- Email: [email protected]
Integrated optical devices fabricated by MBE
Ping. Czym/Growth ~ . Vol.2, pp. 86-113. ¢, Peqlan~ Prm Ltd, 1g'/9.PrinzlKIin Gnmt Bdtain.
0146-3E~179/0701-0086t06.00/0
INTEGRATED OPTICAL DEVICES FABRICATED BY MBE R. D. Burnham and D. R. Scifres XEROX PARC, Palo Alto, CA 94304, U.S.A.
(Submitted March 1979]
I.
INTRODUCTION
Approximately IO y e a r s ago, S. E. M i l l e r ( I ) a t B e l l Telephone L a b o r a t o r i e s o u t l i n e d a p r o p o s a l f o r t a k i n g v a r i o u s d i s c r e t e and b u l k y o p t l c a l components, which f o r many y e a r s had been spread out over a l a r g e o p t i c a l bench, and m i n i a t u r i z i n g them on a common r i g i d s u b s t r a t e . The i n c o r p o r a t i o n o f such d i s c r e t e d e v i c e s i n t o an o p t i c a l c i r c u i t would e l i m l n a t e m e c h a n i c a l and thermal s t a b i l i t y problems, a l o n g w i t h the need f o r c o s t l y and heavy o p t i c s and e x p e n s i v e b u l k y o p t i c a l t a b l e s . The s i z e of t h e s e m i n i a t u r i z e d components a t l e a s t i n one or two d i m e n s i o n s needed to be o n l y as l a r g e as the o p t i c a l c a r r i e r wave (O.4-10 ~m) and thus were amenable to mass p r o d u c t i o n u s i n g s t a n d a r d p h o t o l i t h o g r a p h i c t e c h n i q u e s . I t was proposed a t t h a t time t h a t the m i n i a t u r i z a t i o n of ~uch components onto a common s u b s t r a t e , a c h i p , be called an integrated optical circuit in view of its electrical counterpart the integrated circuit. Today, several components of such an integrated optical circuit have indeed been i n t e g r a t e d onto a common s u b s t r a t e and have been d e m o n s t r a t e d to a c h i e v e improved s t a b i l i t y , s m a l l s i z e , l i g h t w e i g h t , and low power r e q u i r e m e n t s . I n f a c t , some c h i p s c o n s i s t i n g of a r r a y s of l a s e r s now appear to be a r e a l i t y f o r o p t i c a l communications a p p l i c a t i o n s and no doubt o t h e r a p p l i c a t l o n s w i l l f o l l o w . F i b e r o p t i c c c ¢ ~ m n i c a t i o n s has p r o v i d e d the p r i m a r y s t i m u l u s f o r i n t e g r a t e d o p t i c s research. I n t e g r a t e d o p t i c s coupled w i t h f i b e r o p t i c s i s expected to f i n d a p p l i c a t i o n i n communications where the d a t a r a t e s exceed 300 Mb/sec and where s p a t l a l m u l t i p l e x i n g i s e x t e n s i v e l y u s e d . This a p p l i c a t i o n may r e q u i r e s i n g l e mode, CW l a s e r s o u r c e s modulated a t r a t e s up to 10 GHz, low l o s s s i n g l e mode f i b e r s and GHz response photodectectors. M o n o l i t h i c and h y b r i d a r e two approaches i n which one could make i n t e g r a t e d o p t i c s . The h y b r i d approach uses s e v e r a l m a t e r i a l s which a r e chosen to o p t i m i z e each of the components. This o b v i o u s l y can p r e s e n t s e r i o u s problems s i n c e g e n e r a l l y p r o c e s s e s used to f a b r i c a t e the d i f f e r e n t d e v i c e s might be i n c o m p a t i b l e . The m o n o l i t h i c approach uses one m a t e r i a l which works w e l l f o r a l l the components. S i n c e the b e s t l a s e r , m o d u l a t o r , waveguide and d e t e c t o r a r e n o t n e c e s s a r i l y found i n one m a t e r l a l , compromises w i l l have to be made. F o r t u n a t e l y , however, some s e m i c o n d u c t o r s appear to be s u i t a b l e f o r m o n o l i t h i c o p t i c a l i n t e g r a t i o n . In fact, b e c a u s e of the d u a l r o l e t h a t d i r e c t bandgap s e m i c o n d u c t o r s p l a y i n b o t h g e n e r a t i n g and d e t e c t i n g o p t l c a l r a d i a t i o n , such m a t e r i a l s a r e v i r t u a l l y the o n l y ones s u i t a b l e for monolithic integration.
95
96
R. D. Burnham and D. R. Scifres
2.
GaAs AND RELATED ALLOYS
One of the most promising and most worked on semiconductor materials is GaAs and its alloys with columns III-V. GaAs technology has progressed to the point that high quality fairly large area substrates (4-5 cm in diameter) can be produced. Substrates of other III-V compounds of similar size are either not available or the cost is prohibitive. The strong concentration on the GaAs family is due to its versatility in terms of its electrlcal and optical properties and to the fact that it is the only materials system in which all the important optical functions -- light generation, guiding, modulation, and detection -- have been achieved. This makes it possible, in principle, to make monolithic integrated optical circuits out of the GaAs family with suitable addition of impurities, just as integrated circuits are made of silicon. An attractive feature of GaAs for integrated optics is that its alloys with other compounds of column III-V elements have refractive indexes and bandgaps that vary with the amount of the other compound added. For example, the refractive index of Gal_xAlxAs at a wavelength of 0.9 ~m decreases almost linearly with x from 3.59 for x = 0 to 2.97 for x = I (see Fig. i). It is clear then that waveguides can be made by epitaxial growth of layers of different x on top of each other. For such structures to have good device performance, it is important that the lattice constants of adjoining layers be well matched. If that is not the case, metallurgical imperfections, such as dislocations, will occur at the interfaces, giving rise to such deleterious features as optical scattering centers and electron-hole recombination centers. Gal-~AlxAs is particularly attractive because the lattice constant for GaAs is 5.653 A and for AlAs, 5.661 A; thus, the largest possible difference between lattice constants in this family is only 0.14% 3
.
7
~
2
.
2
300K 3.6[--
J
42.1
%% 3.5
\
2.0
\ \ \
x
1.9
%%
3.4
x
al
\ 3.3
% %
X z LiJ 3.2
1.8 ~ %
%. %
> t(J
%
%. %. %
M. 3.1
i¢1
1.7 ~ >
%
%. %. ,% %. %.
3.0
. . . .
2.9
REFRACTIVE INDEX
w
1.6 ~ %. %. . 1,5 %. %. 1.4
ENERGY GAP
[}.0
0.5
1.3 1.0
MOLE FRACTION AlAs, x
Fig. i.
Variation of refractive index and bandgap of Gal_ x AI As with x. x
Integrated Optical Devices Fabricated by MBE
97
The interconnection of optical components via waveguides is the central problem of monolithic integrated optics for two main reasons: first, the minute cross sections of the waveguides (10"8-10 -7 cm 2) requite strict dimensional tolerances, and second, optical circuits consisting of active and passive waveguide components need adjacent regions with different material properties such as composition, bandgap, index of refraction, doping, etc. These requirements can only be satisfied by taking the properties and limitations of the crystal growth techniques into account.
3.
MOLECULAR BEAM EPITAXY
There are three growth techniques which could be used for integrated optical circuits based on single crystal semiconductors, they are: liquid phase epitaxy (LPE), chemical vapor phase ep[taxy (VPE), and molecular beam epitaxy (MBE). Even though more work has been done with optical integration by LPE, MBE appears to be the mosL promising growth technique because of its overall versatility. VPE has recently received considerable recognition because of advances made by Dupuls and Dapkus (2) using metalorganics but it still lacks the overall flexibility required to make extremely complicated monolithic integrated optical circuits compared to MBE. The initial pioneering work on MBE was done by A. Y. Cho at about the same time that S. E. Miller proposed integrated optics. Today one of the major research goals is to merge these two technologies so as to produce a variety of integrated optical circuit elements. The advantages of growing GaAs/Gal_xAlxAs integrated optical circuits by MBE are many: (a) Since the growth is performed in ultra high vacuum with the availability of in situ diagnostic instruments, one can be assured that the desired surface conditions have been reached before comencement of the growth; (b) the MBE is controlled by kinetics rather than by a diffusion process and thus allows one to grow layers with practlcally any predetermined doping profile (3) and variation of composition (4,5); (c) large uniform layers (3 cm x 3 cm) with featureless surfaces may be grown by MBE; (d) the low epitaxial temperature (5800C) may prevent dopant diffusion both within the epitaxial layer and from the substrate into the epitaxial layer; and (e) the slow growth rate (I ~m/hr) allows for essentially monolayer <5 A thickness control (6-9). The major problem with MBE for integrated optics applications has been centered on the growth of high quality Gal_xAlxAs. This review article will therefore deal mainly with the various MBE techniques that have been used towards the goal of making monolithic integration in the GaAs/ Gal_xAlxAs system. We refer the reader to other review articles (I0,ii) and books (12,13) on a broader coverage of integrated optics which discuss other growth techniques and considerably more theory than will be presented here.
4.
WAVEGUIDES
First, consider the optical wavegulde, which is perhaps the most basic of all integrated optics components. Most of these guides are planar, i.e~, those with rectangular geometry. The guiding property can be understood by studying the heterostructure illustrated in Fig. 2. As shown, the waveguide consists of a sandwich of three different materlal compositions, each of which is transparent to the propagating light-wave. A necessary condition for waveguiding in these heterostructure layers is n2 > n I, n3, where n is the refractive index. The waveguiding property can be thought to result from the phenomenon of total internal reflection Consider a ray of light starting in material 2 and propagating toward the 2-3 interface. If the angle between the normal to the interface and the direction of lightwave propagation 823 is larger than the critical angle for total internal reflection
rR. D. Burnham and D. R. Scifres
HETE RO&TRUCTURE WAVEGUIDES
X m
13 HIGH INDEX
:l
LOW,, ,EX
MATERIAL ' LOW INDEX
I
LOW,NDEX
MATERIAL 2 MATERIAL 3 HIGH INDEX LOW INDEX
"1
n2
., "
"
n3
~,TE 0
~~,TE 1 /
~
TE2
Schematic diagram of a heterostructure way, guide.
Fig. 2.
[ ec = sin -I (n3/n2)], the light will be totally reflected upon striking this interface. The same holds true for a light ray propagating toward the 1-2 interface. The light rays then follow a ziH-zag path making the same angle e23 with each reflection. However, because of interference between rays that have traveled different paths, only those rays making certain angles will survive. These are called the modes of the guides. In terms of wave optics, each mode has a different electric field distribution across the guide and propagation constant along the guide. The propagation constant (8) can be determined from wave theory or from the ray-interference pattern. In either case, the wavelength of light in the guide (~g) is
g and the wave is said to see an equivalent refractive index neq given by neq =
Xo/Xg
where A is the free-space light wavelength. Since the guided wave exists in all three m~dia with nl, n2, and n3, the equivalent index is a sort of average satisfying n I , n 3
<
n
eq
<
n 2
•
I n t e g r a t e d O p t i c a l Devices F a b r i c a t e d by MBE
99
The e l e c t r i c - f i e l d d i s t r i b u t i o n s (12) of t h r e e modes, TEO, TEl, and TE2 a r e i l l u s t r a t e d i n F i g . 2; each has a d i f f e r e n t n e q . Guides c o n s i s t i n g of a Gal-xAlxAs l a y e r sandwiched between two l a y e r s of Gal_yAlyAs of lower i n d e x , thus y > x a r e examples of a o n e - d i m e n s i o n a l g u i d e . S i n c e - t h e r e a r e two c o n t a c t s between l a y e r s o f d i s s i m i l a r c o m p o s i t i o n , t h e s e a r e d o u b l e h e t e r o s t r u c t u r e (DH) g u i d e s . Such l a y e r e d l i g h t g u i d i n g s t r u c t u r e s can be r e a d i l y grown by any of the t h r e e p r e v i o u s l y m e n t i o n e d c r y s t a l growth t e c h n i q u e s . Merz and Cho (14) compared o p t i c a l l o s s e s f o r 1-2 pm t h i c k DH waveguides of s i m i l a r c o m p o s i t i o n grown by MBE and LPE. They r e p o r t e d l o s s e s l e s s than 1.5 cm- I f o r MBE grown g u i d e s between 1.1 and 1.4 eV, t h e e n e r g y of t h e GaAs DH l a s e r . These 1oases a r e comparable t o , or l e s s t h a n , the l o s s e s observed i n s i m i l a r wave-guide s t r u c t u r e s grown by the LPE. They did p o i n t o u t , however, t h a t d e s p i t e the h i g h degree of l a y e r u n i f o r m i t y and c o n t r o l t h a t MBE can a c h i e v e , t h e s e low l o s s e s a r e n o t u p r i o r i o b v i o u s . The r e a s o n i s t h a t the Gal-xAlxAS s u r f a c e exposed d u r i n g growth i n an MBE system i s h i g h l y r e a c t i v e which makes the l a y e r v e r y s u s c e p t l b l e to c o n t a m i n a t i o n by t r a c e amounts of water vapor and c a r b o n - c o n t a i n i n g g a s e s ; t h e s e or o t h e r c o n t a m i n a n t s might c o n t r i b u t e s i g n i f i c a n t l y to the below gap a b s o r p t i o n . These r e s u l Z s show, however, t h a t i s i t p o s s i b l e to grow l o w - l o s s Gal_xAiAs wave-guides by MBE w i t h t h e p r e s e n t - d a y vacuum t e c h n o l o g y . A f t e r Gossard e t uS. (9) r e p o r t e d the u l t i m a t e i n l a y e r c o n t r o l i n which up to 104 a l t e r n a t i n g s i n g l e or m u l t i p l e monolayers of GaAs and AlAs were s e q u e n t i a l l y d e p o s i t e d by MBE, Merz e t GZ. (15) (see F i g . 3) r e p o r t e d on l o w - l o s s wsveguldes whose c e n t r a l r e g i o n c o n s i s t e d of a l t e r n a t i n g sequences of a p p r o x i m a t e l y n i n e monolayers of GaAs and one monolayer of AlAs. The o u t e r l a y e r c o n s i s t e d of an a l t e r n a t i n 8 sequence of t h r e e monolayers of AIAs and seven monolayers of GaAs. This gave the c e n t r a l r e g i o n a lower a v e r a g e A1 c o m p o s i t i o n (x ~ 0.10) than the o u t e r sequence o f l a y e r s (x - 0 . 3 0 ) . O p t i c a l l o s s e s were ~1 cm-1 h i g h e r than the b e s t ~a0.9Al0.1As random a l l o y waveguides f a b r i c a t e d to date by MBE (14). Cho, Y a r i v , and Yeh (16) r e p o r t e d a n o v e l wavegulde grown by MBE i n which a Ga0.62 A10.38As g u i d i n g l a y e r was bounded on one s i d e by a i r and on the o t h e r s i d e by a periodic layered medium composed of alternating layers of GaAa and Ga0.8Al0.2As. The periodic media present high refelectivity to radiation which satisfies the Bragg condition. Unlike ordinary dielectric waveguldes, confined guiding with arbitrarily low loss is possible even when the guiding layer possesses an index of refraction which is lower than that of the periodic layers. V a r i o u s t y p e s of t h r e e - d i m e n s i o n a l h e t e r o s t r u c t u r e waveguides have been made by MBE. One attractive approach for making three-dimensional waveguldes involves embedding a GaAs waveguide i n a m a t r i x of Gal_xAlxAs. This i s a c h i e v e d by s e l e c t i v e l y e t c h i n g away p o r t i o n s of a p r e v i o u s l y grown p l a n a r wavegulde r e s u l t i n g i n a t h r e e - d l m e n s l o n a l g u i d e and t h e n e p i t a x i a l l y c o v e r i n g the n o n p l a n a r s u r f a c e w i t h Gal_xAlxAs by MBE to c o m p l e t e l y e n c a p s u l a t e the GaAs waveguide ( 1 7 ) . T h i s t e c h n i q u e d e m o n s t r a t e s one of the growth advantages MBE has over LPE. It is difficult to grow on GaAIAs by LPE once the sample is exposed to the ambient because of the reactive nature of AI. But with MBE the exposed GaAIAs surface need only be cleaned by ion milling prior to any regrowth. Another t e c h n q l u e f o r growing 3 - d i m e n s i o n a l g u i d e s i n v o l v e s e t c h i n g c h a n n e l s or mesas and growing over t h ~ - b y M B E ( 1 8 , 1 9 ) . Since t h e beams can be r e g a r d e d as b e i n g p a r a l l e l , the l o c a l a r r i v a l r a t e i s p r o p o r t i o n a l to cos ~, where ~ i s the l o c a l i n c i d e n t a n g l e . When t h e s u r f a c e i s n o t f i a t the i n c i d e n t a n g l e v a r i e s p o i n t by p o i n t on t h e s u r f a c e . One of t h e more i n t e r e s t i n g g u i d e s f a b r i c a t e d by t h i s t e c h n i q u e i s shown i n F i g . 4. I t i n v o l v e s the growth i n c h a n n e l s a l i g n e d a l o n g the (Ti"o) d i r e c t i o n s , the two edges of the u n d e r c u t c h a n n e l s s e r v e as two edge s h a d o w l n g m a s k s p r e v e n t i n g atoms from the i n c i d e n t f l u x a r r i v i n g a t the s u r f a c e r i g h t u n d e r n e a t h them. As a r e s u l t , a s t r i p of e p i l a y e r bounded on a l l a i d e s by s l n g l e - c r y a t a l p l a n e s c o m p l e t e l y d i s c o n n e c t e d from the r e s t of the e p i l a y e r i s grown i n the c h a n n e l .
100
R . D . Burnham and D. R. Scifres
CLADDING
CORE
142 Fig. 3.
Transmission micrograph of core and cladding regions.
Also, because of the directionality of the molecular beam three-dimensional waveguides have been reported by masking molecular beams during epitaxy. As early as 1972 Cho and Reinhart (20) reported on a waveguide structure made by masking with a 50 ~m tungsten wire. The AI effusion cell is directed perpendicular to the substrate so that the wire can form a sharp shadow with a width approximately the diameter of the wire. The Ga effusion cell is situated at an angle with respect to the substrate so that the Ga flux does not fall on the same region of the substrate as does the AI flux. This allows the GaAs beam to reach behind the masking wire where the AI beam is shadowed thus resulting in a strip containing only GaAs. Tsang and llegems (21) more recently reported on single and multilayered stripe mesa waveguides with widths as narrow as i ~m using single-level or multilevel masks. These masks were made from single crystal Si wafers. Silicon was chosen as the mask material because it has a low vapor pressure and is relatively inert towards GaAs at the usual growth temperature (~580°C). This makes it possible to press the Si mask directly against the GaAs substrate without contaminating the substrate. The use of single crystalline Si substrates also allows one to generate mask openings with dimensions 51 Bm by using preferential etchants. Figure 5 shows examples of single and multiple growth through various stripe patterns on single crystal Si masks.
Integrated Optical Devices Fabricated by MBE
. ,-, _
(Jl O)
101
(IT01
= 0.91,~m
(001) ¢ 5.18~u m
(ool
! t(o01 ) =3.18Aim
t (a)
(b)
(c) Fig. 4.
( a ) a n d ( b ) show t h e c ~ o s s - s e c t i o n a l v i e w s o f MBE l a y e r s o v e r two c h a n n e l s o f w i d t h 11 a n d 29 ~m aligned along the 110 d i r e c t i o n on a ( 0 0 1 ) o r i e n t e d GaAs substrate. (c) A SEM picture of the
channel edge showing the surface quality of the as-grown crystal facets.
102
R . D . Burnham and D. R. Scifres
Fig. 5.
Optical mlcrographs of slngle-level (a) and twolevel (b) depositions.
I n t e g r a t e d O p t i c a l D e v i c e s F a b r i c a t e d by ~ E
103
The l a t e s t i n improvements f o r making 3 - d l m e n s i o n a l g u i d e s i n v o l v e s m o l e c u l a r b e a m e p i t a x i a l w r i t i n g of p a t t e r n e d GaAs e p i l a y e r s t r u c t u r e s w i t h c o n t r o l l e d t h i c k n e s s p r o f i l e s , l e n g t h w i s e v a r y i n g t h i c k n e s s e s , f e a t u r e l e s s and o p t i c a l l y smooth s i d e w a l l s ( 2 2 ) . T h i s i s a c h i e v e d by h a v i n g a f i x e d S i mask w i t h p a t t e r n e d o p e n i n g s between t h e e v a p o r a t i n g s o u r c e s and t h e s u b s t r a t e w h i l e t h e s u b s t r a t e i s moved i n a p l a n e p a r a l l e l t o t h e mask d u r i n g growth ( s e e F i g . 6 . ) . Y MOTION
~/MOVING SUBSTRATE I'": '..".": . : . - ~ = . " l--_--_' ----.
((J
,:~'.: ".: :." .'. : = t . , ,','." " . ' . . ' .
r~
I....'...
~'.".'.".'" %'..-.
i.'.'.
-- - - - -.'_--_..__.~ ---
~ ~.[.'.:]
-----
--l~: .__._
-~-X MOT~N
Fig. 6.
B a s i c a r r a n g e m e n t used f o r a c h i e v i n g MBE w r i t i n g .
CaAs e p i l a y e r s t r u c t u r e s w r i t t e n b e h i n d a s e r i e s o f s q u a r e mask o p e n i n g s w i t h d i f f e r e n t s i z e s when t h e m a s k - s u b s t r a t e s e p a r a t i o n i s s m a l l i s shown i n F i g s . 7 a - d . The v a r i o u s t e c h n i q u e s m e n t i o n e d f o r f a b r i c a t i n g waveguides s h o u l d g i v e t h e r e a d e r a flavor for the high degree of versatility t h a t i s a c h i e v a b l e by MBE. I t i s i m p o r t a n t to p o i n t o u t t h n t m a n y o t h e r t y p e s o f waveguide s t r u c t u r e s t h a t have been f a b r i c a t e d by o t h e r growth t e c h n i q u e s can i n p r i n c i p l e be f a b r i c a t e d by MBE.
5.
DISCRETE L A S E E S
Ever s i n c e t h e f i r s t p - n j u n c t i o n GaAs i n j e c t i o n l a s e r s were d e m o n s t r a t e d i n 1962, c o n s i d e r a b l e e f f o r t h a s been expended i n t r y i n g t o r e f i n e t h e s e m i c o n d u c t o r l a s e r to t h e p o i n t t h a t i t i s no l o n g e r j u s t a l a b o r a t o r y c u r i o s i t y b u t a u s e f u l d e v i c e . P r o g r e s s was slow u n t i l Cal_xAlxAs DH l a s e r s were d e v e l o p e d i n 1970 ( 2 3 , 2 4 ) . These DH l a s e r s d e c r e a s e d t h e room t e m p e r a t u r e t h r e s h o l d c u r r e n t two o r d e r s o f magnitude o v e r t h a t o f i D / t i a l l a s e r s made o n l y of CaAs. T h i s tremendous r e d u c t i o n i n t h r e s h o l d c u r r e n t i s a t t r i b u t e d to t h e l a r g e r e f r a c t i v e i n d e x and bandgap d i f f e r e n c e between t h e t h i n Cat_xAlxAs ( a c t i v e ) l a y e r and t h e o u t e r two Cal_yAlyAa l a y e r s when y - x • O.15 and x ~ 0 . 2 5 . The l a r g e r e f r a c t i v e i n d e x d i f f e r e n c e s e r v e s to s i g n i f i c a n t l y c o n f i n e t h e l i g h t g e n e r a t e d i n t h e Cal-xAlxAs ~the a c t i v e r e g i o n ) w h i l e t h e w i d e r bandgap o f t h e Gal-yA1vAs c a u s e s t h e i n j e c t e d c a r r i e r s i n t h e Cal_XAlxAs r e g i o n t o be s l m i l a r l y c o n f i n e d . F o r p r a c t i c a l CW l a s e r s i t i s a l s o n e c e s s a r y t o r e s t r i c t t h e c u r r e n t t o a n a r r o w s t r i p e to l i m i t t h e a c t i v e r e g i o n l a t e r a l l y ( 2 5 ) .
104
R . D . Burnham and D. R. Scifres
Fig. 7.
MBE mask writing through a series of sizes of square mask openings.
Discrete lasers are generally made by cleaving the faces of the semiconductor crystals These cleaved facets form plane, parallel mirror-like surfaces which reflect a portion of the light back into the region of the p-n junction. The reflected light is amplified, and the energy density within the active waveguide of the laser continues to build up to produce an intense laser beam. The first GaAs-Gal_xAlxAS discrete cleaved facet DH lasers prepared by MBE were reported by Cho and Casey (26). The threshold currents of these lasers were 13 times higher than similar stuctures grown by LPE. It was found that the threshold ratio could be decreased to 1.4 if the samples were annealed at 750°C or above. Contamination of the GaAIAs layers was thought to be the major problem. Later Cho et aZ. (27), reported CW laser operation at temperatures as high at lOP C. The threshold currents were still about 1.4 times higher but they did not need to be annealed. This was achieved by eliminating more of the hydrocarbon and water vapor sources in the epltaxial growth system by using only pyrolytic BN effusion cells and installing an additional liquid-nitrogen-cooled panel around the susbstrate area. Also, the AI concentrations at all the GaAs-Gal_xAlxAs interfaces were graded in order to reduce the lattice mismatch at the interfaces because of the reduced MBE growth temperature 580°C. Another problem with making GaAs and GaAIAs lasers by MBE was to find suitable dopants. Impurities like Zn (28), Cd (28) and Mg (29) have high vapor pressure at the growth temperature compared to the group-Ill elements. Thus, they tend to have very short absorption lifetimes on the surface and re-evaporate before any significant incorporation in the grown layer can occur. Te (30,31) also has a high vapor pressure at the growth temperature compared to the group-Ill elements but it has a peculiar problem in that it tends to accumulate on the surface giving nonuniform doping and poor surface morphology. Apparently a stable surface compound (GaTe) is formed. A significant increase in sticking coefficient can be realized by using an ionized dopant beam. Naganuma et aZ.
Integrated Optical Devices Fabricated by MBE
106
(32,33), reported that GaAs layers could be doped heavily p-type -5 X 1019 (cm -3) Zn by ionizing the Zn atoms and weakly accelerating them much like ion implantation. The weakly accelerated Zn ions can penetrate just beneath the growing surface ~10-20
A, and be incorporated i n t o the grown layer by successive growth.
For amphoterlc
impurities such as the group-IV elements in GaAs, the site distribution ratio appears to be determined (34) by surface stoichiometry during growth. As a result, the elements Si (35), Ge (34,35) and Sn (35) are incorporated primarily as donors under the usual As-rich growth conditions with As-stabilized surface structure, whereas for the case of Ge it has been shown that the same element is incorporated primarily as an acceptor under Ga-stabilized growth conditions. So far, however, it has been difficult to achieve routine mirror-like surface appearance under Ga-stablized growth conditions. Ilegems 8t GZ. (36), reported using Mn as a p-type dopant. Even though Mn has a high sticking coefficient it is a ~iiO meV deep acceptor and thus doping levels of only 1018 cm -3 could be achieved. A much more promising impurity is Be (37) where at least 3 x 1019 cm -3 Be can be incorporated in GaAs and Gal_xAl xAs (up to x ~ 0.33). In fact, the Be doping level is simply proportional to the Be arrival rate. At this point in time Sn appears to be the most attractive n-type dopant and either ionized Zn or Be are attractive for heavy p-type doping. Heavy doping is necessary in order to reduce the series resistance of the device to minimize thermal heating which would allow for CW laser operation. We have talked about oxygen contamination in GaAIAs as being very deleterious for making DH Gal_yAiFAs/Gal-xAlxAs lasers. There appears to be a promising aspect to oxygen doped GaAIAs layers and that is they can be made insulating. Casey 8t aZ. (38) demonstrated that a lattice matched single crystal insulator-semiconductor heterojunction can be grown by doping Ga0.5AI0.5As layer with oxygen by MBE. This could have significant impact on monolithic device fabrication. For example, this could eliminate the need for an earlier technique developed by Cho and Ballamy (39) for planar technology in which the active device positions are defined by windows in SiO 2 layers which are deposited prior to MBE growth. By this process, polycrystalline and monocrystalline GaAs are selectively grown producing a pattern which consists of areas of device quality single-crystal material isolated by semiinsulating polycrystalline material. In fact, Lee and Cho (40) reported on single transverse mode injection lasers with embedded stripe layers using an SiO 2 mask. A scanning electron-beammicrograph of the cleaved end face of the laser view at an angle is shown in Fig. 8. Along the same llne Cho et al. (41), demonstrated that semiconducting materials consisting of inlays of different doping or composition in selective areas may ha grown by MBE. They showed it is possibleto fill an etched hole with MBE without formation of voids resulting from facet growth as observed in other growth techniques, Optically pumped laser oscillation from quantum states in thin multilayer heterostructures have been observed (42,43). Energy levels of electrons and holes in a very thin GaAs layer which is sandwiched between wider bandgap layers of Ga0.gAI0. 2 As consist of a discrete number of bound states within the well. The MBE growth technique not only allows for the formation of rectangular potential wells but by programming the variation of the AI content essentially any arbitrary potential can be repeatedly deposited. Lasers taking advantage of quantum size effects may be able to add a whole new dimension to the improvement of laser properties.
6.
LASER INTEGRATION
In order to take full advantage of the concepts of integrated optics it is necessary to fabricate laser structures which can be integrated onto the same substrate as are the other optical and electronic components. Unfortunately, the conventional
106
R. D. Burnham and D. R. Scifres
egrowth 2.5/~m ~
/~m m]
5/~m)
m~
Fig. 8.
SEM of the cross-section of an embedded stripe laser.
method of fabrlcat~ng the mirrors on a semiconductor laser involves cleavage of the crystal. Thus, the laser chip must be broken or cleaved off the substrate in order to provide optical feeback. Within the last several years there have been a number of methods developed for fabricating mirrors or reflecting elements which are compatible with monolithic thin film technology. The first to be realized was distributed feedback (DFB) (44) in which an optical grating is used to diffract incoming light back upon itself. These original devices were grown by LPE. If the period of the grating A is an integer multiple of half-wavelengths of light in the guide (A&) a small amount of light is reflected back into the waveguide conherently at each corrugation, the sum providing sufficient feedback to operate the laser. Distributed-feedback heterostructure lasers have several advantages as discrete devices over conventional cleaved-n~rror lasers. First, they generate a reproducible laser wavelength, which is four times less sensitive to thermal variation than is a conventional cleaved mirror laser (45,46). Second, the laser usually emits in a narrow-bandwldth single-longitudinal (47,48) mode. Third, as mentioned in part A of this section, the grating in a DFB laser also couples out beams which have approximately IOO times improved directionality (ll) over those emitted through the cleaved-end mirror or a conventional laser diode. Thus, these DFB devices, in addition to serving as sources for integrated optical circuits, may also prove useful a s s t a n d - a l o n e d i s c r e t e d e v i c e s . One p r o b l e m t h a t d o e s e x i s t w i t h DFB l a s e r s is that non-radiatlve recombination centers are introduced by the grating. One solution is to physically separate the corrugations from the active region (11). Casey et uS. (49), found that ion milling of the corrugations did not affect the radiative recombination as long as the corrugations were removed 0.30 ~m or more from the layer confining the carriers. This involved a hybrid LPE and MBE growth process (see Fig. 9). MBE is used to grow over the corrugations since growth on a GaAIAs surface is not reproducible using LPE.
Integrated Optical Devices Fabricated by MBE
107
AS sAS
I As ~,S
5p.m
Fig. 9.
SEM cross-section of a DFB laser.
It was suggested by S. Wang (50) that instead of gratings being used to provide d i s t r i b u t e d r e f l e c t i o n t h r o u g h o u t t h e l a s e r c a v i t y , t h e y could be p u t a t the ends of the c a v i t y to a c t as m i r r o r s . L a s e r s of t h i s type a r e c a l l e d DBR ( D i s t r i b u t e d Bragg R e f l e c t i o n ) l a s e r s , and they appear to have more promise than DFB l a s e r s . The removal of the g r a t i n g from the a c t i v e r e g i o n has t h e o b v i o u s advantage of e l i m i n a t i n g the n o n r a d i a t i v e r e c o m b i n a t i o n a s s o c i a t e d w i t h i t , t h u s a l l o w i n g s i m p l e r s t r u c t u r e s and s i m p l e r p r o c e s s i n g f o r the DBR l a s e r . The e x c e l l e n t l a y e r t h i c k n e s s and c o m p o s i t i o n c o n t r o l a f f o r d e d by MBE should prove q u i t e v a l u a b l e i n f a b r i c a t i n g DBR l a s e r s s i n c e such c o n t r o l makes i t p o s s i b l e to d e t e r m i n e a p r i o r i the c o r r e c t g r a t i n g s p a c i n g . The DBR l a s e r should a l s o have a t l e a s t as good f r e q u e n c y and mode c o n t r o l as the DFB l a e r . I n t e g r a t i o n of DH l a s e r s by f a b r i c a t i n g d i s c r e t e p a r a l l e l r e f l e c t o r s w i t h o u t c l e a v i n g the crystal has also been achieved by selective ion sputtering (51), chemical etching (52-53) and epitaxy (54). Although these devices were not grown by MBE, they could have easily been with the possible exception of Ref. 54 since it involves facet formation along with selective epitaxy.
7.
COUPLERS
As mentioned earlier the interconnection of optical components via waveguides is the central problem of monolithic integrated optics and is at the mercy of the crysal grower and the growth technique he chooses to use. MBE appears to be particularly suitable for making structures which allow coupling from one waveguide to another. Conwell and Burnham (11) d i s c u s s e s s e n t i a l l y f i v e k i n d s of c o u p l e r s which a r e : p r i s m , g r a t i n g , b u t t and c o m p o s i t i o n a l , t a p e r e d and d i r e c t i o n a l or e v a n e s c e n t . Prism, g r a t i n g , and b u t t c o u p l e r s a r e m a i n l y used to c o u p l e l i g h t out of t h e o p t i c a l c i r c u i t and w i l l n o t be d i s c u s s e d i n t h i s p a p e r . The t a p e r e d c o u p l e r has r e c e i v e d c o n s i d e r a b l e a t t e n t i o n f o r c o u p l i n g between waveguides. A t a p e r e d c o u p l e r i s made by t a p e r i n g down the end o f a waveguide. When a g u i d e mode t h a t has been u n d e r g o i n g s u c c e s s i v e r e f l e c t i o n s w i t h the a n g l e 8m r e a c h e s t h e t a p e r e d r e g i o n i t s r e f l e c t l o n s
108
R.D.
Burnham and D. R. Scifres
take place at progressively smaller 0m because of the shape of the taper, until 0m becomes smaller than the critical angle. Beyond that point it is no longer confined by the guide, and is refracted into an adjacent waveguide of slightly lower index of refraction. Taper couplers are quite easily made by MBE and are extremely uniform, gradual and reproducible (see Fig. IO). Prior to starting a growth which is to have a taper an 0.25 mm thick knife-edged mask is swung into place i or 2 mm above the surface of the sample. This masks a portion of the substrate from~the Ga and As sources. A GaAs layer grows on the unmasked regions, terminated by linear tapers -200 ~m wide which grew in the penumbra areas of the mask edge. Merz and coworkers report (55) that coupling efficiencies approaching 100% can be routinely achieved by MBE while efficiencies of only ~50% have been achieved by LPE. Composition coupling involves the transition from one composition to another along a waveguide. Various types of composition coupling schemes have been reported, however, none involves MBE. With the larger repertoire of MBE growth techniques there should be no problem in making composition couplers by MBE. On the other hand, with taper couplers approaching 100% efficiency there does not seem to be much incentive to develop composition couplers. When two identical guides are sufficiently close (within about a wavelength of light) that the exponentially decaying (evanescent) light from one extends into the other the energy will oscillate between the guides with a characteristic period L ° called the coupling lenth. Evanescent wave coupling has been used in twin guide lasers (51). Here the active waveguide of the laser is separated by a region of lower index from a passive guide separation being small enough so that the evanescent fields overlap. Twin guide lasers so far have only been fabricated by LPE but it is obvious that MBE is much more suitable fG_ its fabrication since layer thicknesses and index of refraction can be controlled so readily.
8.
SWITCHES, MODULATORS AND DETECTORS
We refer the reader again to the review article by Conwell and Burnham (II) for a more complete coverage of switches, modulators and detectors related to GaAs and its alloys. It suffices to say that although none of these components have been fabricated by MBE techniques these principles are well understood and have been fabricated by other growth techniques and MBE does not seem to present any limitations.
9.
MULTI-COMPONENT INTEGRATION
Although there has been various types of multi-component integration including the integration of a laser, waveguide and detector recently reported by Merz and Logan (56), only one involves growth by MBE. Reinhart and Cho (57) reported growth of a Gal_~lyAs-Gal-xAlxAs laser taper coupled to a passive (more transparent) waveguide grown by MBE. The structure is similar to the schematic shown in Fig. lOa except additional epitaxial layers are grown to facilitate electrical contact. Layer to is the active layer of the devices and layer t0. I to the right of the taper services as the passive waveguide. These particular structures indicated significant absorption losses.
I0.
CONCLUSIONS
MBE has certainly demonstrated that monolayer dimensional control along with a wide range of doping and compositional profiles is achievable. Because of this versatility, MBE is a promising growth technique for the fabrication of extremely compli-
slntegrated Optical Devices Fabricated by MBE
109
AIR
(o)
Fig. IO.
Tapers grown by MBE.
cated integrated optical circuits. However, considerable effort and improvement is needed to identify and suppress the sources of contamination during the growth of GaAs and especially GaAIAs. Already improved system designs are beginning to be implemented such as the incorporation of elaborate vacuum interlocks, improved cryopumping, improved substrate heaters and source heater assemblies. These steps should significantly reduce eomtamlnation during growth. Calawa (58) recently reported that the introduction of hydrogen during MBE growth of GaAs seems to significantly improve the mobility of the material. As soon as MBE grown GaAIAs can reproducibly compare with LPE grown GaAIAs, then significant advances in inte-
110
R. D. Burnham and D. R. Scifres
g r a t e d o p t i c s can be e x p e c t e d . S o l v i n g the c o n t a m i n a t i o n problem i n MBE i s s o r t of analogous to solving the waveguiding and carrier confinement problems for lasers. As was mentioned earlier, this was achieved by making DH lasers. Without it laser progress was slow. Of course, solving the GaAIAs contamination problem may, on the other hand, simply require the use of alloys which do not contain A1. For example, Miller et al. (59), recently reported on InP/GalnAs/InP lattice matched DH lasers grown by MBE. We are grateful to R. Z. Bachrach and A. Y. Cho for many helpful discussions. Acknowledgements are due H. C. Casey, Jr., T. P. Lee, J. L. Merz and W. T. Tsang for supplying photographs of their work. Special thanks go to D. Bernsen for secretarial service.
REFERENCES
i. 2. 3. 4. 5. 6. 7. 8. 9. i0. ii. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
Miller, Bell Syst. Tech. J. 48, 2059 (1969). Dupuis and P. D. Dapkus, Appl-~Phys. Lett. 32, 406 (1978). Cho and F. K. Reinhart, J. Appl. Phys. 45, 1812 (1974). Arthur and J. J. LePore, J. Va~um Sci.-~echnol. 6, 545 (1969). Cho and M. B. Panish, J. Appl I. Phys. 43, 5118 (1972). Chang, L. Esaki, W. E. Howard, R. Ludeke, and G. Schul, J. Vucuum Sci. Technol. i00, 655 (1973). R. Dingle, W. Wiegmann, and C. H. Henry, Phys. Rev. Lett. 33, 827 (1974). R. Dingle, A. C. Gossard, and W. Wiegmann, Phys. Rev. Lett. 34, 1327 (1975). A. C. Gossard, P. M. Petroff, W. Wiegmann, R. Dingle, and A. Savage, Appl. Phys. Lett. 29, 323 (1976). P. K. Tien, Rev. Mcd. Phys. 49, 361 (1977). E . M . Conwell and R. D. Burnham, Ann. Rev. Mater. Sci. 8, 135 (1978). M. K. Barnoski, ed., Introduction to Integrated optics, Plenum, New York (1974) T. Tamir, ed., Integrated Optics, Springer, New York (1975). J. L. Merz and A. Y. Cho, Appl. Phys. Lett. 28, 456 (1976). J. L. Merz, A. C. Gossard, and W. Wiegmann, Ap---pl.Phys. Lett. 30, 629 (1977). A. Y. Cho, A. Yariv, and P. Yeh, Appl. Phys. Lett. 30, 471 (197). J. C. Tracy, W. Wiegmann, R. A. Logan, and F. K. Reinhart, Appl. Phys. Lett. 22, 511 (1973). W. T. Tsang and A. Y. Cho, Appl. Phys. Lett. 30, 293 (1977). S. Nagata, T. Tanaka, and M. Fukai, Appl. Phys-f Lett. 30, 503 (1977). A. Y. Cho and F. K. Reinhart, Appl. Phys. Lett. 21, 355 (1972). W. T. Tsang and M. Illegems, Appl. Phys. Lett. 31, 301 (1977). W. T. Tsang and A. Y. Cho, Appl. Phys. Lett. 32, 491 (1978). Zh. I. Alferov, V. M. Andreev, D. Z. Garbuzov, Yu. V. Zhilgaev, E. P. Morozov, E. L. Portnoi, and V. G. Trofim, Fiz. Tekh. Poluprovodn 4, 1826 (1970). I. Hayashi, M. B. Panish, P. W. Foy, and S. Sumski, Appl. Phys. Lett. 17, 109 (1970). J. E. Ripper, J. C. Dyment, L. A. D'Asaro, and T. L. Paoli, Appl. Phys. Lett. 18, 155 (1971). A. Y. Cho and H. C. Casey, J r . , Appl. Phys. Lett. 2_55,288 (1974). A. Y. Cho, R. W. Dixon, H. C. Casey, Jr., and R. L. Hartman, Appl. Phys. Lett. 28, 501 (1976). ~?. R. Arthur, Surf. Sci. 3-8, 394 (1973). A. Y. Cho and M. B. Panish, J. Appl. Phys. 43, 5118 (1972). J. R. Arthur, Surf. Sci. 43, 449 (1974). A. Y. Cho and J. R. Arthur, Prog. Solid. St. Chem. iO, 157 (1975). M. Naganuma and K. Takahashl, Appl. Phys. Lett. 27, 342 (1975). N. Matsunaga, M. Naganuma, and K. Takahashi, Jap. J. Appl. Phy8. 16, 443 (1976) A. Y. Cho and I. Hayashi, J. Appl. Phys. 42, 4422 (1971). A. Y. Cho, J. Appl. Phys. 46, 1732 (1975). S. R. A. J. A. L.
E. D. Y. R. Y. L.
Integrated Optical Devices Fabricated by ~ E 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59.
111
M. llegems, R. Dingle, and L. W. Rupp, Jr., J. AFpl. P~8. 46, 3059 (1975). M. Ilegems, O. A~l. Phys. 48, 1278 (1977). H. C. Casey, J r . , A. Y. Cho, and E. H. N i c o l l i a n , AFpl. Phys. Left. 32, 678 (1978). A. ¥. Cbo and W. C. Ballamy, J. Appl. Phys. 4._66, 783 (1975). T. P. Lee and A. Y. Cho, Appl. Phys. Lstt. 2_.99, 164 (1976). A. Y. Cho, J. V. Dilorenzo, and G. E. Mahoney, IEEETRrn8. on Electr. Devices ED-24, 1186 (1977). J. P. van der Ziel, R. Dingle, R. C. Miller, W. Wiegnmnn, and W. A. Nordland, Jr., Appl. Phys. Left. 26, 463 (1975). R. C. Miller, R. Dingle, A. C. Gossard, R. A. Logan, W. A. Nordland, Jr., and W. Wiegmann, J. App,. Phys. 47, 4509 (1976). D. R. Sclfres, R. D. Burnham, and W. Streifer, Appl. Phys. Lett. 25, 203 (1974) M. Nakamura, K. Aiki, J. Umeda, A. Katzlr, A. Yariv, and H. W. Yen, IEEE J. QunntT~mElec~. qE-ll, 436 (1975). R. D. Burnham, D. R. Scifres, and W. Streifer, App,. Phys. Lett. 29, 287 (1976). R. D. Burnham, D. R. Scifres, and W. Streifer, IEEE J. Quantum Electr. qE-ll, 439 (1975). D. E. Scifres, R. D. Burnham, and W. Streifer, IEEE Tr~rn8. on Electr. Devices ED-22, 609 (1975). M. Ilegems, H. C. Casey, S. Somekh, and M. B. Panish, J. C r ~ s ~ G ~ t h 3_!1, 158 (1975). S. Wang, IEEE J. Q~o~t~mElectr. qE-IO, 413 (1974) Y. Suematsu, M. Y-m, da, and K. Hayashi, IEEE J. ~unt~m EZeetra. qE-ll, 457 (1975). C. Hurwitz, J. A. Rossi, J. J. Hsieh, C. M. Wolfe, Appl. Phys. Left. 27, 241 (1975). J. L. Merz and R. A. Logan, Appl. Phys. Left. 47, 3503 (1976). J. C. Campbell and D. W. Bellavance, IEEE J. Quuntw, Electr. Eg~!, 253 (1977). J. L. Merz, R. A. Logan, W. Wiegmann, and A. C. Gossard, App,. Phys. Left. 2._66,337 (1975). J. L. Merz and R. A. Logan, Appl. Phys. Left. 30, 530 (1977). F. K. Reinhart and A. Y. Cho, Appl.'Phys. Lett?--3_!l, 457 (1977). A. R. Calawa, 2OthAnnual Electronic Materials Conference, Santa Barbara, California, June 28-30, 1978. B. I. Miller, J. S. McFee, R. J. Martin, and P. K. Tien, Appl. Phy8. Left. 33, 44 (1978).
112
R . D . Burnham and D. R. Scifres
THE AUTHORS
R. D. Burnham
D.R. Scifres
Robert D. Burnham was born in Havre de Grace, Maryland, on 21 March 1944. He receivea his B.S., M.S., and Ph.D. degrees from the University of lllinois~ Urbana, in 1966, 1968, and 1971, respectively. He held an NDEA Fellowship from 1966-1969, and a General Telephone and Electronics Fellowship from 1969-1971. He has been a member of the Research Staff at the Xerox Palo Alto Research Center, Palo Alto, California, since 1971. He is currently in charge of the materials growth of III-V semiconductor lasers by molecular beam epitaxy, liquid phase epitaxy and metalorganic chemical vapor deposition. He has been a co-author of over eighty papers. His most significant contributions were in the area of distributed feedback, quaternary, buried heterostructure, and InGaP lasers. Dr. Burnham is a member of Tau Beta Pi, Eta Kappa Nu, Sigma Tau, the Electrochemical Society, and the AIME.
Integrated Optical Devices Fabricated by MBE
113
Don R. S c i f r e s was born i n L a f a y e t t e , I n d i a n a , on 10 September 1946. He r e c e i v e d t h e B.S. degree w i t h honors i n E l e c t r i c a l Engineering from Purdue U n i v e r s i t y , West L a f a y e t t e , I n d i a n a , i n 1968, and the M.S. and P h . D . ! d e g r e e s i n E l e c t r i c a l Engineering from t h e U n i v e r s i t y o f I l l i n o i s , Urbana, I l l i n o i s , i n 1970 and 1972, r e s p e c t i v e l y . While a t the U n i v e r s i t y of I l l i n o i s he h e l d a U n i v e r s i t y of I l l i n o i s Doctoral Support F e l l o w s h l p and a General Telephone and E l e c t r o n i c s F e l l o w s h l p . In 1972, he joined the Xerox Palo Alto Research Center where he is currently Manager of an Opto-Electronics Group. His research has been involved with developing semiconductor injection lasers as well as incorporating optical and electronic components in the semiconductor laser chip to achieve such integrated optical functions as distributed feedback, transverse mode control, and electronically controlled laser beam deflection. Dr. Scifres is on the Board of Editors for the journal of Fiber and Integrated Optics, is a national lecturer on "Semiconductor Lasers" for the Quantum Electronics group of IEEE, is a Fellow of the Optical Society of America, and a member of IEEE and the American Physical Society.
0146-3E~179/0701-0086t06.00/0
INTEGRATED OPTICAL DEVICES FABRICATED BY MBE R. D. Burnham and D. R. Scifres XEROX PARC, Palo Alto, CA 94304, U.S.A.
(Submitted March 1979]
I.
INTRODUCTION
Approximately IO y e a r s ago, S. E. M i l l e r ( I ) a t B e l l Telephone L a b o r a t o r i e s o u t l i n e d a p r o p o s a l f o r t a k i n g v a r i o u s d i s c r e t e and b u l k y o p t l c a l components, which f o r many y e a r s had been spread out over a l a r g e o p t i c a l bench, and m i n i a t u r i z i n g them on a common r i g i d s u b s t r a t e . The i n c o r p o r a t i o n o f such d i s c r e t e d e v i c e s i n t o an o p t i c a l c i r c u i t would e l i m l n a t e m e c h a n i c a l and thermal s t a b i l i t y problems, a l o n g w i t h the need f o r c o s t l y and heavy o p t i c s and e x p e n s i v e b u l k y o p t i c a l t a b l e s . The s i z e of t h e s e m i n i a t u r i z e d components a t l e a s t i n one or two d i m e n s i o n s needed to be o n l y as l a r g e as the o p t i c a l c a r r i e r wave (O.4-10 ~m) and thus were amenable to mass p r o d u c t i o n u s i n g s t a n d a r d p h o t o l i t h o g r a p h i c t e c h n i q u e s . I t was proposed a t t h a t time t h a t the m i n i a t u r i z a t i o n of ~uch components onto a common s u b s t r a t e , a c h i p , be called an integrated optical circuit in view of its electrical counterpart the integrated circuit. Today, several components of such an integrated optical circuit have indeed been i n t e g r a t e d onto a common s u b s t r a t e and have been d e m o n s t r a t e d to a c h i e v e improved s t a b i l i t y , s m a l l s i z e , l i g h t w e i g h t , and low power r e q u i r e m e n t s . I n f a c t , some c h i p s c o n s i s t i n g of a r r a y s of l a s e r s now appear to be a r e a l i t y f o r o p t i c a l communications a p p l i c a t i o n s and no doubt o t h e r a p p l i c a t l o n s w i l l f o l l o w . F i b e r o p t i c c c ¢ ~ m n i c a t i o n s has p r o v i d e d the p r i m a r y s t i m u l u s f o r i n t e g r a t e d o p t i c s research. I n t e g r a t e d o p t i c s coupled w i t h f i b e r o p t i c s i s expected to f i n d a p p l i c a t i o n i n communications where the d a t a r a t e s exceed 300 Mb/sec and where s p a t l a l m u l t i p l e x i n g i s e x t e n s i v e l y u s e d . This a p p l i c a t i o n may r e q u i r e s i n g l e mode, CW l a s e r s o u r c e s modulated a t r a t e s up to 10 GHz, low l o s s s i n g l e mode f i b e r s and GHz response photodectectors. M o n o l i t h i c and h y b r i d a r e two approaches i n which one could make i n t e g r a t e d o p t i c s . The h y b r i d approach uses s e v e r a l m a t e r i a l s which a r e chosen to o p t i m i z e each of the components. This o b v i o u s l y can p r e s e n t s e r i o u s problems s i n c e g e n e r a l l y p r o c e s s e s used to f a b r i c a t e the d i f f e r e n t d e v i c e s might be i n c o m p a t i b l e . The m o n o l i t h i c approach uses one m a t e r i a l which works w e l l f o r a l l the components. S i n c e the b e s t l a s e r , m o d u l a t o r , waveguide and d e t e c t o r a r e n o t n e c e s s a r i l y found i n one m a t e r l a l , compromises w i l l have to be made. F o r t u n a t e l y , however, some s e m i c o n d u c t o r s appear to be s u i t a b l e f o r m o n o l i t h i c o p t i c a l i n t e g r a t i o n . In fact, b e c a u s e of the d u a l r o l e t h a t d i r e c t bandgap s e m i c o n d u c t o r s p l a y i n b o t h g e n e r a t i n g and d e t e c t i n g o p t l c a l r a d i a t i o n , such m a t e r i a l s a r e v i r t u a l l y the o n l y ones s u i t a b l e for monolithic integration.
95
96
R. D. Burnham and D. R. Scifres
2.
GaAs AND RELATED ALLOYS
One of the most promising and most worked on semiconductor materials is GaAs and its alloys with columns III-V. GaAs technology has progressed to the point that high quality fairly large area substrates (4-5 cm in diameter) can be produced. Substrates of other III-V compounds of similar size are either not available or the cost is prohibitive. The strong concentration on the GaAs family is due to its versatility in terms of its electrlcal and optical properties and to the fact that it is the only materials system in which all the important optical functions -- light generation, guiding, modulation, and detection -- have been achieved. This makes it possible, in principle, to make monolithic integrated optical circuits out of the GaAs family with suitable addition of impurities, just as integrated circuits are made of silicon. An attractive feature of GaAs for integrated optics is that its alloys with other compounds of column III-V elements have refractive indexes and bandgaps that vary with the amount of the other compound added. For example, the refractive index of Gal_xAlxAs at a wavelength of 0.9 ~m decreases almost linearly with x from 3.59 for x = 0 to 2.97 for x = I (see Fig. i). It is clear then that waveguides can be made by epitaxial growth of layers of different x on top of each other. For such structures to have good device performance, it is important that the lattice constants of adjoining layers be well matched. If that is not the case, metallurgical imperfections, such as dislocations, will occur at the interfaces, giving rise to such deleterious features as optical scattering centers and electron-hole recombination centers. Gal-~AlxAs is particularly attractive because the lattice constant for GaAs is 5.653 A and for AlAs, 5.661 A; thus, the largest possible difference between lattice constants in this family is only 0.14% 3
.
7
~
2
.
2
300K 3.6[--
J
42.1
%% 3.5
\
2.0
\ \ \
x
1.9
%%
3.4
x
al
\ 3.3
% %
X z LiJ 3.2
1.8 ~ %
%. %
> t(J
%
%. %. %
M. 3.1
i¢1
1.7 ~ >
%
%. %. ,% %. %.
3.0
. . . .
2.9
REFRACTIVE INDEX
w
1.6 ~ %. %. . 1,5 %. %. 1.4
ENERGY GAP
[}.0
0.5
1.3 1.0
MOLE FRACTION AlAs, x
Fig. i.
Variation of refractive index and bandgap of Gal_ x AI As with x. x
Integrated Optical Devices Fabricated by MBE
97
The interconnection of optical components via waveguides is the central problem of monolithic integrated optics for two main reasons: first, the minute cross sections of the waveguides (10"8-10 -7 cm 2) requite strict dimensional tolerances, and second, optical circuits consisting of active and passive waveguide components need adjacent regions with different material properties such as composition, bandgap, index of refraction, doping, etc. These requirements can only be satisfied by taking the properties and limitations of the crystal growth techniques into account.
3.
MOLECULAR BEAM EPITAXY
There are three growth techniques which could be used for integrated optical circuits based on single crystal semiconductors, they are: liquid phase epitaxy (LPE), chemical vapor phase ep[taxy (VPE), and molecular beam epitaxy (MBE). Even though more work has been done with optical integration by LPE, MBE appears to be the mosL promising growth technique because of its overall versatility. VPE has recently received considerable recognition because of advances made by Dupuls and Dapkus (2) using metalorganics but it still lacks the overall flexibility required to make extremely complicated monolithic integrated optical circuits compared to MBE. The initial pioneering work on MBE was done by A. Y. Cho at about the same time that S. E. Miller proposed integrated optics. Today one of the major research goals is to merge these two technologies so as to produce a variety of integrated optical circuit elements. The advantages of growing GaAs/Gal_xAlxAs integrated optical circuits by MBE are many: (a) Since the growth is performed in ultra high vacuum with the availability of in situ diagnostic instruments, one can be assured that the desired surface conditions have been reached before comencement of the growth; (b) the MBE is controlled by kinetics rather than by a diffusion process and thus allows one to grow layers with practlcally any predetermined doping profile (3) and variation of composition (4,5); (c) large uniform layers (3 cm x 3 cm) with featureless surfaces may be grown by MBE; (d) the low epitaxial temperature (5800C) may prevent dopant diffusion both within the epitaxial layer and from the substrate into the epitaxial layer; and (e) the slow growth rate (I ~m/hr) allows for essentially monolayer <5 A thickness control (6-9). The major problem with MBE for integrated optics applications has been centered on the growth of high quality Gal_xAlxAs. This review article will therefore deal mainly with the various MBE techniques that have been used towards the goal of making monolithic integration in the GaAs/ Gal_xAlxAs system. We refer the reader to other review articles (I0,ii) and books (12,13) on a broader coverage of integrated optics which discuss other growth techniques and considerably more theory than will be presented here.
4.
WAVEGUIDES
First, consider the optical wavegulde, which is perhaps the most basic of all integrated optics components. Most of these guides are planar, i.e~, those with rectangular geometry. The guiding property can be understood by studying the heterostructure illustrated in Fig. 2. As shown, the waveguide consists of a sandwich of three different materlal compositions, each of which is transparent to the propagating light-wave. A necessary condition for waveguiding in these heterostructure layers is n2 > n I, n3, where n is the refractive index. The waveguiding property can be thought to result from the phenomenon of total internal reflection Consider a ray of light starting in material 2 and propagating toward the 2-3 interface. If the angle between the normal to the interface and the direction of lightwave propagation 823 is larger than the critical angle for total internal reflection
rR. D. Burnham and D. R. Scifres
HETE RO&TRUCTURE WAVEGUIDES
X m
13 HIGH INDEX
:l
LOW,, ,EX
MATERIAL ' LOW INDEX
I
LOW,NDEX
MATERIAL 2 MATERIAL 3 HIGH INDEX LOW INDEX
"1
n2
., "
"
n3
~,TE 0
~~,TE 1 /
~
TE2
Schematic diagram of a heterostructure way, guide.
Fig. 2.
[ ec = sin -I (n3/n2)], the light will be totally reflected upon striking this interface. The same holds true for a light ray propagating toward the 1-2 interface. The light rays then follow a ziH-zag path making the same angle e23 with each reflection. However, because of interference between rays that have traveled different paths, only those rays making certain angles will survive. These are called the modes of the guides. In terms of wave optics, each mode has a different electric field distribution across the guide and propagation constant along the guide. The propagation constant (8) can be determined from wave theory or from the ray-interference pattern. In either case, the wavelength of light in the guide (~g) is
g and the wave is said to see an equivalent refractive index neq given by neq =
Xo/Xg
where A is the free-space light wavelength. Since the guided wave exists in all three m~dia with nl, n2, and n3, the equivalent index is a sort of average satisfying n I , n 3
<
n
eq
<
n 2
•
I n t e g r a t e d O p t i c a l Devices F a b r i c a t e d by MBE
99
The e l e c t r i c - f i e l d d i s t r i b u t i o n s (12) of t h r e e modes, TEO, TEl, and TE2 a r e i l l u s t r a t e d i n F i g . 2; each has a d i f f e r e n t n e q . Guides c o n s i s t i n g of a Gal-xAlxAs l a y e r sandwiched between two l a y e r s of Gal_yAlyAs of lower i n d e x , thus y > x a r e examples of a o n e - d i m e n s i o n a l g u i d e . S i n c e - t h e r e a r e two c o n t a c t s between l a y e r s o f d i s s i m i l a r c o m p o s i t i o n , t h e s e a r e d o u b l e h e t e r o s t r u c t u r e (DH) g u i d e s . Such l a y e r e d l i g h t g u i d i n g s t r u c t u r e s can be r e a d i l y grown by any of the t h r e e p r e v i o u s l y m e n t i o n e d c r y s t a l growth t e c h n i q u e s . Merz and Cho (14) compared o p t i c a l l o s s e s f o r 1-2 pm t h i c k DH waveguides of s i m i l a r c o m p o s i t i o n grown by MBE and LPE. They r e p o r t e d l o s s e s l e s s than 1.5 cm- I f o r MBE grown g u i d e s between 1.1 and 1.4 eV, t h e e n e r g y of t h e GaAs DH l a s e r . These 1oases a r e comparable t o , or l e s s t h a n , the l o s s e s observed i n s i m i l a r wave-guide s t r u c t u r e s grown by the LPE. They did p o i n t o u t , however, t h a t d e s p i t e the h i g h degree of l a y e r u n i f o r m i t y and c o n t r o l t h a t MBE can a c h i e v e , t h e s e low l o s s e s a r e n o t u p r i o r i o b v i o u s . The r e a s o n i s t h a t the Gal-xAlxAS s u r f a c e exposed d u r i n g growth i n an MBE system i s h i g h l y r e a c t i v e which makes the l a y e r v e r y s u s c e p t l b l e to c o n t a m i n a t i o n by t r a c e amounts of water vapor and c a r b o n - c o n t a i n i n g g a s e s ; t h e s e or o t h e r c o n t a m i n a n t s might c o n t r i b u t e s i g n i f i c a n t l y to the below gap a b s o r p t i o n . These r e s u l Z s show, however, t h a t i s i t p o s s i b l e to grow l o w - l o s s Gal_xAiAs wave-guides by MBE w i t h t h e p r e s e n t - d a y vacuum t e c h n o l o g y . A f t e r Gossard e t uS. (9) r e p o r t e d the u l t i m a t e i n l a y e r c o n t r o l i n which up to 104 a l t e r n a t i n g s i n g l e or m u l t i p l e monolayers of GaAs and AlAs were s e q u e n t i a l l y d e p o s i t e d by MBE, Merz e t GZ. (15) (see F i g . 3) r e p o r t e d on l o w - l o s s wsveguldes whose c e n t r a l r e g i o n c o n s i s t e d of a l t e r n a t i n g sequences of a p p r o x i m a t e l y n i n e monolayers of GaAs and one monolayer of AlAs. The o u t e r l a y e r c o n s i s t e d of an a l t e r n a t i n 8 sequence of t h r e e monolayers of AIAs and seven monolayers of GaAs. This gave the c e n t r a l r e g i o n a lower a v e r a g e A1 c o m p o s i t i o n (x ~ 0.10) than the o u t e r sequence o f l a y e r s (x - 0 . 3 0 ) . O p t i c a l l o s s e s were ~1 cm-1 h i g h e r than the b e s t ~a0.9Al0.1As random a l l o y waveguides f a b r i c a t e d to date by MBE (14). Cho, Y a r i v , and Yeh (16) r e p o r t e d a n o v e l wavegulde grown by MBE i n which a Ga0.62 A10.38As g u i d i n g l a y e r was bounded on one s i d e by a i r and on the o t h e r s i d e by a periodic layered medium composed of alternating layers of GaAa and Ga0.8Al0.2As. The periodic media present high refelectivity to radiation which satisfies the Bragg condition. Unlike ordinary dielectric waveguldes, confined guiding with arbitrarily low loss is possible even when the guiding layer possesses an index of refraction which is lower than that of the periodic layers. V a r i o u s t y p e s of t h r e e - d i m e n s i o n a l h e t e r o s t r u c t u r e waveguides have been made by MBE. One attractive approach for making three-dimensional waveguldes involves embedding a GaAs waveguide i n a m a t r i x of Gal_xAlxAs. This i s a c h i e v e d by s e l e c t i v e l y e t c h i n g away p o r t i o n s of a p r e v i o u s l y grown p l a n a r wavegulde r e s u l t i n g i n a t h r e e - d l m e n s l o n a l g u i d e and t h e n e p i t a x i a l l y c o v e r i n g the n o n p l a n a r s u r f a c e w i t h Gal_xAlxAs by MBE to c o m p l e t e l y e n c a p s u l a t e the GaAs waveguide ( 1 7 ) . T h i s t e c h n i q u e d e m o n s t r a t e s one of the growth advantages MBE has over LPE. It is difficult to grow on GaAIAs by LPE once the sample is exposed to the ambient because of the reactive nature of AI. But with MBE the exposed GaAIAs surface need only be cleaned by ion milling prior to any regrowth. Another t e c h n q l u e f o r growing 3 - d i m e n s i o n a l g u i d e s i n v o l v e s e t c h i n g c h a n n e l s or mesas and growing over t h ~ - b y M B E ( 1 8 , 1 9 ) . Since t h e beams can be r e g a r d e d as b e i n g p a r a l l e l , the l o c a l a r r i v a l r a t e i s p r o p o r t i o n a l to cos ~, where ~ i s the l o c a l i n c i d e n t a n g l e . When t h e s u r f a c e i s n o t f i a t the i n c i d e n t a n g l e v a r i e s p o i n t by p o i n t on t h e s u r f a c e . One of t h e more i n t e r e s t i n g g u i d e s f a b r i c a t e d by t h i s t e c h n i q u e i s shown i n F i g . 4. I t i n v o l v e s the growth i n c h a n n e l s a l i g n e d a l o n g the (Ti"o) d i r e c t i o n s , the two edges of the u n d e r c u t c h a n n e l s s e r v e as two edge s h a d o w l n g m a s k s p r e v e n t i n g atoms from the i n c i d e n t f l u x a r r i v i n g a t the s u r f a c e r i g h t u n d e r n e a t h them. As a r e s u l t , a s t r i p of e p i l a y e r bounded on a l l a i d e s by s l n g l e - c r y a t a l p l a n e s c o m p l e t e l y d i s c o n n e c t e d from the r e s t of the e p i l a y e r i s grown i n the c h a n n e l .
100
R . D . Burnham and D. R. Scifres
CLADDING
CORE
142 Fig. 3.
Transmission micrograph of core and cladding regions.
Also, because of the directionality of the molecular beam three-dimensional waveguides have been reported by masking molecular beams during epitaxy. As early as 1972 Cho and Reinhart (20) reported on a waveguide structure made by masking with a 50 ~m tungsten wire. The AI effusion cell is directed perpendicular to the substrate so that the wire can form a sharp shadow with a width approximately the diameter of the wire. The Ga effusion cell is situated at an angle with respect to the substrate so that the Ga flux does not fall on the same region of the substrate as does the AI flux. This allows the GaAs beam to reach behind the masking wire where the AI beam is shadowed thus resulting in a strip containing only GaAs. Tsang and llegems (21) more recently reported on single and multilayered stripe mesa waveguides with widths as narrow as i ~m using single-level or multilevel masks. These masks were made from single crystal Si wafers. Silicon was chosen as the mask material because it has a low vapor pressure and is relatively inert towards GaAs at the usual growth temperature (~580°C). This makes it possible to press the Si mask directly against the GaAs substrate without contaminating the substrate. The use of single crystalline Si substrates also allows one to generate mask openings with dimensions 51 Bm by using preferential etchants. Figure 5 shows examples of single and multiple growth through various stripe patterns on single crystal Si masks.
Integrated Optical Devices Fabricated by MBE
. ,-, _
(Jl O)
101
(IT01
= 0.91,~m
(001) ¢ 5.18~u m
(ool
! t(o01 ) =3.18Aim
t (a)
(b)
(c) Fig. 4.
( a ) a n d ( b ) show t h e c ~ o s s - s e c t i o n a l v i e w s o f MBE l a y e r s o v e r two c h a n n e l s o f w i d t h 11 a n d 29 ~m aligned along the 110 d i r e c t i o n on a ( 0 0 1 ) o r i e n t e d GaAs substrate. (c) A SEM picture of the
channel edge showing the surface quality of the as-grown crystal facets.
102
R . D . Burnham and D. R. Scifres
Fig. 5.
Optical mlcrographs of slngle-level (a) and twolevel (b) depositions.
I n t e g r a t e d O p t i c a l D e v i c e s F a b r i c a t e d by ~ E
103
The l a t e s t i n improvements f o r making 3 - d l m e n s i o n a l g u i d e s i n v o l v e s m o l e c u l a r b e a m e p i t a x i a l w r i t i n g of p a t t e r n e d GaAs e p i l a y e r s t r u c t u r e s w i t h c o n t r o l l e d t h i c k n e s s p r o f i l e s , l e n g t h w i s e v a r y i n g t h i c k n e s s e s , f e a t u r e l e s s and o p t i c a l l y smooth s i d e w a l l s ( 2 2 ) . T h i s i s a c h i e v e d by h a v i n g a f i x e d S i mask w i t h p a t t e r n e d o p e n i n g s between t h e e v a p o r a t i n g s o u r c e s and t h e s u b s t r a t e w h i l e t h e s u b s t r a t e i s moved i n a p l a n e p a r a l l e l t o t h e mask d u r i n g growth ( s e e F i g . 6 . ) . Y MOTION
~/MOVING SUBSTRATE I'": '..".": . : . - ~ = . " l--_--_' ----.
((J
,:~'.: ".: :." .'. : = t . , ,','." " . ' . . ' .
r~
I....'...
~'.".'.".'" %'..-.
i.'.'.
-- - - - -.'_--_..__.~ ---
~ ~.[.'.:]
-----
--l~: .__._
-~-X MOT~N
Fig. 6.
B a s i c a r r a n g e m e n t used f o r a c h i e v i n g MBE w r i t i n g .
CaAs e p i l a y e r s t r u c t u r e s w r i t t e n b e h i n d a s e r i e s o f s q u a r e mask o p e n i n g s w i t h d i f f e r e n t s i z e s when t h e m a s k - s u b s t r a t e s e p a r a t i o n i s s m a l l i s shown i n F i g s . 7 a - d . The v a r i o u s t e c h n i q u e s m e n t i o n e d f o r f a b r i c a t i n g waveguides s h o u l d g i v e t h e r e a d e r a flavor for the high degree of versatility t h a t i s a c h i e v a b l e by MBE. I t i s i m p o r t a n t to p o i n t o u t t h n t m a n y o t h e r t y p e s o f waveguide s t r u c t u r e s t h a t have been f a b r i c a t e d by o t h e r growth t e c h n i q u e s can i n p r i n c i p l e be f a b r i c a t e d by MBE.
5.
DISCRETE L A S E E S
Ever s i n c e t h e f i r s t p - n j u n c t i o n GaAs i n j e c t i o n l a s e r s were d e m o n s t r a t e d i n 1962, c o n s i d e r a b l e e f f o r t h a s been expended i n t r y i n g t o r e f i n e t h e s e m i c o n d u c t o r l a s e r to t h e p o i n t t h a t i t i s no l o n g e r j u s t a l a b o r a t o r y c u r i o s i t y b u t a u s e f u l d e v i c e . P r o g r e s s was slow u n t i l Cal_xAlxAs DH l a s e r s were d e v e l o p e d i n 1970 ( 2 3 , 2 4 ) . These DH l a s e r s d e c r e a s e d t h e room t e m p e r a t u r e t h r e s h o l d c u r r e n t two o r d e r s o f magnitude o v e r t h a t o f i D / t i a l l a s e r s made o n l y of CaAs. T h i s tremendous r e d u c t i o n i n t h r e s h o l d c u r r e n t i s a t t r i b u t e d to t h e l a r g e r e f r a c t i v e i n d e x and bandgap d i f f e r e n c e between t h e t h i n Cat_xAlxAs ( a c t i v e ) l a y e r and t h e o u t e r two Cal_yAlyAa l a y e r s when y - x • O.15 and x ~ 0 . 2 5 . The l a r g e r e f r a c t i v e i n d e x d i f f e r e n c e s e r v e s to s i g n i f i c a n t l y c o n f i n e t h e l i g h t g e n e r a t e d i n t h e Cal-xAlxAs ~the a c t i v e r e g i o n ) w h i l e t h e w i d e r bandgap o f t h e Gal-yA1vAs c a u s e s t h e i n j e c t e d c a r r i e r s i n t h e Cal_XAlxAs r e g i o n t o be s l m i l a r l y c o n f i n e d . F o r p r a c t i c a l CW l a s e r s i t i s a l s o n e c e s s a r y t o r e s t r i c t t h e c u r r e n t t o a n a r r o w s t r i p e to l i m i t t h e a c t i v e r e g i o n l a t e r a l l y ( 2 5 ) .
104
R . D . Burnham and D. R. Scifres
Fig. 7.
MBE mask writing through a series of sizes of square mask openings.
Discrete lasers are generally made by cleaving the faces of the semiconductor crystals These cleaved facets form plane, parallel mirror-like surfaces which reflect a portion of the light back into the region of the p-n junction. The reflected light is amplified, and the energy density within the active waveguide of the laser continues to build up to produce an intense laser beam. The first GaAs-Gal_xAlxAS discrete cleaved facet DH lasers prepared by MBE were reported by Cho and Casey (26). The threshold currents of these lasers were 13 times higher than similar stuctures grown by LPE. It was found that the threshold ratio could be decreased to 1.4 if the samples were annealed at 750°C or above. Contamination of the GaAIAs layers was thought to be the major problem. Later Cho et aZ. (27), reported CW laser operation at temperatures as high at lOP C. The threshold currents were still about 1.4 times higher but they did not need to be annealed. This was achieved by eliminating more of the hydrocarbon and water vapor sources in the epltaxial growth system by using only pyrolytic BN effusion cells and installing an additional liquid-nitrogen-cooled panel around the susbstrate area. Also, the AI concentrations at all the GaAs-Gal_xAlxAs interfaces were graded in order to reduce the lattice mismatch at the interfaces because of the reduced MBE growth temperature 580°C. Another problem with making GaAs and GaAIAs lasers by MBE was to find suitable dopants. Impurities like Zn (28), Cd (28) and Mg (29) have high vapor pressure at the growth temperature compared to the group-Ill elements. Thus, they tend to have very short absorption lifetimes on the surface and re-evaporate before any significant incorporation in the grown layer can occur. Te (30,31) also has a high vapor pressure at the growth temperature compared to the group-Ill elements but it has a peculiar problem in that it tends to accumulate on the surface giving nonuniform doping and poor surface morphology. Apparently a stable surface compound (GaTe) is formed. A significant increase in sticking coefficient can be realized by using an ionized dopant beam. Naganuma et aZ.
Integrated Optical Devices Fabricated by MBE
106
(32,33), reported that GaAs layers could be doped heavily p-type -5 X 1019 (cm -3) Zn by ionizing the Zn atoms and weakly accelerating them much like ion implantation. The weakly accelerated Zn ions can penetrate just beneath the growing surface ~10-20
A, and be incorporated i n t o the grown layer by successive growth.
For amphoterlc
impurities such as the group-IV elements in GaAs, the site distribution ratio appears to be determined (34) by surface stoichiometry during growth. As a result, the elements Si (35), Ge (34,35) and Sn (35) are incorporated primarily as donors under the usual As-rich growth conditions with As-stabilized surface structure, whereas for the case of Ge it has been shown that the same element is incorporated primarily as an acceptor under Ga-stabilized growth conditions. So far, however, it has been difficult to achieve routine mirror-like surface appearance under Ga-stablized growth conditions. Ilegems 8t GZ. (36), reported using Mn as a p-type dopant. Even though Mn has a high sticking coefficient it is a ~iiO meV deep acceptor and thus doping levels of only 1018 cm -3 could be achieved. A much more promising impurity is Be (37) where at least 3 x 1019 cm -3 Be can be incorporated in GaAs and Gal_xAl xAs (up to x ~ 0.33). In fact, the Be doping level is simply proportional to the Be arrival rate. At this point in time Sn appears to be the most attractive n-type dopant and either ionized Zn or Be are attractive for heavy p-type doping. Heavy doping is necessary in order to reduce the series resistance of the device to minimize thermal heating which would allow for CW laser operation. We have talked about oxygen contamination in GaAIAs as being very deleterious for making DH Gal_yAiFAs/Gal-xAlxAs lasers. There appears to be a promising aspect to oxygen doped GaAIAs layers and that is they can be made insulating. Casey 8t aZ. (38) demonstrated that a lattice matched single crystal insulator-semiconductor heterojunction can be grown by doping Ga0.5AI0.5As layer with oxygen by MBE. This could have significant impact on monolithic device fabrication. For example, this could eliminate the need for an earlier technique developed by Cho and Ballamy (39) for planar technology in which the active device positions are defined by windows in SiO 2 layers which are deposited prior to MBE growth. By this process, polycrystalline and monocrystalline GaAs are selectively grown producing a pattern which consists of areas of device quality single-crystal material isolated by semiinsulating polycrystalline material. In fact, Lee and Cho (40) reported on single transverse mode injection lasers with embedded stripe layers using an SiO 2 mask. A scanning electron-beammicrograph of the cleaved end face of the laser view at an angle is shown in Fig. 8. Along the same llne Cho et al. (41), demonstrated that semiconducting materials consisting of inlays of different doping or composition in selective areas may ha grown by MBE. They showed it is possibleto fill an etched hole with MBE without formation of voids resulting from facet growth as observed in other growth techniques, Optically pumped laser oscillation from quantum states in thin multilayer heterostructures have been observed (42,43). Energy levels of electrons and holes in a very thin GaAs layer which is sandwiched between wider bandgap layers of Ga0.gAI0. 2 As consist of a discrete number of bound states within the well. The MBE growth technique not only allows for the formation of rectangular potential wells but by programming the variation of the AI content essentially any arbitrary potential can be repeatedly deposited. Lasers taking advantage of quantum size effects may be able to add a whole new dimension to the improvement of laser properties.
6.
LASER INTEGRATION
In order to take full advantage of the concepts of integrated optics it is necessary to fabricate laser structures which can be integrated onto the same substrate as are the other optical and electronic components. Unfortunately, the conventional
106
R. D. Burnham and D. R. Scifres
egrowth 2.5/~m ~
/~m m]
5/~m)
m~
Fig. 8.
SEM of the cross-section of an embedded stripe laser.
method of fabrlcat~ng the mirrors on a semiconductor laser involves cleavage of the crystal. Thus, the laser chip must be broken or cleaved off the substrate in order to provide optical feeback. Within the last several years there have been a number of methods developed for fabricating mirrors or reflecting elements which are compatible with monolithic thin film technology. The first to be realized was distributed feedback (DFB) (44) in which an optical grating is used to diffract incoming light back upon itself. These original devices were grown by LPE. If the period of the grating A is an integer multiple of half-wavelengths of light in the guide (A&) a small amount of light is reflected back into the waveguide conherently at each corrugation, the sum providing sufficient feedback to operate the laser. Distributed-feedback heterostructure lasers have several advantages as discrete devices over conventional cleaved-n~rror lasers. First, they generate a reproducible laser wavelength, which is four times less sensitive to thermal variation than is a conventional cleaved mirror laser (45,46). Second, the laser usually emits in a narrow-bandwldth single-longitudinal (47,48) mode. Third, as mentioned in part A of this section, the grating in a DFB laser also couples out beams which have approximately IOO times improved directionality (ll) over those emitted through the cleaved-end mirror or a conventional laser diode. Thus, these DFB devices, in addition to serving as sources for integrated optical circuits, may also prove useful a s s t a n d - a l o n e d i s c r e t e d e v i c e s . One p r o b l e m t h a t d o e s e x i s t w i t h DFB l a s e r s is that non-radiatlve recombination centers are introduced by the grating. One solution is to physically separate the corrugations from the active region (11). Casey et uS. (49), found that ion milling of the corrugations did not affect the radiative recombination as long as the corrugations were removed 0.30 ~m or more from the layer confining the carriers. This involved a hybrid LPE and MBE growth process (see Fig. 9). MBE is used to grow over the corrugations since growth on a GaAIAs surface is not reproducible using LPE.
Integrated Optical Devices Fabricated by MBE
107
AS sAS
I As ~,S
5p.m
Fig. 9.
SEM cross-section of a DFB laser.
It was suggested by S. Wang (50) that instead of gratings being used to provide d i s t r i b u t e d r e f l e c t i o n t h r o u g h o u t t h e l a s e r c a v i t y , t h e y could be p u t a t the ends of the c a v i t y to a c t as m i r r o r s . L a s e r s of t h i s type a r e c a l l e d DBR ( D i s t r i b u t e d Bragg R e f l e c t i o n ) l a s e r s , and they appear to have more promise than DFB l a s e r s . The removal of the g r a t i n g from the a c t i v e r e g i o n has t h e o b v i o u s advantage of e l i m i n a t i n g the n o n r a d i a t i v e r e c o m b i n a t i o n a s s o c i a t e d w i t h i t , t h u s a l l o w i n g s i m p l e r s t r u c t u r e s and s i m p l e r p r o c e s s i n g f o r the DBR l a s e r . The e x c e l l e n t l a y e r t h i c k n e s s and c o m p o s i t i o n c o n t r o l a f f o r d e d by MBE should prove q u i t e v a l u a b l e i n f a b r i c a t i n g DBR l a s e r s s i n c e such c o n t r o l makes i t p o s s i b l e to d e t e r m i n e a p r i o r i the c o r r e c t g r a t i n g s p a c i n g . The DBR l a s e r should a l s o have a t l e a s t as good f r e q u e n c y and mode c o n t r o l as the DFB l a e r . I n t e g r a t i o n of DH l a s e r s by f a b r i c a t i n g d i s c r e t e p a r a l l e l r e f l e c t o r s w i t h o u t c l e a v i n g the crystal has also been achieved by selective ion sputtering (51), chemical etching (52-53) and epitaxy (54). Although these devices were not grown by MBE, they could have easily been with the possible exception of Ref. 54 since it involves facet formation along with selective epitaxy.
7.
COUPLERS
As mentioned earlier the interconnection of optical components via waveguides is the central problem of monolithic integrated optics and is at the mercy of the crysal grower and the growth technique he chooses to use. MBE appears to be particularly suitable for making structures which allow coupling from one waveguide to another. Conwell and Burnham (11) d i s c u s s e s s e n t i a l l y f i v e k i n d s of c o u p l e r s which a r e : p r i s m , g r a t i n g , b u t t and c o m p o s i t i o n a l , t a p e r e d and d i r e c t i o n a l or e v a n e s c e n t . Prism, g r a t i n g , and b u t t c o u p l e r s a r e m a i n l y used to c o u p l e l i g h t out of t h e o p t i c a l c i r c u i t and w i l l n o t be d i s c u s s e d i n t h i s p a p e r . The t a p e r e d c o u p l e r has r e c e i v e d c o n s i d e r a b l e a t t e n t i o n f o r c o u p l i n g between waveguides. A t a p e r e d c o u p l e r i s made by t a p e r i n g down the end o f a waveguide. When a g u i d e mode t h a t has been u n d e r g o i n g s u c c e s s i v e r e f l e c t i o n s w i t h the a n g l e 8m r e a c h e s t h e t a p e r e d r e g i o n i t s r e f l e c t l o n s
108
R.D.
Burnham and D. R. Scifres
take place at progressively smaller 0m because of the shape of the taper, until 0m becomes smaller than the critical angle. Beyond that point it is no longer confined by the guide, and is refracted into an adjacent waveguide of slightly lower index of refraction. Taper couplers are quite easily made by MBE and are extremely uniform, gradual and reproducible (see Fig. IO). Prior to starting a growth which is to have a taper an 0.25 mm thick knife-edged mask is swung into place i or 2 mm above the surface of the sample. This masks a portion of the substrate from~the Ga and As sources. A GaAs layer grows on the unmasked regions, terminated by linear tapers -200 ~m wide which grew in the penumbra areas of the mask edge. Merz and coworkers report (55) that coupling efficiencies approaching 100% can be routinely achieved by MBE while efficiencies of only ~50% have been achieved by LPE. Composition coupling involves the transition from one composition to another along a waveguide. Various types of composition coupling schemes have been reported, however, none involves MBE. With the larger repertoire of MBE growth techniques there should be no problem in making composition couplers by MBE. On the other hand, with taper couplers approaching 100% efficiency there does not seem to be much incentive to develop composition couplers. When two identical guides are sufficiently close (within about a wavelength of light) that the exponentially decaying (evanescent) light from one extends into the other the energy will oscillate between the guides with a characteristic period L ° called the coupling lenth. Evanescent wave coupling has been used in twin guide lasers (51). Here the active waveguide of the laser is separated by a region of lower index from a passive guide separation being small enough so that the evanescent fields overlap. Twin guide lasers so far have only been fabricated by LPE but it is obvious that MBE is much more suitable fG_ its fabrication since layer thicknesses and index of refraction can be controlled so readily.
8.
SWITCHES, MODULATORS AND DETECTORS
We refer the reader again to the review article by Conwell and Burnham (II) for a more complete coverage of switches, modulators and detectors related to GaAs and its alloys. It suffices to say that although none of these components have been fabricated by MBE techniques these principles are well understood and have been fabricated by other growth techniques and MBE does not seem to present any limitations.
9.
MULTI-COMPONENT INTEGRATION
Although there has been various types of multi-component integration including the integration of a laser, waveguide and detector recently reported by Merz and Logan (56), only one involves growth by MBE. Reinhart and Cho (57) reported growth of a Gal_~lyAs-Gal-xAlxAs laser taper coupled to a passive (more transparent) waveguide grown by MBE. The structure is similar to the schematic shown in Fig. lOa except additional epitaxial layers are grown to facilitate electrical contact. Layer to is the active layer of the devices and layer t0. I to the right of the taper services as the passive waveguide. These particular structures indicated significant absorption losses.
I0.
CONCLUSIONS
MBE has certainly demonstrated that monolayer dimensional control along with a wide range of doping and compositional profiles is achievable. Because of this versatility, MBE is a promising growth technique for the fabrication of extremely compli-
slntegrated Optical Devices Fabricated by MBE
109
AIR
(o)
Fig. IO.
Tapers grown by MBE.
cated integrated optical circuits. However, considerable effort and improvement is needed to identify and suppress the sources of contamination during the growth of GaAs and especially GaAIAs. Already improved system designs are beginning to be implemented such as the incorporation of elaborate vacuum interlocks, improved cryopumping, improved substrate heaters and source heater assemblies. These steps should significantly reduce eomtamlnation during growth. Calawa (58) recently reported that the introduction of hydrogen during MBE growth of GaAs seems to significantly improve the mobility of the material. As soon as MBE grown GaAIAs can reproducibly compare with LPE grown GaAIAs, then significant advances in inte-
110
R. D. Burnham and D. R. Scifres
g r a t e d o p t i c s can be e x p e c t e d . S o l v i n g the c o n t a m i n a t i o n problem i n MBE i s s o r t of analogous to solving the waveguiding and carrier confinement problems for lasers. As was mentioned earlier, this was achieved by making DH lasers. Without it laser progress was slow. Of course, solving the GaAIAs contamination problem may, on the other hand, simply require the use of alloys which do not contain A1. For example, Miller et al. (59), recently reported on InP/GalnAs/InP lattice matched DH lasers grown by MBE. We are grateful to R. Z. Bachrach and A. Y. Cho for many helpful discussions. Acknowledgements are due H. C. Casey, Jr., T. P. Lee, J. L. Merz and W. T. Tsang for supplying photographs of their work. Special thanks go to D. Bernsen for secretarial service.
REFERENCES
i. 2. 3. 4. 5. 6. 7. 8. 9. i0. ii. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
Miller, Bell Syst. Tech. J. 48, 2059 (1969). Dupuis and P. D. Dapkus, Appl-~Phys. Lett. 32, 406 (1978). Cho and F. K. Reinhart, J. Appl. Phys. 45, 1812 (1974). Arthur and J. J. LePore, J. Va~um Sci.-~echnol. 6, 545 (1969). Cho and M. B. Panish, J. Appl I. Phys. 43, 5118 (1972). Chang, L. Esaki, W. E. Howard, R. Ludeke, and G. Schul, J. Vucuum Sci. Technol. i00, 655 (1973). R. Dingle, W. Wiegmann, and C. H. Henry, Phys. Rev. Lett. 33, 827 (1974). R. Dingle, A. C. Gossard, and W. Wiegmann, Phys. Rev. Lett. 34, 1327 (1975). A. C. Gossard, P. M. Petroff, W. Wiegmann, R. Dingle, and A. Savage, Appl. Phys. Lett. 29, 323 (1976). P. K. Tien, Rev. Mcd. Phys. 49, 361 (1977). E . M . Conwell and R. D. Burnham, Ann. Rev. Mater. Sci. 8, 135 (1978). M. K. Barnoski, ed., Introduction to Integrated optics, Plenum, New York (1974) T. Tamir, ed., Integrated Optics, Springer, New York (1975). J. L. Merz and A. Y. Cho, Appl. Phys. Lett. 28, 456 (1976). J. L. Merz, A. C. Gossard, and W. Wiegmann, Ap---pl.Phys. Lett. 30, 629 (1977). A. Y. Cho, A. Yariv, and P. Yeh, Appl. Phys. Lett. 30, 471 (197). J. C. Tracy, W. Wiegmann, R. A. Logan, and F. K. Reinhart, Appl. Phys. Lett. 22, 511 (1973). W. T. Tsang and A. Y. Cho, Appl. Phys. Lett. 30, 293 (1977). S. Nagata, T. Tanaka, and M. Fukai, Appl. Phys-f Lett. 30, 503 (1977). A. Y. Cho and F. K. Reinhart, Appl. Phys. Lett. 21, 355 (1972). W. T. Tsang and M. Illegems, Appl. Phys. Lett. 31, 301 (1977). W. T. Tsang and A. Y. Cho, Appl. Phys. Lett. 32, 491 (1978). Zh. I. Alferov, V. M. Andreev, D. Z. Garbuzov, Yu. V. Zhilgaev, E. P. Morozov, E. L. Portnoi, and V. G. Trofim, Fiz. Tekh. Poluprovodn 4, 1826 (1970). I. Hayashi, M. B. Panish, P. W. Foy, and S. Sumski, Appl. Phys. Lett. 17, 109 (1970). J. E. Ripper, J. C. Dyment, L. A. D'Asaro, and T. L. Paoli, Appl. Phys. Lett. 18, 155 (1971). A. Y. Cho and H. C. Casey, J r . , Appl. Phys. Lett. 2_55,288 (1974). A. Y. Cho, R. W. Dixon, H. C. Casey, Jr., and R. L. Hartman, Appl. Phys. Lett. 28, 501 (1976). ~?. R. Arthur, Surf. Sci. 3-8, 394 (1973). A. Y. Cho and M. B. Panish, J. Appl. Phys. 43, 5118 (1972). J. R. Arthur, Surf. Sci. 43, 449 (1974). A. Y. Cho and J. R. Arthur, Prog. Solid. St. Chem. iO, 157 (1975). M. Naganuma and K. Takahashl, Appl. Phys. Lett. 27, 342 (1975). N. Matsunaga, M. Naganuma, and K. Takahashi, Jap. J. Appl. Phy8. 16, 443 (1976) A. Y. Cho and I. Hayashi, J. Appl. Phys. 42, 4422 (1971). A. Y. Cho, J. Appl. Phys. 46, 1732 (1975). S. R. A. J. A. L.
E. D. Y. R. Y. L.
Integrated Optical Devices Fabricated by ~ E 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59.
111
M. llegems, R. Dingle, and L. W. Rupp, Jr., J. AFpl. P~8. 46, 3059 (1975). M. Ilegems, O. A~l. Phys. 48, 1278 (1977). H. C. Casey, J r . , A. Y. Cho, and E. H. N i c o l l i a n , AFpl. Phys. Left. 32, 678 (1978). A. ¥. Cbo and W. C. Ballamy, J. Appl. Phys. 4._66, 783 (1975). T. P. Lee and A. Y. Cho, Appl. Phys. Lstt. 2_.99, 164 (1976). A. Y. Cho, J. V. Dilorenzo, and G. E. Mahoney, IEEETRrn8. on Electr. Devices ED-24, 1186 (1977). J. P. van der Ziel, R. Dingle, R. C. Miller, W. Wiegnmnn, and W. A. Nordland, Jr., Appl. Phys. Left. 26, 463 (1975). R. C. Miller, R. Dingle, A. C. Gossard, R. A. Logan, W. A. Nordland, Jr., and W. Wiegmann, J. App,. Phys. 47, 4509 (1976). D. R. Sclfres, R. D. Burnham, and W. Streifer, Appl. Phys. Lett. 25, 203 (1974) M. Nakamura, K. Aiki, J. Umeda, A. Katzlr, A. Yariv, and H. W. Yen, IEEE J. QunntT~mElec~. qE-ll, 436 (1975). R. D. Burnham, D. R. Scifres, and W. Streifer, App,. Phys. Lett. 29, 287 (1976). R. D. Burnham, D. R. Scifres, and W. Streifer, IEEE J. Quantum Electr. qE-ll, 439 (1975). D. E. Scifres, R. D. Burnham, and W. Streifer, IEEE Tr~rn8. on Electr. Devices ED-22, 609 (1975). M. Ilegems, H. C. Casey, S. Somekh, and M. B. Panish, J. C r ~ s ~ G ~ t h 3_!1, 158 (1975). S. Wang, IEEE J. Q~o~t~mElectr. qE-IO, 413 (1974) Y. Suematsu, M. Y-m, da, and K. Hayashi, IEEE J. ~unt~m EZeetra. qE-ll, 457 (1975). C. Hurwitz, J. A. Rossi, J. J. Hsieh, C. M. Wolfe, Appl. Phys. Left. 27, 241 (1975). J. L. Merz and R. A. Logan, Appl. Phys. Left. 47, 3503 (1976). J. C. Campbell and D. W. Bellavance, IEEE J. Quuntw, Electr. Eg~!, 253 (1977). J. L. Merz, R. A. Logan, W. Wiegmann, and A. C. Gossard, App,. Phys. Left. 2._66,337 (1975). J. L. Merz and R. A. Logan, Appl. Phys. Left. 30, 530 (1977). F. K. Reinhart and A. Y. Cho, Appl.'Phys. Lett?--3_!l, 457 (1977). A. R. Calawa, 2OthAnnual Electronic Materials Conference, Santa Barbara, California, June 28-30, 1978. B. I. Miller, J. S. McFee, R. J. Martin, and P. K. Tien, Appl. Phy8. Left. 33, 44 (1978).
112
R . D . Burnham and D. R. Scifres
THE AUTHORS
R. D. Burnham
D.R. Scifres
Robert D. Burnham was born in Havre de Grace, Maryland, on 21 March 1944. He receivea his B.S., M.S., and Ph.D. degrees from the University of lllinois~ Urbana, in 1966, 1968, and 1971, respectively. He held an NDEA Fellowship from 1966-1969, and a General Telephone and Electronics Fellowship from 1969-1971. He has been a member of the Research Staff at the Xerox Palo Alto Research Center, Palo Alto, California, since 1971. He is currently in charge of the materials growth of III-V semiconductor lasers by molecular beam epitaxy, liquid phase epitaxy and metalorganic chemical vapor deposition. He has been a co-author of over eighty papers. His most significant contributions were in the area of distributed feedback, quaternary, buried heterostructure, and InGaP lasers. Dr. Burnham is a member of Tau Beta Pi, Eta Kappa Nu, Sigma Tau, the Electrochemical Society, and the AIME.
Integrated Optical Devices Fabricated by MBE
113
Don R. S c i f r e s was born i n L a f a y e t t e , I n d i a n a , on 10 September 1946. He r e c e i v e d t h e B.S. degree w i t h honors i n E l e c t r i c a l Engineering from Purdue U n i v e r s i t y , West L a f a y e t t e , I n d i a n a , i n 1968, and the M.S. and P h . D . ! d e g r e e s i n E l e c t r i c a l Engineering from t h e U n i v e r s i t y o f I l l i n o i s , Urbana, I l l i n o i s , i n 1970 and 1972, r e s p e c t i v e l y . While a t the U n i v e r s i t y of I l l i n o i s he h e l d a U n i v e r s i t y of I l l i n o i s Doctoral Support F e l l o w s h l p and a General Telephone and E l e c t r o n i c s F e l l o w s h l p . In 1972, he joined the Xerox Palo Alto Research Center where he is currently Manager of an Opto-Electronics Group. His research has been involved with developing semiconductor injection lasers as well as incorporating optical and electronic components in the semiconductor laser chip to achieve such integrated optical functions as distributed feedback, transverse mode control, and electronically controlled laser beam deflection. Dr. Scifres is on the Board of Editors for the journal of Fiber and Integrated Optics, is a national lecturer on "Semiconductor Lasers" for the Quantum Electronics group of IEEE, is a Fellow of the Optical Society of America, and a member of IEEE and the American Physical Society.