Millimeter Wave Specific Challenges mmWave Design Challenges Informational

What is the effect of plastic mold compound on antenna performance in a millimeter wave package?

Plastic mold compound (epoxy molding compound, EMC) used to encapsulate mmWave ICs and antenna-in-package modules significantly affects the antenna performance: (1) Frequency shift: the mold compound has a dielectric constant of Dk = 3.0-5.0 (depending on formulation). When the mold compound covers the antenna (patched antenna with a mold compound superstrate): the effective dielectric constant of the antenna increases. The resonant frequency shifts downward: f_new ≈ f_air / sqrt(epsilon_eff), where epsilon_eff ≈ (1 + epsilon_mold) / 2 for a thin mold layer. For Dk_mold = 3.5: epsilon_eff ≈ 2.25. f_new = f_air / sqrt(2.25) = f_air / 1.5. This is a 33% frequency reduction. A patch designed for 28 GHz in air would resonate at approximately 18.7 GHz under mold compound. This shift must be pre-compensated in the antenna design. (2) Bandwidth: the mold compound loading reduces the Q of the antenna (the radiation impedance increases due to the dielectric loading). Lower Q = wider bandwidth. A patch with 5% bandwidth in air may have 8-10% bandwidth under mold compound. This is actually beneficial for wideband 5G signals (the 5G NR n257 band spans 26.5-29.5 GHz, requiring approximately 10% bandwidth). (3) Radiation efficiency: the mold compound has a loss tangent (tan_d = 0.004-0.020 at 28 GHz). The electromagnetic fields passing through the mold compound lose energy to dielectric heating. Efficiency reduction: 5-20% (depending on mold compound thickness and loss tangent). For tan_d = 0.008 and t_mold = 0.5 mm at 28 GHz: the efficiency drops from 85% (air) to approximately 70% (with mold). This 1 dB loss in efficiency directly reduces the EIRP and G/T of the module. (4) Surface wave excitation: the mold compound creates a dielectric slab on top of the antenna ground plane. At mmWave: the slab can support surface wave modes that trap energy and cause inter-element coupling in an array. Surface wave modes are excited when t_mold > lambda_0 / (4 × sqrt(Dk_mold - 1)). For Dk = 3.5 at 28 GHz: t_critical = 10.7 / (4 × sqrt(2.5)) = 1.7 mm. Below 1.7 mm: surface waves are suppressed. Above: surface waves degrade the pattern and reduce efficiency.

Category: Millimeter Wave Specific Challenges

Updated: April 2026

Product Tie-In: mmWave Components, Substrates, Packaging

Mold Compound Antenna Effects

The interaction between the mold compound and the antenna is one of the key co-design challenges in mmWave AiP modules. The antenna cannot be designed independently of the package.

Technical Considerations

(1) Standard EMC (Sumitomo EME-G790, Hitachi CEL-9220): Dk = 3.5-4.5 at 28 GHz. Loss tangent = 0.008-0.020. Formulated primarily for mechanical protection and reliability (CTE matching, moisture resistance). The RF properties are a secondary consideration. These compounds with tan_d > 0.01 cause > 15% efficiency reduction. (2) Low-loss EMC (developed specifically for mmWave AiP): Dk = 3.0-3.5 at 28 GHz. Loss tangent = 0.004-0.008. Filler particles (silica + specialty fillers) optimized for low dielectric loss. These are used by leading 5G module manufacturers (Qualcomm, Samsung). The lower loss tangent reduces the efficiency penalty to 5-10%. (3) Air-cavity packages: eliminate the mold compound over the antenna entirely by creating an air cavity above the antenna elements. The antenna radiates into air (no dielectric loading). The RFIC and feed network are still molded for protection. Advantage: highest antenna efficiency and easiest design (no mold compound compensation needed). Disadvantage: the air cavity is mechanically fragile (thin lid that must survive drop tests). More expensive (requires a lid bonding process). Used in: some high-performance AiP modules where maximum efficiency is critical. (4) Antenna-in-mold: the antenna is embedded within the mold compound (not on the top surface of the substrate). The mold compound surrounds the antenna on all sides. This provides the most uniform dielectric environment (easiest to simulate and compensate). The efficiency is lower (more mold in the field), but the design is more robust to variations in mold thickness.

Performance verification: confirm specifications against the application requirements before finalizing the design
Environmental factors: temperature range, humidity, and vibration affect long-term reliability and parameter drift
Cost vs. performance: evaluate whether the application demands premium components or standard commercial grades
Interface compatibility: verify impedance, connector type, and mechanical form factor match the system architecture
Margin allocation: include sufficient design margin to account for manufacturing tolerances and aging effects

Performance Analysis

(1) Pre-compensation: design the antenna to resonate at a higher frequency in air, so that after mold compound application, it resonates at the target frequency. The pre-compensation factor: f_design = f_target × sqrt(epsilon_eff_with_mold / epsilon_eff_air). For a patch on a substrate with Dk_sub = 3.5 and mold Dk = 3.5 covering the top: epsilon_eff_air ≈ (1 + 3.5)/2 = 2.25. epsilon_eff_mold ≈ (3.5 + 3.5)/2 = 3.5. f_design = f_target × sqrt(3.5/2.25) = f_target × 1.25. Design the patch to resonate 25% higher than the target frequency. (2) Stack-up optimization: add an air gap between the antenna and the mold compound. The air gap reduces the effective dielectric loading and improves the efficiency. Implementation: use a spacer or standoff structure above the antenna, with molding compound on top. The air gap thickness of 0.1-0.3 mm significantly reduces the Dk loading. (3) Antenna topology selection: some antenna types are less sensitive to mold compound: aperture-coupled patches (the coupling slot is on the ground plane below the antenna; the mold compound affects only the radiation side, not the feed side). Slot antennas (the radiation is from the slot in the ground plane; the mold compound above adds loading but the feed is below). Yagi-Uda or endfire arrays (the radiation is along the substrate surface, not through the mold compound; the mold compound effect is reduced).

Common Questions

Frequently Asked Questions

How do I test the antenna under mold compound?

Two approaches: (1) Pre-mold measurement: test the antenna on a bare module (before encapsulation) using an over-the-air (OTA) setup. Compare with EM simulation. Then mold the module and re-test. The pre-mold vs post-mold comparison directly shows the mold compound effect. (2) Simulation-based prediction: create an accurate EM model of the antenna with the mold compound (using measured Dk and loss tangent of the specific mold compound at 28 GHz). Validate the model with measurement of a test antenna. Once correlated: use the simulation model to predict the performance of the final design. (3) Material characterization: measure the mold compound Dk and loss tangent vs frequency using a split-cavity or microstrip ring resonator method. Do this on the actual production mold compound batch (properties can vary between batches). Feed the measured values into the antenna simulation.

Does the mold compound thickness variation matter?

Yes, significantly. Typical mold compound thickness variation: ±50-100 um (due to die height variation, substrate warpage, and mold flow non-uniformity). For a 0.5 mm nominal thickness at 28 GHz: ±0.1 mm variation changes the effective Dk loading by ±5-10%. This shifts the antenna resonant frequency by ±1-3%. For a 28 GHz patch with 10% bandwidth: a ±3% shift moves the center frequency by ±840 MHz (out of the 1 GHz passband margin). To mitigate: (1) Design the antenna with wider bandwidth (> 15%) to absorb the thickness variation. (2) Work with the mold vendor to tighten the thickness tolerance (±30-50 um is achievable with controlled molding). (3) Use a topology (like aperture-coupled patch) that is less sensitive to superstrate thickness variation.

Can I use conformal coating instead of mold compound?

Conformal coatings (parylene, silicone, or acrylic) are much thinner than mold compound (10-50 um vs 200-500 um). The thin layer has minimal dielectric loading effect on the antenna: Dk loading: negligible (the thin layer contributes < 1% to the effective dielectric constant). Efficiency impact: < 1% (very low loss through the thin layer). Disadvantages: conformal coating provides much less mechanical protection than mold compound (does not protect against drop shock or moisture as effectively). Not suitable as the primary protection for consumer devices (which must pass drop testing, humidity testing, and thermal shock). Conformal coating can be used for: indoor AP modules (less mechanical stress), military modules with a metal lid (the coating protects the die and bond wires, the lid provides mechanical protection), and laboratory prototypes.

Keep Reading