proceedings of the institution of mechanical engineers, part b:...

httppibsagepubcomManufacture

Engineers Part B Journal of Engineering Proceedings of the Institution of Mechanical

httppibsagepubcomcontent222135The online version of this article can be found at

DOI 10124309544054JEM840

2008 222 35Proceedings of the Institution of Mechanical Engineers Part B Journal of Engineering ManufactureP V Petkov S S Dimov R M Minev and D T Pham

Laser milling Pulse duration effects on surface integrity

Published by

httpwwwsagepublicationscom

On behalf of

Institution of Mechanical Engineers

can be found atManufactureProceedings of the Institution of Mechanical Engineers Part B Journal of EngineeringAdditional services and information for

httppibsagepubcomcgialertsEmail Alerts

httppibsagepubcomsubscriptionsSubscriptions

httpwwwsagepubcomjournalsReprintsnavReprints

httpwwwsagepubcomjournalsPermissionsnavPermissions

httppibsagepubcomcontent222135refshtmlCitations

What is This

- Jan 1 2008Version of Record gtgt

at Cardiff University on April 4 2012pibsagepubcomDownloaded from

Laser milling pulse durationeffects on surface integrityP V Petkov S S Dimov R M Minev and D T Pham

Manufacturing Engineering Centre Cardiff University Cardiff UK

The manuscript was received on 15 February 2007 and was accepted after revision for publication on 24 July 2007

DOI 10124309544054JEM840

Abstract Laser milling of engineering materials is a viable alternative to conventional methodsfor machining complex microcomponents The laser source employed to perform such micro-structuring has a direct impact on achievable surface integrity At the same time the trade-offsbetween high removal rates and the resulting surface integrity should be taken into accountwhen selecting the most appropriate ablation regime for performing laser milling In this paperthe effects of pulse duration on surface quality and material microstructure are investigatedwhen ablating a material commonly used for manufacturing microtooling inserts For bothmicro- and nanosecond laser regimes it was estimated that the heat-affected zone on the pro-cessed surface is within 50mm When performing ultra-short pulsed laser ablation the effectsof heat transfer are not as evident as they are after processing with longer laser pulse durationsAlthough some heat is dissipated into the bulk when working in pico- and femtosecond regimesit is not sufficient to trigger significant structural changes

Keywords laser micromachining micromachining laser pulse duration

1 INTRODUCTION

The laser milling of engineering materials has becomea viable alternative to conventional methods for pro-ducing tooling inserts for microreplication or formachining microfeatures in components By applyingthis technologymaterial is removed in a layer-by-layerfashion to produce the desired three-dimensionalstructures Direct interfaces to three-dimensionalcomputer-aided design (CAD) modelling packagesexist to assist in the machining of complex free-formsurfaces Being a non-contact material removal pro-cess some of the main advantages of laser milling arethat the process does not suffer from any problemsassociated with tool breakage does not require inclu-sion of collision checking routines in machining pro-grammes and it is easy to access areas that are verydeep in cavities Also if ultra-short pulsed lasers areutilized almost any material can be machined andthe thermal load is significantly reduced resulting inhigh surface integrity

Laser radiation can be delivered to the workpiece inan ordered sequence of pulses with a predeterminedpulse length (duration) and repetition rate (frequency)This allows the accumulated energy to be released inrelatively short time intervals which is a prerequisitefor the formation of extremely high peak powersAdditionally the laser beam can be focused on a spotwith very small dimensions from submicrometre to50mm which results in a significant energy density(fluence) and intensity (power density) in the spotarea Therefore an extremely high density can beachieved in the laserndashmaterial interaction zone thatcould not be achieved by any conventional machiningtechnology This explains the capability of lasermillingto process materials that are difficult to machine [1]In addition such a high fluence is very importantwhen producing microstructures that require a highsurface finish and hence atom cluster and atomicprocessing In particular to carry out machining atsuch a scale it is necessary to remove material withunits from 1ndash100nm to 001ndash1nm with a correspond-ing increase of the specific processing energy from103 ndash 104 (Jcm3) to 105 ndash 106 (Jcm3) [2]which is attain-able with ultra-short pulse laser ablation

The laser source employed has a direct impact onachievable surface integrity In recent years a wide

Corresponding author Manufacturing Engineering Centre

Cardiff University Queenrsquos Building The Parade Newport

Road Cardiff CF24 3AA UK email PetkovPVcfacuk

SPECIAL ISSUE PAPER 35

JEM840 IMechE 2008 Proc IMechE Vol 222 Part B J Engineering Manufacture


range of laser sources has become commercially avail-able Laser pulse durations may vary from microse-conds to a few femtoseconds [3] In this researchthe effects of pulse duration on surface integrity areinvestigated when ablating a material commonlyused for manufacturing microtooling inserts In parti-cular a method for analysing the effects of pulse dura-tion on surface integrity is proposed that takes intoaccount not only the resulting surface roughness afterlaser milling but also the changes of the materialmicrostructure as a result of the exercised thermalload by different laser sources Thus instead of asses-sing only qualitatively the resulting surface integritythrough this method it is possible to conduct a quan-titative analysis of the material grain morphologyand thus to judge more precisely the heat penetrationdepth as a function of pulse duration

The paper starts with a discussion of the physicalphenomena that take place during laser milling withdifferent pulse durations Then the set-ups and themethod used to carry out this experimental study areoutlined and the results of the metallographic andsurface profile analyses are provided Finally conclu-sions are made on the effects of pulse duration onthe resulting surface integrity

2 MATERIAL REMOVAL MECHANISMS

When pulsed laser machining is performed the actualprocess of ablating a material takes place within thepulse Several mechanisms exist for material removaldepending on the laser pulse duration and somematerial specific time parameters [4ndash6] The followingimportant material-dependent time constants inregard to the substrate material and the laser sourceare considered

(a) te the electron cooling time(b) ti the lattice heating time(c) tL the laser pulse duration

As a rule teti and for most materials ti is in thepicosecond range According to the laser pulse dura-tion three different ablation regimes can be defined

(a) femtosecond tLltte ltti(b) picosecond telttL ltti(c) nanosecond and longer pulses te lttilttL

The femtosecond and picosecond ablation mechan-isms are similar and are illustrated in Fig 1

In these two regimes the laser radiation isinitially absorbed locally in the electron systembecause the ions are heavier and cannot follow thefast oscillations of the electromagnetic field [7] Thecollisions between the energetic electrons and thenthe electrons and the atomic lattice result in their

thermalization However only a small fraction ofenergy can be transmitted by each electronndashlatticecollision due to the large mass difference betweenelectrons and ions Thus a multiple of electronndashphonon relaxation time has to pass to achievethermodynamic equilibrium between the electron sys-tem and the atomic lattice Therefore if tL is muchshorter than the time required to reach this thermo-dynamic equilibrium the ablation process can beregarded as a direct solidndashvapour transition (sub-limation) with negligible thermal conduction intothe substrate and almost no heat-affected zone(HAZ) [8ndash11] In particular each pulse creates somelsquosolid plasmarsquo a substance consisting of loosely boundions and electrons which leaves the substrate afterthe end of the pulse by expanding in a highly ionizedstate The electrons are lighter and are the first toleave the substrate followed by the ions The latterare all positively charged and repel one another whichfacilitates their removal from the substrate Duringthis expansion the solid plasma takes away most ofthe energy and consequently the thermal load onthe substrate is very low In the picosecond regimein spite of the formation of amolten zone and the exis-tence of some heat conduction the dominant removalmechanism is still a solidndashvapour transition [7]

In general to perform atom cluster and atomicprocessing with pulsed lasers tL should be shorterthan the time necessary to achieve thermodynamicequilibrium between the electron system and theatomic lattice For example for metals with strongelectronndashphonon coupling such as steel tL should

Laser pulse

Lens

No recastlayer

No surfacedebris

No melt zone

Hot denseionelectronsubstance(ie plasma)

Minimal heat transferto the surrounding material

No microcracks

No shock wave

Plasma plume

No damagecaused toadjacentstructures

Fig 1 Femto- and picosecond pulsed laser ablation

36 P V Petkov S S Dimov R M Minev and D T Pham

Proc IMechE Vol 222 Part B J Engineering Manufacture JEM840 IMechE 2008


be in the range from 3 to 5ps while for aluminiumand copper materials with weak coupling it needsto be one or two orders of magnitude higher [7]A further reduction of tL would not bring additionalbenefits in terms of a material machining responseNon-linear effects due to interactions between theultra-short laser pulse and atmospheric gas in thefocal region occur that lead to a wavefront disruptionof the beam profile distortion and increased beamdivergence In particular these are the side effectswhen performing laser ablation in the femtosecondregime [7]

For nanosecond and longer pulses the processconditions are summarized in Fig 2 In this casethe absorbed energy from the laser pulse melts thematerial and heats it to a temperature at which theatoms gain sufficient energy to enter into a gaseousstate There is enough time for a thermal wave to pro-pagate into the material Evaporation occurs from theliquid state of the material The molten material ispartially ejected from the cavity by the vapour andplasma pressure but a part of it remains near the sur-face held by surface tension forces After the end of apulse the heat quickly dissipates into the bulk of thematerial and a recast layer is formed [12]

Secondary effects of machining regimes with nano-second and longer pulses are HAZ a recast layermicrocracks shock wave surface damage and debrisfrom ejected material Additionally the vaporizedmaterial forms plasma almost at the start of thepulse and it is sustained throughout it Due to theplasma shielding effect (absorption and defocusing

of the pulse energy) a higher irradiance (fluence) isrequired for deeper penetration [12]

In the case of ultra-short pulsed laser ablation theplasma is formed after the end of the pulse whichmeans that the shielding effect is avoided It is impor-tant to note that for femto- and picosecond regimesthe fluence should only vary within predefined limitsfor different materials Exceeding these limits canlead to undesirable secondary effects [12]

For optimal machining results a proper matchbetween the laser source and the material should beachieved Generally higher absorption efficiencyleads to a more effective laser milling process Anumber of ways exist to increase laser absorptivityin particular creating an appropriate surface finishprior to laser milling or applying a suitable surfacecoating Laser ablation efficiency can also beincreased by performing the laser milling process atelevated temperatures or under water [13]

3 EXPERIMENTAL SET-UPS AND METHOD

A series of experiments was conducted to assess theimpact of the laser pulse duration on surface integrityof a substrate Two main effects were studied in par-ticular changes in material microstructure and sur-face quality by carrying out metallographic andsurface profile analyses In particular to estimatethe thermal load exercised on the substrate the pro-cessed areas were analysed for phase transformationsand changes in the grain structure

Four different laser milling systems were employedhaving femto- pico- nano- and microsecond pulsedurations respectively to ablate a field with dimen-sions 1 middot 1mm The characteristics of the laser sourcesemployed in this experimental study are shown inTable 1 The experiments were conducted at four dif-ferent sites within a day on the same workpiece andincluded the following

1 Familiarization with the material The four part-ner organizations involved in this study did nothave experience with the selected material forthe trials Thus some test features were producedto find the best processing window within theavailable timeframe It should be stressed thatthese may not be the optimal parameters butthe effects on surface integrity of the substratecould be considered representative for perform-ing ablation in these four different regimes

2 Machining of a series of 1middot 1mm fields A few teststructures were produced on each system byvarying laser milling parameters within the iden-tified processing window However the availabletime did not allow the analysis of the machined

Laser pulse

Lens

Recastlayer

Surfacedebris

Melt zone

Heat affectedzone

Microcracks

Ejected moltenmaterial

Damagedadjacentstructure

Heat transfer

Shock wave

Plasma plume

Fig 2 Nanosecond and longer pulse laser ablation

Laser milling pulse duration effects on surface integrity 37



surfaces to be carried out immediately after thetests Therefore as was already indicated theobtained surface roughness may not be the bestachievable with these four laser sources Forfurther analysis in this research the fields withthe best surface roughness for each of the fourstudied ablation regimes were selected

The experiments were conducted on a BS EN ISO4957ndashX40CrMoV5-1 tool steel workpiece (035C1Si 5Cr 14Mo 1V) This material wasselected because it is commonly used to manufacturetooling inserts for microinjection moulding and

hot embossing and thus to endure many thermalcycles The material properties of the X40CrMoV5-1tool steel are provided in Table 2 The workpieceused in this experimental study was polished beforeit was processed with the four different laser sourcesin succession

After completing the machining all fields on theworkpiece were cleaned in an ultrasonic bath withlight degreaser to preserve the topology of the result-ing surfaces The fields were inspected with a whitelight profiling microscope before dicing the substratein pieces Then for a better edge retention the pieceswere embedded in an epoxy-based resin

Table 1 Laser sources characteristics

Laser type Laser sourceLaser processparameters

Roughness achievedRa (mm)

A Femtosecond laser sourceSP Hurricane(amplified Tisapphire)

Wavelengthfrac14 800nmRepeat ratefrac14 5 kHz

Powerfrac14 20mWScanning speedfrac14 100mmminNumber of passesfrac14 4Stepfrac14 001 mmFluencefrac14 025 Jcm2

035

Pulsefrac14 130 fsBeam diameterfrac14 6mmFocal distancefrac14 75mmSpot sizefrac14 15mm

B Picosecond laser sourceStacatto (Lumera)

Wavelengthfrac14 1064nmRepeat ratefrac14 50 kHzPulsefrac14 12ps

Powerfrac14 100mWScanning speedfrac14 100mmsNumber of passesfrac14 10Stepfrac14 0002mmFluencefrac14 113 Jcm2

029

Beam diameterfrac14 2mmFocal distancefrac14 100mmSpot sizefrac14 15mm

C Nanosecond CVL MOPA(Oxford lasers)

Powerfrac14 10WScanning speedfrac14 100mms

086

Wavelengthfrac14 511nm Number of passesfrac14 10Repeat ratefrac14 10 kHz Stepfrac14 001mmPulsefrac14 17ns Fluencefrac14 2 Jcm2


D Microsecond Foba (Lasertech)Wavelengthfrac14 1064nmRepeat ratefrac14 305 kHzPulsefrac14 10msBeam diameterfrac14 8mmFocal distancefrac14 100mmSpot sizefrac14 45 mm

Powerfrac14 52WScanning speedfrac14 305mmsNumber of passesfrac14 10Stepfrac14 001mmFluencefrac14 18 Jcm2

218

Table 2 Material properties of BH13

C Si Mn Cr Mo V

038 100 040 500 130 100

At temperature

Physical properties 200 C 400 C 600 C

Density (kgdm3) 775 770 765Coefficient of thermal expansion(per C from 0 C)

119middot 106 124middot 106 128middot 106

Thermal conductivity (calcm s C) 600middot 103 624middot 103 636middot 103

Modulus of elasticity (Nmm2) 184000 175000 154000




Finally the specimens were polished and developedwith picral (recommended for structures consisting offerrite and carbides) and natal (the most commonetchant for revealing alpha grain boundaries of Fecarbon and alloy steels) reagents in order to analysethe material microstructure In particular this wasdone to highlight the boundaries of the ferrite grains(a-phase) and carbide sets An analysis of the mater-ial microstructure was carried out employing theBuehlerndashOmnimet software [14] In Fig 3 examplesof micrographs depicting the grain structure of theanalysed area and a printout showing the number ofgrains and their maximum minimum and mean dia-meters are provided

The changes in the grain structure were the maincriterion for estimating the heat-affected zones Thematerial microstructure of the workpiece was uniformbefore performing any processing After the ablation agrain refinement was observed in the area surround-ing the machined surface Such changes are the resultof the thermal wave propagation into the substratewhich is immediately followed by a quick coolingdown at the end of the pulse In particular to analyse

the affected regions in this experimental study theywere split into three zones taking into account theextent of these changes Zone 1 covers the area wherethe most of the heat was absorbed and therefore thechanges are clearly visible In zone 2 some changescan still be observed but at the same time there is asteady decrease of the thermal impact Finally inzone 3 the material microstructure can be consideredto be the same as in non-processed areas of thesubstrate

To make the comparison of microstructurechanges easier it was assumed that these three char-acteristic zones cover the same area in depth formicro- and nanosecond and for pico- and femtose-cond ablation regimes respectively In particularthe three zones were set to be equal for

(a) long pulsed lasers (micro- and nanosecondregimes) zone 1 below 15mm in depth zone 2from 15 to 50mm and zone 3 above 50mm

(b) short pulsed lasers (pico- and femtosecondregimes) zone 1 below 10mm zone 2 from 10 to30mm and zone 3 above 30mm

(a) the material microstructure resulting from ablationwith the ms laser under polarized light

(c) a printout of grain structure analysis

(b) the highlighted grain structure employing theBuehler-Omnimet software

Fig 3 Typical results




A quantitative assessment of the microstructurechanges was carried out by calculating the numberof grains in each zone and their maximum mini-mum and mean diameters with the BuehlerndashOmnimet software

4 RESULTS

41 Surface roughness

The surface maps of fields laser milled with differentpulse durations were studied in order to understandthe effects of the four ablation mechanisms on theresulting surface roughness As was mentioned insection 3 the fields with the best surface roughnessfor each of the four studied ablation regimes wereselected for further analysis In Fig 4 the three-dimensional surface maps of the four studied fieldsare presented In addition surface profiles were cre-ated to analyse the effects of pulse duration on theresulting surface topography They are shown in Fig 5

All roughness measurements were taken using awhite light profiling microscope The size of thescanned areas was chosen according to ISO 42881996

and ISO 115621996 [15] The parameter used toevaluate the surface roughness was the arithmeticmean roughness (Ra) because relative heights inmicrotopographies are more representative especiallywhen measuring flat surfaces

In Fig 6 the surface profiles of the fields machinedwith the ps and fs laser sources are superimposed fordirect comparison

42 Material microstructure

Micrographic pictures were obtained in polarizedlight in order to enhance the appearance of the crys-tallographically identical ferrite grains The area andequivalent circular diameter of each individual grainwere calculated using the BuehlerndashOmnimet soft-ware as was explained in section 3 Based on thesedata it was possible to estimate the average grainsizes and thus to have a quantitative measure forassessing the thermal effects on the processed sur-faces and ultimately to judge the thermal load exer-cised on the substrate in each ablation regime Aqualitative analysis of the resulting grain structureafter performing laser milling with long and shortpulsed lasers is provided in Fig 7

(a) fs pulse duration (b) ps pulse duration

(c) ns pulse duration (d) micros pulse duration

Fig 4 Three-dimensional surface maps




The changes of the material microstructures inthe three characteristic zones after processing indifferent ablation regimes can be summarized asfollows

1 Microsecond pulse duration Figure 8(a) showsthe studied three characteristic zones In zone 1(0ndash15mm) the mean diameter of the grains wasestimated to be approximately 13mm and themaximum diameter measured was 75mm Inzone 2 (15ndash50mm) the mean diameter was equal

to 25mm while the maximum diameter was115mm Finally above 50mm no changes in thegrain structure were identified The mean andmaximum diameters were 79 and 36mm respec-tively the same as in unprocessed areas on thesubstrate

2 Nanosecond pulse duration The three studiedzones in the micrograph are shown in Fig 8(b)In zone 1 the estimated mean diameter of thegrains was approximately 145mmwhile the max-imum diameter measured was 98mm In zone 2from 15 to 50mm the mean and maximum dia-meters were 28 and 155mm respectively Againabove 50mm there were no more changes in thegrain structure The mean and maximum dia-meters were 78 and 33mm

3 Picosecond pulse duration In Fig 9(a) a micro-graph depicting the three characteristic zonesused for analysing the thermal load of short pulsedlasers is provided The results obtained showedthat mean and maximum diameters of the grainsin zones 1 and 2 were 23 and 41mm and 93and 215mm correspondingly No changes in thegrain sizes were observed in zone 3 above30mm In particular the measured mean andmaximum diameters were equal to 82 and31mm which were the same as those for unpro-cessed areas of the substrate

4 Femtosecond pulse duration The analysis of thematerial microstrucrure was carried out again bysplitting the micrograph in three zones as shownin Fig 9(b) In zone 1 the estimated mean dia-meter of the grains was approximately 16mmwhile the maximum diameter measured was82mm In zone 2 from 10 to 30mm the meanand maximum diameters were 4 and 175mmrespectively Again above 30mm from the ablatedsurface there were no changes in the grain struc-ture and the mean and maximum diameterswere 82 and 31mm correspondingly

5 DISCUSSION


As expected the roughness of the field processedwith the ms laser was the highest Ra 218mm Thesurface profile after machining with the ns lasersource was significantly better in particular theroughness was reduced to Ra 086mm However theresults produced working in ps and fs regimes werenot expected Initially it was anticipated that inthese two ablation regimes a shortening of the pulseduration would lead to a better machining responsein particular surface finish The surface roughnessmeasured on the surface ablated with the fs laser

(a) fs pulse duration

(b) ps pulse duration

(c) ns pulse duration

(d) micros pulse duration

Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers

Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers

Fig 5 Surface profiles




was Ra 035mm compared to Ra 029mm achievedwith the ps one This could be explained with non-linear effects that are typical when processing mat-erials at this regime and also with the specific

machining response of the tooling steel to theselected processing parameters


Pulse duration is a major factor affecting the surfaceintegrity of processed areas In particular it is impor-tant to understand the effects of heat dissipation intothe regions nearest to the machined surface In thisresearch these effects were studied by analysing thechanges in material grain structure and thus indir-ectly to make a judgement about the specific thermalload of each ablation regime

Based on the grain size refinement observed in theareas processed with ms and ns lasers it was esti-mated that the temperature in the affected zones 1and 2 reached more than 800ndash900 C before the heatstarted to dissipate into the substrate Thus the tem-perature was sufficiently high to initiate an austenite(g) transformation which was followed by a g atransformation with cooling rates much higher thanthose in a conventional heat treatment This resultedin the creation of a non-equilibriummicrostructure inthe material in particular a higher stress level smallera grain sizes and carbides precipitated within the agrains At the same time the cooling rate was nothigh enough to initiate a martensite transformationMartensite transformations were observed only insome areas exposed to extreme conditions where asignificant deterioration of surface integrity wasobserved together with formation of large torch-likerecast zones as shown in Fig 10 The microhardnessmeasurements carried out in these areas resulted invalues around 550MHV (see Fig 10(b)) that are typicalfor quenched structures Although in this case themartensite structures were an undesired effect thetrials demonstrated that laser systems could be usedfor performing controlled surface modifications

As expected the material microstructures formedafter processing with ultra-short laser pulses showedless phase transformations than those created by

Comparison Chart

-5-4-3-2-10123

0 28 55 83 110 138 165 193 220 248 275 303 330 358 385Length micrometers

Hei

ght

mic

rom

eter

s

picosecond

femtosecond

Fig 6 A direct comparison of the surface profiles of the fields machined with the ps and fs laser sources

(a) The changes of maximum and mean grain diameters in the threestudied zones after processing with long pulsed lasers

(b) The changes of maximum and mean grain diameters in the threestudied zones after processing with short pulsed lasers

0

5

10

15

20

25

30

35

40

Zone1 Zone2 Zone3

Microsecond laserMean diameter micromMicrosecond laserMax diameter micromNanosecond laserMean diameter micromNanosecond laserMax diameter microm

0

5

10

15

20

25

30

35

Zone1 Zone2 Zone3

Picosecond laserMean diameter microm

Picosecond laserMax diameter microm

Femtosecond laserMean diameter microm

Femtosecond laserMax diameter microm

Gra

ins

diam

eter

mic

rom

etre

sG

rain

s di

amet

er m

icro

met

res

Fig 7 The changes of maximum and mean grain dia-meters in the three studied zones




Zone 1

Zone 2

Zone 3 Zone 3

Zone 2

Zone 1

(a) micros laser (b) ns laser

Fig 8 A micrograph depicting the three characteristic zones after machining

Zone 3

Zone 1

Zone 2

Zone 1

Zone 2

Zone 3

(a) ps laser (b) fs laser


Torch like martensite structure

(a) Martensite structures (b) Micro hardness chart

90

140

190

240

290

340

390

440

490

540

0 25 50 75 100 125 150Depth microm

MH

V0

025

free surface

nanosecond laser

Fig 10 Martensite torch-like structures




performing ablation with longer pulses This can beeasily explained with the specific characteristics ofthese two distinctive ablation regimes In particularthe material undergoes a direct solidndashvapour transi-tion in the case of ps and fs laser pulses comparedto the solidndashmeltndashvapour transitions when exposedto longer pulses The meltndashvapour proportion deter-mines the amount of heat that is dissipated into thesubstrate and eventually causes secondary effectssuch as microcracks phase transformations andgrain size changes As reported by Breitlung et al[7] the meltndashvapour ratio depends on pulse durationand fluence and decreases with the reduction of theinteraction time The presence of melt instigatesmore intensive heat transfer to the substrate andsubsequently a larger HAZ

In ps and fs laser ablation regimes the overallenergy transfer is very small and thus the changesof the microstructure are almost negligible A directde-sublimation of the atoms occurs and the energyis immediately taken away from the substrate Inspite of that some changes in material microstruc-ture can still be observed in the micrographs forboth ablation regimes In the case of ps laser ablationthey are more evident (see Fig 9(a)) while for the fsregime if there are any they are only within 1ndash2mmin depth (Fig 9(b))

6 CONCLUSIONS

In this research the effects of pulse duration of fourdifferent laser sources on surface integrity are investi-gated In particular an attempt is made to assess theimpact of four distinctly different laser regimes onsurface quality and material microstructure Theseare the issues that have to be taken into accountwhen considering the trade-offs between high removalrates and the resulting surface integrity This is a par-ticular dilemma when selecting the most appropriateablation regime for performing microstructuring

During laser milling applying different ablationmechanisms the material goes through several phasetransitions that have a direct impact on surface integ-rity of the processed area Thus the relevant materialcharacteristics are transition energies such as eva-poration energy and melting energy In additionthermal conductivity is a key material factor affectingthe resulting surface integrity In particular thisaffects the dissipation of the absorbed energy intothe bulk of the material and the energy losses andhence determines the size of the HAZ

The following generic conclusions could be drawnfrom this experimental study

1 For both ms and ns laser milling it was estimatedthat the HAZ on the ablated surface was within

50mm However there were some differences ingrain size refinements when comparing theresulting microstructures The melt phase duringms laser processing was bigger and more heatwas transferred into the substrate leading to for-mation of a finer grain structure

2 When performing ultra-short pulsed laser abla-tion the effects of heat transfer are not evidentas was the case with longer laser pulse durationsAlthough some heat is transferred into the bulk itis not sufficient to trigger significant structuralchanges Heat penetration is much smaller andgrain refinement is minimal The effects of pulseduration on the resulting material microstructureare more evident in the micrograph of the fieldexposed to ps laser ablation than that of thearea which underwent processing with fs laserpulses

3 Due to the ablation mechanism that is in placewhen applying ultra-short pulses significantimprovements of surface roughness can beachieved by applying ps and fs pulse lasers Inthis research a marginally better surface qualitywas achieved when performing laser millingwith a ps laser source This could be explainedwith non-linear effects that are typical for proces-sing materials at fs regimes and also with thespecific machining response of the tooling steelto the selected processing parameters especiallythe laser wavelength

These generic conclusions again underline theexisting trade-offs between the resulting surfaceintegrity and removal rates Therefore it is requiredto look for the best compromise when selecting theoptimum laser source for each specific applicationTaking into account the specific requirements ofmicrotooling applications in particular as high aspossible surface quality and relatively small volumesof material that have to be removed ultra-shortpulsed laser ablation regimes present a viable solu-tion Furthermore this research suggests that pspulse lasers offer some advantages over fs lasersources when they are utilized for machining micro-cavities in tooling steel Taking into account that thefluence of the ps laser source is four times higherthan that of the fs laser it can be expected thatthrough further process optimization an even bettersurface quality could be achieved

ACKNOWLEDGEMENTS

The research reported in this paper was fundedunder the MicroBridge programme supported bythe Welsh Assembly Government and the UK Depart-ment of Trade and Industry the EPSRC Programme




lsquoThe Cardiff Innovative Manufacturing ResearchCentrersquo and the ERDF programme lsquoMicro ToolingCentrersquo Also it was carried out within the frameworkof the EC FP6 Networks of Excellence lsquoMulti-MaterialMicro Manufacture (4M) Technologies and Appli-cationsrsquo and lsquoInnovative Production Machines andSystems (IPROMS)rsquo The authors gratefully acknow-ledge the support given to the Networks by theEuropean Commission

The authors would like to thank Dr MartynKnowles and Dr Dimitris Karnakis of Oxford LasersSteven Wheeler of Lumera and Dr Nadeem Rizvi ofUK Laser Micromachining Centre for their help inconducting this experimental study

REFERENCES

1 Lasertech GmbH Presentations operating manualGildemeister Lasertec GmbH Tirolerstrasse 85 D 87459Pfronten Germany 1999

2 Taniguchi N (Ed) Nanotechnology integrated proces-sing systems for ultra-precision and ultra-fine products1996 (Oxford University Press) ISBN 0 19 8562837

3 Fraunhofer Institut Lasertechnik (ILT) website httpwwwiltfhgdeenglasertypenhtml Last visited170106

4 Shirk M D and Molian P A A review of ultrashortpulsed laser ablation of materials J Laser Applics1998 10(1) 18ndash28

5 Chichkov B N Momma C Nolte S vonAlvensleben F and Tuennermann A Femtosecondpicosecond and nanosecond laser ablation of solidsAppl Physics 1996 A63 109ndash115

6 Momma C Nolte S Chichkov B N vonAlvensleben F and Tunnermann A Precise laserablation with ultrashort pulses Appl Surf Sci 1997109ndash110 15ndash19

7 Breitlung D Ruf A and Dausinger F Fundamentalaspects in machining of metals with short and ultra-short laser pulses Proc SPIE 2004 5339 49ndash63

8 Kautek W and Kruger J Femtosecond pulse laserablation of metallic semiconducting ceramic and bio-logical materials Proc SPIE 1994 2207 600ndash610

9 Preuss S Demchuk A and Stuke M Sub-picosecondUV laser ablation of metals Appl Physics 1995 A6133ndash37

10 von der Linde D and Sokolowski-Tinten K The phy-sical mechanisms of short-pulse laser ablation ApplSurf Sci 2000 154ndash155 1ndash10

11 Leong K Drilling with lasers Ind Laser Solutions forMfg 2000 15(9) 39


13 Geiger M Becker W Rebhan T Hutfless J andLutz N Increase of efficiency for the XeCl excimer laserablation of ceramics Appl Surf Sci 1996 96ndash98309ndash315

14 BuehlerndashOmnimet software15 Surface metrology guide website httpwwwpredev

comsmgstandardshtm Last visited 020207




Laser milling pulse durationeffects on surface integrityP V Petkov S S Dimov R M Minev and D T Pham

Manufacturing Engineering Centre Cardiff University Cardiff UK

The manuscript was received on 15 February 2007 and was accepted after revision for publication on 24 July 2007

DOI 10124309544054JEM840

Abstract Laser milling of engineering materials is a viable alternative to conventional methodsfor machining complex microcomponents The laser source employed to perform such micro-structuring has a direct impact on achievable surface integrity At the same time the trade-offsbetween high removal rates and the resulting surface integrity should be taken into accountwhen selecting the most appropriate ablation regime for performing laser milling In this paperthe effects of pulse duration on surface quality and material microstructure are investigatedwhen ablating a material commonly used for manufacturing microtooling inserts For bothmicro- and nanosecond laser regimes it was estimated that the heat-affected zone on the pro-cessed surface is within 50mm When performing ultra-short pulsed laser ablation the effectsof heat transfer are not as evident as they are after processing with longer laser pulse durationsAlthough some heat is dissipated into the bulk when working in pico- and femtosecond regimesit is not sufficient to trigger significant structural changes

Keywords laser micromachining micromachining laser pulse duration

1 INTRODUCTION

The laser milling of engineering materials has becomea viable alternative to conventional methods for pro-ducing tooling inserts for microreplication or formachining microfeatures in components By applyingthis technologymaterial is removed in a layer-by-layerfashion to produce the desired three-dimensionalstructures Direct interfaces to three-dimensionalcomputer-aided design (CAD) modelling packagesexist to assist in the machining of complex free-formsurfaces Being a non-contact material removal pro-cess some of the main advantages of laser milling arethat the process does not suffer from any problemsassociated with tool breakage does not require inclu-sion of collision checking routines in machining pro-grammes and it is easy to access areas that are verydeep in cavities Also if ultra-short pulsed lasers areutilized almost any material can be machined andthe thermal load is significantly reduced resulting inhigh surface integrity

Laser radiation can be delivered to the workpiece inan ordered sequence of pulses with a predeterminedpulse length (duration) and repetition rate (frequency)This allows the accumulated energy to be released inrelatively short time intervals which is a prerequisitefor the formation of extremely high peak powersAdditionally the laser beam can be focused on a spotwith very small dimensions from submicrometre to50mm which results in a significant energy density(fluence) and intensity (power density) in the spotarea Therefore an extremely high density can beachieved in the laserndashmaterial interaction zone thatcould not be achieved by any conventional machiningtechnology This explains the capability of lasermillingto process materials that are difficult to machine [1]In addition such a high fluence is very importantwhen producing microstructures that require a highsurface finish and hence atom cluster and atomicprocessing In particular to carry out machining atsuch a scale it is necessary to remove material withunits from 1ndash100nm to 001ndash1nm with a correspond-ing increase of the specific processing energy from103 ndash 104 (Jcm3) to 105 ndash 106 (Jcm3) [2]which is attain-able with ultra-short pulse laser ablation

The laser source employed has a direct impact onachievable surface integrity In recent years a wide

Corresponding author Manufacturing Engineering Centre

Cardiff University Queenrsquos Building The Parade Newport

Road Cardiff CF24 3AA UK email PetkovPVcfacuk

SPECIAL ISSUE PAPER 35














Laser pulse

Lens

No recastlayer

No surfacedebris

No melt zone



No microcracks

No shock wave

Plasma plume

















Laser pulse

Lens

Recastlayer

Surfacedebris

Melt zone

Heat affectedzone

Microcracks



Heat transfer

Shock wave

Plasma plume















035





029




086





218


C Si Mn Cr Mo V

038 100 040 500 130 100

At temperature























4 RESULTS




















5 DISCUSSION







Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers

Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers











Comparison Chart

-5-4-3-2-10123


Hei

ght

mic

rom

eter

s

picosecond

femtosecond




0

5

10

15

20

25

30

35

40

Zone1 Zone2 Zone3


0

5

10

15

20

25

30

35

Zone1 Zone2 Zone3





Gra

ins

diam

eter

mic

rom

etre

sG

rain

s di

amet

er m

icro

met

res





Zone 1

Zone 2

Zone 3 Zone 3

Zone 2

Zone 1



Zone 3

Zone 1

Zone 2

Zone 1

Zone 2

Zone 3





90

140

190

240

290

340

390

440

490

540

0 25 50 75 100 125 150Depth microm

MH

V0

025

free surface

nanosecond laser







6 CONCLUSIONS









ACKNOWLEDGEMENTS







REFERENCES






























Laser pulse

Lens

No recastlayer

No surfacedebris

No melt zone



No microcracks

No shock wave

Plasma plume

















Laser pulse

Lens

Recastlayer

Surfacedebris

Melt zone

Heat affectedzone

Microcracks



Heat transfer

Shock wave

Plasma plume















035





029




086





218


C Si Mn Cr Mo V

038 100 040 500 130 100

At temperature























4 RESULTS




















5 DISCUSSION







Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers

Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers











Comparison Chart

-5-4-3-2-10123


Hei

ght

mic

rom

eter

s

picosecond

femtosecond




0

5

10

15

20

25

30

35

40

Zone1 Zone2 Zone3


0

5

10

15

20

25

30

35

Zone1 Zone2 Zone3





Gra

ins

diam

eter

mic

rom

etre

sG

rain

s di

amet

er m

icro

met

res





Zone 1

Zone 2

Zone 3 Zone 3

Zone 2

Zone 1



Zone 3

Zone 1

Zone 2

Zone 1

Zone 2

Zone 3





90

140

190

240

290

340

390

440

490

540

0 25 50 75 100 125 150Depth microm

MH

V0

025

free surface

nanosecond laser







6 CONCLUSIONS









ACKNOWLEDGEMENTS







REFERENCES






























Laser pulse

Lens

Recastlayer

Surfacedebris

Melt zone

Heat affectedzone

Microcracks



Heat transfer

Shock wave

Plasma plume















035





029




086





218


C Si Mn Cr Mo V

038 100 040 500 130 100

At temperature























4 RESULTS




















5 DISCUSSION







Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers

Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers











Comparison Chart

-5-4-3-2-10123


Hei

ght

mic

rom

eter

s

picosecond

femtosecond




0

5

10

15

20

25

30

35

40

Zone1 Zone2 Zone3


0

5

10

15

20

25

30

35

Zone1 Zone2 Zone3





Gra

ins

diam

eter

mic

rom

etre

sG

rain

s di

amet

er m

icro

met

res





Zone 1

Zone 2

Zone 3 Zone 3

Zone 2

Zone 1



Zone 3

Zone 1

Zone 2

Zone 1

Zone 2

Zone 3





90

140

190

240

290

340

390

440

490

540

0 25 50 75 100 125 150Depth microm

MH

V0

025

free surface

nanosecond laser







6 CONCLUSIONS









ACKNOWLEDGEMENTS







REFERENCES





























035





029




086





218


C Si Mn Cr Mo V

038 100 040 500 130 100

At temperature























4 RESULTS




















5 DISCUSSION







Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers

Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers











Comparison Chart

-5-4-3-2-10123


Hei

ght

mic

rom

eter

s

picosecond

femtosecond




0

5

10

15

20

25

30

35

40

Zone1 Zone2 Zone3


0

5

10

15

20

25

30

35

Zone1 Zone2 Zone3





Gra

ins

diam

eter

mic

rom

etre

sG

rain

s di

amet

er m

icro

met

res





Zone 1

Zone 2

Zone 3 Zone 3

Zone 2

Zone 1



Zone 3

Zone 1

Zone 2

Zone 1

Zone 2

Zone 3





90

140

190

240

290

340

390

440

490

540

0 25 50 75 100 125 150Depth microm

MH

V0

025

free surface

nanosecond laser







6 CONCLUSIONS









ACKNOWLEDGEMENTS







REFERENCES

































4 RESULTS




















5 DISCUSSION







Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers

Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers











Comparison Chart

-5-4-3-2-10123


Hei

ght

mic

rom

eter

s

picosecond

femtosecond




0

5

10

15

20

25

30

35

40

Zone1 Zone2 Zone3


0

5

10

15

20

25

30

35

Zone1 Zone2 Zone3





Gra

ins

diam

eter

mic

rom

etre

sG

rain

s di

amet

er m

icro

met

res





Zone 1

Zone 2

Zone 3 Zone 3

Zone 2

Zone 1



Zone 3

Zone 1

Zone 2

Zone 1

Zone 2

Zone 3





90

140

190

240

290

340

390

440

490

540

0 25 50 75 100 125 150Depth microm

MH

V0

025

free surface

nanosecond laser







6 CONCLUSIONS









ACKNOWLEDGEMENTS







REFERENCES

























5 DISCUSSION







Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers

Hei

ght

mic

rom

eter

sH

eigh

t m

icro

met

ers











Comparison Chart

-5-4-3-2-10123


Hei

ght

mic

rom

eter

s

picosecond

femtosecond




0

5

10

15

20

25

30

35

40

Zone1 Zone2 Zone3


0

5

10

15

20

25

30

35

Zone1 Zone2 Zone3





Gra

ins

diam

eter

mic

rom

etre

sG

rain

s di

amet

er m

icro

met

res





Zone 1

Zone 2

Zone 3 Zone 3

Zone 2

Zone 1



Zone 3

Zone 1

Zone 2

Zone 1

Zone 2

Zone 3





90

140

190

240

290

340

390

440

490

540

0 25 50 75 100 125 150Depth microm

MH

V0

025

free surface

nanosecond laser







6 CONCLUSIONS









ACKNOWLEDGEMENTS







REFERENCES

























Comparison Chart

-5-4-3-2-10123


Hei

ght

mic

rom

eter

s

picosecond

femtosecond




0

5

10

15

20

25

30

35

40

Zone1 Zone2 Zone3


0

5

10

15

20

25

30

35

Zone1 Zone2 Zone3





Gra

ins

diam

eter

mic

rom

etre

sG

rain

s di

amet

er m

icro

met

res





Zone 1

Zone 2

Zone 3 Zone 3

Zone 2

Zone 1



Zone 3

Zone 1

Zone 2

Zone 1

Zone 2

Zone 3





90

140

190

240

290

340

390

440

490

540

0 25 50 75 100 125 150Depth microm

MH

V0

025

free surface

nanosecond laser







6 CONCLUSIONS









ACKNOWLEDGEMENTS







REFERENCES



















Zone 1

Zone 2

Zone 3 Zone 3

Zone 2

Zone 1



Zone 3

Zone 1

Zone 2

Zone 1

Zone 2

Zone 3





90

140

190

240

290

340

390

440

490

540

0 25 50 75 100 125 150Depth microm

MH

V0

025

free surface

nanosecond laser







6 CONCLUSIONS









ACKNOWLEDGEMENTS







REFERENCES





















6 CONCLUSIONS









ACKNOWLEDGEMENTS







REFERENCES





















REFERENCES



















proceedings of the institution of mechanical engineers, part b:...

Documents