Skip to content

Effect of tool material on machining performance of soda-lime glass using EDM-fabricated microtools

FieldValue
JournalMicro & Nano Manufacturing
Published2026-05-12
AuthorsJae Yeon Kim, Seon Jong Kim, S W Lee, Chan Bin Oh, Bo Hyun Kim
DOI10.1007/s44374-026-00019-8

The PCD microtool produced the strongest soda-lime glass result. It limited maximum chipping to 9.39 µm and maintained average thrust near 0.14 N during drilling. Its mean critical depth of cut reached 2.9 µm, and it did not fracture after 10 holes.

PCD combined the lowest hole chipping with the highest critical depth of cut among the tested tools.

Tool materialMaximum edge chippingMean critical depth of cutDrilling outcome
WC-Co71.92 µm1.24 µmFractured during the first hole
PCBN65.29 µm1.69 µmFractured during the first hole
PCD9.39 µm2.9 µmNo fracture after 10 holes

Drilling used a feed rate of 1 µm/s, tool speed of 30,000 rpm, and peck cycle of 5 µm. Critical depth tests used a feed rate of 50 µm/s and the same rotational speed. Both operations used a cutting fluid mist.

The comparison covered complete WEDG-fabricated tools under the selected conditions. It did not isolate tool composition from the surface morphology produced during fabrication.

WEDG formed a different abrasive surface for each material

Section titled “WEDG formed a different abrasive surface for each material”

The authors fabricated each microtool by wire electrical discharge grinding, or WEDG. A 200 µm brass wire served as the electrode, and commercial EDM oil served as the dielectric fluid. The starting rods were 1 mm in diameter.

The control system advanced the tool at 2 µm/s during normal discharge. On a detected short circuit, it withdrew the tool by 1 µm or held it in place. The final cylindrical tools measured 100 µm in diameter and 350 µm in length. Fabrication used 100 V and a capacitance of 10,000 pF. A CCD camera tracked the process in real time.

MaterialEDM machining timeMaterial removal rateBottom-surface Sa
PCBN186 s7.09 × 10^-3 mm³/min0.33 µm
WC-Co215 s6.14 × 10^-3 mm³/min0.49 µm
PCD235 s5.62 × 10^-3 mm³/min0.61 µm

PCBN experienced fewer short circuits and maintained steadier tool position than WC-Co or PCD. It also had the shortest machining time and highest material removal rate. WC-Co and PCD required more setbacks during the discharge process.

The authors calculated MRR for a 300 µm diameter and 350 µm height reduced to 100 µm diameter. Discharge craters overlapped to form protrusions on the tool surface. These protrusions acted as abrasive grains during glass machining.

Measured hardness was 14 GPa for WC-Co, 35 to 45 GPa for PCBN, and 60 GPa for PCD. Their grain sizes were 0.2 µm, 1 to 2 µm, and about 1 µm, respectively. These differences affected how the protrusions wore and cut.

PCD limited chipping and thrust during microdrilling

Section titled “PCD limited chipping and thrust during microdrilling”

Microdrilling used 15 mm × 15 mm × 1 mm soda-lime glass substrates. Cutting fluid was supplied through minimum quantity lubrication at 4 bar and 6 ml/h. A tool dynamometer recorded thrust, while optical and confocal imaging measured hole geometry and surface condition. Edge chipping width was derived from the maximum and minimum hole diameters.

The PCD tool produced 9.39 µm maximum chipping. PCBN produced 65.29 µm chipping, and WC-Co produced 71.92 µm. PCD maintained a relatively low and stable thrust force, averaging approximately 0.14 N. PCBN reached a maximum of 5.07 N, while WC-Co reached 4.48 N.

The authors relate the high forces to dense fine abrasive grains on the smoother PCBN and WC-Co surfaces. They state that these grains increased rubbing and plowing loads. They also attribute the smaller PCD chipping to its relatively rough tool bottom.

PCD’s 60 GPa hardness and wear resistance helped its protrusions retain sharper geometry during drilling. The authors state that lower hardness accelerated abrasive wear and edge blunting in the WC-Co and PCBN tools. That geometry change increased machining loads and promoted brittle fracture.

The authors tilted the glass by 0.21° to find the critical depth of cut. A digital dial gauge verified the height variation before three grooves were machined along the incline. The first visible surface crack defined the critical depth.

PCD produced critical depths of 2.83, 3.09, and 2.78 µm. Its average was 2.9 µm. PCBN produced 1.88, 1.75, and 1.45 µm, averaging 1.69 µm. WC-Co produced 1.29, 1.11, and 1.33 µm, averaging 1.24 µm.

Each groove stayed crack-free below its critical depth, and cracks appeared above it. The authors attribute PCD’s higher limit to its hardness and resistance to abrasive wear. Wear flattens protrusions on softer tools and raises grinding force. That mechanism increased the risk of brittle fracture for WC-Co.

PCBN produced the lowest crack-free surface roughness

Section titled “PCBN produced the lowest crack-free surface roughness”

Surface roughness was measured after grinding at depths from 0.1 to 0.5 µm. These depths remained below each tool’s critical depth, so the measured surfaces were free of brittle fracture.

PCBN produced the lowest roughness, with Sa and Sq both equal to 0.02 µm. PCD reached Sa = 0.02 µm and Sq = 0.03 µm. WC-Co was rougher at Sa = 0.03 µm and Sq = 0.04 µm.

The authors link PCBN’s result to its dense field of fine protrusions. These grains removed a smaller layer thickness and produced a smoother surface. PCD had fewer protrusions, but its wear resistance kept them in cutting geometry.

Prior work cited by the paper reported Sa = 0.009 µm on glassy carbon with an EDMed PCD tool. The present PCD result was Sa = 0.02 µm on soda-lime glass. These values are not direct benchmarks because the workpiece and study conditions differ. The cited literature also included EDMed PCD tools for fused silica and an R-shaped PCD tool for sapphire.

WC-Co and PCBN tools fractured during the first drilled hole. PCD survived all 10 holes without fracture. After 10 holes, its bottom surface had flattened through wear. The measured Sa fell from 0.61 µm before machining to 0.38 µm after machining.

The wear data show that PCD’s stability did not eliminate wear. PCD also had the slowest EDM fabrication time and lowest MRR in the material comparison. Tool selection therefore involved a tradeoff between fabrication efficiency and machining durability.

The results support PCD when low chipping, stable thrust, and repeated drilling take priority. They do not establish PCD as the best choice for every glass grade or machining condition.

Testing covered one soda-lime substrate geometry and one principal feed and speed setting for each operation. The comparison also included the EDM-formed surface, not only nominal composition. Bulk hardness, grain density, binder, and protrusion morphology changed together.

Edge-radius reporting is inconsistent. Table 2 gives 6.12 µm for WC-Co, 3.12 µm for PCBN, and 7.22 µm for PCD. The conclusion instead gives 4.56, 3.12, and 6.21 µm, in the same order. The paper also reports average PCD thrust but not its maximum thrust. Critical depth was repeated three times for each tool, but no uncertainty interval or significance test is given.

The data favor PCBN when tool fabrication time and crack-free roughness take priority. It produced the fastest EDM cycle and lowest crack-free roughness, but its thrust peaked at 5.07 N. It fractured during the first drilled hole. WC-Co gave the largest chipping at 71.92 µm and also fractured during the first hole.

Within the tested conditions, PCD therefore suits soda-lime microstructures where entrance integrity and repeated hole production govern.

View Original Abstract

Abstract Glass materials are widely used in optical lenses, micro-electromechanical systems (MEMS), and micro-total analysis systems (µ-TAS) owing to their excellent optical transparency and chemical stability. However, their high hardness and intrinsic brittleness make conventional mechanical machining prone to brittle fracture, resulting in cracks, surface defects, and edge chipping. Ductile mode machining of glass can be achieved using microtools when appropriate tool materials and machining conditions are carefully controlled. In this study, microtools made of tungsten carbide (WC-Co), polycrystalline cubic boron nitride (PCBN), and polycrystalline diamond (PCD) were fabricated using electrical discharge machining (EDM), and their tool fabrication characteristics were systematically investigated. The machining performance of these microtools was evaluated through microhole drilling experiments on soda-lime glass to assess chipping behavior and thrust force, as well as microgroove machining on an inclined substrate to determine the critical depth of cut (CDC). The experimental results revealed that the PCBN microtool exhibited the most stable fabrication quality, whereas the PCD microtool demonstrated superior machining performance. In particular, the PCD tool effectively suppressed brittle fracture, resulting in the lowest chipping and the highest CDC among the tested tools. These findings indicate that PCD is the most effective tool material for achieving stable ductile mode machining of glass materials.