EVA 發泡為什麼會收縮？

Why Does EVA Foam Shrink?

PART 1

從 POE 熔點、DCP 交聯密度到冷卻定型的完整分析

A Complete Analysis: From POE Melting Point and DCP Crosslink Density to Cooling and Stabilization

並解析 IUS‑4065 如何改善超軟配方的二次收縮與尺寸穩定

Including How IUS-4065 Improves Secondary Shrinkage and Dimensional Stability in Ultra-Soft Formulations

一、先釐清：EVA 發泡的「收縮」有三個時間尺度

I. First, Clarify the Issue: EVA Foam "Shrinkage" Occurs on Three Time Scales

研發現場常把所有尺寸變小都稱為收縮，但不同時間點對應的主導機理並不相同。只有先分清發生時間，才有可能對症調整配方。

In R&D practice, every reduction in dimensions is often called shrinkage, but the dominant mechanism differs depending on when it occurs. Only by identifying the timing first can the formulation be adjusted appropriately.

圖 1｜EVA／POE 發泡收縮的六項耦合機理。

Figure 1 | Six coupled mechanisms of EVA/POE foam shrinkage.

二、泡孔為什麼在冷卻後自然想變小？

II. Why Do Foam Cells Naturally Tend to Contract After Cooling?

1. 泡孔氣體壓力隨溫度下降

1. Cell-Gas Pressure Falls as Temperature Decreases

發泡時泡孔內部處於高溫、高壓狀態。若先用理想氣體近似，P ≈ nRT/V；當溫度 T 下降、氣體莫耳數 n 又因擴散而減少，泡孔內壓便下降。若泡孔壁的彈性回復力與外界大氣壓大於內部支撐，泡孔體積 V 會縮小。

During foaming, the gas inside the cells is at high temperature and high pressure. Using the ideal-gas approximation, P ≈ nRT/V. As temperature T falls and the number of gas moles n also 

decreases through diffusion, the internal cell pressure drops. If the elastic restoring force of the cell walls plus external atmospheric pressure exceeds the internal support, cell volume V decreases.

2. 發泡氣體外逸與空氣回滲不同步

2. Blowing-Gas Out-Diffusion and Air Back-Diffusion Are Not Synchronized

ADC 分解氣體、CO₂ 或其他氣體對聚合物的溶解度與擴散係數，通常與空氣不同。冷卻初期若內部氣體較快逸出，而外界空氣尚未充分回滲，便形成暫時負壓；高倍率、薄泡孔壁及高溫停放會放大這個效應。

The solubility and diffusion coefficients of ADC decomposition gases, CO₂, and other gases in the polymer generally differ from those of air. If the internal gas escapes faster during initial cooling than 

outside air can diffuse back in, a temporary negative pressure develops. High expansion ratios, thin cell walls, and storage at elevated temperature amplify this effect.

3. 黏彈性回復與殘留應力

3. Viscoelastic Recovery and Residual Stress

發泡膨脹會使鏈段與泡孔壁產生取向。只要交聯網路或結晶支撐不足，這些鏈段就會朝較低自由能狀態回復，表現為長寬回縮或厚度下降。

Foam expansion orients molecular chains and cell walls. If the crosslinked network or crystalline support is insufficient, these chains recover toward a lower-free-energy state, appearing as contraction in length and width or a loss of thickness.

三、POE 熔點：為什麼材料越軟，二次收縮可能越大？

III. POE Melting Point: Why Can Softer Materials Exhibit Greater Secondary Shrinkage?

EVA 加入 POE，通常是為了降低硬度、改善柔韌性與回彈。然而許多超軟 POE 具有較低密度、較低結晶度及較低熔融區間。其晶區在常溫可充當「可逆的物理交聯點」；一旦貼合或後加工溫度接近熔融區間，晶區支撐減弱，交聯網路便必須獨自承擔泡孔與尺寸。

POE is usually added to EVA to reduce hardness and improve flexibility and resilience. However, many ultra-soft POEs have lower density, lower crystallinity, and a lower melting range. Their crystalline regions can act as "reversible physical crosslink points" at room temperature. Once 

lamination or post-processing temperatures approach the melting range, crystalline support weakens and the crosslinked network must carry the cell structure and maintain dimensions on its own.

1. 低熔點／低結晶度：有利柔軟與低溫觸感，但高溫下物理支撐較早消失。

Low melting point/low crystallinity: promotes softness and a soft touch at low temperature, but physical support disappears earlier at elevated temperature.

2. EVA 的 VA 含量：VA 越高通常結晶度與熔點下降，會與軟 POE 的低熔融區間疊加。

VA content of EVA: higher VA content generally lowers crystallinity and melting point, compounding the low melting range of soft POE.

3. POE 添加比例：比例越高，連續相形態、熔體黏度與 DCP 反應分配都可能改變。

POE loading: as the proportion rises, continuous-phase morphology, melt viscosity, and the distribution of DCP reactions may all change.

4. 二次加工溫度：即使設備設定僅 55–60°C，厚件中心與接觸面局部溫度、停留時間仍需實測。

Secondary-processing temperature: even if the equipment is set to only 55-60°C, the local temperature and dwell time at the core of a thick part and at contact surfaces must still be measured.

圖 2｜供應商資料顯示：DF 640、ENGAGE 8842 熔點低於 50°C；IUS‑4065 約 64°C。數值須以最新版 TDS、DSC 與批次實測覆核。

Figure 2 | Supplier data indicate melting points below 50°C for DF 640 and ENGAGE 8842, versus approximately 64°C for IUS-4065. Values must be verified against the latest TDS, DSC data, and batch testing.

四、DCP 交聯密度：太低會縮，太高也可能不穩

IV. DCP Crosslink Density: Too Low Causes Shrinkage, but Too High Can Also Be Unstable

DCP 受熱分解產生自由基，引發 EVA／POE 鏈段間形成共價交聯。交聯網路的任務，是在發泡劑大量產氣、泡孔快速成長時提供足夠熔體強度，並在冷卻後固定泡孔幾何。

When heated, DCP decomposes to generate free radicals, initiating covalent crosslinks between EVA/POE chains. The crosslinked network must provide sufficient melt strength while the blowing agent generates large volumes of gas and the cells grow rapidly, and it must fix the cell geometry after cooling.

圖 3｜有效交聯密度存在最佳窗口，並非 DCP 添加越多越穩定。

Figure 3 | Effective crosslink density has an optimum window; adding more DCP does not necessarily improve stability.

交聯不足時

When Crosslinking Is Insufficient

1. 泡孔壁承載力不足，容易併孔、破泡或開模回縮。

The cell walls lack load-bearing strength, making cell coalescence, rupture, or demolding shrink-back more likely.

2. 自由鏈段比例高，後熟化與二次加熱時黏彈回復較大。

A high proportion of free chains produces greater viscoelastic recovery during post-curing and secondary heating.

3. 低熔點 POE 稀釋 EVA 網路後，若 DCP 未重校正，問題更明顯。

The problem becomes more pronounced when low-melting POE dilutes the EVA network and DCP has not been recalibrated.

交聯過高時

When Crosslinking Is Excessive

1. 泡孔成長受限，發泡倍率下降，密度與硬度上升。

Cell growth is restricted, the expansion ratio falls, and density and hardness rise.

2. 鏈段延伸能力下降，撕裂、延伸率與動態疲勞可能受損。

Chain extensibility decreases, potentially impairing tear strength, elongation, and dynamic-fatigue performance.

3. 開模膨脹與內部應力增加；尺寸未必更好，反而可能翹曲。

Demolding expansion and internal stress increase; dimensions may not improve and warpage can instead occur.

在橡膠平台區，可用 G ≈ νeRT 理解剪切模數 G 與有效網鏈密度 νe 的關係。但對半結晶 EVA／POE 泡棉，晶區與填料也會貢獻模數，因此應以 MDR／流變曲線、凝膠率、平衡溶脹及最終泡孔形貌交叉判讀，而不能只以 DCP phr 代表交聯密度。

In the rubbery plateau region, G ≈ νeRT can be used to understand the relationship between shear modulus G and effective network-chain density νe. In semicrystalline EVA/POE foams, however, crystalline regions and fillers also contribute to the modulus. MDR/rheology curves, gel content, equilibrium swelling, and final cell morphology should therefore be interpreted together; DCP phr alone cannot represent crosslink density.