In my extensive experience with investment casting, commonly known as lost wax casting, I have observed that the sodium silicate-based ceramic shell process has been utilized in our industry for over five decades. Despite numerous improvements in shell-making materials, hardeners, and process controls over the years, making it a relatively mature technology, its inherent shortcomings compared to silica sol shell systems remain pronounced. The high-temperature strength of a sodium silicate shell is only a fraction, sometimes even one-tenth or less, of that of a silica sol shell. This deficiency necessitates the application of more layers, resulting in a shell thickness typically 2 to 3 times greater than that of its silica sol counterpart. The increased thickness directly translates to higher consumption of raw and auxiliary materials per ton of cast product, elevated energy consumption during firing, greater shell weight, operational difficulties in shell building, and ultimately, lower production efficiency in lost wax casting operations.
Although the sodium silicate shell process suffers from lower strength and poorer environmental performance due to effluent concerns, cost constraints ensure its persistence in the foundry landscape for the foreseeable future. Therefore, the primary developmental directions for this process are to further enhance its strength and reduce pollutant emissions. In this context, my research focused specifically on strengthening the backup layer coatings of sodium silicate shells. The objective was to achieve the high-temperature strength level of silica sol shells without major alterations to the existing shell-making process and without increasing the total material cost. This was to be accomplished by introducing specific strengthening agents into the coatings, thereby enabling the goal of reducing the number of shell layers and overall weight in lost wax casting.
The core challenge in lost wax casting with sodium silicate binders lies in the residual sodium oxide (Na₂O) content. Even after proper chemical hardening, a sodium silicate shell retains approximately 0.3% to 0.5% Na₂O, which is more than 2.5 times the amount found in silica sol shells. This free Na₂O drastically reduces the viscosity of the glassy liquid phase formed at high temperatures during metal pouring, severely compromising the shell’s hot strength. The high-temperature strength ($\sigma_{HT}$) of a ceramic shell can be conceptually related to the viscosity ($\eta$) of its liquid phase and the primary crystalline phase, as shown in the following relationship:
$$ \sigma_{HT} \propto \eta \cdot f(\text{Crystalline Phase}) $$
where a higher viscosity generally correlates with greater strength. The free Na₂O acts as a flux, significantly lowering $\eta$. Therefore, my research strategy was to enhance the liquid phase viscosity by introducing agents that could consume or neutralize the free Na₂O or otherwise reinforce the silicate network.
Review of Domestic Research in Lost Wax Casting Shells
Prior investigations within the field of lost wax casting in our region have concentrated on strengthening sodium silicate shells through modifications to the backup layer flour, the type of hardener, and process control parameters. Since the 1970s, researchers from institutions like Beijing Yongding Machinery Factory pioneered the use of kaolinitic refractory clay and bauxite flour to replace quartz flour in backup layer coatings, which substantially increased shell strength, enabling single-shell pouring without sand support. Subsequent studies, such as those by Zhao et al., successfully developed a thin-walled, high-strength sodium silicate shell with fewer layers using Guiyang bauxite flour (with 85.77% Al₂O₃) as the refractory material and ammonium chloride as the hardener. For a 8.1 kg casting, only 5 coating layers were required. Zhang et al. incorporated rare earth oxide La₂O₃ into a refractory clay coating strewed with quartz sand, reporting a 40% to 44% improvement in high-temperature strength, allowing a reduction from 6 to 5 layers. Research by Zhou et al. examined the influence of Al₂O₃ content in refractory flour and sodium silicate modulus, finding that an Al₂O₃ content between 55% and 65% yielded an optimal high-temperature bending strength of 3.09–3.60 MPa at 900°C, an 84.8% increase. Hu demonstrated that using bauxite flour with 80–83% Al₂O₃ could improve shell strength, reduce run-out defects, and allow the elimination of one coating layer. The common thread in these lost wax casting studies was the introduction of strengthening materials like Al₂O₃ or La₂O₃ to enhance the viscosity of the high-temperature liquid phase, thereby improving performance.
Experimental Design for Shell Strengthening
My experimental approach was grounded in principles from both the glass industry and prior lost wax casting research. Materials that increase glass viscosity, such as Al₂O₃, SiO₂, and ZrO₂, were considered. While B₂O₃ can lower high-temperature viscosity, borate ions can repair broken siloxane bonds, promoting a denser network structure. For this lost wax casting study, I selected active aluminum-based (denoted as AlnM₃), silicon-based (SinM₄), and boron-based (BnM₃) materials as potential strengthening agents. Their selection was based on dispersibility, compatibility with sodium silicate binder, and cost. Two strengthening schemes were devised and validated.
Scheme 1: Single-Agent Strengthening. Individually adding AlnM₃, SinM₄, or BnM₃ to the backup layer coating at specific ratios, informed by related industry studies and cost.
Scheme 2: Composite Strengthening. Combining AlnM₃, SinM₄, and BnM₃ in defined proportions to form a composite strengthening agent added to the coating.
The shell-making process for test specimens adhered to the standard JB/T 2980.2-1999, measuring the bending strength at 980°C. The process mirrored production conditions. The binder was sodium silicate with a density of 1.30 g/cm³. Hardening was done in an aluminum chloride solution for 20 minutes, followed by 20 minutes of air drying. Firing was at 980°C for 1 hour. For each test condition, five specimens were evaluated, and the average strength was calculated. The key shell-making parameters for the specimens are summarized in Table 1.
| Coating Layer | Coating Type | Coating Viscosity (s) | Stucco Material | Notes |
|---|---|---|---|---|
| Face Coat | Silica Sol + Quartz Flour | 55-65 | 40-100 mesh Quartz Sand | – |
| 2nd Layer | Sodium Silicate + Mullite Flour | 30-35 | 30-60 mesh Quartz Sand | Pre-wetted with sodium silicate before coating |
| 3rd Layer | Sodium Silicate + Calcined Kaolin Flour | 30-35 | 16-30 mesh Quartz Sand | – |
| 4th Layer | Sodium Silicate + Calcined Kaolin Flour | 30-35 | 16-30 mesh Quartz Sand | – |
| Seal Coat | Sodium Silicate + Calcined Kaolin Flour | 30-35 | No Stucco | – |
The calcined kaolin flour used had an Al₂O₃ content of 40-45%. For comparison, control specimens were made using the standard sodium silicate process (no strengthening agent) and a full silica sol process. Their average 980°C bending strengths were 2.95 MPa and 9.12 MPa, respectively, highlighting the significant gap in lost wax casting shell performance.
Results and Analysis of Strengthening Trials
Single-Agent Strengthening
The results for specimens with individual strengthening agents are presented in Table 2. All additives provided some improvement in high-temperature strength. The most effective single agent was 3% AlnM₃, which increased strength by 53.2% compared to the unmodified sodium silicate shell baseline. The active AlnM₃ readily reacts with free Na₂O at high temperatures to form nepheline phase (Na₂O·Al₂O₃·2SiO₂), which has a refractoriness around 1526°C, thereby improving strength. The reaction can be represented as:
$$ \text{Na}_2\text{O} + \text{Al}_2\text{O}_3 + 2\text{SiO}_2 \xrightarrow{\text{High T}} \text{Na}_2\text{O}\cdot\text{Al}_2\text{O}_3\cdot2\text{SiO}_2 $$
The ultra-fine SinM₄ powder contributes by increasing the flour-to-binder ratio and promoting sintering. BnM₃ acts as a modifier for sodium silicate, helping to repair the siloxane network. To further elucidate the role of AlnM₃ in consuming Na₂O, I prepared two sets of specimens that were not chemically hardened (containing ~1.5-2% Na₂O). One set included 4% AlnM₃, the other had no additive. The load-displacement curves during bending tests at 980°C showed a dramatic difference: the AlnM₃-reinforced specimen failed at a displacement of 0.22 mm, indicating higher rigidity, while the untreated specimen exhibited a large displacement of 2.56 mm, signaling a weak, low-viscosity liquid phase predominant structure.
| Test No. | AlnM₃ Addition (wt.%) | SinM₄ Addition (wt.%) | BnM₃ Addition (wt.%) | Bending Strength at 980°C (MPa) | Notes |
|---|---|---|---|---|---|
| 1-1 (Baseline) | 0 | 0 | 0 | 2.78 | Standard sodium silicate shell |
| 1-2 | 0.1 | 0 | 0 | 3.27 | – |
| 1-3 | 2.0 | 0 | 0 | 4.05 | – |
| 1-4 | 3.0 | 0 | 0 | 4.26 | AlnM₃ is nano-sized |
| 1-5 | 0 | 0.2 | 0 | 3.52 | – |
| 1-6 | 0 | 0.5 | 0 | 3.34 | SinM₄ is sub-nano sized |
| 1-7 | 0 | 0 | 0.2 | 3.95 | – |
| 1-8 | 0 | 0 | 0.4 | 3.39 | – |
Composite Strengthening
Building on the single-agent results, I designed a composite strengthening experiment using a three-factor, two-level orthogonal array to optimize the combination of AlnM₃, SinM₄, and BnM₃. The factor levels are defined in Table 3.
| Factor | Level 1 | Level 2 | Description |
|---|---|---|---|
| A: AlnM₃ Addition (wt.%) | 2.0 | 3.0 | Nano-sized active alumina-based material |
| B: SinM₄ Addition (wt.%) | 0.2 | 0.5 | Sub-nano sized silica-based material |
| C: BnM₃ Addition (wt.%) | 0.2 | 0.4 | Boron-based modifier |
The experimental matrix and the resulting average bending strengths at 980°C are presented in Table 4. The strength values achieved through composite strengthening were remarkable, exceeding 9.6 MPa and reaching up to 11.23 MPa. This represents more than a threefold increase over the standard sodium silicate shell strength and matches or surpasses the 9.12 MPa benchmark of the silica sol shell in lost wax casting. Analyzing the orthogonal array, the primary factor influencing strength was C (BnM₃ addition), followed by B (SinM₄ addition), and then A (AlnM₃ addition). The optimal combination from a pure strength perspective was A₂B₁C₁ (3.0% AlnM₃, 0.2% SinM₄, 0.2% BnM₃). However, considering economic efficiency and the minor difference contributed by factor A, the combination A₁B₁C₁ (2.0% AlnM₃, 0.2% SinM₄, 0.2% BnM₃) was deemed the most practical and cost-effective for industrial application in lost wax casting.
| Experiment Run | A: AlnM₃ (wt.%) | B: SinM₄ (wt.%) | C: BnM₃ (wt.%) | Bending Strength at 980°C (MPa) |
|---|---|---|---|---|
| 1 | 1 (2.0%) | 1 (0.2%) | 1 (0.2%) | 11.23 |
| 2 | 1 (2.0%) | 2 (0.5%) | 2 (0.4%) | 9.60 |
| 3 | 2 (3.0%) | 1 (0.2%) | 2 (0.4%) | 10.37 |
| 4 | 2 (3.0%) | 2 (0.5%) | 1 (0.2%) | 10.96 |
Mean Analysis:
Mean for A at Level 1: (11.23 + 9.60)/2 = 10.42 MPa
Mean for A at Level 2: (10.37 + 10.96)/2 = 10.67 MPa
Range for A: 10.67 – 10.42 = 0.25 MPa
Mean for B at Level 1: (11.23 + 10.37)/2 = 10.80 MPa
Mean for B at Level 2: (9.60 + 10.96)/2 = 10.28 MPa
Range for B: 10.80 – 10.28 = 0.52 MPa
Mean for C at Level 1: (11.23 + 10.96)/2 = 11.10 MPa
Mean for C at Level 2: (9.60 + 10.37)/2 = 9.99 MPa
Range for C: 11.10 – 9.99 = 1.11 MPa
Primary Factor Order: C > B > A. Optimal for strength: A₂B₁C₁. Recommended for balance: A₁B₁C₁.
The synergistic effect in composite strengthening can be modeled conceptually. The overall strengthening effect ($S$) might be expressed as a function of the individual contributions:
$$ S = k_A \cdot [\text{AlnM}_3] + k_B \cdot [\text{SinM}_4] + k_C \cdot [\text{BnM}_3] + k_{AB} \cdot ([\text{AlnM}_3][\text{SinM}_4]) + \ldots $$
where $k$ coefficients represent the efficacy of each agent and their interactions. The experimental data suggests significant positive interaction, particularly between the Na₂O-consuming AlnM₃ and the network-modifying BnM₃.
Critical Process Control Parameters for Strengthened Shells in Lost Wax Casting
Implementing this composite strengthening technology in mass production for lost wax casting revealed specific process sensitivities that must be controlled to ensure consistent shell quality.
1. Basicity and Temperature of the Hardener Solution: The facility uses crystalline aluminum chloride as the hardener. For environmental reasons, the bath is used continuously for years with only replenishment, leading to sodium salt accumulation. If the basicity (often measured as alkalinity or pH-related parameter) of the bath reaches the upper process limit, its hardening capability deteriorates severely, risking shell cracking during dewaxing. The strengthening agents, particularly the fine powders, can slightly retard the hardening reaction. Therefore, maintaining the hardener basicity below 15% is crucial. In winter, bath temperature must also be controlled to ensure adequate reaction kinetics for proper shell hardening in lost wax casting.
2. Flour-to-Binder Ratio of the Backup Layer Coating: The quality of commercially supplied refractory flour can vary due to differences in ore type, calcination temperature, and raw flour content. This variability can affect the optimal flour-to-binder (F:B) ratio of the coating. Merely controlling coating viscosity is insufficient. For quartz-sand-based shells in lost wax casting, if the F:B ratio falls below 1.1, the effectiveness of the strengthening agents diminishes. The relationship between coating consistency ($C_c$), often related to viscosity, and the F:B ratio can be described empirically:
$$ C_c \approx \alpha \cdot (\text{F:B}) + \beta $$
where $\alpha$ and $\beta$ are constants dependent on the flour properties. Maintaining an F:B ratio between 1.1 and 1.3 is recommended for optimal strengthened shell performance.
3. Shell Pouring Temperature: Strength tests indicated that at high temperatures (e.g., 980°C), the strength difference between quartz-sand and mullite-sand shells is minimal. However, upon cooling, the difference becomes pronounced due to the cristobalite phase transformations in quartz. When the shell temperature drops below approximately 300°C (near the α-cristobalite to β-cristobalite inversion at 270°C), the strength of quartz-based shells may become insufficient to withstand the metallostatic pressure during pouring in lost wax casting. Therefore, it is essential to ensure the shell mold temperature remains at or above 300°C during metal pouring to leverage its designed high-temperature strength.
Industrial Application and Results in Lost Wax Casting
The implementation of the composite strengthening agent with the optimal formula (2.0% AlnM₃ + 0.2% SinM₄ + 0.2% BnM₃) into production for lost wax casting has yielded transformative results. The number of backup layers was successfully reduced from the original 4–6 layers down to just 2–3 layers. This directly led to a shell weight reduction of 30% to 40%. The decrease in material consumption per casting and the reduction in manual labor for coating and stuccoing operations significantly improved production efficiency in lost wax casting by over 25%. Despite achieving a high-temperature strength of 11 MPa, comparable to silica sol shells, the residual strength of these strengthened sodium silicate shells after casting remains very low (around 0.2 MPa), ensuring easy knockout and eliminating any sand removal difficulties commonly associated with some high-strength binder systems in lost wax casting.
The economic and operational benefits are substantial for foundries engaged in lost wax casting. The reduction in refractory flour, sand, binder, and energy for firing thicker shells translates to lower direct costs. The lighter shells are easier to handle, reducing ergonomic strain and potential for damage during handling. Furthermore, the shorter shell-building cycle (fewer layers to apply and dry) increases the throughput of the shell-making department, a critical bottleneck in many lost wax casting shops.
Conclusions
Through systematic research focused on the lost wax casting process, I have developed and validated an effective composite strengthening technology for sodium silicate-based ceramic shells. The key findings are:
- Strength Achievement: The synergistic combination of active aluminum-based (AlnM₃), silicon-based (SinM₄), and boron-based (BnM₃) strengthening agents in the backup layer coating dramatically enhances the high-temperature performance of quartz-sand stuccoed sodium silicate shells. The bending strength at 980°C can reach 11 MPa, equaling that of silica sol shells used in lost wax casting.
- Process Control Imperatives: Stable production of these high-strength, thin-walled shells in lost wax casting requires strict control of three key parameters:
- Hardener bath basicity must be maintained below 15%.
- The flour-to-binder ratio of the backup coating should be kept between 1.1 and 1.3.
- Shell molds should be poured at a temperature not lower than 300°C to maintain adequate strength during metal filling.
- Significant Application Benefits: The practical implementation of this technology in lost wax casting enables a reduction of 2 to 3 layers in the shell build, leading to a 30–40% decrease in shell weight and raw material consumption. Shell-making productivity improves by more than 25%, offering a compelling path to enhance the competitiveness of the sodium silicate shell process within the broader lost wax casting industry.
This research underscores the potential for continuous innovation in established foundry processes like lost wax casting. By addressing the fundamental issue of free Na₂O through tailored material science, the performance gap between cost-effective sodium silicate shells and premium silica sol shells can be effectively closed, providing foundries with a more versatile and economical option for producing high-integrity castings via the lost wax casting method.