Comparative Experiment of Dry Etching and Wet Etching: A Technical Evaluation of Etch Performance an

This technical article presents a systematic comparative experiment between dry etching and wet etching, two fundamental pattern transfer processes in semiconductor and microfabrication manufacturing. The evaluation covers etch rate, uniformity, anisotropy, selectivity, critical dimension control, and surface quality, supported by standardized characterization methods performed in accredited third-party laboratories. Engineers, process developers, and quality managers will gain practical guidance for selecting the optimal etching strategy for their device requirements.

Technical Overview

Etching is a core process step in microelectronics, MEMS, optoelectronics, and advanced packaging, responsible for selectively removing material to transfer lithographic patterns into underlying films or substrates. The two dominant process families are wet etching, which relies on liquid chemical etchants, and dry etching, which utilizes plasmas or reactive gas-phase chemistries. A rigorous comparative experiment between the two methods quantifies their respective strengths in etch rate, directional control, selectivity, uniformity, and surface integrity.

Wet etching typically proceeds through purely chemical reactions between the etchant solution and the target material. For amorphous or polycrystalline films, wet etching is generally isotropic, etching equally in all directions, which limits its applicability to feature sizes above roughly one to three micrometers. However, for single-crystal silicon, alkaline etchants such as potassium hydroxide (KOH) or tetramethylammonium hydroxide (TMAH) exhibit crystallographic anisotropy, attacking the (110) planes significantly faster than the (111) planes. This enables precise, self-limiting three-dimensional structures widely used in MEMS fabrication.

Dry etching encompasses reactive ion etching (RIE), inductively coupled plasma (RIE/ICP), deep reactive ion etching (DRIE), and physical techniques such as ion milling. These processes combine chemical radicals with directional ion bombardment, producing anisotropic profiles with vertical sidewalls suitable for sub-micron and nanoscale features. The trade-offs include potentially lower throughput per wafer, higher equipment cost, ion-induced surface damage, and possible polymer or residue formation.

A well-designed comparative experiment controls key variables including material stack, masking scheme, etch endpoint, and metrology method, allowing direct, quantitative comparison under ISO/IEC 17025-aligned laboratory conditions and relevant SEMI, ASTM, and JIS test guidelines.

Testing Samples

Bare Wafers and Blanket Dielectric Films

Unpatterned silicon wafers coated with thermally grown silicon dioxide (SiO2), low-pressure chemical vapor deposited silicon nitride (Si3N4), or polysilicon films serve as baseline samples for etch rate, uniformity, and selectivity measurements. Blanket films eliminate masking effects and allow large-area gravimetric, profilometric, and ellipsometric thickness determination before and after etching.

Patterned Wafers and Photolithographic Test Structures

Photoresist-patterned wafers with line-and-space gratings, contact hole arrays, and dense/isolated feature mixes are used to evaluate critical dimension (CD) bias, sidewall profile angle, pattern fidelity, and CD uniformity. These samples directly reveal the anisotropy differences between plasma-based dry etching and solution-based wet etching.

MEMS, Compound Semiconductor, and Specialty Substrates

Silicon-on-insulator (SOI) substrates, bulk silicon for DRIE through-wafer etching, III-V compound semiconductors such as gallium arsenide (GaAs) and gallium nitride (GaN), metal films including aluminum and copper, and glass or quartz substrates are tested to assess process compatibility across device platforms.

Testing Items

The comparative experiment evaluates the following performance metrics for both wet and dry etching processes:

• Etch rate (angstrom per minute or micron per minute) for each material layer
• Within-wafer and wafer-to-wafer etch uniformity, typically expressed as a percentage of the mean
• Etch selectivity to masking photoresist and to underlying stop layers such as SiO2 or Si3N4
• Anisotropy index, sidewall profile angle, and undercut distance obtained from cross-sectional imaging
• Critical dimension bias and post-etch CD uniformity across pattern density variations
• Surface roughness of etched floors and sidewalls, quantified as Ra or RMS values
• Surface residues, polymer deposition, micromasking defects, and metallic contamination
• Etch-induced damage, lattice defects, and hydrogen or fluorine passivation of silicon surfaces
• Bubble adhesion, wetting behavior, and drying defects specific to wet processing
• Time-dependent etch depth linearity and process window stability

Testing Methods

Etch Rate, Uniformity, and Selectivity Measurement

Film thickness before and after etching is measured using spectroscopic ellipsometry for transparent films, reflectometry for opaque layers, and stylus profilometry step-height measurement at nine or more points per wafer following the standard radial mapping convention. Selectivity is calculated as the ratio of target film etch rate to mask or stop layer etch rate, derived from parallel test wafers. For wet bench evaluation, gravimetric weight-loss measurement per ASTM procedures supplements optical thickness data, particularly for thick films and bulk micromachining where material removal exceeds several micrometers.

Profile, Anisotropy, and CD Evaluation

Cross-sectional imaging using scanning electron microscopy (SEM) provides sidewall angle, undercut, trench depth, and profile shape for both etching methods. Critical dimension measurements are performed with CD-SEM or optical overlay metrology before and after etch to quantify CD bias and line-edge roughness. For deep silicon structures produced by DRIE, etch depth, scalloping amplitude from the Bosch process, and verticality are extracted from dedicated cleaved samples.

Surface Quality and Contamination Analysis

Atomic force microscopy (AFM) measures nanoscale surface roughness of etched surfaces, enabling comparison between plasma-damaged surfaces and chemically etched surfaces. X-ray photoelectron spectroscopy (XPS) and energy-dispersive X-ray spectroscopy identify residual fluorocarbon polymers, oxidation states, and chemical termination after each process. Time-of-flight secondary ion mass spectrometry detects trace metallic contamination introduced by wet chemistries or plasma chamber components. Contact angle goniometry characterizes surface hydrophilicity or hydrophobicity, which affects subsequent cleaning, adhesion, and bonding steps. Optical inspection and defect review microscopy map micromasking sites, residues, and pattern collapse across full wafers.

Experimental Design and Data Analysis

The comparison follows a structured design of experiments with fixed input variables, including film stack, mask layout, target depth, and endpoint criteria. Each process is characterized over at least three etch times to confirm depth linearity, with replicate wafers to quantify run-to-run repeatability. Results are reported with measurement uncertainty budgets traceable to ISO/IEC 17025 requirements, providing statistically defensible process selection data.

Applications

The comparative data support etching process selection across a broad range of industries and device categories:

• CMOS and advanced logic manufacturing, where dry plasma etching is indispensable for sub-micron vertical profiles and self-aligned structures
• MEMS sensors and actuators, where KOH/TMAH wet etching and DRIE both produce high-aspect-ratio silicon structures for accelerometers, gyroscopes, and microphones
• Power semiconductor devices, including trench MOSFETs and IGBTs requiring deep, low-damage silicon etching
• RF and photonic devices on GaAs, GaN, and indium phosphide substrates, where plasma chemistries such as chlorine- or methane-based etchants control fine features
• Display and micro-LED fabrication requiring precise pattern transfer with minimal undercut
• Advanced packaging and wafer-level integration, where wet spray etching and isotropic dry release etches remove sacrificial layers
• Surface preparation, residue removal, and wafer reclaim, where wet cleans complement dry plasma processing

FAQ

What is the fundamental difference between dry etching and wet etching?

Wet etching uses liquid chemical solutions that remove material through purely chemical reactions, typically isotropic in amorphous films, while dry etching uses plasma or gas-phase chemistries combining chemical radicals with directional ion bombardment, enabling anisotropic, vertical profiles at sub-micron dimensions.

Which method provides better anisotropy and CD control?

Dry etching provides superior anisotropy because ion directionality constrains material removal to the vertical axis, achieving sidewall angles approaching 90 degrees. Wet etching of non-crystalline films undercuts the mask laterally, limiting CD control, although crystallographic wet etching of silicon offers excellent self-limiting geometry for MEMS structures.

Which process is faster and more cost-effective?

Wet etching generally offers higher throughput, lower equipment cost, and excellent selectivity for many material systems, making it economical for large features, film removal, and surface conditioning. Dry etching carries higher capital and per-wafer costs but is mandatory where dimensional precision and profile control dominate.

How is etch rate measured in the comparative experiment?

Eth rate is determined by measuring film thickness or step height before and after a timed etch using spectroscopic ellipsometry, reflectometry, or stylus profilometry, supplemented by gravimetric analysis for thick-film and bulk etching. Multiple points per wafer establish uniformity maps.

What sample types are accepted for third-party comparative testing?

Accredited third-party laboratories accept bare and patterned wafers from 100 mm to 300 mm, wafer fragments, III-V substrates, SOI stacks, and packaged process coupons, provided mask layout, film stack specifications, and target etch depth documentation accompany the samples.

Which standards govern etching process characterization?

Evaluation protocols reference SEMI film and wafer specifications, ASTM thickness and gravimetric measurement standards, ISO surface chemical analysis standards, and JIS semiconductor metrology standards, ensuring results are comparable across laboratories and process lines.

Can dry and wet etching be combined in a single process flow?

Yes, hybrid flows are common. Dry plasma etching defines fine vertical features, followed by wet cleaning or isotropic wet etch steps that remove residues, polymers, or sacrificial layers. The comparative experiment data help engineers position each technique optimally within the integrated process sequence.