80Th Conference On Glass Problems S K Sundaram Full Chapter

80th Conference on Glass Problems S.
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.
Limit of Liability/Disclaimer of Warranty

In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of
information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the
information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among
other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher
and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the
accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any
implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales
representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product
is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors
endorse the information or services the organization, website, or product may provide or recommendations it may make. This
work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies
contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers
should be aware that websites listed in this work may have changed or disappeared between when this work was written and
when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages,
including but not limited to special, incidental, consequential, or other damages.
Library of Congress Cataloging-in-Publication Data is available.
Names: Conference on Glass Problems (80th : 2018 : Columbus, Ohio) |

American Ceramic Society, issuing body.
Title: 80th conference on glass problems / The American Ceramic Society.
Description: Hoboken, NJ : Wiley-American Ceramic Society, 2021. | Series:
Ceramic transactions; volume 268 | Includes index.
Identifiers: LCCN 2020036165 (print) | LCCN 2020036166 (ebook) | ISBN
9781119744900 (cloth) | ISBN 9781119744917 (adobe pdf) | ISBN
9781119744924 (epub)
Subjects: LCSH: Glass─Congresses. | Glass manufacture─Congresses. |
Glass─Defects─Congresses. | Glass melting─Congresses.
Classification: LCC TP786 .C66 2021 (print) | LCC TP786 (ebook) | DDC
666/.1─dc23
LC record available at https://lccn.loc.gov/2020036165
LC ebook record available at https://lccn.loc.gov/2020036166
Cover design by Wiley
Printed in the United States of America.
10 9 8 7 6 5 4 3 2 1
Contents
Foreword vii
Preface ix
Acknowledgments xi
PLENARY SESSION
Future Of Oxy-Fuel Glass Melting: Oxygen Production, Energy Efficiency, Emissions 3

and CO2 Neutral Glass Melting
Hisashi Kobayashi
Peculiar Wear Behavior of Soda Lime Silicate Glass in Humid Air And Its Implications 13
Nisha Sheth and Seong H. Kim
MELTING AND COMBUSTION

A Glass Problem Solved 25
James Uhlik
Electric Power Adjustment in Glass Furnace with Varivolt Transformer 35

Marc Bernard
Synchronized Oxy-Fuel Boost Burners for Zero-Port Performance Optimization in Float 37

Glass Melting Furnaces
Michael J. Gallagher, Mark D. D’Agostini, and William J. Horan
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
80th Conference on Glass Problems v

BATCH, ENVIRONMENTAL, AND MODELING
Designing Furnace Feed Systems That Work 97

Roger A. Barnum and Scott A. Clement
Bag Filter And Catalyst (SCR) – Does This Fit Together? 109
Ruediger Margraf
Cullet – Another Step Towards Glass Sustainability 123

Christopher J. Hoyle, Kevin L. Fulkerson, and Brian J. Naveken
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
SENSORS AND CONTROL
Application of Advanced Sensors in the Glass Industry 149

Lieke de Cock
153
Paulus Schreuders
HTXTM – High Performance, High Temperature Thermocouple Wire 159

Declan Tucker, Jonathan Pearce, Ritesh Rawal, and Les Hutton
Not Just A Pretty Picture – In-Furnace Thermal Imaging 167

Mark Bennett, Neil Simpson, S. Fiona Turner, and John Naughton
80th Conference on Glass Problems vi

Foreword
The 80th Glass Problem Conference (GPC) was organized by the Kazuo Inamori
School of Engineering, The New York State College of Ceramics, Alfred
University, Alfred, NY 14802 and The Glass Manufacturing Industry Council
(GMIC), Westerville, OH 43082. The Program Director was S. K. Sundaram,
Inamori Professor of Materials Science and Engineering, Kazuo Inamori School of
Engineering, The New York State College of Ceramics, Alfred University, Alfred,
NY 14802. The Conference Director was Robert Weisenburger Lipetz, Executive
Director, Glass Manufacturing Industry Council (GMIC), Westerville, OH 43082.
The GPC Advisory Board (AB) included the Program Director, the Conference
Director, and several industry representatives. The Board assembled the technical
program. Donna Banks of the GMIC coordinated the events and provided support.
The Conference started with a half-day plenary session followed by technical
sessions. The themes and chairs of four technical sessions were as follows:
Melting and Combustion

Uyi Iyoha, Praxair, Inc., Peachtree City, GA, Jan Schep, Owens-Illinois, Inc.,
Perrysburg, OH, and Justin Wang, Guardian Industries, Auburn Hills, MI
Batch, Environmental, and Modeling

Phil Tucker, Johns Manville, Littleton, CO and Chris Tournour, Corning Inc.,
Corning, NY
Refractories
Larry McCloskey, Anchor Acquisition, LLC, Lancaster, OH and Eric Dirlam,
Ardagh Group, Muncie, IN
Sensors and Control

Adam Polycn, Vitro Architectural Glass, Cheswick, PA and Glenn Neff, Glass
Service USA, Inc., Stuart, FL
80th Conference on Glass Problems vii

Preface
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
80th Conference on Glass Problems ix

Acknowledgements
It is my great pleasure to acknowledge the dedicated service, advice, and team

spirit of the members of the GPC AB in planning this Conference, inviting key
speakers, reviewing technical presentations, chairing technical sessions, and
reviewing manuscripts for this publication:
Kenneth Bratton─Bucher Emhart Glass, Windsor, CT

Chris Bloom─Owens Corning, Granville, OH
Weijian Chen─Libbey Glass, Toledo, OH
Eric Drilam─Ardagh Glass, Muncie, IN
Uyi Iyoha─Praxair Inc., Peachtree City, GA
Robert Lipetz─Glass Manufacturing Industry Council, Westerville, OH
Larry McCloskey─Anchor Acquisition, LLC, Lancaster, OH
Glenn Neff─Glass Service USA, Inc., Stuart, FL
Adam Polcyn─Vitro Architectural Glass, Cheswick, PA
Jan Schep─Owens-Illinois, Inc., Perrysburg, OH
Christopher Tournour─Corning Incorporated, Corning, NY
Phillip Tucker─Johns Manville, Littleton, CO
James Uhlik─Toledo Engineering Co., Inc., Toledo, OH
Justin Wang─Guardian Industries, Auburn Hills, MI
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.
80th Conference on Glass Problems xi

PLENARY SESSION

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.
80th Conference on Glass Problems 4

ECONOMICS OF OXY-FUEL FIRING
The economics of oxy-fuel conversions depends mainly on the fuel savings achievable, fuel cost
and oxygen cost for operating cost comparison. The value of fuel savings needs to be greater than
the cost of oxygen to achieve a net saving in the operating cost when other oxy-fuel benefits such as
NOx reduction and production rate increase are not considered. Many air fired furnaces less than
100 tpd capacity were converted to oxy-fuel firing as the small waste heat recovery systems,
especially recuperators, are not very efficient and large fuel savings of 40-50% were achieved. For
large container and float glass furnaces with efficient regenerators about 10 to 15% fuel savings
have been achieved on the furnace campaign average. The economic drivers for the conversion
were capital cost savings of eliminating regenerators, especially for green field projects, furnace
capacity increase, and NOx reduction.
Figure 1 compares the fuel and oxygen costs for a generic 300 mtpd (metric ton per day)
container glass furnace with 50% cullet and no electric boosting at three different fuel costs
of 5, 10 and 15 dollars per MMBtu HHV (million Btu in higher heating value) and at a
constant oxygen cost of $50/ston (short ton). For the typical natural gas composition
represented as methane, one short ton of O2 is required to combust about 12 MMBtu
(HHV) of natural gas. The baseline specific fuel consumption for the air fired furnace with
regenerators is 4 MMBtu/ston and fuel and oxygen savings of 10% and 30% are assumed
for oxy-fuel firing without heat recovery (“Oxy”) and oxy-fuel firing with heat recovery
(“Oxy-HR”) respectively. At the low fuel cost of $5/MMBtu the baseline melting cost is
$20/ston of glass. With 10% fuel savings the specific fuel cost and the oxygen cost are $18
and $15 per ston of glass respectively. The combined cost of $33 per ston is higher than the
baseline cost. Even with 30% fuel savings achieved by using a flue gas heat recovery
system with oxy-fuel firing the combined cost of melting increases by $5.7/ston. At
$10/MMBtu the melting cost of Oxy-HR becomes slightly less than the air case and at a
high fuel cost of $15/MMBtu about $5/ston savings in melting cost is projected with Oxy-
HR. This example illustrates that achieving high fuel savings and reducing oxygen cost are
key economic factors to make oxy-fuel glass melting as the economic choice in the future.
Figure 1. fuel and oxygen costs for a generic 300 mtpd (metric ton per day) container with 50%
cullet and no electric boostingg

OXYGEN GENERATION TECHNOLOGY
Oxygen for glass furnace conversion applications is typically supplied by the on-site oxygen
generation system using a vacuum-pressure swing adsorption (VPSA) technology and backed up
with delivered liquid (LOX) in a storage tank. Most of oxy-fuel fired flat glass furnaces use
cryogenic air separation systems as the nitrogen co-produced from the cryogenic separation
system is used for the float tank. Cost of oxygen is highly site specific and involves many factors
including the plant size, electric power cost, VPSA equipment, manufacturing, construction,
engineering and operation. Electric power cost is typically the most important cost element and
constitutes about 1/3 to 2/3 of the overall oxygen cost depending on the power cost and the plant
size. Advancements in the air separation technology have reduced the power required to separate
air to make oxygen (kwh/ton O2) over the years. Figure 2 shows a historical trend and
expectation for future in a relative scale by assigning 100% for year 2019. Specific power
consumption was about 40% higher in 2000. The trend is likely to continue and further
improvements of 20-30% reduction are expected in the next 5 to 10 years.
Figure 2. Historical trend of specific electric power consumption to make VPSA oxygen
ENERGY EFFICIENCY IMPROVEMENTS FOR FUTURE

In a large oxy-fuel fired glass furnace without flue gas heat recovery about 25-30% of the heat generated
by combustion is lost in the sensible heat of flue gas and the balance is available in the furnace to melt
glass batch/cullet and to compensate for the wall heat losses. Figure 3 shows projected improvements on
specific fuel consumption for a 300 tpd oxy-fuel fired container glass furnace with 50% cullet and no
electric boost. Without any flue gas heat recovery the baseline fuel consumption is 3.7 GJ LHV/t (lower
heating value per metric ton or 3.52 MMBtu HHV/ston) for this state-of-the-art oxy-fuel fired furnace.
The electric energy consumption for oxygen production is assumed to be 250 kWh/t O2 generated and the
primary fuel required to produce the electricity was assumed to be 9 MJ/kWh at 40% electric power
generation efficiency. With the assumed primary fuel required to make oxygen overall fuel consumption
becomes 4.3 GJ/t. With a full flue gas heat recovery system to reduce flue gas temperature to 250 C
specific fuel consumption of 2.85 GJ/t is projected and the oxygen requirement is reduced proportionally.

Further improvements in fuel consumption requires new innovations to reduce wall heat losses and power
consumption for oxygen generation. If 50% reduction in wall heat losses and 30% reduction in the power
consumption for oxygen generation are achieved in the future, fuel consumption is projected to be 2.43
GJ/t without considering the fuel energy to make oxygen and 2.70 GJ/t with the assumed primary fuel for
oxygen generation added. If the electric power is generated by renewable sources such as wind and solar,
no fuel is required to produce oxygen. We may consider 2.5 GJ/t (2.38 MMBtu HHV/ston) as the goal for
future.
Various flue gas heat recovery systems for oxy-fuel fired furnaces have been developed and
installed in several commercial furnaces. They include batch-cullet preheaters, oxygen-fuel preheaters and
thermo-chemical regenerators. Fuel consumption as low as 3 GJ/t has been achieved in commercial
container glass furnaces12-18. Thus, good flue gas heat recovery is achievable today and the main
engineering challenge is to reduce the capital costs to improve the economics. The goal of reducing wall
heat losses by 50% may be more challenging with the current furnace design. Heavy insulation layers have
already been applied to furnace walls and crown and the total wall thickness of hot face refractory and the
insulation layers are more than 800 mm for some furnaces. Adding more insulation layers does not appear to
be practical. In the typical furnace design about a third of the total wall heat losses is from uninsulated wall
areas19, especially from the air-cooled metal lines. Innovative designs to reduce heat losses from uninsulated
wall areas such as metal lines, skew blocks and tuck stones are required. Another design approach to reduce
wall heat loss per ton of glass is to design a rapid melting system with very small total wall areas. The
maximum pull rate of the current container furnace design is about 4 tpd/m2, often assisted with electric
boost. If the specific pull rate is increase to about 10 tpd/m2, then the total wall area would be reduce to
about ½. In a pilot scale test of a rapid melter with oxy-fuel roof fired burners melting rates as high as 25
tpd/m2 have been demonstrated by operating the furnace temperature above 1700 C20. A pilot scale rotary
glass melting furnace with spray water cooled wall was demonstrated to withstand the very high furnace
temperature20. Submerged combustion melter with water cooled walls is another approach to rapid
melting21. Although water cooled walls have very high specific heat loss, small wall area may make the total
wall heat losses less than that of the current furnace design.
Figure 3. Potential improvements in specific fuel consumption for a 300 mtpd container furnace with
50% cullet.

CURRENT CO2 EMISSIONS FROM GLASS PRODUCTION
Figure 4 compares CO2 emissions from a 300 tpd container glass furnace with 50% cullet using 1,000 kW
of electric boost. Specific fuel consumptions assumed are 4.0 GJ/t for air firing, 3.8 GJ/t for oxy-fuel firing
(“OXY”), 3.1 GJ/t for oxy-fuel firing with heat recovery (“OXY-HR”), and 800 kWh/t for full electric
melting21. The CO2 emission from decomposition of soda ash and limestone/dolomite in the batch
contributes about 100 kg/t for each case. CO2 emission from natural gas combustion contributes about 210
kg/t for the air fired furnace, 200 kg/t for OXY and 160 kg/t for OXY-HR. CO2 emissions from electric
power generation range from zero for renewable power such as wind and solar to as high as 1.0 kg/kWh
from an old coal fired power plant. The average CO2 emission in the U.S. was about 0.5 kg/kWh in 2015
and about 0.4 kg/kWh in EU in 2011. In this analysis CO2 emission rate of 0.4 kg/kWh was assumed. 1,000
kW of electric boost contributes 32 kg/t of CO2 emission. The electric power required for oxygen generation
also contributes to the overall CO2 emission. At an assume electric power requirement of 250 kWh per ton
of oxygen generation additional CO2 emissions of 29 and 23 kg/t come from OXY and OXY-HR
respectively. For the full electric melter with 800 kWh/t of power consumption CO2 emission becomes 320
kg/t and the total CO2 emission 420 kg/t, which is higher than those from the natural gas fired furnaces by
about 15 to 35%.
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.
HYDROGEN AND BIOMASS DERIVED FUELS FOR REDUCING CO2 EMISSIONS

Hydrogen is considered as the fuel of future to eliminate CO2 emissions from combustion of
hydrocarbon fuels. Large scale demonstration projects have been announced in EU. For example,
existing natural gas distribution pipelines are proposed to be converted to distribute compressed
hydrogen in Leeds, UK24. On a volume basis hydrogen has a lower heating value (LHV) of 10.8

MJ/Nm3 @ 0 C (275 Btu/scf @ 60F) as compare with 35.8MJ/Nm3 (913 Btu/scf) for methane.
For the same heat input the volume flow rate of hydrogen becomes 3.2 times greater than that of
methane or typical natural gas. For oxy-fuel combustion hydrogen has an advantage over natural
gas as the oxygen requirement for combustion to generate the same amount of heat (LHV) is
reduced by about 17%. Hydrogen has a wider flammability range and a higher adiabatic flame
temperature as compared with methane. For glass furnace combustion applications hydrogen
flame is non-luminous, but the radiation heat transfer is excellent because the water vapor
concentration of combustion gases is near 100%, which enhances gas emissivity of oxy-
hydrogen flame and combustion gases. The conversion of the existing natural gas firing system
to hydrogen firing requires fuel piping and burner modifications, but no fundamental heat
transfer issues are expected with a proper oxy-hydrogen burner design. As experienced in the
conversion of air-fired furnace to oxygen firing, higher water vapor concentration in the furnace
atmosphere is expected to increase the water dissolution into glassmelt and to enhance the fining
reactions7-8. Reduction of batch fining agents is recommended23 to compensate for the enhanced
fining reactions and to control foam generation.
Table 1 compares hydrogen and biomass derived fuels with natural gas (as pure methane)
for glass melting applications25. For each fuel specific fuel and oxygen consumptions were
calculated for a typical 300 mtpd container furnace with 50% cullet and compared. The baseline
specific natural gas and oxygen consumptions were 3.87 GJ/t and 220 Nm3/t respectively.
Approximate fuel costs are also compared assuming $8/GJ LHV for natural gas which was a
typical cost in EU in 2018. Most of hydrogen used today is made from natural gas by the steam-
methane reforming (SMR) process. Typical hydrogen yield is about 70% based on the lower
heating value of hydrogen (LHV) as compared with the LHV of natural gas input. The cost of
hydrogen made from natural gas is roughly twice that of natural gas and assumed to be $15/GJ.
The cost of CO2 capture and sequestration (CCS) from the SMR process is not included. Both
hydrogen and oxygen can also be made by the electrolysis of water using renewable electric
power. According to a report by IEA26 the project cost of hydrogen in 2030 made by electrolysis
is about $2 to $4 per kg of H2 as compared with about $1.4 to $3 per kg of H2 made from the
SMR process with CCS. The large variations are mostly due to regional variations in the
renewable power cost and the natural gas cost. The specific fuel consumption with oxy-hydrogen
firing increases by about 1% as compared with oxy-methane firing and the specific oxygen
consumption is reduced by 15%. The main challenge of using hydrogen for glass melting is its
projected high cost. As shown in Figure 1 oxy-fuel firing is likely to be more favored than air
firing at $2 per kg of H2. ($2 per kg of H2 corresponds to $16.7 per GJ LHV or $17.6 per MMBtu
LHV).
Digester gas is produce from anaerobic digestion of organic materials such as pig and cow
manures and municipal waste water sludge. Typical composition is about 60% methane and 40%
carbon dioxides by volume. The existing natural gas firing system can be modified to fire
compressed digester gas. Both the specific fuel and oxygen consumptions are projected to increase
by 10%. Digester gas is typically priced less than natural gas and assumed to be $6 per GJ, but the
availability is limited to the proximity of local digesters. Liquid fuels produced from biomass
include ethanol and pyrolysis oil. Biomass based ethanol is produced from corn, sugar cane and
other crops in large quantities for gasoline blends. In the U.S. over 1 million bbl/day of corn
ethanol was produced in 2017. It is enough to produce 400 million tpy of container glass at 4 GJ/t,
which is more than three times the world container glass production. Fuel and oxygen requirements
are similar to the natural gas fired system. Although the existing natural gas firing system can be

modified to fire ethanol, high cost of fuel is the major hurdle at present. Pyrolysis oil is
produced from wood and other biomass by a thermal decomposition process optimized to
produce liquid products containing various oxygenated hydrocarbon compounds and
water. Heating value is about 16 MJ LHV/kg with 25% water content corresponding to
about 60% of LHV of ethanol (27.7 MJ/kg). When fired with oxygen about 20% more heat
input is required as compared with natural gas.
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

hydrogen and biomass derived fuels can be used with oxygen for glass melting. There are no major
technical barriers in using hydrogen for glass melting and oxygen-firing is likely to be favored over
air-firing due to expected high hydrogen costs and the reduced specific oxygen requirement with
hydrogen combustion.
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.

14. Goruney,T., Arzan,N., et.al., “Oxy-fuel Tableware Furnace with Novel Oxygen and Natural
Gas Preheating System”, 77th Conference on Glass Problems, Columbus, OH, November 8,
2016.
15. Barrickman, L., Leone, P., “Leone Industries: Experience with Cullet Filter /Preheater,” 67th
Conference on Glass Problems, Columbus, OH, October 31 to November 1, 2006.
16. Lubitz,G., Beutin, E.F., Leimkuhler, J., “Oxy-fuel fired furnaces in combination with batch
and cullet preheating”, Proceedings of Novem Energy Efficiency in Glass Industry. Ed.
Th. Nohlmans, L. Moonen, and R. Beerkens. Amsterdam. The Netherlands l8.-I9. May 2000
17. Schroeder, R.W., Kwamya, J.D., Leone, P., Barrickman, L., “Batch and Cullet Preheating
and Emissions Control on Oxy-Fuel Furnaces,” 60th Conference on Glass Problems,
University of Illinois at Urbana, October 19 to 20, 1999.
18. Snyder, W.J., Chamberland, R.P., Steigman, F.N., and Hoyle, C.J., “Economic Aspects of
Preheating Batch and Cullet for Oxy-Fuel Fired Furnaces,” 61st Conference on Glass
Problems, Ohio State University, October 17 to 18, 2000.
19. Lindig, M,. “Energy savings and furnace design”, 73rg Conference on Glass
20. Kobayashi,H., LeBlanc,J.L.,”Oxy-fuel Fired Rapid Glass Melting Furnace”, Third
International Conference in Advances in the Fusion and Processing of Glass, New Orleans,
LA, June 10-12 1992
21. Rue,D., and Brown, J.T., “Submerged Combustion Melting of Glass” International Journal of
Applied Glass Science 2 [4] 262–274 (2011)
22. Lindig, M,. “Glass melting technology addressing the environment trend - Electro”,
Glastechnische Tagung Der DGG, 29-31May, 2017.
23. U.S. Patent U.S. Patent 5,922,097 (July 13,1999), “Water Enhanced Fining Process- A
Method to Reduce Toxic Emissions from Glass Melting Furnaces”, H. Kobayashi and R.G.C.
Beerkens
24. “The H21 Leeds City Gate project report”, July 2016, https://futureofgas.uk/
documents/h21-leeds-citygate-project/
25. Kobayashi, H. and Chakravarti,S., “Oxy-Fuel Combustion toward CO2 Neutral Glass
Production” Glass Trend Seminar by CelSian, Wurzburg, April 2018
26. International Energy Agency, “The Future of Hydrogen – Seizing today’s opportunities”
Report prepared by IEA for the G20, Japan, June 2019

PECULIAR WEAR BEHAVIOR OF SODA LIME SILICATE GLASS IN HUMID AIR AND
ITS IMPLICATIONS
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.
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.
Figure 1. Architectural glazing product lifecycle.
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.

From a scientific perspective, this issue can be described as a tribological problem—the
study of the interaction of sliding surfaces including friction, lubrication, and wear. It becomes
critical to understand how thermal treatments, such as heat strengthening, influence the wear
behavior of the glass surface with various counter surface materials (organic and inorganic) in
various chemical environments. The purpose of this paper is to review the scientific work on this
topic and discuss future directions that the glass community ought to explore.
WEAR OF THE GLASS SURFACE

Wear depends on not only the intrinsic properties of glass, but also various extrinsic
conditions of the tribo-test system: contact pressure, speed, cycles, lubrication, chemical
environment, contact geometry, and generated wear debris as well as surface conditions of the
substrate and counter-surface. [2–5] Based on these conditions, moderate or severe wear types can
be produced. Under moderate wear conditions, surface damage is induced by a smooth counter
surface under a mechanical load that is much less than the indentation damage threshold. In this
way, the surface damage produced is entirely through the interfacial shear along a direction
tangential to the surface. It is assumed that minimal sub-surface damage occurs during moderate
wear in humid environments. However, this assumption needs to be tested and verified for each
tribosystem independently by, for example, using cross-sectional microscopy.
In contrast, severe wear type tends to occur at much larger contact pressures resulting in
both indentation-induced deformation and deformation due to shear force (i.e. scratch). As a result,
the loading rate also influences the wear behavior of the tribosystem. Various deformation and
fracture types occur below the sliding contact area due to the spatially-dependent magnitudes of
shear stress extending into the sub-surface.[6] As a result, surface abrasion, chipping, and surface
and sub-surface cracking are observed due to high normal and shear loads. An example of a severe
wear type is a scratch resulting from mechanical damage from a relatively sharp point of contact
between the counter-surface and substrate. The applied normal load, geometry and surface
condition of the counter-surface, and the surface condition of the substrate determine the width of
the wear track due to chipping that occurs along a scratch.
SODA LIME GLASS RESISTANCE TO MODERATE WEAR

Referred to as Reye Archard Khrushchov wear criterion, it is intuitively believed that
harder materials scratch softer materials. In dry environments (0% relative humidity (RH)) and
under moderate wear conditions, hard counter-surfaces, such as Si3N4, Al2O3, stainless steel, or
Pyrex borosilicate glass, can mechanically abrade soda lime silicate glass substrates. This
mechanical abrading leaves deep and rough wear tracks even if the counter-surface is smooth and
the applied contact load is lower than the indentation damage threshold of the substrate glass.[7]
In a humid environment under moderate wear conditions, the Reye Archard Khrushchov
relationship breaks down—moderate wear behavior becomes complex and hardness is no longer
the dominant factor determining wear outcomes. Glass substrates that have significantly higher
hardness, mechanical strength, and chemical durability are not necessarily more resistant to surface
damages during moderate wear test conditions in humid environments. As humidity increases, the
substrate wear resistance decreases for fused quartz,[8] alkali borosilicate (Schott BF33),[9]
barium boroaluminosilicate (Schott AF45),[9] sodium aluminosilicate,[10] and K+-exchanged
sodium aluminosilicate (Corning Gorilla-2),[10] against a borosilicate (Pyrex) counter-surface. In
contrast, as the humidity increases, soda lime glass shows increasing wear resistance. Near the
saturation vapor pressure, soda lime glass shows high wear resistance with minimal surface
damage regardless of the counter-surface materials, such as Si3N4, Al2O3, stainless-steel,[7] and

Pyrex borosilicate,[1,8–10] that are polished flat. It is interesting to note that regardless of the soda
lime glass manufacturer, composition or thickness, wear-resistance at high humidity under
moderate wear conditions was observed. Soda lime glasses tested include melt-derived green
container glass from Owens-Illinois, 0.7- and 1.0-mm thick float glass from AGC, 4 mm float
glass from PPG (now Vitro), 10 mm thick Guardian glass, and microscope slide glass from an
unknown manufacturer. Table 1 describes the moderate wear characteristics of various substrates
and counter-surface materials near saturation vapor pressure (90% RH).
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
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.
INFLUENCE OF SODA LIME GLASS PROCESSING CONDITIONS ON WEAR BEHAVIOR

While sodium may play a critical role in high wear resistance of soda lime glass under
moderate wear conditions in high humidity environments, this may not be exclusively due to a
compositional effect. A recent study compared the moderate wear behavior of the air and tin-side
of soda lime float glass that underwent various thermal treatments: annealing, heat strengthening,
and industrial thermal tempering.[1] Heat strengthening and thermal treatments create a
compressive stress casing thereby increasing the fracture toughness and mechanical strength of the
glass.
While heat strengthened and thermally tempered glasses had high indentation fracture
resistance, both treatments resulted in poor wear-resistance at 90% RH. Figure 2 summarizes the
wear behavior for air-side and tin-side SLS float glass as a function of relative humidity and
thermal treatment. In dry environments, moderate wear conditions resulted in mechanical abrasion
of the soda lime float glass surface regardless of treatment or air/tin side. At 40% RH, the wear
depth decreased by 50-70 times compared to wear depth at 0% RH and wear debris adhered onto

the borosilicate counter-surface. The side of the float glass did not appear to influence wear
outcomes.
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.
At 90% relative humidity (RH) conditions, significant differences in wear behavior

between thermal treatments and air/tin sides of the float glass were observed. The air-side annealed
soda lime float glass showed greater wear resistance compared to the tin-side. Thermally tempered
soda lime glasses, regardless of the side of the float glass side, became more damaged at the same
wear test condition. It is interesting to note that for the tin-side, the heat strengthened and thermal
tempered surfaces showed larger variances in wear depth. In these cases, it is speculated that the
initial wear debris generated influenced the wear behavior outcomes. Interestingly, the borosilicate
ball counter-surface was polished flat regardless of the substrate type, suggesting that the wear
debris generated was pushed out from the system. Statistically significant interactions between
humidity, float glass side (air vs. tin), and thermal treatments were observed at high humidity.
These results cast doubt on the hypothesis that highly mobile sodium ions are the main
determinant for high wear resistance under moderate wear conditions govern by shear force.[1]
Sodium has greater leachability in soda lime glasses with higher fictive temperatures due to the
larger free volume.[17] Since heat strengthened and thermally tempered glasses have a higher
fictive temperature, sodium modifiers should have a greater leachability. If the mobility of sodium
modifiers played a critical role for high wear resistance under moderate wear conditions at 90%
RH, then it would have been anticipated that heat strengthened and thermally tempered glasses
have higher wear resistance. Yet, the opposite is observed—annealed glasses have the highest wear
resistance while heat strengthened and thermally tempered are susceptible to shear-induced
damage.[1] Annealing allows for structural relaxation of the silicate network allowing for a
narrower bond parameter distribution while glasses with higher fictive temperatures will have a
broader bond parameter distribution.[18]
It can then be hypothesized that strained Si-O bonds are more susceptible to shear-induced
hydrolysis resulting in greater damage under moderate wear conditions at high humidity.[1,14]
While residual stress may influence wear behavior, no correlation was observed between the
magnitude of residual stress and wear behavior at 90% RH.[1] Nevertheless, it may be difficult to
separate the impact of residual stress and strained silicate network since residual stress generated
by thermal treatments is dependent on strained Si-O bonds.
These findings then suggest that in addition to glass composition, processing conditions of
the soda lime glass also affect the wear behavior. Furthermore, it also demonstrates that
strengthening strategies, such as thermal tempering, do not necessarily demonstrate high wear
resistance since mechanical behavior under one loading condition is not predictive of the material’s
response under another loading condition.[10] Like most mechanical testing, wear testing requires
a large sample size for a statistically significant conclusion.
WEAR BEHAVIOR OF SODA LIME GLASSES UNDER SEVERE WEAR CONDITIONS

Like moderate wear, severe wear behavior also depends on glass composition and material
processing conditions. The large normal load applied under severe wear conditions generates
lateral cracks that can propagate to the surface or intersect with other sub-surface cracks resulting
in chipping. Chipping due to severe wear increases the width of the wear track and volume of
removed material. In contrast, the width of the wear track under moderate wear conditions does
not appear to be influenced by the thermal processing treatment. This indicates that under moderate
wear conditions, lateral cracks were either not initiated, do not propagate to the surface or do not
intersect with possible sub-surface cracking.
In the presence of water, it was found that thermally tempered glass was more susceptible
to time-delayed chipping along the wear track as compared to annealed glass. Over time, chipping
widens the damaged region by 10 times making the scratch on tempered glass more visible.[2] In
comparison, annealed glasses experience minimal crack growth over time even in the presence of
water.[2]
While lateral cracks are generated in both annealed and thermally tempered glass, the
lateral crack propagation to the surface (resulting in chipping) appears to differ. This may be due
to sub-critical crack growth (SCCG) which is a mechanochemical event involving chemically-
assisted crack propagation under applied stresses.[19,20] Commonly referred to as stress
corrosion, SCCG depends on water transport to the flaw site and subsequent hydrolysis of the
silicate network. The transport of water and hydroxylation kinetics depend strongly on the network
structure. [21–23] Higher fictive temperature glasses, or glasses with more open structures, have
faster water transport through the glass surface.[17] It can then be hypothesized that thermally
tempered glass, or glasses that have high fictive temperatures and more open network structures,
are more susceptible to time-dependent chipping along the wear track in the presence of molecular
water.
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.

Scratches and wear can act as stress concentrators, ultimately compromising the mechanical
strength of the glass. It is critical to understand that mechanical behavior under a given loading
condition (e.g. indentation along the surface normal direction) does not necessarily predict the
mechanical behavior under another loading condition (i.e., interfacial shear stress).
While compositional and processing strategies have been developed to create stronger soda
lime glass products, these solutions may not provide higher resistance to damage under shear
stress. For example, while thermally and chemically tempered float glass has higher fracture
strength due to the compressive stress casing, they are highly susceptible to both moderate and
severe wear conditions. Wear behavior is complex and appears to depend on the interaction of
multiple variables including substrate composition, counter-surface composition, silicate network
structure, environmental conditions, and material processing. More work needs to be done to
identify material and environmental parameters that govern wear, understand the contribution and
interaction of these variables, and gain a mechanistic understanding of wear, especially in humid
environments. Eventually, this will bring us closer to damage-resistant glass.
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.

[10] N.D. Surdyka, C.G. Pantano, S.H. Kim, Environmental effects on initiation and
propagation of surface defects on silicate glasses: scratch and fracture toughness study,
Appl. Phys. A. 116 (2014) 519–528. doi:10.1007/s00339-014-8552-7.
[11] J. Luo, W. Grisales, M. Rabii, C.G. Pantano, S.H. Kim, Differences in surface failure
modes of soda lime silica glass under normal indentation versus tangential shear: A
comparative study on Na + /K + -ion exchange effects, J. Am. Ceram. Soc. 102 (2018)
1665–1676. doi:10.1111/jace.16019.
[12] H. He, J. Luo, L. Qian, C.G. Pantano, S.H. Kim, Thermal Poling of Soda-Lime Silica
Glass with Nonblocking Electrodes - Part 2: Effects on Mechanical and Mechanochemical
Properties, J. Am. Ceram. Soc. 99 (2016) 1231–1238. doi:10.1111/jace.14080.
[13] J. Luo, H. Huynh, C.G. Pantano, S.H. Kim, Hydrothermal reactions of soda lime silica
glass – Revealing subsurface damage and alteration of mechanical properties and
chemical structure of glass surfaces, J. Non. Cryst. Solids. 452 (2016) 93–101.
doi:10.1016/j.jnoncrysol.2016.08.021.
[14] S.H. Hahn, H. Liu, S.H. Kim, A.C.T. Van Duin, Atomistic Understanding of Surface
Wear Process of Sodium Silicate Glass in Dry versus Humid Environments, ACS Appl.
Mater. Interfaces. Submitted (2019) am-2019-181968.
[15] J. Luo, H. He, N.J. Podraza, L. Qian, C.G. Pantano, S.H. Kim, Thermal Poling of Soda-
Lime Silica Glass with Nonblocking Electrodes-Part 1: Effects of Sodium Ion Migration
and Water Ingress on Glass Surface Structure, J. Am. Ceram. Soc. 99 (2016) 1221–1230.
doi:10.1111/jace.14081.
[16] N. Sheth, D. Ngo, J. Banerjee, Y. Zhou, C.G. Pantano, S.H. Kim, Probing Hydrogen-
Bonding Interactions of Water Molecules Adsorbed on Silica, Sodium Calcium Silicate,
and Calcium Aluminosilicate Glasses, J. Phys. Chem. C. 122 (2018) 17792–17801.
doi:10.1021/acs.jpcc.8b04233.
[17] S. Amma, J. Luo, S.H. Kim, C.G. Pantano, Effects of fictive temperature on the leaching
of soda lime silica glass surfaces, J. Am. Ceram. Soc. 100 (2017) 1424–1431.
doi:10.1111/jace.14754.
[18] J. Luo, Y. Zhou, S.T. Milner, C.G. Pantano, S.H. Kim, Molecular dynamics study of
correlations between IR peak position and bond parameters of silica and silicate glasses:
Effects of temperature and stress, J. Am. Ceram. Soc. 101 (2018) 178–188.
doi:10.1111/jace.15187.
[19] R. Danzer, T. Lube, P. Supancic, R. Damani, Fracture of ceramics, Adv. Eng. Mater. 10
(2008) 275–298. doi:10.1002/adem.200700347.
[20] M. Ciccotti, Stress-corrosion mechanisms in silicate glasses, J. Phys. D. Appl. Phys. 42
(2009). doi:10.1088/0022-3727/42/21/214006.
[21] J. Yeon, A.C.T. van Duin, ReaxFF Molecular Dynamics Simulations of Hydroxylation
Kinetics for Amorphous and Nano-Silica Structure, and Its Relations with Atomic Strain
Energy, J. Phys. Chem. C. 120 (2016) 305–317. doi:10.1021/acs.jpcc.5b09784.
[22] E.A. Leed, C.G. Pantano, Computer modeling of water adsorption on silica and silicate
glass fracture surfaces, J. Non. Cryst. Solids. 325 (2003) 48–60. doi:10.1016/S0022-
3093(03)00361-2.
[23] E.A. Leed, J.O. Sofo, C.G. Pantano, Electronic structure calculations of physisorption and
chemisorption on oxide glass surfaces, Phys. Rev. B - Condens. Matter Mater. Phys. 72
(2005) 1–11. doi:10.1103/PhysRevB.72.155427.

MELTING AND
COMBUSTION
A GLASS PROBLEM SOLVED
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:
Fig. 1 – Defect level on October 1
TECO analysis of the above provided information showed, generally speaking, no

temperatures, flows or other instrumented and recorded data that was unusual or inconsistent with
good furnace operation. So plans were quickly made to perform a site visit and operations audit as
soon as possible. Upon arrival on Dec. 16th, the online glass ribbon defect analysis system showed
the following:
Fig. 2 – Defects on December 16

What is quickly of note is the periodic nature of increased levels of defects, also being
spread more across the ribbon. This is Clue 2:
Fig. 3 – December 16 defects over 24 hours
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:

Fig. 4 – Port fires
Fig. 5 – Fire comparison
Fig. 6 – Typical combustion space flows

At client’s furnace:
Fig. 7 – Client’s furnace flows representation
Fig. 8 – Client furnace fires
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.
WHY LEFT SIDE?

From extensive TECO modeling and practical experience, certain areas of particular float
glass furnaces are the first to come under scrutiny in such a situation as is described. Left side areas
to be carefully inspected are the glass level bowl and the left side of the waist coolers:

Fig. 9 – Typical float furnace glass flows
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.
Fig. 10 – Furnace glass level bowl
THE WAIST AREA

In the waist area it was noticed that there was some wear behind the left side cooler. The
plant had properly addressed this by placing an auxiliary cooler in the wear location. However, the
placed cooler was rather shallow, and a deeper cooler was quickly fabricated and substituted. It is
known that wear in this area can be severe at times, as these example cold inspection pictures
show:

Fig. 11 – Waist sidewall wear
Fig. 12 – Client’s left side waist area at cooler
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.

Fig. 13 – Furnace metal line; waist cooler build-up
Over the next days, due to float furnace glass flows taking up to three days to stabilize, the
defects slowly diminished:
Fig. 14 – Defects after fires adjustment, December 19-21
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!

Fig. 15 – Defect level at end of TECO’s visit
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.

ELECTRIC POWER ADJUSTMENT IN GLASS FURNACE WITH VARIVOLT
TRANSFORMER
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.
As an electromechanical device based on transformer technology the VARIVOLT main

advantages are : stepless variation of delivered power through stepless control of the secondary
voltage, high efficency, high power factor and long lifetime. The limitations of these advantages
however is a slower power adjustment time which may not be adequate on some special
applications requesting very frequent power adjustments or/and .very short reaction time.
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 on motor

Addition of Variable Reactor on the secondary side.
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.
SYNCHRONIZED OXY-FUEL BOOST BURNERS FOR ZERO-PORT PERFORMANCE
OPTIMIZATION IN FLOAT GLASS MELTING FURNACES
Michael J. Gallagher, Mark D. D’Agostini, and William J. Horan

Air Products and Chemicals, Inc.
Allentown, PA, USA
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
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.

Figure 2. CFD modeling results showing the effect of recirculation patterns near the boost burners
(far left side in the figure) that can cause the boost burner flames to deflect.
Air Products recently commercialized HRx burner was used as the boost burner during the trial of
the synchronized boosting system. The burner was designed to achieve a very high degree of
staging (95%) with high momentum and luminosity and is a good fit for oxy-boosting applications
[4]. The HRx burner has a high level of adjustability, including two individual valves that control
the direction and amount of staging oxygen. The HRx burner replaced the Cleanfire® HRi™
Advanced Boost burner in this project, therefore the results presented in this paper will show the
effect of the transition from the HRi Advanced Boost burner to the HRx burner, with synchronized
boosting. The impact of various furnace operational parameters will be discussed, including the
effect on energy consumption, local furnace temperatures, and glass defects.
THE CLEANFIRE® HRx™ BURNER
The Air Products Cleanfire® HRx™ burner was a key component in the synchronized boosting
system mainly because of its high degree of adjustability in controlling flame properties such as
momentum and luminosity. The HRx burner is a flat flame oxy-fuel burner designed for the glass
industry that has several features which include increased flame radiation for high fuel efficiency,
foam reduction capability for higher quality glass production, low NOx emissions, and optional
Instrumentation and sensors for remote performance monitoring. Figure 3 shows a photo of the
Cleanfire HRx burner from burner block hot face.

Figure 3. Photo of Cleanfire® HRx™ burner from the burner block hot face.
The HRx burner block has three ports; a central precombustor port where the fuel and primary
oxygen initiate combustion and the flame is rooted and stabilized, and upper and lower oxygen
staging ports. The HRx burner has unique oxygen staging capability where the staging oxygen can
be directionally controlled and diverted through either the upper or lower (or split between both)
staging ports that surround the primary precombustor. The modes of oxygen staging include Foam
Control mode, Melt mode, and Split mode. Such directional control of staging oxygen provides
several benefits including adjustment of flame length, momentum and luminosity. Figure 4 depicts
the various staging modes for the HRx burner. Oxygen staging also prevents NOx formation by
delaying mixing of oxygen and natural gas resulting in a lower initial flame temperature where
most NOx is produced. The HRx burner is also equipped with a secondary valve, called the primary
O2 valve, that controls the amount of primary and staging oxygen that is distributed through the
ports of the burner block. The primary oxygen valve essentially controls the amount of oxygen
staging used, although the valve works in reverse of a typical staging valve.
Split Mode. In Split mode, an equal amount of oxygen is directed to both the upper and lower
oxygen staging ports. This results in a shorter, brighter, and more stable flame. The Split mode can
be especially useful in turbulent locations of oxy-fuel furnaces (i.e. near the flue) and for oxy-
boosting applications.
Melt Mode. In Melt mode, oxygen is directed to the lower oxygen staging port of the burner block,
which is below the primary flame. The flame will develop a bright bottom surface due to thermal
radiation caused by the localized combustion of staging oxygen with the gases in the lower surface
of the flame. The high radiation produced in Melt mode is directed downward towards the glass
surface and has been shown to accelerate the melting process.
Foam Control Mode. In Foam Control mode, oxygen is directed to the upper oxygen staging port
of the burner block which is above the primary flame. The resulting flame appears to have a sooty

bottom edge that contains reducing gases made up of primarily carbon monoxide (in
concentrations of several percent). The reducing atmosphere created by the flame extends out
above the glass surface and acts to dissipate surface foam.
Primary Oxygen Valve. The HRx burner includes an additional valve that controls the amount of
primary oxygen that flows through the main burner nozzle with the natural gas. When the primary
oxygen valve is 100% open, approximately 75% of the total incoming oxygen to the burner is
passing through the primary nozzle. This condition enhances mixing between oxygen and natural
gas in the main nozzle and generally creates a short, stable flame with high momentum.
Additionally, with the primary valve 100% open, oxygen back pressure will decrease significantly,
and the burner tip temperature will decrease due to the additional oxygen flow around the burner
tip. In the opposite condition, where the primary valve is closed, there is approximately 5% of the
incoming oxygen passing through the main nozzle with the remainder (95%) distributed to the
upper, lower, or split staging ports, which is dependent on the selected staging mode, i.e. Foam
Control, Melt, or Split mode, respectively. When the primary valve is closed, it is possible to
achieve the maximum amount of oxygen staging and the flame length will become its maximum
for the current firing rate. In addition, NOx emissions will decrease, and flame luminosity will
increase, due to the burning of intermediate soot that is formed.
Figure 4. Various staging modes of the HRx burner

SYNCHRONIZED BOOSTING SYSTEM
The synchronized boosting system was designed to allow for automatic adjustment of the boost
burner flame properties, particularly length, luminosity and momentum in conjunction with each
regenerator reversal cycle in air-fired side port furnaces. The system ensures that each of the oxy-
fuel boost burners’ flame quality is always maximized and mitigates the negative effects of
turbulence due to the regenerator reversal cycles. Automatic control of the flame properties is
achieved through pneumatic actuation of the staging mode and primary oxygen valves on the HRx
boost burners. A PLC-based control system works in conjunction with the plant DCS to change
valve positions to preset locations with each regenerator reversal cycle. The optimum burner
settings are determined through visual observation of the boost burner flames and optical
temperature measurements of the charge end wall and breast walls near the boost burners. The
synchronized boosting system allows furnace engineers to customize each burner’s settings for
each firing direction and local atmosphere conditions. Using both the flexibility of the HRx boost
burners and a system to switch burner settings with the reversal, we can optimize the boosting
operation. The HRx Synchronized Boosting system offers the potential to enhance the benefits of
oxy-boosting, which include higher glass production and furnace bottom temperatures, lower glass
defects, and greater fuel savings.
RESULTS OF COMMERCIAL DEMONSTRATION
A trial of the synchronized boosting system was conducted at a 650 ton per day air-fired
regenerative side port float glass furnace. The furnace previously used HRi Advanced Boost
burners for a period of several years. In the trial, the HRi burners were replaced with the HRx
burner including the synchronized boosting system. Results comparing the performance of the HRi
Advanced Boost burners against the HRx Synchronized Boosting system will be discussed in this
section and include: energy consumption (MMBTU/ton), average bottom temperatures, average
crown temperature, as well as the impact on glass defects. It should be noted that most of the results
presented here were normalized to the HRi Advanced Boost burner data as a baseline case to
protect the confidentiality of the float glass plant’s data. The furnace pull rate remained nearly
constant throughout the trial period, but the cullet ratio changed slightly by about 2 percent. Table
1 shows the average cullet as a percentage of the batch material used for each case. The boost
burner firing rates remained constant prior to and throughout the trial at 8.25 MMBTU/hr per
burner. Also, the air-fired regenerative furnace in this trial was operating with the Glass Services
Expert System (ES III) control system with all data sets presented here.
Table 1. Average cullet as a percentage of charged material for each of the boost burner cases
evaluated in the trial.
Burner Type: HRi Advanced HRx Burner with
Boost Burner Synchronized Boost
Average Cullet (%): 18 20

Another random document with
no related content on Scribd:
“Levi Stucker ain’t no fool. He tole me and William he heard Ma’y
Ann and Marthy Ann whisperin’ and plannin’ in dere room nights till
he was sure dey was a hatchin’ mischief, ef dey hadn’t already
hatched more snakes dan dey could kiver, so I ’low’d I’d go and
ramshackle dat room o’ theirn, and onder de baid, Marse Jim, ’twixt
de sackin’ and de cotton baid, way onder de very middle, I found dis
bonnet what I bin lookin’ fur ever since grindin’ time. Now, Marse
Jim, dere ain’t no use in talkin’ to dem gals; dere ain’t no use in
readin’ no caterkism to ’em, nor in Miss Liza telling no more tales to
’em ’bout dat liar Anifera, or sum sich name. No use in whippin’ ’em,
nudder. If I’se whipped dem two niggers once fur not lookin’ fur dis
bonnet when I sont ’em to, I’se whipped ’em forty times. Dat didn’t
make ’em find what they hid demselves, and it ain’t going to do ’em
no good now. Marse Jim, you jist got to skeer de very life outen ’em,
and send ’em to de canefields. Dey is rascals and rogues.”
“Well, Charlotte,” he responded, “put the bonnet on this side, out
of sight, and bring those children here. I’ll see what I can do.”
As Charlotte left he turned to his tender-hearted wife and told her,
“It is important those little negroes should have a lesson that would
be of some use. Charlotte is right on the subject of moral suasion as
far as those little imps are concerned, so don’t let your kindness and
sympathy interfere with my conduct of the case. Keep in the
background, and I will give them a lesson they will not soon forget.”
“I can’t imagine what could have induced those children to make
way with that bonnet,” said Miss Liza, meditatively, as she looked at
the crumpled wreck on the floor.
“Perhaps mischief, perhaps accident. The thing is to make them
acknowledge the theft. Entrenched as they are behind a whole
barricade of lies and deceit, the thing is to make them capitulate,”
replied the husband.
“Cum right in; don’t be modest now. Marse Jim sont fur you,” was
heard in Charlotte’s bantering tone, as she appeared in the doorway,
half-leading, half-dragging the reluctant culprits, who already began
to sniff a coming battle. With some difficulty she marshaled them
before the master and stood close at hand ready to offer moral
support if the court of inquiry gave any signs of weakening, or to cut
off retreat on the part of the little darkies if they became too alarmed
to “stand fire.”
“Well, Mary and Martha, where have you been?” inquired Marse
Jim, in his blandest and most conciliatory tone.
“Down in de orchard lookin’ for aigs fur Marm Charlotte.” “And we
was findin’ some when she hollowed fur us to cum to de house.” “De
Dominicker hen got nest in de haige.” “She’s settin’, too.”
“Hold on, hold on, don’t both of you talk at once. I didn’t ask about
the hen’s nest. Have you been all over the orchard in the hot sun?”
“Yes, sir.” “Yes, sir, we goes anywhar fur Marm Charlotte.” “She
sont us.” “Yes, sir, she sont us fur aigs.” “An’ we was findin’ sum too.”
“Dat Dominicker hen——”
With uplifted restraining hand he said: “Hush, don’t both talk at
once. Let me talk some. Did you go away down there without your
bonnet?”
“We ain’t got no bonnet.” “Me and Ma’y Ann don’t wear bonnets,
Marse Jim.”
“Yes, you have a bonnet. Isn’t this your bonnet?” the master said,
in his quiet, inquiring tone, holding up before their bulging eyes the
dilapidated wreck that they had not dared look at in all the months
they had buried it out of sight. Ma’y Ann steadfastly turned her face
away from the ghost. She bit her lips, but uttered not a word.
“No, Marse Jim—I—I—er, Marse Jim, I feel sick, sick,” stammered
Marthy, as she trembled so she almost fell.
“Sick! Give me your hand.” She quickly recovered, and clasped
the tawny paws behind her back. “Give me your hand; let me feel
your pulse.” Reluctantly she proffered the hand. “There, now,” he
said, letting the limp little hand fall to her side. “You feel chilly, don’t
you? Go sit down on that step.” Marthy sidled slowly away, tears
welling her eyes and her whole frame shaken with suppressed sobs.
“Stop dat cryin’; nobody ain’t doin’ nuthin’ to you; stop dat
foolishness and listen to what Marse Jim is a sayin’ to you two
onreasonable rapscallions,” said Charlotte, in a severe tone. She
held Mary Ann (who was making ready to fly at the first opportunity)
by the back of her neckband.
“Let Martha alone, Charlotte, she is weakening; we’ll talk about the
bonnet to Mary Ann, she knows.”
“No, Marse Jim, I ’clar I never see dat bonnet in all my life; I ’clar I
never did. I ’clar——”
“Hush,” said the master in a stern voice, “let me ask a question or
two, and only answer what I ask.”
“Tell de truth, too,” ejaculated Charlotte, “onless you want de
debbil to kotch you.”
“Give me your hand.” The child clutched at her cotton skirt with
both hands. He reached out, quietly and forcibly took one skinny little
black paw in his firm grasp. Drawing the shrinking, reluctant child
toward him, he fixed his eyes upon her averted face. “Now look me
right in the eye; everybody does that to people who are talking to
them; look me in the eye. What made you hide that bonnet? Look at
me when I am talking to you.”
“I didn’t neber see dat bonnet b’fore. I ’clar——”
“Stop, look at me; don’t look at Martha, she’s better.” The child’s
eyes dropped. “Don’t look at the floor, look me in the eye.”
“Marse Jim, slap her; make her look at you.”
“Be quiet, Charlotte; she’s going to tell, I want to help her,” replied
the imperturbable inquisitor in his blandest tones. Still holding the
reluctant hand and drawing the figure more closely to him, he said,
“You say you never saw this bonnet? How came it in your bed?”
There was a long pause. The little negro at last gathered herself
up, and, with a gleam of inspiration, exclaimed: “Marse Jim, de rats
put it dar—de rats runs all over dat floor nights. Me and Marthy Ann
jist hears ’em jist toting things all around. Rats put it dar, Marse Jim,
big rats.”
“Dat’s a lie,” said Charlotte, positively. “Nary rat on dat floor. Marse
Jim, you jist foolin’ way your time on dese niggers.”
The baffled master turned toward the crouching figure on the
steps. She was still trembling, her face buried in her hands. He saw
she was ready to confess, but he was determined Mary Ann should
acknowledge also.
“Have you a mammy, Mary Ann?” he inquired.
“No, Marse Jim; I ain’t got no mammy; I ain’t never had no
mammy, and my daddy, he’s daid, and I ain’t——”
“Hush, I didn’t ask all that. If you haven’t a mammy there’s no one
to care if you die. I am sure I don’t want little girls round the house
that steal and lie. Nobody else would have you; nobody would buy
you, and I can’t keep you here. It’s come to a pretty pass when a
lady can’t lay her bonnet on the bed without you two little imps taking
it and hiding it for months, and lying about it right straight along. You
have no mammy to cry for you, and I don’t want you, and Miss Liza
don’t want you. What can be done with you?”
Martha sobbed, on the veranda step, and Mary looked defiant, but
no response came to that repeated inquiry. After a pause, Mary Ann
bridled up; the matter in question seemed to be taking a broader
range; the bonnet seemed to be merging in generalities, and might in
time sink into the other question of what can be done with them.
Martha’s courage also revived, so she could respond to the inquiry of
her parentage.
“I ain’t neber had no daddy, and my mammy she’s married to long
Phil now.”
The planter shifted his legs, looked abroad in a meditative way,
then turned to the charge.
“Well, now, you girls want to tell us all you know about this,”
holding up again before them the battered brim and crushed poppies
and long, dingy ribbons. Martha buried her face again, and Mary was
suddenly interested in the gambols of a squirrel in the pecan tree.
Neither culprit would look at the evidence of their guilt. “What will
become of you? I can’t keep you and nobody will buy a rogue;
nobody wants you.”
“My mammy wants me, Marse Jim,” whimpered the scared
Martha.
“No, your mother is Nancy, isn’t she? She’s a good woman and
don’t want a rogue and a liar tied to her all her days.” Another long
pause. “Come here, Martha, both of you stand by Charlotte and hold
her hands. I will give you one more chance. Which—one—of—you—
stole—that bonnet? Did both of you do it together? Who hid it? What
made you do it?” There was a pause between the questions, not one
word of response. Martha’s tears dropped on her little naked foot,
while Mary Ann looked vacantly at the nimble squirrel in apparent
indifference, not a muscle of her face giving any evidence of
emotion.
“Marse Jim,” said Charlotte, whose impatience increased as she
saw signs of action on the part of the inquisitor. “Marse Jim, what
you gwine to do? It’s no use er whippin’ dese gals; dere hides is like
cowhide and whippin’ ain’t no good noways fur liars. Killin’ is good
for such.”
The planter rose from his chair, straightened his tired limbs and
kicked the bonnet out of his way. “Bring them along, Charlotte. I’ll
see what I can do.”
Charlotte, with a firm grasp of each child, followed the tall leader,
who, as he turned into the hall, tossed a nod and a significant wink to
his wife. She obediently rose and followed. In all the interview the
mistress had remained a passive but interested spectator, feeling
sure that at a critical moment a signal from her husband would afford
her an opportunity to intervene. The master led his followers straight
to the well-house, under whose vine-clad arbor reposed the dripping
bucket, attached by a windlass to an endless chain.
“I think it best to drown them,” he quietly remarked. The little group
filled the arbor. William and Billy, the gardener; Delia, the laundress;
Lucy, the maid; Sawny, the “woodpile boy” and Oliver, who “went wid
de buggy,” attracted by the spectacle, gathered around the outskirts.
The story of the finding of the long lost bonnet had spread over the
yard and premises; fragments had even wafted to “the quarters,”
with the mysterious rapidity and certainty that always attended a
household event in the old plantation days.
“Mary Ann first,” said the master, as catching her suddenly and
firmly by the neckband of her dress and imprisoning her struggling
legs by wrapping her skirts tightly around them, he held her over the
well-hole, head a little down. The struggles and writhings of the child
were of no avail in the grasp of the strong man. “I want you to tell the
truth and promise never to tell another lie before I drop you down this
well.” The child squirmed and screamed in the relentless clutch,
swearing entire ignorance of the whole matter. Charlotte felt she
must pile on the agony, so she saw “de debbil down dar wid his
pitchfork, ready to ketch her.” That vision was too much for the now
thoroughly alarmed little darky.
“I tuck it, Marse Jim, I tuck it,” she screamed.
“Will you ever steal again?” still holding her over the well, where in
her own little reflection in the placid water she was convinced to her
dying day she had seen “de debbil.”
“Neber, neber, ’fore God, neber agin.”
“Never tell another lie if I let you off?”
“Neber, Marse Jim; neber’s long as I lib. Please the Lord and Miss
Liza, I’ll be a good little nigger; neber lie agin if you’ll lemme off dis
time.”
While that harrowing scene was being enacted with the most
determined and refractory of the little witches, and the spectators on
the outskirts were convulsed with laughter—every one of them had
at one time or another been suspected of the theft—Martha, the
tearful, was on her knees, holding despairingly to Miss Liza’s skirts
and imploring her “Jist to save me dis time, I’ll be good, I’ll neber tell
anoder lie. I’se got a mammy dat will cry fur me, and I don’t want ter
die. Oh! save me frum de debbil,” she screamed, when Charlotte’s
voice proclaimed him at the bottom of the well. “Don’t let de debbil
have your good little nigger.”
Confessions and promises being obtained, Mary Ann was placed
upon her feet. Four little black legs flew down the backyard; two little
guinea-blue skirts flipped over the cowyard fence and two little dusky
spots vanished in the distance. William called after them to “clip it
’fore de debbil gits outen dat well.” Charlotte held her sides with
outbursts of laughter that had been held in painful restraint.
“De debbil done skeer ’em more en Marse Jim,” Sawny remarked,
as he shambled back to the woodpile.
“I think, my dear,” said the planter, linking his arm into that of his
wife and returning to the library with her, “I think those children had a
lesson that may last them all their lives. They had to be scared into a
confession.”
“I hated to see them badgered,” she replied. “I dropped a few tears
over Martha myself—perhaps,” with a smile, “she thought I was
scared too.”
Charlotte came in and picked up the wreck. “Miss Liza, I’se goin’
to take dis bonnet, jist as it is, all tousled up and mashed and I’m
gwine to make Ma’y Ann war it one day and Marthy Ann de next
clean till dey gits sick o’ bonnets; dey shall war it till de chillen come
home Sat-day. I ’spose dere’ll be sum laffin’ done when de chillen
sees Ma’y Ann wid dat bonnet tied on her haid.”
Another winter had come and gone, and June was again filling the
old plantation with its intoxicating odors and delicious melody. The
little room on the back porch was darkened by a heavy curtain at the
only window. A table drawn up by the rough wooden bed, made gay
by a patchwork quilt, held a few medicine bottles, a cup and spoon;
also a tumbler of pink and white roses. The quiet mistress moved
about noiselessly, occasionally putting her cool hand upon the brow
of the little sick negro, or gently stroking the thin, black fingers that
lay listlessly upon the bright coverlet.
“Miss Liza, whar Ma’y Ann?” The lady turned her face from the
questioner. After a moment’s hesitation she replied, cheerfully:
“She’s all right, Martha.”
“Miss Liza, whar is she? Whar Ma’y Ann?”
“She’s down by the quarters now,” was the unsatisfactory
response. The weary patient closed her eyes for a few moments, but
it was evident that with the first consciousness, following a severe
illness, the child’s thoughts turned to her old companion.
“She ain’t bin here sence I was tuk sick.” After a pause, “I want ter
talk to Ma’y Ann ’bout sumthin’.”
“Tell me,” said the mistress, soothingly, “what it was you wanted to
see Mary for.”
Both the little negroes had been ill of scarlet fever. The children of
the household had not been allowed for weeks to come home for
their Saturday holidays. Martha fell ill first, and Mary was removed
into the room formerly occupied by Levi Stucker, where she soon fell
a victim to the disease. The mistress and Charlotte only were
allowed to minister to the invalids. Mary, the robust one of the two,
the more mischievous, the one apparently better equipped for a
struggle with disease, succumbed, after a few days of delirium. The
busy hands were stilled, the flying feet arrested, the voluble tongue
silenced, at the touch of the Angel of Death. The little body was
carried past the “quarters” and beyond, to the negroes’ “burying
ground,” where it lay in peaceful shadows of the trees the romping
children loved so well. Martha lingered long on the mysterious
border, fitfully fighting an apparently hopeless battle, the more
tenderly and faithfully nursed by Mammy Charlotte, as the warm-
hearted, childless woman realized the frail tenure of life held by the
little negro whom she had ruled in varying moods of sternness and
tenderness, untempered with judgment. With the fretful peevishness
of convalescence, the sick child whined repeated desires to know
“Whar Ma’y Ann?”
“What is it you want to tell Mary Ann to-day, when she is not here?
Can’t you tell me?” said the patient watcher.
“I jist want ter see her; I’se gwine ter tell you ’bout dat bonnet, Miss
Liza, and she ain’t here, and I mout die; sometimes folkses dies of
broke laigs, and my laigs is broke. I want Ma’y Ann ter know I ain’t
goin’ outen dis world wid dat bonnet on my soul.”
The mistress drew closer to the bedside, stroked and patted the
attenuated hand in a soothing way to quiet and compose the restless
invalid.
“Maybe it’s jist as good Ma’y Ann ain’t here, Miss Liza. I kin tell de
tale better’n when she is here to jine in.” After a pause, apparently to
marshal her thoughts more clearly, the child proceeded: “Dat time
Miss Ellen cum here, she tuk outen her trunk a red bonnet, and she
sed she had two on ’em jist alike, dat her chillen had wore out, and
she fotched ’em fur me and Ma’y Ann. I was in dar and seed de
bonnet, and you tuk hit, don’t you ’member, Miss Liza? You tuk hit
and sed no, Ma’y Ann and me had no use fur bonnets, and you
know’d two pore little white gals at your church dat didn’t have none,
and you was goin’ ter give ’em to dem. I went out and tole Ma’y Ann
all ’bout hit, and she ’low’d if we had bonnets we cud go to church
too. Well, we talked tergedder ’bout dose bonnets, and we plan we’d
take ’em ennyhow, fust time we seed ’em. Well, one night Ma’y Ann
runned right in here, in dat very door. I was in here den. I shet de
door and stood against it, and onder her apern she had de bonnet.
She didn’t find only one, but she grabbed dat. I tole her dat was the
bery one Miss Ellen took outen her trunk, and me and Ma’y Ann, we
tried it on our haids, ’fore dat bery piece o’ lookin’ glass stickin’ on de
wall dere, and we ’greed ter watch till we kotch de udder one, so we
hid it in dat trunk dar, behind you, Miss Liza, and ev’ry day we tried
hit on. I want ter tell you all ’bout hit ’fore Ma’y Ann gits back frum de
quarters. I dun know how long we kep’ hit in dat trunk, ontil one day
dere was a awful fuss, eberybody skeered up, lookin’ fur your
bonnet, dat was missin’. Me and Ma’y Ann was glad. We couldn’t
find one of our bonnets now your’n wuz gone, too.”
“Didn’t you know you had taken my bonnet?” said the mistress,
who was at last seeing through the mystery.
“Jist let me tell you de whole thing, Miss Liza. I bin layin’ here long
time thinkin’ de straight uv hit, so Ma’y Ann can’t bodder me when I
telled it to you. Ma’y Ann is dat sondacious she most make you
b’lieve anythin’. No, Miss Liza, we never thought dat till one day I
hear Miss Ellen say how nice dem red bonnets she brung did look on
de Quiggins gals at church. Den Marm Charlotte, she begun agin
’bout your bonnet bein’ missed and she searchin’ fur hit all de time,
and I hear her tell Sawny it wuz red and had black flowers on hit. Me
and Ma’y Ann took de bonnet outen de trunk dat night and dere wuz
de black flowers, jist like she sed, den we know’d you had give Miss
Ellen’s bonnets to the Quigginses, and Ma’y Ann had stole your’n.
We hefted dis baid and put de bonnet under hit, and, please Gord,
Miss Liza, I neber seed dat bonnet agin till Marse Jim shuck hit at us
dat day.”
“Why didn’t you come tell me what you had done, and why you
had done it, when you first found it out?”
“Miss Liza, we was afeerd. Marm Charlotte kep’ sayin’ whoever
had dat bonnet wud be hung, and de odder negroes talked back.
Thank de Lord, dey never seed hit, so Ma’y Ann and me didn’t dar
let on.”
“Didn’t you expect it would be found out some day?”
“Yes’em, I ’spec we did.”
XXX
WHEN LEXINGTON WON THE RACE
Every Kentucky woman loves a horse, and when Lexington was

entered in the great State stake in 1854 a crowd of the crême de la
crême of the Blue Grass country clamored to be present at the race.
The St. Charles Hotel, then in the hands of those genial hosts,
Messrs. Hall and Hildreth, was crowded for the event, beyond its
capacity, for when that Kentucky contingent of women, unheralded
and unexpected, swarmed into its broad parlor and halls, even the
servants’ quarters, so near the roof that the only light admitted was
skylight, were put into requisition. There was enough Blue Grass
blood in my family to compel a rush to the city, and we had a “sky
parlor,” right next to the one occupied by Gen. John H. Morgan
(simply “John” then. He won his spurs and title a decade or so later)
and his Kentucky wife. It took us “forever and a day” to mount the
stairs to our roosts, and we were so tired when we arrived that we
actually found the quarters acceptable.
All the Breckinridges, Wards, Flournoys, Johnsons and Hunts in
Kentucky were more or less financially interested in the superb racer.
Those who did not own one drop of Lexington’s blood, nor one hair
of his tail, “put their money” on the horse, and therewith a financial
interest was created. Every man, it seemed, in the place, that could
spare the time, wanted to see the great race. “Lee Count,” as a good
many Kentuckians call Le Comte, was the most prominent rival of
their boasted and beloved Lexington, and he showed mettle that
astonished even those blind partisans, and added zest to the
wagers. Ladies had never been in evidence at a horse race in
Louisiana. The bare idea was a shock to the Creole mind, that
dominated and controlled all the fashionable, indeed, all the
respectable, minds in New Orleans at that day. But the Kentucky
belles had minds of their own. Every mortal one of them felt a
personal interest, and a personal pride, and a personal ambition in
that Kentucky horse, though probably not ten out of the scores who
rushed to see him race had ever seen him before, and when he did
appear on the paddock he had to be pointed out to those
enthusiastic admirers.
What a host of dashing, high-bred, blue-blooded Kentucky women
swarmed the parlors, halls, rotunda of that, the finest hotel in all the
land! How they talked, in the soft, Southern accent, so peculiarly
their own! How they laughed! How they moved about, seemingly
knowing everybody they met. How they bet! Gloves, fans, money,
too, on their horse, when they found any one in all the crowd that
was not a “Lexington horse” man. Those bright women dominated
everything in their enthusiasm. I recall a host of them.
There was a lamentable scarcity of conveyances. Those Kentucky
people who had never felt the lack of vehicles and horses, had
apparently made small provision for travel to the course, so at the
moment of departure, when a large party was almost driven to
despair, Messrs. Hall and Hildreth ordered out the hotel stage, which
was one of the “nine-passenger” type. A nine-passenger coach, one
of the kind that was in vogue in the days of Pickwick, afforded seats
inside for nine persons, and could accommodate as many outside as
chose to pile on. The celerity with which those Kentucky women
filled that coach and the Kentucky men covered the top was a sight
worth seeing. No doubt when that stage rattled and bumped over the
cobblestones, en route to Metaire, many a cautious Creole mamma
made her innocent mam’zelles repair to the backyard while she
hastily closed the shutters. It was like a circus van, though no circus
had ever paraded those decorous streets.
Richard Tenbroeck (also a Kentuckian), who was associated in the
management of the course, was on hand to receive the merry crowd
from his own State, furnish it with grandstand seats and make it
welcome in every way. According to my recollection the Kentucky
women were the only females present, so very unfashionable it was
for ladies to go to races in the extreme South. There may have been
some demi-mondaines scattered here and there, in inconspicuous
places.
The race, the only one I had ever witnessed, was tremendously
exciting, and as the gallant horses swept round the last lap,
Lexington, ever so little, in the lead, the uproar became quite
deafening. One of the Johnson women, beautiful and enthusiastic,
sprang upon the bench and said to her equally excited escort, “Hold
me while I holler.” He threw his strong arms about her and steadied
her feet. “Now, holler”—and never did I hear the full compass of the
female voice before, nor since. Such excitement, as we all know, is
contagious, and it continued for days after the great achievement
that put dear old Lexington in the front rank, and filled the
pocketbooks of his owners, abettors and admirers.
Of course, this race was practically an all-day venture, and,
equally of course, people got hungry; and throats, most particularly
Kentucky throats, awfully dry. Mr. Tenbroeck provided liberally for
such a contingency, so a luncheon was served al fresco, with lots of
champagne, which latter did not dampen the ardor of those terribly
dry throats. We assembled in little groups around the viands, and
there were jokes and puns and stories that varied the monotony of
horse talk, that had dominated every other topic for days. In all the
circles there was fun and frolic. Kentuckians can be very hilarious.
The unique vehicle that carried our party back to the hotel rocked
and tumbled tipsily along. The sprightly crowd that departed in a
somewhat steady condition in the forenoon were sleepily tired when
they gained their sky parlors later in the day. A brief rest must have
revived them, for as we passed through the hall to a rather late
breakfast the following morning, trays of empty glasses and bottles,
flanked by freshly blacked boots and shoes, afforded evidence that
more refreshments had been absorbed later, and the parties had
returned to the Land of Nod.
XXXI
LOUISIANA STATE FAIR FIFTY YEARS AGO
It was in 1859 or 1860—I cannot fix the exact dates of many

events immediately prior to the war, for the rush of an overwhelming
waste carried dates, as everything else, away, but it was before the
war that several enterprising and advanced citizens of Louisiana
planned and organized and “resolved” themselves into a committee
to stimulate the indolent agricultural population to a more active life,
by inaugurating a series of State agricultural and mechanical
exhibitions, patterned as near as might be on the annual State and
county fairs of Kentucky, Missouri and other enterprising agricultural
States. Mr. John A. Dougherty, Major Sam Hart, George W. Ward,
John Perkins, my husband, Mr. James McHatton and his brother
Charles, Wm. A. Pike and others whose names escape me now,
secured from the United States government, through the joint efforts
of Hon. John Slidell and J. P. Benjamin, United States Senators from
Louisiana, and Thomas Green Davidson, Representative of the Sixth
District, temporary use of the then practically abandoned Barracks in
Baton Rouge, as being the most available site in the State for the
purpose of an experimental fair. Only a corporal’s guard had been
stationed there, to furl and unfurl the flag and to fire the evening gun,
as evidence that the grounds were United States property. In those
precincts and under those auspices, were held the first and the last
and only “Louisiana State Agricultural and Mechanical Fair.”
There came from New Orleans many exhibitors of farming
implements and products; from plantations, whose owners happened
to be “wide awake,” cattle, horses, sugar, molasses, and all such;
from the small farmer who occasionally read the papers, and thereby
kept in touch with the march of events, pigs and poultry; and from
the homes of enterprising women, all sorts of fancy work and
domestic articles. There were quite handsome prizes of silver, worth
competing for, offered by the managers. The parade ground was
ample to “show off” harness horses. An area was fenced off for
cattle, and side-show places assigned for pigs and poultry. The
Barrack buildings, two stories in height, surrounding the enclosure,
offered abundant room for the exhibit of farming utensils, harness,
etc. Rooms were appropriated for the luncheons and lounging places
of friends and guests.
The first two days were rather disappointing, so few people
understood just what was being attempted, but the number of the
exhibitors increased day by day, so that, before the final day, the
managers had reason to be enthusiastic at the success and
consequent promise for future State fairs.
Old Mr. Kleinpeter, of the high lands, entered a sow with a litter of
nine pigs, whereupon Granville Pierce “went one better” with a sow
and fourteen pigs. To be sure, the pigs varied in size, and people
made merry over the pig exhibit! From the “Cottage” plantation
(Cottage, by the way, was a tremendous big house) came a
hogshead of prize open kettle brown sugar. Immediately “Whitehall”
plantation saw it could beat that—and next day a hogshead of the
“Whitehall” brand was entered. It was thus the project expanded to
creditable dimensions. An enterprising lady who had won a silver
spoon prize at a similar fair in the West, entered a dressy bonnet,
made entirely of fine corn shucks; bows, flowers, feathers and all!
Whereupon, a smart miss from Grosse Tête sent three home-made
sun bonnets. The domestic exhibit thus resolved itself into a
competitive show. A Jew in town had met with indifferent success in
a sewing machine venture (sewing machines were in their immaturity
then, and not coveted by women who had domestics to order), till the
happy thought of a chance at the fair. Soon there was a sewing
machine on exhibition—a “Finkle and Lyon”—I don’t forget the make,
now happily out of existence, for in an evil moment, moved by the
Jew’s persuasive eloquence, I invested in a “Finkle and Lyon” which
I quickly found could only be made to “run” by copious drenchings of
olive oil, aided by the warm rays of the sun!
All the citizens of Baton Rouge entertained guests for the fair
week, the Harney House and other small hostelries being totally
inadequate. Several New Orleans merchants showed great interest
in the venture. Cuthbert Slocomb entered a fine exhibit of plows,
hoes and other farming tools, that were in his line of trade. So, also,
did the firm of Slark, Day and Stauffer; Henderson & Gaines sent of
their stock, as also did many others whose business brought them in
contact with the agricultural world. The cattle display was quite
surprisingly good, as were also the harness horses. The
inexperienced judges of such stock were often criticised for their
decisions, but the people were amiable and in a mood to enjoy
everything.
Such an outpouring from the “Cajin” settlements on the river, and
on Bayou Tête and Bayou Fordoche, and such other communities of
small pretensions, and still smaller achievements, never, I am sure,
had invaded Baton Rouge before. It was as “good as a play” to
watch their interest and enthusiasm, to see the greetings of families
and friends, who lived beyond the reach of a ramshackle voiture and
a worn-out horse. I do not recall the season of the year that immortal
fair occurred, but it must have been in late winter, for I remember a
small dish of radishes on my lunch table, such a rarity that Col.
Sparks ate every one. How one does recall, after a lapse of years,
such insignificant things! Some of the bon vivants, like Dr. French,
Mr. Bonnecage, and Dr. Harney, regretted that the enterprise was
not postponed till artichokes and river shrimp were in season.
It seems almost immediately after that I accompanied my delegate
husband to that ill-starred Democratic convention in Charleston, and
almost the next day that the Hon. J. P. Benjamin made his soul-
stirring speech in Congress, that magnificent burst of impassioned
oratory, whose prediction was never verified; almost the next day
that Hon. John Slidell returned to Louisiana a sad, despondent man,
and old Tom Green Davidson hobbled back to Baton Rouge on his
crutches, so full of bitterness and hate—almost the next day that the
flag that waved so gloriously over the parade ground where the
hopes and aspirations of those enterprising citizens took flight, was
hauled down.——
And after that—the Deluge!
XXXII
THE LAST CHRISTMAS
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

80Th Conference On Glass Problems S K Sundaram Full Chapter

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

80Th Conference On Glass Problems S K Sundaram Full Chapter

Uploaded by

Copyright:

Available Formats

80th Conference on Glass Problems S.

Limit of Liability/Disclaimer of Warranty

Library of Congress Cataloging-in-Publication Data is available.

Names: Conference on Glass Problems (80th : 2018 : Columbus, Ohio) |

Cover design by Wiley

Printed in the United States of America.

Future Of Oxy-Fuel Glass Melting: Oxygen Production, Energy Efficiency, Emissions 3

MELTING AND COMBUSTION

Electric Power Adjustment in Glass Furnace with Varivolt Transformer 35

Synchronized Oxy-Fuel Boost Burners for Zero-Port Performance Optimization in Float 37

80th Conference on Glass Problems v

Designing Furnace Feed Systems That Work 97

Cullet – Another Step Towards Glass Sustainability 123

SENSORS AND CONTROL

Application of Advanced Sensors in the Glass Industry 149

HTXTM – High Performance, High Temperature Thermocouple Wire 159

Not Just A Pretty Picture – In-Furnace Thermal Imaging 167

80th Conference on Glass Problems vi

Melting and Combustion

Batch, Environmental, and Modeling

Sensors and Control

80th Conference on Glass Problems vii

80th Conference on Glass Problems ix

It is my great pleasure to acknowledge the dedicated service, advice, and team

Kenneth Bratton─Bucher Emhart Glass, Windsor, CT

80th Conference on Glass Problems xi

80th Conference on Glass Problems 4

80th Conference on Glass Problems 5

ENERGY EFFICIENCY IMPROVEMENTS FOR FUTURE

80th Conference on Glass Problems 6

80th Conference on Glass Problems 7

HYDROGEN AND BIOMASS DERIVED FUELS FOR REDUCING CO2 EMISSIONS

80th Conference on Glass Problems 8

80th Conference on Glass Problems 9

80th Conference on Glass Problems 10

80th Conference on Glass Problems 11

80th Conference on Glass Problems 12

Figure 1. Architectural glazing product lifecycle.

80th Conference on Glass Problems 14

WEAR OF THE GLASS SURFACE

SODA LIME GLASS RESISTANCE TO MODERATE WEAR

80th Conference on Glass Problems 15

INFLUENCE OF SODA LIME GLASS PROCESSING CONDITIONS ON WEAR BEHAVIOR

80th Conference on Glass Problems 17

At 90% relative humidity (RH) conditions, significant differences in wear behavior

WEAR BEHAVIOR OF SODA LIME GLASSES UNDER SEVERE WEAR CONDITIONS

80th Conference on Glass Problems 19

80th Conference on Glass Problems 20

80th Conference on Glass Problems 21

TECO analysis of the above provided information showed, generally speaking, no

Fig. 2 – Defects on December 16

80th Conference on Glass Problems 26

Fig. 3 – December 16 defects over 24 hours

80th Conference on Glass Problems 27

Fig. 5 – Fire comparison

Fig. 6 – Typical combustion space flows

Fig. 7 – Client’s furnace flows representation

Fig. 8 – Client furnace fires

WHY LEFT SIDE?

80th Conference on Glass Problems 29

Fig. 10 – Furnace glass level bowl

THE WAIST AREA

80th Conference on Glass Problems 30