• Aucun résultat trouvé

SPATIALLY RESOLVED ABSORPTION AND DETECTION OF MICROSCOPIC IMPURITIES IN OPTICAL THIN FILMS BY PHOTOTHERMAL DETECTION

N/A
N/A
Protected

Academic year: 2021

Partager "SPATIALLY RESOLVED ABSORPTION AND DETECTION OF MICROSCOPIC IMPURITIES IN OPTICAL THIN FILMS BY PHOTOTHERMAL DETECTION"

Copied!
7
0
0

Texte intégral

(1)

HAL Id: jpa-00223240

https://hal.archives-ouvertes.fr/jpa-00223240

Submitted on 1 Jan 1983

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

SPATIALLY RESOLVED ABSORPTION AND DETECTION OF MICROSCOPIC IMPURITIES IN

OPTICAL THIN FILMS BY PHOTOTHERMAL DETECTION

J. Abate, R. Roides

To cite this version:

J. Abate, R. Roides. SPATIALLY RESOLVED ABSORPTION AND DETECTION OF MICRO-

SCOPIC IMPURITIES IN OPTICAL THIN FILMS BY PHOTOTHERMAL DETECTION. Journal

de Physique Colloques, 1983, 44 (C6), pp.C6-497-C6-502. �10.1051/jphyscol:1983681�. �jpa-00223240�

(2)

JOURNAL DE PHYSIQUE

Colloque C6, suppl6ment au nO1O, Tome 44, octobre 1983 page C6- 497

S P A T I A L L Y RESOLVED ABSORPTION AND DETECTION OF MICROSCOPIC I M P U R I T I E S I N OPTICAL T H I N F I L M S BY PHOTOTHERMAL DETECTION

J.A. Abate and R. Roides B. P. O i l , Inc., and

Laboratory for Laser Energetics, University of Rochester, 250 East River Road, Rochester, New York 14623, U . S . A.

R6sum6

-

On d6crit une mgthode de mesure d'absorption r6solue spatiallement qui permet la localisation dfimpuret6s absorbantes dans des films di6lec- triques. Une corr6lation entre les sites absorbants et les donunages est 6ta- blie partiellement.

Abstract

-

Spatially resolved absorption measurements allowing for the lo- cation of micron size absorbing impurities in dielectric thin films using photothermal detection is described. A partial correlation between localized absorption sites and damage is made.

As the laser fusion community proceeds t o frequency convert Nd:glass laser systems t o shorter wavelengths, t h e importance of t h e performance of optical coatings a t these wavelengths becomes of greater interest. Of particular interest t o the LLE program i s t h e performance of these coatings at t h e tripled frequency of Nd:glass, 351 nm. The fluence level a t which the optical coatings can transport the UV beam will have a major impact on t h e size, cost, and energy-on-target on all UV upgrades t o t h e OMEGA laser. We have therefore developed a n experimental system t o better characterize t h e damage process in thin film coatings.

Presently the most successful theoryl,2 describing t h e damage of optical thin films by high power laser radiation attributes t h e damage process t o isolated microscopic impurities, or defects, in the deposited coating. In this theory, t h e impurity absorbs the incident radiation causing its temperature t o rise, eventually producing melting, vaporization, or f r a c t u r e of the material surrounding t h e impurity.

This impurity model can explain many of the observed scaling e f f e c t s seen in thin film damage studies. The observed decrease in damage threshold with decreasing wavelength can be attributed t o t h e increase in Mie absorption with decreasing particle size, and t h e well- known result t h a t Mie scattering i s strongest f o r particles whose size is comparable t o the incident light wavelength. The observation t h a t t h e damage threshold increases with thinner films can be explained by t h e exclusion of larger, easier t o damage, impurities as t h e physical thickness of the film is reduced. The observation of very high damage thresholds for small incident laser spot sizes is consistent with damage being caused by localized impurities: when the spot size is small compared t o t h e mean distance between impurities, i t is possible t o avoid these weaker areas and measure higher thresholds. In addition, i t has been shown for oxide and fluoride films t h a t t h e dependence of damage and thermal properties on laser pulse width is consistent with t h e impurity model.1

Very little is known, however, concerning t h e e x a c t nature or density of these damage- causing impurities. It has been determined from damage morphology3-5 t h a t t h e impurities must be small in size, typically several microns or less in diameter. Figure 1 shows a photomicrograph of t h e typical morphology seen a s a result of laser-induced damage. In this work we have set out t o establish a direct connection between the localized absorption properties of a thin film and t h e appearance of damage in areas of locally high absorption.

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1983681

(3)

JOURNAL DE PHYSIQUE

Fig. 1 Photomicrograph of typical damage t o a T a g 5 - S i 0 2 high reflectance (HRI coating. Note in particular t h e micron-sized pits.

Detection of Impurities

Several attempts have been made t o use photothermal e f f e c t s t o investigate surface and sub-surface structure in solids and in thin films.6-11 Photothermal imaging has developed into a very powerful non-destructive testing technqiue. All previous methods have lacked sufficient spatial resolution t o d e t e c t t h e objects of interest t o laser-damage studies, 'of size 5 1 urn. Some defects have been seen in thin f i 1 m s , 7 - ~ ~ but their size, typically 20 microns or larger, is not consistent with t h e experimental observation of micron-size pits in damage coatings a t threshold (see Fig. 1).

I

Fig. 2 Block diagram of the "Mirage" apparatus for photothermal deflection.

(4)

The apparatus we have used t o l o c a t e micron-size absorbing impurities in dielectric thin films is shown schematically in Fig. 2. I t relies on t h e photothermal defiectionll of a probe laser t o measure t h e absorption in a localized a r e a of energy from t h e pump laser.

The pump beam consists of modulated 0.351 pm light from an argon-ion laser focused t o a 1.0+ 0.2 diameter spot by a 60X microscope objective. The sample of interest is kinematically mounted on a computer-controlled X-Y translation stage, with a minimum step size of 0.4 p. A HeNe laser probe beam has been used in t w o geometries: skimming across the sample (Fig. 3a), and reflecting off the sample surface a t a shallow angle (Fig.

3b1.10

IPI,oIo T h e r m a l D ~ ~ I I e ~ l o o n ) (Pholo-Thermal Delleclion Using

Rellecled Probe Beam)

llcclcd obe

Fig. 3 Use of the "Mirage" e f f e c t for mapping defects in a thin film sample. Pump laser energy is absorbed by a defect with two consequences: (i) t h e surrounding air is heated, leading t o refractive index perturbations; and (ii) thermal expan- sion in the absorption region causes a "bump" t o form on t h e sample surface.

Depending on t h e geometry chosen, the probe laser is either refracted (case (a)) or reflected a s well a s refracted (case (b)).

The probe beam is focused t o a spot size of 75 pm in t h e vicinity of the pump beam focus. The two probe beam geometries have been compared and t h e second (reflective) method has shown a factor of ten greater sensitivity. This most likely c a n be attributed t o a new photothermal effect first discussed by Olmstead et a1.12

In t h e skimming geometry (Fig. 3a), t h e probe beam is deflected by a refractive index gradient, caused by t h e temperature gradient s e t up when t h e pump beam is absorbed locally by an impurity in the sample. This e f f e c t is also present in t h e reflective geometry (Fig.

3b), but here the probe beam is also deflected off t h e "bump" caused by t h e local heating and resulting thermal expansion of t h e substrate. The modulated position of the probe laser is measured by a position sensor (United Detector Technologies SC-25) connected t o a s e t of lock-in amplifiers arranged in quadrature t o measure both t h e amplitude and phase of the position signal. A DEC 11/23 computer controls the position of t h e sample and reads t h e output of the lock-in amplifiers. False color absorption maps of t h e coatings of interest a r e produced on a Chromatics color graphics display. The spatial resolution obtainable using this technique is illustrated in Fig. 4, where scans of absorbing sites of C r spots of various diameters

(5

5 p,

5

1 pm, and

5

0.4 prn) a r e shown. The 0.4 prn sites a r e clearly resolvable.

(5)

C6-500 JOURNAL DE PHYSIQUE

When t h e cross-section of t h e absorbing volume becomes smaller than t h e diameter of the pump beam, only a portion of the incident energy is transformed into heat and t h e resulting thermal and acoustic waves. The ultimate detectivity in this case will depend not only on the absorption coefficient and t h e thickness of t h e inclusion, but also on the ratio of the cross-sections of t h e pump beam and the absorbing region, t h e distance between t h e inclusion and the coating surface interface, and t h e thermal properties of the material in which the inclusion is embedded. It should be clear t h a t the p u m p b e a m spot size must be kept as small a s possible in order t o detect t h e absorbing inclusions in thin films with maximum sensitivity.

Fig. 4 Absorption maps of C r impurity particles of various diameters (d) (a) d 5 5 m

(b) d

5

1

w

(c) d (0.4 prn

Impurities and Damage

In order t o demonstrate a connection between absorptive impurities and damage, we have made absorption maps of several coatings and then exposed them t o high intensity laser light on the LLE UV damage-testing facility.13 Photomicrographs and absorption maps of sample coatings were taken both before and following the damage test. , Typical examples are shown in Figs. 5 and 6. In many cases the absorption map before damage shows small areas of high absorption t h a t a r e not visible in the darkfield photomicrographs (See Fig. 5).

When tile coating shown in Fig. 5 was exposed t o a laser irradiance approximately 20% above its damage threshold, several localized areas on t h e coating a r e seen t o have damaged in the photomicrograph of Fig. 6a.

(6)

Fig. 5 (a) Photomicrograph and (b damage test.

Fig. 6 (a) Photomicrograph and (b) absorption map of Ta205-Si02 HR coating a f t e r damage test.

In addition, t h e absorption map of t h e s a m e a r e a (Fig. 6b) shows clearly increased absorption in t h e region which showed a d e f e c t in Fig. 5b. There a r e also several areas which show damage in Fig. 6b but no strong absorption signature in Fig. 5b. In these cases, i t i s uncertain whether our sensitivity i s insufficient t o d e t e c t t h e damage-causing impurities, o r whether other mechanisms besides linear absorption a r e acting t o initiate s o m e of t h e damage. A t t e m p t s a r e being made t o increase t h e sensitivity of out apparatus t o further explore this question.

Conclusion

We have taken a f i r s t s t e p towards understanding t h e role of absorptive inclusions in t h e damage process of thin films. We have demonstrated t h a t photothermal deflection microscopy c a n d e t e c t submicron absorbing impurities in optical coatings. In t h e future, we plan t o identify t h e e x a c t nature of these damage-causing impurities. I t is hoped t h a t optical coatings of increased damage resistance will be obtained by t h e use of measures t o control these impurities in t h e manufacturing process.

(7)

JOURNAL DE PHYSIQUE

References

WALKER, T.W., GUENTHER, A.H. and NIELSEN, P.E., TEEE J. Quant. Elect.

17,

2053 (1981).

See references 21-48 in (1).

LOWDERMILK, W.H., MILAM, D., RAINER, F., in "Laser Induced Damage in Optical Materials", H.E. Bennett, A.J. Glass and A.H. Guenther, and B.E. Newnam, Eds., Washington, D.C., US, GPO, NBS Special Publ. 568, 391 (1979).

CARNIGLIA, C.K., APFEL, J.H., ALLEN, T.H., TYTTLE, T.A., LOWDERMILK, W.H., MILAM, D., AND RAINER, F., in "Laser Induced Damage in Optical Materials", H.E.

Bennett, A.J. Glass, A.H. Guenther, B.E. Newnam, Eds., Washington, D.C., US, GPO, NBS Special Publ. 568, 377 (1979).

WALKER, T.W., GUENTHER, A.H., AND BIELSEN, P.E., IEEE,

J.

Quant. Elect. l7, 2041 (1981).

WONG, Y.H., THOMAS, R.L., HAWKINS, G.F Appl. Phys. Lett. 32, 538 (1978).

FREESE, R.P. AND TEEGARDEN, K.J., n .: *'Laser 1nducedU~amage in Optical Materials". H.E. Bennett. A.J. Glass. A.H. Guenther. B.E. Newnam. Eds.. Washington. " , DC. US. GPO. NBS ~ o e c i a l Publ. 568: , 313 - - - (19791. .- '

ATLAS,' D.s.,'M.s.

~ h e s i s ,

institute

of Optics, university of Rochester, 1981,

ABATE, J.A., JACOBS, S.D., PISKOROWSKI, W.S., AND POTTENGER, T.P.. CLEO.

1982 Conference, Phonix, Arizona, 1982.

MUNDY, W.C., HUGHES, R.S., AND CARNIGLIA, C.K. in "Laser Induced Damage in Optical Materialstf, H.F. Bennett, A.H. Guenther, D. Milam, B. Newnam, Eds., t o be published 1982.

BOCCARA, A.C., FOURNIER, D., AND BADOZ, J., Appl. Phys. Lett. 36, 130 (1980).

OLMSTEAD, M., AMER, N.M., FOURNIER, D. AND BOCCARA, A.C. Appl. Phys. (to be published).

ABATE, J.A., ROIDES, R., JACOBS, S.D., PISKOROWSKI, W., AND CHIPP, T., in

"Laser Induced Damage in Optical Materials", NBS Special Publication, 1982 (to be published).

Références

Documents relatifs

Periodic current flowing down the thin film track results in periodic expansion due to Joule heating; this expansion can then be detected using an optical heterodyne

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des

The objective of this experimental work was to assess the impact damage in Flax/PLA-PP composites and propose a non-contact, non-destructive testing setup based

They allow to calculate through equation (25) the normalizes stiffness variation showed in the figures (4) - (4) for the case of single and double damage respectively. The

The following acronyms are used: MI (Michelson interferometer), EFPI (extrinsic Fabry – Perot interferometer), DHR (Differiential Helmholtz), QTF (Quartz Tuning Fork), mR

h = z/a is the dimensionless height of the dipoles above the plane of their images, a a characteristic length for the net (e. the nearest neighbour dis- tance

ABSTRACT: The existing bridges become increasingly older, which leads to the fact that bridge inspections, in the future, will be even more important. The assessment of the

Raman scattering by the lattice of ionic crystals containing