransacking the distant woods for holly
branches and magnolia boughs, enough for the ballroom as well as
the wedding supper table, the family were busy with the
multitudinous preparations for the annual dance, for which Arlington,
with its ample parlors and halls, and its proverbial hospitality, was
noted far and wide.
The children made molasses gingerbread and sweet potato pies,
and one big bride’s cake, with a real ring in it. They spread the table
in the big quarters nursery, and the boys decorated it with greenery
and a lot of cut paper fly catchers, laid on the roast mutton and pig,
and hot biscuits from the big house kitchen, and the pies and cakes
of the girls’ own make. The girls proceeded to dress Nancy
Brackenridge, pulling together that refractory satin waist which,
though it had been “let out” to its fullest extent, still showed a sad
gap, to be concealed by a dextrous arrangement of some discarded
hair ribbons. Nancy was black as a crow and had rather a startling
look in that dazzling white satin dress and the pure white flowers
pinned to her kinks. At length the girls gave a finishing pat to the
toilet, and their brothers pronounced her “bully,” and called Marthy
Ann to see how fine her mammy was.
As was the custom, the whole household went to the quarters to
witness the wedding. Lewis, the plantation preacher, in a cast-off
swallow-tail coat of Marse Jim’s that was uncomfortably tight,
especially about the waist line, performed the ceremony. Then my
husband advanced and made some remarks, to the effect that this
marriage was a solemn tie, and there must be no shirking of its
duties; they must behave and be faithful to each other; he would
have no foolishness. These remarks, though by no means elegant,
fitted the occasion to a fraction. There were no high flights of
eloquence which the darky mind could not reach, it was plain,
unvarnished admonition.
The following morning, Christmas Day, the field negroes were
summoned to the back porch of the big house, where Marse Jim,
after a few preliminary remarks, distributed the presents—a head
handkerchief, a pocketknife, a pipe, a dress for the baby, shoes for
the growing boy (his first pair, maybe), etc., etc., down the list. Each
gift was received with a “Thankee, sir,” and, perhaps, also a remark
anent its usefulness. Then after Charlotte brought forth the jug of
whisky and the tin cups, and everyone had a comforting dram, they
filed off to the quarters, with a week of holiday before them and a trip
to town to do their little buying.
James Alexander McHatton
The very last Christmas on the old plantation we had a tree. None
of us had ever seen a Christmas tree; there were no cedars or pines,
so we finally settled upon a tall althea bush, hung presents on it, for
all the house servants, as well as for the family and a few guests.
The tree had to be lighted up, so it was postponed till evening. The
idea of the house servants having such a celebration quite upset the
little negroes. I heard one remark, “All us house niggers is going to
be hung on a tree.” Before the dawn of another Christmas the
negroes had become discontented, demoralized and scattered, freer
than the whites, for the blacks recognized no responsibilities
whatever. The family had abandoned the old plantation home. We
could not stand the changed condition of things any longer, and the
Federals had entered into possession and completed the ruin. Very
likely some reminiscent darky told new-found friends, “All de house
niggers was hung on a tree last Christmas.” I have heard from
Northern lips even more astonishing stories of maltreated slaves
than a wholesale hanging.
Frequently before the holidays some of the negroes were
questioned as to what they would like to have, and the planter would
make notes and have the order filled in the city. That, I think, was the
custom at Whitehall plantation. I was visiting there on one occasion
when a woman told Judge Chinn she wanted a mourning veil. “A
mourning veil!” he replied. “I thought you were going to marry Tom
this Christmas?” “I is, marster, but you know Jim died last grinding,
and I ain’t never mourned none for Jim. I want to mourn some ’fore I
marries ag’in.” I did not remain to see, but I do not doubt she got the
mourning veil and had the melancholy satisfaction of wearing it
around the quarter lot a few days before she married Tom.
After the departure of our happy negroes, whose voices and
laughter could be heard long after the yard gate was closed and they
had vanished out of sight, we rushed around like wild to complete
preparations for the coming ball guests. They began to arrive in the
afternoon from down the coast and from the opposite side of the
river. Miles and miles some of them drove in carriages, with
champagne baskets, capital forerunners of the modern suit case,
tied on behind, and, like as not, a dusky maid perched on top of it;
poor thing, the carriage being full, she had to travel in that precarious
way, holding on for dear life. Those old-time turtle-back vehicles had
outside a small single seat for the coachman only. Parties came also
in skiffs, with their champagne baskets and maids. Long before time