Microelectronics Processing - American Chemical Society

---

Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

10 Semiconductor Processing Problems Solved by Wet (Solution) Chemistry Marjorie K. Balazs Balazs Analytical Laboratory, Sunnyvale, CA, 94086

This chapter gives explicit examples of how the techniques of wet (solution) chemistry can be applied to the production of integrated circuits. The quality control for processed thin films, chemicals, and pure water, along with microcontamination analysis, to resolve production problems are discussed. These examples indicate that wet chemical techniques are the only ones available for absolute standardization and measurement of trace metals and their effect on the devices produced by current very-large-scale-integration (VLSI) technology.

THE SEMICONDUCTOR INDUSTRY USES CHEMICALS

a n d c h e m i s t r y to p r o d u c e its products b u t does not e m p l o y large n u m b e r s of chemists or c h e m i c a l engineers. C l e a r l y , the omission of these professionals has h i n d e r e d the advancement a n d r e d u c e d the c o m p e t i t i v e edge of the s e m i c o n d u c t o r i n dustry. T h e lack of c h e m i c a l expertise has l e d to a neglect of advanced c h e m i c a l processes that can establish a f u n d a m e n t a l u n d e r s t a n d i n g of i n t e grated-circuit (IC) p r o d u c t i o n . T h u s some of the most accurate a n d sensitive methods for the m e a s u r e m e n t a n d identification of organic a n d inorganic materials for process quality c o n t r o l , analysis of m i c r o c o n t a m i n a t i o n , i n c o m i n g - m a t e r i a l evaluation, or p r o d u c t i o n p r o b l e m evaluation have b e e n i g nored. A l t h o u g h e l e c t r i c a l engineers a n d physicists have resorted to creative, i n t e r e s t i n g , a n d often v e r y precise e l e c t r i c a l measurements to investigate p r o b l e m s or establish c o n t r o l parameters for I C p r o d u c t i o n , these methods do not y i e l d direct c h e m i c a l information c o n c e r n i n g materials or processes.

0065-2393/89/0221-0505$06.00/0 © 1989 American Chemical Society

In Microelectronics Processing; Hess, D., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1989.

506

MICROELECTRONICS PROCESSING: C H E M I C A L ENGINEERING ASPECTS

Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

E l e c t r o n b e a m techniques have a i d e d electrical measurements greatly, b u t these methods often lack sensitivity (X-ray a n d A u g e r spectroscopy a n d E S C A [electron spectroscopy for c h e m i c a l analysis]) a n d accuracy ( S I M S [secondary-ion mass spectrometry], etc.), two attributes that are of p r i m e i m p o r t a n c e i n I C process technology. F o r t u n a t e l y , materials can b e a n a l y z e d w i t h b o t h accuracy a n d sensitivity b y w e t c h e m i c a l analysis. C u r r e n t l y , A m e r i c a n s e m i c o n d u c t o r manufacturers are l o s i n g the battle w i t h Japan a n d K o r e a i n the p r o d u c t i o n a n d sales o f r e l i a b l e , l o w - p r i c e d I C devices or c h i p s . T h e failures a n d l o w yields that have h i n d e r e d U . S . efforts are partially caused b y the inadequate u n d e r s t a n d i n g of the c h e m i c a l p r o c esses i n s e m i c o n d u c t o r p r o d u c t i o n a n d the lack of good q u a l i t y controls for chemicals a n d processes. F o r the U n i t e d States to w i n this w o r l d c o m p e t i t i o n , w e m u s t have a better u n d e r s t a n d i n g of the materials u s e d i n m a k i n g I C s , that is, t h e i r exact composition a n d trace c o n t a m i n a t i o n analysis b e l o w p a r t - p e r - b i l l i o n levels, a n d w e must b e m o r e critical about the accuracy of the measurements w e make.

Wet Chemical Analysis in Semiconductor Manufacturing W e t c h e m i c a l analysis is especially useful to s e m i c o n d u c t o r I C manufacturers i n the f o l l o w i n g five areas: 1.

m e a s u r e m e n t of dopant concentrations i n d i e l e c t r i c t h i n films,

2. evaluation o f a l u m i n u m metallization a n d other t h i n films s u c h as silicides a n d t i t a n i u m - t u n g s t e n ( T i W ) , 3. d e t e r m i n a t i o n of metals i n u l t r a p u r e water, 4. q u a l i t y c o n t r o l o f chemicals a n d t h e i r evaluation d u r i n g a n d after use, a n d 5. use of w e t c h e m i s t r y i n c o n j u n c t i o n w i t h other i n s t r u m e n t a l methods for the r e s o l u t i o n of m i c r o c o n t a m i n a t i o n p r o b l e m s .

Analysis ofPSG and PBSG Wet Chemical Methods. W e t c h e m i c a l methods may be u s e d r o u t i n e l y to d e t e r m i n e the phosphorus content (total, P 0 , P 0 , a n d P H ) of phosphosilicate glass ( P S G ) a n d phosphoborosilicate glass ( P B S G ) d i e l e c t r i c t h i n films, the total b o r o n content of P B S G , a n d the s i l i c o n a n d c o p p e r contents of a l u m i n u m films. M e t h o d s for the d e t e r m i n a t i o n of other elements critical to s e m i c o n d u c t o r m a n u f a c t u r i n g are still b e i n g d e v e l o p e d . 2

3

2

5

3

W e t c h e m i c a l c o l o r i m e t r i c methods are the choice for the accurate d e t e r m i n a t i o n of the c o m p o s i t i o n of d o p e d t h i n films. Since early 1960, colori m e t r y has b e e n u s e d to measure the percentage of phosphorus i n P S G ( I ,

In Microelectronics Processing; Hess, D., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1989.

10.

BALAZS

Semiconductor

Processing

507

Problems

2) a n d the percentages of phosphorus a n d b o r o n i n P B S G . C o l o r i m e t r y is the o n l y m e t h o d that gives accurate a n d absolute results; it is the p r i m a r y m e t h o d of d e t e r m i n i n g the q u a n t i t y of dopant i n t h i n films, a n d conseq u e n t l y , it has b e e n u s e d to standardize p h o s p h o r u s - d o p i n g processes. B e cause w e t c h e m i c a l methods have no matrix effects, the accuracy w i t h these methods far exceeds that b y X - r a y , E S C A , A u g e r , or F o u r i e r t r a n s f o r m Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

infrared ( F T I R ) spectroscopy, w h i c h are methods that show m a t r i x effects. W e t c h e m i c a l methods i n v o l v e sophisticated sample p r e p a r a t i o n a n d standardization w i t h N a t i o n a l B u r e a u of Standards reference materials b u t are not difficult for the analytical c h e m i s t n o r necessarily t i m e c o n s u m i n g ( F i g u r e 1). T h e t i m e from sample preparation to final results for various analytical methods, such as G F A A (graphite furnace atomic absorption), I C P (inductively c o u p l e d plasma spectroscopy), I C P - M S ( I C P - m a s s s p e c t r o m etry), a n d c o l o r i m e t r y , ranges f r o m 0.5 to 5.0 h , d e p e n d i n g o n the t e c h n i q u e used. C o l o r i m e t r y is the m e t h o d of choice because of its extreme accuracy. T y p i c a l results of the c o l o r i m e t r i c analysis of d o p e d oxides are s h o w n i n Tables I a n d I I , w h i c h show the accuracy a n d p r e c i s i o n of the measurements. Because w e t c h e m i s t r y gives absolute results a n d is m o r e sensitive c o m p a r e d w i t h e l e c t r o n b e a m methods, it is useful i n m e a s u r i n g the u n i f o r m i t y of a dopant across a wafer. E S C A , X - r a y spectroscopy, a n d A u g e r spectrosc o p y can measure u n i f o r m i t y m o r e q u i c k l y , b u t these methods give r e l a t i v e , not absolute, measurements. F u r t h e r m o r e , calculations for w e t c h e m i c a l analyses are d o n e i n three significant figures ( F i g u r e 2). H o w e v e r , because the accuracy of dopant measurements i n t h i n films is generally ± 3 % of the actual v a l u e , data are r e p o r t e d m o r e often i n two significant figures. N e v e r theless, the r e l i a b i l i t y of the results from w e t analysis g i v e n i n two significant figures is m u c h greater than that of results from electron b e a m e q u i p m e n t . Instrumental Methods. E n g i n e e r s i n the I C i n d u s t r y prefer to use X - r a y or F T I R spectroscopy to d e t e r m i n e the quantities of phosphorus i n t h i n films because of the speed of these methods. These spectroscopic m e t h ods are satisfactory for a relative i n d i c a t i o n of the dopant l e v e l i n t h i n films or additives to m e t a l l i z a t i o n layers, b u t they do have serious drawbacks. X ray spectroscopy is seriously affected b y matrix effects a n d can easily be off b y ± 1 5 - 2 0 % of the actual concentration of dopant i n t h i n films i f the e q u i p m e n t is not p r o p e r l y c a l i b r a t e d against a m a t e r i a l that has b e e n a n a l y z e d b y w e t techniques. X - r a y spectroscopy is f u r t h e r affected b y the film thickness a n d the dopant profile throughout the film. F T I R spectroscopy is a m o r e accurate m e t h o d for a q u i c k analysis of the phosphorus i n t h i n films. H o w e v e r , the P 0 peak i n F T I R is seriously affected i f the t h i n film becomes h y d r a t e d . S o m e loss of sensitivity is caused b y the m o v e m e n t of part of the peak from the P 0 region to the P 0 · H 0 r e g i o n u p o n h y d r a t i o n of the t h i n film ( F i g u r e 3). T h i s shift affects the m e a s u r e d phosphorus content of a t h i n film. A l s o , the lack of a significant 2

5

2

5

2

In Microelectronics Processing; Hess, D., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1989.

2

5

In Microelectronics Processing; Hess, D., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1989.

Weigh

wafer section with f i l m >

f

Strip film

layer wafer section

Reweigh

Wt.

Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

of

film

10.

BALAZS

Semiconductor Processing Problems

509

Table I. Phosphorus Analysis of Sections of P B S G Films

Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

Wafer Oxide Thickness (A) 6100 5500 6300 4900 3100 5400 7000 5300 4700 5900 4800 3800 3400

Wt. % Ρ 1

2

1.78 1.89 1.91 1.98 2.48 3.63 3.69 4.39 4.41 5.00 5.20 5.94 5.98

1.79 1.98 1.96 2.04 2.43 3.60 3.74 4.34 4.45 5.02 5.16 5.74 6.02

3

1.96

4.36 4.60

Maximum Variation 0.01 0.09 0.05 0.08 0.05 0.03 0.05 0.05 0.19 0.02 0.04 0.20 0.04

Table II. Boron Analysis of Sections of P B S G Films Wafer Oxide Thickness (Â) 6000 3600 3300 5000 7900 5100 7300 4400 5400 4000 5400

Wt. % Β 1

2

Maximum Variation

1.35 1.51 1.86 2.56 3.54 3.68 3.77 4.99 5.27 5.36 5.80

1.64 1.46 1.83 2.64 3.61 3.84 3.61 5.11 5.29 5.19 5.78

0.29 0.05 0.03 0.08 0.16 0.20 0.16 0.12 0.02 0.17 0.02

b o r o n peak makes F T I R spectroscopy less reliable for the m e a s u r e m e n t o f b o r o n i n P B S G films. T h e d e t e r m i n a t i o n o f specific phosphorus c o m p o u n d s i n t h i n films is important. O n l y t h r o u g h wet c h e m i c a l analysis was it possible to first discover the presence a n d t h e n to accurately measure t h e quantities of P 0 , P 0 , a n d p h o s p h i n e f o u n d i n p l a s m a , p l a s m a - e n h a n c e d , L P O - L T O (low-pressure o x i d e - l o w - t e m p e r a t u r e oxide), a n d C V D (chemical vapor deposition) p r o c esses (3). M e t h o d s such as X - r a y o r F T I R spectroscopy w o u l d have seen a l l phosphorus atoms a n d w o u l d have characterized t h e m as totally useful p h o s p h o r u s . I n p l a s m a a n d plasma-enhanced C V D films, p h o s p h i n e is totally useless i n d o p i n g processes. 2

5

2

3

A s i m i l a r p r o b l e m exists w i t h b o r o n - d o p e d oxides o r P B S G s . I n this case, b o r o n is f r e q u e n t l y f o u n d as v e r y small s u b m i c r o m e t e r crystals sitting

In Microelectronics Processing; Hess, D., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1989.

In Microelectronics Processing; Hess, D., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1989.

Figure 2. Wafer mapping showing good and poor wafers.

Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

10.

BALAZS

Semiconductor Processing Problems

0.0 ι 1600.0

1

1500.0

1

1000.0

511

r 400.0

Wavenumber Figure 3. Calibration

standard for 3.0 wt % P.

on top of the P B S G . T h e s e crystals make the film look somewhat hazy. B y u s i n g wet c h e m i c a l techniques, this material can be r e m o v e d f r o m the top of the film a n d quantified. A second piece of a wafer can be analyzed for total b o r o n . B y subtracting surface b o r o n from total b o r o n , one can d e t e r m i n e the useful amount of b o r o n i n the t h i n film that w i l l assist i n l o w e r i n g the m e l t i n g point of the oxide d u r i n g reflow. Oftentimes the quantity of b o r o n left o n the surface sufficiently reduces the quantity of b o r o n i n the t h i n film to affect the t e m p e r a t u r e at w h i c h reflow w i l l take place.

Analysis of Other Thin Films Silicon in Aluminum Films. Table III illustrates the t y p i c a l results of measurements of silicon i n an a l u m i n u m film c o m p a r e d w i t h that i n various targets used to make those t h i n films. Calculations can be c a r r i e d out easily i n three significant figures. N o other m e t h o d of m e a s u r i n g silicon i n a l u ­ m i n u m has the sensitivity or the a b i l i t y to c o m e w i t h i n less than 2 0 % of the actual value. Suicides and Titanium-Tungsten. T h e d e t e r m i n a t i o n of the ratio of metals to s i l i c o n i n m e t a l silicides or of t i t a n i u m to tungsten i n t i t a -

In Microelectronics Processing; Hess, D., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1989.

512

MICROELECTRONICS PROCESSING: C H E M I C A L ENGINEERING ASPECTS

Table III. Typical Results of Analysis of % Si in Aluminum Sample

% Si

Al Film 1

0.98

2

1.01

Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

Target 1

0.97

2

1.01

3

1.01

n i u m - t u n g s t e n materials is e v e n m o r e difficult. A l t h o u g h w e t c h e m i c a l t e c h ­ niques have b e e n u s e d b y process engineers for the analysis o f these t h i n films for o v e r a decade, these analyses w e r e not r o u t i n e procedures i n s e m ­ iconductor-processing plants. R o u t i n e w e t c h e m i c a l procedures for the a n a l ­ ysis of silicides have b e e n d e v e l o p e d a n d are n o w u s e d extensively. T h e results o f t w o studies o f silicide films are p r e s e n t e d i n Tables I V and V. F o r the results shown i n Table I V , m o l y b d e n u m silicides w e r e d e ­ p o s i t e d o n vitreous carbon disks. O n e set was s t r i p p e d w i t h base (2 Ν N a O H ) , a n d the other set was s t r i p p e d w i t h a c i d ( H F - H N 0 - H 0 , 1:1:10). T h e study o f c h r o m i u m silicides (Table V) gave s i m i l a r results. 3

2

T o verify the accuracy of an analytical c h e m i c a l p r o c e d u r e , a m a t e r i a l balance or verification b y other techniques m u s t be o b t a i n e d . F o r the analysis of m o l y b d e n u m a n d c h r o m i u m silicides, a m a t e r i a l balance a n d verification of results b y three techniques w e r e the goals. B o t h acidic a n d basic s t r i p p i n g procedures w e r e u s e d i n sample p r e p a r a t i o n . T w o different c o l o r i m e t r i c procedures a n d atomic absorption techniques w e r e u s e d for quantitative determinations. F o r b o t h m o l y b d e n u m a n d c h r o m i u m silicides, the different analytical procedures gave comparable results b u t no m a t e r i a l balance (ac­ countability for 100% of the materials i n a sample). T h e s e results i n d i c a t e d that o u r data w e r e accurate b u t that other elements w e r e present that w e r e not b e i n g m e a s u r e d . H y d r o g e n , oxygen, or n i t r o g e n was suspected to be the m i s s i n g e l e m e n t . A u g e r analysis r e v e a l e d the presence of a p p r o x i m a t e l y 1 5 % oxygen i n samples of m o l y b d e n u m s i l i c i d e . Because A u g e r analysis cannot give accurate data nor measure h y d r o g e n , an absolute m a t e r i a l b a l ­ ance was not o b t a i n e d . M e a s u r e m e n t s of these elements w o u l d r e q u i r e a quantitative gas c h r o m a t o g r a p h i c - m a s s spectrometric study a n d the d e s i g n of special testing chambers.

Analysis of Water W a t e r is one of the most w i d e l y u s e d c o m m o d i t i e s i n the p r o d u c t i o n of integrated c i r c u i t s , a n d the q u a l i t y of water is e x t r e m e l y i m p o r t a n t i n this

In Microelectronics Processing; Hess, D., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1989.

10.

BALAZS

Semiconductor

513

Processing Problems

Table IV. Analysis of Molybdenum Suicides Sample

Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

Al A2 A3 Bl B2

Etchant

% Mo

% Si

Total % Mo + % Si

base acid acid acid acid

51.8 50.3 51.5 60,5 58.8

35.9 35.2 36.3 28.8 25.7

87.7 85.5 87.8 89.3 84.5

Empirical

Formula

M01.oSi2.4O17 M01.oSi2.401.4 M01.oSi1.501.2 MoioSiisOie

Table V. Analysis of Chromium Suicides Sample

% Cr

% Si

Av % Si

Total % Cr + % Si

Al A2 Bl B2 CI C2 Dl D2 Target

31.7

50.2 52.9 53.7 53.9 55.2 56.0 45.7 46.3 43.9

51.6

83.3

53.8

78.1

55.6

81.0

46.0

65.1

24.3 25.4 19.1 50.3

94.2

i n d u s t r y . N u m e r o u s papers have b e e n w r i t t e n about this subject. T h e q u a l i t y of water d u r i n g t h e past decade has i m p r o v e d significantly. T a b l e V I shows the latest specifications for p u r e water. C l e a r l y , t h e attainable levels of p u r i t y have i m p r o v e d t r e m e n d o u s l y . U n t i l 1985, t h e levels of m e t a l l i c constituents w e r e not m e a s u r e d e v e n i n h i g h - q u a l i t y water b e l o w t h e p a r t - p e r - b i l l i o n range. W i t h t h e advent of more-sensitive tools for w e t c h e m i c a l analysis, such as I C P - M S (inductively c o u p l e d plasma spectroscopy-mass spectroscopy), s u b - p a r t - p e r - t r i l l i o n levels o f m e t a l l i c constituents i n water c o u l d b e analyzed. I n 1985, t h e results o f a study d o n e at Balazs A n a l y t i c a l L a b o r a t o r y ( B a l L a b ) (4) a n d i n Plessey Research (5) l e d to t h e specification of m e t a l l i c materials i n water. A t that t i m e , t h e metals that w e r e analyzed w e r e those thought most l i k e l y to b e present i n a n d deleterious to I C s . Because b o t h i o n chromatography a n d G F A A are v e r y t i m e c o n s u m i n g , t h e evaluation o f o n l y n i n e or t e n metals a n d six to eight n o n m e t a l l i c species at the p a r t - p e r t r i l l i o n l e v e l is a l l that one w o u l d d o n o r m a l l y to get results w i t h i n a r e a sonable t i m e . W i t h o u t sample concentration, analysis at this l e v e l w i l l be at the limits of these methods. T o better detect m e t a l l i c residuals i n h i g h - p u r i t y water, a n e w m e t h o d was a p p l i e d that allows the d e t e r m i n a t i o n o f trace elements i n p u r e water i n t h e p a r t - p e r - t r i l l i o n l e v e l . T a b l e V I I gives the results of a study r e c e n t l y c o m p l e t e d at B a l - L a b . P u r e water samples f r o m seven sites, B a l - L a b (site A) a n d six I C producers (sites B - G ) , w e r e obtained o n the same day a n d

In Microelectronics Processing; Hess, D., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1989.

514

MICROELECTRONICS PROCESSING: C H E M I C A L ENGINEERING ASPECTS

analyzed s e m i q u a n t i t a t i v e l y b y I C P - M S to d e t e r m i n e t h e elements (from l i t h i u m to u r a n i u m ) present. Standards w e r e p r e p a r e d , a n d t h e samples w e r e t h e n a n a l y z e d q u a n t i t a t i v e l y to d e t e r m i n e t h e exact amount o f each m e t a l i n t h e seven samples. M o r e than 30 trace elements w e r e d e t e c t e d b y I C P - M S . O f these trace e l e m e n t s , 11 w e r e p r o m i n e n t a n d 20 w e r e q u i t e measurable. T h e e n t i r e analysis, i n c l u d i n g the preparation o f standards, took Downloaded by UNIV OF CALIFORNIA SANTA BARBARA on September 25, 2015 | http://pubs.acs.org Publication Date: May 5, 1989 | doi: 10.1021/ba-1989-0221.ch010

less t h a n 4 h . Table VI. Specifications and Guidelines for Semiconductor-Pure Water r

SPECIFICATIONS ~\ /

GUIDELINES

c ο

item

Φ

8

Q.-Q

Φ

CO