ltOptical properties of Al at λ=300 nmgt

Work function of Al 408 eV (~30388 nm) (from Handbook of Chemistry and Physics)

The wavelength of the incident beam must be shorter than 30388 nm for electron emission

Absorption rate of Al depending on thickness

Absorption = 1 ndash Reflection ndash Transmission

A = 1 ndash R ndash T

At λ=300 nm

Film thickness (A) A R T

40 7 19 74

80 10 43 47

120 11 62 27

160 10 74 16

200 94 815 91

240 89 860 51

280 85 884 31

320 82 900 18

360 81 909 10

400 81 914 05

500 79 920 01

Absorption rate has maximum value (11 ) at 12 nm thickness then keeps going down as

we get thicker film It seems like we do not get 30 of absorption rate from Al thin film

Reflectance from Aluminum

Skin Depth of Aluminum at 300 nm is 2592 nm

nmm

H

s

m

H

s

m

f

5922105922

)()104()1

1099319(

10652

9

714

8

ρ = resistivity (Ωm)

μ = permeability (4π10-7 Hm) note H = henries = Ωs

reference httpwwwrfcafecom

JOURNAL OF THE OPTICAL SOCIETY OF AMERICA

Optical Constants and Reflectance and Transmittance of EvaporatedAluminum in the Visible and Ultraviolet

G HASS AND J E WAYLONISU S Army Engineer Research and Development Laboratories Fort Belvoir Virginia

(Received February 13 1961)

The paper presents new measurements of the optical constants and of the reflectance and transmittanceof evaporated aluminum films produced from the purest grade of aluminum with extremely fast evaporationat about 10-5 mm Hg The optical constants were determined by Drudes polarimetric method from opaquefilms and by reflectance transmittance and true thickness measurements of semitransparent films Thefollowing results were obtained

X (my)nkRo

220 260 300 340 380 436 492 546 578 650014 019 025 031 037 047 064 082 093 130235 285 333 380 425 484 550 599 633 711

918 920 921 923 926 926 922 916 915 907

For extremely fast evaporated films the optical constants were found to remain unchanged down to filmthicknesses of less than 100 A Data are given on the reflectance and transmittance of semitransparentaluminum films on glass and fused quartz as a function of wavelength from 220 my to 650 mgi

INTRODUCTION

EVAPORATED aluminum films are the most fre-quently used coatings for front-surface mirrors and

for interferometry in the ultraviolet Measurements oftheir optical properties have therefore been reportedby various authors1 -7 But only recently it was dis-covered that films with highest reflectance especially inthe ultraviolet and vacuum ultraviolet can only beproduced by extremely fast evaporation and by the useof the purest grade of aluminum8 9 It is the purpose ofthis paper to report on new measurements of the opticalconstants and of the reflectance and transmittance ofevaporated aluminum films produced from purest alumi-num with extremely high rates of deposition and todiscuss in addition the effect of speed of evaporationsubstrate temperature and vapor-incidence angle on thereflectance of opaque aluminum in the visible andultraviolet

EXPERIMENTAL TECHNIQUES

The vacuum apparatus used for producing aluminumfilms at extremely high rates of deposition has beendescribed in a previous paper1 The evaporations weremade in an 18-in vacuum system in which a pressure of1X 10-5 mm Hg could be maintained during the filmdeposition Ultrasil and glass plates mounted about18 in above the evaporation sources were used as sub-strates They were cleaned by a high-voltage dc glow

Presented at the Meetings of the Optical Society of AmericaBoston Massachusetts October 12-141960

I J Strong Astrophys J 83 401 (1936)2 H OBryan J Opt Soc Am 26 122 (1936)3 W Walkenhorst Z tech Physik 22 14 (1941)4 G Hass Optik 1 2 (1946)5 M F Crawford W M Gray A L Schwawlow and F M

Kelly J Opt Soc Am 39 888 (1949)6 L G Schulz J Opt Soc Am 44 357 (1954)7 L G Schulz and F R Tangherlini J Opt Soc Am 44 362

(1954)8 G Hass J Opt Soc Am 45 945 (1955)9 G Hass W R Hunter and R Tousey J Opt Soc Am

46 1009 (1956) ibid 47 1070 (1957)

discharge using 90 ma at 5 kv for 1-5 min A Japanese-fan-type shutter was used to avoid contamination of thesubstrates during the outgassing of the evaporationsources and their charges The shutter which was placedclose to the targets was opened when the evaporationreached constant speed and it was closed when a filmof desired thickness was obtained The evaporationsource for producing aluminum films at high-depositionrates consisted of six helical coils of 62-mil tungstenwire which were heated simultaneously They wereoutgassed and freed from impurities before beingcharged with aluminum With this arrangement alumi-num films could be deposited at a deposition rate ofmore than 500 Asec

The reflectance R and transmittance T of aluminumfilms were measured from 220 to 1000 myz with a modifiedBeckman model DK spectrophotometer All reflectanceand transmittance values reported in this paper weredetermined with an accuracy of about 03o of theincident light and the measurements were made at closeto normal incidence The true thickness t of the filmswas measured by a multiple-beam interferometermethod as described by Tolansky 0

The measured R and T values were corrected for theeffect of the back-surface reflectance of the substrateplates and then used to determine the optical constantsn and k = nK of semitransparent aluminum films Toobtain the optical constants from R T and I measure-ments a graphical method was applied which uses agreat number of curves displaying R and T as a functionof tX for films of various n and k on a nonabsorbing sub-strate of n= 15 The graphs were calculated from theequations developed by Hadley and Dennison 2 Theoptical constants of opaque aluminum films were de-

10 S Tolansky Multiple Beam Interferometry of Films andSurfaces (Clarenden Press Oxford England 1948)

1L N Hadley and D M Dennison J Opt Soc Am 37451 (1947)

12 The authors are indebted to L N Hadley and his co-workersfor the preparation of the graphs used in this paper for the de-termination of n and k of semitransparent aluminum films

719

VOLUME 51 NUMBER 7 JULY 1961

G HASS AND J E WAYLONIS V 51

termined by Drudes polarimetric method from 435 to650 mA The theory and various applications of thismethod have been discussed extensively in the litera-ture In the arrangement used here monochromaticlight polarized at an azimuth angle of 450 was reflectedat an angle = 70deg from the aluminum film and theellipticity produced by reflection from the metal coatingwas determined A Babinet-Soleil compensator and aGlan-Thomson prism were employed to measure thephase difference A of the two components of the ellipti-cally polarized light and of the azimuth V of the restoredplane of linear polarization Drudes equation in itsrigorous form was used to calculate the optical con-stants from A and o measurements

To obtain the true optical constants with Drudesequations A and b of an oxide-free aluminum have tobe used Since all measurements were made after thealuminum films had been exposed to air for a certainamount of time the measured A and values werecorrected for the effect of the oxide film formed onaluminum in air up to the time the measurements weremade Oxidation curves for aluminum published in arecent paper were used to calculate the corrections forA and sb The maximal errors of the optical constantsdetermined by Drudes polarimetric method amountedto about -L4 for n and 462 for k

RESULTS

A Optical Constants and Reflectanceof Aluminum Films

Table I lists the optical constants and the calculatedand measured reflectance of aluminum films All films

TABLE I The optical constants and the calculated and measuredreflectance of evaporated aluminum produced under optimumevaporation conditions from 200 to 650 mu

ReflectanceX (mU) n k calc meas

Calculated from R T and t220 014 235 918 915240 016 260 921 919260 019 285 920 922280 022 313 922 923300 025 335 921 923320 028 356 922 924340 031 380 923 925360 034 401 924 925380 037 425 926 925400 040 445 926 924450 051 500 925 922546 080 592 916 916

Drudes polarimetric method436 047 484 926 923492 064 550 922 919546 082 599 916 916578 093 633 915 914650 130 711 907 905

P P Drude Wiedem Ann 64 159 (1898)14 P H Berning G Hass and R P Madden J Opt Soc Am

50 586 (1960)

100

90

50

40

30

II

) _

IIo

IIAl

20 20_

I-2xlo-5mmH 1

1a 3 I4 5 6(0

X in mu

-6 sec

-135 sec

-180 secI

1-2x O-4mmHg

3 4 5 600X in mu

4 n mu

FIG 1 Effect of pressure and speed of evaporation on the reflect-ance of 600 to 700-A thick Al films from 220 to 600 mu

were produced under optimum conditions using purestgrade aluminum (better than 9999) and depositionrates of more than 300 Asec at 1X 10-5 mm Hg Ahigh-speed shutter was employed to produce uniformsemitransparent films at high deposition rates

In the wavelength region below 436 mu the opticalconstants were determined only from reflectance trans-mittance and true thickness measurements of semi-transparent films while at longer wavelengths bothdescribed methods were used The optical constants ob-tained from spectrophotometric measurements of semi-transparent films show good agreement with thosedetermined by Drudes polarimetric method on opaquecoatings It was also found that the optical constants ofaluminum films evaporated at an extremely high rateremain unchanged down to film thicknesses of about100 A The values for n and k decrease rather constantlywith decreasing wavelength The reflectance of opaquealuminum calculated from n and k agrees well with thedirectly measured one and is about 2 higher than theresults reported by Schulz and Tangherlini7 Films withsuch high reflectance in the ultraviolet (915 to 925)can only be produced by fast evaporation at low pres-sures Such films are in addition more compact thanslowly deposited ones and show therefore less oxidationand very small decreases in reflectance during exposureto air At 220 mu where the effect of aging is mostpronounced one-year-old aluminum mirrors still showedabout 90 reflectance The 15 reflectance decreaseat 220 mgt over a one-year period can be explained bythe formation of a 25- to 30-A-thick oxide film Such athin oxide film has only a very small effect on the re-flectance of aluminum in the visible (less than 03reflectance decrease at 550 mu)

The fact that high speed of evaporation is the mostimportant basic factor for producing aluminum filmswith highest reflectance in the ultraviolet is illustratedin Fig 1 The data presented are for three groups offilms evaporated at three different pressures The firstgroup was deposited at 1-2X 10- mm Hg the secondat 1-2X 10-4 mm Hg and the third at 1X10-3 mm Hg

720 Vol 51

I

VISIBLE AND ULTRAVIOLET 721

All films were just opaque or 600-700 A thick and con-densed on glass plates at a distance of about 20 in fromthe evaporation sources Additional filaments wereadded to obtain high deposition rates at 1 X 10-3 mm HgAll reflectances were measured within one hour after thedeposition of the coatings The time for preparing theopaque coatings is noted for each curve The curvesshow clearly that the speed of evaporation has a greatinfluence on the ultraviolet reflectance of aluminum anda much smaller effect on its reflectance at longer wave-lengths An increase of the evaporation time for anopaque coating from 7 sec (which is still not the opti-mum evaporation speed) to 180 sec at 1-2X 10-5 mmHg decreases the reflectance at 220 mgu from 91 to62 but produces only a reflectance decrease of 1 at550 mt The influence of firing speed is even greater athigher pressures However films with very high ultra-violet reflectance can still be made at 10-4 mm Hg andeven at 10-3 mm Hg At higher pressures however theultraviolet reflectance decreases more rapidly with in-creasing evaporation time than it does at lower pres-sures An opaque coating deposited in 180 sec at 10-4

mm Hg reflects only about 40 at 220 mAt and a coatingproduced at 10-3 mm Hg shows only 12 reflectance atthe same wavelength The conclusion is that fast evapo-ration at low pressure is most important for producingaluminum films with highest reflectance in the ultra-violet and that only such films should be used to deter-mine the optical constants of aluminum

High speed of evaporation is not the only factoraffecting the reflectance of evaporated aluminum Sub-strate temperature during the evaporation film thick-ness and angle of vapor incidence have also a greatinfluence on the reflectance of evaporated coatingsespecially at shorter wavelengths

Experience has shown that most materials includingaluminum have better adherence if they are evaporatedonto heated substrates With increasing substrate tem-

TABLE II The effect of substrate temperature during theevaporation on the reflectance of about 700-A thick Al films de-posited at 1-2X 10-5 mm Hg at various deposition rates

Percent reflectanceSubstrate temperature (0C)

X (mA) 300 500 1000 1500 2000

A Deposition rate -300 Asec220 915 913 900 879 838260 922 921 915 904 884300 923 923 919 912 902400 924 923 920 917 915500 917 917 915 914 913600 910 909 908 907 905

B Deposition rate -10-15 Asec220 890 889 860 652 466260 912 910 899 735 567300 918 917 912 792 649400 920 919 917 874 779500 916 915 915 895 839600 909 909 908 896 867

TABLE III The effect of vapor incidence on the reflectance ofevaporated aluminum films of two different thicknesses

Percent reflectanceAngle of vapor incidence

O (normal incidence) 300 600t-600 A t 2000 A t 600 A t -2000 A 1-600 A t 2000

A Optimum evaporation conditions(Deposition rate 300 Asec p =1 X10-5 mm Hg)

220 915 912 914 907 904 750260 922 919 920 917 915 822300 923 922 921 918 918 855400 924 922 921 919 918 885500 917 917 916 916 915 900600 910 909 909 908 908 896

B Poor evaporation conditions(Deposition rate 10 Asec =1 X10-4 mm Hg)

220 835 659 826 605 814 318260 884 741 882 705 876 425300 905 795 904 759 896 532400 913 868 913 832 910 720500 910 888 910 875 908 813600 903 893 902 890 900 847

perature however grain size and therefore surfaceroughness increase and this results in a decreasedspecular reflectance especially of shorter wavelengthsTable II shows the effect of substrate temperature onthe reflectance of just opaque aluminum films depositedat 10-5 mm Hg The results are given for two groups offilms one deposited at a rate of about 300 Asec and theother deposited at 10-15 Asec For substrate tempera-tures higher than 50C the specular reflectance of allfilms decreases with increasing temperature The effectis however much more pronounced in the ultravioletand for films produced at low deposition rates For therapidly evaporated films (deposition rate -300 Asec)a substrate temperature increase from 300 to 200TC re-duces the reflectance at 220 mAu from 915 to about 82and at 500 mpk from 917 to 913 Films produced atlower rates show much greater reduction in reflectancewith increasing substrate temperature Films depositedat a rate of 10-15 Asec for example show a reflectancedecrease from 89 to 46 at 220 mAt and from 916 to839 at 500 m4 when the substrate temperature isincreased from 30deg to 200TC The data presented supportthe following conclusions Whenever aluminum filmsof highest ultraviolet reflectance are required the sub-strate temperature during the evaporation should notbe higher than 50TC and if higher substrate tempera-tures for increased adhesion are applied the films shouldbe evaporated as fast as possible

The effect of vapor incidence on the reflectance ofevaporated aluminum has been studied by Holland 5

He showed that diffuse reflecting surfaces are formedmore easily as the vapor incidence angle and the filmthickness are increased Table III shows more completedata of the effect of vapor incidence on the ultravioletand visible reflectance of evaporated aluminum films oftwo different thicknesses Two groups of films wereinvestigated one prepared under optimum conditions

1 L Holland J Opt Soc Am 43 376 (1953)

July 1961 EVAPORATED ALUMINUM IN

G HASS AND J E WAYLONIS

TABLE IV Calculated reflectance and transmittance of Al films evaporated under optimum conditions onto transparent substratesof n = 15 for various wavelengths as a function of film thickness (Calculated values agree with directly measured ones for film thick-nesses gt 100 A back surface antireflected)

Film Wavelength (mu)thickness 220 300 400 546 650

(A) R T R T R T R T R T

40 14 82 19 74 25 65 33 51 38 4280 33 60 43 47 52 36 60 24 63 18

120 52 40 62 27 70 19 74 12 75 9160 67 25 74 16 79 11 81 7 82 5200 763 152 815 91 849 59 856 35 854 26240 824 91 860 51 881 33 881 20 875 14280 862 54 884 31 900 19 895 11 888 08320 885 32 900 18 911 11 904 05 896 04360 898 19 909 10 917 06 909 04 900 03400 906 11 914 05 921 04 912 02 903 02500 915 03 920 01 925 lt01 915 lt01 906 lt01

(deposition rate 300 Asec at X 10-5 mm Hg) and theother deposited under poor evaporation conditions (de-position rate 10 Asec at 1X 10-4) In each group thefilms measured were about 600 A and 2000 A thick Theresults show that the reflectance decrease with increas-ing angle of vapor incidence depends upon wavelengthfilm thickness and evaporation conditions The effectis much more pronounced at shorter wavelengths andfor films prepared under poor conditions and increasesin all cases with increasing film thickness An increaseof vapor incidence angle from 00 (normal incidence) to60deg has only a very small effect on the reflectance of600-A-thick coatings but greatly decreases the re-flectance of 2000-A-thick films in the ultraviolet A2000-A-thick film deposited at normal incidence underoptimum conditions reflects almost as well as a justopaque coating even at 220 myi The reflectance at220 mju however drops for a 2000-A thick film from912 to 750 when the vapor incidence angle is in-creased from 0deg to 600 This effect is much greater forfilms produced under poor evaporation conditions Here2000-A thick films evaporated at normal incidencealready show a much lower reflectance than just opaquecoatings and the increase in vapor incidence angle from0deg to 600 drops the reflectance of a 2000-A thick filmfrom 659 to 312 at 220 mu and from 888 to 813at 500 mu This indicates that mirror coatings for theultraviolet should be evaporated at close to normalincidence and that the coatings should not be thicker

than absolutely necessarycidence are unavoidable

if higher angles of vapor in-

B Reflectance and Transmittance ofSemitransparent Films

Semitransparent aluminum films are especially im-portant for interferometry in the ultraviolet becausethey show in this wavelength region higher reflectanceand lower absorptance than any other metallic coatingTable IV shows the calculated reflectance and trans-mittance of evaporated aluminum for various wave-lengths and various film thicknesses The optical con-stants n and k listed in Table I were used for thecalculations For films produced with deposition rates ofmore than 100 Asec at 1 X 10-5 mm Hg the calculatedR and T values agreed well with the directly measuredones down to film thicknesses of about 100 A The re-sults show that for the same film thickness the trans-mittance increases strongly with decreasing wavelengthA 160-A thick film for example transmits 25 at220 mAu and only 5 at 546 mit

Films deposited at low deposition rates show muchhigher absorptance and lower reflectance and are there-fore unsuitable for interferometry in the ultraviolet Apreviously reported statements that semitransparent Alfilms produced at high deposition rates show a mottledappearance could not be verified All films producedunder the above described conditions were extremelyuniform in appearance

722 Vol 51

• Optical properties of Al at 300nmpdf
• Al_Reflection_Transmissionpdf
• Optical Properties of Aluminumpdf

Reflectance from Aluminum

Skin Depth of Aluminum at 300 nm is 2592 nm

nmm

H

s

m

H

s

m

f

5922105922

)()104()1

1099319(

10652

9

714

8

ρ = resistivity (Ωm)

μ = permeability (4π10-7 Hm) note H = henries = Ωs

reference httpwwwrfcafecom

JOURNAL OF THE OPTICAL SOCIETY OF AMERICA

Optical Constants and Reflectance and Transmittance of EvaporatedAluminum in the Visible and Ultraviolet

G HASS AND J E WAYLONISU S Army Engineer Research and Development Laboratories Fort Belvoir Virginia

(Received February 13 1961)

The paper presents new measurements of the optical constants and of the reflectance and transmittanceof evaporated aluminum films produced from the purest grade of aluminum with extremely fast evaporationat about 10-5 mm Hg The optical constants were determined by Drudes polarimetric method from opaquefilms and by reflectance transmittance and true thickness measurements of semitransparent films Thefollowing results were obtained

X (my)nkRo

220 260 300 340 380 436 492 546 578 650014 019 025 031 037 047 064 082 093 130235 285 333 380 425 484 550 599 633 711

918 920 921 923 926 926 922 916 915 907

For extremely fast evaporated films the optical constants were found to remain unchanged down to filmthicknesses of less than 100 A Data are given on the reflectance and transmittance of semitransparentaluminum films on glass and fused quartz as a function of wavelength from 220 my to 650 mgi

INTRODUCTION

EVAPORATED aluminum films are the most fre-quently used coatings for front-surface mirrors and

for interferometry in the ultraviolet Measurements oftheir optical properties have therefore been reportedby various authors1 -7 But only recently it was dis-covered that films with highest reflectance especially inthe ultraviolet and vacuum ultraviolet can only beproduced by extremely fast evaporation and by the useof the purest grade of aluminum8 9 It is the purpose ofthis paper to report on new measurements of the opticalconstants and of the reflectance and transmittance ofevaporated aluminum films produced from purest alumi-num with extremely high rates of deposition and todiscuss in addition the effect of speed of evaporationsubstrate temperature and vapor-incidence angle on thereflectance of opaque aluminum in the visible andultraviolet

EXPERIMENTAL TECHNIQUES

The vacuum apparatus used for producing aluminumfilms at extremely high rates of deposition has beendescribed in a previous paper1 The evaporations weremade in an 18-in vacuum system in which a pressure of1X 10-5 mm Hg could be maintained during the filmdeposition Ultrasil and glass plates mounted about18 in above the evaporation sources were used as sub-strates They were cleaned by a high-voltage dc glow

Presented at the Meetings of the Optical Society of AmericaBoston Massachusetts October 12-141960

I J Strong Astrophys J 83 401 (1936)2 H OBryan J Opt Soc Am 26 122 (1936)3 W Walkenhorst Z tech Physik 22 14 (1941)4 G Hass Optik 1 2 (1946)5 M F Crawford W M Gray A L Schwawlow and F M

Kelly J Opt Soc Am 39 888 (1949)6 L G Schulz J Opt Soc Am 44 357 (1954)7 L G Schulz and F R Tangherlini J Opt Soc Am 44 362

(1954)8 G Hass J Opt Soc Am 45 945 (1955)9 G Hass W R Hunter and R Tousey J Opt Soc Am

46 1009 (1956) ibid 47 1070 (1957)

discharge using 90 ma at 5 kv for 1-5 min A Japanese-fan-type shutter was used to avoid contamination of thesubstrates during the outgassing of the evaporationsources and their charges The shutter which was placedclose to the targets was opened when the evaporationreached constant speed and it was closed when a filmof desired thickness was obtained The evaporationsource for producing aluminum films at high-depositionrates consisted of six helical coils of 62-mil tungstenwire which were heated simultaneously They wereoutgassed and freed from impurities before beingcharged with aluminum With this arrangement alumi-num films could be deposited at a deposition rate ofmore than 500 Asec

The reflectance R and transmittance T of aluminumfilms were measured from 220 to 1000 myz with a modifiedBeckman model DK spectrophotometer All reflectanceand transmittance values reported in this paper weredetermined with an accuracy of about 03o of theincident light and the measurements were made at closeto normal incidence The true thickness t of the filmswas measured by a multiple-beam interferometermethod as described by Tolansky 0

The measured R and T values were corrected for theeffect of the back-surface reflectance of the substrateplates and then used to determine the optical constantsn and k = nK of semitransparent aluminum films Toobtain the optical constants from R T and I measure-ments a graphical method was applied which uses agreat number of curves displaying R and T as a functionof tX for films of various n and k on a nonabsorbing sub-strate of n= 15 The graphs were calculated from theequations developed by Hadley and Dennison 2 Theoptical constants of opaque aluminum films were de-

10 S Tolansky Multiple Beam Interferometry of Films andSurfaces (Clarenden Press Oxford England 1948)

1L N Hadley and D M Dennison J Opt Soc Am 37451 (1947)

12 The authors are indebted to L N Hadley and his co-workersfor the preparation of the graphs used in this paper for the de-termination of n and k of semitransparent aluminum films

719

VOLUME 51 NUMBER 7 JULY 1961

G HASS AND J E WAYLONIS V 51

termined by Drudes polarimetric method from 435 to650 mA The theory and various applications of thismethod have been discussed extensively in the litera-ture In the arrangement used here monochromaticlight polarized at an azimuth angle of 450 was reflectedat an angle = 70deg from the aluminum film and theellipticity produced by reflection from the metal coatingwas determined A Babinet-Soleil compensator and aGlan-Thomson prism were employed to measure thephase difference A of the two components of the ellipti-cally polarized light and of the azimuth V of the restoredplane of linear polarization Drudes equation in itsrigorous form was used to calculate the optical con-stants from A and o measurements

To obtain the true optical constants with Drudesequations A and b of an oxide-free aluminum have tobe used Since all measurements were made after thealuminum films had been exposed to air for a certainamount of time the measured A and values werecorrected for the effect of the oxide film formed onaluminum in air up to the time the measurements weremade Oxidation curves for aluminum published in arecent paper were used to calculate the corrections forA and sb The maximal errors of the optical constantsdetermined by Drudes polarimetric method amountedto about -L4 for n and 462 for k

RESULTS

A Optical Constants and Reflectanceof Aluminum Films

Table I lists the optical constants and the calculatedand measured reflectance of aluminum films All films

TABLE I The optical constants and the calculated and measuredreflectance of evaporated aluminum produced under optimumevaporation conditions from 200 to 650 mu

ReflectanceX (mU) n k calc meas

Calculated from R T and t220 014 235 918 915240 016 260 921 919260 019 285 920 922280 022 313 922 923300 025 335 921 923320 028 356 922 924340 031 380 923 925360 034 401 924 925380 037 425 926 925400 040 445 926 924450 051 500 925 922546 080 592 916 916

Drudes polarimetric method436 047 484 926 923492 064 550 922 919546 082 599 916 916578 093 633 915 914650 130 711 907 905

P P Drude Wiedem Ann 64 159 (1898)14 P H Berning G Hass and R P Madden J Opt Soc Am

50 586 (1960)

100

90

50

40

30

II

) _

IIo

IIAl

20 20_

I-2xlo-5mmH 1

1a 3 I4 5 6(0

X in mu

-6 sec

-135 sec

-180 secI

1-2x O-4mmHg

3 4 5 600X in mu

4 n mu

FIG 1 Effect of pressure and speed of evaporation on the reflect-ance of 600 to 700-A thick Al films from 220 to 600 mu

were produced under optimum conditions using purestgrade aluminum (better than 9999) and depositionrates of more than 300 Asec at 1X 10-5 mm Hg Ahigh-speed shutter was employed to produce uniformsemitransparent films at high deposition rates

In the wavelength region below 436 mu the opticalconstants were determined only from reflectance trans-mittance and true thickness measurements of semi-transparent films while at longer wavelengths bothdescribed methods were used The optical constants ob-tained from spectrophotometric measurements of semi-transparent films show good agreement with thosedetermined by Drudes polarimetric method on opaquecoatings It was also found that the optical constants ofaluminum films evaporated at an extremely high rateremain unchanged down to film thicknesses of about100 A The values for n and k decrease rather constantlywith decreasing wavelength The reflectance of opaquealuminum calculated from n and k agrees well with thedirectly measured one and is about 2 higher than theresults reported by Schulz and Tangherlini7 Films withsuch high reflectance in the ultraviolet (915 to 925)can only be produced by fast evaporation at low pres-sures Such films are in addition more compact thanslowly deposited ones and show therefore less oxidationand very small decreases in reflectance during exposureto air At 220 mu where the effect of aging is mostpronounced one-year-old aluminum mirrors still showedabout 90 reflectance The 15 reflectance decreaseat 220 mgt over a one-year period can be explained bythe formation of a 25- to 30-A-thick oxide film Such athin oxide film has only a very small effect on the re-flectance of aluminum in the visible (less than 03reflectance decrease at 550 mu)

The fact that high speed of evaporation is the mostimportant basic factor for producing aluminum filmswith highest reflectance in the ultraviolet is illustratedin Fig 1 The data presented are for three groups offilms evaporated at three different pressures The firstgroup was deposited at 1-2X 10- mm Hg the secondat 1-2X 10-4 mm Hg and the third at 1X10-3 mm Hg

720 Vol 51

I

VISIBLE AND ULTRAVIOLET 721

All films were just opaque or 600-700 A thick and con-densed on glass plates at a distance of about 20 in fromthe evaporation sources Additional filaments wereadded to obtain high deposition rates at 1 X 10-3 mm HgAll reflectances were measured within one hour after thedeposition of the coatings The time for preparing theopaque coatings is noted for each curve The curvesshow clearly that the speed of evaporation has a greatinfluence on the ultraviolet reflectance of aluminum anda much smaller effect on its reflectance at longer wave-lengths An increase of the evaporation time for anopaque coating from 7 sec (which is still not the opti-mum evaporation speed) to 180 sec at 1-2X 10-5 mmHg decreases the reflectance at 220 mgu from 91 to62 but produces only a reflectance decrease of 1 at550 mt The influence of firing speed is even greater athigher pressures However films with very high ultra-violet reflectance can still be made at 10-4 mm Hg andeven at 10-3 mm Hg At higher pressures however theultraviolet reflectance decreases more rapidly with in-creasing evaporation time than it does at lower pres-sures An opaque coating deposited in 180 sec at 10-4

mm Hg reflects only about 40 at 220 mAt and a coatingproduced at 10-3 mm Hg shows only 12 reflectance atthe same wavelength The conclusion is that fast evapo-ration at low pressure is most important for producingaluminum films with highest reflectance in the ultra-violet and that only such films should be used to deter-mine the optical constants of aluminum

High speed of evaporation is not the only factoraffecting the reflectance of evaporated aluminum Sub-strate temperature during the evaporation film thick-ness and angle of vapor incidence have also a greatinfluence on the reflectance of evaporated coatingsespecially at shorter wavelengths

Experience has shown that most materials includingaluminum have better adherence if they are evaporatedonto heated substrates With increasing substrate tem-

TABLE II The effect of substrate temperature during theevaporation on the reflectance of about 700-A thick Al films de-posited at 1-2X 10-5 mm Hg at various deposition rates

Percent reflectanceSubstrate temperature (0C)

X (mA) 300 500 1000 1500 2000

A Deposition rate -300 Asec220 915 913 900 879 838260 922 921 915 904 884300 923 923 919 912 902400 924 923 920 917 915500 917 917 915 914 913600 910 909 908 907 905

B Deposition rate -10-15 Asec220 890 889 860 652 466260 912 910 899 735 567300 918 917 912 792 649400 920 919 917 874 779500 916 915 915 895 839600 909 909 908 896 867

TABLE III The effect of vapor incidence on the reflectance ofevaporated aluminum films of two different thicknesses

Percent reflectanceAngle of vapor incidence

O (normal incidence) 300 600t-600 A t 2000 A t 600 A t -2000 A 1-600 A t 2000

A Optimum evaporation conditions(Deposition rate 300 Asec p =1 X10-5 mm Hg)

220 915 912 914 907 904 750260 922 919 920 917 915 822300 923 922 921 918 918 855400 924 922 921 919 918 885500 917 917 916 916 915 900600 910 909 909 908 908 896

B Poor evaporation conditions(Deposition rate 10 Asec =1 X10-4 mm Hg)

220 835 659 826 605 814 318260 884 741 882 705 876 425300 905 795 904 759 896 532400 913 868 913 832 910 720500 910 888 910 875 908 813600 903 893 902 890 900 847

perature however grain size and therefore surfaceroughness increase and this results in a decreasedspecular reflectance especially of shorter wavelengthsTable II shows the effect of substrate temperature onthe reflectance of just opaque aluminum films depositedat 10-5 mm Hg The results are given for two groups offilms one deposited at a rate of about 300 Asec and theother deposited at 10-15 Asec For substrate tempera-tures higher than 50C the specular reflectance of allfilms decreases with increasing temperature The effectis however much more pronounced in the ultravioletand for films produced at low deposition rates For therapidly evaporated films (deposition rate -300 Asec)a substrate temperature increase from 300 to 200TC re-duces the reflectance at 220 mAu from 915 to about 82and at 500 mpk from 917 to 913 Films produced atlower rates show much greater reduction in reflectancewith increasing substrate temperature Films depositedat a rate of 10-15 Asec for example show a reflectancedecrease from 89 to 46 at 220 mAt and from 916 to839 at 500 m4 when the substrate temperature isincreased from 30deg to 200TC The data presented supportthe following conclusions Whenever aluminum filmsof highest ultraviolet reflectance are required the sub-strate temperature during the evaporation should notbe higher than 50TC and if higher substrate tempera-tures for increased adhesion are applied the films shouldbe evaporated as fast as possible

The effect of vapor incidence on the reflectance ofevaporated aluminum has been studied by Holland 5

He showed that diffuse reflecting surfaces are formedmore easily as the vapor incidence angle and the filmthickness are increased Table III shows more completedata of the effect of vapor incidence on the ultravioletand visible reflectance of evaporated aluminum films oftwo different thicknesses Two groups of films wereinvestigated one prepared under optimum conditions

1 L Holland J Opt Soc Am 43 376 (1953)

July 1961 EVAPORATED ALUMINUM IN

G HASS AND J E WAYLONIS

TABLE IV Calculated reflectance and transmittance of Al films evaporated under optimum conditions onto transparent substratesof n = 15 for various wavelengths as a function of film thickness (Calculated values agree with directly measured ones for film thick-nesses gt 100 A back surface antireflected)

Film Wavelength (mu)thickness 220 300 400 546 650

(A) R T R T R T R T R T

40 14 82 19 74 25 65 33 51 38 4280 33 60 43 47 52 36 60 24 63 18

120 52 40 62 27 70 19 74 12 75 9160 67 25 74 16 79 11 81 7 82 5200 763 152 815 91 849 59 856 35 854 26240 824 91 860 51 881 33 881 20 875 14280 862 54 884 31 900 19 895 11 888 08320 885 32 900 18 911 11 904 05 896 04360 898 19 909 10 917 06 909 04 900 03400 906 11 914 05 921 04 912 02 903 02500 915 03 920 01 925 lt01 915 lt01 906 lt01

(deposition rate 300 Asec at X 10-5 mm Hg) and theother deposited under poor evaporation conditions (de-position rate 10 Asec at 1X 10-4) In each group thefilms measured were about 600 A and 2000 A thick Theresults show that the reflectance decrease with increas-ing angle of vapor incidence depends upon wavelengthfilm thickness and evaporation conditions The effectis much more pronounced at shorter wavelengths andfor films prepared under poor conditions and increasesin all cases with increasing film thickness An increaseof vapor incidence angle from 00 (normal incidence) to60deg has only a very small effect on the reflectance of600-A-thick coatings but greatly decreases the re-flectance of 2000-A-thick films in the ultraviolet A2000-A-thick film deposited at normal incidence underoptimum conditions reflects almost as well as a justopaque coating even at 220 myi The reflectance at220 mju however drops for a 2000-A thick film from912 to 750 when the vapor incidence angle is in-creased from 0deg to 600 This effect is much greater forfilms produced under poor evaporation conditions Here2000-A thick films evaporated at normal incidencealready show a much lower reflectance than just opaquecoatings and the increase in vapor incidence angle from0deg to 600 drops the reflectance of a 2000-A thick filmfrom 659 to 312 at 220 mu and from 888 to 813at 500 mu This indicates that mirror coatings for theultraviolet should be evaporated at close to normalincidence and that the coatings should not be thicker

than absolutely necessarycidence are unavoidable

if higher angles of vapor in-

B Reflectance and Transmittance ofSemitransparent Films

Semitransparent aluminum films are especially im-portant for interferometry in the ultraviolet becausethey show in this wavelength region higher reflectanceand lower absorptance than any other metallic coatingTable IV shows the calculated reflectance and trans-mittance of evaporated aluminum for various wave-lengths and various film thicknesses The optical con-stants n and k listed in Table I were used for thecalculations For films produced with deposition rates ofmore than 100 Asec at 1 X 10-5 mm Hg the calculatedR and T values agreed well with the directly measuredones down to film thicknesses of about 100 A The re-sults show that for the same film thickness the trans-mittance increases strongly with decreasing wavelengthA 160-A thick film for example transmits 25 at220 mAu and only 5 at 546 mit

Films deposited at low deposition rates show muchhigher absorptance and lower reflectance and are there-fore unsuitable for interferometry in the ultraviolet Apreviously reported statements that semitransparent Alfilms produced at high deposition rates show a mottledappearance could not be verified All films producedunder the above described conditions were extremelyuniform in appearance

722 Vol 51

• Optical properties of Al at 300nmpdf
• Al_Reflection_Transmissionpdf
• Optical Properties of Aluminumpdf

Skin Depth of Aluminum at 300 nm is 2592 nm

nmm

H

s

m

H

s

m

f

5922105922

)()104()1

1099319(

10652

9

714

8

ρ = resistivity (Ωm)

μ = permeability (4π10-7 Hm) note H = henries = Ωs

reference httpwwwrfcafecom

JOURNAL OF THE OPTICAL SOCIETY OF AMERICA

Optical Constants and Reflectance and Transmittance of EvaporatedAluminum in the Visible and Ultraviolet

G HASS AND J E WAYLONISU S Army Engineer Research and Development Laboratories Fort Belvoir Virginia

(Received February 13 1961)

The paper presents new measurements of the optical constants and of the reflectance and transmittanceof evaporated aluminum films produced from the purest grade of aluminum with extremely fast evaporationat about 10-5 mm Hg The optical constants were determined by Drudes polarimetric method from opaquefilms and by reflectance transmittance and true thickness measurements of semitransparent films Thefollowing results were obtained

X (my)nkRo

220 260 300 340 380 436 492 546 578 650014 019 025 031 037 047 064 082 093 130235 285 333 380 425 484 550 599 633 711

918 920 921 923 926 926 922 916 915 907

For extremely fast evaporated films the optical constants were found to remain unchanged down to filmthicknesses of less than 100 A Data are given on the reflectance and transmittance of semitransparentaluminum films on glass and fused quartz as a function of wavelength from 220 my to 650 mgi

INTRODUCTION

EVAPORATED aluminum films are the most fre-quently used coatings for front-surface mirrors and

for interferometry in the ultraviolet Measurements oftheir optical properties have therefore been reportedby various authors1 -7 But only recently it was dis-covered that films with highest reflectance especially inthe ultraviolet and vacuum ultraviolet can only beproduced by extremely fast evaporation and by the useof the purest grade of aluminum8 9 It is the purpose ofthis paper to report on new measurements of the opticalconstants and of the reflectance and transmittance ofevaporated aluminum films produced from purest alumi-num with extremely high rates of deposition and todiscuss in addition the effect of speed of evaporationsubstrate temperature and vapor-incidence angle on thereflectance of opaque aluminum in the visible andultraviolet

EXPERIMENTAL TECHNIQUES

The vacuum apparatus used for producing aluminumfilms at extremely high rates of deposition has beendescribed in a previous paper1 The evaporations weremade in an 18-in vacuum system in which a pressure of1X 10-5 mm Hg could be maintained during the filmdeposition Ultrasil and glass plates mounted about18 in above the evaporation sources were used as sub-strates They were cleaned by a high-voltage dc glow

Presented at the Meetings of the Optical Society of AmericaBoston Massachusetts October 12-141960

I J Strong Astrophys J 83 401 (1936)2 H OBryan J Opt Soc Am 26 122 (1936)3 W Walkenhorst Z tech Physik 22 14 (1941)4 G Hass Optik 1 2 (1946)5 M F Crawford W M Gray A L Schwawlow and F M

Kelly J Opt Soc Am 39 888 (1949)6 L G Schulz J Opt Soc Am 44 357 (1954)7 L G Schulz and F R Tangherlini J Opt Soc Am 44 362

(1954)8 G Hass J Opt Soc Am 45 945 (1955)9 G Hass W R Hunter and R Tousey J Opt Soc Am

46 1009 (1956) ibid 47 1070 (1957)

discharge using 90 ma at 5 kv for 1-5 min A Japanese-fan-type shutter was used to avoid contamination of thesubstrates during the outgassing of the evaporationsources and their charges The shutter which was placedclose to the targets was opened when the evaporationreached constant speed and it was closed when a filmof desired thickness was obtained The evaporationsource for producing aluminum films at high-depositionrates consisted of six helical coils of 62-mil tungstenwire which were heated simultaneously They wereoutgassed and freed from impurities before beingcharged with aluminum With this arrangement alumi-num films could be deposited at a deposition rate ofmore than 500 Asec

The reflectance R and transmittance T of aluminumfilms were measured from 220 to 1000 myz with a modifiedBeckman model DK spectrophotometer All reflectanceand transmittance values reported in this paper weredetermined with an accuracy of about 03o of theincident light and the measurements were made at closeto normal incidence The true thickness t of the filmswas measured by a multiple-beam interferometermethod as described by Tolansky 0

The measured R and T values were corrected for theeffect of the back-surface reflectance of the substrateplates and then used to determine the optical constantsn and k = nK of semitransparent aluminum films Toobtain the optical constants from R T and I measure-ments a graphical method was applied which uses agreat number of curves displaying R and T as a functionof tX for films of various n and k on a nonabsorbing sub-strate of n= 15 The graphs were calculated from theequations developed by Hadley and Dennison 2 Theoptical constants of opaque aluminum films were de-

10 S Tolansky Multiple Beam Interferometry of Films andSurfaces (Clarenden Press Oxford England 1948)

1L N Hadley and D M Dennison J Opt Soc Am 37451 (1947)

12 The authors are indebted to L N Hadley and his co-workersfor the preparation of the graphs used in this paper for the de-termination of n and k of semitransparent aluminum films

719

VOLUME 51 NUMBER 7 JULY 1961

G HASS AND J E WAYLONIS V 51

termined by Drudes polarimetric method from 435 to650 mA The theory and various applications of thismethod have been discussed extensively in the litera-ture In the arrangement used here monochromaticlight polarized at an azimuth angle of 450 was reflectedat an angle = 70deg from the aluminum film and theellipticity produced by reflection from the metal coatingwas determined A Babinet-Soleil compensator and aGlan-Thomson prism were employed to measure thephase difference A of the two components of the ellipti-cally polarized light and of the azimuth V of the restoredplane of linear polarization Drudes equation in itsrigorous form was used to calculate the optical con-stants from A and o measurements

To obtain the true optical constants with Drudesequations A and b of an oxide-free aluminum have tobe used Since all measurements were made after thealuminum films had been exposed to air for a certainamount of time the measured A and values werecorrected for the effect of the oxide film formed onaluminum in air up to the time the measurements weremade Oxidation curves for aluminum published in arecent paper were used to calculate the corrections forA and sb The maximal errors of the optical constantsdetermined by Drudes polarimetric method amountedto about -L4 for n and 462 for k

RESULTS

A Optical Constants and Reflectanceof Aluminum Films

Table I lists the optical constants and the calculatedand measured reflectance of aluminum films All films

TABLE I The optical constants and the calculated and measuredreflectance of evaporated aluminum produced under optimumevaporation conditions from 200 to 650 mu

ReflectanceX (mU) n k calc meas

Calculated from R T and t220 014 235 918 915240 016 260 921 919260 019 285 920 922280 022 313 922 923300 025 335 921 923320 028 356 922 924340 031 380 923 925360 034 401 924 925380 037 425 926 925400 040 445 926 924450 051 500 925 922546 080 592 916 916

Drudes polarimetric method436 047 484 926 923492 064 550 922 919546 082 599 916 916578 093 633 915 914650 130 711 907 905

P P Drude Wiedem Ann 64 159 (1898)14 P H Berning G Hass and R P Madden J Opt Soc Am

50 586 (1960)

100

90

50

40

30

II

) _

IIo

IIAl

20 20_

I-2xlo-5mmH 1

1a 3 I4 5 6(0

X in mu

-6 sec

-135 sec

-180 secI

1-2x O-4mmHg

3 4 5 600X in mu

4 n mu

FIG 1 Effect of pressure and speed of evaporation on the reflect-ance of 600 to 700-A thick Al films from 220 to 600 mu

were produced under optimum conditions using purestgrade aluminum (better than 9999) and depositionrates of more than 300 Asec at 1X 10-5 mm Hg Ahigh-speed shutter was employed to produce uniformsemitransparent films at high deposition rates

In the wavelength region below 436 mu the opticalconstants were determined only from reflectance trans-mittance and true thickness measurements of semi-transparent films while at longer wavelengths bothdescribed methods were used The optical constants ob-tained from spectrophotometric measurements of semi-transparent films show good agreement with thosedetermined by Drudes polarimetric method on opaquecoatings It was also found that the optical constants ofaluminum films evaporated at an extremely high rateremain unchanged down to film thicknesses of about100 A The values for n and k decrease rather constantlywith decreasing wavelength The reflectance of opaquealuminum calculated from n and k agrees well with thedirectly measured one and is about 2 higher than theresults reported by Schulz and Tangherlini7 Films withsuch high reflectance in the ultraviolet (915 to 925)can only be produced by fast evaporation at low pres-sures Such films are in addition more compact thanslowly deposited ones and show therefore less oxidationand very small decreases in reflectance during exposureto air At 220 mu where the effect of aging is mostpronounced one-year-old aluminum mirrors still showedabout 90 reflectance The 15 reflectance decreaseat 220 mgt over a one-year period can be explained bythe formation of a 25- to 30-A-thick oxide film Such athin oxide film has only a very small effect on the re-flectance of aluminum in the visible (less than 03reflectance decrease at 550 mu)

The fact that high speed of evaporation is the mostimportant basic factor for producing aluminum filmswith highest reflectance in the ultraviolet is illustratedin Fig 1 The data presented are for three groups offilms evaporated at three different pressures The firstgroup was deposited at 1-2X 10- mm Hg the secondat 1-2X 10-4 mm Hg and the third at 1X10-3 mm Hg

720 Vol 51

I

VISIBLE AND ULTRAVIOLET 721

All films were just opaque or 600-700 A thick and con-densed on glass plates at a distance of about 20 in fromthe evaporation sources Additional filaments wereadded to obtain high deposition rates at 1 X 10-3 mm HgAll reflectances were measured within one hour after thedeposition of the coatings The time for preparing theopaque coatings is noted for each curve The curvesshow clearly that the speed of evaporation has a greatinfluence on the ultraviolet reflectance of aluminum anda much smaller effect on its reflectance at longer wave-lengths An increase of the evaporation time for anopaque coating from 7 sec (which is still not the opti-mum evaporation speed) to 180 sec at 1-2X 10-5 mmHg decreases the reflectance at 220 mgu from 91 to62 but produces only a reflectance decrease of 1 at550 mt The influence of firing speed is even greater athigher pressures However films with very high ultra-violet reflectance can still be made at 10-4 mm Hg andeven at 10-3 mm Hg At higher pressures however theultraviolet reflectance decreases more rapidly with in-creasing evaporation time than it does at lower pres-sures An opaque coating deposited in 180 sec at 10-4

mm Hg reflects only about 40 at 220 mAt and a coatingproduced at 10-3 mm Hg shows only 12 reflectance atthe same wavelength The conclusion is that fast evapo-ration at low pressure is most important for producingaluminum films with highest reflectance in the ultra-violet and that only such films should be used to deter-mine the optical constants of aluminum

High speed of evaporation is not the only factoraffecting the reflectance of evaporated aluminum Sub-strate temperature during the evaporation film thick-ness and angle of vapor incidence have also a greatinfluence on the reflectance of evaporated coatingsespecially at shorter wavelengths

Experience has shown that most materials includingaluminum have better adherence if they are evaporatedonto heated substrates With increasing substrate tem-

TABLE II The effect of substrate temperature during theevaporation on the reflectance of about 700-A thick Al films de-posited at 1-2X 10-5 mm Hg at various deposition rates

Percent reflectanceSubstrate temperature (0C)

X (mA) 300 500 1000 1500 2000

A Deposition rate -300 Asec220 915 913 900 879 838260 922 921 915 904 884300 923 923 919 912 902400 924 923 920 917 915500 917 917 915 914 913600 910 909 908 907 905

B Deposition rate -10-15 Asec220 890 889 860 652 466260 912 910 899 735 567300 918 917 912 792 649400 920 919 917 874 779500 916 915 915 895 839600 909 909 908 896 867

TABLE III The effect of vapor incidence on the reflectance ofevaporated aluminum films of two different thicknesses

Percent reflectanceAngle of vapor incidence

O (normal incidence) 300 600t-600 A t 2000 A t 600 A t -2000 A 1-600 A t 2000

A Optimum evaporation conditions(Deposition rate 300 Asec p =1 X10-5 mm Hg)

220 915 912 914 907 904 750260 922 919 920 917 915 822300 923 922 921 918 918 855400 924 922 921 919 918 885500 917 917 916 916 915 900600 910 909 909 908 908 896

B Poor evaporation conditions(Deposition rate 10 Asec =1 X10-4 mm Hg)

220 835 659 826 605 814 318260 884 741 882 705 876 425300 905 795 904 759 896 532400 913 868 913 832 910 720500 910 888 910 875 908 813600 903 893 902 890 900 847

perature however grain size and therefore surfaceroughness increase and this results in a decreasedspecular reflectance especially of shorter wavelengthsTable II shows the effect of substrate temperature onthe reflectance of just opaque aluminum films depositedat 10-5 mm Hg The results are given for two groups offilms one deposited at a rate of about 300 Asec and theother deposited at 10-15 Asec For substrate tempera-tures higher than 50C the specular reflectance of allfilms decreases with increasing temperature The effectis however much more pronounced in the ultravioletand for films produced at low deposition rates For therapidly evaporated films (deposition rate -300 Asec)a substrate temperature increase from 300 to 200TC re-duces the reflectance at 220 mAu from 915 to about 82and at 500 mpk from 917 to 913 Films produced atlower rates show much greater reduction in reflectancewith increasing substrate temperature Films depositedat a rate of 10-15 Asec for example show a reflectancedecrease from 89 to 46 at 220 mAt and from 916 to839 at 500 m4 when the substrate temperature isincreased from 30deg to 200TC The data presented supportthe following conclusions Whenever aluminum filmsof highest ultraviolet reflectance are required the sub-strate temperature during the evaporation should notbe higher than 50TC and if higher substrate tempera-tures for increased adhesion are applied the films shouldbe evaporated as fast as possible

The effect of vapor incidence on the reflectance ofevaporated aluminum has been studied by Holland 5

He showed that diffuse reflecting surfaces are formedmore easily as the vapor incidence angle and the filmthickness are increased Table III shows more completedata of the effect of vapor incidence on the ultravioletand visible reflectance of evaporated aluminum films oftwo different thicknesses Two groups of films wereinvestigated one prepared under optimum conditions

1 L Holland J Opt Soc Am 43 376 (1953)

July 1961 EVAPORATED ALUMINUM IN

G HASS AND J E WAYLONIS

TABLE IV Calculated reflectance and transmittance of Al films evaporated under optimum conditions onto transparent substratesof n = 15 for various wavelengths as a function of film thickness (Calculated values agree with directly measured ones for film thick-nesses gt 100 A back surface antireflected)

Film Wavelength (mu)thickness 220 300 400 546 650

(A) R T R T R T R T R T

40 14 82 19 74 25 65 33 51 38 4280 33 60 43 47 52 36 60 24 63 18

120 52 40 62 27 70 19 74 12 75 9160 67 25 74 16 79 11 81 7 82 5200 763 152 815 91 849 59 856 35 854 26240 824 91 860 51 881 33 881 20 875 14280 862 54 884 31 900 19 895 11 888 08320 885 32 900 18 911 11 904 05 896 04360 898 19 909 10 917 06 909 04 900 03400 906 11 914 05 921 04 912 02 903 02500 915 03 920 01 925 lt01 915 lt01 906 lt01

(deposition rate 300 Asec at X 10-5 mm Hg) and theother deposited under poor evaporation conditions (de-position rate 10 Asec at 1X 10-4) In each group thefilms measured were about 600 A and 2000 A thick Theresults show that the reflectance decrease with increas-ing angle of vapor incidence depends upon wavelengthfilm thickness and evaporation conditions The effectis much more pronounced at shorter wavelengths andfor films prepared under poor conditions and increasesin all cases with increasing film thickness An increaseof vapor incidence angle from 00 (normal incidence) to60deg has only a very small effect on the reflectance of600-A-thick coatings but greatly decreases the re-flectance of 2000-A-thick films in the ultraviolet A2000-A-thick film deposited at normal incidence underoptimum conditions reflects almost as well as a justopaque coating even at 220 myi The reflectance at220 mju however drops for a 2000-A thick film from912 to 750 when the vapor incidence angle is in-creased from 0deg to 600 This effect is much greater forfilms produced under poor evaporation conditions Here2000-A thick films evaporated at normal incidencealready show a much lower reflectance than just opaquecoatings and the increase in vapor incidence angle from0deg to 600 drops the reflectance of a 2000-A thick filmfrom 659 to 312 at 220 mu and from 888 to 813at 500 mu This indicates that mirror coatings for theultraviolet should be evaporated at close to normalincidence and that the coatings should not be thicker

than absolutely necessarycidence are unavoidable

if higher angles of vapor in-

B Reflectance and Transmittance ofSemitransparent Films

Semitransparent aluminum films are especially im-portant for interferometry in the ultraviolet becausethey show in this wavelength region higher reflectanceand lower absorptance than any other metallic coatingTable IV shows the calculated reflectance and trans-mittance of evaporated aluminum for various wave-lengths and various film thicknesses The optical con-stants n and k listed in Table I were used for thecalculations For films produced with deposition rates ofmore than 100 Asec at 1 X 10-5 mm Hg the calculatedR and T values agreed well with the directly measuredones down to film thicknesses of about 100 A The re-sults show that for the same film thickness the trans-mittance increases strongly with decreasing wavelengthA 160-A thick film for example transmits 25 at220 mAu and only 5 at 546 mit

Films deposited at low deposition rates show muchhigher absorptance and lower reflectance and are there-fore unsuitable for interferometry in the ultraviolet Apreviously reported statements that semitransparent Alfilms produced at high deposition rates show a mottledappearance could not be verified All films producedunder the above described conditions were extremelyuniform in appearance

722 Vol 51

• Optical properties of Al at 300nmpdf
• Al_Reflection_Transmissionpdf
• Optical Properties of Aluminumpdf

JOURNAL OF THE OPTICAL SOCIETY OF AMERICA

Optical Constants and Reflectance and Transmittance of EvaporatedAluminum in the Visible and Ultraviolet

G HASS AND J E WAYLONISU S Army Engineer Research and Development Laboratories Fort Belvoir Virginia

(Received February 13 1961)

The paper presents new measurements of the optical constants and of the reflectance and transmittanceof evaporated aluminum films produced from the purest grade of aluminum with extremely fast evaporationat about 10-5 mm Hg The optical constants were determined by Drudes polarimetric method from opaquefilms and by reflectance transmittance and true thickness measurements of semitransparent films Thefollowing results were obtained

X (my)nkRo

220 260 300 340 380 436 492 546 578 650014 019 025 031 037 047 064 082 093 130235 285 333 380 425 484 550 599 633 711

918 920 921 923 926 926 922 916 915 907

For extremely fast evaporated films the optical constants were found to remain unchanged down to filmthicknesses of less than 100 A Data are given on the reflectance and transmittance of semitransparentaluminum films on glass and fused quartz as a function of wavelength from 220 my to 650 mgi

INTRODUCTION

EVAPORATED aluminum films are the most fre-quently used coatings for front-surface mirrors and

for interferometry in the ultraviolet Measurements oftheir optical properties have therefore been reportedby various authors1 -7 But only recently it was dis-covered that films with highest reflectance especially inthe ultraviolet and vacuum ultraviolet can only beproduced by extremely fast evaporation and by the useof the purest grade of aluminum8 9 It is the purpose ofthis paper to report on new measurements of the opticalconstants and of the reflectance and transmittance ofevaporated aluminum films produced from purest alumi-num with extremely high rates of deposition and todiscuss in addition the effect of speed of evaporationsubstrate temperature and vapor-incidence angle on thereflectance of opaque aluminum in the visible andultraviolet

EXPERIMENTAL TECHNIQUES

The vacuum apparatus used for producing aluminumfilms at extremely high rates of deposition has beendescribed in a previous paper1 The evaporations weremade in an 18-in vacuum system in which a pressure of1X 10-5 mm Hg could be maintained during the filmdeposition Ultrasil and glass plates mounted about18 in above the evaporation sources were used as sub-strates They were cleaned by a high-voltage dc glow

Presented at the Meetings of the Optical Society of AmericaBoston Massachusetts October 12-141960

I J Strong Astrophys J 83 401 (1936)2 H OBryan J Opt Soc Am 26 122 (1936)3 W Walkenhorst Z tech Physik 22 14 (1941)4 G Hass Optik 1 2 (1946)5 M F Crawford W M Gray A L Schwawlow and F M

Kelly J Opt Soc Am 39 888 (1949)6 L G Schulz J Opt Soc Am 44 357 (1954)7 L G Schulz and F R Tangherlini J Opt Soc Am 44 362

(1954)8 G Hass J Opt Soc Am 45 945 (1955)9 G Hass W R Hunter and R Tousey J Opt Soc Am

46 1009 (1956) ibid 47 1070 (1957)

discharge using 90 ma at 5 kv for 1-5 min A Japanese-fan-type shutter was used to avoid contamination of thesubstrates during the outgassing of the evaporationsources and their charges The shutter which was placedclose to the targets was opened when the evaporationreached constant speed and it was closed when a filmof desired thickness was obtained The evaporationsource for producing aluminum films at high-depositionrates consisted of six helical coils of 62-mil tungstenwire which were heated simultaneously They wereoutgassed and freed from impurities before beingcharged with aluminum With this arrangement alumi-num films could be deposited at a deposition rate ofmore than 500 Asec

The reflectance R and transmittance T of aluminumfilms were measured from 220 to 1000 myz with a modifiedBeckman model DK spectrophotometer All reflectanceand transmittance values reported in this paper weredetermined with an accuracy of about 03o of theincident light and the measurements were made at closeto normal incidence The true thickness t of the filmswas measured by a multiple-beam interferometermethod as described by Tolansky 0

The measured R and T values were corrected for theeffect of the back-surface reflectance of the substrateplates and then used to determine the optical constantsn and k = nK of semitransparent aluminum films Toobtain the optical constants from R T and I measure-ments a graphical method was applied which uses agreat number of curves displaying R and T as a functionof tX for films of various n and k on a nonabsorbing sub-strate of n= 15 The graphs were calculated from theequations developed by Hadley and Dennison 2 Theoptical constants of opaque aluminum films were de-

10 S Tolansky Multiple Beam Interferometry of Films andSurfaces (Clarenden Press Oxford England 1948)

1L N Hadley and D M Dennison J Opt Soc Am 37451 (1947)

12 The authors are indebted to L N Hadley and his co-workersfor the preparation of the graphs used in this paper for the de-termination of n and k of semitransparent aluminum films

719

VOLUME 51 NUMBER 7 JULY 1961

G HASS AND J E WAYLONIS V 51

termined by Drudes polarimetric method from 435 to650 mA The theory and various applications of thismethod have been discussed extensively in the litera-ture In the arrangement used here monochromaticlight polarized at an azimuth angle of 450 was reflectedat an angle = 70deg from the aluminum film and theellipticity produced by reflection from the metal coatingwas determined A Babinet-Soleil compensator and aGlan-Thomson prism were employed to measure thephase difference A of the two components of the ellipti-cally polarized light and of the azimuth V of the restoredplane of linear polarization Drudes equation in itsrigorous form was used to calculate the optical con-stants from A and o measurements

To obtain the true optical constants with Drudesequations A and b of an oxide-free aluminum have tobe used Since all measurements were made after thealuminum films had been exposed to air for a certainamount of time the measured A and values werecorrected for the effect of the oxide film formed onaluminum in air up to the time the measurements weremade Oxidation curves for aluminum published in arecent paper were used to calculate the corrections forA and sb The maximal errors of the optical constantsdetermined by Drudes polarimetric method amountedto about -L4 for n and 462 for k

RESULTS

A Optical Constants and Reflectanceof Aluminum Films

Table I lists the optical constants and the calculatedand measured reflectance of aluminum films All films

TABLE I The optical constants and the calculated and measuredreflectance of evaporated aluminum produced under optimumevaporation conditions from 200 to 650 mu

ReflectanceX (mU) n k calc meas

Calculated from R T and t220 014 235 918 915240 016 260 921 919260 019 285 920 922280 022 313 922 923300 025 335 921 923320 028 356 922 924340 031 380 923 925360 034 401 924 925380 037 425 926 925400 040 445 926 924450 051 500 925 922546 080 592 916 916

Drudes polarimetric method436 047 484 926 923492 064 550 922 919546 082 599 916 916578 093 633 915 914650 130 711 907 905

P P Drude Wiedem Ann 64 159 (1898)14 P H Berning G Hass and R P Madden J Opt Soc Am

50 586 (1960)

100

90

50

40

30

II

) _

IIo

IIAl

20 20_

I-2xlo-5mmH 1

1a 3 I4 5 6(0

X in mu

-6 sec

-135 sec

-180 secI

1-2x O-4mmHg

3 4 5 600X in mu

4 n mu

FIG 1 Effect of pressure and speed of evaporation on the reflect-ance of 600 to 700-A thick Al films from 220 to 600 mu

were produced under optimum conditions using purestgrade aluminum (better than 9999) and depositionrates of more than 300 Asec at 1X 10-5 mm Hg Ahigh-speed shutter was employed to produce uniformsemitransparent films at high deposition rates

In the wavelength region below 436 mu the opticalconstants were determined only from reflectance trans-mittance and true thickness measurements of semi-transparent films while at longer wavelengths bothdescribed methods were used The optical constants ob-tained from spectrophotometric measurements of semi-transparent films show good agreement with thosedetermined by Drudes polarimetric method on opaquecoatings It was also found that the optical constants ofaluminum films evaporated at an extremely high rateremain unchanged down to film thicknesses of about100 A The values for n and k decrease rather constantlywith decreasing wavelength The reflectance of opaquealuminum calculated from n and k agrees well with thedirectly measured one and is about 2 higher than theresults reported by Schulz and Tangherlini7 Films withsuch high reflectance in the ultraviolet (915 to 925)can only be produced by fast evaporation at low pres-sures Such films are in addition more compact thanslowly deposited ones and show therefore less oxidationand very small decreases in reflectance during exposureto air At 220 mu where the effect of aging is mostpronounced one-year-old aluminum mirrors still showedabout 90 reflectance The 15 reflectance decreaseat 220 mgt over a one-year period can be explained bythe formation of a 25- to 30-A-thick oxide film Such athin oxide film has only a very small effect on the re-flectance of aluminum in the visible (less than 03reflectance decrease at 550 mu)

The fact that high speed of evaporation is the mostimportant basic factor for producing aluminum filmswith highest reflectance in the ultraviolet is illustratedin Fig 1 The data presented are for three groups offilms evaporated at three different pressures The firstgroup was deposited at 1-2X 10- mm Hg the secondat 1-2X 10-4 mm Hg and the third at 1X10-3 mm Hg

720 Vol 51

I

VISIBLE AND ULTRAVIOLET 721

All films were just opaque or 600-700 A thick and con-densed on glass plates at a distance of about 20 in fromthe evaporation sources Additional filaments wereadded to obtain high deposition rates at 1 X 10-3 mm HgAll reflectances were measured within one hour after thedeposition of the coatings The time for preparing theopaque coatings is noted for each curve The curvesshow clearly that the speed of evaporation has a greatinfluence on the ultraviolet reflectance of aluminum anda much smaller effect on its reflectance at longer wave-lengths An increase of the evaporation time for anopaque coating from 7 sec (which is still not the opti-mum evaporation speed) to 180 sec at 1-2X 10-5 mmHg decreases the reflectance at 220 mgu from 91 to62 but produces only a reflectance decrease of 1 at550 mt The influence of firing speed is even greater athigher pressures However films with very high ultra-violet reflectance can still be made at 10-4 mm Hg andeven at 10-3 mm Hg At higher pressures however theultraviolet reflectance decreases more rapidly with in-creasing evaporation time than it does at lower pres-sures An opaque coating deposited in 180 sec at 10-4

mm Hg reflects only about 40 at 220 mAt and a coatingproduced at 10-3 mm Hg shows only 12 reflectance atthe same wavelength The conclusion is that fast evapo-ration at low pressure is most important for producingaluminum films with highest reflectance in the ultra-violet and that only such films should be used to deter-mine the optical constants of aluminum

High speed of evaporation is not the only factoraffecting the reflectance of evaporated aluminum Sub-strate temperature during the evaporation film thick-ness and angle of vapor incidence have also a greatinfluence on the reflectance of evaporated coatingsespecially at shorter wavelengths

Experience has shown that most materials includingaluminum have better adherence if they are evaporatedonto heated substrates With increasing substrate tem-

TABLE II The effect of substrate temperature during theevaporation on the reflectance of about 700-A thick Al films de-posited at 1-2X 10-5 mm Hg at various deposition rates

Percent reflectanceSubstrate temperature (0C)

X (mA) 300 500 1000 1500 2000

A Deposition rate -300 Asec220 915 913 900 879 838260 922 921 915 904 884300 923 923 919 912 902400 924 923 920 917 915500 917 917 915 914 913600 910 909 908 907 905

B Deposition rate -10-15 Asec220 890 889 860 652 466260 912 910 899 735 567300 918 917 912 792 649400 920 919 917 874 779500 916 915 915 895 839600 909 909 908 896 867

TABLE III The effect of vapor incidence on the reflectance ofevaporated aluminum films of two different thicknesses

Percent reflectanceAngle of vapor incidence

O (normal incidence) 300 600t-600 A t 2000 A t 600 A t -2000 A 1-600 A t 2000

A Optimum evaporation conditions(Deposition rate 300 Asec p =1 X10-5 mm Hg)

220 915 912 914 907 904 750260 922 919 920 917 915 822300 923 922 921 918 918 855400 924 922 921 919 918 885500 917 917 916 916 915 900600 910 909 909 908 908 896

B Poor evaporation conditions(Deposition rate 10 Asec =1 X10-4 mm Hg)

220 835 659 826 605 814 318260 884 741 882 705 876 425300 905 795 904 759 896 532400 913 868 913 832 910 720500 910 888 910 875 908 813600 903 893 902 890 900 847

perature however grain size and therefore surfaceroughness increase and this results in a decreasedspecular reflectance especially of shorter wavelengthsTable II shows the effect of substrate temperature onthe reflectance of just opaque aluminum films depositedat 10-5 mm Hg The results are given for two groups offilms one deposited at a rate of about 300 Asec and theother deposited at 10-15 Asec For substrate tempera-tures higher than 50C the specular reflectance of allfilms decreases with increasing temperature The effectis however much more pronounced in the ultravioletand for films produced at low deposition rates For therapidly evaporated films (deposition rate -300 Asec)a substrate temperature increase from 300 to 200TC re-duces the reflectance at 220 mAu from 915 to about 82and at 500 mpk from 917 to 913 Films produced atlower rates show much greater reduction in reflectancewith increasing substrate temperature Films depositedat a rate of 10-15 Asec for example show a reflectancedecrease from 89 to 46 at 220 mAt and from 916 to839 at 500 m4 when the substrate temperature isincreased from 30deg to 200TC The data presented supportthe following conclusions Whenever aluminum filmsof highest ultraviolet reflectance are required the sub-strate temperature during the evaporation should notbe higher than 50TC and if higher substrate tempera-tures for increased adhesion are applied the films shouldbe evaporated as fast as possible

The effect of vapor incidence on the reflectance ofevaporated aluminum has been studied by Holland 5

He showed that diffuse reflecting surfaces are formedmore easily as the vapor incidence angle and the filmthickness are increased Table III shows more completedata of the effect of vapor incidence on the ultravioletand visible reflectance of evaporated aluminum films oftwo different thicknesses Two groups of films wereinvestigated one prepared under optimum conditions

1 L Holland J Opt Soc Am 43 376 (1953)

July 1961 EVAPORATED ALUMINUM IN

G HASS AND J E WAYLONIS

TABLE IV Calculated reflectance and transmittance of Al films evaporated under optimum conditions onto transparent substratesof n = 15 for various wavelengths as a function of film thickness (Calculated values agree with directly measured ones for film thick-nesses gt 100 A back surface antireflected)

Film Wavelength (mu)thickness 220 300 400 546 650

(A) R T R T R T R T R T

40 14 82 19 74 25 65 33 51 38 4280 33 60 43 47 52 36 60 24 63 18

120 52 40 62 27 70 19 74 12 75 9160 67 25 74 16 79 11 81 7 82 5200 763 152 815 91 849 59 856 35 854 26240 824 91 860 51 881 33 881 20 875 14280 862 54 884 31 900 19 895 11 888 08320 885 32 900 18 911 11 904 05 896 04360 898 19 909 10 917 06 909 04 900 03400 906 11 914 05 921 04 912 02 903 02500 915 03 920 01 925 lt01 915 lt01 906 lt01

(deposition rate 300 Asec at X 10-5 mm Hg) and theother deposited under poor evaporation conditions (de-position rate 10 Asec at 1X 10-4) In each group thefilms measured were about 600 A and 2000 A thick Theresults show that the reflectance decrease with increas-ing angle of vapor incidence depends upon wavelengthfilm thickness and evaporation conditions The effectis much more pronounced at shorter wavelengths andfor films prepared under poor conditions and increasesin all cases with increasing film thickness An increaseof vapor incidence angle from 00 (normal incidence) to60deg has only a very small effect on the reflectance of600-A-thick coatings but greatly decreases the re-flectance of 2000-A-thick films in the ultraviolet A2000-A-thick film deposited at normal incidence underoptimum conditions reflects almost as well as a justopaque coating even at 220 myi The reflectance at220 mju however drops for a 2000-A thick film from912 to 750 when the vapor incidence angle is in-creased from 0deg to 600 This effect is much greater forfilms produced under poor evaporation conditions Here2000-A thick films evaporated at normal incidencealready show a much lower reflectance than just opaquecoatings and the increase in vapor incidence angle from0deg to 600 drops the reflectance of a 2000-A thick filmfrom 659 to 312 at 220 mu and from 888 to 813at 500 mu This indicates that mirror coatings for theultraviolet should be evaporated at close to normalincidence and that the coatings should not be thicker

than absolutely necessarycidence are unavoidable

if higher angles of vapor in-

B Reflectance and Transmittance ofSemitransparent Films

Semitransparent aluminum films are especially im-portant for interferometry in the ultraviolet becausethey show in this wavelength region higher reflectanceand lower absorptance than any other metallic coatingTable IV shows the calculated reflectance and trans-mittance of evaporated aluminum for various wave-lengths and various film thicknesses The optical con-stants n and k listed in Table I were used for thecalculations For films produced with deposition rates ofmore than 100 Asec at 1 X 10-5 mm Hg the calculatedR and T values agreed well with the directly measuredones down to film thicknesses of about 100 A The re-sults show that for the same film thickness the trans-mittance increases strongly with decreasing wavelengthA 160-A thick film for example transmits 25 at220 mAu and only 5 at 546 mit

Films deposited at low deposition rates show muchhigher absorptance and lower reflectance and are there-fore unsuitable for interferometry in the ultraviolet Apreviously reported statements that semitransparent Alfilms produced at high deposition rates show a mottledappearance could not be verified All films producedunder the above described conditions were extremelyuniform in appearance

722 Vol 51

• Optical properties of Al at 300nmpdf
• Al_Reflection_Transmissionpdf
• Optical Properties of Aluminumpdf

G HASS AND J E WAYLONIS V 51

termined by Drudes polarimetric method from 435 to650 mA The theory and various applications of thismethod have been discussed extensively in the litera-ture In the arrangement used here monochromaticlight polarized at an azimuth angle of 450 was reflectedat an angle = 70deg from the aluminum film and theellipticity produced by reflection from the metal coatingwas determined A Babinet-Soleil compensator and aGlan-Thomson prism were employed to measure thephase difference A of the two components of the ellipti-cally polarized light and of the azimuth V of the restoredplane of linear polarization Drudes equation in itsrigorous form was used to calculate the optical con-stants from A and o measurements

To obtain the true optical constants with Drudesequations A and b of an oxide-free aluminum have tobe used Since all measurements were made after thealuminum films had been exposed to air for a certainamount of time the measured A and values werecorrected for the effect of the oxide film formed onaluminum in air up to the time the measurements weremade Oxidation curves for aluminum published in arecent paper were used to calculate the corrections forA and sb The maximal errors of the optical constantsdetermined by Drudes polarimetric method amountedto about -L4 for n and 462 for k

RESULTS

A Optical Constants and Reflectanceof Aluminum Films

Table I lists the optical constants and the calculatedand measured reflectance of aluminum films All films

TABLE I The optical constants and the calculated and measuredreflectance of evaporated aluminum produced under optimumevaporation conditions from 200 to 650 mu

ReflectanceX (mU) n k calc meas

Calculated from R T and t220 014 235 918 915240 016 260 921 919260 019 285 920 922280 022 313 922 923300 025 335 921 923320 028 356 922 924340 031 380 923 925360 034 401 924 925380 037 425 926 925400 040 445 926 924450 051 500 925 922546 080 592 916 916

Drudes polarimetric method436 047 484 926 923492 064 550 922 919546 082 599 916 916578 093 633 915 914650 130 711 907 905

P P Drude Wiedem Ann 64 159 (1898)14 P H Berning G Hass and R P Madden J Opt Soc Am

50 586 (1960)

100

90

50

40

30

II

) _

IIo

IIAl

20 20_

I-2xlo-5mmH 1

1a 3 I4 5 6(0

X in mu

-6 sec

-135 sec

-180 secI

1-2x O-4mmHg

3 4 5 600X in mu

4 n mu

FIG 1 Effect of pressure and speed of evaporation on the reflect-ance of 600 to 700-A thick Al films from 220 to 600 mu

were produced under optimum conditions using purestgrade aluminum (better than 9999) and depositionrates of more than 300 Asec at 1X 10-5 mm Hg Ahigh-speed shutter was employed to produce uniformsemitransparent films at high deposition rates

In the wavelength region below 436 mu the opticalconstants were determined only from reflectance trans-mittance and true thickness measurements of semi-transparent films while at longer wavelengths bothdescribed methods were used The optical constants ob-tained from spectrophotometric measurements of semi-transparent films show good agreement with thosedetermined by Drudes polarimetric method on opaquecoatings It was also found that the optical constants ofaluminum films evaporated at an extremely high rateremain unchanged down to film thicknesses of about100 A The values for n and k decrease rather constantlywith decreasing wavelength The reflectance of opaquealuminum calculated from n and k agrees well with thedirectly measured one and is about 2 higher than theresults reported by Schulz and Tangherlini7 Films withsuch high reflectance in the ultraviolet (915 to 925)can only be produced by fast evaporation at low pres-sures Such films are in addition more compact thanslowly deposited ones and show therefore less oxidationand very small decreases in reflectance during exposureto air At 220 mu where the effect of aging is mostpronounced one-year-old aluminum mirrors still showedabout 90 reflectance The 15 reflectance decreaseat 220 mgt over a one-year period can be explained bythe formation of a 25- to 30-A-thick oxide film Such athin oxide film has only a very small effect on the re-flectance of aluminum in the visible (less than 03reflectance decrease at 550 mu)

The fact that high speed of evaporation is the mostimportant basic factor for producing aluminum filmswith highest reflectance in the ultraviolet is illustratedin Fig 1 The data presented are for three groups offilms evaporated at three different pressures The firstgroup was deposited at 1-2X 10- mm Hg the secondat 1-2X 10-4 mm Hg and the third at 1X10-3 mm Hg

720 Vol 51

I

VISIBLE AND ULTRAVIOLET 721

All films were just opaque or 600-700 A thick and con-densed on glass plates at a distance of about 20 in fromthe evaporation sources Additional filaments wereadded to obtain high deposition rates at 1 X 10-3 mm HgAll reflectances were measured within one hour after thedeposition of the coatings The time for preparing theopaque coatings is noted for each curve The curvesshow clearly that the speed of evaporation has a greatinfluence on the ultraviolet reflectance of aluminum anda much smaller effect on its reflectance at longer wave-lengths An increase of the evaporation time for anopaque coating from 7 sec (which is still not the opti-mum evaporation speed) to 180 sec at 1-2X 10-5 mmHg decreases the reflectance at 220 mgu from 91 to62 but produces only a reflectance decrease of 1 at550 mt The influence of firing speed is even greater athigher pressures However films with very high ultra-violet reflectance can still be made at 10-4 mm Hg andeven at 10-3 mm Hg At higher pressures however theultraviolet reflectance decreases more rapidly with in-creasing evaporation time than it does at lower pres-sures An opaque coating deposited in 180 sec at 10-4

mm Hg reflects only about 40 at 220 mAt and a coatingproduced at 10-3 mm Hg shows only 12 reflectance atthe same wavelength The conclusion is that fast evapo-ration at low pressure is most important for producingaluminum films with highest reflectance in the ultra-violet and that only such films should be used to deter-mine the optical constants of aluminum

High speed of evaporation is not the only factoraffecting the reflectance of evaporated aluminum Sub-strate temperature during the evaporation film thick-ness and angle of vapor incidence have also a greatinfluence on the reflectance of evaporated coatingsespecially at shorter wavelengths

Experience has shown that most materials includingaluminum have better adherence if they are evaporatedonto heated substrates With increasing substrate tem-

TABLE II The effect of substrate temperature during theevaporation on the reflectance of about 700-A thick Al films de-posited at 1-2X 10-5 mm Hg at various deposition rates

Percent reflectanceSubstrate temperature (0C)

X (mA) 300 500 1000 1500 2000

A Deposition rate -300 Asec220 915 913 900 879 838260 922 921 915 904 884300 923 923 919 912 902400 924 923 920 917 915500 917 917 915 914 913600 910 909 908 907 905

B Deposition rate -10-15 Asec220 890 889 860 652 466260 912 910 899 735 567300 918 917 912 792 649400 920 919 917 874 779500 916 915 915 895 839600 909 909 908 896 867

TABLE III The effect of vapor incidence on the reflectance ofevaporated aluminum films of two different thicknesses

Percent reflectanceAngle of vapor incidence

O (normal incidence) 300 600t-600 A t 2000 A t 600 A t -2000 A 1-600 A t 2000

A Optimum evaporation conditions(Deposition rate 300 Asec p =1 X10-5 mm Hg)

220 915 912 914 907 904 750260 922 919 920 917 915 822300 923 922 921 918 918 855400 924 922 921 919 918 885500 917 917 916 916 915 900600 910 909 909 908 908 896

B Poor evaporation conditions(Deposition rate 10 Asec =1 X10-4 mm Hg)

220 835 659 826 605 814 318260 884 741 882 705 876 425300 905 795 904 759 896 532400 913 868 913 832 910 720500 910 888 910 875 908 813600 903 893 902 890 900 847

perature however grain size and therefore surfaceroughness increase and this results in a decreasedspecular reflectance especially of shorter wavelengthsTable II shows the effect of substrate temperature onthe reflectance of just opaque aluminum films depositedat 10-5 mm Hg The results are given for two groups offilms one deposited at a rate of about 300 Asec and theother deposited at 10-15 Asec For substrate tempera-tures higher than 50C the specular reflectance of allfilms decreases with increasing temperature The effectis however much more pronounced in the ultravioletand for films produced at low deposition rates For therapidly evaporated films (deposition rate -300 Asec)a substrate temperature increase from 300 to 200TC re-duces the reflectance at 220 mAu from 915 to about 82and at 500 mpk from 917 to 913 Films produced atlower rates show much greater reduction in reflectancewith increasing substrate temperature Films depositedat a rate of 10-15 Asec for example show a reflectancedecrease from 89 to 46 at 220 mAt and from 916 to839 at 500 m4 when the substrate temperature isincreased from 30deg to 200TC The data presented supportthe following conclusions Whenever aluminum filmsof highest ultraviolet reflectance are required the sub-strate temperature during the evaporation should notbe higher than 50TC and if higher substrate tempera-tures for increased adhesion are applied the films shouldbe evaporated as fast as possible

The effect of vapor incidence on the reflectance ofevaporated aluminum has been studied by Holland 5

He showed that diffuse reflecting surfaces are formedmore easily as the vapor incidence angle and the filmthickness are increased Table III shows more completedata of the effect of vapor incidence on the ultravioletand visible reflectance of evaporated aluminum films oftwo different thicknesses Two groups of films wereinvestigated one prepared under optimum conditions

1 L Holland J Opt Soc Am 43 376 (1953)

July 1961 EVAPORATED ALUMINUM IN

G HASS AND J E WAYLONIS

TABLE IV Calculated reflectance and transmittance of Al films evaporated under optimum conditions onto transparent substratesof n = 15 for various wavelengths as a function of film thickness (Calculated values agree with directly measured ones for film thick-nesses gt 100 A back surface antireflected)

Film Wavelength (mu)thickness 220 300 400 546 650

(A) R T R T R T R T R T

40 14 82 19 74 25 65 33 51 38 4280 33 60 43 47 52 36 60 24 63 18

120 52 40 62 27 70 19 74 12 75 9160 67 25 74 16 79 11 81 7 82 5200 763 152 815 91 849 59 856 35 854 26240 824 91 860 51 881 33 881 20 875 14280 862 54 884 31 900 19 895 11 888 08320 885 32 900 18 911 11 904 05 896 04360 898 19 909 10 917 06 909 04 900 03400 906 11 914 05 921 04 912 02 903 02500 915 03 920 01 925 lt01 915 lt01 906 lt01

(deposition rate 300 Asec at X 10-5 mm Hg) and theother deposited under poor evaporation conditions (de-position rate 10 Asec at 1X 10-4) In each group thefilms measured were about 600 A and 2000 A thick Theresults show that the reflectance decrease with increas-ing angle of vapor incidence depends upon wavelengthfilm thickness and evaporation conditions The effectis much more pronounced at shorter wavelengths andfor films prepared under poor conditions and increasesin all cases with increasing film thickness An increaseof vapor incidence angle from 00 (normal incidence) to60deg has only a very small effect on the reflectance of600-A-thick coatings but greatly decreases the re-flectance of 2000-A-thick films in the ultraviolet A2000-A-thick film deposited at normal incidence underoptimum conditions reflects almost as well as a justopaque coating even at 220 myi The reflectance at220 mju however drops for a 2000-A thick film from912 to 750 when the vapor incidence angle is in-creased from 0deg to 600 This effect is much greater forfilms produced under poor evaporation conditions Here2000-A thick films evaporated at normal incidencealready show a much lower reflectance than just opaquecoatings and the increase in vapor incidence angle from0deg to 600 drops the reflectance of a 2000-A thick filmfrom 659 to 312 at 220 mu and from 888 to 813at 500 mu This indicates that mirror coatings for theultraviolet should be evaporated at close to normalincidence and that the coatings should not be thicker

than absolutely necessarycidence are unavoidable

if higher angles of vapor in-

B Reflectance and Transmittance ofSemitransparent Films

Semitransparent aluminum films are especially im-portant for interferometry in the ultraviolet becausethey show in this wavelength region higher reflectanceand lower absorptance than any other metallic coatingTable IV shows the calculated reflectance and trans-mittance of evaporated aluminum for various wave-lengths and various film thicknesses The optical con-stants n and k listed in Table I were used for thecalculations For films produced with deposition rates ofmore than 100 Asec at 1 X 10-5 mm Hg the calculatedR and T values agreed well with the directly measuredones down to film thicknesses of about 100 A The re-sults show that for the same film thickness the trans-mittance increases strongly with decreasing wavelengthA 160-A thick film for example transmits 25 at220 mAu and only 5 at 546 mit

Films deposited at low deposition rates show muchhigher absorptance and lower reflectance and are there-fore unsuitable for interferometry in the ultraviolet Apreviously reported statements that semitransparent Alfilms produced at high deposition rates show a mottledappearance could not be verified All films producedunder the above described conditions were extremelyuniform in appearance

722 Vol 51

• Optical properties of Al at 300nmpdf
• Al_Reflection_Transmissionpdf
• Optical Properties of Aluminumpdf

VISIBLE AND ULTRAVIOLET 721

All films were just opaque or 600-700 A thick and con-densed on glass plates at a distance of about 20 in fromthe evaporation sources Additional filaments wereadded to obtain high deposition rates at 1 X 10-3 mm HgAll reflectances were measured within one hour after thedeposition of the coatings The time for preparing theopaque coatings is noted for each curve The curvesshow clearly that the speed of evaporation has a greatinfluence on the ultraviolet reflectance of aluminum anda much smaller effect on its reflectance at longer wave-lengths An increase of the evaporation time for anopaque coating from 7 sec (which is still not the opti-mum evaporation speed) to 180 sec at 1-2X 10-5 mmHg decreases the reflectance at 220 mgu from 91 to62 but produces only a reflectance decrease of 1 at550 mt The influence of firing speed is even greater athigher pressures However films with very high ultra-violet reflectance can still be made at 10-4 mm Hg andeven at 10-3 mm Hg At higher pressures however theultraviolet reflectance decreases more rapidly with in-creasing evaporation time than it does at lower pres-sures An opaque coating deposited in 180 sec at 10-4

mm Hg reflects only about 40 at 220 mAt and a coatingproduced at 10-3 mm Hg shows only 12 reflectance atthe same wavelength The conclusion is that fast evapo-ration at low pressure is most important for producingaluminum films with highest reflectance in the ultra-violet and that only such films should be used to deter-mine the optical constants of aluminum

High speed of evaporation is not the only factoraffecting the reflectance of evaporated aluminum Sub-strate temperature during the evaporation film thick-ness and angle of vapor incidence have also a greatinfluence on the reflectance of evaporated coatingsespecially at shorter wavelengths

Experience has shown that most materials includingaluminum have better adherence if they are evaporatedonto heated substrates With increasing substrate tem-

TABLE II The effect of substrate temperature during theevaporation on the reflectance of about 700-A thick Al films de-posited at 1-2X 10-5 mm Hg at various deposition rates

Percent reflectanceSubstrate temperature (0C)

X (mA) 300 500 1000 1500 2000

A Deposition rate -300 Asec220 915 913 900 879 838260 922 921 915 904 884300 923 923 919 912 902400 924 923 920 917 915500 917 917 915 914 913600 910 909 908 907 905

B Deposition rate -10-15 Asec220 890 889 860 652 466260 912 910 899 735 567300 918 917 912 792 649400 920 919 917 874 779500 916 915 915 895 839600 909 909 908 896 867

TABLE III The effect of vapor incidence on the reflectance ofevaporated aluminum films of two different thicknesses

Percent reflectanceAngle of vapor incidence

O (normal incidence) 300 600t-600 A t 2000 A t 600 A t -2000 A 1-600 A t 2000

A Optimum evaporation conditions(Deposition rate 300 Asec p =1 X10-5 mm Hg)

220 915 912 914 907 904 750260 922 919 920 917 915 822300 923 922 921 918 918 855400 924 922 921 919 918 885500 917 917 916 916 915 900600 910 909 909 908 908 896

B Poor evaporation conditions(Deposition rate 10 Asec =1 X10-4 mm Hg)

220 835 659 826 605 814 318260 884 741 882 705 876 425300 905 795 904 759 896 532400 913 868 913 832 910 720500 910 888 910 875 908 813600 903 893 902 890 900 847

perature however grain size and therefore surfaceroughness increase and this results in a decreasedspecular reflectance especially of shorter wavelengthsTable II shows the effect of substrate temperature onthe reflectance of just opaque aluminum films depositedat 10-5 mm Hg The results are given for two groups offilms one deposited at a rate of about 300 Asec and theother deposited at 10-15 Asec For substrate tempera-tures higher than 50C the specular reflectance of allfilms decreases with increasing temperature The effectis however much more pronounced in the ultravioletand for films produced at low deposition rates For therapidly evaporated films (deposition rate -300 Asec)a substrate temperature increase from 300 to 200TC re-duces the reflectance at 220 mAu from 915 to about 82and at 500 mpk from 917 to 913 Films produced atlower rates show much greater reduction in reflectancewith increasing substrate temperature Films depositedat a rate of 10-15 Asec for example show a reflectancedecrease from 89 to 46 at 220 mAt and from 916 to839 at 500 m4 when the substrate temperature isincreased from 30deg to 200TC The data presented supportthe following conclusions Whenever aluminum filmsof highest ultraviolet reflectance are required the sub-strate temperature during the evaporation should notbe higher than 50TC and if higher substrate tempera-tures for increased adhesion are applied the films shouldbe evaporated as fast as possible

The effect of vapor incidence on the reflectance ofevaporated aluminum has been studied by Holland 5

He showed that diffuse reflecting surfaces are formedmore easily as the vapor incidence angle and the filmthickness are increased Table III shows more completedata of the effect of vapor incidence on the ultravioletand visible reflectance of evaporated aluminum films oftwo different thicknesses Two groups of films wereinvestigated one prepared under optimum conditions

1 L Holland J Opt Soc Am 43 376 (1953)

July 1961 EVAPORATED ALUMINUM IN

G HASS AND J E WAYLONIS

TABLE IV Calculated reflectance and transmittance of Al films evaporated under optimum conditions onto transparent substratesof n = 15 for various wavelengths as a function of film thickness (Calculated values agree with directly measured ones for film thick-nesses gt 100 A back surface antireflected)

Film Wavelength (mu)thickness 220 300 400 546 650

(A) R T R T R T R T R T

40 14 82 19 74 25 65 33 51 38 4280 33 60 43 47 52 36 60 24 63 18

120 52 40 62 27 70 19 74 12 75 9160 67 25 74 16 79 11 81 7 82 5200 763 152 815 91 849 59 856 35 854 26240 824 91 860 51 881 33 881 20 875 14280 862 54 884 31 900 19 895 11 888 08320 885 32 900 18 911 11 904 05 896 04360 898 19 909 10 917 06 909 04 900 03400 906 11 914 05 921 04 912 02 903 02500 915 03 920 01 925 lt01 915 lt01 906 lt01

(deposition rate 300 Asec at X 10-5 mm Hg) and theother deposited under poor evaporation conditions (de-position rate 10 Asec at 1X 10-4) In each group thefilms measured were about 600 A and 2000 A thick Theresults show that the reflectance decrease with increas-ing angle of vapor incidence depends upon wavelengthfilm thickness and evaporation conditions The effectis much more pronounced at shorter wavelengths andfor films prepared under poor conditions and increasesin all cases with increasing film thickness An increaseof vapor incidence angle from 00 (normal incidence) to60deg has only a very small effect on the reflectance of600-A-thick coatings but greatly decreases the re-flectance of 2000-A-thick films in the ultraviolet A2000-A-thick film deposited at normal incidence underoptimum conditions reflects almost as well as a justopaque coating even at 220 myi The reflectance at220 mju however drops for a 2000-A thick film from912 to 750 when the vapor incidence angle is in-creased from 0deg to 600 This effect is much greater forfilms produced under poor evaporation conditions Here2000-A thick films evaporated at normal incidencealready show a much lower reflectance than just opaquecoatings and the increase in vapor incidence angle from0deg to 600 drops the reflectance of a 2000-A thick filmfrom 659 to 312 at 220 mu and from 888 to 813at 500 mu This indicates that mirror coatings for theultraviolet should be evaporated at close to normalincidence and that the coatings should not be thicker

than absolutely necessarycidence are unavoidable

if higher angles of vapor in-

B Reflectance and Transmittance ofSemitransparent Films

Semitransparent aluminum films are especially im-portant for interferometry in the ultraviolet becausethey show in this wavelength region higher reflectanceand lower absorptance than any other metallic coatingTable IV shows the calculated reflectance and trans-mittance of evaporated aluminum for various wave-lengths and various film thicknesses The optical con-stants n and k listed in Table I were used for thecalculations For films produced with deposition rates ofmore than 100 Asec at 1 X 10-5 mm Hg the calculatedR and T values agreed well with the directly measuredones down to film thicknesses of about 100 A The re-sults show that for the same film thickness the trans-mittance increases strongly with decreasing wavelengthA 160-A thick film for example transmits 25 at220 mAu and only 5 at 546 mit

Films deposited at low deposition rates show muchhigher absorptance and lower reflectance and are there-fore unsuitable for interferometry in the ultraviolet Apreviously reported statements that semitransparent Alfilms produced at high deposition rates show a mottledappearance could not be verified All films producedunder the above described conditions were extremelyuniform in appearance

722 Vol 51

• Optical properties of Al at 300nmpdf
• Al_Reflection_Transmissionpdf
• Optical Properties of Aluminumpdf

G HASS AND J E WAYLONIS

TABLE IV Calculated reflectance and transmittance of Al films evaporated under optimum conditions onto transparent substratesof n = 15 for various wavelengths as a function of film thickness (Calculated values agree with directly measured ones for film thick-nesses gt 100 A back surface antireflected)

Film Wavelength (mu)thickness 220 300 400 546 650

(A) R T R T R T R T R T

40 14 82 19 74 25 65 33 51 38 4280 33 60 43 47 52 36 60 24 63 18

120 52 40 62 27 70 19 74 12 75 9160 67 25 74 16 79 11 81 7 82 5200 763 152 815 91 849 59 856 35 854 26240 824 91 860 51 881 33 881 20 875 14280 862 54 884 31 900 19 895 11 888 08320 885 32 900 18 911 11 904 05 896 04360 898 19 909 10 917 06 909 04 900 03400 906 11 914 05 921 04 912 02 903 02500 915 03 920 01 925 lt01 915 lt01 906 lt01

(deposition rate 300 Asec at X 10-5 mm Hg) and theother deposited under poor evaporation conditions (de-position rate 10 Asec at 1X 10-4) In each group thefilms measured were about 600 A and 2000 A thick Theresults show that the reflectance decrease with increas-ing angle of vapor incidence depends upon wavelengthfilm thickness and evaporation conditions The effectis much more pronounced at shorter wavelengths andfor films prepared under poor conditions and increasesin all cases with increasing film thickness An increaseof vapor incidence angle from 00 (normal incidence) to60deg has only a very small effect on the reflectance of600-A-thick coatings but greatly decreases the re-flectance of 2000-A-thick films in the ultraviolet A2000-A-thick film deposited at normal incidence underoptimum conditions reflects almost as well as a justopaque coating even at 220 myi The reflectance at220 mju however drops for a 2000-A thick film from912 to 750 when the vapor incidence angle is in-creased from 0deg to 600 This effect is much greater forfilms produced under poor evaporation conditions Here2000-A thick films evaporated at normal incidencealready show a much lower reflectance than just opaquecoatings and the increase in vapor incidence angle from0deg to 600 drops the reflectance of a 2000-A thick filmfrom 659 to 312 at 220 mu and from 888 to 813at 500 mu This indicates that mirror coatings for theultraviolet should be evaporated at close to normalincidence and that the coatings should not be thicker

than absolutely necessarycidence are unavoidable

if higher angles of vapor in-

B Reflectance and Transmittance ofSemitransparent Films

Semitransparent aluminum films are especially im-portant for interferometry in the ultraviolet becausethey show in this wavelength region higher reflectanceand lower absorptance than any other metallic coatingTable IV shows the calculated reflectance and trans-mittance of evaporated aluminum for various wave-lengths and various film thicknesses The optical con-stants n and k listed in Table I were used for thecalculations For films produced with deposition rates ofmore than 100 Asec at 1 X 10-5 mm Hg the calculatedR and T values agreed well with the directly measuredones down to film thicknesses of about 100 A The re-sults show that for the same film thickness the trans-mittance increases strongly with decreasing wavelengthA 160-A thick film for example transmits 25 at220 mAu and only 5 at 546 mit

Films deposited at low deposition rates show muchhigher absorptance and lower reflectance and are there-fore unsuitable for interferometry in the ultraviolet Apreviously reported statements that semitransparent Alfilms produced at high deposition rates show a mottledappearance could not be verified All films producedunder the above described conditions were extremelyuniform in appearance

722 Vol 51

• Optical properties of Al at 300nmpdf
• Al_Reflection_Transmissionpdf
• Optical Properties of Aluminumpdf

• Optical properties of Al at 300nmpdf
• Al_Reflection_Transmissionpdf
• Optical Properties of Aluminumpdf