Brief Description of the Drawings
FIG. 1 is a cross-section of the bulk structure of this invention;
FIG. 2 is a cross-section of the bulk structure of this invention as shown in FIG. 1 wherein stripe-like openings are formed in the bulk structure;
FIG. 3A is a cross-section of a double heterojunction laser of this invention;
FIG. 3B is a cross-section of a double heterojunction buried laser device combining the laser of FIG. 3A and a portion of the support structure of FIG. 2; and
FIG. 4 is a cross-section of a multiple double heterojunction buried laser device of this invention having a plurality of the DH laser devices of FIG. 3B.
Detailed Description of a Preferred Embodiment
In making a multiple double heterojunction (DH) buried laser device of this invention a bulk structure is constructed of a plurality of semiconductor layers deposited in succession on a substrate. The bulk structure then has a plurality of stripe-like openings cut into the bulk structure by conventional photolithographic techniques. The DH lasers are grown into the stripe-like openings by depositing a plurality of semiconductor layers therein. After removal of extraneous material, electrical contact layers are deposited on the multiple DH buried laser device. At this point in the fabrication, the individual DH-buried laser devices can be separated and used individually or as an array. The bulk structure and lasers are composed of a Group III-V compound or alloys of such compounds. The present invention will be described using, in particular, InP and InGaAsP as the materials for the various layers using appropriate n-type or p-type dopants in the appropriate layers. Although various epitaxial techniques can be used for depositing the layers, a preferred technique is liquid phase epitaxy (LPE).
A suitable apparatus and method for depositing the layers by liquid phase epitaxy is shown and described in U.S. Pat. No. 3,753,801 to H. F. Lockwood et al entitled, "METHOD OF DEPOSITING EPITAXIAL SEMICONDUCTOR LAYERS FROM THE LIQUID PHASE", issued Aug. 21, 1973 which is incorporated herein by reference. Basically, the apparatus includes a furnace boat having a plurality of spaced wells in the top surface thereof and a substrate-carrying slide movable through the furnace boat and across the bottom of each of the wells. Each of the wells contains a liquid source from which a layer is to be deposited. The slide carries the substrate into each of the wells where it is brought into contact with the source to permit the layer to be deposited on the substrate or other layers already deposited. For depositing thin layers, small amounts of the solution are used in the well and a weight is placed on the solution to spread the solution completely across the substrate.
Each of the source solutions is made-up of the desired semiconductor material and a suitable conductivity modifier dissolved in a metal solvent. The solutions are heated to a temperature at which the semiconductor material and the conductivity modifier are completely dissolved in the solvent. The substrate is moved into each of the wells in succession and, while in contact with the solution in the well, the temperature of the solution is lowered to precipitate out the semiconductor material and deposit it on the surface of the substrate. The deposition of the semiconductor material also carries along some of the conductivity modifier. Prior to placing the substrate in the solution, the solution is generally saturated with the semiconductor material by means of a source wafer of the semiconductor material which is either carried by the slide ahead of the substrate or is within the well on top of the solution.
Referring to FIG. 1, bulk structure 19 includes a substrate 10 of n-type InP doped with tin. The same reference numerals will be used in all the Figures to identify identical elements. A first layer 12 of n-type InP doped with tellurium is deposited upon substrate 10. The purpose of protective first layer 12 is more fully described in U.S. patent application Ser. No. 384,291, entitled, "InP:Te PROTECTIVE LAYER FOR REDUCING SUBSTRATE DISSOCIATION," filed on June 2, 1982 by the present inventor. This application discloses that first layer 12 of indium phosphide doped with tellurium on an indium phosphide substrate substantially reduces the formation of indium droplets thereon and pitting caused by dissociation of phosphorous from the InP substrate during the LPE technique. The deposition of first layer 12 involves forming a melt solution of In, InP, and Te, homogenizing the melt solution, wiping substrate 10 with an indium solution, and depositing a layer of InP doped with Te upon substrate 10 after the indium wipe. This first layer 12 is deposited by supercooling the growth melt from about 645.degree. C. to about 635.degree. C. The concentration of Te ranges from about high 10.sup.18 to low 10.sup.19 atoms per cubic centimeter.
Following layer 12, a second layer 14 of p-type InP doped with zinc is deposited therein. A third layer 16 of n-type InP doped with tellurium is deposited on second layer 14. When a positive voltage is applied across a laser 34 as shown in FIG. 3A, the bulk p-n junction will be reverse biased and stop the flow of current. This feature is required to confine current flow through structures to be defined hereinafter.
The tellurium doping in first layer 12 and third layer 16 substantially block the diffusion of zinc from second layer 4. The diffusion barrier to zinc is more fully described in U.S. patent application, Ser. No. 372,364, entitled, "DIFFUSION BARRIER FOR LONG WAVELENGTH LASER DIODES," filed on Apr. 27, 1982 by the present inventor. These layers include Group VI elements such as sulfur, selenium, or tellurium for dopants in the n-type layers. These dopants when properly introduced into the n-type layers, which for example can be heterojunction layers in a laser, provide a higher donor concentration than the acceptor doping levels which are controlled by the amount of zinc added by weight to the p-type layer solution. These n-type layers 12 and 16 act as a barrier to the diffusion of acceptors into the n-type layers. The concentration of Group VI dopants is about high 10.sup.19. Referring to FIG. 1, it is seen that first layer 12 and third layer 16 are doped with tellurium. Second layer 14 having zinc is between layers 12 and 16 thus there is no diffusion of zinc therefrom and there is an abrupt interface between first, second, and third layers 12, 14, and 16 respectively. A capping fourth layer 18 of n-type InP doped or undoped is deposited on third layer 16.
After the semiconductor layers 12 to 18 are deposited on substrate 10, bulk structure 19 is removed from the furnace boat. A masking layer 20, shown in FIG. 2, such as silicon dioxide, is then deposited on the surface of fourth layer 18. Silicon dioxide layer 20 can be deposited by the conventional technique of chemical vapor deposition, such as heating silane in an atmosphere containing oxygen at a temperature at which the silane decomposes to form silicon oxide which then deposits on fourth layer 18. As shown in FIG. 2, a plurality of stripe shaped openings 22 are then formed in masking layer 20. This is achieved by forming a photoresist layer on masking layer 20 and then developed to form openings 22 using conventional photolithographic techniques, and then etching away the exposed portion of masking layer 20. Further etching removes semiconductor material from fourth, third, second, and first layers 18, 16, 14, and 12, respectively, to a depth where a bottom 24 of openings 22 is in n-type layer 12. Bottom 24 can not stop in p-type layers because a reverse biased junction would result from the semiconductor material deposited in openings 22 or result in a higher device series resistance, as will be shown.
Referring to FIG. 3A double heterojunction laser 34 such as the one shown are formed in openings 22. A related structure is shown and described in U.S. Pat. No. 4,233,090, mentioned above. In the present DH laser 34, an n-type InP first layer 26 is deposited on a substantially planar bottom 24 of openings 22. First layer 26 is doped with tellurium. As noted above, InP doped with tellurium served to stop the dissociation of phosphorous from InP thus preventing the formation of indium droplets. If bottom 24 occured on the surface of substrate 10 of InP, the above feature especially prevents the indium droplets; otherwise, layer 26 is substantially adjacent first layer 12 of InP doped with tellurium. A laser region second layer 28 forms the active region of laser 34, second layer 28 is normally about 2000 angstroms thick. In this embodiment, a quaternary alloy of InGaAsP forms laser region second layer 28. If other alloys or materials are used to form second layer 28 such as GaAs, the adjacent heterojunction layers can be composed of AlGaAs, for example. A heterojunction third layer 30 of p-type InP doped with the acceptor zinc is deposited on laser region second layer 28. A capping fourth layer 32 of p-type InGaAsP doped with zinc is deposited on heterojunction third layer 30 which is at a lower bandgap and higher acceptor concentration so that zinc diffusion therefrom increases the acceptor concentration of the underlying p-layer surface of laser 34 for lower series resistance.
FIG. 3B shows DH buried laser 34 in bulk structure 19 forming a DH buried laser 35. The width of laser 34 is about 0.5 mils. Of critical importance is the requirement that laser region second layer 28 abut highly doped third layer 16 of bulk structure 19. Highly doped layers 16, 26 and 30 enhance both optical and electrical confinement due to controlability of their energy bandgap and refractive index properties which can also be varied. This configuration confines current to flow through laser 34 when it is buried in bulk structure 19 and lasing light is also confined to laser region second layer 28 therein because of the changes in the index of refraction at the interfaces to layer 28. Possible dopants to achieve high dopant levels are Group VI elements such as tellurium, selenium, or sulfur.
After DH lasers 34 are deposited into openings 22, masking layer 20 is removed with a suitable etchant. The etchant used should have little or no effect on capping fourth layer 18 of bulk structure 19 and capping fourth layer 32 of laser 34.
Referring to FIG. 4, metal contacts 39 and 45 are then deposited on the surface of capping fourth layer 18 of bulk structure 19, capping fourth layer 32 of lasers 34, and substrate 10. Metal contacts 39 and 45 may be of any metal which forms an operable ohmic contact with the particular material involved. For n-type InP substrate 10, a metal contact 45 includes a layer 44 of tin alloyed to InP substrate 10, followed by a layer 42 of nickel and lastly coated with a layer 40 of gold. These have been found to be suitable. For contact 39, a layer 36 of nickel is deposited onto fourth layers 18 and 32 and covered with a thin layer 38 of gold is suitable.
In use, a multiple DH buried laser device 52 can be operated as an array wherein a positive lead 48 and a negative lead 46 supply current to operate lasers 34 which can be connected to a plurality of optical fibers (not shown).
In order to operate the laser 34 as a single device, individual DH-buried lasers can be cleaved in a conventional manner along dotted lines 50. An alternative method could place conventional current barriers (not shown) between lasers 34 in bulk structure 19. Thus individual input lines (not shown) could operate each laser 34 independently of one another while these lasers 34 remain on a common substrate 10.
Obviously, many modifications and variations of the present invention are possible in light of the above teachings and it is therefore understood that, within the scope of the disclosed inventive concept, the invention may be practiced otherwise than specifically described.