Water-soluble salts or electrolytes are just some of the mineral contaminants found in everyday tap water used by foundries across the U.S for their green sand systems. Bentonite is the main bonding compound of green sand systems, and for it to have quality bonding, it first must be activated with the use of an agent. A good activating agent must fulfill three requirements:  

Be distinctly soluble in water, having a solubility of at least 1 g/100 g of water.

Be composed of anions, which react with cations (Na+) to give insoluble compounds.

Supply small, univalent cations for ion exchange.

Previous research studied the effect of these electrolytic salts on the physical characteristics of green sand, in particular wet tensile strength. Wet tensile strength testing helps determine the tensile strength of the green sand’s condensation zone after the mold has been poured. Previous research has shown that an increase in the electrolytic salts, found in well water, led to a decrease in wet tensile strength when compared to samples of green sand containing distilled water, where the wet tensile was not affected.

Past research led to the question of whether these electrolytes are causing the decreasing strength within the clay-water relationship. This possibly could be due to the anionic (positively charged ions) nature of sodium bentonite clay and its necessary cationic (negatively charged ions) exchange between the electrolytic salts within the water. Anions and cations react together to form neutralized compounds. An example of this reaction is that of Sodium (Na+) and Chlorine (Cl-) where the two react together forming your basic table salt (NaCl).  

Past research conducted on the effect of individual electrolytic salts on the wet tensile strength of green sand showed salts had an immediate effect on the wet tensile strength, with the wet tensile strength decreasing initially. However, as time went on, the wet tensile strength was determined to increase. This was determined to be due to the presence of the particular electrolytes within the system and not the actual concentration.

Recently, trials were conducted at the University of Northern Iowa Metal Casting Center (Cedar Falls, Iowa) to evaluate the effect of contaminants commonly found in tap and well water in the U.S. on green sand properties. Sixteen water samples were obtained from different regions of the country and a series of experiments were designed to evaluate the effect of the measured contaminants.

The water samples received were sent to an external lab to test for contaminants using an Inductively Coupled Plasma Mass Spectrometer (ICP-MS).

The lab determined sodium, chlorine, magnesium and calcium were the major contaminants detected above the method detection limit (MDL). Figure 1 shows the average contaminants detected in the water samples, sorted by region. From the results, the southwest region was observed to have higher concentrations of sodium, chlorine and calcium, when compared to the other regions. The magnesium concentrations detected were lower when compared to the other three contaminants, though still over the MDL. Other contaminants, such as phosphorous and potassium were detected below MDL and hence, were not significant.

Low and high levels of the four contaminants were selected from the results shown in Figure 2. Nine experiments were conducted, including one with distilled water, which served as the control sample. The water samples were created by doping distilled water with the contaminants.

For each experiment, a new batch of green sand was prepared using 7% Western Bentonite, based on sand weight, and 20% seacoal, based on clay weight. A three-screen 55 GFN Silica sand was used as the base aggregate.

Class 30 iron plate castings were poured to evaluate and compare the performance properties of the nine green sand mixtures. The casting design offered a sand:metal ratio of 5:1.  

pH of Doped Water Samples
The pH of the doped water samples was measured. Adding the elements at their low levels slightly increases the pH, when compared to the control sample. From the pH of samples 2, 3 and 4, it can be seen adding higher levels of Mg, Ca and Cl2 considerably increases the pH. However, the addition of sodium can be observed to have a larger effect on the pH, when compared to the other contaminants. All water samples with high sodium levels were observed to have pH values of approximately 11.65-11.89.

Green Sand Properties
The green compression strength results are shown in Figure 2. Similar strengths were measured for all samples, with the exception of sample 3. Sample 3 was recorded to have an average green strength of 17 psi. Sample 3 can be observed to have a high chlorine level and the other contaminants at low levels.
The wet tensile strength results are shown in Figure 3. A trend similar to green compression strength can be observed. Wet tensile strength results were measured to be similar for all samples, except samples 3 and 4. Samples 3 and 4 were observed to have wet tensile strengths of 0.303 N/cm2 and 0.330 N/cm2. The other samples were measured to have wet tensile strengths in the range of 0.380-0.410 N/cm2.

Figure 4 shows the green shear strength results measured for all samples. The control sample was measured to have higher green shear strength when compared to the other samples, at 6.33 psi. Similar to the trend seen in green compression and wet tensile strength results, sample 3 was measured to have comparatively lower green shear strengths, at 3.817 psi. Samples 4, 7 and 8 were also observed to have lower shear strengths, when compared to the control sample, ranging from 4.5 – 4.7 psi.

The permeability results are shown in Figure 5. The control sample can be observed to have significantly higher permeability, at 206.3, when compared to the other samples. Samples 2 and 3 were measured to have AFS permeability of ~125 while the other samples were measured to have permeability in the range of 138 – 150.

Figure 6 shows the loss on ignition (LOI) results for all samples. With the exception of sample 3, the other samples can be observed to have similar LOI, ranging from 1.07 – 1.14%. Sample 3 was measured to have a slightly higher LOI of 1.37%.

The bulk density results for all samples are shown in Figure 8. Similar results were observed for all samples, ranging from 93.4 lbs./ft3 to 94.6 lbs./ft3.
While the research indicated there were some detrimental effects of the contaminants on green sand properties, especially in green compression, wet tensile and green shear strength, little to no differences were observed in the test castings.

Some contaminants also might have positive side effects, such as sodium or calcium on improving green sand properties. However, further research with a larger number of replicates will be required to conclusively prove the preliminary results.   

This article is based on the paper “Effect of Water Quality on Green Sand Properties” (18-009) presented at the 122nd Metalcasting Congress. The study was an AFS-funded research project.

Click here to see this story as it appears in the June 2018 issue of Modern Casting