silicatos de sodio liquidos / sÓlidos de sodio n... · 2009-01-26 · contenido de oxido de sodio...

1 - 7

SILICATOS DE SODIO LIQUIDOS / SÓLIDOS

Los silicatos de sodio líquido son soluciones en agua manufacturadas a partir de proporciones variadas de oxido de sodio (Na2O) y oxido de silicio (SiO2). Dependiendo de su composición dan un amplio rango de propiedades físicas y químicas. Además de los silicatos de sodio líquidos, SIDESA tiene una amplia variedad de especialidades para satisfacer las necesidades de todos los usuarios que incluyen metasilicatos de sodio pentahidratado y anhidro, silicatos en polvo G y GD, hidro y cero geles, zeolitas, etc.,... FABRICACIÓN

Los silicatos de sodio se producen fundiendo a altas temperaturas, carbonato de sodio (Na2CO3) con

arena sílice especialmente seleccionada. El producto resultante es un cristal amorfo (VIDRIO PRIMARIO) que puede ser disuelto por procesos especiales para producir soluciones en gran variedad de formas. LIQUIDOS

Teóricamente, los óxidos de sodio y silicio pueden ser combinados en toda proporción. Sin embargo,

los productos líquidos actuales no deben exceder en una relación molar de SiO2 a Na2O de 3.5 a 1 o inferiores a 1.6 a 1. En la tabla 1 se muestran los principales productos líquidos:

TABLA 1 - SOLUCIONES DE SILICATO DE SODIO SIDESA

Nombre del

producto

Relación en peso

SiO2/Na2O

% Na2O

% SiO2

Densidad a 68°F (20°C)

*Be g/cm3

pH

Viscosidad en Centipoises

(20°C)

Características

STIXO 3.25 9.2 29.9 42.5 1.41 11.3 200 a 6,000 Líquido neutro viscoso N 3.22 8.9 28.7 41.0 1.38 11.3 Líquido neutro O 3.22 9.2 29.6 42.2 1.41 11.3 Más concentrado que N K 2.88 11.1 31.9 47.0 1.47 11.5 650 a 3,800 Líquido pesado

A-20 2.35 13.4 31.4 50.0 1.53 12.0 Líquido alcalino opalescente RU 2.40 13.8 33.2 52.0 1.55 12.0 1,150 a 4,000 Líquido denso

D-50 2.00 14.7 29.5 50.5 1.53 12.7 Líquido alcalino B-W 1.60 17.3 27.8 52.8 1.57 13.4 Líquido alcalino fluido

SÓLIDOS

Para aquellos clientes que desean comprar silicato de sodio en forma de pastillas sólidas con el

propósito de disolverlo para su uso, SIDESA ofrece los productos enlistados en la tabla 2. Ya que se necesita equipo y procedimientos especiales para disolver el silicato de sodio, es conveniente que se ponga en contacto con el departamento de Servicio Técnico y Ventas de SIDESA para los detalles.

TABLA 2: SILICATOS DE SODIO SÓLIDOS SIDESA

Nombre del producto

Relación en peso SiO2/Na2O

% Na2O % SiO2 % H2O Densidad g/cm3 Características del tamaño de partícula

SS 3.22 23.5 75.5 0 1.40 pastillas gruesas SS-C 2.00 33.0 66.0 0 1.40 pastillas gruesas

ASISTENCIA TÉCNICA Sidesa cuenta con un Departamento de Servicio Técnico el cual lo asistirá en caso de dudas o requerimientos adicionales EN APLICACIONES a los teléfonos: 5227-68-00 o sin costo 01-800-90-685-00. CERTIFICACIONES

El silicato de sodio cuenta con certificaciones internacionales y nacionales tales como: • NSF (National Sanitation Foundation) de Estados Unidos para toxicología en agua potable • CERTIMEX (Certificación Mexicana, S.C) antes IMTA

2 - 7

DENSIDAD

En la industria de silicatos la densidad ha sido expresada en términos de grados Baumé que pueden

ser convertidos a gravedad específica dividiendo 145 entre 145 menos los grados Baumé (tabla 3). Nuestras medidas son hechas con hidrómetros de intervalo angosto diseñados especialmente para medir a una temperatura estándar de 20°C. La temperatura tiene un efecto sobre la densidad de las soluciones de silicato, cuando la temperatura se incrementa la densidad decrece (tabla 4).

TABLA 3: DENSIDAD / GRAVEDAD ESPECIFICA ° Be EQUIVALENTES

TABLA 4: DENSIDADES DE SILICATOS A VARIAS TEMPERATURAS

Grados Baumé

Gravedad específica

Grados Baumé Gravedad específica

Temperatura °C

N Densidad °Be

O Densidad ° Be

35.0 1.3182 48.0 1.4948 10 41.5 42.6 36.0 1.3303 49.0 1.5104 21 41.0 42.2 37.0 1.3426 50.0 1.5268 32 40.6 41.7 38.0 1.3551 51.0 1.5426 38 40.3 41.4 39.0 1.3679 52.0 1.5591 49 39.9 41.0 40.0 1.3810 53.0 1.5761 60 39.4 40.5 41.0 1.3942 54.0 1.5934 42.0 1.4078 55.0 1.6111 43.0 1.4216 56.0 1.6292 44.0 1.4356 57.0 1.6477 45.0 1.4500 58.0 1.6667 46.0 1.4646 59.0 1.6860 47.0 1.4796 60.0 1.7059

VISCOSIDAD

La viscosidad es una propiedad física importante de las soluciones de silicato soluble. Desde el punto

estándar de aplicaciones la viscosidad de las soluciones de silicato de sodio es una función de la relación, concentración y temperatura. La comparación de viscosidades de soluciones de silicato de sodio de varias relaciones, muestra que las viscosidades de soluciones más silíceas (relación más alta) aumentan más rápidamente con un incremento en concentración que aquellas de silicatos más alcalinos. pH

El pH de las soluciones de silicato está íntimamente relacionado con la concentración y la relación de peso. El pH decrece cuando se incrementa el contenido de sílice. Análisis potenciométricos con ácidos muestran que el pH alto de las soluciones de silicato se mantiene hasta que el álcali es completamente neutralizado. La capacidad de amortiguamiento (la habilidad de una solución a resistir cambios en el pH) aumenta cuando se incrementan las proporciones de sílice soluble. Sin embargo, aún las soluciones de silicato diluido mantendrá un pH relativamente constante a pesar de agregar ácido. ANÁLISIS

La densidad de las soluciones de silicato de sodio, se mide generalmente con un hidrómetro. Ya que

las soluciones de silicato se expanden cuando se calientan, todas las medidas deben hacerse a 20°C. El hidrómetro debe ser ajustado y bajado lentamente en la solución de silicato. No deje caer el hidrómetro en el líquido. Cuando el hidrómetro se equilibra se toma una lectura lo más cercana en precisión a 0.1°Be. El contenido de oxido de sodio de los silicatos de sodio es determinado por un análisis de titulación volumétrica sobre una muestra con ácido clorhídrico estándar en la que se utiliza como indicador, púrpura o anaranjado de metilo, siendo en todo caso útil una mezcla de xileno-cyanole. El contenido de SiO2 es determinado por métodos gravimétricos. Se disuelve una muestra en agua, se acidifica con HCl y se deshidrata en un baño a vapor hasta que se seca. Se separa el precipitado, se calcina y se pesa como SiO2, aunque también existen técnicas volumétricas adecuadas.

3 - 7

MANEJO Entrega en pipas

Los silicatos líquidos se transportan en pipas (camiones tanque) en las que el extremo de conexión de la pipa debe ser roscado o soldado para prevenir le entrada de material extraño. La línea de llenado debe estar equipada con una válvula de cierre cerca del extremo soldado. La línea de llenado no debe estar a más de 1mt. del nivel del piso y el diámetro de salida de la manguera es de 2 pulgadas = 5cm. generalmente. Carros tanque

Existen tres métodos para descargar carros tanque; gravedad, bombeo y presión de aire,. Los preferidos son descarga por gravedad o por bombeo. Descarga por gravedad o bombeo.- Por norma utilice anteojos de seguridad y guantes de neopreno mientras descarga las soluciones de silicato de sodio. Revise siempre la válvula de salida y las conexiones de descarga para evitar fuga. Ventile la campana o casco abriendo el domo en la parte superior. En los carros tanques sin domo, abra la válvula de liberación de presión. Abra la válvula de salida y vacíe el carro, inspeccione el interior del carro. Descarga por presión de aire.- Este método no es recomendado. Si fuese necesario, es inminente un ensamble de control de aire para descargar con seguridad. Debe aplicarse muy cuidadosamente la presión de aire- PRECAUCION: NO DEBE EXCEDER LA PRESIÓN DE AIRE DE 25 psi DURANTE LA DESCARGA. El término de la descarga será señalado por un ligero soplido en la línea de descarga. Después de que se ha cerrado la presión de aire al tanque y se ha liberado la presión, inspeccione el interior del carro para asegurarse de que está vacío. Tambores

SIDESA ofrece silicatos de sodio en tambores no retornables de 200lts. Para información acerca de estos productos, póngase en contacto con nuestro departamento de ventas. Bombas, válvulas y tuberías

Las bombas de tipo centrífugo o rotatorias se han encontrado satisfactorias y dependiendo del trabajo que se vaya a realizar para manejar varios grados de silicatos, se requiere succión de anegación. Las bombas centrífugas requieren generalmente menos mantenimiento ya que el cuello empacado no es sujeto a una descarga de presión total. Las bombas rotatorias son utilizadas para servicios que requieren acción positiva y son de auto preparación. Toda construcción de acero al carbón es satisfactoria para las bombas de silicato con tuberías de hierro negro que cumplen con todos los requerimientos usuales para las líneas de diseño. Para la mayoría de las instalaciones de silicato, el diafragma de hule o válvulas de tipo compuerta son satisfactorios. Se utilizan generalmente válvulas de compuerta mientras que se evita el uso de las válvulas de globo. Almacenamiento

Las soluciones de silicato de sodio se evaporan lentamente cuando son expuestas al aire, así que los tanques deben estar cerrados pero ventilados. Los tanques deben ser construidos de acero al carbón, hierro o concreto y deben tener espacio suficiente para que un hombre tenga acceso para la inspección y mantenimiento. Si es posible los tanques deben estar localizados en un punto cercano a su lugar de consumo. El punto de congelación es cercano al del agua. En consecuencia, en climas fríos, los tanques de almacenaje deben mantenerse en edificios con calefacción. En soluciones de silicato que han sido congeladas y luego deshieladas, contendrán silicato altamente concentrado en el fondo y una solución relativamente diluida en la parte superior, debido al fenómeno físico de separación. Se puede volver a mezclar, agitándola vigorosamente a bombeándola hasta que la solución tenga todas las características del original. Las soluciones sobre 60 °Be no se separan o congelan pero se vuelven sólidas y frágiles. Del mismo modo, no se separan en porciones diluida y concentrada cuando se calientan a temperatura ambiente. Si se requiere calentamiento puede usarse cinta eléctrica o serpentines de vapor convencionales. Se prefiere el calentamiento exterior al interior ya que lo minimiza por zona y previene algunas variaciones en la concentración del silicato dentro del tanque. En cualquier momento se debe evitar la evaporación por calentamiento prolongado. Las soluciones de silicato pueden ser almacenadas por largos períodos de tiempo en tambores cerrados, recipientes de acero u otros recipientes hechos de material no reactivo. No deben utilizarse recipientes de aluminio, acero galvanizado o zinc ya que existe la posibilidad de formar gas de hidrógeno e incendiar o hacer explotar el recipiente. CALIDAD, SERVICIO Y CAPACIDAD DE PRODUCCIÓN.

4 - 7

Los usuarios actuales y los prospectos encuentran datos de valor en los boletines publicados en SIDESA así como por nuestra asociada PQ a solicitud individual. Ofrecemos también sugerencias que pueden ahorrar valiosísimo tiempo a investigadores y usuarios; el conocimiento técnico de PQ está basado en más de un siglo de experiencia en la manufactura de silicato y su uso. Para conveniencia de los usuarios de silicatos solubles en cantidades menores a la de carros tanques o pipas, los silicatos líquidos se comercializan también por distribuidores químicos en las principales ciudades de la República Mexicana. Pida el nombre del distribuidor más cercano a usted. Cuando necesite información sobre silicatos, muestras para evaluación o sugerencias para almacenaje y sistemas de distribución póngase en contacto con SIDESA. REACCIONES QUIMICAS DE LOS SILICATOS SIDESA Formulación SOL y GEL

Los silicatos de sodio reaccionan con compuestos ácidos. Cuando las soluciones de concentraciones

relativamente alta se acidifican, los aniones de silicato soluble se polimerizan hasta formar un “gel”. Cuando se acidifican sílices disueltos de concentraciones relativamente diluidas se pueden formar “soles” activados. El grado de polimerización de los aniones de las soluciones de silicato de sodio depende de la concentración de la solución, temperatura pH y otros factores. La gelación ocurre muy rápidamente al pH neutro. Puede ocurrir retrasos en los tiempos de gelación (soles inestables) en rangos de pH 8-10 y 2-5. La formación gel es generalmente muy rápida en el rango intermedio (5-8). Los soles de sílice coloidal pueden prepararse a partir de silicatos de sodio por medio de un intercambio de iones, diálisis y otros medios a sílice activada, que es utilizada en tratamiento de agua de desperdicio industrial o municipal. El sol de sílice coloidal puede usarse para el curtido de pieles, reforzamiento de polímeros sintéticos, terminado de telas y cubiertas. El gel de sílice se prepara neutralizando una solución de silicato con ácido mineral. El gel húmedo es triturado, desalinizado y secado para preparar desecantes absorbentes, agentes transmisores y bases de catalizadores. La neutralización del silicato con soluciones ácidas o gases, forma capas de gel de baja solubilidad pero que son de alguna manera frágiles y temporales por naturaleza. Reacciones de precipitación

Las soluciones de silicato de sodio reaccionan con cationes polivalentes disueltos para formar

precipitados. Dependiendo de las condiciones de reacción, como pH, concentración y temperatura resultarán silicatos de metal insoluble o sílice hidratada con óxidos de metal absorbidos o hidróxidos. Este tipo de reacciones puede usarse para formar pigmentos o compuestos que pueden utilizarse como extensores o selladores por medio de un intercambio de iones, catalizadores, absorbentes y otros productos. El cloruro de calcio reacciona instantáneamente con soluciones de silicato. La reacción es un mecanismo efectivo también para insolubilizar una capa o cubierta de silicato. Las superficies de cemento portland son endurecidas y se hacen menos porosas cuando se aplica una solución de silicato a pisos o paredes. Los aluminosilicatos son un mecanismo efectivo también para insolubilizar una capa o cubierta de silicato. Las superficies de cemento portland son endurecidas y se hacen menos porosas cuando se aplica una solución de silicato a pisos o paredes. Los aluminosilicatos de sodio se forman por reacciones entre compuestos de aluminio y silicato de sodio. Los productos resultantes pueden servir por medio de un intercambio de iones como suavizante de agua, zeolitas sintéticas o mallas moleculares. La extensión e índice de la reacción de silicatos con varias sales metálicas depende de la naturaleza de la sal y de su estructura molecular y física. Por ejemplo, los carbonatos de calcio mineral, tales como calcita muestran una interacción limitada con silicato soluble, mientras que el carbonato de calcio precipitado muestra una alta reactividad. Interacción con compuestos orgánicos

Relativamente pocos compuestos orgánicos son compatibles con soluciones de silicato soluble

concentrado. Solventes polares sensibles pueden causar una separación de fase o deshidratación. En mezclas con sustancias oleofílicas inmiscibles en agua, el silicato se separa en la fase acuosa, aunque en el caso de formulaciones de detergente líquido, este fenómeno puede superarse agregando un hidrótropo adecuado o una emulsión estabilizadora. Pocos compuestos como la glicerina, sorbitol de azúcar y etilenglicol con miscibles y algunas veces se usan como humectantes y ayudan a plastificar la película de silicato. Los agentes de ésteres orgánicos son utilizados para producir un retraso en el tiempo de gelación de las soluciones de silicato. La alcalinidad de la solución de silicato es consumida por la hidrólisis de estos ésteres por un período de tiempo extenso.

5 - 7

APLICACIONES COMUNES SÓLIDOS Y LIQUIDOS SIDESA

Los silicatos de sodio tienen muchas propiedades útiles que no comparten otras sales alcalinas. Esto, junto con el

hecho de que tienen bajo costo, da como resultado un amplio campo de uso en diferentes industrias. Son utilizados en la industria como adhesivos, detergentes, ingredientes en compuestos de limpieza, cementos, binders, capas protectoras y peculiares, ayuda coagulante, anticorrosivos, bases de catalizadores, defloculadores, insumos químicos, zeolitas, etc. Las diferentes propiedades y características funcionales de los silicatos solubles pueden ser utilizadas para resolver eficiente y económicamente muchos problemas que surgen en procesos industriales y químicos. Las investigaciones de PQ Corporation, base tecnológica de SIDESA han ampliado el conocimiento de las propiedades del producto y su utilización hasta incluir un amplio rango de aplicaciones. Aún más, continuamente se están estudiando nuevos usos. Si usted tiene un problema en el que el silicato pueda ayudar a resolver, le invitamos a consultarnos. Para una referencia rápida de los usos clasificados en las principales industrias consumidoras, a continuación se le muestra una explicación breve. DETERGENTES

Las soluciones de silicato de sodio de SIDESA han sido utilizadas como un ingrediente en la fabricación de detergentes en polvo vía torre de secado, por muchos años. Las soluciones de silicato son fácilmente agregadas a la pasta detergente y ayudan a controlar la viscosidad para la producción de un polvo detergente de la densidad deseada. El silicato de sodio actúa también como binder para dar el grado adecuado de “dureza” a la partícula esférica del detergente, sin perjudicar la solubilidad del polvo en agua. En las formulaciones de detergente el silicato de sodio tiene un número de propiedades vigorizantes que intensifican la actuación del sistema detergente. Estas propiedades son: Humectación.-Las moléculas del surfactante son agentes de reducción de la tensión superficial de los líquidos permitiendo que las superficies duras de los tejidos de fábrica se mojen apropiadamente durante el lavado. Emulsificación.- Es la dispersión de manchas grasosas en rocío fino que las mantienen suspendidas en la solución del lavado. Defloculacion.- Promueve la ruptura de manchas o partículas inorgánicas en partículas finas en la solución del lavado. Antiredeposición.- Previene que la mugre o la suciedad flote sin que se redeposite a las superficies limpias. Poder buffer.- Gracias a su alcalinidad tiene la habilidad para controlar el pH de la solución detergente en presencia de mugre ácida, amortiguando los cambios. Prevención de la corrosión.- o la protección de superficies de metal sensible a los efectos corrosivos de otros ingredientes del detergente. Poder secuestrante y suavizante.- Remueve los minerales y por lo tanto la dureza del agua de lavado, mediante un mecanismo de intercambio iónico en el que se generan compuestos que se eliminan durante el enjuague, principalmente de calcio, magnesio, fierro y manganeso.

En la formulación de detergente de silicatos tienen la habilidad de ayudar en los efectos sobre la superficie de las telas humectándolas en conjunto con los elementos orgánicos, ayudando a mantener las pequeñas suciedades o partículas grasosas en suspensión en el líquido de lavado, o ayudando a remover las manchas de las superficies. La alcalinidad de los silicatos de sodio los capacita para neutralizar las manchas ácidas y promover la emulsificación de grasas y aceites mientras dispersa o solubiliza proteínas. Los silicatos tienen una alta capacidad de amortiguamiento en relación a otras sales alcalinas que estabiliza el pH al nivel deseado en presencia de compuestos ácidos o en dilución. ADHESIVOS

Los silicatos más siliceos (relación 2.8 - 3.2) son muy útiles como adhesivos o binders, debido a un contenido más alto de sílice polimérica. Estos materiales pegan por la remoción de pequeñas cantidades de agua, los cuales los convierten de un líquido a un sólido. Las ventajas de los adhesivos de silicato de sodio incluyen un buen extendimiento y contacto, un buen pegado, un índice de ajuste controlable en rangos amplios y la formación de una capa rígida, fuerte, permanente de un sellado que es resistente a jalones, bichos (plagas), calor y moderadamente resistente al agua. Los silicatos para adhesivos son generalmente embarcados y listos para su uso, pero para aplicaciones especiales pueden ser modificados por ciertos aditivos tales como arcilla, caseína y otros materiales inorgánicos. Se utilizan para papel, madera, metal, hojas metálicas y otros materiales, excepto plásticos. CEMENTOS/BINDERS

Cuando los silicatos son combinados con ingredientes de cemento, reaccionan químicamente para formar masas con fuertes propiedades binders. Una gran variedad de cementos se hacen con silicatos, tanto en polvo como en solución. Los silicatos son ingredientes importantes en las especialidades refractarias autofraguantes y morteros químicamente resistentes. Las ventajas de los silicatos solubles como binders son: 1)Resistencia a la temperatura, 2)Resistencia a los ácidos, 3)Resistencia a disolventes después de su uso, 4)Facilidad de manejo, 5)Seguridad, 6)Bajo costo. PELICULAS Y RECUBRIMIENTOS.

Las películas secas de silicatos no se afectan por aceites, sebo y grasas minerales, o de otras clases. Cuando se aplican en producto de papel o madera son resistentes al fuego y plagas y a prueba de grasas. Se puede incrementar la resistencia al agua agregando óxidos de metal pesado, agentes insolubles de carbonato, polímeros inorgánicos o selladores minerales como la mica. Los silicatos forman capas protectoras en los metales para controlar la corrosión en las líneas de agua.

6 - 7

TRATAMIENTO DE AGUA

Cuando se utilizan los silicatos con alumbre, sales férricas u otros coagulantes en el tratamiento de agua cruda se incrementa la velocidad de formación de floculo, su tamaño, densidad y resistencia. El añadir silicato de sodio, puede ser un método sencillo y económico de controlar el hierro y/o manganeso excesivo, contenido en muchos sistemas de abastecimiento de agua. Muchos abastecimientos de agua natural, con presencia de hierro y/o manganeso, que fallan cediendo a los procedimientos regulares para removerlos, pueden desarrollar el silicato durante este tratamiento. FUNDICION

El silicato de sodio es un binder bien conocido para arenas de vaciado. Es compatible con el almidón y la dextrina utilizados en la preparación de materiales de fundición. Los silicatos SIDESA reemplazan exitosamente los aceites vegetales y resinas más costosos utilizados para formar ciertos tipos de moldes. Los compuestos de moldes. Los compuestos de moldes basados en silicatos dan una capa dura a temperaturas moderadas, porosidad apropiada para la liberación de gases calientes y buena colapsibilidad. Para métodos de producción rápida, la aplicación de silicatos de sodio en combinación con reacciones ácidas de gas se emplean en el proceso común de bióxido de carbono (CO2). También se pueden lograr moldes de fundición de autoendurecimiento con un binder hecho de la mezcla de cemento portland o ésteres orgánicos con silicato de sodio. BENEFICIO MINERAL.

Los silicatos solubles son utilizados en numerosos procesos para la concentración de minerales. El uso principal del silicato de sodio en la flotación mineral, es como depresor y dispersor de minerales silíceos que no se desean. En general sólo se necesitan pequeñas cantidades de silicato, similares a las concentraciones utilizadas en las operaciones de limpieza. Algunas veces la habilidad del silicato, para reaccionar o precipitar iones de metal pesado para formar una capa de iones de sodio en una superficie particular, es importante para preparar esa superficie o bien para una separación selectiva. El silicato de sodio también ayuda a prevenir la corrosión, reduciendo así el desgaste en los equipos de molienda. TEXTILES

Los silicatos de sodio tienen muchas aplicaciones en los talleres textiles. En las operaciones de blanqueado, la característica estabilizadora de silicato reduce el índice de descomposición del agente blanqueador. Este control previene el daño que ocurría en la fibra si se permite un cambio de pH. Los silicatos de sodio son aplicados en los pretratamientos de tela e hilo para remover cera, grasa y motas de algodón. La elección apropiada de la relación de silicato y sólidos produce una mejor limpieza y evita que se redeposite la suciedad. AGLOMERACION

Las características adhesivas naturales del silicato, junto con su habilidad de proporcionar una fijación rápida y un fuerte secado, así como la de suministrar una capa resistente al agua a bajo costo, han resultado en su utilización en las aplicaciones de pelletizado, briqueteado y sintetizado de minerales. Sirve también como ayuda en el proceso de nodulación y estiramiento. Para la aglomeración de materiales que son extraídos, procesados o recuperados en formas muy pequeñas, elimina el polvo y mejora la transferencia de calor en las operaciones de secado, calcinación y fundición. La eliminación de estos polvos puede reducir o eliminar la contaminación de aire y agua, así como las capas o anillos internos en un horno de secado y mejora las características de almacenaje y transportación. El pelletizado es un proceso en el cual el material pequeño se forma en bolas húmedas. El silicato combinado con las partículas finas ayuda a la formación del pellet e incrementa la fuerza de los gránulos formados en las etapas de secado y cocción. En los procesos de estiramiento, compresión y aglomerado, el silicato actúa como lubricante y/o binder, de manera que se mejoran las características de fluidez y propiedades físicas del producto. DILUYENTES AGREGADOS

Cuando se utiliza en la pasta de los cementos la solución de silicato de sodio deflocula el mineral de arcilla y los componentes de cal de pasta. Esto se logra mediante la combinación de efectos, que incluyen: absorción de la superficie del agregado; donación de iones de sodio (Na+); y precipitación de iones de calcio (Ca++) en la pasta. El efecto real de este proceso de defloculación es una reducción de la cantidad de agua necesitada para mantener la pasta en un rango de viscosidad bombeable. Se prefiere el silicato de sodio como defloculador en la preparación de soluciones estables de arcilla fusible para el vaciado en los poros de los moldes de yeso. En la arcilla fusible, la cual es fácilmente bombeable, la sílice actúa como un estabilizador para cualquier álcali presente. Ya que se necesita menos agua, la arcilla fusible es más densa y por lo tanto la pieza resultante más fuerte. Se necesita menos tiempo de cocción en el molde y hay un menor encogimiento conforme se secan las piezas moldeadas. En el proceso de arcilla cruda, la acción depresora del silicato es utilizada para separar impurezas como cuarzo, feldespato, mica, óxido de hierro, etc. INYECCIÓN QUÍMICA

Para la inyección química o solidificación del subsuelo, puede ser utilizado el silicato de sodio en las siguientes situaciones: (1) donde las formaciones del subsuelo tienen fuerza insuficiente para soportar la carga requerida, tales como bajo las paredes o cimientos, (2) donde los poros del subsuelo, permeables al agua, permiten la inundación de minas, pozos y túneles (3) para prevenir la pérdida del agua en presas, (4) para sellar poros en los trabajos de concreto o ladrillo como se pueden encontrar en desagües o construcciones subterráneas. En el “Proceso de Joosten” se inyectan por separado silicato de sodio y cloruro de calcio en el subsuelo. Se combinan para formar un gel insoluble que fragua y sella los poros y suelos con baja resistencia.

7 - 7

TRATAMIENTO DE CONCRETO

El silicato de sodio ofrece dos aplicaciones diferentes para aumentar la durabilidad del concreto. Se puede aplicar una solución de silicato de sodio como agente curante a la superficie de la capa fresca de concreto después de que el área ha sido cubierta y mantenida húmeda durante 24 horas. La aplicación de silicato cierra los poros de la superficie sellándolos mientras están húmedos. Para tratar el concreto, después de que está completamente seco o endurecido, se aplica el silicato hasta penetrar el concreto. La cal y otros ingredientes en el concreto fresco reaccionan lentamente con la solución penetrante de silicato, formando un gel insoluble en los poros del concreto. Se incrementa la resistencia al uso, agua, grasa o ácido.

INDUSTRIAS QUE UTILIZAN SILICATOS DE SODIO DE SIDESA INDUSTRIA FUNCIÓN DEL SILICATO BENEFICIO PRINCIPAL CERAMICA Cementos refractarios Fundición Diluyente de pasta Refinado de arcilla

Binder Defloculador Defloculador Defloculador

Fraguado de aire Sólidos altos Reducción de agua Mejora fluidez

CONSTRUCCION Endurecimiento de concreto Cementos a prueba de ácido Cementos refractarios Aislamiento térmico Solidificación del suelo

Reacción química, sellado. Binder Binder Adhesivo, formación de película Reacción gel

A prueba de grasa y polvo, resistente al ácido Fácil de usar, económico Capa dura, excelente acción térmica, resistente al ácido Capa a prueba de fuego Binder económico

PETROLEO Lodo de perforación Prevención de corrosión Rompimiento de emulsión

Control coloidal Reacción química Reacción química

Controla formación geológica Eficaz, reduce costo Rompe emulsión

PAPEL Tratamiento de agua cruda Aditivo de caja maestra Cubiertas Blanqueo con peróxido de pasta Adhesivos para laminación y etiquetado Tratamiento de agua pura Destintado

Floculación Floculación. Formación de película Reacción química Adhesión Floculación Detergencia

Mayor claridad en efluente Retiene finos y cargas en la línea A prueba de grasa, resistente a la humedad Conserva el peróxido, produce pasta más blanca Capas fuertes, económico Incremento de tamaño de floculo, clarificación mejorada Remoción de tinta

CARTON Tambores de fibra Tubos espirales

Adhesión Adhesión

Agrega rigidez, bajo costo Agrega rigidez, bajo costo

TEXTIL Blanqueo con peróxido Entintado

Reacción química Amortiguador de pH

Conserva peróxido, aumenta blancura Fijación de tinta, menores costos de proceso

COMPUESTOS DE LIMPIEZA/DETERGENTES Detergentes en polvo Jabones líquidos y limpiadores

Binder, inhibidor de corrosión y defloculador Defloculador y Amortiguador de pH

Ayuda de proceso en torre de secado y aglomeración. Protección a la corrosión y detergencia Detergencia y protección e inhibición a la corrosión

TRATAMIENTO DE AGUA Tratamiento de agua cruda y desperdicio. Prevención de corrosión en líneas de agua Control de contenido de plomo y cobre Estabilización de fierro y manganeso

Floculante Formación de película Reacción química Reacción química

Incrementa el tamaño y acelera la formación del floculo Película Protectora inhibe la corrosión del metal Reduce niveles de metales tóxicos Mejora el sabor, elimina agua roja

METALES Fundición porosa Cubiertas de varillas de soldadura Flotación de mineral Moldes de fundición y binders Polvos de fundición Pelletizado Briqueteado

Impregnación Binder Defloculador Binder Aglomeración Binder Binder

Sella fugas y llena huecos Buen binder y acción de flux Agente de separación y control de corrosión Ajuste rápido Elimina polvo, mejora las condiciones ambientales Ayuda e incrementa la formación del pellet Mejora características de flujo y propiedades cohesivas

TRATAMIENTO DE DESECHOS Solidificación y estabilización

Reacción química, binder

Reducción de porosidad y tiempo de fijación

NOTA La información contenida en este documento es una guía para el cliente. Las recomendaciones se hacen sin ninguna garantía. Antes de usar los productos se recomienda que el cliente determine la adecuación del material en su proceso y este asume el riesgo y consecuencias del mismo. No sugerimos la violación de ninguna propiedad intelectual o permitir practicar cualquier patente de invención sin una licencia.

PQ CORPORATION

CORPORATE HEADQUARTERSPO Box 840Valley Forge, PA 19482-0840Phone: 800-944-7411

IN CANADANational SilicatesPhone: 416-255-7771

IN MEXICOSilicates y Derivados, S.A.Phone: 52-555-227-6801

IN EUROPEPQ EuropePhone: 31-33-450-9030

IN AUSTRALIAPQ Australia Pty. Ltd.Phone: 61-3-9708-9200

IN TAIWANPQ Silicates Ltd.Phone: 886-2-2383-0515

INDUSTRIALCHEMICALS

DIVISION

Judy LaRosa ThompsonTechnical Service Representative, The PQ Corporation, Conshohocken, PA

Barry E. ScheetzProfessor of Materials, The Pennsylvania State University, University Park, PA

Michael R. SchockResearch Chemist, Environmental Protection Agency, Cincinnati, OH

Darren A. LytleEnvironmental Engineer, Environmental Protection Agency, Cincinnati, OH

Patrick J. DelaneyMunicipal Market Manager, Carus Chemical, Peru, IL

Presented at the Water Quality Technology Conference, November 9-12, 1997, Denver, CO

I. INTRODUCTIONAccording to the EPA’s 1991 Lead and Copper Rule (LCR), water distributionsystems are required to have corrosion control measures completelyinstalled by January 1997 for large systems (> 50,000 people), by January1998 for medium systems, and by January 1999 for small systems (< 3300people). The water systems are obligated to evaluate the methods listedbelow in order to find the optimal method for their particular system. Theymust ensure that the lead and copper concentrations at the consumers’ tapsmeet certain “action levels”; the 90th percentile of all sample measurementsmust not exceed 0.015 mg/L for lead and 1.3 mg/L for copper.

Possible treatment methods for corrosion control include:1) Adjust pH and alkalinity to decrease corrosivity of water.2) Saturation of calcium carbonate to promote precipitation of carbonate

layer within pipes.3) Add phosphate corrosion inhibitor.4) Add silicate corrosion inhibitor.

Of these methods, the effectiveness and the mechanism by which silicatesinhibit the corrosion of pipes are perhaps the least understood even thoughsilicates have been used regularly for this purpose since the late 1920’s.There has been a confusing assortment of both positive and negativeresults, and consequently, many contradictory statements have been pub-lished concerning this method. For example, in the chapter on corrosioninhibitors in “Internal Corrosion of Water Distribution Systems” [1], the follow-ing statements can be found: “Silicates are generally not effective,” (p.570);

Sodium Silicate CorrosionInhibitors: Issues ofEffectiveness and Mechanism

® Registered trademarks of PQ Corporation

Report 12

PQ Corporation is a privately held globalenterprise operating in 19 countries, withannual revenues in excess of $500 million.PQ is a leading producer of silicate, zeolite,and other performance materials servingthe detergent, pulp and paper, chemical,petroleum, catalyst, water treatment,construction, and beverage markets.

Potters Industries, PQ’s wholly ownedsubsidiary, is a leading producer ofengineered glass materials serving thehighway safety, polymer additive, fineabrasive, and conductive product markets.

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

and, “Test applications... have shown a generally excellent protection with sil-icate inhibitors,” (p.553). The mechanism of corrosion inhibition by silicates isalso in question. Again from the same reference: “The principal benefit ofusing silicates appears to be the increase in pH after dosage,” (p.552); and,“Surface analysis indicates that protection by the silicates occurs when a thinlayer forms over a layer of corroded metal,” (p.552). Such contrary state-ments point to the need for a more definitive evaluation of the effectivenessof silicates to control corrosion in water distribution systems especially con-sidering the requirements of the LCR.

In response to this need, this paper provides a review of various topics relat-ed to the use of soluble sodium silicates to inhibit corrosion of metal pipes.These topics include: metal corrosion and inhibition, characterization of sili-cate species in solution, known mechanisms of silicate film formation, reviewof experimental studies, and results of surface analysis of pipes. Some par-ticular issues that are addressed include: effect on iron and manganesesequestration, the effectiveness of silicate treatment with and without pre-existing corrosion products, the effect of water quality on silicate film forma-tion, and the pH effect of silicate treatment. It is hoped that this review willbring a measure of harmony to previous seemingly contradictory results andthat a contribution can be made to the provision of safer drinking water.

II. BACKGROUND

Metal Corrosion and Corrosion InhibitionCorrosion proceeds when all of the components of a “corrosion cell” arepresent. These include: 1) an anode, 2) a cathode, 3) an electrical connec-tion between the anode and cathode, i.e., the metal itself, and 4) an elec-trolyte to serve as the transport medium for ions. The anode and cathode areareas on the metal surface whose difference in electrical potential sponta-neously drives both the anodic oxidation reaction (M = M+ + e-) and thecathodic reduction reaction (A + e- = A-). In the case of a single metal, theelectrical potential difference may result from inhomogeneities on the surface(such as defects, impurities, or different crystal structures) or from momen-tary or systematic differences in the concentrations of electrolyte, dissolvedoxygen, or dissolved hydrogen [2]. In the case of two different metals oralloys, a galvanic cell forms, meaning that one metal or alloy, due to its morenegative electrical potential, tends to be the anode and is oxidized while theother acts as the cathode. For example, copper is cathodic to iron, zinc, andlead. Generally speaking, the locations of the anode and the cathode mayconstantly change, resulting in a uniform corrosion across the pipe surface.On the other hand, the sites may be fixed, become large in area, and causethe formation of pits, crevices, or tubercles.

It is common in the literature to see terms like “plumbosolvency” or “cupro-solvency” to describe the solubility of a metal under various water conditions(e.g., pH, alkalinity, dissolved inorganic carbonate). Metals might not neces-sarily be thought of as being “soluble” in water; however, dissolution impliesthat ions are transferred from a solid into a solution. In the case of ionic com-pounds such as salts, the dissolving crystal decomposes. But in the case ofmetals, dissolution (also known as leaching) occurs electrochemically viareactions that are non-oxidative, oxidative, or reductive, all with or withoutcomplex formation. Therefore, “plumbosolvency” indicates the fact that lead

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

can get into solution to some extent, whereas “lead corrosion” denotes howlead gets into solution (i.e., by electrochemical process).

Besides anodic and cathodic reactions, a variety of other reactions can takeplace to form numerous types of corrosion products. For instance, in the ironsystem, any of the following species may form, disappear, and reformthroughout the corrosion process: Fe2+, Fe3+, FeOH2+, Fe(OH)2

+, FeOOH(s),Fe(OH)2(s), Fe(OH)3(s), Fe3O4(s), FeCO3(s). In the case of copper, which is notusually susceptible to galvanic corrosion because it is cathodic to most ofthe common plumbing metals, corrosion proceeds in the presence of oxidiz-ers, e.g., 2 Cu(s) + O2(g) + 4H+ = 2Cu2+ + 2H2O. And with bicarbonate, 2 Cu2+

+ HCO3-+ 2H2O = Cu2(OH)2CO3(s). Depending on the exact conditions, other

corrosion products may include CuO and Cu2O or a great variety of others:Cu2OH2CO3, CuCl, Cu2CO3, Cu2SO4, Cu2S, Cu4(OH)6SO4, Cu3OH4SO4,CuO●CuSO4, Cu(OH)2, CuCO3●Cu(OH)2, Cu2OH3Cl, Cu2OH3NO3, Cu3(PO4)2,CuH2SiO4. The specific water conditions will influence which of these phasesoccurs [3].

Corrosion is also influenced by the water quality. The concentrations of dis-solved oxygen, chloride, sulfate, organics, and suspended solids plus thepH, alkalinity, and hardness may all have some effect on the corrosionprocess [2]. Theoretical predictions and experimental observations agreethat plumbosolvency, for example, decreases with increasing pH [4-6].

One way to inhibit corrosion in water distribution systems, then, is to makewater less corrosive, for example, by increasing the pH. Another strategy isto eliminate one or more of the corrosion cell components. Arpaia [7]explains that the pipe surfaces may be modified by protective deposits which1) act as a physical barrier between the surface and the electrolyte to inhibitthe flow of electrons and/or the diffusion of reacting species; and/or 2) havean electrochemical effect by displacing reactive species from the surface andthereby changing the electrical potential difference. For instance, oxide,hydroxide, or carbonate passivation layers may form on pipe surfaces due toappropriate conditions in the system (e.g. high CaCO3 levels in the water), orthe formation of protective layers may be induced by the addition of corro-sion inhibitors such as phosphates or silicates. The corrosion of lead may becurbed by the formation of a protective layer of lead carbonate or basic leadcarbonate (Pb3(CO3)3(OH)2) which serves as a physical barrier, limiting theuptake of lead [5, 8].

Soluble Sodium Silicate SolutionsBesides corrosion control in water distribution systems, soluble sodium sili-cates are utilized in a number of other applications. Some of these includethe following: engine antifreeze (to inhibit corrosion), detergents (to inhibitcorrosion, improve wetting, disperse oily soils, deflocculate inorganics, bufferwash water), paper adhesives, fire- and grease-proof films and coatings, orebeneficiation (to inhibit corrosion, disperse siliceous matter, aids in separa-tion of minerals), textiles (buffer in bleaching and dying operations, removeoil and dust from cotton), binders for refractory materials, stabilization andsolidification of hazardous waste, and concrete surface treatment to hardenand improve acid-resistance. It is interesting to note that a number of theseapplications utilize the corrosion inhibiting properties of sodium silicate.

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

Liquid sodium silicate, also known as waterglass, is produced by meltinghigh purity silica sand along with sodium carbonate at 1100-1200oC. Themolten glass is combined with water resulting in a solution that may consistof 24-36 weight % SiO2 with an SiO2/Na2O weight ratio between 1.60 and3.22. For water treatment applications, the SiO2/Na2O ratio is usually 3.22;however a ratio of 2.00 may be used in the case of very acidic water.Manufacturer’s recommend a start-up dosage of about 24 mg/L SiO2 for 30to 60 days followed by an incremental decrease to a maintenance dosageof about 8-12 mg/L. A high startup dosage is critical to gaining control overcorrosion. The protective layer should be formed as quickly as possiblebecause if only parts of the pipe surfaces are protected while others are stillexposed, the current density at the exposed parts will be magnified and cor-rosion may become worse before it gets better.

A sodium silicate solution consists of monomeric and polymeric species. Theconcentrations of monomer and polymer depend on the silica content andthe SiO2/Na2O ratio of the solution. To illustrate, a concentrated solution hav-ing a SiO2/Na2O ratio of 1.0 or 0.5 mainly consists of SiO3

-2 and HSiO-;whereas solutions with higher SiO2/Na2O ratios are characterized byincreasing polymer concentration and increasing polymer size (up to 30nmdiameter). Furthermore, the polymer is in equilibrium with the solublemonomer Si(OH)2 according to the following [9]:

SinO(4n-nx)/2(OH)nx + [(4n-nx)/2]H2O = nSi(OH)4

n = degree of polymerization, depends on SiO2/Na2Ox = OH/Si in polymer (x decreases as n increases)

The polymers are equiaxed and approximately uniform in size. In largerpolymers, the interior Si are linked by bridging oxygen while exterior Si maybe bonded to at least one OH-. In addition, the polymeric particles may beionized [9]:

SinO(4n-nx)/2(OH)nx + zOH- = SinO(4n-nx)/2(OH)nx-zOz-z + zH2O

So, the polymeric particles are in equilibrium with Si(OH)4, as stated above,and the Si(OH)4, in turn, is in equilibrium with monomeric silicate ions, forexample [9]:

OH- + Si(OH)4 = (HO)3SiO- + H2OOH- + (HO)3SiO- = (HO)2SiO2

-2 + H2O

Therefore, an equilibrium exists between ionic silica and colloidal, or poly-merized, silica. Arpaia expresses it this way:

nSiO3-2 + 3nH2O = (H2SiO3)nnH2O + 2nOH-

In general, as concentrated alkali metal silicate solutions are diluted (to alower limit of ~330 ppm), the pH and [OH-] are reduced, and silicate ionshydrolyze to form larger polymeric species along with a lower SiO2/Na2Oratio silicate [9]. As stated by Iler [10](pg.20): “When a solution of soluble sil-icate, which is always highly alkaline, is neutralized by acid to a pH belowabout 10.7, the silicate ions decompose to silicic acid [Si(OH)4], which thenpolymerizes to silica.”

However, for very dilute solutions (< 300 ppm SiO2) as in the case of drink-ing water applications (4-25 ppm SiO2), essentially complete depolymeriza-tion occurs and monomer (i.e., Si(OH)4 and HSiO3

- [9]) is the dominant

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

species. For example, upon adding a sodium silicate solution (SiO2/Na2O =3.22, ~400,000 ppm SiO2) to water, Lehrman and Shuldener [11, 12]observed that for dilutions of 14, 70, 140, and 350 ppm SiO2, only a few per-cent of the silica at most was in colloidal form whereas the majority wasmolybdate-reactive (i.e., monomer). According to the authors, these equilib-ria were reached very rapidly, “probably instantaneously.” For dilutions of700 and 1400 ppm SiO2, the percent colloidal silica increased over theseven days of the experiment up to 21 and 29%, respectively.

The kinetics of depolymerization of soluble sodium silicates in very dilutesolutions has been studied by Dietzel and Usdowski [13]. They obtained anempirical pseudo second-order law to describe the depolymerization:

[T]/[P]t = [T]kDt + [T]/[P]t=0

[T] = Total silica concentration[P]t = concentration of polymer at time tkD = reaction rate constant for depolymerization

For the case of a waterglass WG 37/40 (SiO2/Na2O = 3.15, 25.2 wt.% SiO2)diluted to 23 mg/L SiO2 at 20oC and a constant pH of 5.74, they found kD =0.3428 L/mol●sec and, by extrapolating to t = 0, [T]/[P]t=0 = 1.123. They alsoobserved that the depolymerization rate increased with pH and decreasedwith increasing final concentrations of silica (over 3.8 to 23 mg/L SiO2). For23mg/L SiO2, kD varies with pH according to: kD = pH11.61(5.662x10-10). So fora pH of 8.0, kD = 17.29 L/mol●sec. Substituting this into the above equationand assuming a similar [T]/[P]t=0 [13], it will take about 22.4 min. for the 23mg/L SiO2 solution to depolymerize to 90% monomer, and about 4.2 hoursto attain 99% monomer. Such values agree with the previous statementsthat depolymerization in very dilute sodium silicate solutions occurs rapidly.

The fact that negatively charged monomeric silicate species are present invery dilute solutions of sodium silicates is significant, especially consideringthe charged nature of pipe surfaces due to the presence of anodes, cath-odes, or other ionic corrosion products. No doubt, the possibility and realityof silicate film formation on pipe surfaces within water distribution systems islinked to these characteristics.

Iron and Manganese SequestrationOne of the challenges facing water systems is having to control a variety ofwater quality parameters. For instance, compliance with the LCR requires apH of 7-10; however, the polyphosphates that are commonly used to controlred and black water problems are less effective with increasing pH [14].Therefore, chemicals that serve in more than one capacity of water treat-ment are desirable. In fact, silicate, in combination with chlorine, is effectivefor Fe and Mn sequestration [15, 16]. As a result, silicate treatment maysimultaneously provide corrosion inhibition and Fe and Mn sequestration.

III. THE ROLE OF SILICATE IN THE FORMATION OF PROTECTIVE FILMS

Silicate Usage and Observations up to 1945The formation of protective silicate layers was observed as early as 1920 bySpeller who, based on experiment, suggested that the natural silica present

INDUSTRIALCHEMICALS

DIVISION

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

in water precipitated onto pipe surfaces via reaction with colloidal ferrichydrate. Speller further suggested that water could be purposefullyendowed with the properties that promote the formation of such corrosioninhibiting films [17]. Perhaps the first occasion of adding sodium silicate toa water supply was reported by Thresh in 1922. In this attempt to preventfurther occurrences of lead poisoning in an English moorland community,the silicate treatment not only controlled the dissolution of lead but alsosolved the red water problems by inhibiting iron corrosion. The treatmentalso was used successfully in other English communities having similarwater problems. Besides water distribution systems, sodium silicates wereadded to water systems in small apartment buildings, office buildings, andlaundries in New York, Boston, and Pittsburgh in the early 1920's. Thiswater treatment served as an alternative to mechanically or chemicallydeaerating corrosive waters. In 1923 Texter [18] reported that a "self-heal-ing" protective film formed in hot water systems upon additions of solublesodium silicates.

Stericker [17] summarized in 1945 what was known at that time about uti-lization of sodium silicates as a corrosion inhibitor. The selected items inthe following list are taken from his summary:

1. Large amounts of silicate speed up the formation of the protective film but increase the chances of removing accumulated products of corrosionfrom old piping.

2. The preferred silicate for waters in which the pH is > 6.0 is 3.22 SiO2/Na2O.

3. Larger amounts of silica are required when the water contains considerable amounts of chlorides.

4. Because of the negative charge on silica, it will migrate to anodic areas where at least a part of the charge will be neutralized.

5. The silicate films are electric insulators.6. In stagnant water, the supply of silica is soon exhausted.7. If only part of the area is protected, the remainder takes all the attack of

the corrosive medium. Therefore it is important to use enough inhibitor.

The Mechanism of Film Formation in Galvanized Iron and BrassThe mechanism of silicate film formation in Zn-containing metal pipes,specifically yellow brass and galvanized iron, was determined by Lehrmanand Shuldener [11, 12]. These authors showed that after a year of continu-ous addition of sodium silicate (8-12 ppm SiO2) to hot water, a thin gelati-nous film formed on the pipe surfaces. The film consisted of two layers.The layer adjacent to the pipe was white and mainly composed of zinchydroxide and/or zinc silicate with trace amounts of carbonate, and thelayer adjacent to the water was darker in color and contained silica gelalong with precipitates of Fe2O3 and organic material. By chemical analysis,the overall compositions of the films that formed in each kind of pipe wereas follows:

Brass: 67% silica and organic matter, 4%Cu, trace Bi, 9%Fe, 6% Zn, 1% Ca,<1% Mg, and 13%Na. (Metals were present as hydroxides and/or silicates.)Galvanized iron: 63% silica and organic matter, trace Cu, 10% Fe, 20%Zn, 1% Ca, <1% Mg., and 6% Na. (Metals were present as hydroxidesand/or silicates.)

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

Although the overall compositions of the layers were different for the twodifferent pipes, Lehrman and Shuldener indicated that the compositions ofthe white layers adjacent to the metal were very similar. In fact, the layerswere similar even for the different kinds of pipes that had carried silicatetreated (8-12ppm SiO2) hot and cold waters over 7 months to 20 years.

In brass pipe, the film was described as being evenly distributed, tan-col-ored, and adherent; whereas in galvanized iron pipe, the film was evenlydistributed, loose, rust-colored, and included white blobs which were deter-mined to be Zn(OH)2. For both types of pipe, analysis by X-ray diffraction(XRD) revealed that the predominant crystalline phases present in the filmswere ßFe2O3●H2O and quartz. (Quartz deposits were common for that par-ticular water even without silicate treatment.) In order to crystallize theamorphous material in the films, they were heated at 1000oC for 30 min.This resulted in the crystallization of ∞-cristobalite (SiO2) plus either cop-per oxide in the film from brass pipe or zinc oxide in the film from galva-nized iron pipe.

Based on the above results and additional experiments, the authors deter-mined that the mechanism of film formation consisted of two steps. First,positively charged corrosion products, namely zinc hydroxide, developedon the pipe surface to become the bottom layer of the total film. Secondly,negatively charged silica species were adsorbed, specifically bychemisorption, and formed the gelatinous, upper layer. This silica gel then“mechanically enmeshes particles” such as iron and hardness precipitatesand organic matter.

In addition to identifying the mechanism of film formation, other experi-ments in this work revealed the importance of having corrosion productspresent in order for silica to adhere to pipe surfaces. In saturated solutionsof hydroxides of Fe, Zn, and Cu (with no solid precipitate present), the con-centration of silica did not decrease over eight weeks indicating that thesilica species were not reacting with any of the metal hydroxides in solu-tion. However, when solid hydroxide was present in the saturated solutions,the silica concentrations did decrease. The decrease was greatest for zinchydroxide followed by ferric hydroxide and then cupric hydroxide. In addi-tion, it was observed that the original hydroxide solids became stickier,adhering strongly to the bottom of the bottle. Furthermore, pieces of blackiron and galvanized iron removed silica from sodium silicate solutions (ini-tially 10 and 292 ppm SiO2) over six and eight weeks, respectively. (In1952 they reported [12], “Magnetite Fe2O3 removes silica from very dilutesodium silicate solutions (8.5 and 290 ppm SiO2) to a larger extent thanzinc hydroxide at room temperature. However, the work has not pro-gressed far enough to indicate the mechanism.”) A yellow brass sample didnot remove silica, however neither did this sample corrode over the courseof the experiment. All of the above results indicate that the presence ofsolid corrosion products is another variable to be considered when design-ing experiments or interpreting results.

In a subsequent paper [19], Shuldener and Lehrman investigated the roleof the bicarbonate ion on corrosion in the presence of silicate for waters at160oF (71oC). Again they showed that silica was removed from water byiron and zinc corrosion products and was deposited on the specimens’ sur-

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

faces. This process occurred more quickly for zinc than for iron. But inaddition, they observed that the bicarbonate ion (HCO3

-) reversed thepotential from zinc being anodic to cathodic. Silicate, however, has theopposite effect, making the zinc more anodic, and therefore counteractsthe bicarbonate ion. The final potential, then, is determined by the relativeamount of each species in solution. For lower concentrations of silica andin the presence of bicarbonate, they deduced that the potential was movedtoward iron being anodic, and because iron corrosion products are slowerto remove silica, rust formed. They concluded that bicarbonate ionreversed the potential between zinc and iron to cause the formation ofrust, but that in the presence of silicate, the corrosion rated decreased.(E.g., 2.0 ppm SiO2, 4 ppm HCO3

-, at pH 7.1 proved to be a more corro-sive water than 8.5 ppm SiO2, 14 ppm HCO3

-, at pH 7.1.)

IV. CORROSION CONTROL STUDIES USING SODIUM SILICATES

In 1945 Stericker noted that there was a history of laboratory tests notmatching field experience for sodium silicate treatment [17]. He believedthat perhaps some factors had not been considered when the experimentswere designed. This section reviews several studies that have utilizedsodium silicate and evaluates the experimental design, the experimentalprocedure, and the resulting conclusions.

A. York Water District, York, Maine [20]

Data was collected over 1991 on this system consisting of 40% 50-100year old unlined cast-iron pipes and 60% cement-lined cast and ductileiron pipes. The average flow rate varies from 5MLD in the winter to 11MLDin the summer. The source of water is a surface water supply character-ized as soft (Ca < 1 mg/L), low alkalinity (8-10 mg/L CaCO3), pH 8.3-8.8,with low turbidity (<0.10 NTU), low color (<10CU), temperature of 13oC(range 4-24oC), 0.03 mg/L Fe, 0.06 mg/L mn, and 0.05 mg/L Al. The wateralso has a natural silica content of approximately 4 mg/L SiO2, which wasmonitored during the study. Regular treatment includes additions of alu-minum sulfate and NaOH for coagulation; chlorination for disinfection; andadditions of NaOH to raise the pH back to 8.3-8.8 (after chlorination) andammonia gas to promote the formation of monochloramine. Consumershad red water complaints due to an iron content in the range 0.40-1.9mg/L. For example, there were 15 complaints from 6/90-12/90.

Over 12/90 to the first week of 1/91, the average metal concentrationswere 83 ± 145 µ/L Pb, 0.33 ± 0.55 mg/L Fe, and 0.15 ± 0.13 mg/L Cu.Sodium silicate (SiO2/Na2O = 3.22) was added at 15-16 mg/L SiO2 for thefirst two months followed by a dosage of 9-14 mg/L SiO2. The pH, alkalini-ty, Ca, Pb, Cu, and Fe were measured at twelve points throughout the dis-tribution system. First-draw samples were taken after a 6-12 hour stagna-tion period. After a five min. flush, second- and third-draw samples weretaken. The flow rate over the course of the experiment was 130-280 millionliters per month.

For the period 5/91-12/91, average metal contents for the first-draw samplesshowed reductions compared to the previous measurements: 16 ± 9 micro-gram/L Pb, 0.10-1.37 mg/L Fe (compared to the range 0.10-1.9 mg/L Fe

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

measured previously), and “slight” reductions in Cu. According to thereport, some sites consistently had reductions in Pb content while othersremained relatively constant or increased. In general the iron contentsdecreased but then increased over the course of the study. The authorssuggested that this increase was due to decreasing flow rate, although itappears to be better correlated with increasing water temperature. Redwater complaints, however, were eliminated. It was observed that the pHwas lower at dead ends (7.52 ± 0.38) compared to central points (8.17 ±0.05). Also there were lower silica concentrations (16.0 ± 1.2 vs. 17.8 ±0.53 mg/L SiO2, includes natural silica) and lower alkalinity (5 mg/L vs. 10mg/L CaCO3) at dead ends. Consequently, the highest concentrations oflead were found at dead-end locations. In addition, the average silica con-tent was lower for unlined cast-iron mains (15.6 ± 1.5) than for other typesof pipes (17.5 ± 0.71), and second-draw silica was being taken out ofsolution (e.g., by being adsorbed onto pipe surfaces) by home plumbingsystems. The authors also noted that it was very difficult to maintain con-stant pH and silica content of the finished water because of the lowbuffering capacity and variations in the coagulation process. It was alsoevident that the problems of low silica content and low pH at the dead-ends were of some concern. They suggested monitoring silicate treatmentfor two or three years in order to gain a better understanding of any fluc-tuations that occur with a particular season.

B. City of Rochester Water Bureau [21]

Both a pipe loop and a field study were reported by the City of RochesterWater Bureau. In this case, zinc orthophosphate and KOH treatmentswere analyzed along with sodium silicate and a control. The initial waterhad an alkalinity of 60 mg/L CaCO3, pH 7.5-8.5, and hardness 80 mg/LCaCO3.

Loop Study: The pipe loop study, performed on AWWARF test racks,consisted of three different phases. In the first five weeks (4/22-6/7/92),no chemicals were added, and the four test racks were studied to see iftheir corrosion characteristics were similar. Only the soldered Cu loop Curesults were significantly lower in the rack that was to be treated withKOH. In phase II (6/8-12/10/92), treatment began: 1) KOH, pH to 8.5; 2)zinc orthophosphate, 1 mg/L as PO4 for two weeks and then 0.4 mg/LPO4 for the remainder; 3) sodium silicate (3.22 SiO2/Na2O), 20 mg/L SiO2

for two weeks followed by 22 weeks at 12 mg/L SiO2. In phase III (12/11-2/25/93), the silicate dose was changed to 8 mg/L SiO2 while the otherswere treated as in phase II.

For the phase II lead and copper results (i.e., Pb loop Pb results, solderCu loop Cu and Pb results, and Cu loop Cu results), the phosphate wasmost effective, followed by silicate, for reducing lead contents, while sili-cate was best, followed by phosphate, for reducing copper contents(based on median measurements). Both the phosphate and the silicatewere better than the control for Pb and Cu uptake. The KOH also showedimprovement over the control except for the case of the soldered Cu loopPb results. Based on the mean corrosion rates of the steel coupons, theKOH loop was the best followed by the control, silicate, and phosphateloops. The pH of each of the loops during this phase generally ranged as

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

follows: 8.5-9.25 for silicate; 8.25-8.75, KOH; 7.75-8.5, control; 7.5-8.25,phosphate.

In phase III which used a smaller silicate dosage, the silicate treatmentdid not perform well. Although it still resulted in significantly lower Pb lev-els than the control for Pb loop Pb results, it was not significantly differ-ent for the soldered copper loop Pb results. (No copper results werereported for phase III.)

The pipe loop studies showed that all of the treatments were effective atreducing lead levels from lead pipes, with zinc orthophosphate being thebest followed by sodium silicate. In addition, these two treatments werethe only ones to significantly reduce lead from soldered copper pipes.

Samples of copper pipe from this experiment were sent to ThePennsylvania State University to be analyzed by x-ray photoelectronspectroscopy (XPS), a method especially suited for studying thin films.(Please refer to the paper, “XPS Characterization of Films Formed onDistribution Systems Using Additives to Control Pb/Cu Levels in DrinkingWater,” for details of the analysis.) The analysis revealed that Si had beenincorporated into the film on the pipe treated with silicate; thus confirmingthe theory that silicate forms some kind of layer on the pipe surface.

Field Test: Two hydraulically isolated sections of the city’s distributionsystem were selected for the field study. Measurements were taken at 20different sites within each section , all of which had lead service lines tosingle family residences. Every two weeks, lead levels were measuredfrom first-draw samples. Before treatment began, the two sections weremonitored (5/4-7/27/93). The mean lead concentration in the test areawas 24ppb which was statistically higher than that of the control (21ppb).From 7/29 to 11/12/93, a silicate dose of 7-10 mg/L SiO2 was added tothe test area. During this time there was no statistical difference in leadlevels between the test and control areas. (Although this may be consid-ered an improvement over the pretreatment results.) The mean lead lev-els dropped in both cases to 11.3 ppb for the control and 9.7 ppb for thesilicate. However, the lead action level was exceeded in both cases.From 11/13/93 to 1/26/94, the silicate dosage was raised to 12 mg/LSiO2. For this period, only the silicate test area met the lead action level,and it was statistically less than the control area having an average leadconcentration of 7.3 ppb compared to 10.9 ppb. For two months the sili-cate treatment was stopped, and, as expected, there was a smallincrease in lead uptake in the test area. The average was 8.6 ppb whichwas not statistically different from the control average of 9.7 ppb.Although the lead levels decreased in the control area over the course ofthe experiments, the lead action level was exceeded in every set ofmeasurements. The same test area was then treated with KOH. (Thezinc orthophosphate was not used in the field study because zinc wouldcause a sludge disposal problem and phosphate could stimulate algaegrowth in the system's reservoirs.) Because of control problems, the pHvaried from about 8.4-9.1 throughout the approximately 10 weeks of thetreatment. During this time, the lead concentrations between the test (7.2ppb) and the control (9.1 ppb) areas were not statistically different.

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

In general, the City of Rochester Water Bureau concluded that a 7-12mg/L silicate treatment reduced lead levels at the consumers’ taps with adose of 12 mg/L SiO2 being the most effective. However, they planned tocontinue investigating pH adjustment to 8.5 with KOH since it also seemedeffective and could be less expensive.

C. University of South Carolina, Dept. of Civil Eng. [22]

The water used in this study was a ground water, low in alkalinity (0-2mg/L CaCO3), acidic (pH 5.2-6.3), soft (1-3 mg/L CaCO3), conductivity13.0-19.0 µmhos/cm, < 0.05 µg/L Pb, 0.4-1.4 mg/L nitrate, < 0.05 mg/L flu-oride, < 0.05 mg/L Zn, 0.03 mg/L Cu, 0.1 mg/L Mn, 0.05 mg/L Fe, 2.1-3.2mg/L Cl, 22-23oC. This water also had a natural silica content of 6.3-7.1mg/L SiO2. Each of the seven lines in this batch-mode, fill and draw type,system consisted of a 4L polyethylene solution tank from which waterflowed into an acrylic sleeve that housed ten copper couplings. These cou-plings had been partially coated with a 50:50 Pb:Sn solder and were elec-trically insulated from each other. For the first five weeks (4/28-5/17/93), alllines were treated with a high test hypochlorite to maintain a 0.5 mg/Lchlorine content. This was done to accelerate corrosion of the coupons sothat corrosion products would be present when sodium silicate (SiO2/Na2O= 3.22) and NaOH treatments began. From 5/18 to 10/6/93, lines 1-3 wereadjusted to 15, 20, and 25 mg/L SiO2, respectively. (These concentrationsincluded the background silica.) Line 4 received chlorinated water andserved as the control. The pH of lines 5-7 was adjusted with NaOH so thatthe pH of line 5 matched that of line 1 (pH 6.6), line 6 matched line 2 (pH7.0), and line 7 matched line 3 (pH 7.6). From 10/7 to 12/1/93, the silicateconcentration in line 1 was increased to 30 mg/L SiO2, and the pH in line 5was raised correspondingly (pH 8.6). There were no changes in the treat-ments of the other lines during this time.

Every week, copper and lead concentrations were measured in 125 mlsamples that were drawn from the columns after 8 and 68 hr. stagnationperiods. On other days, the columns were gently flushed with at least twothrough-put volumes of treated water. The pH and silica contents wereanalyzed daily. Water samples from the solution tanks were analyzedtwice/week to monitor alkalinity, total dissolved solids, conductivity, dis-solved oxygen, and temperature. The effects of stagnation were observedby monitoring the water quality in the tanks and in the coupon assemblies.The corrosion rates were determined by coupon weight loss after they hadbeen removed at the end of the study.

The copper up-take was minimal in all cases. The lead up-take wasreduced by 75, 81, and 87% over the control in the 20, 25, and 30 mg/LSiO2 lines, respectively, somewhat better than the corresponding NaOHlines by 12, 8, and 0%. However, for the last (i.e., seventh) month of treat-ment, the improvement over NaOH increased to 18, 11, and 7%, respec-tively. These results suggest that silicate offers some benefit besides thatof raising the pH. Also, with increasing silicate dose, there is less differ-ence between silicate treatment and the corresponding NaOH treatment;though with time, silicate provides increasing protection relative to NaOH.It may be that because of the higher pH resulting from the higher silicatedosage, the formation of lead corrosion products is slowed. However, with

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

time, they do form, and the silicate can interact with the corrosion products(adsorb onto them) and provide additional protection. To verify this, therate of lead corrosion product formation must be known at different pH’s.

According to a visual inspection of the samples, a “non-uniform, fibrousfilm” developed on most of the coupons. It was more prevalent on the sol-der surface but did “extend” to the copper surface as well. ConsequentXPS analysis at Penn State University on samples from Loops 3 and 7revealed that, as with the Rochester sample, Si was prevalent in the filmfrom the silicate treated pipe. A smaller amount of Si was also found in theNaOH treated pipe, presumably from the natural silica.

Although the stagnant feature of this experimental set-up is unique com-pared to other studies which utilize the AWWARF pipe racks, it was sug-gested that further investigation be performed under flowing conditions.Flow tends to enhance the performance of silicates because, apparently,more silica is likely to come in contact with the pipe surfaces compared tostagnant conditions which rely solely on diffusion.

D. City of Portland, Bureau of Water Works [23]

This water is low in alkalinity (6.0 mg/L CaCO3), soft (7.2 mg/L CaCO3),with a pH of ~ 6.8, and a natural silica content ~ 9 mg/L SiO2. TheAWWARF test racks included galvanized pipes and Pb-Sn soldered cop-per pipes. Copper and steel coupons were also incorporated. From 7/1 to8/8/93, water flowed 23hr/day during a pretreatment phase. After this(through 2/14/94), flow was restricted to 3.5hr/day, and the following treat-ments were administered:

1) pH9 (lime), alk. 20mg/L CaCO3 (actual pH was ~9.2)2) pH8 (lime), alk. 25mg/L CaCO3

3) Trisodium phosphate, 0.5 mg/L as P, lime to adjust pH to ~7.54) Sodium silicate, 10mg/L SiO2 (pH ranged from 8-9 from 8/23-10/25,

then it was usually ~9.2 for 11/3/93 to 2/14/94)5) Control

Every week a 1L sample was drawn after an 8 hr. stagnation period, andmetals uptake was analyzed. Of the five treatments, #1 and #4 performedthe best and are described below:

Source Metal Metal Uptake #4 % reduction over #5 #1 % reduction over #5

Pb solder Pb 75 68Brass Pb 92 92

Cu pipe Cu 95 87Brass Cu 90 87Brass Zn 96 93

Galvanized steel Zn 78 80

E. Las Vegas Valley Water District, Southern Nevada Water System [24]

This water was characterized by: alkalinity 125 mg/L CaCO3, pH 7.8, hard-ness 90 mg/L CaCO3, total dissolved solids 680 mg/L, 9 mg/L Si, 248-259mg/L SO4, 0.0017-0.0067 mg/L PO4, 16oC. Copper test loops were giventhe treatments listed below for six months and compared to a control. In

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

each case, there was 8 hr. of flow per day. Also shown are copper and leadreductions over the control. (The source of lead is not clear from this report.)

Treatment Dosage (mg/L) % Pb Reduction % Cu Reduction

1) Sodium polyphosphate 3.74 72 232) Silicate polyphosphate 4.44 -55 273) Zinc orthophosphate 0.93 67 -614) Sodium silicate 3.99 59 145) Blended orthophosphate 4.00 70 496) NaOH 7.65 22 36

The conclusions of this study were to continue testing only the threebest—sodium polyphosphate, blended orthophosphate, and zincorthophosphate—although it was admitted that silicate may have per-formed better if a higher start-up dosage had been used. For the final 3months, the silicate reduced Pb uptake by 64%. Had higher start-up andmaintenance dosages been used, and if a pretreatment phase hadallowed some corrosion product to form before silicate was added, the sili-cate treatment would have probably provided additional reductions inmetal up-take.

F. City of Newark, NJ, Division of Water/Sewer Utility [25]

The City of Newark Water Utility employed a two-cell corrosion test deviceand AWWARF test racks to investigate treatments of lime, orthophosphate,orth-poly blended phosphate, and silicate. The water quality was describedas: alkalinity 22-35 mg/L as CaCO3, pH 6.2-6.9, hardness 24-45 mg/L asCaCO3, turbidity 1.0-2.5 NTU, color 10-25 CU, 0.1-0.3 mg/L Fe, 5-25 mg/Lsulfate, 4.0 mg/L SiO2, 0.01-0.06 mg/L Mn, DIC 48-49 as CaCO3, 6-32mg/L chloride, 1-20oC.

In the two-cell device, steel coupons were placed in each of the cells. Rawwater flowed through the first cell and then was mixed with the inhibitor.The treated water then flowed through the second cell. The % reduction ofPb leaching for each of the inhibitors was as follows:

1) 15% for lime, pH 8.5-9.02) 50% for zinc orthophosphate (dose not specified)3) 40% for ortho-poly blended phosphate (dose not specified)4) 45% for silicate (1 month 20 mg/L, 1 month 9 mg/L)

These tests were followed by AWWARF pipe loop experiments. For sixweeks three racks were run with no treatment in order to flush them, toallow corrosion to begin, and to allow the water quality and metal contentto stabilize. By the end of this period, the water had an alkalinity 19.2 mg/LCaCO3, pH 7.3, and Ca hardness 23.0 mg/L CaCO3. After some initial fluc-tuations from pipe to pipe and rack to rack, the lead and copper concentra-tions leveled off to the following: soldered copper pipe Pb was 47.5-47.9ppb, soldered copper pipe Cu was 0.31 ppm, and lead pipe Pb was 428.5-437.0 ppb. The flow rate was one gal./min., and 1 L samples were drawnevery week following an 8 hr. stagnation period. The following treatmentswere applied after the stabilization period:1) Non-zinc orthophosphate: 1 mg/L, weeks 7-25 of the study2) Sodium silicate: 24 mg/L SiO3 for weeks 7-9; 18 mg/L for wks. 10-13;12

mg/L for wks. 14-21, 9 mg/L for wks. 22-33, 8 mg/L for wks. 34-383) Control: weeks 7-40

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

After the first 25 weeks of the study, the orthophosphate treatment wasstopped. That rack was cleaned, and then water with no treatment wasrun through from weeks 26 to 29. After this, ortho-poly blended phosphatewas fed at a dose of 1 mg/L (as ortho) from weeks 30 to 42. The averagewater quality during the course of treatment was : alkalinity 31.0 mg/L (allracks), Ca hardness 37.0 mg/L (all racks), and pH of 7.4 for the phos-phate treatments, 8.4 for 8 mg/L SiO2 dose, and 9.0 for the 24 mg/L SiO2

dose.

The blended phosphate did not provide effective corrosion control, andalthough the orthophosphate was considerably better, the water utilitydecided against phosphate treatment because of anticipated problemswith algae growth in its open reservoirs. The silicate treatment proved tobe very effective. It reduced the Pb content from lead pipe 65% over thecontrol and Pb from soldered copper pipe 60% over control. With respectto lead control, it was described as being more effective and more consis-tent than the orthophosphate treatment. All three treatments reduced cop-per concentrations. In the end, the water utility recommended sodium sili-cate to control corrosion starting with a dosage of 18-20 mg/L and thendecreasing to an 8 mg/L maintenance dose.

G. Greater Vancouver Water District [26]

The combination of acid rain and the absence of lime in Canadian lakesresult in acidic surface water. Greater Vancouver is no exception. Its waterhas a pH of 6.0-6.3 which drops to 5.4-5.9 after chlorination. The alkalinityof 1.5-3.7mg/L CaCO3 also drops to 0.5-2.0 mg/L after chlorination. Inaddition, the dissolved oxygen content is near to saturation levels. Allthese factors contribute to produce a rather corrosive water.

The following treatments were administered in this pilot plant study whichwas conducted over 12 months, 3/91-3/92:

1) Control, raw water2) Treated control; pH 8, alk. 20 mg/L3) Silicate/orthophosphate blend, 5 mg/L (0.18 mg/L as P, 0.83 mg/L as

SiO2 ); pH 8, alk. 20 mg/L4) Sodium silicate, 12 mg/L; pH 8, alk. 20 mg/L5) Zinc orthophosphate 1.5 mg/L (0.13 mg/L as P, 0.13 mg/L as Zn); pH 8,

alk. 20 mg/L6) Zinc orthophosphate, 4.5 mg/L (0.37 mg/L as P,0.37 mg/L as Zn); pH 8,

alk. 20 mg/L7) Zinc orthophosphate, 4.5 mg/L (0.37 mg/L as P,0.37mg/L as Zn);

pH7.5, alk. 10-12 mg/L

Loops 2-7 were disinfected with chloramine (2.5 mg/L). The alkalinity wasadjusted with NaHCO3 . For loops 2, 3, and 5-7, the pH was raised withCa(OH)2 . And, although the report does not say so, MacQuerrie’s thesisclearly says that the pH of loop 4, the silicate loop, was lowered from avalue >9 to 8 with addition of HCl [27].

Each loop consisted of the following: cast iron and copper pipe couponinserts (4” long, 1” ID), copper and mild steel corrosometer probes, 84’ oflead soldered copper plumbing coils downstream of the coupon inserts

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

(50:50 Pb:Sn solder every 4’, velocity 2.6’/sec., metal content from thissection was measured from 24 hr. standing water), 500g 50:50 Pb:Snsolder coils contained in plastic canisters downstream from the plumbingcoils, and brass faucets. Each loop had a total flow of 6 hr./day. Thewater temperature varied with the season. Any fluctuations that may haveoccurred in the water quality were not reported.

The silicate treatment (#4) provided comparatively good protection in sev-eral categories. The 12 month average copper coupon corrosion rate waslowest for treatments #6 and #7 (~0.0024 mm/yr) followed by #4 and #5(0.0047 mm/yr). The 12 month average cast iron corrosion rate was low-est for #4 (0.1350 mm/yr) followed by #6 (0.1510 mm/yr) and #7 (0.1600mm/yr). The copper corrosometer probe corrosion rates were lowest for#6 (0.0014 mm/yr) followed by #7 (0.0023 mm/yr) and #4 (0.0035 mm/yr).The mild steel corrosometer probe corrosion rates were lowest for #6(0.217 mm/yr) followed by #7 (0.265 mm/yr) and #4 (0.308 mm/yr). Thecopper uptake from the plumbing coils and the faucets was generally low-est for #2 followed by #4, and #6 followed by #4, respectively. The uptakeof lead from the plumbing coils was generally lowest for #2 followed by #1and #4. In the case of zinc uptake, only #2, 3, and 4 reduced the Zn levelbelow that of #1, the raw water case.

One interesting note is that there were several coincidental spikes in thedata for different metals and different loops. Coincidental spikes occurredfor copper from solder coils (loops 5, 6, and 7 only), zinc from soldercoils, lead from solder coils (except loop 4), copper from plumbing coils,and lead from plumbing coils. This phenomenon, however, may be gener-ated by the open nature of the different test coupons (i.e., corroding cop-per coupons upstream from lead and zinc).

At this point, it must be clear that silicate treatment performs best whenthere has been 1) a pretreatment phase to allow the formation of somecorrosion products and 2) a higher start-up dosage to facilitate a speediertransition from pipe surfaces being partially protected (i.e., highly suscep-tible to corrosion) to fully protected. Neither of these conditions were pro-vided in this study. Furthermore, the addition of HCl to a water distributionsystem is, to state it mildly, unrealistic, and undoubtedly this detractedfrom the performance of the silicate. In spite of these shortcomings, thesilicate treatment performed better than many or most of the other treat-ments especially for the cases of reducing the corrosion rate of cast ironpipe coupons, reducing copper uptake from the plumbing coils andfaucets, and reducing lead uptake from the plumbing coils.

V. DISCUSSION

With respect to the studies A through G given above, sodium silicate wasproven to be very effective at reducing metal uptake in all cases except Eand G. In E a very low dosage of 4 mg/L SiO2 was used for the durationof the six month experiment. In addition, there was no pretreatmentphase to allow for the formation of corrosion products which appear to benecessary for silicate film formation. Study G not only lacked a pretreat-ment phase, but HCl and NaHCO3 were both added to the silicate loopand undoubtedly reduced its effectiveness. (Recall the article by

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

Shuldener and Lehrman [19] which discussed the adverse effect of thebicarbonate ion with silicate treatment.)

In many of the studies reported above, references were given indicatingthat silicate was not an effective corrosion inhibitor. The three referencesthat were usually cited included Sheiham and Jackson [28], Boffardi [29],and Schock and Wagner [30]. Sheiham and Jackson analyzed thedeposits in lead pipes, studied the theoretical solubility of lead for differentwater conditions, and performed experiments to measure the lead uptakefrom lead pipes with different water conditions. The lead pipes were 4-6.5”long, 1/2” diameter. Water was fed continuously through them, and the“mean residence time” of water in the pipe was about 30 min. In one case,sodium metasilicate (SiO2/Na2O weight ratio = 1) was added continuouslyat a level of 10 mg/L SiO2 at a location immediately before water enteredthe pipe. As a variation, the silicate treated water was first fed through tub-ing for two days and then entered the lead pipe. These two tests were runon new pipes at pH 6.5 and on older pipes, which had been removed fromservice, at pH 7.5. A low alkalinity moorland water with a pH of 7.5 wasused. Addition of sodium silicate raises water pH anywhere from 1-2 unitsdepending on the dose and soda content. Considering the original pH ofthe water, it is apparent that after silicate treatment the pH was adjusteddown to 6.5 and 7.5 thereby losing any benefit that the silicate would haveprovided in raising the pH. (Details of the procedure indicate that the pHwas lowered by adding CO2(g).) In comparing the results that they obtainedfor silicate treatment vs. phosphate treatment, Sheiham and Jackson con-cluded that, compared to the control, the silicate had little effect but phos-phate was “much more attractive.” Since these tests were conductedunder conditions that misrepresent real service, one cannot conclude byextrapolation that silicate treatment is ineffective in the field.

Boffardi [29] is another reference that portrays silicates as ineffective:“Their effectiveness has not matched that of phosphate-based treatmentsfor protecting iron and steel. Silicate treatments are not recommended forcontrol of lead solubility in distribution systems.” Boffardi, however, pro-vides neither experimental evidence nor reference citation to support thesestatements. Since Sheiham and Jackson is one of three references listedat the end of the paper, it may be supposed that their report was influentialin forming Boffardi’s opinion.

Schock and Wagner [30] report on experiments performed by the DrinkingWater Research Div. of the USEPA in 1979 and 1980. In these experi-ments, sodium silicate was added at 10 and 20 mg/L SiO2 to a soft, lowalkalinity (20 mg/L CaCO3) water. The pH for both silicate treatments anda control was 8.2. Obviously for the three to be the same, the pH was low-ered for the silicate treated water. Three recirculating systems were usedwith new pipes. The lead levels in the control rose to about 0.175 mg/L Pbafter 60 days and then fell to about 0.140 mg/L Pb by the end of 75 days.The 10 and 20 mg/L SiO2 cases exhibited slightly better results. The leadlevels at 75 days were about 0.105 mg/L Pb and 0.095 mg/L Pb, respec-tively, with peaks at 0.110 and 0.105. By the end of 90 days, the levelswere about 0.095 mg/L Pb for both. After the first runs, the water wasemptied from the system and water of the same quality was added. Forthe second run, the control peaked at about 0.125 mg/L Pb and then

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

decreased to about 0.065 mg/L Pb by 105 days. At 105 days, the lead lev-els for 10 and 20 mg/L SiO2 were about 0.060 and 0.072 mg/L Pb afterhaving peaked at 0.063 and 0.095, respectively. The last measurement forthe 10 mg/L SiO2 was about 0.050 at 115 days, and that of the 20 mg/LSiO2 was about 0.040 mg/L Pb at almost 260 days. The authors conclud-ed that in order to achieve significant control, a treatment of 20 mg/L SiO2

would have to be administered.

Of those studying corrosion in water distribution systems, some believethat sodium silicate inhibits corrosion by raising the pH of the water; oth-ers hold the opinion that silicates form a protective layer against corrosion.The former implies a neutralization approach to corrosion control, and thelatter, a passivation approach [27]. If raising the pH is the only benefit toadding silicate, then its effectiveness is limited to suppressing those reac-tions in which H+ is a reactant and to changing the solubilities of metal,corrosion products, etc. However, if silicate forms a thin protective layer,then there should be additional benefits. This was the case in an experi-ment reported by Wehle [31] in which a dose of 7 ppm SiO2 reduced thecorrosion rate of galvanized steel significantly more than a NaOH treat-ment at the same pH. The work of Duffek and McKinney [32] also directlycompared the effects of silicate treatment with NaOH treatment adjustedto the same pH. For a range of 3 mg/L SiO2 (pH 7.9) to 500 mg/L (pH9.3), the NaOH treatments were comparatively ineffective at preventingcorrosion. In the study by the University of S. Carolina (described above)in which silicate treatments were analyzed alongside companion treat-ments of the same pH, the silicate performed more effectively for leaduptake although to a lesser degree with increasing silicate dosage andpH. However, with time, the differences between the corresponding treat-ments increased. Considering the fact that corrosion products must bepresent in order for silica to adsorb onto the surface of iron, copper, andzinc-containing metal pipes, perhaps the same is true for lead. In thatcase, the results are consistent with the idea that at higher silicate con-centrations, the pH is raised causing the corrosion of lead to slow. Sincethe corrosion products that are needed for a silicate film to form are slowto appear, the benefits of a silicate passivation layer are not experienceduntil later. Since the silicate layer depends on the presence of corrosionproducts, once these products are covered, the layer will cease to accu-mulate.

VI. CONCLUSIONS

Historically, silicate’s role as a corrosion inhibitor has received mixedreviews. This is due, in part, to the mystery surrounding how it works andalso because of the wide range of experimental findings. However, newevidence provided by XPS analysis indicates that a silicate film does formon the interior of pipes. This film may help to inhibit corrosion as indicatedby experiments in which silicate treatments were compared to NaOH ofthe same pH - silicate always provided superior corrosion control. In addi-tion, silicate’s poor or mediocre performance in some studies must bereconsidered because of misconceptions about how silicate functions. Insome cases, the pH of silicate treated water was actually lowered andthen compared to other treatments. Also, it is apparent that silicate filmsrequire the presence of corrosion products in order to form on pipes.

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

Some experiments, then, had a negative influence on the effectiveness ofsilicate because of their design. In order to clarify silicate’s efficacy, furtherstudy is warranted especially considering its potential to simultaneouslycontrol corrosion and red and black water problems.

VII. REFERENCES

1. Ryder, R.A. and I. Wagner (1985). “Corrosion Inhibitors.” Internal Corrosion of Water Distribution Systems. Denver. AWWA Research Foundation. 513.

2. Snoeyink, V.L. and A. Kuch (1985). “Principles of Metallic Corrosion in Water Distribution Systems.” Internal Corrosion of Water Distribution Systems. Denver, AWWA Research Foundation. 1.

3. Edwards, M., J.F. Ferguson, and S.H. Reiber (1994). “The Pitting Corrosion of Copper.” Jour.AWWA 86 (7):74.

4. Lee, R.G., W.C. Becker, D.W. Collins (1989). “Lead at the Tap: Sources and Control.” Jour.AWWA 81:52.

5. Schock, M.R. and M.C. Gardels (1983). “Plumbosolvency Reduction by High pH and Low Carbonate-Solubility Relationships.” Jour. AWWA 75:87.

6. Karalekas, P.C., C.R. Ryan, and F.B. Taylor (1983). “Control of Lead, Copper, and Iron Pipe Corrosion in Boston.” Jour. AWWA 75:92.

7. Arpaia, M. Corrosion and Scale Inhibitors for Drinking-Water Pipes. International Committee on Corrosion and Protection of Underground Pipelines.

8. Kirmeyer, G.J. and G.S. Logsdon (1983). “Principles of Internal Corrosion and Corrosion Monitoring.” Jour. AWWA 75:78.

9. Iler, R.K. (1979). The Chemistry of Silica. New York. John Wiley and Sons.

10. Iler, R.K. (1955.) The Colloid Chemistry of Silica and Silicates. Ithica, New York, Cornell University Press.

11. Lehrman, L. and H.L. Shuldener (1951). “The Role of Sodium Silicate in Inhibiting Corrosion by Film Formation on Water Piping.” Jour. AWWA 43:175.

12. Lehrman, L. and H.L. Shuldener (1952). “Action of Sodium Silicate as a Corrosion Inhibitor in Water Piping.” Ind. Eng. Chem. 44:1765.

13. Dietzel, M. and E. Usdowski (1995). “Depolymerization of Soluble Silicate in Dilute Aqueous Solutions.” Col. Poly. Sci. 273 (6):590.

14. Clement, J.A., M.R. Schock, and D.A. Lytle (1994). “Controlling Lead and Copper Corrosion and Sequestering of Iron and Manganese.” Critical Issues in Water and Wastewater Treatment. Proceedings from National Conference on Environmental Engineering. Published by Amer. Soc. Civ. Eng. 1.

15. Robinson, R.B., G.D. Reed, and B. Frazier (1992). “Iron and Manganese Sequestration Facilities Using Sodium Silicate.” Jour.AWWA 84:77.

16. Dart, F.J. and P.D. Foley (1972). “Silicate as Fe, Mn Deposition Preventative in Distribution Systems.” Jour. AWWA 64:244.

17. Stericker, W. (1945). “Protection of Small Water Systems from Corrosion.” Ind. Eng. Chem. 37:716.

18. Texter, C.R. (1923). “The Prevention of Corrosion in Hot Water Supply Systems and Boiler Economizer Tubes.” Jour. AWWA 10:764.

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

19. Shuldener, H.L. and L. Lehrman (1957). “Influence of Bicarbonate Ion on Inhibition of Corrosion by Sodium Silicate in a Zinc-Iron System.” Jour. AWWA 49:1142.

20. Michniewicz, D.C., J.A. Clement, T.S. Gimpel, and M.S. Schock (1993). “Sodium Silicate for the Simultaneous Control of Lead, Copper and Iron Based Corrosion.” Report on York Water District, York, ME.

21. Schantz, L.G. (1994) “Summary of Corrosion Control Studies.” Report by the City of Rochester, Rochester Water Bureau, Water Quality Operations, Rochester, NY.

22. McAnally, A.S. (1994). “Evaluation of Sodium Silicate for Lead/Copper Corrosion Mitigation Utilizing a Lab-Scale Treatability Study.” Dept. of Civil Eng., University ofSouth Carolina, Columbia, SC.

23. Watson, M. (1994). “Lead and Copper Rule Corrosion Control Study.” Report by the City of Portland, Bureau of Water Works and Participating Wholesale Customers, Portland, OR.

24. Gifford, G.F. and J. Fronk. “Pipe Loop Testing for Corrosion Control Optimization.” Report by Las Vegas Valley Water District, Southern Nevada Water System, Boulder City, NV.

25. City of Newark report on “Corrosion Optimization Study.” (1994) Division of Water/Sewer Utility, Little Falls, NJ.

26. MacQuarrie, D.M., D.S. Mavnic, and D.G. Neden (1993). “Greater Vancouver Water District Drinking Water Corrosion Inhibitor Testing, Part I: Weight Loss Coupons and Part II: Metal Mobility.”

27. MacQuarrie, D.M. (1993). GVWD Corrosion Control Initiative - Phase II Inhibitor Chemical Testing at Seymour Dam. Master's Thesis. The University of British Columbia.

28. Sheiham, I. And P.J. Jackson (1981). “The Scientific Basis for Control of Lead in Drinking Water by Water Treatment.” J.Inst.Water Eng.Sci. 35(6):491.

29. Boffardi, B.P. (1988). “Lead in Drinking Water - Causes and Cures.” Public Works (November):67.

30. Schock, M.R. and I. Wagner (1985). “The Corrosion and Solubility of Lead in Drinking Water.” Internal Corrosion of Water Distribution Systems. Denver. AWWAResearch Foundation. 213

31. Wehle, V. (1982). “Influence of Phosphates and/or Silicates on the Corrosion Behaviour of Drinking Water Towards Installation Materials.” The Institute of Water Engineers and Scientists, Scientific Section Symposium on the Internal Corrosion of Iron Mains and Copper Services. London, The Chameleon Press Ltd. 54.

32. Duffek, E.F. and D.S. McKinney (1956). “New Method of Studying Corrosion Inhibition of Iron with Sodium Silicate.” J. Electrochem. Soc. 103 (12):645.

Reprinted from Proceedings of 1997 AWWA Water Quality TechnologyConference, by permission. Copyright® 1997, American Water Works Association

Printed with the permission of WQTC.

Although the information and suggestions in this brochure("information") are believed to be correct, PQ Corporation makesno representations or warranties as to the completeness or accu-racy of the information. The information is supplied upon the fol-lowing conditions: The persons receiving the information willdetermine its suitability for their purposes; PQ Corporation will notbe responsible for damages of any nature whatsoever resultingfrom the use of, or reliance upon, the information or the materials,devices or products to which the information refers; Noinformation is to be construed as a recommendation to use anyproduct, process, equipment or formulation in conflict with anypatent; PQ Corporation makes no representation or warranty,express or implied, that the use thereof will not infringe anypatent; and NO REPRESENTATIONS OR WARRANTIES,EITHER EXPRESS OR IMPLIED, OF MERCHANTABILITY, FIT-NESS FOR A PARTICULAR PURPOSE OR OF ANY OTHERNATURE ARE MADE HEREUNDER WITH RESPECT TO INFOR-MATION OR THE MATERIALS, DEVICES OR PRODUCTS TOWHICH THE INFORMATION REFERS.

Copyright © 2003 by PQ Corporation.

All rights reserved. No part of this publication may be reproduced,stored in a retrieval system or transmitted in any form or by anymeans electronic, mechanical, photocopying, recording or other-wise, without the prior permission of the publisher and copyrightholder.

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

PQ Corporation is a privately held globalenterprise operating in 20 countries, withannual revenues in excess of $500 million.PQ is a leading producer of silicate, zeolite,and other performance materials servingthe detergent, pulp and paper, chemical,petroleum, catalyst, water treatment,construction, and beverage markets.

Potters Industries, PQ’s wholly ownedsubsidiary, is a leading producer ofengineered glass materials serving thehighway safety, polymer additive, fineabrasive, and conductive product markets.

B.E. ScheetzMaterials Research Laboratory, The Pennsylvania State University,

University Park, PA

J. LaRosa ThompsonPQ Corporation, R&D Center, Conshohocken, PA and

P.J. DelaneyCarus Chemical, Co., Peru IL

Presented at the Water Quality Technology Conference, November 9-12, 1997, Denver, CO

INTRODUCTIONSilicates have been used by the drinking water utilities for decades(Shuldener and Sussman, 1960). Initially, silicates were used for “red” watercomplaints in the distribution system. Control of zinc [from galvanized iron]and aluminum corrosion by silicates has also been described by earlyresearchers (Lehrman and Shuldener, 1951, 1952). More recently, silicateshave been found effective for reducing “red” and “black” water complaintsresulting from the oxidation of naturally occurring iron and manganese ingroundwaters.

The Lead and Copper Rule states that silicate treatment, phosphate treatment,and pH and alkalinity adjustment, are best available methods for controllinglead and copper in drinking water. However, detailed field and experimentaldata of the use of silicates for lead and copper control are scarce. In caseswere silicate treatment has been or is being employed, application appears tobe more of an art based on previous iron corrosion control experiences ratherthan on solid scientific knowledge. In the past, silicate manufacturers haveattempted to sell their product on the theory that silicates form a smooth,glassy, protective coating on the surface of distribution system materials.Typically, an initial “start-up” dosage is recommended, followed by a drop to a“maintenance” dose after “passivation” has occurred. These mechanisms andrecommendations lack sufficient scientific evidence to support them, however.Questions about the mechanism of corrosion control, water quality conditionsfavoring silicate use, and dosage requirements along with the lack of welldocumented experiences and chemical cost have led to a general lack ofconsideration by water utilities.

Probably the largest problem encountered through observations andliterature reviews is that the mechanism by which silicates work is unclear.The problem primarily arises from the difficulty in clearly separating pH

XPS Characterization of FilmsFormed on Distribution SystemsUsing Additives to ControlPb/Cu Levels in Drinking Water


Bulletin 17

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

and silicate effects resulting from sodium silicate addition. “N” sodium silicate(which is most commonly used in drinking water for corrosion control) is abasic solution (pH > 11). Depending on silicate dosage and a water’s alkalin-ity, it can increase the pH of the source water by more than 1 pH unit. A pHincrease of that magnitude would more than likely be beneficial to corrosioncontrol. Also, documented identification of silica-based films or solids onlead, copper, and brass surfaces that have been exposed to silicateinhibitors over significant time periods are very difficult to locate.

The most significant question that needs to be answered is whether silicateis incorporated into protective films on copper and lead surfaces. If so, it willbe important to identify the composition, form, and nature of the film, theconditions that favor its’ formation, and the mechanism by which the filmforms. A literature review shows that most of the previous detailed examina-tions of silica-based films on metal surfaces have been done on aluminum,zinc (or galvanized), and iron surfaces rather than lead and copper (Lehrmanand Shuldener, 1951, 1952; McCune, 1959; Lane et al., 1977a, 1977b;Shuldener and Lehrman, 1957). In addition, hot water is often used. Thefilm-forming mechanism in most of the cases is described as adsorption ofsilica to existing -oxide or -hydroxide films on the metal surface as men-tioned. The general thought is that the silica acts as a webbing on the sur-face, filling in the voids on the film and adding a thin layer to the surface.Theoretically, the film reinforcement provided by the silica results in addition-al protection and corrosion reduction. These films are typically described asbeing very thin. Because silica apparently adsorbs onto existing films, onceall the existing surface sites are occupied, the film does not build on itself,and for this reason manufacturers suggest dropping to a lower maintenancesilica dosage after silica “start-up” dosage has reduced metal levels. If thesilica film is not completed quickly, partial coverage may make corrosionworse because uncovered sites would have more concentrated corrosiongoing on.

Laboratory studies are often inadequate in supplying sufficient, usefulinformation primarily because of time constraints. There is also theproblem that the silicate deposit may be difficult to identify if it does notform a distinct crystalline solid phase. Further, it may affect the formationof other solid phases on the pipe surface, so general characterization ofall corrosion products may be necessary to discover the mechanism ofaction of the silicate. Examining pipe surfaces that have been extractedfrom distribution systems employing silicate treatment and have relativelywell documented water quality records will be critical in identifying amechanism.

There are a variety of techniques available to identify films on thesurface of pipes. Deposition of films within water pipes that result fromeither natural precipitation or precipitation induced by the addition ofagents to control the concentrations of regulated aqueous metal ions havebeen characterized by x-ray diffraction (XRD), scanning electronmicroscopy (SEM), and energy dispersive x-ray analysis (EDX) in anattempt to understand the chemistry and structure of the films. However,no one method provides a full understanding of the nature of the films.The objective of the research presented in this paper was two-fold: first,utilize x-ray photoelectron spectroscopy (XPS) to detect and characterizefilms on the interior of distribution pipes; and secondly, use XPS to

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

quantify elemental compositions of the films in order to supplement thedata from the more conventional characterization methods.XPS was chosen as an analytical tool in order to attempt to retrievemore detailed compositional information from films. With this technique,all elements, with the exception of H and He, can be detected and canprovide semi- and quantitative results on the elemental composition of theinterrogated sample. It is a non-destructive technique so the samples areavailable for additional analysis. XPS characterizes films to a depth of 0.5to 5nm and can be used for depth profiling to about 5nm. It has a depthresolution of a few nm and a lateral resolution of 75mm to 5mm. XPS hasan advantage over conventional XRD and SEM/EDX because it examinesa much smaller volume of material and is, therefore, especially suited forfilms. The disadvantage of XPS, however, is that it operates under a veryhard vacuum.

BASIS FOR THE XPS ANALYSISAnalysis of the experimental results of the XPS data is based onthe Einstein Photoelectric Law:

KE = hu -BE

where:KE = kinetic energy of the photoelectronhu = energy of the photonBE = binding energy of the electron.

By controlling the energy of excitation photon and measuring the kineticenergy of the expelled photoelectron, the energy associated with the bond-ing electrons from the elements in the surface of the test specimen can bedetermined. Because the technique examines the electrons directly, thesemethods can be used to identify the valence state of elements, the immedi-ate ligand environment around an atom, i.e. CH2, CH3. CF3 and CO appearas distinct carbons in the XPS analysis of CF3COCH2CH3 and can be usedto imply structure to the bonding pattern of the element, i.e. bridging vs.non-bridging.

SAMPLE SELECTION AND EXPERIMENTAL CONDITIONSSamples: Two sets of samples were examined; one from a controlled exper-iment conducted at the University of South Carolina’s experimental copperpipe loop and the other from a copper service line in Rochester, NY. TheRochester samples also represented a copper pipe loop that functionedwithout the addition of any additives, a control, then followed by the use of aphosphate admixture and finally by the addition of silicate. The samplesused in this study are summarized in Table 1.

TABLE 1. SUMMARY OF TEST SPECIMENS

Location Sample ID Source Pipe Type Treatment Type

Univ SC PQ3; PQ7 pipe loop copper silicate:high & low

Rochester, NY PQR1 pipe loop copper nonePQR2 pipe loop copper phosphatePQR3 pipe loop copper silicate

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

Instrumentation: The six samples used in this study were characterizedwith the aid of an environmental scanning electron microscope (ESEM), aconventional SEM equipped with energy dispersive x-ray analyses, x-raydiffraction and with x-ray photoelectron spectroscopy (XPS). Both SEM’sprovided photographs of the surface structures of the specimens. The envi-ronmental SEM was chosen for the initial characterization of the interior ofthe pipes and was operated at 10 torr of water vapor pressure as a precau-tion to prevent modification to the surfaces resulting from exposure to thehigh vacuum environment of the conventional SEM and the XPS.

The XPS utilized soft x-rays generated from a magnesium Kα target as theexcitation source. The Kratos instrument was operated at an accelerationpotential of 14 KeV with an anode current of 20mA. The sample area sur-veyed for the XPS analysis measured 3mm x 7mm. Occasionally specimenswere examined with a 700mm magnification which corresponded to a circu-lar iris opening of 1 mm. Data was collected in 0.3eV increments with adwell time at each step of 500 milliseconds. Quantification was conductedon the observed elements as detailed in Table 4.

The technique has the capability of differentiating between Si and Si bondedto adjacent oxygens. Although in the above table the analysis was reportedas Si, and is used in this report in that manner, it is clearly bonded to anoxygen network.

Table 2 summarizes the controlled conditions of the USC pipe loop experi-ment. In order to eliminate the potential pH effects of the addition ofaqueous silicate to the pipe loop, the pH was independently controlled atapproximately 7.7 while almost tripling the silicate concentrations betweentest #7 and test #3. Table 3 presents the composition of the waters anddosages of phosphate and silicate from the Rochester, NY specimens.

TABLE 2. CONDITIONS OF THE UNIVERSITY OF SOUTH CAROLINA PIPELOOP EXPERIMENT

Avg. pH Alkalinity Hardness TDS Silicatemg/L mg/L mg/L mg/L

CaCO3 CaCO3 SiO2

Loop 3 - PQ3 7.7 29.2 1.7 23.4 24.8Loop 7 - PQ7 7.6 18.7 1.7 16.5 7.1

TABLE 3. CONDITIONS OF THE ROCHESTER, NY SERVICE LINE SAMPLE

Sample ID pH Alkalinity Hardness Treatmentmg/L mg/L

CaCO3 CaCO3

PQR1 7.5-8.5 60 80 5 wks no additions

PQR2 7.5-8.5 60 80 ZnPO4 2wks @ 1mg/L33 wks @ 0.4mg/L

PQR3 7.5-8.5 60 80 silicate 2wks@20mg/L22wks@12mg/L11wks@8mg/L

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

RESULTSUniversity of South Carolina Samples: Sections approximately 1cmsquare were cut from the copper pipes which were removed from the pipeloop and forwarded for analysis. Each section was carefully flattened forcharacterization. Both samples exhibited a noticeable film on the interiorsurface of the entire pipe including the sampled section. The amount of thefilm varied between the two specimens; PQ 3 exhibiting a more heavilydeveloped film that graded from light to heavy and sample PQ 7 possesseda film that was just noticeable with the naked eye and exhibited distinctcolor zonations varying of white to yellow. Figure 1 is a micrograph takenfrom PQ 7 which reveals the lightly developed film and damage to the filmthat was caused by flattening the copper pipe section. A more detailed closeup of this surface in Figure 2 reveals the open porous nature of the film.Detailed examination of the film developed on PQ3, Figure 3 designatedlight zone, shows the development of a base film [cracking similar to Figure1 is readily visible] with secondary 1000 to 2000mm foil-like growths. As thefilm develops into a heavy encrustation, the secondary foil-like growthsthicken and colas as shown in Figure 4. A more detailed examination of thefoil-like growths begin to reveal micron-size objects that appear to be sub-euhedral crystal growth, Figure 5. X-ray mapping and EDX analysis of thefilms supports a composition that is substantially Si (see Figure 6). X-ray dif-fraction of these samples exhibited the scattering halo typical of amorphousmaterials.

TABLE 4. ELECTRONIC STRUCTURE AND BINDINGENERGIES FOR SELECTED ELEMENTS

Element electron structure binding energy (eV)

O 1s 534.4C 1s 290-285Si 2p 105Cu 2p 938Pb 4f 141Cl 2p 201

FIGURE 1: Silica-rich coating on interiorof University of South Carolina’s experi-mental pipe loop, sample ID-PQ7. Thecracking was introduced when the curvedcopper pipe specimen was flattened.

FIGURE 2: An enlargement of the surfaceof the film for sample PQ7 revealing the

porous nature of the surface. Scale bar isequal to 1 micron.

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

The results of XPS analyses for the surfaces of both of thesesamples and for each of the observed zonations are reported in Table 5.

FIGURE 3: Silica-rich coating on encrustedinterior of University of South Carolina’sexperimental pipe loop, sample ID-PQ3.1000-2000uum sized foil-like deposits haveformed on the film surface.

FIGURE 4: Heavy encrustation of PQ3exhibiting thickening and colasing of the

foil-like growths.

FIGURE 5: Detailed image of the foil-likegrowths on PQ3 revealing micron-sizedsub-euhedral grain growth.

FIGURE 6: Si X-ray map of the imagepresented in Figure 5.

TABLE 5. XPS RESULTS FOR UNIVERSITY OF SOUTH CAROLINA PIPE LOOPSAMPLES [MOLE %]

Sample ID O C Si Cu Pb Cl

PQ 3

heavy zone 66.5 7.3 21.6 2.6 0.5 1.5light zone 67.3 9.0 17.5 4.8 1.1 —

PQ 7

yellow zone 43 41 10.0 2.6 0.5 2.7white zone 53 28 10.5 4.1 1.5 2.3

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

Qualitatively, the sample with the heaviest encrustation corresponded tothe specimen with the highest silica concentration in the circulating water.Furthermore, there appears to be an inverse correlation to the alkalinity ofthe circulating water and the carbon [presumably as CO3

2- ] concentrationof the films. The copper and lead concentrations of both films appear tobe similar.

Rochester, NY Samples: This set of samples represented a long-termexperiment from the service lines of the City of Rochester, NY. The80-week experiment was conducted in three phases. pH, alkalinity andhardness were monitored during the entire test period and are reportedonly in summary term that are presented in table 3; pH 7.5-8.5, alkalinityof 60 mg/L CaCO3 equivalent and hardness of 80 mg/L CaCO3 equivalent.Initially the system was run for five weeks without the addition of anyCu/Pb suppressing ingredients. The second phase utilized zincorthophosphate for two weeks at a dosage rate of 1 mg/L followed by 33weeks at a rate of 0.4 mg/L. The final stage of the experiment utilized thesoluble silicate at a dosage rate of 20 mg/L for two weeks followed by 12mg/L for 22 weeks and finally at 8 mg/L for 11 weeks. At the end of eachof the three intervals, a fresh section of copper pipe was removed fromthe distribution system for analysis.

SEM characterization of the control and the zinc orthophosphatesamples revealed very little while the soluble silicate sample appearedsimilar to that reported in Figure 1. The XPS data are summarized inTable 6. As would be anticipated, the control sample and the zincorthophosphate samples contained no detectable silica but bothpossessed comparable levels of copper and lead. From these analysesthe corresponding oxygen for both samples were also comparable. Incontrast, the soluble silicate sample contained correspondingly lessoxygen and the silicone concentrations appeared comparable to levelsobserved in the laboratory pipe loop samples at comparable dosages. Itis interesting to note that no phosphorous was identified in the zincorthophosphate treated specimens.

A feature of XPS that can be exploited is the interpretation of thenext-nearest-neighbor environment of surface atoms in these samples.Figure 7a, b and c presents the actual spectral data for the bondingoxygen atoms associated with samples in which soluble silicate was usedas a treatment agent. In panels A and B, it is clear that the O 1s bindingenergy is perturbed as represented by shoulders on the emission peaks.That is, there appears to be two unique types of oxygen. Panel C,exhibits an O 1s spectra that is more typical of a single type of oxygen.Hochella and Brown (1988) have interpreted the presence of shoulders onthe O 1s emission peaks as being attributed to non-bridging and bridging

TABLE 6. XPS ANALYTICAL RESULTS FOR THE CITY OFROCHESTER, NY SERVICE LINE SAMPLES

Sample ID O C Si Cu Pb

control-PQR1 45.6 50.9 -- 3.3 0.08phosphate-PQR2 50.1 46.3 -- 3.5 0.02silicate-PQR3 74.8 10.0 10.0 5.1 0.04

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

Inte

nsi

ty

/ C

ou

nts

+ 1

000

0 1s55

50

25

20

550 548 546 544 542 540 538 536 534 532 530 528 526 524 522 520 518 516Binding Energy / cV

B

oxygen atoms. They attribute the bridging oxygen to the larger bindingenergy and the non-bridging oxygen to the smaller binding energy. Thedata in panel C would suggest that the silica tetrahedra in that particularfilm is essentially fully coordinated, one silica linked to another via an oxy-gen ion. When other cations, such as copper and lead in these samples,are incorporated into the films, the bonding pattern is disrupted whichresults in some silica being coordinated to an oxygen which is not in turnconnected to another silica. In principal, the relative proportion of the twotypes of oxygen in the films could then be determined which would assistin the interpretation of the structural nature of the film.

CONCLUSIONSThe objective of this study was to begin to supply the necessaryinformation that will serve to develop a theoretical understanding of howsoluble silicate functions to suppress corrosion and Cu/Pb in drinkingwater. The use of x-ray photoelectron spectroscopy has provided initial

FIGURE 7: XPS spectra for 01s atomscollected on silicate films showing thepresence of non-bridging oxygenatoms in panels A and B [top to bottom]and just non-bridging oxygen atoms inpanel C [bottom]. Panel D schematicallydisplays the bridging and non-bridgingcomponents of the 01s oxygen atoms.

Inte

nsi

ty

/ C

ou

nts

+ 1

000

0 1s15

10

5


A

Inte

nsi

ty

/ C

ou

nts

+ 1

000

0 1s40

30

20


C

D

535 527Binding energy (eV)

nbo

bo

PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

insight into this process. This study has shown that on all pipe loop andservice line specimens examined that have been treated with solublesilicate, detectable silica can be found on the interior surfaces of the pipesat the scale of several nanometers, suggesting a film development. Inthose specimens, ESEM and SEM/EDX examination, typically to a depthof several microns, also confirmed the presence of films on the speci-mens surfaces.

Data from the XPS has presented a compositional picture of the filmspossessing a silicone concentration varying between 10 and 20 molepercent and oxygen concentration between 65 and 75 mole percent.Additional support for the presence of a film can be derived from the next-nearest-neighbor behavior of the oxygen atoms that were characterized inthe surfaces as containing both non-bridging and bridging oxygens.

Clearly more needs to be done in order to establish the exact mechanismfor how soluble silicate works to minimize corrosion and Cu/Pb in munici-pal water systems.

REFERENCESLehrman, L. and H.L. Shuldener, “The role of soluble silicate in inhibiting corro-sion by film formation on water piping,”, J. AWWA, Vol. 43, 175-188 (1951).

Lehrman, L. and H.L. Shuldener, “Action of soluble silicate as a corrosioninhibitor in water piping,” Industrial & Engineering Chemistry, Vol. 44, 1765-1769(1952).

McCune, H.W., “Corrosion product and inhibitor films on aluminum,” J. Electro-chem. Soc., Vol 106, #1, 345-342 (1959).

Lane, R.W. T.E. Larson and S.W. Schilskly, “Silicate treatment inhibits corrosionof galvanized steel and copper alloys,” Materials Performance, Vol 12, #4, 32-37(1977).

Lane, R.W. T.E. Larson and S.W. Schilskly, “The effect of pH on the silicate treat-ment of hot water in galvanized piping, J.AWWA. Vol. 69, 457-461 (1977).

Shuldener, H.L., and L. Lehrman, “Influence of bicarbonate ion on inhibition ofcorrosion by soluble silicate in a Zn-Fe system,” J. AWWA, Vol. 49, 1432-1438(1958).

Shuldener, H.L. and S. Sussman, “Thirty years experience with silicate as a cor-rosion inhibitor in water system,” Corrosion, Vol. 16,#7, 126-130 (1960).

Reprinted from Proceedings of 1997 AWWA Water Quality Technology Conference,by permission. Copyright® 1997, American Water Works Association

Although the information and suggestions in this brochure("information") are believed to be correct, PQ Corporation makesno representations or warranties as to the completeness or accu-racy of the information. The information is supplied upon the fol-lowing conditions: The persons receiving the information willdetermine its suitability for their purposes; PQ Corporation will notbe responsible for damages of any nature whatsoever resultingfrom the use of, or reliance upon, the information or the materials,devices or products to which the information refers; Noinformation is to be construed as a recommendation to use anyproduct, process, equipment or formulation in conflict with anypatent; PQ Corporation makes no representation or warranty,express or implied, that the use thereof will not infringe anypatent; and NO REPRESENTATIONS OR WARRANTIES,EITHER EXPRESS OR IMPLIED, OF MERCHANTABILITY, FIT-NESS FOR A PARTICULAR PURPOSE OR OF ANY OTHERNATURE ARE MADE HEREUNDER WITH RESPECT TO INFOR-MATION OR THE MATERIALS, DEVICES OR PRODUCTS TOWHICH THE INFORMATION REFERS.


All rights reserved. No part of this publication may be reproduced,stored in a retrieval system or transmitted in any form or by anymeans electronic, mechanical, photocopying, recording or other-wise, without the prior permission of the publisher and copyrightholder.

SILICATO DE SODIO PARA TRATAMIENTO DE HIERRO Y MANGANESO EN AGUA POTABLE.

El hierro y el manganeso que se encuentran en el agua potable se consideran, generalmente no dañinos para la salud. Sin embargo, causan problemas estéticos cuando se encuentran en concentraciones que exceden a los límites permisibles ( Fe 0.3 ml/L, Mn 0.15 mg/L ). Estos problemas incluyen el agua coloreada, manchas en los accesorios del baño y promueven y acentúan el crecimiento bacteriano dentro del agua. El hierro y el manganeso pueden tratarse de vario métodos. Existen tratamientos convencionales como oxidación, precipitación y remoción, combinado con la aereación, sedimentación, filtración y posible adición química. Estos métodos eliminan el problema, pero se consideran muy costosos. El método propuesto es el de adicionar simultáneamente en el agua de pozo diversas dosis de silicato de sodio tipo N y suficiente gas cloro o hipoclorito. De esta forma, el hierro no forma precipitados ni da color o turbidez al agua por varios días. El silicato de sodio no es peligroso para la salud, ya que, inclusive, se utiliza en el tratamiento de aguas como floculante. Una técnica alternativa para el tratamiento, es el de secuestrar los iones fierro y manganeso; secuestrar significa prevenir la formación de color y turbiedad sin realmente remover el fierro y el manganeso. Este mecanismo involucra la formación de partículas coloidales que, por su pequeño tamaño y características, no generan color ni turbiedad. Las ventajas del método silicato de sodio/cloro comparadas con los métodos convencionales son: • No se producen lodos residuales, • No se requiere de instalaciones complejas; y además • El silicato que pudiera agregarse en exceso inhibirá la corrosión en el sistema, ya

que ésta es otra de sus cualidades. Este método, aunque es poco conocido y relativamente nuevo, se está utilizando en gran parte de las plantas tratadoras de Canadá y Estados Unidos de Norteamérica. Diversos trabajos experimentales han demostrado en que debe adicionarse el silicato de sodio, teniéndose como conclusión que, primero debe adicionarse el silicato y, de 15 a 30 segundos después, el cloro preferentemente en forma de hipoclorito. Para poder determinar la cantidad que debe adicionarse al pozo a tratar, es necesario hacer diferentes pruebas experimentales tales como pruebas de jarras y de filtros, así como un análisis químico, ya que estas variables determinarán la dosis de silicato a utilizar. 1

PRUEBAS DE ESTABILIZACIÓN DE HIERRO Y MANGANESO Este procedimiento está diseñado para determinar la concentración óptima de silicato de sodio que se requiere para la estabilización de hierro y manganeso en aguas subterráneas o de pozo. El procedimiento se realiza en un total de 2 a 5 horas, y el equipo necesario que se recomienda es el siguiente: • 6 jarras o vasos de plástico de un litro de capacidad. • 1 muestra de silicato de sodio diluido (0.250) (1% en peso de concentración de sílice

que puede llevarse a cabo de la siguiente manera: 1 parte en peso de silicato de sodio tipo N con 27 partes en peso de agua.)

• 1 muestra (0.250 litros) de una solución de cloruro al 1% en peso de cloro. • 1 matraz Kitazato y un filtro. • 6 piezas de papel filtro libre de hierro ( se recomiendan filtros de fibra de vidrio, que

son estándar para análisis de sólidos suspendidos en agua con retención efectiva 1.5 micrones, velocidad de flujo 13 ml/s y espesor de 0.033 cm.)

• 2 pipetas graduadas con 0.05 de precisión. PROCEDIMIENTO En el laboratorio, se adicionan a las jarras 0, 0.1, 0.2, 0.4, 0.6 y 0.8 gramos de solución de silicato al 1%, posteriormente, se les adiciona la cantidad de cloro utilizada normalmente por el usuario. Ya en el pozo, se llena cada una de las jarras con el agua de pozo bombeada. Asegúrese de que la muestra se tome antes del proceso normal de cloración y antes de que el agua esté expuesta al oxígeno atmosférico no más de 10 segundos. Tape las muestras inmediatamente y agítelas. Deje que las muestras se estabilicen (esto tal vez requiera de una a tres horas), agítelas ocasionalmente; posteriormente, filtre cada una de las muestras con diferentes filtros cada una. RESULTADOS Y EVALUACIÓN. Si ninguno de los papeles filtro presenta un residuo coloreado, podemos obtener dos conclusiones: 1. Hay poco o no hay hierro presente en el agua. 2. No se adicionó suficiente cloro. Debe de haber una concentración de silicato a la cual se reduce sustancialmente el residuo de hierro en el papel filtro como precipitado coloreado; entonces, no deberá de colorear la porcelana blanca o manchar la ropa. Como alternativa para evaluar los resultados, se puede medir el color de cada una de las muestras, siendo la dosis adecuada de silicato la correspondiente a la muestra, cuyo nivel de color sea el deseado.

2

SIDESA produce los silicatos solubles en hornos de alta temperatura, mezclando carbonatos de sodio o potasio con arena silícea. Como resultado de ésta reacción, se obtiene silicato de sodio sólido que, posteriormente, puede disolverse y dar lugar a una gran variedad de productos con diferentes propiedades. SIDESA recomienda el uso del silicato de sodio “N” para aplicaciones en tratamientos de agua, ya que su presentación facilita su embarque, almacenamiento y manejo. Propiedades típicas del silicato de sodio “N” Relación en peso SiO2 / Na2 O 3.22 % SiO2 28.7 % Na2 O 8.9 Densidad a 20°C, Be 41.0 Densidad a 20°C, g/cm3 1.38 pH 11.3 Características Líquido viscoso ACEPTACIÓN ECOLÓGICA. Los silicatos de sodio son considerados como productos no tóxicos y no corrosivos. No representan ningún riesgo a la salud o al medio ambiente. El uso del silicato de sodio para tratamiento de agua municipal está aprobado en Estados Unidos por American Water Works Association (AWWA), Food and Drug Administration (FDA), Enviromental Protection Agency (EPA) y la National Sanitation Foundation (NSF). En México está aprobado por la Secretaría de Salud. ASISTENCIA TÉCNICA. Para información técnica adicional relativa a las aplicaciones del silicato de sodio en tratamiento de agua, póngase en contacto con su agente de ventas o con el Departamento de Servicio Técnico el cual lo asistirá en caso de dudas o requerimientos adicionales EN APLICACIONES a los teléfonos: (55) 5227-68-00 o sin costo 01-800-90-685-00. NOTA La información contenida en este documento es una guía para el cliente. Las recomendaciones se hacen sin ninguna garantía. Antes de usar los productos se recomienda que el cliente determine la adecuación del material en su proceso y este asume el riesgo y consecuencias del mismo. No sugerimos la violación de ninguna propiedad intelectual o permitir practicar cualquier patente de invención sin una licencia.

3

Although the information and suggestions in this brochure ("information") are believed to be correct, PQ Corporation makes no representations or warranties as to the completeness or accuracy of the information. The information issupplied upon the following conditions: The persons receiving the information will determine its suitability for their purposes; PQ Corporation will not be responsible for damages of any nature whatsoever resulting from the use of,or reliance upon, the information or the materials, devices or products to which the information refers; No information is to be construed as a recommendation to use any product, process, equipment or formulation in conflict with anypatent; PQ Corporation makes no representation or warranty, express or implied, that the use thereof will not infringe any patent; and NO REPRESENTATIONS OR WARRANTIES, EITHER EXPRESS OR IMPLIED, OFMERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE OR OF ANY OTHER NATURE ARE MADE HEREUNDER WITH RESPECT TO INFORMATION OR THE MATERIALS, DEVICES OR PRODUCTS TO WHICH THEINFORMATION REFERS.


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, without the prior permission of the publisher and copyright holder.

6Reasons to ChoosePQ.Sodium Silicate asYour Corrosion Inhibitor

®

It WorksSodium silicate is a proven corrosion inhibitor,effective on a variety of metals, including lead,zinc and copper. Since first being used in 1921to control lead in English drinking water, sodiumsilicate remains the time-tested solution tocorrosion inhibition.

It Works—Two WaysSodium silicate reacts with metal surfaces toform a protective barrier against corrosion. Italso increases water pH, another importantcorrosion control mechanism.

It Works—SafelyJust add sodium silicate to your water distributionsystem. It’s safe. And it’s easy to handle, store,and apply.

It Works—EasilyUse sodium silicate as a corrosion inhibitor, andyou won’t need to think about other treatments.Sodium silicates contribute no zinc or phospho-rous to drinking water.

It Works—EffectivelySodium silicate is the simple, natural way to inhibitcorrosion. Many water systems in the UnitedStates and Canada are using this environmentallysafe, readily soluble liquid.

It Works—For YouBe sure to include PQ sodium silicate in your nextcorrosion control study.

PQ: The Hard Working Silicate ExpertsWhen you’re ready to make short work of corrosion with sodium silicates, give the experts at PQ acall. No other company combines such consistent product quality with strong technical support—we can show you how to efficiently and effectively use sodium silicates in your system. Then wecan supply you high-quality sodium silicates at highly competitive prices. And with 17 plants in theU.S. and Canada, you receive prompt delivery in the quantities you need. Call today to find out moreabout why PQ sodium silicates are the best choice in corrosion inhibition. 610-651-4200

PQ Corporation, recently acquired by JPMorgan Partners, is a leading producer of silicate, zeolite, and other performance materials serving thedetergent, pulp and paper, chemical, petroleum, catalyst, water treatment, construction, and beverage markets. It is a global enterprise, operatingin 19 countries on five continents, and along with its chemical businesses, includes Potters Industries, a wholly owned subsidiary, which is aleading producer of engineered glass materials serving the highway safety, polymer additive, metal finishing, and conductive particle markets.

INTRODUCTIONSoluble silicates are economical, effective, and environmentally responsiblechemicals which have been used for more than 70 years to protect metalsfrom the corrosive effects of water.1 They are classified as corrosioninhibitors because they can deposit protective films onto various metal sur-faces, isolating the metal from any further corrosive attack, and becausethey raise water pH which can make it less corrosive to metals. Silicates donot contribute zinc or phosphorous to treated water.

Laboratory and field experience has shown that silicate corrosion inhibitorsare effective in many different types of water. Protection is provided in bothacidic and alkaline water. In harder water slightly more silicate is needed toachieve the same degree of corrosion inhibition, since some of the injectedsilica may react with hardness ions before it has a chance to bond ontometal surfaces

PQ® silicates have been successfully used for corrosion control in water sup-plies of municipalities, industrial plants, oil refineries, textile mills, private res-idences, apartment buildings, and office buildings. Their corrosion inhibitionproperties also make them key ingredients in such products as automobileantifreeze and laundry detergents.

WHAT ARE SOLUBLE SILICATES?PQ Corporation manufactures sodium and potassium silicates. These solu-ble silicates are produced by fusing high purity silica sand and sodium car-bonate (or potassium carbonate) at temperatures of 1000 - 1500oC. Theresulting product is an amorphous glass that can be dissolved in water toproduce silicate solutions, sometimes referred to as “waterglass.” The silicain a silicate solution is present as both monomeric and polymeric anionicspecies (Figure 1), that exist in equilibrium with each other.2

Ratio and silicate concentration are two important factors that influence whatspecies are present in solution. At concentrations typical for corrosion con-trol, the silica monomer predominates.3

The proportion of silica to alkali in a sodium silicate is expressed as theweight ratio SiO2/Na2O. It is one of the main characteristics that influencesproduct properties and distinguishes one product from another.

PQ manufactures liquid sodium silicates which range in ratio from 1.60 to3.22. Typically, 2.00 or 3.22 ratio sodium silicate solutions, containing 25 to30% SiO2, are used for municipal water treatment.

PQ® Soluble Silicates:For Protection of WaterSystems From Corrosion


THE PQ CORPORATION



IN MEXICOSilicates y Derivados, S.A.Phone: 011-52-55-5227-6801




Bulletin 37-3INDUSTRIALCHEMICALS

DIVISION

INDUSTRIALCHEMICALS

DIVISION

Terminology: Silica vs. SilicateThe term “silica” refers to the compound silicon dioxide (SiO2.) Water is typi-cally analyzed for silica as part of normal water quality analysis procedures.Silica levels are usually reported as mg SiO2/L.

“Silicate” is a generic term for compounds that contain silicon, oxygen, andone or more metals. They can be naturally occurring or synthetic. As anexample, PQ’s synthetic sodium silicate compounds can be represented bythe generalized formula Na2O●xSiO2, where “x” varies from 1.60 to 3.22 forcommercial products.

Naturally occurring silicate minerals make up nearly 90% of the earth’s crust.4

Dissolved silica is a minor but ubiquitous constituent of the hydrosphere.5

Commercial soluble silicates have a higher degree of polymerized silicaspecies than natural dissolved silica because of higher concentrations; how-ever, when diluted they depolymerize to molecular species that are indistin-guishable from natural dissolved silica.6

PHYSIOLOGICAL AND ENVIRONMENTALEFFECTS OF SODIUM SILICATESodium silicate adds silicate anions, together with sodium and hydroxyl ions,to water.

Silica is found to some extent in all natural waters and is believed to be eco-logically harmless. The charged, polymeric nature of the silica found in syn-thetic silicate solutions is responsible for its reaction with metals and corro-sion inhibition properties.

The sodium content of water will increase slightly with sodium silicate addi-tion. This issue has been raised as a concern in some instances. At the high-est dosages recommended for potable water treatment (24 mg SiO2/L), PQ’sN® sodium silicate will contribute less than 5.6 mg Na/L to the water.

THE PQ CORPORATION







Na+

O-

Si

HO OH

O-

Na+

Na+

O- OH

HO O Si

Si O

O- O OH

O Si SiNa+

Si O O-O-OH

Na+Na+

OHO-

Na+monomer branched ring

FIGURE 1. TWO SILICA STRUCTURES FOUND INSODIUM SILICATE SOLUTIONS

Furthermore, when using N silicate at normal maintenance dosages of 4-12mg SiO2/L, the sodium contribution is 0.9-2.8 mg Na/L, respectively. Whenother sodium silicates are used, the sodium contribution will be differentdepending on the weight ratio of SiO2/Na2O. If no sodium addition is tolera-ble, potassium silicates offer an alternative.

Neither sodium nor potassium silicate corrosion inhibitors contribute phos-phorus or metals such as zinc to the ecosystem. These are concerns withother corrosion inhibitors, especially phosphorus-based types.

pH EFFECTSSodium silicates are alkaline chemicals. Treating water at typical levels of4-24 mg SiO2/L may raise the water pH anywhere from 0.1 to 2.0 pH units ormore. The actual pH increase will depend on overall water quality and sili-cate dosage. See Figure 2 for example. Increases in pH will generally helpminimize corrosion and will provide a synergistic effect along with depositionof monomolecular silica film.

APPROVALS FOR USEThe use of sodium silicates for the control of corrosion in municipal watersystems is approved by the American Water Works Association and theAmerican National Standards Institute (refer to ANSI/AWWA Standard B404).Sodium silicate also has Food and Drug Administration (FDA) unpublished“generally recognized as safe” (GRAS) status as a corrosion preventative inwater (at levels below 100 mg/L).7

The U.S. Environmental Protection Agency (EPA) recognizes that silicateinhibitors may be effective in controlling corrosion of lead and copper inpotable water systems.8

THE NATURE OF SILICATE CORROSION INHIBITIONStudies have shown that soluble silicates are reactive with cationic metalsand metal surfaces.9 This phenomenon is the basis by which silicates inhibitcorrosion and is illustrated at the top of the following page.10

THE PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

FIGURE 2. pH RESPONSE WITH SILICATE ADDITION

LAKE TAHOE WATER pH RESPONSE TO SODIUM SILICATE ADDITION

10

9

8

7

6

Measured Water pH

Silicate Addition (ppm SiO2)3.22 Wt. Ratio Na-Silicate Added

0 2 4 6 8 10 12

■■

■ ■■ ■

■■

■ ■ ■

Monomeric and polymeric silica, introduced into a water distribution systemas sodium silicate solution, is carried by flowing water to all parts of the dis-tribution system. At the dilution levels used for water treatment, the majorityof the silica depolymerizes to a reactive monomer form. The monomeric sili-ca, which can be represented by (SiO3)2-, is adsorbed onto metal pipe sur-faces at anodic areas, forming a thin monomolecular film on the interior ofthe pipe. This prevents any further corrosive reaction at the anode.

Corrosion will be inhibited when an anodic reaction of the type proposedbetween ferrous iron and silica (shown as equation (2) in Figure 4) occursin place of the reaction which forms ferric hydroxide (shown as equation (4)in Figure 4). Protective films can be formed by such a reaction.

Microscopic and x-ray examination of the film formed at the metal surfaceshow two layers, with most of the silica in the surface layer adjacent to thewater. When the hydrous metal oxide film has been covered with a silicafilm, silica deposition stops.11 The film does not build on itself, and thereforewill not form excessive scale. The film is an electrical insulator and blocksthe electrochemical reactions of corrosion, yet it is thin enough that it doesnot obstruct water flow.12

Corrosion protection with PQ sodium silicates can be achieved by modifyingthe SiO2 content of the water. Therefore the key water property is SiO2 con-

INDUSTRIALCHEMICALS

DIVISION

THE PQ CORPORATION







OHM

OH

O

M

FIGURE 3. DEPOSITION OF SILICA ON METAL SURFACE

+ Si (OH)4 =

OM

O

O

M

Si

OH

OH

FIGURE 4. POSTULATED ELECTROCHEMICAL REACTIONS OF IRON INWATER, WITH AND WITHOUT ADDED SILICA

IRON PIPE WALLIRON PIPE WALL

e-

CATHODE ANODE

(3) 2 H+ + 2e- H2 (1) Fe2++ 2e-

(2) Fe2++ SiO32-+ 2H2O FeOSi (OH)3++ H+

INNER IRON PIPE SURFACE

ANODE REACTION WITHOUT ADDED SILICA: (4) 4Fe2++ 3O2 + 6H2O 4Fe (OH)3

tent, not pH or calcium level, as in other types of corrosion control practices.Alkalinity, pH, and water hardness may influence the effectiveness of sili-cate treatment.

In many cases, natural SiO2 found in water probably has already “reacted”(i.e., adsorbed) with other metals in the water and may not be effective inreacting with metal pipes. Therefore, a fresh source of “reactive” SiO2 isneeded (as from soluble silicate solutions).

LENGTH OF CARRYWhen silicate treatment is first started, the system near the point of applica-tion is coated immediately; then the film continues to form at greater dis-tances. Eventually, with continued feeding and sufficient water flow rate, thetreatment becomes effective throughout the entire system.

Once the film has been created, the protection is maintained as long as sili-cate treatment is continued; if stopped, the protection is gradually lost. Ifdamaged, the film is self healing, as long as the silicate feed is continued.

METALS PROTECTEDSilicate treatment is effective in controlling corrosion of many differentmetals and in systems made up of a variety of metals.

METALS THAT HAVE BEEN SUCCESSFULLY PROTECTED BY TREATMENT OF SOLUBLE SILICATES:

■ Lead■ Copper■ Cast Iron and

Ferrous Metals■ Steel

PROTECTION OF CEMENTICIOUS MATERIALSAddition of silicate to water systems can protect cementicious materialsfrom long term deterioration. The silicate reacts with available calcium toform insoluble calcium-silicate compounds. Studies have shown that silicatetreatment may reduce the breakdown of asbestos-cement surfaces, thusprolonging the life of the material and minimizing the release of fibrilasbestos.13

COST OF TREATMENTThe cost for silicate treatment will vary, depending on the amount of silicatebeing fed into the water system, and the quantities purchased.

INDUSTRIALCHEMICALS

DIVISION

THE PQ CORPORATION







■ Galvanized Steel■ Bronze■ Red Brass■ Yellow Brass■ Nickel Alloys

A GENERAL EQUATION FOR DETERMINING THE COST OFSILICATE TREATMENT IS:

[8.34] [a] [b]

[c]

$

MG water=

a = desired SiO2 dose (in mg/L or ppm)b = $/pound silicate solutionc = wt. % SiO2 in silicate (decimal format)

TESTING THE EFFECTIVENESS OF SILICATE TREATMENTThe effectiveness of silicate corrosion inhibitors can be evaluated in anumber of ways. Options include:

1. Monitoring levels of dissolved metals (e.g., Fe, Cu, Pb) in water samples before and after treatment.

2. Conducting pipe-loop studies based on U.S. Army Construction Engineering Research Laboratory (USA-CERL) and/or American Water Works Association Research Foundation (AWWARF) procedures.

3. Conducting coupon testing based on ATSM Test Method D-2688, “Corrosivity of Water in the Absence of Heat Transfer.”

Results from a coupon test are presented in Figure 5. The water source wasa surface water in southern California. The treatment levels of silicate andzinc orthophosphate used were cost equivalent.These results are valid for the specific water tested and actual test condi-tions. The effectiveness of silicates on other waters may vary.

Current theory postulates that the presence of some corrosion product onthe metal surface will enhance the effectiveness of the silicates since thereaction is believed to be between silica species and cationic metals (Figure4). It is therefore recommended that coupon specimens and pipe loops beexposed to untreated water for a “conditioning” period prior to passivation.This practice will also be more representative of conditions that exist in actu-al water systems where corrosion has already occurred.

During lab or pilot testing, as in actual field use, a relatively high passivationdose of SiO2 is required initially for at least 30 days in order to establish theprotective film on metal surfaces.

PQ technical service personnel are available to assist in evaluations ofsilicate corrosion inhibitors.

INDUSTRIALCHEMICALS

DIVISION

THE PQ CORPORATION







STEEL CORROSION RATES

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

Corrosion Rate (MPY)

Water Treatment Method37 Days Exposure, 4 ppm SiO2, O.5 ppm Zn1010 Mild Steel/MPY=Mils Per YearZOP=Zinc Orthophosphate

Untreated Untreated N Silicate ZOP

FIGURE 5. WATER PLANT TRIAL RESULTS

SUGGESTED PQ SILICATES FOR MUNICIPAL WATER TREATMENTEither PQ’s N® or D® sodium silicate solutions can be used for corrosioncontrol. The choice will depend on water quality and operator preference.Since D silicate is slightly more alkaline than N, it may be preferred in lowerpH waters (e.g., water with pH below 6.0).

TYPICAL PROPERTIES OF PQ SODIUM SILICATES

PQ Product Name N® D®

Wt. Ratio (SiO2/Na2O) 3.22 2.00

%Na2O 8.9 14.7

%SiO2 28.7 29.4

Density @20oC (oBe') 41.0 50.5

(lb./gal.) 11.6 12.8

(g/cm3) 1.38 1.53

pH 11.3 12.7

Viscosity @20oC (cps) 180 400

OTHER SILICATESIn addition to solutions, PQ offers silicates in powder and glass form. Theproducts (e.g., SS® or METSO® ) can be mixed with other inhibitors to pro-duce blended products.

SILICATE DOSAGESAlthough there is natural silica (SiO2) in many waters, it may or may nothave any inhibiting effect and is generally not considered in determining thedosage of the silicate treatment.

PASSIVATION DOSAGENormally, relatively high dosages of silicate are required during the first 30to 60 days of treatment in order to form the initial protective coating. Thisinitial silicate dosage is referred to as a passivation dosage and should be24 mg/L above the background silica level.

The actual amount of time required to establish the initial coating willdepend on the amount of silicate injected, water quality, water flow rates,and system length.

MAINTENANCE DOSAGEAfter the first 30 to 60 days of treatment, or once film formation has beenverified (i.e., by SiO2 balance), the dosage can be reduced to a mainte-nance dose. It is advisable to reduce the silica dose incrementally, forexample by lowering it to 12 mg SiO2/L for a period of time (e.g., 30 days),then 8 mg SiO2/L, and possibly even down to 4 mg/L. Silica balancesshould be done over the system as the dosage is decreased in order toverify that the protective film remains intact.

INDUSTRIALCHEMICALS

DIVISION

THE PQ CORPORATION







CALCULATING SILICATE DOSAGESIt is important to remember that commercial sodium silicate solutions con-sist of various proportions of silica (expressed as weight percent SiO2),alkali (expressed as Na2O), and water. The required silicate injection isbased on the %SiO2 in the silicate, the water flow rate through the system,and the desired incremental SiO2 level.

The following general equation can be used to determine silicate solutionfeed rates in gallons per day:

[8.34] [a] [b] = gallons silicate solution

[c] [d] day

a = water flow rate (MGD)b = desired SiO2 dose (mg/L or ppm)c = wt. %SiO2 in silicate (decimal format)d = density of silicate (lb./gal.)

Example:

a = 2MGDb = 24 mg SiO2/Lc = 0.287 (wt. %SiO2)d = 11.6 lb./gal.

[8.34] [2] [24] = 120.2 gal. silicate solution

[0.287] [11.6] day

Alternatively, the following more specific relationships can be used:

Using N® SilicateTo obtain: 1 mg SiO2/LInject: 2.50 gal. N® per MG Water

For example, to get an incremental dosage of 8 mg SiO2/L into a waterstream flowing at 10 million gallons per day, 200 gallons per day N silicatemust be injected (2.5 x 8 x 10 = 200).

Using D® SilicateTo obtain: 1 mg SiO2/LInject: 2.25 gal. D® per MG Water

For example, to get an incremental dosage of 8 mg SiO2/L into a waterstream flowing at 0.5 million gallons per day, 9 gallons per day D silicatemust be injected (2.25 x 8 x 0.5 = 9).

In situations with low water flow rates it may be necessary to dilute thesodium silicate solution in order to get feed rates within the operating rangeof typical metering pumps.

INDUSTRIALCHEMICALS

DIVISION

THE PQ CORPORATION







DETERMINING WHEN A PROTECTIVE FILM HAS BEEN FORMEDIN A DISTRIBUTION SYSTEM BY SILICA BALANCEOne way to monitor formation of the protective silica film is to do a silica(SiO2) balance around the distribution system. This involves first determiningthe normal background level of SiO2 in the water to be treated, and monitor-ing it periodically to make sure any changes are recognized.

The silicate treatment will involve introducing a known incremental amountof SiO2 (e.g., 24 mg/L). Once treatment begins, water samples at strategicend points of the distribution system are periodically checked for SiO2. Oncethe incremental level of SiO2 at the ends of the distribution system matchesthe amount being introduced, no more SiO2 is being consumed (i.e., becom-ing part of the protective film). This means the initial protective coating hasbeen formed, and the treatment dosage can gradually be reduced to amaintenance dosage which will maintain the protective film.

It is important to continuously feed the silicate maintenance dose in order topreserve the film. If silicate treatment is stopped, the film will gradually dis-solve.

DETERMINATION OF SILICATest kits for determination of silica, as well as continuous monitoring equip-ment, are available from firms such as:

Hach Chemical CompanyP.O. Box 389Loveland, Colorado 80539

Classical laboratory procedures for the determination of silica by colorimetricand photometric techniques may be found in standard texts on inorganicanalysis.

Such procedures are published in “Standard Methods for the Examination ofWater and Waste Water,” American Public Health Asssociation, Washington,D.C., or in current ASTM Book of Standards (ATSM Designation D-859,“Standard Test Method for Silica in Water”).

Test methods for silica determination should be reviewed concerning possi-ble interference effects that may influence the accuracy of results.

METHODS OF FEEDINGSodium silicate solutions can be easily added to water by means of positivedisplacement metering pumps. Piston, plunger, diaphragm and gear pumpsare suitable. Chemical feed rates can be adjusted by changing the lengthand speed of the piston or diaphragm stroke, or by adjusting the gearspeed.

Silicate solution should be injected at a point near the discharge of the watertreatment plant, after final filtration, (i.e., at the clear well). With groundwater,the silicate can be injected at the well site. The silicate feed pump can bewired to activate based on water flow rates, or in conjunction with water trans-port pumps, in order to maintain the proper proportion of silicate to water.

INDUSTRIALCHEMICALS

DIVISION

THE PQ CORPORATION







INDUSTRIALCHEMICALS

DIVISION

THE PQ CORPORATION







Aside from a suitable storage tank and feed pump, no special equipment isrequired to use silicate. The silicate solution can be stored in drums, day-tanks, or if delivered in tanker trucks, in bulk storage tanks. It can be fed atdelivered concentration or diluted with water. If diluted, this must be consid-ered when calculating dosages.

The internals of the silicate feed pumps should be flushed with water whenthey are taken out of service (e.g., during routine maintenance.) For addi-tional information refer to PQ bulletin 17-70, “Storage and Handling of PQLiquid Sodium and Potassium Silicates.”

SAFETYSodium silicate solutions are inorganic compounds. They are not explosiveor flammable and are not classified as hazards in U. S. Department ofTransportation shipping regulations.

Due to their moderate alkalinity, care should be taken to prevent contact ofsilicate solutions with eyes and skin. Personnel involved in unloading and/orhandling silicate solutions should use protective clothing and equipment toprevent accidental contact. This equipment generally consists of hard hat,chemical goggles or face shield, alkali-resistant gloves, safety footwear andcoveralls.

If silicate is splashed into the eyes, flush immediately with warm water andseek medical attention. If splashed on the skin, sodium silicate solutionshould be washed off with water, preferably warm water. If allowed to remainin contact with the skin, irritation may result.Wet silicate spills are slippery and will dehydrate to form glass-like films;therefore, it is recommended that minor spills be rinsed with water immedi-ately.

Some dried deposits of liquid silicate may form sharp edges (e.g., if spilledaround openings of sample jars), and care should be taken in such casessince skin cuts may result. Such cuts should be washed with water andgiven appropriate medical attention to prevent infection.

Additional safety information is contained in Material Safety Data Sheetssupplied by PQ Corporation.

SUMMARY TABLE1. Soluble silicates are economical, effective, and environmentally responsible chem-

icals that have been used for more than 70 years to protect metals from the corrosive effects of water. They do not contribute zinc or phosphorus to the environment.

2. Silicates are approved as direct additives to potable water. They are nonhazar-dous, nontoxic, and nonflammable. They do not impart any taste or odor to water.

3. American Water Works Association Standard for Liquid Sodium Silicate (ANSI/AWWA B404) recognizes the use of sodium silicate in water treatment.

4. The U.S. Environmental Protection Agency recognizes that silicates may be effective in controlling lead and copper corrosion in potable water systems.

5. At the dilutions typical in water treatment, most of the added silica is in the monomeric form.

6. The silica in sodium silicate solutions carries a negative charge and will migrate to anodic areas where it can react with metallic ions and form a protective film that inhibits corrosion.

7. The alkali present in silicate will typically raise pH. Increases in pH generally lead to decreased corrosion rates.

8. The film does not build on itself and will not obstruct water flow.

9. In areas of low water flow the supply of silica may eventually be exhausted within the effective range of the electrical forces around the anode. A sufficient water flow is required to supply additional silica.

10. In areas of low flow, the pH contribution of the silicate may also be reduced.

11. If only part of the area is protected, the remainder takes all the attack of the corrosive medium. Therefore it is important to use enough inhibitor.

12. The efficacy of the silicate treatment may vary with the type of metal.

13. The treatment has inhibited corrosion in systems where two dissimilar metals are in contact.

14. A passivation dose of 24 mg SiO2/L is recommended during the first 30-60 daysof treatment in order to quickly establish the protective film.

15. After the protective film has been formed, it can be maintained by feeding less silicate. In most waters a maintenance dosage of 8 mg SiO2/L is effective.

16. The minimum recommended maintenance dosage is 4 mg SiO2/L

17. Higher dosages of silicate have been shown to give better protection from corrosion.

18. Higher dosages of silicate speed the formation of the protective film.

19. Optimum silicate dosage will depend on specific water chemistry and system characteristics.

20. The preferred sodium silicate solution for treating water with pH greater than 6.0is 3.22 ratio SiO2/Na2O. For waters with pH of 6.0 or lower, 2.00 ratio SiO2/Na2O is preferred.

21. Silicate solutions from PQ are supplied ready to use, or may be diluted.

22. Contact PQ Corporation for product samples or additional information.

ABBREVIATIONS

MGD: Million Gallons per Daymg/L: Milligram per Litercps: Centipoiseppm: Parts per Million

INDUSTRIALCHEMICALS

DIVISION

THE PQ CORPORATION







FACTORS FOR UNIT CONVERSIONS

Quantity Equivalent Values

Mass 1 kg=1000g= 0.001 metric ton = 2.20462 lb

Volume 1 m3 = 1000 liters = 106 cm3 = 106 ml

= 22.83 imperial gallons = 27.41 gallons

Concentration 1 mg/L = 1 ppm = 0.0001%

Viscosity 1 centipoise = 1 millipascal-second = 0.01 poise

LITERATURE REFERENCES1. Thresh, J. C., Analyst, p. 459, 1922.

2. Iler, R. K., The Chemistry of Silica, Wiley-Interscience, New York, NY, p.127, 1979.

3. Katsanis, E. P., Esmonde, W. B., and Spencer, R. W. "Soluble Silicate Corrosion Inhibitors in Water Systems," Materials Performance, Vol.25, No.5, p. 19, 1986.

4. Encyclopedia of Chemical Technology, 3rd ed., s. v. "silicon compounds (syn inorganic silicates)."

5. Ibid.

6. Ibid.

7. Select Committee on GRAS Substances, Evaluation of the Health Aspects of Certain Silicates as Food Ingredients, SCOGS-61, Federation of American Societies for Experimental Biology, NTIS Publication 301-402/AS,Springfield, Va., p.4, 1979.

8. 40 CFR 141.82, National Primary Drinking Water Regulations-Description of Corrosion Control Treatment Requirements, revised July 1, 1991.

9. Falcone, Jr., J. S., Soluble Silicates, American Chemical Society Symposium Series No. 194, Falcone, Jr., J. S., ed. p. 133, 1982.

10. ller, R. K., The Chemistry of Silica, Wiley-Interscience, New York, NY, p 83, 1979.

11. Vail, J. G., Soluble Silicates, Their Properties and Uses, Vol. 2, Technology, Reinhold Publishing, New York, NY, p.251, 1952.

12. Ibid., p. 256.

13. Schock, M. R. and Buelow, R. W., "The Behavior of Asbestos-Cement Pipe Under Various Water Quality Conditions: Part 2, Theoretical Considerations, " AWWA J., Vol. 73, No. 12, p. 646, 1981.

INDUSTRIALCHEMICALS

DIVISION

THE PQ CORPORATION







METSO®, SS®, N, and D® are registered trade-marks of the PQ Corporation.

Information herein is accurate to the best of ourknowledge. Suggestions are made without war-ranty or guarantee of results. Before using, usershould determine the suitability of the product forhis intended use and user assumes the risk andliability in connection therewith. We do not sug-gest violation of any existing patents or give per-mission to practice any patented invention with-out a license.

37-3/1002

silicatos de sodio liquidos / sÓlidos de sodio n... · 2009-01-26 · contenido de oxido de sodio...

Documents