Study of the effect of green sand properties on thermal conductivity variation of sand mould

Thermal conductivity of moulding sands is very poorly known. In addition, the study of thermal conductivity variation of moulding sand as a function of its properties is a crucial step in the prediction of moulding defects. So, this study investigates the effects of moisture content, compactibility and permeability of green sand on its thermal conductivity variation, with a specific focus on the combination of bentonite, sand and water for thermal characterization of the transient hot-bridge (THB) method. Three levels of active clay content were selected: 8.1%, 8.98% and 10.03%. For each level of active clay, different mixes were made by varying the amount of added water. Compactibility, permeability and thermal conductivity tests were carried out for each combination of active clay and moisture contents. The experimental results show that thermal conductivity increases with water content and compactibility for different active clay contents, whereas thermal conductivity decreases with permeability, and this variation is highly significant especially for 8.07% of active clay content.


Introduction
Sand casting is one of the most flexible manufacturing processes, as it is used for most metals and alloys, especially those with high melting temperatures such as iron, copper and nickel alloys.It remains the most important method for producing castings, particularly when the required parts are large in number and have a complex shape.
Modern foundries use numerical simulation, which has now become an indispensable tool for understanding all the details of the foundry, aiming to predict casting defects.Indeed, all input parameters must be correctly defined to adequately reproduce the real casting, including the chemical composition of the metal, the casting temperature, the mould material, the thermo-physical properties of the mould and the cast metal, and boundary conditions such as heat transfer coefficients [1].The determination of these properties allows better control of sand defects [2].On the other hand, it is well known that the amount and rate of heat transfer between the molten metal and the sand and at room temperature determine the structure and properties of the casting.Both parameters depend on the thermo-physical properties of the molten alloy and on the mould.The thermo-physical properties of sand moulds are significantly influenced by the characteristics of the sand.However, these relationships are often not well understood.Additionally, the software available typically utilizes average values found in literature, which can result in inaccuracies in calculations.It is important to note that contemporary foundry engineering requires the incorporation of numerical simulations for heat and mass transfer processes in the design of casting and solidification procedures [3].One of the key input parameters for these numerical algorithms is the thermo-physical properties of the mould materials.
S. Guharaja et al. [4] have analysed different parsameters that affect the green sand casting process: Moisture content, green strength, mould hardness and permeability.Using Taguchi's method, they optimized these different parameters by studying their effects on casting defects.Dhruval Patel et al. [5] have also used Taguchi parametric design to optimize the mould properties by studying the relation between these properties and the input parameters: shape and drain size, clay content, moisture and clay content.Prafulla Kumar Sahoo et al. [6] have studied the effect of different additives, i.e. coal dust, coir fibre and bamboo dust on the characteristics of green sand mould such as green and dry compressive strength, green and dry shear strength and permeability.Archana Kumari et al. [7] have developed a predictive models to find optimal parameters of sand casting process using Taguchi method.They have studied the effect of moisture, clay content, coal dust quantity and the mulling time on moulding properties such as permeability and green compressive strength.Finally, they have explored an optimization technic based on multi-objective genetic algorithm.In order to analyse and reduce green casting defects, Shohanuzzaman Shohan and Fardim Sufian [8] have developed a new method based on Taguchi method and solid casting simulation technique.They have obtained as results products with far fewer defects and reduced surface roughness.G. Soleniki et al. [9] have developed an instrument to measure the thermal conductivity of moulding sand as a function of temperature up to 300 °C in order to study the effect of sand granulometry on solidification process.S. I. Bakhtiyarov et al. [10] have identified bulk density and thermal expansion of zircon sand, olivine sand and silica sand cores using a the technic of computer-controlled dual-pushrod dilatometer.Ewa Majchrzak and Bohdan Mochnacki [11] have identified the thermal properties of the mould-metal system based on the inverse problem.Won-Jin Cho et al. [12] have developed an empirical model for the identification of the thermal conductivity and have investigated the effect of solid fraction, water content and dry density on thermal conductivity of bentonite and the sand-bentonite mixture.Min Wang et al. [13] have determined the effective conductivity of the compacted sand-bentonite additive mixture by proposing simplifications at the microstructure of the mixture such as shape, size and orientation of pores.R. R. Kundu et al. [14] have studied the effects of bentonite, water and curing time on the properties of the green sand mould, such as permeability, compactibility and compressive strength.
The aim of this section is to explore how variations in bentonite and water impact the properties of moulding sand.Despite the availability of new technologies for metal casting, sand casting remains prevalent due to its cost-effectiveness in raw materials, the versatility in casting sizes and compositions and the potential for recycling casting sand.This casting method is highly flexible as it accommodates a wide range of metals and alloys with high melting points like iron, copper and nickel.Extensive research has been conducted to determine the optimal values for parameters influencing the casting process, with the goal of enhancing casting quality through various methodologies.The optimal configurations for these parameters signify the most efficient levels for each factor, optimizing the process response.

Sample preparation
Green sand is a mixture of silica grains, clay and water.The percentage of each component varies according to the type of casting alloys.In this study, after drying, the sand was mixed with increasing water content.We used a laboratory mixer which is specific to these tests.For each water content value, the desired characteristics and properties are performed.These tests will be repeated for each active clay content.We selected three levels of active clay content: 8.1%, 8.98% and 10.03%.A wooden mould was made for preparing samples with dimensions of 120 × 60 × 20 mm3.The samples were fabricated using the existing rammer in the TUNICAST sand laboratory [15] (Fig. 1).

Experimental procedure
The sand is mixed with proportional quantities of bentonite and water according to the AB standard.Then compactibility, compressive strength, and permeability are measured.Samples of the THB method with dimensions 110 × 60 × 20 mm 3 are manufactured using a mould specified for this type of sample.

The transient hot-bridge method
A thermoelectric sensor coupled to a current source and placed between two identical samples is used in the experiment (see Fig. 2).The THB sensor consists of a nickel sheet that is 7.5 µm thick and is arranged in a circuit with four parallel tandem strips (see Fig. 3).Two strips are placed at a distance from D1 in the centre, and two more are placed at the end of the sensor on either side (at a distance from D2).Every strip is considered a resistor in electrical terms.With equal resistances, the eight strips connected to make the combined circuit constitute a Wheatstone bridge.This nickel sheet, which measures 120 × 60 × 0.055 mm 3 , is placed between two sheets of kapton.

Analytical model
The following formula is used to determine the temperature differential that is produced between the sample's outside and centre: where T inn is the temperature variation on the inner ribbon (K) and T out is the temperature variation on the outer ribbon (K).
where Di is the ribbon width (m).
The temperature difference is then as follows [16]: (1) 2 Experimental setup for the THB method

Fig. 3 THB sensor circuit
The signal of the THB sensor is proportional to ΔT [17]: With: U s : Sensor output signal (V).R eff : Effective sensor resistance at measurement temperature (Ω).I B : Sensor current (A).α: Temperature coefficient (K −1 ).The basis of this approach is an analysis of the slope's change over time.In fact, this slope is the temperature derivative with respect to the logarithm of time, and it corresponds to the following: With: m(t) is the dimensionless slope.λ: Thermal conductivity of sample (W m −1 K −1 ).q: Linear heat flux density (W m −1 ).The maximal slope is detected at t = t max : With.
From Eq. (2.13), we can identify the thermal conductivity:

Results and discussion
The moulding sand's thermal conductivity has a particular effect on the cooling rate of a casting.The resultant microstructure is greatly influenced not only by other elements (such as feeding technology, degassing, melting cleaning and grain refining) but also by the cooling rate.Higher cooling rates typically result in finer grains and smaller secondary dendritic arms in the microstructure, which enhances its mechanical characteristics.
The characterization results are presented in Table 1.In Fig. 4, thermal conductivity was plotted as a function of sand characteristics.It was observed that conductivity increased with water content for different active clay contents (Fig. 4a) [18].Conversely, thermal conductivity decreased with increasing active clay content (AA).The same direction of variation was noted as a function of compactibility (AS) (Fig. 4b).The (5) variation in conductivity as a function of permeability (P) was highly significant (Fig. 4c).Specifically, for AA = 8.07%, conductivity increased significantly from 0.7 (± 0.003) to 0.8 (± 0.003) and means to 14.28% with a slight decrease in permeability of 3.76%.For AA = 10.03%,thermal conductivity increases by 20% with a low permeability decrease of 8.14%.Heat transfer in the sand occurs mainly through contact between solid particles, while the addition of water increases the contact between sand grains [19].For samples with  low moisture, the water usually forms bridges between the particles.As the water content increases, the particles start to stick together due to the surface tension pulling the water bridges towards the particles.The formation of water droplets between the particles enhances the interaction between them, which then improves the heat conduction through the material [20].Furthermore, bentonite has the ability to expand, which may help it absorb water more easily.This water then solidifies into a sticky substance that fills the pores around the sand grains [21].Sand-bentonite mixture contains three different types of pores-inter-aggregate, intra-aggregate and inter-laminar as Fig. 5 illustrates [22].When the mixture (sand + bentonite + additives) is mixed with water, the water first fills the intra-aggregate pores and causes hydration and swelling of the bentonite particles due to its high-water absorption, and subsequently, the bentonite grains evolve into a gel filling the inter-aggregate pores.This step is dominated by the intra-aggregate regulatory aspiration.However, with increasing water content, the inter-aggregate pores are progressively filled with water, and the intra-aggregate pores play a dominant role in aspiration, increasing thermal conductivity and the heat removal rate.The thermal conductivity of the moulding sand influences the cooling rate of a moulded part.The resultant microstructure is significantly influenced by the cooling rate in addition to other parameters (e.g.feeding technique and degassing).Higher cooling rates usually create microstructures with finer grains and smaller secondary dendritic arms, which enhance the mechanical characteristics of the material.

Conclusion
The present work deals with thermal conductivity variation of moulding sand as function of its properties: compactibility and permeability for different combination of active clay and water contents.Three levels of active clay content were selected, and for each level, three percentages of water were added.Permeability, compactibility and thermal conductivity were measured for every combination.The THB method was used for thermal conductivity measurement.
It can be concluded through the experimental measurements.Thermal conductivity increases with water content for different active clay contents.Conversely, thermal conductivity decreased with increasing active clay content.This variation is explained by the formation of water droplets between particles which Fig. 5 Schematic representation of an unsaturated mixture of bentonite and sand [22] reinforces the interaction between them, improving heat conduction through the material.Sand grains adhere to each other in the presence of bentonite and water, generating voids between the sand grains coated with bentonite.By applying a compaction force, the total void is reduced.In this way, an increase in bonding helps to achieve greater compactibility and, in turn, better thermal conduction.
As function of permeability, thermal conductivity is inversely proportional specifically; for AA = 8.07%, it increases significantly by14.28% with a slight decrease in permeability of 3.76%.For a constant percentage of bentonite, the percentage of water is not sufficient to bind all the bentonite.As the percentage of water increases, so does binding.More binding means that the grains move closer together, and inter-aggregate voids are reduced.As a result, the surface area from which gases escape is reduced, and permeability decreases, increasing heat conduction between the grains.
These findings contribute to understanding how material composition impacts heat transfer and cooling rates in sand casting.This ultimately affects the microstructure and mechanical properties of the final cast metal parts.Further research in this area could lead to improved methodologies for optimizing casting processes and enhancing product quality in metal casting industries.

Fig. 4
Fig. 4 Thermal conductivity variation as a function of sand characteristics: a as a function of H, b as a function of AS and c as a function of P

Table 1
Characterization results