Lane Shaw and David Yu, Director of Technology & Business Development and Technology Manager of Plastics and Polymers for Huber Carbonates09.01.18
GCC and Abrasivity
Abrasion characteristics of any material are a function of hardness, concentration and particle size.
Hardness is typically measured on the Mohs scale, where a diamond is 10 and talc is 1 (Figure 1). A higher Mohs hardness will scratch or abrade a material with a lower number. Calcium carbonate is a soft mineral with a Mohs hardness of 3.
Since ground calcium carbonate is a naturally mined material, impurities are present as they are in all ore bodies.
Almost all contain some dolomite (CaMg(CO3)2) with a Mohs of 4, while others contain mica (Mohs 3) or graphite (Mohs 1-2). None of these softer minerals are of concern for abrasion.
However, almost all GCC has some silica (Mohs 7) present. This may or may not present a concern depending on the concentration and its particle size, as we discuss in greater detail later.
Finally, the particle size of the GCC is an important parameter. Coarse-grit sandpaper is far more abrasive than fine-grit sandpaper, even though the grit is composed of the same material. Therefore, the median particle size and distribution of the GCC is critical, as is the particle size and distribution of the impurity.
There are several ways to deal with silica content in the production of GCC. Many ore bodies produce very white GCC but are high in silica. To address this, ores are first milled to a 20-45 µm particle size and undergo froth floatation. In this process, the GCC is slurried with water and a floatation chemical that binds to the silica. Air is bubbled through the mixture and the froth that forms carries the silica with it while the GCC sinks. While not 100% effective, it does remove the majority of the silica. Since the product is already wet, economies of ultra-fine grinding favor wet milling to the desired particle size.
If required by application, the GCC slurry is then dried, which is an energy intensive process.
An alternative production method is to start with an ore body with inherently low silica and dry mill it as found. This production method usually results in a slightly less bright product but has favorable economics.
Experimental
Various 3 µm GCC products were obtained from producing locations in the U.S. All products were mined from ore in that geographic region with the state identified. The exception is “CAR” which is Caribbean sourced ore that was milled at Huber’s GCC manufacturing facility in Quincy, IL. The products were measured for acid insolubles (primarily silicas and silicates). Most exhibited a low level of acid insolubles, whether wet ground (WG) or dry ground (DG), reflecting the underlying quality of the ore deposit or the efficiency of the wet floatation process (Figure 2). The obvious exception is the Hubercarb® M3 calcium carbonate produced in Marble Falls, TX. This sample was selected as a lot on the very high side of the acid insoluble range encountered during normal operations.
To examine the abrasivity of GCC products, the samples from various ore sources and milling methods throughout North America were tested. A Brabender® torque rheometer with #6 rotors was modified with a removable soft brass (D65 hardness) pin through one of the rotors (Photo 1).
A standard twin screw, tin-stabilized high fill (30 phr calcium carbonate) PVC pipe recipe was run for 16,000 cycles. To minimize the effects of PVC thermal degradation, 10 runs of 20 minutes each was required with 80 rpm rotor speed for each data point.
Examining the effect of GCC particle size with a fixed ore source, a comparison of Hubercarb G3 (3 µm median) with Hubercarb G260 (20 µm median) calcium carbonate from Marble Hill, Georgia, clearly shows that the larger particle size calcium carbonate abrades more than the finer one, as would be expected (Figure 3).
When all 3 µm products were compared for abrasion, there did not appear to be a wide data spread (Figure 4). It did not make a significant difference as to the ore source or the grinding method. The products from ore sources low in acid insolubles (like Hubercarb W3N calcium carbonate) tended to be on the low end of the range, and those with high acid insolubles (Hubercarb M3 calcium carbonate) tended to be on the high end of the range, but the Alabama wet ground material was also at the upper end of the narrow range despite a lower amount of acid insolubles.
The data points out an inconsistency because the expected result of the Hubercarb M3 calcium carbonate would be high abrasion. However, this study backed up experimental evidence where the Marble Falls, TX, Hubercarb M3 calcium carbonate has been used for decades without abrasion issues.
To further explore this inconsistency, a study of the particle sizes of the actual acid insoluble (silica fraction) was undertaken. The acid insoluble fraction of each product was measured for particle size (Figure 5). In all GCC products, except one, the median size of the acid insoluble fraction exceeded that of the GCC. This is logical since the GCC products are either dry or wet centrifugally classified and the density of silica (2.65 g/cc) is essentially the same as calcium carbonate (2.71 g/cc). Being the harder of the two materials, silica will be less likely to fracture and results in a slightly larger particle size.
The exception was the Hubercarb M3 calcium carbonate grade from Marble Falls, which had a median particle size of only 1.2 microns. As demonstrated earlier, a finer particle sized material of a given hardness results in lower abrasion. Thus, the relative higher amount of silica in HuberCarb M3 is offset by its very fine particle size.
Morphology
Scanning electron microscopy of the silica fractions was informative. The micrographs (Figure 6) show both the 20 µm and 3 µm median D(50) acid insoluble fractions from the Marble Hill, GA, marble and the Marble Falls, TX, limestone. These clearly show the silica fraction from Georgia as well defined crystals which would be resistant to fracture. However, the Texas material appears porous and friable. It is simple to visualize how easily it would fracture into much smaller particles.
When all factors are taken into account, a statistical correlation to both the amount and particle size of the acid insoluble was developed (Figure 7). A good correlation with an R2 of 0.89 resulted.
Summary
In summary, when evaluating the abrasion characteristics of a naturally occurring mineral, one must take into consideration both the abrasivity of the mineral and of the impurities. Abrasion is a function of concentration, hardness and particle size of the material and/or impurity. A mineral with a high concentration of a hard impurity is not necessarily more abrasive. It is also highly dependent on the particle size of the impurity.
Abrasion characteristics of any material are a function of hardness, concentration and particle size.
Hardness is typically measured on the Mohs scale, where a diamond is 10 and talc is 1 (Figure 1). A higher Mohs hardness will scratch or abrade a material with a lower number. Calcium carbonate is a soft mineral with a Mohs hardness of 3.
Since ground calcium carbonate is a naturally mined material, impurities are present as they are in all ore bodies.
Almost all contain some dolomite (CaMg(CO3)2) with a Mohs of 4, while others contain mica (Mohs 3) or graphite (Mohs 1-2). None of these softer minerals are of concern for abrasion.
However, almost all GCC has some silica (Mohs 7) present. This may or may not present a concern depending on the concentration and its particle size, as we discuss in greater detail later.
Finally, the particle size of the GCC is an important parameter. Coarse-grit sandpaper is far more abrasive than fine-grit sandpaper, even though the grit is composed of the same material. Therefore, the median particle size and distribution of the GCC is critical, as is the particle size and distribution of the impurity.
There are several ways to deal with silica content in the production of GCC. Many ore bodies produce very white GCC but are high in silica. To address this, ores are first milled to a 20-45 µm particle size and undergo froth floatation. In this process, the GCC is slurried with water and a floatation chemical that binds to the silica. Air is bubbled through the mixture and the froth that forms carries the silica with it while the GCC sinks. While not 100% effective, it does remove the majority of the silica. Since the product is already wet, economies of ultra-fine grinding favor wet milling to the desired particle size.
If required by application, the GCC slurry is then dried, which is an energy intensive process.
An alternative production method is to start with an ore body with inherently low silica and dry mill it as found. This production method usually results in a slightly less bright product but has favorable economics.
Experimental
Various 3 µm GCC products were obtained from producing locations in the U.S. All products were mined from ore in that geographic region with the state identified. The exception is “CAR” which is Caribbean sourced ore that was milled at Huber’s GCC manufacturing facility in Quincy, IL. The products were measured for acid insolubles (primarily silicas and silicates). Most exhibited a low level of acid insolubles, whether wet ground (WG) or dry ground (DG), reflecting the underlying quality of the ore deposit or the efficiency of the wet floatation process (Figure 2). The obvious exception is the Hubercarb® M3 calcium carbonate produced in Marble Falls, TX. This sample was selected as a lot on the very high side of the acid insoluble range encountered during normal operations.
To examine the abrasivity of GCC products, the samples from various ore sources and milling methods throughout North America were tested. A Brabender® torque rheometer with #6 rotors was modified with a removable soft brass (D65 hardness) pin through one of the rotors (Photo 1).
A standard twin screw, tin-stabilized high fill (30 phr calcium carbonate) PVC pipe recipe was run for 16,000 cycles. To minimize the effects of PVC thermal degradation, 10 runs of 20 minutes each was required with 80 rpm rotor speed for each data point.
Examining the effect of GCC particle size with a fixed ore source, a comparison of Hubercarb G3 (3 µm median) with Hubercarb G260 (20 µm median) calcium carbonate from Marble Hill, Georgia, clearly shows that the larger particle size calcium carbonate abrades more than the finer one, as would be expected (Figure 3).
When all 3 µm products were compared for abrasion, there did not appear to be a wide data spread (Figure 4). It did not make a significant difference as to the ore source or the grinding method. The products from ore sources low in acid insolubles (like Hubercarb W3N calcium carbonate) tended to be on the low end of the range, and those with high acid insolubles (Hubercarb M3 calcium carbonate) tended to be on the high end of the range, but the Alabama wet ground material was also at the upper end of the narrow range despite a lower amount of acid insolubles.
The data points out an inconsistency because the expected result of the Hubercarb M3 calcium carbonate would be high abrasion. However, this study backed up experimental evidence where the Marble Falls, TX, Hubercarb M3 calcium carbonate has been used for decades without abrasion issues.
To further explore this inconsistency, a study of the particle sizes of the actual acid insoluble (silica fraction) was undertaken. The acid insoluble fraction of each product was measured for particle size (Figure 5). In all GCC products, except one, the median size of the acid insoluble fraction exceeded that of the GCC. This is logical since the GCC products are either dry or wet centrifugally classified and the density of silica (2.65 g/cc) is essentially the same as calcium carbonate (2.71 g/cc). Being the harder of the two materials, silica will be less likely to fracture and results in a slightly larger particle size.
The exception was the Hubercarb M3 calcium carbonate grade from Marble Falls, which had a median particle size of only 1.2 microns. As demonstrated earlier, a finer particle sized material of a given hardness results in lower abrasion. Thus, the relative higher amount of silica in HuberCarb M3 is offset by its very fine particle size.
Morphology
Scanning electron microscopy of the silica fractions was informative. The micrographs (Figure 6) show both the 20 µm and 3 µm median D(50) acid insoluble fractions from the Marble Hill, GA, marble and the Marble Falls, TX, limestone. These clearly show the silica fraction from Georgia as well defined crystals which would be resistant to fracture. However, the Texas material appears porous and friable. It is simple to visualize how easily it would fracture into much smaller particles.
When all factors are taken into account, a statistical correlation to both the amount and particle size of the acid insoluble was developed (Figure 7). A good correlation with an R2 of 0.89 resulted.
Summary
In summary, when evaluating the abrasion characteristics of a naturally occurring mineral, one must take into consideration both the abrasivity of the mineral and of the impurities. Abrasion is a function of concentration, hardness and particle size of the material and/or impurity. A mineral with a high concentration of a hard impurity is not necessarily more abrasive. It is also highly dependent on the particle size of the impurity.